No CrossRef data available.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 17 October 2022
The White Tiger Hall conference, held in the fourth year (79 c.e.) of the Jianchu 建初 reign in the Eastern Han, was a significant event in both politics and classical learning during and after that time. As the summary of the conference, composed after its conclusion, the Baihu tong 白虎通 is the main resource for investigating the details of this conference. Clarifying the formation process of the Baihu tong is helpful to elicit information regarding the White Tiger Hall conference from the findings recorded in its text. By tracing the history of the court conferences as an administrative institution and considering the particular nature of manuscript compilation, textual genre formats, and literary circulation during the Han, this paper suggests that the Baihu yizou 白虎議奏 referred to in the sources represents the compilation of the positions of the different debaters during the conference by Chunyu Gong 淳于恭 that was eventually sent to Emperor Zhang 章帝 for his approval; the Baihu tongde lun 白虎通德論 would be the corpus of the final rulings that had already been compiled before the conference ended and then edited by Ban Gu 班固. Later, the Emperor instructed his archivists to compose the Baihu tong by condensing the Baihu tongde lun. According to its formation process, the Baihu tong is the work of a collection of experts, rather than a compilation by a single person. Evidence shows that, although Emperor Zhang could weigh in on the court discussions (chengzhi linjue 稱制臨決), he could not ignore the consensus, nor could he simply mandate that the conference participants agree with him. In this regard, the Baihu tong cannot be considered a synthesis of the court's findings, establishing a single court ideology. Rather, it is best to see the text we have now as evidence of vigorous debates among the conference participants, including the Emperor himself and a range of other officials. In conclusion, the best way to uncover the facts about the White Tiger Hall conference via the Baihu tong is to reverse the process of textual formation, to glean information about the probable historical basis for the disputes recorded in the text.
東漢建初四年（79）所舉行的白虎觀會議在當時和稍後的政治、經學領域都是一次重要事件。會議之後，根據會議結論所編撰的《白虎通》是當前探尋白虎觀會議細節的主要資源。而弄清《白虎通》文本的生成過程，則對於學者們利用其中的記錄以發掘白虎觀會議的具體信息有著直接的幫助。本文將《白虎通》文本的生成過程置於漢代會議文書的生成和流轉之下進行考察，兼以文本形態的比對，認爲《白虎議奏》是會議參與者的議文經過淳于恭彙集奏上，等待章帝批答的文本總集合。而《白虎通德論》很可能是所有議奏經過章帝批答，在會後由班固纂集而成。其後，章帝又命史臣將《白虎通德論》壓縮編輯成《白虎通》。一旦清楚了這一過程，就可以明白《白虎通》並非一人之著述，其中凝聚著會議參與者集體的努力。證據顯示，儘管章帝可以在會上”稱制臨決,”但他既不能忽略大臣們的共識，也難以強行令大臣們附議他的觀點。緣此，《白虎通》的內容不能被視爲當時朝廷的共識，特別是不能被認爲建立起了一個統一的意識形態。相反，它是一個會議各方力量博弈的產物。要通過《白虎通》真正接近白虎觀會議的細節，惟有與其文本生成的過程反向而行，將抽象的問答還原爲具體的歷史場景，從文本的縫隙之中發掘歷史的信息。
I would like to thank Prof. Michael Nylan at UC Berkeley for fully guiding my study of the Baihu Tong, Kevin J. Turner at Peking University for greatly improving my written English, as well as Andrew Hardy at UC Berkeley and two anonymous reviewers for their critical comments and suggestions.
