Hostname: page-component-7dd5485656-2pp2p Total loading time: 0 Render date: 2025-10-31T15:23:03.603Z Has data issue: false hasContentIssue false
  • English
  • Français

On the Identity of Shang Di and the Origin of the Concept of a Celestial Mandate (TIAN MING)*

Published online by Cambridge University Press:  26 March 2015

Sarah Allan
Sarah Allan*
Affiliation:
Dept. of Asian and Middle Eastern Languages and Literatures, HB 6191 Dartmouth College Hanover, NH 03755 USA
Get access Rights & Permissions [Opens in a new window]

Extract

This article reexamines the hypotheses of Guo Moruo and H.G. Creel that Shang Di was the high god of the Shang and Tian, that of the Zhou. It proposes that Shang Di was originally the spirit of the pole star. As such, it was the one celestial body which was higher than the ten suns, with whom the Shang ancestors were identified. Tian was not a high god, but quite literally, the sky. The sky was the location of the Shang Di and the other ancestral spirits, so it came to serve as a euphemism for Shang Di or, more broadly, for Shang Di and all the celestial phenomena and spirits who were under his aegis. The primary distinction between the Shang and Zhou was not that Shang Di was particular to the Shang, but that the Shang rulers identified themselves with the ten suns. Shang Di, as the pole star, was acknowledged by both Shang and Zhou as the highest of the spirits. Tian, as the sky, was understood primarily as the celestial bodies that inhabit it. As in later time, the sky was a spiritual force associated with patterns of time, which were revealed in the movements of the celestial bodies. Thus, the original tian ming (“celestial mandate” or “mandate of heaven”) was, quite literally, an astronomical sign, a “command” seen in the sky during the reign of Wen, whose son Wu founded the Zhou Dynasty.

Information

Type
Research Article
Information
Early China , Volume 31 , 2007 , pp. 1 - 46
Copyright
Copyright © Society for the Study of Early China 2007

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Article purchase

Temporarily unavailable

Footnotes

*

I would like to thank Constance Cook, the members of the Early China Seminar at Columbia University, and the anonymous readers for Early China, for many helpful suggestions. I would also like to thank Xing Wen for providing the handwritten characters in the text.

References

1. Allan, Sarah “Tian as Sky: The Conceptual Implications,” in En suivant la Voie Royale: Mélanges offerts en hommage à Léon Vandermeersch, ed. Gernet, Jacques and Marc Kalinowski (Études thématiques, 7, Paris: École Française d'Extrême-Orient, 1997), 225–30Google Scholar.

2. Liya, Gu [Herrlee.G. Creel], “Shi tian Yanjing xuebao 18 (1935), 59–71 Google Scholar; Moruo, Guo Xian-Qin tiandao guan zhi jinzhan (Shanghai: Shangwu, 1936), 1–18 Google Scholar; Creel, Herrlee G., The Origins of Statecraft in China, vol. 1: The Western Zhou Empire (Chicago: Chicago University Press, 1970)Google Scholar, Appendix C: “The Origin of the Deity Tien,” 493–506. Creel, also promulgated this theory in The Birth of China: A Survey of the Formative Period of Chinese Civilization (New York: Frederick Ungar, 1937)Google Scholar.

3. See, for example, Keightley, David N., The Ancestral Landscape: Time, Space, and Community in Late Shang China (Berkeley: Institute of East Asian Studies, 2000), 123 Google Scholar. That Keightley did not feel the need to provide a citation is an indication of the extent to which this hypothesis is now generally held to be valid. I have also previously accepted it in earlier writings; see, for example, Allan, Sarah, The Shape of the Turtle: Myth, Art and Cosmos in Early China (Albany, New York: State University of New York Press, 1991), 39 Google Scholar, with n74, and 59, and thereafter.

4. See Sarah Allan, “Tian as Sky,” 225–30.

5. Baocong, Qian , “Tai Yi kao ,” Yanjing xuebao 12 (1932), 2449–78Google Scholar; see also Allan, Sarah, “The Great One, Water, and the Laozi: New Light from Guodian,” Toung Pao 89.4/5 (December 2003), 237–85CrossRefGoogle Scholar.

6. Sun, Xiaochun and Kistemaker, Jacob, The Chinese Sky during the Han: Constellating Stars and Society (Leiden: E.J. Brill, 1997), 82–83 Google Scholar. The Great Di (da di) is identified with Tai Yi in the Suoyin commentary of the Shiji (Beijing: Zhonghua, 1969), 27.1289–90Google ScholarPubMed (“Tianguan shu” ).

