Skip to main content Accessibility help

Han Thanatology and the Division of “Souls”*

  • K.E. Brashier (a1) and 白瑞旭


Modern scholarship has repeatedly maintained that the separation of the hun and po souls at death was a popular Han belief, but a re-examination of the received literature suggests that hunpo dualism was at best only a scholastic model. Sources ranging from the Zuo zhuan to Han medical texts depict the hunpo as an inherent part of the properly functioning body complex, and any deficiency in the hunpo did not necessarily result in death but in distress and disease. Grave stele texts, which also never distinguish between a hun and a po suggest a different dualism—that between the hun or po with its corporeal associations on the one hand and the more rarefied shen on the other. This dualism may have found a practical expression in ancestral worship because the hunpo and body were generally confined to the cemetery but the mobile shen enjoyed its sacrifices at the lineage shrine.



Hide All

I would like to express my appreciation to Wu Hung for first arousing my interest in this topic, to Mark Edward Lewis, David McMullen, Joseph McDermott, and John EC. Mof fett for providing valuable direction and for reading earlier drafts, and to Donald Harper and the assessors of Early China for their extensive comments and criticisms. All errors in judgement and fact are solely my own.



Hide All

1. No English word exactly fits hunpo 魂魄, but instead of “soul,” the concept of harmonia seems to be a closer match to the scholastic interpretation of hunpo dualism. The Pythagoreans are said to have defined the soul as a harmonia within the body, a blending or attunement of opposing forces just as the body itself is compounded out of contraries. This attunement reflects the greater cosmology, but when the body is unduly stressed, the harmonia is threatened. And at death, the harmonia perishes with the body. See Guthrie, W.K.C., The Earlier Presocratics and the Pythagoreans (Cambridge: Cambridge University Press, 1988), 306–19. In the early Chinese scholastic tradition, the hun and po together represent a mixing of opposite forces (i.e. yang and yin respectively) and they divide at death. Discord damages the harmonia in Pythagorean thought just as anxiety damages the hunpo (see below). Theories on the harmonia in-elude balanced numerological implications such as a fascination with the symbolic power of ratios and a veneration of the number ten. Discussions on the hunpo in later sources such as the Taiping jing 太平經 and Baopuzi 抱朴子 also apply a ratio of the hun to the po—three to seven respectively—so that ten of these entities existed within the living person. See Ming, Wang 王明, Taiping jing hejiao 太平經合校 (Beijing: Zhonghua, 1960), 8 (“Taiping jinque” 太平金闕) and 627 (“Wei fumu buyi jue” 爲父母不易訣); Ming, Wang, Baopuzi neipian jiaoshi 抱朴子內篇校釋 (Beijing: Zhonghua, 1980), 299 (“Dizhen” 地眞).

2. 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 47.2 (1987), 374–75.

3. , , “O Soul, Come Back,” 378. Many scholars have since echoed Yū's assertion that hunpo dualism was a facet of popular belief. For example, in his 1991 Ph.D. dissertation, Lo Yuet Keung acknowledges his indebtedness to Yū, and when writing on Wang Yi 王逸 (the second-century commentator to the Chu ci 楚辭), states, “It is clear that Wang Yi, like most people of the Han period, identified hun and po, the dual soul that sustains human life, with the yin-yang principle/” See Keung, Lo Yuet, “The Destiny of the ‘Shen’ (Soul) and the Genesis of Early Medieval Confucian Metaphysics (221-587 A.D.)” (Ph.D. diss., University of Michigan, 1991), 21.

4. Loewe, Michael, “The Religious and Intellectual Background,” in The Cambridge History of China, vol. 1, ed. Twitchett, Denis and Loewe, Michael (Cambridge: Cambridge University Press, 1986), 718.

5. Loewe, Michael, Ways to Paradise (London: George Allen and Uuwin, 1979), 9.

6. Seidel, Anna, “Tokens of Immortality in Han Graves,” Numen 29 (1982), 107.

7. Seidel, Anna, “Traces of Han Religion in Funeral Texts Found in Tombs,” in Dōkyō to shūkyō bunka 道敎宗敎文化, ed. Kan'ei, Akizuki 秋月觀暎 (Tokyo: Hirakawa, 1987), 2157.

8. Seidel, Anna, “Post-mortem Immortality, or: The Taoist Resurrection of the Body,” in Gilgul: Essays on Transformation, Revolution and Permanence in the History of Religions, ed. Shaked, S., et al. (Leiden: E.J. Brill, 1987), 227.

9. Muzhou, Pu (Mu-Chou Poo) 蒲慕少卜, Muzangyu shengsi: Zhongguogudai zongjiao zhi xingsi 墓葬與生死:中國古代宗敎之省思 (Taibei: Lianjing, 1993), 216.

10. In a recent review of Robinet's, IsabelleTaoist Meditation: The Mao-shan Tradition of Great Purity (Albany: State University of New York Press, 1993), Susan Whitfield writes that many facts about this hunpo dualism at different times in the history of Chinese religion are still unknown, but that “discussion of these seems to have gone out of fashion, occurring mainly in the works of sinologists of previous generations”; see Journal of the Royal Asiatic Society, 3rd ser,, vol. 4.3 (1994), 455.

11. The difference between scholasticism and general belief can be demonstrated in the Christian tradition's similar attempts at precisely defining the soul. For example, Origen (c. 185-254) is said to have envisioned the soul as wrapped in a corporeal envelope made of pneuma, thus explaining the appearance of phantoms and leading other early church scholars to ask. if this limbless, organless entity is spherical, cubical or polygonal. Much more recently, the Doctrine Commission of the Church of England has described the soul as a vastly complex “information-bearing pattern” of the body to be remembered by God, that pattern serving as a carrier of memories and of personality. While these definitions are part of our literary heritage on the science of death, we cannot regard them as ” general beliefs”. See Crouzel, Henri, Origen (Edinburgh: T&T Clark, 1989), 252–53; The Doctrine Commission of the Church of England, The Mystery of Salvation (London: Church House Publishing, 1995), 12, 191.

12. , , “Soul, Come Back,” 374; Xidan, Sun 孫希旦, Li ji jijie 禮記集解 (Taibei: Wen shi zhe, 1973), 652. The commentary by Zheng Xuan 鄭玄 (127–200 C.E.) is silent on these terms.

13. Suetoshi, Ikeda, “Konpaku kō” 魂魄考, Tōhō shūkyō 東方宗敎 3 (1953), 1.

14. Needham, Joseph, Science and Civilisation in China, vol. 5.2 (Cambridge: Cambridge University Press, 1974), 87.

15. Lin, Paul J., A Translation of Lao Tzu's Tao Te Ching and Wang Pi's Commentary, Michigan Papers in Chinese Studies no. 30 (Ann Arbor, 1977), 10.

16. Xidan, Sun, Li ji jijie, 534.

17. Xidan, Sun, Li ji jijie, 1117.

18. Shike, Chen 陳士河, Kongzijiayu shuzheng 孔子家語疏證 (jicheng, Congshu 叢書集成 ed.), 119 (“Aigong wen zheng” 哀公問政).

19. Tai, Zhong 鍾泰, Zhuangzi fawei 莊子發微 (Shanghai: Shanghai guji, 1988), 30 (“Qiwu lun” 齊物論).The Eastern Han divination text Yilin 易林 similarly balances the hun with the xing in the following verse:

年衰歲暮 At year's end and seasons' close,

精魂遊去 The quintessential hun wanders away,

形容消枯 The form declines and withers.

喪子相呼 Children in mourning cry out to one another.

See Jiao shi Yilin 焦氏易林 (jicheng, Congshu, ed.), 239 (“Zhen: daguo” 大過).