1. Fan Ye, Hou Han shu (Beijing: Zhonghua, 1965), 3.138.
2. Hou Han shu, 79A.2546.
3. Hong Ye 洪業 claims that the present Baihu tong is not the work deriving from the White Tiger Hall conference but a later compilation that was forged between 213 and 245. However, Tjan Tjoe Som 曾珠森 believes that Hong’s conclusion is invalid, because Hong’s arguments are based on only one quotation, which cannot impugn the general authenticity of the whole book. Nowadays, most researchers take Tjan’s side. See further “Baihu tong yinde xu” 白虎通引得序, in Hong Ye, Hong Ye lunxue ji 洪業論學集 (Beijing: Zhonghua, 1981), 31–36; Tjan Tjoe Som, Po Hu T’ung: The Comprehensive Discussions in the White Tiger Hall (Leiden: Brill, 1949 and 1952), Vol. 1, 30–31.
4. An example can be found in Chapter 8 (“Diguo yishi xingtai de chongjian—Banyan ‘Guoxian’ jichu de Baihu tong sixiang” 帝國意識形態的重建——扮演”國憲”基礎的白虎通思想), in Lin Congshun 林聰舜, Handai ruxue biecai: Diguo yishi xingtai de xingcheng yu fazhan 漢代儒學別裁:帝國意識形態的形成與發展 (Taipei: Taida, 2013), 213–62.
5. For instance, Hou Wailu 侯外廬 emphasizes that considering the operation of the White Tiger Hall Conference, as the “national fundamental law” 國憲), Baihu tong was finally approved by the emperor who acted as the patriarch and the religious leader. See Wailu, Hou, et al. , Zhongguo sixiang tongshi 中國思想通史 (Beijing: Renmin, 1957), Vol. 2, 226–27Google Scholar.
6. Hou Han shu, 3.138.
7. Hou Han shu, 40B.373.
8. Hou Han shu, 79A.2546.
9. Yong, Cai, Cai Zhonglang ji 蔡中郎集, in Yingyin Wenyuange Siku quanshu 景印文淵閣四庫全書 (Taibei: Shangwu, 1986), 1063 ce, 2.171Google Scholar.
10. Zhuang Shuzu, “Baihu tongyi kao,” in Chen Li 陳立, Baihu tong shuzheng 白虎通疏證, ed. Wu Zeyu 吳則虞 (Beijing: Zhonghua, 1994), Appendix 2.605. Wu, in the process of putting the Baihu tong shuzheng in order, also gathered together eight essays on the compilation, including Zhuang Shuzu’s and Liu Shipei’s work as appendices; unfortunately those appendices do not have the normal juan numbers. Zhuang’s essay is the second of the eight; hence my way of referring to it.
11. In early China, “bamboo bundle” (juan) was a unit referring to the length of writing material, but “chapter” (pian) was a unit referring to a complete article. See Li Ling 李零, Jianbo gushu yu xueshu yuanliu 簡帛古書與學術源流 (Beijing: Shenghuo dushu xinzhi sanlian shudian, 2007), 130. A bamboo bundle could contain several short chapters or parts of a long chapter, or it could be exactly equal to a complete chapter. See, Chen Mengjia 陳夢家, Hanjian zhuishu 漢簡綴述 (Beijing: Zhonghua, 1980), 305.
12. Wei Zheng 魏徵 et al., Sui shu (Beijing: Zhonghua, 1973), 32.937.
13. The present Baihu tong has 43 chapters.
14. Liu Shipei, “Baihu tongyi yuanliu kao,” in Baihu tong shuzheng, ed. Wu, Appendix 7.783–84.
15. Yong Rong 永瑢 et al., Siku quanshu zongmu (Beijing: Zhonghua, 1965), 118.1015.
16. Yao Zhenzong, “Hou Han Yiwen zhi,” in Xuxiu Siku quanshu 續修四庫全書 (Shanghai: Shanghai guji, 2002), 914 ce, 1.242.
17. Sun Yirang, “Baihu tongyi kao xia” 白虎通義考下, in Sun Yirang’s Zhouqing shulin 籀廎述林, ed. Xu Jialu 許嘉璐 (Beijing: Zhonghua, 2010), 1.46.