7. Shiji, 27.1291. Zhongguo guxing tu , ed. Meidong, Chen (Shenyang: Liaoning jiaoyu, 1996), 6, 39 Google Scholar.

8. For the four fang as quadrates rather than simply directions, see Allan, The Shape of the Turtle, 75–76. For discussion of the use of the dipper as a pointer in cosmographs, see Sarah Allan,“ The Great One, Water, and the Laozi,” 246–49.

9. See Puett, Michael J., To Become a God: Cosmology, Sacrifice and Self-Divinization in Early China (Cambridge: Harvard University Press, 2002)Google Scholar, for a discussion of ideas of self-divinization in the Warring States and Han Periods.

10. Pankenier, David W., “A Brief History of Beiji (Northern Culmen), with an Excursus on the Origin of the Character di ,” Journal of the American Oriental Society, 124.2 (2004), 211–36CrossRefGoogle Scholar; Pankenier, , “The Cosmo-political Background of Heaven's Mandate,” Early China 20 (1995), 121–76CrossRefGoogle Scholar. Although the star closest to the pole changed over periods of thousands of years because of the precession of the equinoxes, as Pankenier discusses in the latter paper, there was always a Pole Star by which buildings and tombs were oriented. This is clear from Pankenier's discovery that when the nearest star to the pole changed in the late Neolithic and early Bronze Age (the early second millennium b.c.e.), the north-south orientation of buildings also changed.

11. Pankenier, “The Cosmo-political Background of Heaven's Mandate,” 164–71.

12. Pankenier, , “Astronomical Dates in Shang and Western Zhou,” Early China 7 (1981–82), 2–37 Google Scholar. See also his two-part article, “The Bamboo Annals Revisited: Problems of Method in Using the Chronicle as a Source for the Chronology of Early Zhou,” Bulletin of the School of Oriental and African Studies 55.2 (1992), 272–97CrossRefGoogle Scholar; 55.3 (1992), 498–510.

13. Pankenier, “The Cosmo-political Background of Heaven's Mandate,” 122.

14. Earlier versions of these arguments may be found in Allan, Sarah, “Sons of Suns: Myth and Totemism in Early China,” Bulletin of the School of Oriental and African Studies 44.2 (1981), 290–326 CrossRefGoogle Scholar, and “The Myth of the Xia Dynasty,” Journal of the Royal Asiatic Society 2 (1984), 242–56Google Scholar.

15. This argument does not preclude the possibility that there were a people, who called themselves the “Xia” and had suffered a defeat by the ancestors of the late Shang kings, or that the name “Xia” referred to inhabitants of Erlitou , Yanshi , in Henan Province, but, at present, there is no evidence for such an identification. For discussion of the role of this site, see my article, “Erlitou and the Foundation of Chinese Civilization,” Journal of Asian Studies, 66.2 (2007), 461–97CrossRefGoogle Scholar.

16. Pankenier argues for the authenticity of the records in the “new text” Annals on the grounds of what he takes to be a remarkable coincidence between the dates of the planetary conjunctions recorded as marking the foundation of the Xia, Shang, and Zhou Dynasties, and real astronomical events. However, he also calculates the dates for the founding of these dynasties on the basis of the real astronomical events. This argument impresses me as circular and I see no evidence of a literary or oral context in which such records would have been transmitted before the Western Zhou Dynasty. The recent publication of a Warring States period Chu script bamboo slip manuscript in the Shanghai Museum collection, Bao Shuya yu Xi Peng zhi jian , which refers to a solar eclipse as late as the mid-seventh century b.c.e. that cannot be matched with the historical records of actual celestial phenomena, should strike a cautionary note about the astronomical records in Warring States texts. See Shanghai bowuguan cang Zhanguo Chu zhushu , ed Chengyuan, Ma (Shanghai: Shanghai guji, 2000–), 5.165–91Google Scholar; Xueqin, Li “Shi shi Chu jian Bao Shuya yu Xi Peng zhi jian ” Wenwu 2006.9, 90–96 Google Scholar.

17. See also David Pankenier, “The Cosmo-political Background of Heaven's Mandate,” 220–24.

18. Summary accounts of Shang Di's role and powers, upon which the following discussion is based, include Mengjia, Chen , Yinxu buci zongshu (Beijing: Kexue, 1956), 561–82Google Scholar; Houxuan, Hu , “Yin buci zhong de shang di he wang di ,” Lishi yanjiu 1959.9, 23–50, and 1959.10, 89–110Google Scholar; Kunio, Shima Inkyo bokuji kenkyū (Hirosaki: Hirosaki Daigaku Chūgoku Kenkyūkai, 1958), 188–216 Google Scholar; Kiyoshi, Akatsuka , Chūgoku kodai noshūkyō to bunka: In ōchō no saishi (Tokyo: Kadokawa, 1977), 470–530 Google Scholar; Michiharu, Itō, “Religion and Society,” in Michiharu, Itō and Takashima, Ken'ichi, ed. Arbuckle, Gary F., Studies in Early Chinese Civilization (Osaka: Kansai Gaidai University, 1996), 1.4–7Google Scholar.

