1. The most complete report of this excavation is the Hunan Provincial Museum and Institute of Archeology, CAS, Changsha Mawangdui yihao Han-mu 長沙馬王堆一號漢墓, 2 vols. (Beijing: Wenwu, 1973)Google Scholar. For an English summary, see Buck, David, “The Han Dynasty Tomb at Ma-wang-tui,” World Archaeology 7.1 (1975), 30–45 CrossRefGoogle Scholar.

2. Silbergeld, Jerome, “Mawangdui, Excavated Materials, and Transmitted Texts: A Cautionary Note,” Early China 8 (1982–1883), 83 Google Scholar.

3. Silbergeld, , “Mawangdui,” 86 Google Scholar.

4. See Yi li (Shisanjing zhushu ed.), 35.1128–29Google Scholar. Li ji (Shisanjing zhushu ed.), 44.1572 Google Scholar. See, too, Legge, James, Li Chi, 2 vols. (New York: University Books, Inc., 1967), vol. 2, 175 Google Scholar. Scholars commonly consider the Yi li to have been written during the later Eastern Zhou, while the Li ji was compiled during the Han. But the descriptions of a funerary ritual in these two texts provide complementary materials for reconstructing the ritual.

5. Yi li, 35.1128–29Google Scholar. Li ji, 44.1572 Google Scholar.

6. Dunyuan, Liu, “Mawangdui Xi-Han bohua zhong de ruogan shenhua wenti” 馬王堆西漢帛畫中的若干神話問題, Wenshizhe 文史哲, 1978.4, 63–72 Google Scholar. Scholars who hold similar opinions include Yu Weichao 俞偉超 and Ying-shih Yü; see Weichao, Yu, “Mawangdui yihao Han-mu bohua neirong kao” 馬王堆一號漢墓帛畫內容考, in Weichao, Yu, Xian-Qin Liang-Han kaoguxue lunji 先秦兩漢考古學論集, (Beijing: Wenwu, 1985), 154–56Google Scholar; Yü, Ying-shih, “‘O Soul, Come Back!’ A Study in the Changing Conceptions of the Soul and Afterlife in Pre-Buddhist China,” Harvard Journal of Asiatic Studies 42.2 (1987), 363–95CrossRefGoogle Scholar.

7. Li ji, 44.1572 Google Scholar.

8. Ying-shih Yü explains the symbolism of this ritual: ”… at the moment when death first occurs, the living cannot bear to believe that their beloved one has really left them for good. The living must first assume that the departure of the hun-soul is only temporary. It is possible, then, that if the departed soul can be summoned back the dead may be brought back to life. A person can be pronounced dead only when the fu ritual has failed to achieve its purpose … “But he still considers this ceremony part of the death ritual: “It was the first of a series of rituals to be performed to the newly dead”; “‘O Soul, Come Back!’” 365.

9. See Eliade, Mircea, Shamanism: Archaic Techniques of Ecstasy, tr. Trask, Willard R. (New York: Pantheon Books, 1964), 215–258 Google Scholar; Frazer, James G., The Golden Bough (rpt. New York: MacMillan Publishing Co., 1963), 206–220 Google Scholar; Clements, Forrest E., Primitive Concepts of Disease (Berkeley: University of California Press, 1932)Google Scholar; Hawkes, David, The Songs of the South (New York: Penguin Books, 1985), 219–223 Google Scholar.

10. Based on translation by Hawkes, , Songs of the South, 224 Google Scholar.

11. Eliade, , Shamanism, 218–220 Google Scholar; Mikhailowske, V. M., “Shamanism in Siberia and European Russia,” Journal of the Royal Anthropological Institute 24 (1894), 69—70 Google Scholar.

12. Hawkes, , Songs of the South, 220 Google Scholar. The example was originally quoted by Frazer, J. in The Golden Bough, 212 Google Scholar.

13. This confusion is partly caused by the identical term, sizhe 死者, used in the ritual canons for both a “dying person” and a “dead person.”