18. First Preface, Baihu tong (Baojing tang ed.), in Congshu jicheng chubian (Beijing: Shangwu, 1936), 4–5.
19. Liu, “Baihu tongyi yuanliu kao,” Appendix 7.785.
20. Hou Han shu, 3.138.
21. Hou Han shu, 40B.1373.
22. Hou Han shu, 79A.2546. The translations of these three passages are based on Tjan Tjoe Som, Po Hu T’ung: The Comprehensive Discussions, Vol. 1, 5–7. But I have also altered them in places for greater precision and to maintain consistency.
23. See, for example, “Jiōshō to chōgi” 丞相と朝議, Chapter 2 (Part 1), in Ōba Osamu 大庭脩, Shin Kan hōseishi no kenkyū 秦漢法制史の研究 (Tōkyō: Sōbunsha, 1982), 46–50; Nagata Hidemasa 永田英正, “Kandai no shūgi ni tsuite” 漢代の集議について, Tōhō gakuhō 43 (1972), 97–136; Rong Yuanda 榮遠大, “Han Jin jiyi zhidu chutan” 漢晉集議制度初探, Nanchong shiyuan xuebao (Zhexue shehui kexue ban) 1989.1, 118–24; Chapter 3 (“Zhongyang juece xitong” 中央決策系統), in Meng Xiangcai 孟祥才, Zhongguo zhengzhi zhidu tongshi 中國政治制度通史 (Beijing: Renmin, 1996), Vol. 3 (Qin Han juan), 118–53; Liao Boyuan 廖伯源, “Qin Han chaoting zhi lunyi zhidu” 秦漢朝廷之論議制度, in Qin Han shi luncong 秦漢史論叢 (Beijing: Zhonghua, 2008), 130–69; Li Xueming 李學銘, “Dong Han zhongyang jiyi zhidu zhi tantao” 東漢中央集議制度之探討, Xinya xuebao 29 (2011), 1–64; and Qin Tao 秦濤, Lüling shidai de “Yishi yi zhi”: Handai jiyizhi yanjiu 律令時代的“議事以制”: 漢代集議制研究 (Beijing: Zhongguo fazhi, 2018).
24. Ban Gu, Han shu (Beijing: Zhonghua, 1962), 1A.76.
25. Because the emperor replies to the memorials, there are the two characters of zhiyue, which is why they are often mistaken as a “decision” (zhishu 制書), but this is not the case. Cai Yong’s Du duan 獨斷 says that edicts have three types, of which the second type is: “when the ministers have something to memorialize and the Director of the Secretariat memorializes it for them, at the end of the memorial there is zhiyue, and the Son of Heaven responds to this by saying ‘Approval’ or by ‘Hand this down to such-and-such officials’, and so on, then this, too, is called an edict.” Therefore, it is known that the memorials become a second kind of edict after having the emperor’s replies attached with the phrase zhiyue. See, Cai Yong, Du duan, in Yingyin wenyuange siku quanshu, 850 ce, 1.78. For more on this issue see Dai Guoxi 代國璽, “Handai gongwen xingtai xintan” 漢代公文形態新探, Zhongguoshi yanjiu 2015.2, 23–49. Additionally, the Du duan translation is from Enno Giele, but I have modified his translations in some places for precision and to maintain consistency. The same is the case for all following Du duan translations. See, Giele, Enno, Imperial Decision-Making and Communication in Early China: A Study of Cai Yong’s Du duan (Wiesbaden: Harrassowitz Verlag, 2006), 234–35Google Scholar.
26. Han shu, 44.2141.
27. Du duan, 1.79 (mod. from Giele’s trans., as above, 186).
28. Han shu, 71.3043.
29. The yi represents in many cases the exposition of the discussions, which discuss policy proposals.
30. Cai Zhonglang ji, 2.173.
31. Because this yi is seen in Cai Zhonglang ji, it was probably written by Cai Yong himself and Academician Ren Min just appended his name to it to show his support.