19. Hu Houxuan “Yin buci zhong de shang di he wang di,” 1959.9,48; Chen Mengjia, Yinxu buci zongshu, 572; Allan, The Shape of the Turtle, 101; Pankenier, “The Cosmo- political Background of Heaven's Mandate,” 171 and n101.

20. In oracle bone inscriptions, the object often appears before the verb, sometimes marked by wei or ; this leads to some ambiguity in the fragments, but I have not found any clear exceptions to this observation.

21. See Shima Kunio, Inkyo bokuji kenkyū, 188–216; Eno, Robert, “Was there a High God Ti in Shang Religion?” Early China 15 (1990), 1–26 CrossRefGoogle Scholar.

22. See Sarah Allan, “Sons of Suns,” 290–326. Most of this material is included in The Shape of the Turtle, 19–56, and earlier research is cited in both works. Particularly important is Donggui, Guan , “Zhongguo gudai shi ri zhi shenhua zhi yanjiu” Zhongyang yanjiuyuan Lishi yuyan yanjiusuo jikan 33 (1962), 287–330 Google Scholar. See also Itō Michiharu, “Religion and Society,” in Itō Michiharu and Ken'ichi Takashima, Studies in Early Chinese Civilization, 1.40, for the identity of the suns and the ancestors and further citations.

23. Shi ji, 32.1481 (“Qi Taigong shijia” ).

24. Shi ji, 2.84–88 (“Xia benji” ); 3.98–105 (“Yin benji” ); 4.128–69 (“Zhou benji” ).

25. Jiaguwen heji , ed. Moruo, Guo et al. (Beijing: Zhonghua, 1982)Google Scholar (hereafter, Heji), 35355,35356, 36168, 36169, 36176, 36177.

26. Chen Mengjia Yinxu bud zongshu, 440, argued that di was a temple altar or tablet, citing the “Qu li” chapter of the Liji , which says that “the tablet (zhii ) set up in the temple is called di.” While it is certainly possible that the status marked by the title di had a physical representation in the temple, the use of the adjective wenwu (“cultural and martial”) before di in the oracle bone inscriptions suggests that the term di referred to the ancestor rather than to a tablet.

27. Xigui, Qiu “Guanyu Shang dai de zongzu zuzhi yu guizu he pingmin liangge jieji de chubu yanjiu” in Xigui, Qiu, Gudai wenshi yanjiu xintan (Nanjing: Jiangsu guji, 1992), 296–342 Google Scholar (see especially pp. 298–300); rpt. from Wenshi 17 (1982). Hu Houxuan, “Yin buci zhong de shang di he wang di,” argued that di is not used as an ancestral title until the third period (Lin Xin , Geng Ding ), but Qiu points out that in first period inscriptions, Wu Ding's father, Fu Xiao Yi , is called Fu Yi Di Qiu relates di to di . Mingchong, Huang “Jiaguwen, jinwen suojian yi shi ri mingming zhe de jitong ‘qubiezi’” Zhongyang yanjiuyuan Lishi yuyan yanjiusuo jikan 76.4 (2005), 625–709 Google Scholar, also takes di as a status marker and discusses its function in detail. Huang argues that di may in some cases be female, but his argument here relies upon parsing the graphs Wen di mu ri xin after the second graph rather than the third in the inscription on the Yu fangding .

28. If huang zu di kao consists of two parallel phrases, di would modify kao. Chen Chusheng cites the Han Dynasty work, Du duan (attributed to Cai Yong ), which has the gloss, . This graph may be read as da (“great”) or as tai (also written as ), commonly used as an epithet indicating ancestral status. See Chusheng, Chen, Jinwen changyong zidian (Xi'an: Shaanxi renmin, 1989), 10 Google Scholar. Examples of this phrase are found in three inscriptions: Yin Zhou jinwen jicheng (Beijing: Zhonghua, 1984) 5.2705 Google Scholar (early Western Zhou), 7.4097 (middle Western Zhou) and 5.2743,2744 (late Western Zhou). The expression is also found in 8.4129 (late Western Zhou); however, in this inscription the graph representing di is written with the mouth at the bottom. This graph is normally used for the di rite.