20. Xidan, Sun, Li ji jijie 266; repeated in Han shu 漢書 (rpt. Taibei: Zhonghua, 1986), 36.1953, and elsewhere. The other occurrence of hun is in the compound hunpo in “Liyun” (Sun Xidan, Li ji jijie, 539), the soul which sacrificers intended to please. Yet not only is this reference unique, it makes no parallelism, and it may be a loose usage of the term (see below). The rarity of these terms in the Liji, a work much devoted to funerary ritual, is notable in itself. If hunpo dualism were genuinely common in early China, one would expect the Li ji to reflect this popularity.

21. Lo, , “The Destiny of the ‘Shen,’24, wrote that “no later than the end of the second century, po lost its eschatological denotations and became more or less identical with the physical body,” thus losing its “life principle.” If these two Liji passages are reiterating the same parallelism, then the po in this passage also seems closely related to the body itself (i.e. the “bones and flesh”) and seems less like a “soul” assodated with the body. Seidel justifiably stated that the term “sour should not even be used in reference to hun and po, although she herself still used the term (Seidel, , “Postmortem Immortality,” 226).

22. Xidan, Sun, Liji jijie, 649–52.

23. , , “Soul, Come Back,” 375; Hawkes, David, The Songs of The South (Harmondsworth: Penguin Books, 1985), 224; Xingzu, Hong 洪興祖, Chu ci buzhu 楚辭注 (Zhongguo gudian wenxue jiben congshu 中國古典文學基本叢書 [Beijing: Zhonghua, 1983]), 198. Yü uses other popular sources to describe the afterlife abodes of hun and po. For heaven he uses the Mawangdui MÜS funeral banner (“O Soul, Come Back,” 368-69), but this description has been rightly questioned by Hung, Wu, “Art in a Ritual Context: Rethinking Mawangdui,” Early China 17 (1992), 111–44. Again for heaven, Yü uses the Taiping jing description of four celestial departments (“O Soul, Come Back,” 382-84), but the Taiping jing never depicts this place as an actual abode for the hun, nor is it contrasted with an earthly abode for the po. The Taiping jing regularly uses the terms hun and hunpo loosely; for example, the hunshen 魂神 or hunpo of a bad person faced accusations in the underworld. See Ming, Wang, Taiping jing hejiao, 74 (“Nuli wei shan fa” 努力爲善法) and 78 (“Fenjie benmo fa” 分解本末法). For the earthly afterlife abode, Yü cites excavated documents addressed to the underworld bureaucracy, such as an inventory sent to the lord of the grave excavated at Mawangdui (“O Soul, Come Back,” 384-85). However, this document does not refer to a po; like the Taiping jing, it also does not juxtapose this underworld against any heavenly hun abode. Nor is it even certain that this document belongs to the class of Han passports addressed to the underworld officialdom; as Zhou Shiyi 周士一 and Chen Kefeng 陳可風 translate it, “In the twelfth year, in the second month, on the twenty-fourth day, Chamberlain in charge of funeral service. A memorial about the funerary articles. Please check up on the list on delivery.” Thus it seems to be addressed to living officials responsible for verifying the grave goods in the tomb at the time of burial. See Juyou, Fu 傅舉有and Songchang, Chen 陳松長, Mawangdui Han mu wenwu, 馬王堆漢墓文物,trans. Shiyi, Zhou and Kefeng, Chen (Changsha: Hunan, 1992), 37; see also, Songchang, Chen, “Mawangdui sanhao Han mu mudu sanlun”馬 王堆三號漢墓木牘散論, Wenwu 文物 1994.6, 6469.

24. Xingzu, Hong, Chu ci buzhu, 83 (here translated by Hawkes, , The Songs of The South, 117).

25. Xingzu, Hong, Chu ci buzhu, 216–17.

26. A legitimate question to raise at this point is, if hunpo and lisan are compound terms, shouldn't the constituent parts of each compound still retain individual meanings? The tendency of classical Chinese is, of course, to decipher each character with a separate definition, but this is not always possible. W.A.C.H. Dobson describes the link between words in such a compound as a “simple connection (morphological hendiadys),” He wrote, “Pairs of words with identical or similar meaning, but which vary in the total extensions and intensions of their meanings, create a class of compound words the meaning of which is confined to that of the meanings shared in common by its elements.” Dobson's first example of morphological hendiadys is, coincidentally, lisan, and I believe hunpo would often—but not always—fall into this class as well, the shared meanings of hun and po being confined to the mental faculties upon which the body depends. See Dobson, W.A.C.H., Late Archaic Chinese (Toronto: University of Toronto Press, 1959), 12.

However, we should by no means altogether cease analyzing the individual meanings of hun and po, for some types of sources (such as medical texts) indeed distinguish between them. It is the context which dictates how much detail we attribute to the terms in the same way that, in English, folklorist or Halloween “spirits” are vaguely conceived but a Holy “Spirit” is more specifically conceived.

27. Loewe, , Ways to Paradise, 9.

28. Bojun, Yang 楊׳(白峻, Chunqiu Zuo zhuan zhu 春秋左傳注 (Beijing: Zhonghua, 1981), 1292–93 (Zhao 昭 7). In his commentary, Du Yu 杜預 (222–284 CE.) writes, “The po is the form” (po xing ye 魄,形也), which may be how the Liji interprets po as well.

29. lishixi, Beijing daxueLunhengxiaozu, zhushi, Lunheng zhushi 論衡注釋 (Beijing: Zhonghua, 1979) 1220–25 (“Siwei” 死偽).

30. , , “O Soul, Come Back,” 372–73.

31. lishixi, Beijing daxueLunhengxiaozu, zhushi, Lunheng zhushi, 1223. Zichan often appears in the Lunheng, but his description of the hunpo is never discussed.

32. Liqi, Wang 王禾器, Yantie lun jiaozhu 暨鐵論校注 (jicheng, Xinbian zhuzi 子集成 ed. [Beijing: Zhonghua, 1992]), 488 (“Zhu Qin” 誅秦) and 500 (”Xiyu” 西域).

33. Yu, Su, Chunqiu fanlu yizheng 春秋繁露義證(Taibei: He lu tushu, 1974), 142b143a (”Sandai gaizhi zhiwen” 三代改制質文).

34. Tai, Zhong, Zhuangzi fawei, 492–93 (“Zhi beiyou” 知北游).Gui 歸 “return” is a common early euphemism for death or burial.

35. Graham, A.C., Chuang-tzu; The Inner Chapters (London: Mandala, 1991), 133.

36. Powers, Martin J., Art and Political Expression in Early China, (New Haven: Yale University Press, 1991), 53.

37. The term po only occurs in one other passage of the Zhuangzi, as part of the compound zaopo 糟魄(dregs); see Tai, Zhong, Zhuangzi fawei, 305 (“Tiandao” 天道). In a different passage, Graham seems to render 鬼 as po 魄, thereby translating it as “anima,” but even if he is correct, the passage expresses no hunpo dualism. See Graham, , Chuang-tzu; The Inner Chapters, 260; Tai, Zhong, Zhuangzi fawei, 284 (“Tiandao” ).

38. A computer database search of these histories and their commentaries in the Zhonghua editions allowed me to locate each occurrence of the terms hun and po and study their context.

39. Seidel, , “Traces of Han Religion,” 28.

40. Seidel, , “Traces of Han Religion,” 30. This passage would contradict Yü's belief that the po is never summoned (, , “O Soul, Come Back,” 375).