14. Hawkes summarizes these different opinions in Songs of the South, 221–22.

15. Chu ci sizhong 楚辭四種 (Shanghai: Shijie shuju, 1936), 119 Google Scholar.

16. Chuci jizhu 楚辭集註, ed. Xi, Zhu (Shanghai: Shanghai guji chubanshe, 1979), 133 Google Scholar.

17. Li ji, 44.572 Google Scholar.

18. Yu Weichao claims that the Mawangdui banner was used to summon the soul of the dead and was called a wu or fu 憮, which was a piece of cloth used to cover the corpse; “Mawangdui yihao Han-mu bohua neirong kao,” 154. This argument does not agree with the record in Li ji that an object used in summoning the soul should not be employed to cover or dress the corpse. In fact, there is no direct textual evidence for the use of a wu in summoning the soul. We can also reject the popular opinion which links the Mawangdui banner with a soul-recalling “garment” and identifies it as a feiyi 飛衣, “flying garment.” As Silbergeld has pointed out, the inventory from the Ma-wangdui tomb records two feiyi, not one, and the measurement of the painting does not agree with that of either; “Mawangdui,” 84.

19. For an interpretation of the ancient Chinese theory of the two souls, see Yü, , “’Soul, Come Back!’” 369–78Google Scholar.

20. For a more detailed discussion of the funeral rituals recorded in ancient texts, see Gongrou, Chen 陳公柔,” ’“Shisang li,’ ‘Jixi li” zhong suo jizai de sangzang zhidu”“ 士喪禮, 即夕禮” 中所記載的喪葬制度, Kaogu xuebao 考古學報 1956.4, 67–84 Google Scholar.

21. Liji, 44.576 Google Scholar.

22. Yi li, 35.1129 Google Scholar.

23. Zheng Xuan explains the funerary banner: “Ming 銘 means to inscribe the dead person's name (ming 名) on the banner.” Therefore, he commented on Zhou li 周禮: “Nowadays people also write the character ming (“inscription” or “inscribed”) as ming (name)”; Zhou li (Shisanjing zhushu ed.), 25.812 Google Scholar. Here I translate the term as “Name Banner” because what is “inscribed” on it should be the dead person's “name.” This translation reflects the nature of this funerary object more precisely than “inscribed banner,” Zheng Xuan's other explanation of the banner, “Ming 銘 means a ‘banner of distinction’ (mingjing 明旌), “also agrees with this translation; Yi li, 35.1130 Google Scholar.

24. Yi li, 35.1130 Google Scholar.

25. Yi li, 37.1139–40Google Scholar. See, too, Yong, Ma 馬雍, “Lun Changsha Mawangdui yihao Han-mu chu tu bohua de mingcheng he zuoyong” 論長沙馬主堆一號漢墓出土帛畫的名稱和作用, Kaogu 考古 1973.2, 118–125 Google Scholar.

26. Li ji, 5.1269 Google Scholar; My translation is based on that by Legge; Li Chi, vol.1, 117 Google Scholar.

27. Gu, Ban 班固, Baihu tong 白虎通 (congkan, Sibu ed.), 10.16b Google Scholar.

28. Xuan, Zheng says: “The Name Banner is a banner inscribed with the name of the deceased. Nowadays people call it a jiu ”; Zhou li, 25.812 Google Scholar. Similarly, the Xiao erya 小爾雅 says: “An empty coffin is called a qin 櫬; when it contains the corpse it is called a jiu”; see Zhongwen dazidian 中文大字典, 10 vols., ed. Yin, Lin 林尹 and Ming, Gao 高明, (Taibei: Zhonghua xueshuyuan, 1973), vol. 5, 7071, 7479 Google Scholar. The term jiu thus refers only to the innermost coffin, not to outer coffins.

29. See Changsha Mawangdui yihao Han-mu, 43–45; Zhimin, An 安志敏, “Changsha xin faxian de Xi-Han bohua shitan” 長沙新發現的西漢帛畫試探, Kaogu 1973.1, 43–53 Google Scholar; and especially, Ma Yong, “Lun Changsha Mawangdui yihao Han-mu chutu bohua.” Ying-shih Yü rejects this identification, arguing, “It is somewhat puzzling that in spite of the fact that a ming-ching (mingjing) is by definition ‘inscribed’ and that all the ming-ching excavated from Han tombs so far invariably bear the names of the dead, both An and Ma still insist on identifying the two uninscribed T-shaped paintings as ‘inscribed funerary banners’”; “‘O Soul, Come Back!’” 369 n. 13. Indeed, as Yü points out, two banners from Gansu are both inscribed with the names of the deceased and thus differ from the Mawangdui banner that bears the portrait of the deceased. But as I have explained earlier, the character ming 銘 (inscribed) was also written as ming 名 (name) during the Han. The main function of the banner, to ‘distinguish’ (ming 明) the deceased, could be fulfilled by either a written name or a pictorial image. Besides the two banners from Mawangdui tombs nos. 1 and 3, another painted banner from Jinqueshan 金雀山 tomb no. 9 in Shandong can also be identified as a Name Banner; see Wenwu 文物 1977.11, 26 Google Scholar. The classification of these five examples into a single category is further supported by a common feature: the name or portrait of the deceased always appears under the images of the sun and moon painted in the upper part of the banner.