29. Jinwen gulin , ed. Fagao, Zhou (Hong Kong: Xianggang Zhong-wen daxue, 1974), 235–61Google Scholar (1.0035).

30. Cook, Constance “Auspicious Metals and Southern Spirits: An Analysis of Chu Bronze Inscriptions,” Ph.D. dissertation, University of California at Berkeley, 1990 (Ann Arbor: University Microfilm), 107 Google Scholar.

31. References to Di Yi are found in the Zhou Yi , “Gui mei” and “Tai” hexagrams; Shang shu , Zhou shu : “Duo fang” and “Jiu gao” . There are no references in the “Song” or “Da ya” sections of the Shi jing . This analysis is based upon the Academia Sínica database found at http://www.sinica.edu.tw.

32. See Shi ji, 38.1607 (“Song Weizi shijia” ). Weizi Qi is called Weizi Kai in the Shi ji, because of the taboo placed on the personal name of Han Jingdi , r. 156–141 b.c.e.

33. Chen Mengjia Yinxu buci zongshu, 580.

34. Qiu Xigui “Guanyu Shang dai de zongzu zuzhi yu guizu he pingmin liangge jieji de chubu yanjiu,” 298–300.

35. See Allan, The Shape of the Turtle, 31–36, for a discussion of these figures.

36. Heji, 1657.

37. Heji, 1402.

38. In my analysis of the graphs for the attached tables, I have generally assumed that the graph refers to the performance of the d/-rite unless it is in the position of a noun or an ancestral title, but there are many examples that are not clear. Of particular interest are the inscriptions recorded in Heji 22073 and 22075, which do record offerings to Di, including not only rites to be performed but animals to be offered (see Table 3). It is noteworthy, however, that the graphs in these two Li diviner-group inscriptions are highly unusual variants. See Yinxu jiagu keci leizuan , ed. Xiaosui, Yao and Ding, Xiao (Beijing: Zhonghua, 1989), 422 Google Scholar, which classifies the graph as di . See also Xiaotun nandi jiagu ed. yanjiusuo, Zhongguo shehui kexueyuan kaogu (Beijing: Zhonghua, 1983), 1147 Google Scholar (hereafter, Xiaotun).

39. A challenge to the predominant view that few, if any, offerings were made to Shang Di is found in Shi Shaohua , Zhao Dehua , Zheng Xinzhu , and Zhao Fuyong , “Luelunbuci zhong suojian ji shang di zhi li” , Zhongguo guwenzi daxi: jiagu wenxian jicheng 485–502 (rpt. from Taizhong jiagu xuehui , jiaguwen lunwenji , 1993). One problem with their analysis is that they do not take into account the grammar of oracle bone inscriptions in which the object is often inverted and placed before the verb. For example, the expression fang di () is interpreted in this article as a fang-rite performed to Di, rather than a di-rite performed to the four fang (directional quadrates). Most of their examples can, in my opinion, be better resolved in this manner, although there remain some problematic divinations. Shima Kunio, Inkyo bokuji kenkyū, 100ff., takes a small square, conventionally transcribed as ding , as representing di. For a critique, see Itō Michiharu, “Religion and Society,” in Itō Michiharu and Ken'ichi Takashima, Studies in Early Chinese Civilization, 2.11n5.

40. Zhenglang, Zhang , “Ge Qi you de zhenwei wenti” , Gugong bowuyuan xuekan 82 (1998.4), 1–5 Google Scholar. The same issue of Gugong bowuyuan xuekan includes articles by Zhang Guangyu , Sun Zhichu , Lian Shaoming , and Zhu Fenghan discussing the vessels and their inscriptions (pp. 6–16). In Zhang Zhenglang's article, originally delivered at a paleography conference held at Qingdao, Shandong Province, in 1987, he discusses the history of the vessels. Zhang thought that the inscription on the bottom might have been added to the Ersi Ge Qi you. He also mentions that he discussed this with Hu Houxuan and other scholars early on. These scholars did not use the inscriptions in their research. However, a number of prominent scholars thought the vessels and their inscriptions were genuine, e.g., Ding Shan , Guo Moruo, Dong Zuobin , Chen Mengjia, Yu Xingwu , and Li Xueqin. These vessels are also known as Zuoce Zhi zi you . This is the name used by Bagley, Robert in Shang Ritual Bronzes in the Arthur M. Sackler Collections (Cambridge, MA: Harvard University Press, 1987)Google Scholar. For his discussion, see pp. 526–28 and 535n28–31.