41. Seidel, , “Traces of Han Religion,” 30. Elsewhere in the same year she wrote, “the hun as well as the po components of man, his whole social persona and individual being must descend under the earth” (Seidel, , “Post-mortem Immortality,” 228). Her position has taken root and has been subsequently cited by other scholars. For example, see von Falkenhausen, Lothar, “Sources of Taoism: Reflections on Archaeological Indicators of Religious Change in Eastern Zhou China,” Taoist Resources 5.2 (1994), 6.

42. Like Seidel, Stevan Harrell distinguishes between elitist hunpo dualism and common religion in his field study on the concept of “soul” in modern Taiwan: “So while it is undeniably true that some Chinese folk believers know about the distinction between hun and po on a theoretical level, it seems to fit badly, if at all, with the way they think about ‘souls’ acting or behaving. So we must reject the idea of two ‘souls’ as a basically analytical construct which has little bearing on behavior of folk believers” (Harrell, Stevan, “The Concept of Soul in Chinese Folk Religion,” The Journal of Asian Studies 38.3 [1979], 522).

43. Her data is mostly derived from the collection compiled by On, Ikeda 池田溫, “Chūgoku rekidai boken ryakukō” 中國歷代墓券略考, Tōyō bunka kenkyūjo kiyō 東洋文化硏究所紀要 86 (1981), 193278. See also Itaru, Tomiya 富谷至, “Kōsen no koku no tochi baibai” 黃泉の國の土地賣買, Kenkyū shūroku (Osaka daigaku)硏究集錄 (大 大學) 36 (1988), 132. Po occurs nowhere else in these passages, and hun only occurs in several titles of the underworld bureaucracy, such as the “Chief of the official's hostel at the gate of the hun” (hunmen tingzhang 魂門亭長). Interestingly, the “gate of the hun”, is also the name of an acupuncture-moxibustion node on the body (see footnote 61).

44. For an example of synonomia, Knechtges, David, The Han Rhapsody: A Study of the Fu of Yang Hsiung (53 B.C.-A.D. 18) (Cambridge: Cambridge University Press, 1976), 2324, translates the following statement by Su Qin 蘇秦 in the Zhanguo ce戰國策 which describes hereditary rulers as “Confused by words, befuddled by language, inundated by discourse, drowned by speech.” In the case of synonomia, it is difficult to read separate meanings into such pairings.

45. Other ordinances use yinyang cosmology in a different manner, the yang identified with the living and the yin with the dead. See On, Ikeda, “Chûgoku rekidai boken ryakukō,” 273, no. 7.

46. Hou Han shu 後漢書 (rpt. Beijing: Zhonghua, 1973), 56.1832.

47. Shao, Ying 應幼, Han guan yi (jicheng, Congshu ed.), 43.

48. This traditional attribution, which sources since the Tang generally maintain, is very much in doubt, but the Han origin of the text is less questioned. See Jiaxi, Yu 余嘉錫, Siku tiyao bianzheng 四庫提要辨證 (Beijing: Zhonghua, 1985), 741–58.

49. Jiao shi Yilin, 175 (“Jiaren: lü” 家人:旅).

50. For examples, see Hou Han shu, 32.1121; and Sanguo zhi 三國志 (rpt. Beijing: Zhonghua, 1975), 28.783.

51. Shuping, Wu 吳樹平, Fengsu tongy i jiaoshi 風俗通義校釋 (Tianjin: Tianjin renmin, 1980), 428 (“Yiwen” 佚文).

52. For an example of this practice, see Han shu, 67.2908. See also Thorp, Robert L., “Mountain Tombs and Jade Burial Suits: Preparations for Eternity in the Western Hanx” in Ancient Mortuary Traditions of China: Papers on Chinese Ceramic Funerary Sculptures, ed. Kuwayama, George (Los Angeles: Far Eastern Art Council/Los Angeles County Museum of Art, 1991), 34.

53. Hou Han shu, 15.582-84. A bodyless grave had at least one precedent in the Yellow Emperor's grave, said to contain only his clothing after he became an immortal. See Shiji 史記 (rpt. Taibei: Zhonghua, 1985), 28.1396.

54. Guowei, Wang 王國維, Shui jing zhu jiao 水經注校 (Shanghai: Renmin, 1984), 254 (“Ji shui” 濟水); and Shi ji, 8.342 (in commentary by Zhang Shoujie 張守節 [fl. 737]).

55. Ho, Wai-kam, “Hun-p'ing: The Urn of the Soul,” The Bulletin of the Cleveland Museum of Art 48.2 (1961), 2634; Hung, Wu, “Buddhist Elements in Early Chinese Art,” Artibus Asiae 47.3/4 (1986), 263352. On the jars in general see Mingliang, Xie 謝明良, “Sanguo Hang Jin shiqi yueyao qingci suojian de foxiang zhuangshi” 三國兩晉時期越窯靑瓷所見的佛像裝節, Gugong xueshu jikan 故宮學術季刊 3.1 (1985), 3568; and Ichirō, Kominami 、南––良, “Tsubokata no uchū” 壺型の宇宙, Tōhō gakuhō 東方學幸艮 61 (1989), 165221. A common motif on these jars is a hole in the base with a snake-like creature crawling into it. In the legends of Gaozu summoning his mother's hun (see footnote 54), the hun comes in the form of a snake.

56. Kejun, Yan 嚴可均, Quart Shanggu Sandai Qin Han Sanguo Liuchao wen 全上古三代秦漢三國六朝文 (Beijing: Zhonghua, 1958), 1781, 2172, 2190, 2195–96, 2271 (Quan Jin wen” 全晉文); Jin shu 晉書 (rpt. Beijing: Zhonghua, 1974), 83.2166. After the lengthy debate, the court decided that a dualism existed between grave and shrine (a topic to be taken up below), but there was a danger of confusing the summoned hun with the shen, and so it banned the practice that Emperor Guangwu had sanctioned. The jars that are thought to have housed the hun (see footnote 55) mostly predate the Jin's southward migration and could not have been a consequence of it, as some have argued.

57. The hurt's departure from the body at death was sometimes employed as an argument in favor of thrifty burials in the standard histories. During Emperor Ling's reign (168-189 C.E.), an upright scholar called Zhao Zi 趙咨 wrote to his son that, “As for the dead, their primal vital energy (yuanqi 元氣) leaves the body and the pure hun (zhenhun ) disperses, going back to the primary constitution of things and returning to the origin”; because nothing would be left in the grave, he begged for thrift in his own burial (Hou Han shu, 39.1314-15). The scholars Yang Wangsun 楊王孫 and Cui Yuan 崔瑗 put forward the same logic (Han shu, 67.2908; and Hou Han shu, 52.1724). In all these cases, they are arguing for a single entity leaving the body which would in turn disperse, not a duality of entities leaving the body.

58. Wendian, Liu 劉文典, Huainan honglie jijie 淮南鴻烈集解 (Taibei: Wen shi zhe, 1992), 270–71 (“Zhushu xun” 主術訓). The translation of the last line is problematic. Following the Wenzi, Roger Ames renders it, “Constantly being a companion to the process of transformation, it is beyond his comprehension (te)” (Ames, Roger T., The Art of Ruler ship: A Study of Ancient Chinese Political Thought [Albany: State University of New York Press, 1994], 169). As for the “mysterious abode,” it is located within the body, for the Huainanzi states:


If the hun and po occupy their dwellings and the quintessential spirit protects its root, then as for death and life, there will be no such transformations within the self, which is thus referred to as perfected spirituality.