32. The Du duan: “Presentations do not require a ‘head.’ At the beginning of a written presentation, those who submit it state, ‘Your minister so-and-so reports the following.’ At the end, they state, ‘Truly fearful and terrified, I keep knocking my head to the floor, doubly deserving capital punishment.’ At the end of the text on the accompanying boards appended to the left they add at the bottom, ‘Submitted by Your official so-and-so from such-and-such office.’” See, Du duan, 1.79 (mod. from Giele’s trans., as above, 135).
33. The Du duan states that memorials and petitions require a “head,” but that the presentations do not require a head. “Require a head” refers to the ministers leave a few blank bamboo strips when presenting their memorials and petitions to the emperor so that he can attach his words of approval. See Dai Guoxi, “Handai zhangzou wenshu ‘Xutou’ yu ‘Yanxing’ wenti kaolun” 漢代章奏文書”需頭”與”言姓”問題考論, Lanzhou xuekan 2017.8, 31–41.
34. The beginning of the Juyan 居延 Han bamboo text of “Yuankang wunian zhaoshu ce” 元康五年(61 b.c.e.)詔書冊 reorganized by Ōba Osamu says: “The Imperial Counselor Ji 吉 dares speak on risk of death: the Chancellor Xiang 相 submits the Superintendent of Ceremonial’s writings: … …” This clearly reveals the order in which Han dynasty memorials were submitted. See Ōba Osamu, Kankan kenkyū 漢簡研究 (Kyōto: Dōhōsha, 1992), 19.
35. Qian, Sima 司馬遷, Shi ji 史記 (Beijing: Zhonghua, 1982), 101.2741Google Scholar. This translation is Watson’s, modified for precision and to maintain consistency. See, Burton Watson, Records of the Grand Historian of China: Han Dynasty, Vol. 1 (Hong Kong: Columbia University Press, 1993), 523.
36. The conference reports quoted below have the Han chancellors directly submitting the yi file for the imperial inspection; naturally the chancellors would have put their names at the beginning, to show that they were responsible for the compilation.
37. Han shu, 8.272.
38. Here the Zhaomu system referred to the correct sequence in which the imperial ancestral temples should be placed: they should be placed alternately to the west and the east sides of the original founder of the line until a total of seven had been built. For more, see Loewe, Michael, “Imperial tombs” in China’s Early Empire, ed. Nylan, Michael and Loewe, Michael (Cambridge: Cambridge University Press, 2010), 215Google Scholar.
39. Han shu, 73.3118–19. The Zhonghua edition of the Han shu punctuates this citation such that it ends Wei’s direct quote with the main position and then reports the other dissenting views in the narrator’s voice. This is a misunderstanding due to a confusion regarding the textual format of court conferences during the Han.
40. Han shu, 25B.1253–55.
41. Shi ji, 112.2950 (mod. from Watson’s trans., as above, 188).
42. See Qin, Lüling shidai de “Yishi yi zhi”, 135–36.
43. Han shu, 3.96.
44. Han shu, 25b.1251.
45. Hou Han shu, Zhi 9.3193.
46. Hou Han shu, Zhi 9.3193.
47. Hou Han shu, Zhi 9.3193–94.
48. Han shu, 44.2141.
49. Bielenstein, Hans, The Restoration of the Han Dynasty, Vol. 4 (The Government) (Stockholm: Elanders Boktryckeri Ab. Kungsbacka, 1979), 59Google Scholar.