41. See Meng, Ding and Min, Jian , “Ge Qi you de X shexian jiance fenxi” , Gugong bowuyuan xuekan 83 (1999.1), 83–85 Google Scholar. The X-rays confirm that the inscription on the Ersi Ge Qi you was cast as part of the original vessel. There no longer seems any valid reason to suspect these vessels and they are included in virtually all modern indices, collections of rubbings, and other such compilations.

42. The last two characters, which I read as Shang Di, are read by some scholars as shang xia di . This reading is based upon the assumption that the line above di is part of a joined character, shangxia, in which one line has been omitted. See, for example, Yin Zhou jinwen jicheng shiwen , ed. yanjiusuo, Zhongguo shehui kexueyuan kaogu (Hong Kong: Xianggang Zhongwen daxue Zhongguo wenhua yanjiusuo , 2001), 4.156 Google Scholar (no. 5412). However, the closely related Sisi Ge Qi you (157, no. 5413) refers to Wenwu di and the character for di there also has two lines at the top, a form commonly found in oracle bone inscriptions of the period. Thus this line should be part of di, not the lower part of a joined character and the phrase should be read as Shang Di, as it is, for example, in Jinwen yinde: Yin-Shang Xi-Zhou juan , ed. daxue, Huadong shifan zhongxin, Zhongguo wenzi yanjiu yu yingyong (Nanning: Guangxi jiaoyu, 2001), 156 Google Scholar (no. 2823). I suspect that the common reading of shang-xia is influenced by the theory that there were no sacrifices to Shang Di.

43. Xia Shang Zhou duandai gongcheng 1996–2000 nian jieduan chengguo baogao: jianben 1996–2000 , ed. Shang, Xia zu, Zhou duandai gongcheng zhuanjia (Beijing: Shijie tushu, 2000), 58 Google Scholar.

44. Yingguo suocang jiagu ji , ed. Xueqin, Li, Wenxin, Qi and Lan, Ai [Sarah Allan], 4 vols, published in 2 parts (Beijing: Zhonghua, 1985, 1991)Google Scholar (hereafter Yingcang).

45. As recorded in Yinxu Huayuanzhuang dongdi jiagu , ed. yanjiusuo, Zhongguo shehui kexueyuan kaogu (Shanghai: Yunnan renmin, 2003)Google Scholar.

46. The Jiaguwen bian , ed. yanjiusuo, Zhongguo shehui kexueyuan kaogu (Beijing: Zhonghua, 1965)Google Scholar, and other such compendia list all known variants, so standard graphs and rare or unique variants are not clearly distinguished. This can be misleading and has sometimes resulted in hypotheses of the origin and development of particular characters based upon unusual graphic forms. As a matter of principle, any hypotheses of the formal evolution of a character should be based upon a pattern of occurrences and common graphic forms.

47. Abstracted from Tao, Wang, “Shang Ritual Animals: Colour and Meaning,” Part 1, Bulletin of the School of Oriental and African Studies 10.2 (2007), 317 Google Scholar, fig. 2. My periodization of the inscriptions is based upon Xueqin, Li and Yushang, Peng , Yinxu jiagu fenqi yanjiu (Shanghai: Shanghai guji, 1996)Google Scholar.

48. Yin Zhou jinwen jicheng 10.5412,10.5413; Jinwen zongji , ed. Yiping, Yan (Taibei: Yiwen, 1983), 7.5491.3, 7.5492.3Google Scholar (Ersi and Sisi Biqi you ,see also n.40,41, 42 above). For the Ban fang ding, see Xueqin, Li, “Shilun xin faxian de Ban fang ding he Rong Zhong fang ding” , Wenwu 2005.9, 59–5, 69Google Scholar.

49. Ambiguous examples, possibly due to unclear rubbings or inattentive calligraphy, include: Heji 862,14148,14172, 30391; the middle inscription has an additional line at the top. These are marked in italics in Table 2. Heji, 21074, 21076, 21077, 34158, 34160, and Yingcang, 1137, are all used in the sense of di .

50. Heji, 34149.

51. Heji, 21074, 21076,21077,34158, 34160; Yingcang, 1137; this reading is discussed below.

52. Zhouyuan jiaguwen , ed. Wei, Cao (Beijing: Shijie tushu, 2002), 1 Google Scholar (H11:1), 62 (H11:82), 84 (H11:123). See also Quanfang, Chen , Zhiyi, Hou , and Min, Chen , Xi-Zhou jiawen zhu (Shanghai: Xuelin, 2003), 1, 21, 47Google Scholar. In this work, the graph is taken as a joined character (hewen) meaning shang di, but there is only a single graph and there is no supporting evidence.