See Wendian, Liu, Huainan honglie jijie, 227 (“Jingshen xun” 精神訓).Other Han works with clear hunpo dualism include the Yuejue shu 越絕書, an Eastern Han work akin to a local gazetteer, and the Baihu tong 白虎通, usually ascribed to Ban Gu 班固 (32-92 C.E.) but most likely written during or just after the Eastern Han. See Zumou, Yue 樂祖謀, Yuejue shu (Shanghai: Shanghai guji, 1985), 9394; Zeyu, Wu 吳則虞, Baihu tong shuzheng 白虎通疏證 (Beijing: Zhonghua, 1994), 389–90 (“Xingqing” ׳׳性情); and Som, Tjan Tjoe, Po Hu Tung: The Comprehensive Discussions in the White Tiger Hall, vol. 2 (Leiden: EJ. Brill 1952), 571. The much later Baopuzi by Ge Hong 葛洪(c.283-343) depicts the process of becoming a Daoist immortal, and some of its language is reminiscent of these Huainanzi passages. Within a list of fantastic feats that result in one's ability to whip on the wind and clouds, mount the great void and live forever, Ge Hong includes the prerequisite of “seizing the hun and controlling the po, causing the bones to be saturated and the frame to be light” (juhun zhiipo, gutian tiqing 拘魂制魄, 骨塡體輕).In like manner, the Taiping jing states, “Thus he composes his hun and pacifies his po, protects his embryonic state and treasures his spirit, affixes his quintessence and saturates his blood, solidifies his secretions and congeals his muscles” (yushi lianhun hepo, shoutai baoshen, lujing tianxue, guye ningjin 於是敛魂和魄, 守胎寶神,錄精塡血, 固液凝筋.). See Ming, Wang, Baopuzi neipian jiaoshi, 100 (“Zhi li” 至理); Ming, Wang, Taiping jing hejiao, 2 (“Taiping jinque”).

59. xiaozu, Mawangdui Han mu boshu zhengli, Mawangdui Han mu boshu 馬王堆漢墓帛書, vol. 4 (Beijing: Wenwu, 1985), 152. The translation is from Harper's, DonaldEarly Chinese Medical Literature: The Mawangdui Medical Manuscripts (London: Kegan Paul, forthcoming), except that he renders hun as “ethereal-spirit” and po as “earthly-spirit.” To my knowledge, the Mawangdui texts, like the Huangdi neijing texts, do not clearly distinguish an ethereal-earthly dualism in the terms hun and po, although as noted above, the Huainanzi does, so Harper's reading is plausible.

60. Zhongyi, Nanjing Zhongyi xueyuan xi, Huangdi neijing lingshu y ishi 黃帝內經靈樞譯 (Shanghai: Shanghai kexue jishu, 1986), 335 (“Tiannian” 天). The pair of “blood and vital energies” or “blood and air” by the time of Confucius refers to physical powers within the body, whereas the yingwei 營衛 are types of circulation functionally distinguished. Sivin, Nathan, Traditional Medicine in Contemporary China (Ann Arbor: Center for Chinese Studies, the University of Michigan, 1987), 4653, explains the vital energies of ying and wei when he writes, “The former, the purer (relatively yang) fraction extracted from air and aliment, circulates within the tract system; the latter, less pure, makes the round of the body outside the tracts, and defends its periphery.”

61. Zhongyi, Nanjing Zhongyi xueyuan xi, Huangdi neijing lingshu yishi, 7980 (“Benshen” 本神).There exist variations on this chart in other texts. See Shōha-chi, Nakamura 中村璋八, Wuxing dayi jiaozhu 五行大義校註 (Taibei: Wuling, 1986), 101–16 (“Lun pei zangfu׳/論配藏府); and Kalinowski, Marc, Cosmologie et divination dans la Chine ancienne: Le Compendium des cinq agents (Wuxing dayi, VIe siècle) (Paris: École Française d'Extrême-Orient, 1991), 289.

The Mingtang明堂uses the terms hun and po in two of the 163 names of bodily nodes at which acupuncture and moxibustion can be applied. Known as the pohu 魄戶 “door of po” and the hunmen 魂門 “gate of hun,” they are listed alongside the shentang 神堂 “hall of the spirit,” the yishse意舍 “the abode of thought,” and the zhishi 志室 “the room of will”—all located with additional nodes along the spinal column. See Longxiang, Huang 黃肯直祥 and Xuetai, Wang 王雪菩, Huangdi mingtang jing jijiao 黃帝明堂經輯校 (Beijing: Zhongguo yiyao keji, 1988), 4349.

62. Zhongyi, Nanjing Zhongyi xueyuan xi, Huangdi neijing lingshu yishi, 7980 (“Benshen”). I follow the commentators' interpretation. Note that qi 氣 has been translated in two different ways: as “vital energy,” of which any of the five organs can have too little or too much, and as “breath,” which is the tangible entity of the lungs alone.

63. Zhongyi, Nanjing Zhongyi xueyuan xi, Huangdi neijing lingshu yishi, 275 (“Yinxie fameng” 淫邪發夢).

64. The Huainanzi describes one who has become rooted into the cosmos as follows:


Therefore, his sleep is dreamless and his intelligence unmanifest; the po is not pressed downward and the hun does not ascend.

See Wendian, Liu, Huainan honglie jijie, 229 (“Jingshen xun”). Much later, Tao Hongjing 陶弘景 (456-536) would even identify the po-phantom or poyao 魄妖 (seven spirits that actively sought the dissolution of the mortal frame) as one of the three causes of foul dreams. See Harper, Donald, “Wang Yen-shou's Nightmare Poem,” Harvard Journal of Asiatic Studies 47.1 (1987), 272.

65. xi, Nanjing Zhongyi xueyuan Zhongyi, Huangdi neijing lingshu yishi, 514 (“Dahuo lun” 大惑論). In his commentary to the Liji, Zheng Xuan also equates the possession of hun and po with a sharpness of the faculties, defining the function of po as “a perceptive clarity of the ears and eyes” (ermu zhi congming 耳目之聰明; Xidan, Sun, Liji jijie 1117 [“Jiyi”]).

66. Xiang, Liu, Xin xu (jicheng, Congshu ed.), 8889 (“Zashi” 雜事). This is the only usage of the terms hun and po in the Xin xu.

67. Jiiao shi Yilin, 53 (“Fou: tongren” 否:同人).

68. Jiao shi Yilin, 124 (“Yi: fou” 頓:否). For other examples of the hun and po disrupted in an emotional situation, see Shi ji, 105.2790; Han shu, 87A.3526; Han shu, 87A.3536; and Hou Han shu, 84.2802.

69. Shan, Li 李善, Wen xuan文選 (Shanghai: Shanghai guji, 1986), 1.17 (“Liangdu fu” 兩都賦).

70. Shan, Li, Wen xuan, 3.133 (“Dongjing fu” 東京賦); translated bby Knechtges, David R., Wen xuan or Selections of Refined Literature, vol. 1 (Princeton: Princeton University Press, 1982), 309.

71. Shan, Li, Wen xuan, 2.69 (“Xijing fu” 西京賦); Han shu, 87A.3549. The hun of both humans and animals can be shaken by sad music; see Shan, Li, Wen xuan, 4.158 (“Nandu fu” 南都賦). A Han poem about a crow shot by a boy also describes its hunpo as ascending to heaven; see Song shu 宋書(rpt. Beijing: Zhonghua, 1974), 21.607. The hunpo of a mythical leader of the Ba巴by the name Lord Lin (Lin jun 糜君)even became an animal (a white tiger) at his death; see Hou Han shu, 86.2840; Sage, Steven F., Ancient Sichuan and the Unification of China (Albany: State University of New York, 1992), 5051.