30. See Jia's, comment in Yi li, 35.1130 Google Scholar.

31. For a summary of these various opinions, see Loewe, Michael, Ways to Paradise (London: George Allen & Unwin, 1979), 34 Google Scholar. Loewe himself divides the painting into two sections and thus follows the opinion of Shang Zhitan 商志譚 and Ma Yong; see Zhitan, Shang, “Mawangdui yihao Hanmu ‘feiyi’ shishi” 馬王堆一號漢墓“非衣”試釋, Wenwu 1972.9, 43–47 Google Scholar; Yong, Ma, “Lun Changsha Mawangdui yihao Han-mu chutu bohua,” 122–123 Google Scholar.

32. In reviewing the scholarship on the painting, Silbergeld also follows this four-level division; “Mawangdui,” 79.

33. See Yong, Ma, “Lun Changsha Mawangdui yihao Han-mu chutu bohua,” 121–22Google Scholar.

34. In addition to textual evidence, a detail of the Mawangdui painting may indicate that the image of Lady Dai portrays her jiu-corpse. An oblique board with rows of spiral patterns is painted almost directly underneath the portrait. Because of the spiral patterns, this image cannot represent “steps” or a “causeway.” It more likely depicts a lingchuang 等床 or pi 裨, which is a rectangular board with ornate openwork designs frequently found in Chu tombs in the Changsha area. Such a board always appears inside an innermost coffin to support the corpse. For an excavated example, see bowuguan, Hubeisheng Jingzhou diqu, Jiangling Yutaishan Chu mu 江陵雨台山楚墓 (Beijing: Wenwu, 1984), fig. 43Google Scholar.

35. This opinion, first proposed by Watson, William, is elaborated by Loewe in his Ways to Paradise, 45–46 Google Scholar.

36. See Hawkes, , Songs of the South, 74, 225 Google Scholar.

37. For more detailed analyses of this section, see Zuoyun, Sun 孫作雲, “Changsha Mawangdui yihao Han-mu chutu huafan kaoshi” 長沙馬王堆一號漢墓出土畫幡考釋, Kaogu 1973.1, 54–61 Google Scholar; Loewe, , Ways to Paradise, 47–59 Google Scholar.

38. Qian, Sima, Shi ji 史 (shuju, Zhonghua ed.), 6.265 Google Scholar.

39. This opinion is represented by Zuoyun, Sun, “Changsha Mawangdui yihao Han-mu chutu huafan,” 57–58 Google Scholar.

40. First proposed by Zhitan, Shang (“Mawangdui yihao Han-mu ‘feiyi’,” 44 Google Scholar), this opinion was then adopted by Loewe, (Ways to Paradise, 37 Google Scholar).

41. Weichao, Yu, “Mawangdui yihao Han-mu bohua neirong kao,” 156 Google Scholar.

42. Hawkes, , Songs of the South, 224–225 Google Scholar.

43. For Han pictorial language, see my discussion of “episodic” and “iconic” representations in The Wu Liang Shrine: The Ideology of Early Chinese Pictorial Art (Stanford: Stanford University Press, 1989), 133–135 Google Scholar. Wang Yi suggested that Qu Yuan's “Heavenly Questions” were inspired by temple murals that depicted stories and myths. But even if we accept his opinion, such murals could still be “episodic” representations that either illustrated “culminating’ events or integrated a number of events into “concentrated” scenes. These two types of pictures dominate Han narrative art. Jonathan Chaves has contended that a mural in a first-century tomb near Luoyang may depict various episodes of the famous story called “Two peaches killed three gentlemen”; “A Han Painted Tomb at Loyang,” Artibus Asiae 30 (1968), 5–27 Google Scholar. But this contention still needs more evidence. A typical sequential narrative consisting of multiple scenes only appeared in the Six Dynasties period, as exemplified by the “Nymph of the Luo River” 洛神賦圖, attributed to Gu Kaizhi 顧愷之 (c. 346–407).