53. jiaguwenzi jishi , ed. Xiaoding, Li (Nangang: Zhongyang yanjiuyuan Lishi yuyan yanjiusuo, 1965), 0025–31 Google Scholar; Jiagu wenzi gulin , ed. Xingwu, Yu (Beijing: Zhonghua, 1996), 1082–86Google Scholar (no. 1132); Jiagu wenzi gulin, 2502–11 (no. 2516); Jinwen gulin, 36 (1.0004).

54. See Itō Michiharu and Ken'ichi Takashima, Studies in Early Chinese Civilization, 1.145.

55. David W. Pankenier, “A Brief History of Beiji (Northern Culmen), 211–36.

56. Examples include: Heji, 8, 368, 371, 405, 418, 456, 478, 505, 905, 940, 974, 1140, 2334, 2580, 3504, 3506, 3671, 5662 (loop at top), 7061, 8649, 10001, 10939, 10976, 11018, 11842, 12855, 14130, 14159, 14295, 14298–300, 14303–5, 14307–8, 14310, 14313, 14320, 14323, 14326–8, 14332, 14360, 14363, 14470, 14531, 14773/ 157°3’ 15950–59, 15962, 15964–68, 15970, 15972–85, 17252, 19243, 21073, 21084, 21387, 22088, 30590, 35720; Yingcang, 12, 86, 149, 1223–29, 1751.

57. See Sarah Allan, Shape of the Turtle, 74–98.

58. See Jiagu wenzij ishi 1539–44.

59. See Lan, Ai [Sarah Allan], “Ya-xing yu Yin ren de yuzhouguan” , Zhongguo wenhua 4 (1991), 31–47 Google Scholar. See also Sarah Allan, Shape of the Turtle, 78. Unfortunately, my argument in the English book is somewhat garbled by a printing error in line 12. This paper was first presented at an international conference in Anyang in 1985, which is published as “Tan Yin dai de yuzhouguan yu zhanbu” , Yinxu bowuyuan yuankan 1 (1989), 189–98Google Scholar, but this publication omits some sections of the original text.

60. For discussion of the axis mundi and the importance of centrality in Shang thought, see Sarah Allan, Shape of the Turtle, 44–45. I also argue therein that the Shang understood the earth as having a central quadrate with four mythical regions to the north, south, east, and west.

61. See Jiagu wenzi gulin, 2502–3.

62. Heji, 14302, 14309, 14312, 14345, 14370, 15960–61, 15963, 19710. Heji, 264, is ambiguous.

63. Heji, 32063,34145,34153–54; Xiaotun, 2161,3664,4524. Heji 21080 is similar, but has a loop at the top of the central vertical. Heji 34158 is similar, but the central vertical does not bisect the upper triangle.

64. Heji, 32012,34145.

65. Heji, 32012, 34155, 34157, Xiaotun, 804.

66. Heji, 34156, 34615. See also Chin-hsiung, Hsu, Oracle Bone Inscriptions from the White and Other Collections (Toronto: Royal Ontario Museum, 1979)Google Scholar, no. 1565 (abbreviated hereafter as White).

67. Cf. Heji, 33230 and 33231, which have the expression mi di written as two characters.

68. Heji, 21387.

69. Heji, 14360.

70. It is also possible that the quadrates were ruled by di. Heji, 34156, and White, 1565, have divinations with the expression yu bei di , which could be taken as “to the northern Di.” However, if they are so interpreted, neither of these divinations names a rite, so they are more likely to be inverted phrases, meaning, “To the North, offer the di-rite.” In any case, the cosmology suggested by the performance to the four quadrates found in the four directions corresponds with my analysis of the di-graph.

71. Liji jijie , ed. Xidan, Sun (Beijing: Zhonghua, 1989), 45.1192 Google Scholar (“Ji fa” , 13).

72. Lunyu jishi , ed. Shude, Cheng et al. (Beijing: Zhonghua, 1990), 5.164–7Google Scholar (3.10).

73. See Sarah Allan, Shape of the Turtle, 53–55.

74. See Mao shi buzheng , ed. Qitao, Long (Taibei: Lixing, 1970), 25.1701 Google Scholar.

75. Lunyu jishi, 5.168–72 (3.11).

76. For compilations of various scholarly interpretations of this character, see jiagu wenzi gulin, 210–14 (0198), Jimuen gulin, 24–35 (1.0003); Jinwen gulin bu (Taibei: Zhongyang yanjiuyuan Lishi yuyan yanjiusuo , 1982), 1.93–105Google Scholar.