72. Bojun, Yang, Chunqiu Zuo zhuan zhu, 765 (Xuan 宣 15).

73. Bojun, Yang, Chunqiu Zuo zhuan zhu, 1456 (Zhao 昭 25). Whereas Zhao Tong loses his po, the weepers at the banquet lose their hunpo, another example of the inter-changeability of these terms. In the thirty-first year of Duke Xiang 襄, a ruler who had not yet reached the age of fifty rambled on in a confused manner as if he were eighty. Although the passage does not refer to his hun or po departing, this confusion accords with the Lingshus description of eighty as being the age when the po departs. See Bojun, Yang, Chunqiu Zuo zhuan zhu, 1183.

74. For a discussion on the increasing awareness of the personality's opacity in the fourth century B.C.E. (including the case of Boyou), see Zoeren, Steven Van, Poetry and Personality: Reading, Exegesis, and Hermeneutics in Traditional China (Stanford: Stanford University Press, 1991), 5568.

75. Shouqi, Chen 陳壽視, Shangshu dazhuan jijiao 尙書大傳輯校 (xubian, Huang Qing jingjie 皇淸經解續編ed•), 2.9b. The Shang shu dazhuan as it currently stands is a collection of fragments from a second century B.C.E. work that seems to have been lost by the Song Dynasty. The passage, which Zheng Xuan here glosses, uses the term ke 腐 for “illness” and not bing 病as in its parallel passages. Zheng Xuan explains that a hunpo deficiency does not make itself manifest upon the body and so is not called a bing, and although his distinction may be nothing more than a textual rationalization, it lends credence to interpreting hunpo deficiency as a “mental illness.” Zheng Xuan cites the Zuo zhuan passage of heaven taking Boyou's po in the twenty-ninth year of Duke Xiang, after which Boyou dies a year later (Bojun, Yang, >Chunqiu Zuo zhuan zhu, 1168).

76. Hou Han shu, “Zhi” 志, 4.3111 (commentary of Liu Zhao 劉昭).

77. Qiyou, Chen 陳奇猷, Hanfeizi jishi 韓非子集釋 (Beijing: Zhonghua, 1958), 579 (“Neichu shuo xia” 內儲說下); Bodde, Derk, Festivals in Classical China: New Year and Other Annual Observances During the Han Dynasty 206 B.C.–A.D. 220 (Princeton: Princeton University Press, 1975), 275–77.

78. In Derk Bodde's translation of this passage (Festivals in Classical China, 275-77), he renders this event as summoning the souls of the dead, but I am hesitant to accept this interpretation for three reasons. First, the decoction made from Chinese thorough-wort (which he identifies as Eupatorium Chinense) was a medicine to restore mental equilibrium, and so this passage fits well among the other examples that correlate hun and po as necessities of good mental health. Second, this passage refers to an annual event of late spring or early summer, not to the soul-summoning ritual of the funeral process. Third, Bodde himself admits that he is puzzled by the phrase xu po 續魄or “perpetuating their po” but if the subject of this passage is living people (i.e. those going down to the Zhen and Wei rivers), then perpetuating the po accords with other texts, such as the Laozi and Huainanzi, which advocate maintaining one's po to attain longevity. See also footnote 82.

79. Zhongyi, Nanjing Zhongyi xueyuan xi, Huangdi neijing lingshu yishi, 337 (“Tian nian”). Lo, , “The Destiny of the ‘Shen,’20, rightly concludes from this, “As such, the loss of po would not threaten a person's life.” Lo suggests this was an Eastern Han development, shifting away from hunpo dualism, but in my opinion, hunpo dualism was never that strong before or during the Han.

80. Xidan, Sun, Li ji jijie, 1037–40 (“Sang daji” 喪大記). Wu Hung recently reviewed the early hun-summoning ceremony to demonstrate that it was not exclusively a death ritual but an attempt to recall the energies of one still alive although experiencing severe disease (Wu, , “Art in a Ritual Context,” 112–16). Yet as seen above, the ritual of hun-summoning seems to have evolved over the course of the Han to include the hun of the dead, as in the case of interring the summoned-hun of Emperor Guangwu's sister Yuan.

81. A century after the Han Dynasty, Ge Hong demarcated between disease and death by the degree of hunpo deficiency. He wrote:

人無賢愚,皆知己身之有魂魄,魂魄分去則人病,盡去則人死. 故分去則術家有拘錄之法,盡去則禮典有招呼之義.

No matter whether virtuous or foolish, everyone knows that one's own body possesses a hunpo. When the hunpo partially leaves, the person then sickens, and when it completely leaves, the person then dies. Therefore when it partially leaves, a specialist of the arts has a method of seizing and affixing; when it completely leaves, the ritual canon has its ceremony of summoning and beckoning.

See Ming, Wang, Baopuzi neipian jiaoshi, 1920 (“Lun xian” 論仙).

82. Harold D. Roth explores approaches to psychology in pre-Han and early Han texts, establishing a link between mental stability and possession of the quintessence in the Guanzi 管子. The mind is naturally filled with the quintessence and naturally tends to generate and develop it, but emotions, desires and selfishness deplete it. Breathing techniques will cause the quintessence to flow within the mind and promote longevity, a concept similar to the “dragon breathing” in the “Shiwen” which benefits the hunpo and also promotes longevity. The principle difference between this quintessence and the hunpo seems to be that the former is often described as flood-like, not limited to the confines of the body, whereas the latter is a more discrete inborn entity. In the Huainanzi, it is the sentient shen which is damaged by emotions and serves as the basis of all normal awareness. See Roth, Harold D., “Psychology and Self-Cultivation in Early Taoistic Thought,” Harvard Journal of Asiatic Studies 51 (1991), 599650.

83. For an introduction to Han stelae, see Ebrey, Patricia, “Later Han Stone Inscriptions,” Harvard Journal of Asiatic Studies 40 (1980), 325–53. Artistic and ideological formula dictated much of the content of pre-Han and Han ritual and religious art, so much so that funerary art “markets” can be geographically defined by concentrations of monuments of similar styles. See Powers, Martin J., Art and Political Expression in Early China, 110. Likewise, the Wu Liang 武梁 shrine carvings of the mid-second century C.E. employ conventional motifs found on other reliefs, although they simultaneously express a sense of individualism through the choice of those motifs. See Hung, Wu, The Wu Liang Shrine: The Ideology of Early Chinese Pictorial Art (Stanford: Stanford University Press, 1992), 223. Stelae, just becoming popular at this time, are also a mixture of convention (biography-hymn format, stock phrases, classical allusions) and individualism (the specific biographical details preserved in stone for future generations).

84. Gua, Hong, Li shi, in Shike shiliao xinbian 石亥!!史料新編 (Taibei: Xinwenfeng, 1977), ser. 1, vol. 9, 12.15a. Heaven snatches away the po in the Zuo zhuan and in the Da Dai liji 大戴禮言己.See Bojun, Yang, Chunqiu Zuo zhuan zhu, 1168 (Xiang 29); and Bian, Lu, Da Dai liji (jicheng, Congshu ed.), 198 (“Shaoxian” 少閒).

85. Gua, Hong, Li shi, 12.17a. The stele is undated, but her husband, presumably the author of the stele, died in 173 C.E. (Li, shi, 9.7a). There are no references to the po in Gua's, HongLi xu 録總, in Shike shiliao xinbian, ser. 1, vol. 10.

86. Gua, Hong, Li shi, 12.16b. “Receiving punishments” after death seems a relatively early concept for the Dynasty, Han. Yi Zhou shu 逸周書 (Taibei: Zhonghua, 1980), 8.7a (“Shiji’ 史言己解),refers to a “dark, capital” or xuandu 玄都. Earlier standard histories and the ” Yao dian” 堯典 both refer to a “dark capital” or youdu 幽都 in the far north but never a xuandu. The Chu ci (Xingzu, Hong, Chu ci buzhu, 201 [“Zhao hun”]) places youdu below ground.