44. For the existence of such a dualistic structure in pre-Han art and cosmology, see Chang, K.C., “Some Dualistic Phenomena in Shang Society,” in Early Chinese Civilization: Anthropological Perspectives (Cambridge, Mass.: Harvard University Press, 1976), 93–114 Google Scholar; Art, Myth, and Ritual (Cambridge, Mass.: Harvard University Press, 1983), 56–80 Google Scholar. Such a static structure also characterizes the composition of the carvings on the famous Wu Liang Shrine, in which heavenly omens are depicted on the ceiling, the wall-engravings represent the human world, while the Queen Mother of the West and the King Father of the East are portrayed on the two opposing gables; Wu, , The Wu Liang Shrine, 218–221 Google Scholar.

45. Except for the images depicted on the central axis, only two or three images are not arranged in pairs. Among them is a very interesting image — a female figure holding the crescent moon with both hands. Many scholars identify this figure as the moon goddess Chang'e 婦娥; see, Zuoyun, Sun, “Changsha Mawangdui yihao Han-mu chutu huafan,” 57–58 Google Scholar. But Wang Bomin 王伯敏 and Jean James argue that this figure actually represents Lady Dai's hun-soul, because of the metaphorical correlation established between the soul and the moon in ancient Chinese thought, and because on the Name Banner of Lady Dai's son buried in Mawangdui tomb no. 3 a male replaces the female figure in the heavenly realm; Bomin, Wang, “Mawangdui yihao Han-mu bohua bingwu Chang'e benyue” 馬王堆一號漢墓帛畫並無嫦娥奔月, Kaogu 1979.3, 273–74Google Scholar; James, J., “An Iconographie Study of Two Late Han Funerary Monuments: The Offering Shrines of the Wu Family and the Multichamber Tomb at Holingor” (Ph.D dissertation: University of Iowa, 1983), 17 Google Scholar; “Interpreting Han Funerary Art: The Importance of Context,” Oriental Art 31.2 (1985), 285 Google Scholar. These two scholars, however, have difficulty explaining why the supposed hun-image in the Tomb no. 3 painting is not associated with the moon.

46. See, for example, Wei, Shi 史爲 (Xia Nai 夏), “Changsha Mawangdui yihao Han-mu de guanguo zhidu” 長沙馬王堆一號漢墓的棺槨制度, Kaogu 1972.6, 48–52 Google Scholar; Weichao, Yu, “Mawangdui yihao Han-mu guanzhi de tuiding” 馬王堆一號漢墓棺制的推定, Hunan kaogu jikan 湖南考古集刊, 1982.1, 111–115 Google Scholar.

47. The Shuo wen 說文 dictionary provides this definition: “Black is the color of the north,” while the Li ji states: “Bury (the deceased) in the north with his head also towards the north”; for the definitions and usages of the character xuan and related words, see Zhongwen dazidian, vol. 6, 262–276 Google Scholar.

48. I have discussed the Han concept and depiction of qi in “A Sanpan Shan Chariot Ornament and the Xiangrui Design in Western Han Art,” Archives of Asian Art 37 (1984), 46–48 Google Scholar.

49. These images are discussed by Zuoyun, Sun in a learned article, “Mawangdui yihao Han-mu qiguanhua kaoshi” 馬王堆一號漢墓漆棺畫考釋, Kaogu 1973.4, 247–254 Google Scholar. My identification and classification of these images is based on Sun's views.

50. Shanhai jing jiaozhu 山海經校註, ed. Ke, Yuan 袁河 (Shanghai: Shanghai guji chubanshe, 1980), 45 Google Scholar.

51. These images are identified as Kunlun by Hiroshi, Sōfukawa 曾布川寬, “Konronzan to Shōsenzu” 崑崙山昇仙圖, Tōhōgakuhō 東方學報 51 (1979), 87–102 Google Scholar.

52. I have discussed the development of the Kunlun, myth in The Wu Liang Shrine, 117–126 Google Scholar.

53. My translation is based on that of Hawkes; Songs of the South, 73–74.

54. An, Liu 劉安, Huainan zi (Sibu congkan ed.), 4.26–27 Google Scholar.

55. I have discussed the changing conception of immortality in ancient China in “Beyond the Great Limit: The Cangshan Funerary Narrative,” in Boundaries in China, ed. Hay, John (London: Reaktion Books, forthcoming)Google Scholar.