77. The editors of Xiaotun nandi jiagu, xiace , 1–992, transcribe the character as and note that tian is the object of the sacrifice here and that this is a peculiarity of the Wu diviner group. The character is also transcribed in this manner in Yinxu jiagu keci leizuan, 3.84. See also Jiagu wenzi gulin, 214 (0201), where it is separated, but also transcribed as tian.

78. Xiaotun, 2241.

79. Ji zhen , Heji, 20975; see also Heji, 17985.

80. Heji, 36535, 36544, Yingcang, 2529.

81. Heji, 36422.

82. Heji, 14200–207.

83. E.g., Heji, 14209–10. For a review of these inscriptions and translations of a number of examples, see David Keightley, The Ancestral Landscape, 57–61. Keightley also discusses the identification of “this settlement” with the finds at Xiaotun.

84. Heji, 5843, 11499, 20957, 21022, 27226, 28663.

85. Zhushu jinian yizhu , ed. Yuchun, Zhang (Harbin: Heilong-jiang renmin, 2002), 167 Google Scholar. Pankenier, David, “Astronomical Dates in Shang and Western Zhou,” Early China 7 (1981–82), 2–37 Google Scholar, was the first to use this planetary conjunction to date the beginning of the Western Zhou.

86. Yin Zhou jinwen jicheng, 8.4261; Jinwen zongji, 4.2777; Shang Zhou qingtongqi ming-wen xuan , ed. Chengyuan, Ma (Beijing: Wenwu, 1986–1990), 3.23 Google Scholar. The transcriptions here and below are not direct representations of the ancient graphs, but my readings of the graphs in modern characters. The graph after Wen Wang (7 in line 3) is damaged and difficult to decipher; some scholars take it as de , rather than jian .

87. Yun, Lin , “Tianwang gui ‘wang si yu tianshi’ xinjie” , in Yun, Lin, Lin Yun xueshu wenji (Beijing: Zhongguo dabaike quanshu, 1998), 167–73Google Scholar. This paper was first presented as a conference paper in Xi'an in 1993 and published in Shixue jikan 1993.3. Cf. David Pankenier, “The Cosmo-political Background,” 139–40, who also follows Lin Yun.

88. Sarah Allan Shape of the Turtle, 98–101. For earlier versions, see Ai Lan [Sarah Allan], “Tan Yin dai de yuzhouguan yu zhanbu,” 189–98, and “Ya-xing yu Yin ren de yuzhouguan,” 39.

89. Mao shi buzheng 25.1696 (“Shang song,” “Xuan niao” ). Since the Shang prince, Weizi Qi , was enfeoffed in the state of Song and allowed to continue making ancestral offerings, the reference to cutting off the Shang rites here does not refer to those to the Shang ancestors. I presume it refers to the offerings to Shang Di and suggest that it is further evidence that the Shang did make offerings to Shang Di. I take the graphs and as representing the same word (the name of a rite).

90. Yin Zhou jinwen jicheng, 1.49; Jinwen zongji, 9.7006.

91. Yin Zhou jinwen jicheng, 16.10175; Jinwen zongji, 8.6792; This vessel, which was excavated in 1976 from a pit (H1) at Fufengxian, Famen Gongshe Zhuangbai in Shaanxi Province, probably dates to the reign of Gong Wang . See Xueqin, Li, “Lun Shi Qiang pan jiqi yiyi” , in his Xinchu qingtongqi yanjiu (Beijing: Wenwu, 1990), 73–82 Google Scholar (rpt. from Kaogu xuebao 1978.2 and Ma Chengyuan, Shang Zhou qingtongqi mingwen xuan, 3.153–58 (no. 225). Another middle Western Zhou bell, the Xing zhong (Yin Zhou jinwen jicheng 1.247–50), includes this same line.

92. My interpretation of this line follows Ma Chengyuan. See the preceding note.

93. Mao shi buzheng, 19.1311–2 (“Da ya,” “Sheng min zhi shi” , “Sheng min” ). The Mao commentary takes Di as Gao Xin shi , that is, Di Ku , which corresponds to the genealogy for Hou Ji in the Shi ji. However, this construct is not found in pre-Han texts. Di used on its own (without a name) in Zhou texts normally refers to Shang Di and Zheng Xuan (C.E. 127–200) glosses di here as Shang Di. The earliest references to the Shang origin myth are in the “Shang song” section of the Shi jing (“Xuan niao” and “Chang fa” , 1694–1708). The “Xuan niao” begins “Tian commanded the black bird to descend and generate the Shang” , and goes on to state Di commanded the Shang founder, Tang . The “Chang fa” also states, “Di established his son to generate the Shang” . I understand Di in all of these lines as Shang Di. See Sarah Allan, The Shape of the Turtle, 39–41, 51–55, 59.