87. Gua, Hong, Li shi, 5.5a. Another Li shi example, although of Jin origin, is a tomb restoration project of Xiahou Zhan 夏侯湛 (second half of third century), a couplet of which reads as follows:

有兆者表其墓 Where there was a cemetery, he marked the graves;

經墳者揖其魂 When he crossed through a burial ground, he paid his respects to their hun.

See Gua, Hong, Li shi, 19.20a. For a similar passage, see Jin shu, 38.1125-26.

88. lishixi, Beijing daxueLunhengxiaozu, zhushi, Lunheng zhushi, 1440 (“Jiechu” 解除).

89. For several examples of spirit roads, see Gua, Hong, Li shi, 13.2a9b.

90. Bojun, Yang, Chunqiu Zuo zhuan zhu, 1292–93 (Zhao 7). The meaning of wujing 物精 is unclear, but later in the same passage Zichan explains that because Boyou's clan was powerful, “its practical material has become bountiful, and its acquired quintessence has become plentiful” (qi yongwu ye hong yi, qi qujing ye duo yi 其用物也弘矣, 其取精也多矣). Whatever their precise meaning, they act as a pair. As for jingshuang 精爽, the Zuo zhuan elsewhere states, “The jingshuang of the heart/mind is called the hunpo” (xin zhi jingshuang, shi wei hunpo 心之精爽,是謂魂魄).See Bojun, Yang, Chunqiu Zuo zhuan zhu, 1456 (Zhao 25).

I am tempted to take yang in this passage as a reference to birth (i.e. entering the sunlight). Perhaps the infant is solely corporeal or po in the dark womb and at birth begins inhaling its ethereal nature or hun. The Zhuangzi uses yang in the sense of life where it states, “As for the heart/mind that is near to death, nothing can cause it to return to yang” (Jinsi zhi xin, mo shi fu yang ye 近死之心,莫使復陽也), See Tai, Zhong, Zhuangzi fawei, 31 (“Qiwu lun”). Perhaps in this Boyou passage there is even a juxtaposition between po with its lunar-phase derivations and yang with its solar connotations, but all this is only speculation.

91. Bojun, Yang, Chunqiu Zuo zhuan zhu, 1016 (Xiang 14).

92. Knoblock, John, Xunzi: A Translation and Study of the Complete Works, vol. 1 (Stanford: Stanford University Press, 1988), 252–55. Knoblock cites the Boyou story of the Zuo zhuan in his useful analysis of the term shenming, and he concludes from this story that the shenming “is present from the beginning of life itself,” that it is an inborn quality. Yet this story might also suggest that shenming is a later stage that can be reached if the hunpo is born into a materially and quintessentially rich environment, external factors that Zichan says come from the clan. Thus the hunpo then “reaches the point of spirit illumination” (zhi yu shenming 至於神明). A cumulative process toward an apex of shenming or spirit illumination also occurs in other early Chinese texts such as the Xunzi 荀子 in which constant studies and a day-after-day concentration of mind will lead to an accumulation of what is good until the earnest student penetrates into the shenming; see Xianqian, Wang, Xunzi jijie 荀子集解 (Beijing: Zhonghua, 1988) 443 (“Xing'e” ׳性惡).In the Mawangdui medical text “Shiwen,” the Heavenly Teacher exhorts the Yellow Emperor to examine the rules of yin and yang, “eat yin and secure yang and attain shenming” again indicating that shenming is not necessarily inborn but rather achieved. See Harper, Early Chinese Medical Literature: The Mawangdui Medical Manuscripts, Prolegomena, Section 4 (“Body and Spirit”), for a further discussion. For other examples in which the shenming seems to be an achieved rather than inborn state, see Roth, , “Psychology and Self-Cultivation in Early Taoistic Thought,” 625, 638, 642.

93. Wendian, Liu, Huainan honglie jijie, 227 (“Jingshen xun” ). The passage is translated in footnote 58.

94. There are at least eight examples of this phrase (or close variations of it) in the Li shi. See Gua, Hong, Li shi, 6.4b, 6.18a, 6.21b, 8.7a, 8.10b, 12.11b, 15.3b, and 18.4b. Another four appear in the Hou Han shu, 10A.406, 34.1174,42.1441, and 54.1768.

95. Perhaps the opening couplet to the anonymous hearse-puller song entitled Haoli 蒿里 reflects this lack of consciousness:


Whose land is Haoli? Of the hunpo gathered there, none are virtuous or foolish.

See Qinli, Lu 逯欽立, Xian Qin Han Wei Jin Nanbei chao shi 先秦漢魏晉南北朝詩 (Beijing: Zhonghua, 1984), 9.257.

96. Gua, Hong, Li shi, 6.18a. The translation of hun er youling 魂而有靈 is tentative. I have translated it as a conditional clause for two reasons: it has common variants such as hun ru youling 魂如有靈 (Li shi, 8.7a) and ruo hunpo youzhi 若魂魄有知 (Jin shu, 96.2525); and uncertainty as to whether the dead possessed consciousness was indeed discussed by groups such as the Mohists (Yujiang, Wu 吳毓江, Mozi jiaozhu 墨子校注 [Beijing: Zhonghua, 1993], 338 [“Minggui xia” 明鬼下]) and individuals such as Wang Chong (lishixi, Beijing daxueLunhengxiaozu, zhushi, Lunheng zhushi, 1312–17 [“Bozang” 薄葬]). Xunzi argued that the dead lacked consciousness but still deserved respect; see Xianqian, Wang, Xunzi jijie, 359 (“Li lun” 禮論). Ren Mo 任末, in his brief biography (Hou Han shu, 79.2572), said:


You must place my corpse at the capital gate. If the dead are conscious, my soul will not be ashamed. If they are not conscious, then I will simply become earth.

As to the term “conscious,” ling 靈 and zhi 知 are interchangeable in the above examples, and the Zuo zhuan similarly raises the question whether the ancestors or the dead “possess consciousness” (youzhi 有矢口); see Bojun, Yang, Chunqiu Zuo zhuan zhu, 1160 (Xiang 29) and 1552 (Ding 定 5); see also Han shu, 67.2908. As these sources indicate, ling is a condition in which the deceased can experience emotion, such as being greatly pleased (by a stele inscription) or feeling unashamed (by being placed at the capital gate). Zhang Heng 張衡 (78–139 C.E.) in his “Dulou fu” 髑髏賦 (Poetic exposition on the skull) wrote that when Zhuangzi's skull became ling, he could then hear the spirit but not see its form. See Zhenze, Zhang 張震澤, Zhang Heng shiwen ji jiaozhu 張谂詩文集校注 (Shanghai: Shanghai guji, 1986), 247. However, the English word “conscious” is not an exact fit. If ling is used the same way in all the above passages, the term seems to imply preternatural potency which is conscious, animate and even efficacious. One unique Li shi stele states, “If the ling possess zhi, then may it defend and bless its descendants” (ling ye youzhi, youfu zisun 靈也有知祐福子孫). In this case, ling may be the same as hun because hurtling is a common term for “soul,” again demonstrating the interchangeability of these terms. See Gua, Hong, Li shi, 11.16a.

97. Gua, Hong, Li shi, 10.8b. For other stelae with the body/social identity dichotomy, see Li shi, 10.25b, 11.7a, 12.7a, and 12.9b. Stelae often refer to this dichotomy because they themselves were a vehicle for making identity permanent. This is evident in the following extract from the inscription of Yang Tong 楊統 (d. 168), chancellor of Pei 市:


By engraving the stone and erecting the stele, the inscription of merit is made vastly illustrious. It will be radiant for a hundred-thousand years so that it will never be extinguished.… Establishing one's words so that they do not decay is what our ancestors treasured. Recording one's name on metal and stone is to hand it down to infinity.