56. For decoration of coffins, see Li ji, 45.1583–84Google Scholar.

57. See Yi li, 37.1142–43Google Scholar; Li ji, 8.1293 Google Scholar.

58. The same intention is found in the plan and furnishing of the tomb of Marquis Yi of Zeng. The east chamber of this tomb, which contained the coffin of the Marquis and those of eight young women, seems to correspond to the deceased's living quarters. The north chamber was arranged as an armory, containing a chariot and a range of weapons. The thirteen young females buried in the west chamber identify this place as the concubine's quarters. The central chamber with a vast array of ritual vessels and musical instruments represented the main hall in a household. See Hubei Provincial Museum, Zenghou Yi mu 曾侯乙墓, 2 vols (Beijing: Wenwu chubanshe, 1989), vol. 1, 12–15, 60–75 Google Scholar.

59. Wei, Shi (Nai, Xia), “Changsha Mawangdui yihao Han-mu de guanguo zhidu,” 52 Google Scholar.

60. These items, all found in the space between the third and fourth coffins, include a bronze mirror and sixty-nine figurines made of peach wood. As the excavators of the tomb explain, these objects served the function of warding off evil and protecting the corpse inside the fourth or innermost coffin; see Changsha Mawangdui yihao Han-mu, 100–101, 128.

61. The excavators of the Mawangdui tomb have explained such inconsistencies: the sizes and number of buried goods are sometimes exaggerated in the written record; some clay replicas of utensils listed in the inventory were not actually installed into the tomb; and gifts, wooden figurines, and the clothes of the deceased are omitted from the document; Changsha Mawangdui yihao Han-mu, 154.

62. Gao Ming 高明 has identified these two male images as the chief eunuchs in Lady Dai's family; “Changsha Mawangdui yihao mu ‘guanren’ yong” 長沙馬王堆一號墓“冠人”俑, Kaogu 1973.4, 255–257 Google Scholar.

63. This phenomenon supports Ying-shih Yü's assertion that “ancient Chinese were extremely hunger-conscious about their ancestors in the afterworld,” and that “without sacrificial food, the hungry ancestral spirits would disintegrate more quickly”; “‘Soul, Come Back!’” 378–379.

64. See Changsha Mawangdui yihao Han-mu, 130–155, where the authors of the report compare the inventory list and the actual furnishings of the Mawangdui tomb.

65. Yü, , “’O Soul, Come Back’”, 378 Google Scholar.

66. Hawkes, , Songs of the South, 227–228 Google Scholar.

67. Li ji, 8.1292 Google Scholar; Legge, , Li Chi, vol. 1, 156 Google Scholar.

68. Chi, Li, Anyang (Seattle: University of Washington Press, 1977), 91–92 Google Scholar.

69. Yü, , “‘Soul, Come Back,’” 378 Google Scholar.

70. This object is published in Tokyo National Museum, Kōga bumtnei tenran 黄河文明展覽, (Tokyo: Chūnichi shimbunsha, 1986), fig. 29Google Scholar.

71. For a brief introduction to these two Chu banners, see Wenkun, Huang 黃文昆, “Zhangou bohua” 戰國帛畫, Zhongguo wenwu 中國文物 3 (1980), 31–32 Google Scholar.

72. See Jianjin, Lei 雷建金, ”Jianyangxian Guitoushan faxian bangti huaxiang shiguan” 簡陽縣鬼頭山發現榜題畫像石棺, Sichuan wenwu 四川文物 1988.6, 65 Google Scholar; wenhuaguan, Neijiang shi wenguansuo and Jianyang xian, “Sichuan Jianyangxian Guitoushan Dong-Han yamu” 四川簡陽縣鬼頭山東漢崖墓, Wenwu 1991.3, 20–25 Google Scholar; Dianzeng, Zhao 趙殿增 and Shuguang, Yuan 袁曙光, “‘Tianmen’ kao—jianlun Sichuan Han huaxiang zhuan(shi) de zuhe yu zhuti’” “天門考—兼論四川漢畫像磚 (石)的組合與主題, Sichuan wenwu 1990.6, 3–11 Google Scholar.

73. The inscription, “Cishang renma jie shi taicang” 此上人馬皆食太倉, is frequently found in Han tombs and means that all the human and animal figures portrayed on a funerary structure should eat in the Grand Granary, the greatest depository of food.

74. For a discussion of a standard decorative program of Sichuan sarcophagi, see Hung, Wu, “Myth and Legends in Han Funerary Art: Their Pictorial Structure and Symbolic Meanings as Reflected in Carvings on Sichuan Sarcophagi,” in Stories from China's Past, ed. Lim, Lucy (San Francisco: The Chinese Culture Foundation, 1989), 72–81 Google Scholar.