50. Qin, Lüling shidai de “Yishi yi zhi”, 161.
51. An example of a conference ending without a clear conclusion can be seen in Hou Han shu treatise on rites and music (Liyue zhi 禮樂志) where Emperor Wen finally ended Jia Yi’s 賈誼 (201–169) “discussion,” (yi) without reaching an obvious conclusion because of Zhou Bo’s 周勃 (d. 169 b.c.e.) and Guan Ying’s 灌嬰 (d. 176 b.c.e.) opposition. An example of dismissing the participants of a conference can be seen in the Hou Han shu biography of Dong Zhuo 董卓 (d. 192 c.e.) where the court was discussing whether or not they should abandon Emperor Shao 少帝 (r. 189 c.e.) and choose King Chenliu 陳留王 (181–234) to replace him as the emperor, because Dong Zhuo was in charge of the reigns at the time, he interrupted the conference by storming out in great anger. An example of reconvening a conference can be seen in the Han shu biography of Wei Xuancheng where the removal of the ancestral temples of commanderies and kingdoms was discussed. See, Han shu, 22.1030; Hou Han shu, 72.2324; Han shu, 73.3117–20.
52. No. 8, in the “Tianzi suo fu” 天子所服第八.
53. Han shu, 74.3139–40 (trans. modified from Watson’s trans.), as above, 184.
54. Han shu, 74.3141.
55. The debating positions could be compiled into personal collections and circulate apart from the administrative documents, like the “Da zhai yi” in the Cai zhonglang ji.
56. Hou Han shu, 3.138.
57. The White Tiger Hall was constructed at the White Tiger Gate (白虎門) in the North Palace (北宮) in Luoyang. See, Hou Han shu, 37.1264–65.
58. The Hou Han shu biography of Ying Shao 應劭 includes an obvious example of xiayi. In 185, the Hanyang 漢陽 gangsters led by Bian Zhang 邊章 and Han Sui 韓遂 (d. 215 c.e.) rebelled against the court, in alliance with some “barbarians” described as Qianghu 羌胡, … Captain of the Center, Northern Army (北軍中候) Zou Jing 鄒靖 requested that he be allowed to supply the army with Xianbei 鮮卑 soldiers. This issue was then sent down (i.e., referred to) for discussion to the four offices (四府) of the Supreme Commander (太尉), the Chancellor (司徒), the Minister of Works (司空), and the General-in-chief (大將軍) Han Zhuo 韓卓. Ying Shao violently opposed Han Zhuo’s position and no final conclusion was reached. Therefore, the emperor convened his high-ranking officials to discuss it at court, and all the officials were in agreement with Ying Shao’s position. See, Hou Han shu, 48.1609–10. For more on xiayi, see Lüling shidai de “Yishi yi zhi”, 149–54.
59. As a model from the previous dynasty, the Stone Canal Pavilion conference shares the same details with the White Tiger Hall conference. The Han shu “Rulin zhuan” says that Liangqiu Lin 梁丘臨 “during the Ganlu reign period was ordered to ask about the classicists of the Stone Canal Pavilion conference,” and that “the messengers also included the Guliang 穀梁 master, the Gentlemen of the Palace 中郎 Wang Hai 王亥.” Yan Shigu’s annotation says, “The term ‘Messengers’ refers to those officers whom the edicts ordered to supervise the discussion at that time.” This makes it clear that Emperor Xuan issued edicts to ask his classicists through such messengers. And the “Annals of Xuandi” says, “The Grand Tutor of the Heir Apparent Xiao Wangzhi and others evaluated and memorialized the discussions.” Therefore, in the Stone Canal Pavilion conference Xiao Wangzhi was charge with compiling the various debating positions elicited during the discussions. See, Han shu, 88.3600, 88.3618–19, 8.272.