94. Yin Zhou jinwen jicheng, 1.260, Shang Zhou qingtongqi mingwen xuan, 1.405; Jinwen zongji, 9.7176. This vessel probably dates to the reign of King Li .

95. Mao shi buzheng, 19.1222 (“Da ya,” “Wen Wang zhi shi,” , “Wen Wang” ).

96. Zihui edition, as reprinted in Defoort, Carine, The Pheasant Cap Master (Albany: State University of New York Press, 1997), 345 Google Scholar (10.71a “Tai hong” ).

97. Lunyu jishi (Beijing: Zhonghua, 1990), 3.61 Google Scholar (2.1).

98. Found in the Shishi Xing jing ; see Sun and Kistemaker, Chinese Sky during the Han, 157.

99. The Xing Hou gui is also called the Rong zuo Zhou Gong gui or Zhou Gong gui . Yin Zhou jinwen jicheng, 8.4241; Jinwen zongji, 4.2764. Ma Chengyuan, Shang Zhou qingtongqi mingwen xuan, 3.66, dates the inscription to King Kang. Xueqin, Li and Yunming, Tang , “Yuanshi tongqi yu Xi-Zhou de Xingguo” , Kaogu 1979.1, 56–59 Google Scholar, 88 [58], date it to the reigns of Kings Cheng . or Kang .

100. Ma Chengyuan Shang Zhou qingtongqi mingwen xuan, 3.46.

101. Some scholars read the graphs that I take as “Shang Di” in the Ersi Ge Qi you as “shang xia di,” but the graphs do not support this reading; see figure 1. To my knowledge, there are no other examples of the expression shang xia di in Western Zhou bronze inscriptions or in the received literature. Similarly, the expression xia di does not occur.

102. Cuben zhushi jinian yizhu jijiao; Jinben zhushu jinian shuzheng , ed. Guowei, Wang (Taipei: Shijie, 1957)Google Scholar, shu shang, 37.

103. Yin Zhou jinwen jicheng, 5.2837, Jinwen zongji, 2.1328; Shang Zhou qingtongqi mingwen xuan, 1.62.

104. Some scholars take you as you , “aid, assist,” but I find this difficult to understand grammatically.

105. Yin Zhou jinwen jicheng, 11.6014; Jinwen zongji, 6.4891, Shang Zhou qingtongqi mingwen xuan, 1.32. My transcription is based on that of Li Xueqin, “He zun xinshi as reprinted in Li Xueqin, Xinchu qingtongqi yanjiu, 38–45. Li Xueqin dates the vessel to the 5th year of king Kang.

106. Of the Zhou chapters most generally accepted as having been compiled in the early Western Zhou (“Da gao” , “Kang gao” , “Jiu gao,”, “Shao gao” , “Luo gao” , “Duo shi” , “Jun shi” , “Duo fang”), the mandate is ascribed to Shang Di in “Da gao” and “Jun shi” and as that of tian in the “Da gao,” “Kang gao,” “Shao gao,” and “Luo gao.” See Bernhard Karlgren, The Book of Documents (Stockholm: Museum of Far Eastern Antiquities, 1950), 34, lines 9 and 13, 60, line 3 (shang di ming); 34, line 1, 38, line 3; 47, line 17, 50, line 14 (tian ming). However, the context of Shang Di's command is usually one with references to tian and tian ming occurs in contexts that also refer to Di or Shang Di.

107. Mao shi buzheng, 23.1518; 19.1222 (“Zhou song” , “Si Wen” ; “Da ya,” “Wen Wang”). It is also found in the “Shang song” verses, “Xuan niao” and “Chang fa” (25.1694, 1704).

108. Karlgren, The Book of Documents, 47, line 9 (my translation).

109. See Li Xueqin, “He zun xinshi,” in his Xinchu qingtongqi yanjiu, 42–43.

* The tables below are based upon rubbings in Jiaguwen heji (Beijing: Zhonghua, 1982), ed. Guo Moruo, et al; Xiaotun nandi jiagu , ed. yanjiusuo, Zhongguo shehui kexueyuan kaogu (Beijing: Zhonghua, 1980–83)Google Scholar, and Yingguo suocang jiagu ji , ed. Xueqin, Li, Wenxin, Qi , and Lan, Ai (Sarah Allan) (Beijing: Zhonghua, 1985 Google Scholar [pt. 1], 1991 [pt. 2]). I would like to thank Angela Fang for her help in producing these tables.