See Gua, Hong, Li shi, 7.16b. The penultimate sentence alludes to the Zuo zhuan (see Bojun, Yang, Chunqiu Zuo zhuan zhu, 1088 [Xiang 24]).

98. Shusen, Sui 隋樹森, Gushi shijiushou jishi 古詩十九首集釋 (Hong Kong: Zhonghua, 1989), 1718.

99. Gua, Hong, Li shi, 10.25b.

100. In an earlier issue of this journal, Wu Hung skillfully traces the roles of shrine and tomb, but he argues that when Emperor Ming (r. 57-75 C.E.) abolished the shrine sacrifice in 58 C.E., shrine-tomb dualism disappeared in the Eastern Han, resulting in the shrine becoming nominal and “the total independence of the tomb as the center of ancestral worship” for the remainder of the dynasty (Hung, Wu, “From Temple to Tomb: Ancient Chinese Art and Religion in Transition,” Early China 13 [1988], 101–2; repeated in Monumentality in Early Chinese Art and Architecture [Stanford: Stanford University Press, 1995], 119–21). The cemetery indeed played a much more significant role in the Eastern Han, and its importance is verified by Eastern Han scholars such as Wang Chong (lishixi, Beijing daxueLunhengxiaozu, zhushi, Lunheng zhushi, 1329) who writes, “In ancient ritual there were shrine offerings; in current custom there are grave sacrifices” (guli miaoji, jinsu must 古禮廟祭,今俗墓示已). However, there is much evidence that the lineage shrine did not become purely nominal as a result of the cemetery's new significance. For example, Sima Biao 司馬彪 (240-306) described the Eastern Han's imperial shrine sacrifices (Hou Han shu, “Zhi,” 9.3197) as follows:


By Emperor Ling's reign, altogether eighteen sets of grand-pen animals plus supplemental animals were used at the seasonal sacrifices in the capital to the five tablets in the Gaozu Shrine, the seven tablets in the Shizu Shrine, the three graves of the emperors who had died young, and the three graves of retrospectively honored empresses.

Thus in the seasonal sacrifices, only women and children receive their offerings at the grave; the accumulation of adult emperors is instead still worshipped at the lineage shrine. Cai Yong 蔡昌(133-192 CE.) provides the same information in the Du duan獨斷 (Congshu jicheng, ed.), 21–22, and so in addition to individual grave sacrifices, the imperial lineages of the Western and Eastern Han each possessed an active Luoyang shrine. Furthermore, as recorded in the Hou Han shu (4.167,5.205,6.250,7.288,61.202930), new emperors were presented before these two shrines, and debates were sometimes carried out as to the order of their tablets. The Dongguan Hanji preserves detailed discussions (all post-dating 58 C.E.) on how music and dances were to be offered in worship at these shrines; see Shuping, Wu 吳樹平, Dongguan Han ji jiaozhu 東觀漢言己校注 (Chengchun: Zhongzhou guji, 1987), 165–66. Finally, Cai Yong's prayer to the shrine tablets informing them of their move to Chang'an in 190 C.E. likewise survives; see Cai Zhonglang wenji 蔡中郎文集 (Sibu congkan ed.), 8.4a–5a.

I would suggest that Emperor Ming never abolished the temple sacrifice in 58 C.E. Wu Hung's evidence (Hou Han shu, 2.99) only states that Emperor Ming ceremonially opened court at the tomb of his father. Such a ceremony may have played a part in the larger revitalization of the Zhou image because.King Wen was thought to have carried out a similar activity at his father's grave. See Weiyu, Xu 許維逼, Lü shi chunqiu jishi 呂氏春秋集釋 (Beijing: Beijing shi Zhongguo shudian, 1985), 21.2a-b (“Kai chun” 開舂); lishixi, Beijing daxueLunhengxiaozu, zhushi, Lunheng zhushi, 1228 (“Si wei” 死僞). However, this ceremony does not imply an abolition of shrine sacrifices.

Moreover, the sacrificial separation between lineage shrine and tomb also exists below the imperial level in the Eastern Han. In the Simin yueling by Cui Shi 崔蹇 (c. 110-170 C.E.), this middle-level Eastern Han official describes an estate's various lineage sacrifices with their prior abstentions, purifications, and specific sacrificial goods, but during the second, eighth, and twelfth months an additional sacrifice at the graves is made, before which the grave records or zhongbu 冢簿 are consulted. See Shenghan, Shi 石聲漢׳ Simin yueiing jiaozhu 四民月令校 (Beijing: Zhonghua, 1965), 19, 60 (one edition has jiabu 家簿, which Shi Shenghan believes is a mistake for zhongbu).

In contrast to lineage shrines, an individual's shrine was sometimes built in the cemetery as well, the Wu Liang shrine being our best understood example thanks to Wu Hungf s impressive study. Too few inscriptions survive to determine whether these shrines generally favored the shen as in a lineage shrine, revered the individual's body with its hunpo in the tomb, or made any such distinction. Above (footnote 87), one stele refers to the cemetery's hun-structures, and the cemetery shrine of Xiang Tajun 蘅他君 records that the shrine was erected so that the parents' hunling could rely upon it. For the latter, see Fuyi, Luo 羅示畐頤, “Xiang Tajun shicitang tizi jieshi” 藏他君石祠堂題字解釋, Gugong bowuyuan yuankan 故宮博物院院刊 2 (1960), 178–80. The cemetery stele and shrine (the latter is called a tang 堂 and not a miao 廟 in the Xiang Taj un inscription, Wu Liang stele, and other Li shi stelae) could thus be construed as an extension of the individual's tomb, but here I only speculate amidst scant evidence on the varied and complex Han conceptions of death (for further sources dividing the roles of tomb and lineage shrine, see footnote 117).

101. Han shu, 73.3115; Hou Han shu “Zhi,” 9.3199.

102. fenhui, Zhongguo shufa jia xiehui Shandong, Han bei yanjiu 漢碑硏究(Jinan: Qi Lu, 1990), 354. Such morning and evening inquiries are said to be characteristic of the nobility (gongjia 公家). See Hou Han shu, 31.1119.

103. fenhui, Zhongguo shufa jia xiehui Shandong, Han bei yanjiu, 351.

104. Poo, Mu-chou, “Ideas Concerning Death and Burial in Pre-Han and Han China,” Asia Major, third ser., 2 (1990), 2562; Rawson, Jessica, Ancient China: Art and Archaeology (London: Book Club Associates, 1980), 200204. Recently von Falkehausen, , “Sources of Taoism,” 7, aptly describes Han tombs as “like dollhouses or iaternae magicae for the dead spirits to play with—and never to leave.” Such lavish-ness in the tomb has been used as evidence for hunpo dualism, but as already noted it only indicates an entity within the tomb, not the ym-counterpart to the ascending hun. Moreover, lavish burials pre- and post-date the popularity of hunpo dualism. Whether that entity was trapped in the tomb, had the ability to come and go, or inevitably, unconsciously dissipated was another matter much debated in the Han. For theories on tomb lavishness pleasing the po, see Loewe, Michael, Chinese Ideas of Life and Death: Faith, Myth and Reason in the Han Period (202 B.C.-A.D. 220) (London: George Allen and Unwin, 1982), 114–17, 120–22.

105. For example, Wang Chong writes that grave fineries are to please the jinghun 精魂; see lishixi, Beijing daxueLunheng zhushixiaozu, , Lunheng zhushi, 1309 (“Bozang”).