60. Chen Guoqing 陳國慶, Hanshu Yiwenzhi zhushi huibian 漢書藝文志注釋彙編 (Beijing: Zhonghua, 1983), 51.
61. Sui shu, 32.923.
62. Du You 杜佑, Tongdian (Beijing: Zhonghua, 1988), 89.2455.
63. Han shu, 66.2903: 漢桓寬 … 著《鹽鐵論》六十篇.
64. Xu, Liu 劉煦, Jiu Tang shu 舊唐書 (Beijing: Zhonghua, 1975), 102.3180Google Scholar.
65. In the “Sheng zheng lun,” Zhang Rong wrote a decision on the emperor’s behalf. And Huan Kuan compiled and edited the Yantie lun, which does not contain any sign of approval by the emperor. How did Huan Kuan realize his goal of establishing a model for his own thought? Xu Hanchang 徐漢昌 thinks that the biggest possibility is that Huan Kuan had already directly inserted his own ideas into the conversation, so there is no editor’s position which separately stated his own views. And those short descriptive passages clearly contain the biting criticism of the court officials which is more aligned with the talented worthies 賢良 and literary scholars 文學. Huan Kuan’s “judgment” can be seen in this. See the related discussion in Xu Hanchang, Yantie lun yanjiu 鹽鐵論研究 (Taipei: Wenshizhe, 1983), pian 1, zhang 1, jie 2, 5–9.
66. According to the Hou Han shu, at least twelve classists participated the White Tiger Hall conference, they were Wei Ying, Lou Wang 樓望, Li Yu 李育, Chunyu Gong, Huan Yu 桓郁, Yang Zhong, Liu Xian 劉羨, Lu Gong 魯恭, Jia Kui 賈逵, Ban Gu, Ding Hong 丁鴻, and Cheng Feng 成封. Generally these court conferences convened around 50–100 people, so this list may simply represent the most famous participants.
67. Han shu, 100A.4225.
68. Hou Han shu, 40B.1373.
69. Hou Han shu, 48.1599.
70. Fang, Li 李昉, et al. , Taiping yulan (Beijing: Zhonghua, 1960), 233.1106Google Scholar. Nylan, Michael, Yang Xiong and the Pleasures of Reading and Classical Learning in Han China (New Haven: The American Oriental Society, 2011)Google Scholar, discusses these events.
71. Hou Han shu, 37.1264, 48.1597, 48.1598.
72. See the biography of Ban Gu (Hou Han shu, 40b.1374): “During the time when Su Zong, who liked literature of an elegant style, held the throne, Ban Gu was particularly favored. Often Ban entered the palace to read manuscripts, sometimes even from sunrise to sunset. Every time the Emperor went out on tour, Ban Gu accompanied him, and on the spot he would offer praise-poems in the fu 賦 and song 頌 forms. Once, when the court held an important conference, the Emperor let Ban question his senior ministers, and argue with them in the imperial presence. The Emperor rewarded generously, giving him his patronage.”
73. Examples have been given in the introduction of this article.
74. Zhang Guangbao 張廣保 points out that in the Baihu tong, the power of the emperor is restricted in three ways: virtue, heaven, and institutions. The Baihu tong opens by claiming that the Son of Heaven is a name of honor revealing that it intends to restrict the authority of the Son of Heaven at the very start. See Zhang Guangbao, “Baihu tongyi zhiduhua jingxue de zhuti sixiang” 白虎通義制度化經學的主體思想, in Zhongguo jingxue sixiangshi 中國經學思想史 Vol. 2, ed. Jiang Guanghui 姜廣輝 (Beijing: Zhongguo shehui kexue, 2003), 387–402.
75. In discussing the succession of the Son of Heaven, the first chapter title, Jue 爵, in today’s Baihu tong, requires the new ruler to strictly obey the three-year mourning rite, ceding all powers to the prime minister (Zhongzai 冢宰), who is to be the man in charge of policy making 攝政. Were this really to be put into practice, the dynasty might collapse. Therefore, it is hard to imagine that any emperor, who is not an infant or an idiot, would accept this practice.
76. Hou Han shu, 35.1202–03.
77. Tongdian, 77.2105.
78. Qin Tao’s research has also confirmed this. See, Qin, Lüling shidai de “Yishi yi zhi”, 162.
79. See Tjan, Po Hu T’ung: The Comprehensive Discussions, Vol. 1, 164.