106. Deshi, Huang 黃得時, Xiao jing jinzhu jinyi 孝經今註今譯 (Taibei: Taiwan shangwu, 1987), 33 (“Ganying” 感應), and 36 (“Sangqin”); Xidan, Sun, Li ji jijie, 1118 (“Jiyi”); and lishixi, Beijing daxueLunhengxiaozu, zhushi, Lunheng zhushi, 1449 (“SiYi” 禮運).

107. A tidy father-son line of descent rarely lasted long, especially in the Han royal lineage. At the shrine sacrifices, ninth-generation Emperor Guangwu 光武 (r. 25-57 C.E.) was told to worship eighth-generation Emperor Yuan 元(r. 49-33 B.C.E.) and seventh-generation Xuan 宣 (r. 74-49 B.C.E.) as his own father and grandfather respectively, even though in reality he was only distantly related to them (Hou Han shu, 35.1194). Similarly, Emperor Shang 蕩(r. 106 C.E.), who died as an infant, was to be worshipped as a father by his successor (Hou Han shu, 61.2029-30).

108. For examples of this relative valuation, see Yu, Su, Chunqiu fanlu yizheng, 7.18a (“Sandai gaizhi zhiwen”); Han shu, 73.3129; and Hou Han shu, 61.2029-30.

109. Wei Xuancheng 韋玄成, chancellor under Emperor Yuan, summarized the system as follows (Han shu, 73.3118):


Setting up the four shrines is in order to mark your closeness to recent ascendants. As the ascendants fade, the shrines are decommissioned in succession. The gradual decline from near to far demonstrates that there is an (ultimate) end.

110. For example, it was argued that Emperor Wu 武 (r. 141-87 B.C.E.) should receive perpetual remembrance because his social identity existed beyond the normal limitations of memory, his accomplishments having affected the empire's national borders, religion, and calendars of much later generations (Han shu, 73.3126). In a paper entitled “Death as Controlled Transformation in the Han/' presented at the 1995 Annual Meeting of the Association for Asian Studies, I argued that the court at the end of the Western Han consciously cast aside the tidy concept of mechanically discarding the oldest ascendants. By Emperor Ling's reign, all seven of the previous adult emperors had been granted permanent tablets with perpetual sacrifices, and the permanence of each, like that of Emperor Wu, was justified in terms of their long-lasting accomplishments.

111. According to the Wei commentator Meng Kang 孟康, the qishi (seven origins) in the music title are heaven, earth, humans, and the four seasons; huashi (blossoming origin) is the budding of the myriad things.

112. Han shu, 22.1046.

113. Han shu, 22.1043. As grand master of rituals, Shusun Tong established the ceremonial rules for the ancestral shrines and settled the various ceremonial procedures of the Han as a whole, founding the scholastic ritual tradition of the Han (Shi ji, 99.2724-25).

114. Xidan, Sun, Li ji jijie, 535 (“Liyun” 禮運).

115. Shenghan, Shi, Simin yueling jiaozhu, 1. In Eastern Han Luoyang as well, the spirits came to watch and feast at the imperial ancestral shrines; see Shan, Li, Wen xuan, 3.116 (“Dongjing fu”).

116. Just before the imperial age, Xunzi (Xianqian, Wang, Xunzi jijie, 371 [“Li lun”]) brings together these three elements when he writes:

故葬埋, 敬藏其形也; 祭祀, 敬事其神也; 其銘, 繋世׳敬傳其名也.

Thus the burial is to respectfully hide away the form; the sacrifice is to respectfully serve the spirit; and the inscription, lament as well as the genealogical record are to respectfully transmit the name.

I do not wish to imply that name preservation is limited to the shrine; grave stelae also play a significant role in perpetuating an individual's identity. However, the shrine does seem to play a more significant role in fostering lineage identity as it defines an ascendant relative to all other ascendants.

117. A century after the Han, the erudite Fu Chun 傅純 summarizes this distinction when he advises Emperor Yuan (r. 317-323 C.E.) on the creation of ritual by the sages (Jin shu, 59.1626):

設冢槨以藏形,而事之以凶; 立廟祧以安神,而奉之以吉.送形而往,迎精而還. 此墓廟之大分,形神之異制也.

They built mounds and vaults to store away the form and sacrificed to it by means of mourning sacrifices. They erected shrines and altars to pacify the spirit and made offerings to it by means of celebratory sacrifices. They saw off the form and it departed; they welcomed the quintessence and it came back. This is the great distinction between tomb and shrine and the different regulations of form and spirit.

This Jin shu distinction between xiong凶 (mourning) and ji 吉 (celebratory) sacrifices has earlier origins. For Xunzi, the mourning sacrifice for the dead is an ugly, sad and anxious event, and the celebratory sacrifice of the living is an elegant, musical and happy event. (The Baihu tong similarly limits weeping to the tomb and music to the shrine.) The Li ji describes funeral vessels as mourning vessels, and while music is not to be played at the grave, the ancestral shrine sacrifice is never to be described as a mourning sacrifice. He Xiu 何休 (129-82 C.E.), in his commentary to the Gongyang zhuan 公羊傳, also describes the sacrifice to pacify the spirit as a celebratory sacrifice. See Xianqian, Wang, Xunzi jijie, 361–63 (“Li lun”); Zeyu, Wu, Baihu tong shuzheng, 558 (“Zangbeishou” 葬4匕首); Xidan, Sun, Li ji jijie, 104–5 (“Quli xia” 曲禮卞); and Gongyang zhushu 公羊注疏 (zhushu, Shisan jing 十三經注疏 ed. [Kyoto: Chūbun, 1974]), 13.4a (Wen 文 2).

118. The ranking system consists of twenty levels, the lowest eight of which could be bestowed upon all male commoners apart from slaves. An enthronement, the change of a reign title, the investiture of an heir apparent, or other such occasion might result in a partial or empire-wide bestowal of an extra rank. Approximately two hundred such bestowals are recorded for the Han Dynasty, resulting in the oldest members of a village possessing the highest ranks. The banquets were probably held at the village shrine, thus making it a religious occasion. See Sadao, Nishijima, “The Economic and Social History of Former Han,” in The Cambridge History of China, vol. 1, ed. Twitchett, and Loewe, , 552–53.

119. Correlating social identity with adulthood is, of course, not unique to Han China. The anthropologist Robert Hertz, when writing on Indonesia at the beginning of this century, offers the following insight:

The deaths of children thus provoke only a very weak social reaction which is almost instantaneously completed. It is as though, for the collective consciousness, there were no real death in this case. Indeed, since the children have not yet entered the visible society, there is no reason to exclude them from it slowly and painfully.

See Hertz, Robert, Death and the Right Hand, trans. Rodney, and Needham, Claudia (Glencoe: The Free Press, 1960), 84.

120. The Lunheng and passages in the Shuoyuan 說苑 even argue that such an uncertainty was necessary lest the people turn away from the rituals that instilled within them a sense of hierarchy. See lishixi, Beijing daxueLunhengxiaozu, zhushi, Lunheng zhushi, 1315 (“Bozang” ); Shanyi, Zhao 趙善諮, Shuoyuan shuzheng (Shanghai: Huadong Shifan daxue, 1985), 558 (“Bianwu” 辨物).

* I would like to express my appreciation to Wu Hung for first arousing my interest in this topic, to Mark Edward Lewis, David McMullen, Joseph McDermott, and John EC. Mof fett for providing valuable direction and for reading earlier drafts, and to Donald Harper and the assessors of Early China for their extensive comments and criticisms. All errors in judgement and fact are solely my own.

Related content

Powered by UNSILO

Han Thanatology and the Division of “Souls”*

  • K.E. Brashier (a1) and 白瑞旭


Altmetric attention score

Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed.