I gratefully acknowledge the assistance of Robin Yates, Sarah Allan, Harold Roth, John Major, Michael Puett, Barry Blakely, Constance Cook and David Troyansky, all of whose advice and critical feedback resulted in substantial improvements to this essay. Thanks also to the anonymous readers whose criticism likewise was of invaluable help.

1. “Intellectual history” is both the object and method of this study. As I understand it, “intellectual history” is a mode of historical inquiry that places the production and use of texts at the center of investigation, with a simultaneous dual focus upon interpreting texts both for various forms of “meaning” and as evidence of shifting social contexts.

2. I will refer throughout this essay to “intellectuals” in ancient China, in full awareness of the potential problems and ambiguities such a category arouses. Pressed for a definition, I would offer that an “intellectual” was and is a person who both produces and uses texts. This obviously opens up the possibility that anyone and everyone could be/is an intellectual (even an illiterate person is capable of producing and using oral, pictorial, or performative “texts”). By way of rescuing the category thus derived from intractable vagueness, I would assert that the status of “intellectual,” like almost any generic label that might be applied to individuals in a human society, must be understood as being open to variations of degree. Thus the more time and effort one expends and/or the more one's perceived social role is dependent upon the production and use of texts, the more one inhabits the role of “intellectual.” This relativity informs almost any deployment of the term, thus in juxtaposing (for example) “rulers” and “intellectuals” I do not mean to imply that rulers could not be “intellectuals,” only that (in the instance at hand) they inhabited that role less thoroughly than the figures against whom they are thus counterpoised. The ancient Chinese themselves did not have a term that corresponds exactly to “intellectuals,” but they did develop phrases to differentiate those engaged in intellectual activity, for example “those who study” (xuezhe 學者)or “knights who have a Way” (youdao zhi shi 有道之士).

3. For the formation of a “Masters literature,” see: Denecke, Weibke, The Dynamics of Masters Literature: Early Chinese Thought from Confucius to Han Feizi (Cambridge: Harvard University Asia Center, 2011)Google Scholar. In focusing on Masters texts, I do not mean to imply that (what Donald Harper terms “fang 方-literature”: Harper, Donald, Early Chinese Medical Literature[London: Kegan Paul International, 1998], 8)Google Scholar was irrelevant to the intellectual history of the Warring States or that its authors were not bona fide “intellectuals.” I am confident, given the extent to which we have seen fang-literature and Masters literature that figures who enjoyed the patronage of the Tian kings both produced and used fang texts. The social and cultural overlap between those activities and the production of Masters texts is an important and complex question that requires further investigation.

I feel justified, however, in narrowing the focus of this study to the field of Masters texts, in that, as I will argue below, they were a forum in which certain specific issues of intellectual participation in public life were being intensively debated. The distinction I draw is thus not between texts produced in two completely different social contexts, but between different media produced in overlapping social contexts.

4. For an account of the Tian rise to supremacy in Qi, see: Gesen, Wang 王閣森 and Zhiqing, Tang 唐致卿, Qi guo shi 齊國史 (Ji'nan: Shandong Renmin, 1992), 360–63Google Scholar.For the larger trend of which the Tian usurpation was a part, see: Kuan, Yang 楊寬, Zhan guo shi 戰國史 (Shanghai: Shanghai Renmin, 1980), 139–51Google Scholar.

5. Hsu, Cho-yun, Ancient China in Transition (Stanford: Stanford University Press, 1965), 31–34Google Scholar.

6. Hsu, , Ancient China in Transition, 80–84Google Scholar.

7. Tales of the hubris of Earl Yao of Zhi are found throughout the texts of the Warring States and early Han. See, for example: Zhanguo ce, “Zhao yi” 趙一 (Zhanguo ce zhuzi suoyin 戰國策逐字索引, Institute for Chinese Studies Concordance [Hong Kong: Commercial Press, 1992], 203/103/22-105/14)Google Scholar; Lüshi chunqiu 呂氏春秋 “Dang ran” 當染 (Lüshi chunqiu zhuzisuoyin 呂氏春秋逐字索引, Institute for Chinese Studies Concordance [Hong Kong: Commercial Press, 1994], 2.4/10/2–4)Google Scholar. Unless otherwise noted, all references to Warring States texts in this essay will be to ICS Concordance Series editions. Citations will be in the form: chapter/page/line(s). If the ICS edition is broken into juan 卷 rather than pian 篇, the pian number will be provided in parentheses: juan (pian)/page/line.

8. For the complex of social and cultural institutions that established the identity of state 國 and clan, see: Chang, K.C. 張光直 Art Myth and Ritual: The Path to Political Authority in Ancient China (Cambridge: Harvard University Press, 1983), 9–32Google Scholar.

9. Barry Blakeley, in a personal communication, concurs with this assessment of the Tian usurpation for the early period. The only other instance I have found in which the same dynastic title was held by two different bloodlines transpired in the Five Dynasties Period (907-79 C.E.). During that era's Zhou Dynasty (951–60 C.E.), the founding emperor Guo Wei 郭威 (904–54 C.E.) was succeeded by his nephew by marriage Chai Rong 柴榮 (921–59 C.E.). : Davis, Richard L., Historical Records of the Five Dynasties (New York: Columbia University Press, 2004), 107–8Google Scholar.

10. Qian, Sima, Shi ji 史言己(Beijing: Zhonghua, 1959), 65.2160Google ScholarPubMed.

11. A text entitled the Sima fa still exists, but its relationship to the text compiled by the order of King Wei (if any) is unclear. See: Sawyer, Ralph, The Seven Military Classics of Ancient China (Boulder: Westview Press, 1993), 111–16Google Scholar. Rangju himself was a member of the Tian clan and thus an ancestor of the Tian kings. His biography in the Shi ji evinces many of the same themes and concepts expressive of the interests of the Tian kings analyzed below as encoded in the Guanzi and Yanzi chunqiu, thus I strongly suspect that Sima Qian was drawing from the very Tian patronage text he mentions in composing his biography of Tian Rangju. If that material was ever part of the transmitted Sima fa it is long lost, however.

12. See, for example: Moruo, Guo 郭沫若, “Jixia Huang-Lao xuepai de pipan” 稷下黃老學派的批判, Shi pipan shu 十批判書 (Beijing: Dongfang chubanshe, 1996), 158–59Google Scholar; Rickett, Allyn, Guanzi: Political, Economic, and Philosophical Essays from Early China, a Study and Translation, Volume I (Princeton: Princeton University Press, 1985), 14–24Google Scholar; Li'e, Lin, Xian Qin Qixue kao 先秦齊學考 (Taibei: Commercial Press, 1992), 252–56Google Scholar.

13. See, for example: Zeyu, Wu 吳則虞 Yanzi chunqiu jishi 晏子春秋集釋 (Beijing: Zhong hua shuju, 1962), 23–30Google Scholar; Yuqian, 駢宇騫, Yinque shan zhujian “Yanzi chunqiu” jiao shi 銀雀山竹簡 “晏子春秋” 校釋 (Taibei: Wanjuan lou, 2000), 9–15, 229–41, 257–68Google Scholar; Li'e, Lin, Xian Qin Qixue kao, 248–52Google Scholar.

14. See, for example, Mencius 3.1. When his student Gongsun Chou asks if Mencius will be able to replicate the meritorious service of Zhong, Guan and Yanzi, Mencius replies, “You truly are a person of Qi. You know Guan Zhong and Yanzi, that's all” 子誠齊人也, 知管仲, 晏子而已矣 (Mencius, “Gongsun Chou, shang” 公孫丑上 [Mengzi zhuzi suoyin 孟子逐字索引, Institute for Chinese Studies Concordance (Hong Kong: Commercial Press, 1995), 3.1/14/8])Google Scholar.

15. The genre of military writings (兵書) was obviously closely associated with the Qi court. This is evinced by the lore surrounding Marshal Rangju, and the fact that the putative authors of two of the most famous texts in this genre, Sun Wu and Sun Bin, were said to be natives of Qi. This inference is further corroborated by the discovery of a cache of buried writings in a Han dynasty tomb at Yinqueshan 銀雀山 in Linyi 臨沂, Shandong 山東. Among the writings found in that tomb were a partial text of the Yanzi chunqiu and a text of the Sun Bin bingfa 孫臏兵法. Other writings, unattested to in imperial catalogues, were also found in the cache. All of the texts express themes and concepts quite consonant with those articulated in the received Guanzi and Yanzi chunqiu, thus it is quite possible that these are also examples of Qi patronage texts. See: Yates, Robin D. S., “Texts on the Military and Government from Yinqueshan: Introductions and Preliminary Transcriptions,” in Xinchu jianbo yanjiu 新出簡帛研究, eds. Allan, Sarah 艾藍 and Wen, Xing 邢文 (Beijing: Wenwu, 2004), 334–87Google Scholar. See also: Li'e, Lin, Xian Qin Qixue kao, 264–82Google Scholar.

16. On the dating of the Guanzi in whole and part, see: Genze, Luo 羅根澤, Guanzi tanyuan 管子探源 (Shanghai: Zhonghua shuju, 1931)Google Scholar; Haloun, Gustav, “Legalist Frag-ments: Part I: Kuan-tsi 55 and Related Texts,” Asia Major n.s. 2, pt. 1 (04 1951), 85–120Google Scholar; van der Loon, Piet, “On the Transmission of the Kuan-tzu,” T'oung Pao II 41 (1952), 357–93CrossRefGoogle Scholar. Rickett, , Guanzi, Volume I, 14–15Google Scholar. Osamu, Kanaya 金谷治, Kanshi no kenkyū 官字の研究 (Tokyo: Iwanami Shoten, 1987), 328–29Google Scholar.

17. Luo, , Guanzi tanyuan, 122–42Google Scholar.

18. Haloun, , “Legalist Fragments,” 96Google Scholar.

19. See: Maeder, W., “Some Observations on the Composition of the ‘Core Chapters’ of the Mozi,” Early China 17 (1992), 27–82CrossRefGoogle Scholar; Ling, Li 李零, Guodian Chujian jiaoduji 郭店楚簡校讀記 (Beijing: Beijing daxue, 2002), 190–93Google Scholar; Boltz, Willaim, “The Composite Nature of Early Chinese Texts,” in Text and Ritual in Early China, ed. Kern, Martin. (Seattle: University of Washington Press, 2005), 50–78Google Scholar.

20. The proposition that the proto-Guanzi corpus was sponsored by the Tian clan naturally raises the question of why such writings continued to be composed and collected in the Han, as the current text of the Guanzi evinces. This is not difficult to understand, however. As I will argue below, important constituencies within the Han imperial leadership took much inspiration from the novel notions of statehood conceived under Tian patronage. They would thus have had a natural interest in preserving and adding to the corpus of writings first begun in Qi. See, for example, citations of the Guanzi various debating parties in the Yantie lun 鹽鐵論: Liqi, Wang 王利器, Yantie lun jiaozhu 鹽鐵論校注 (Beijing: Zhonghua, 1992), 3, 43, 423, 567Google Scholar.

21. For example, they consistently portray Guan Zhong in positive terms, which would not be the case if the writings were a truly random collection.

22. Huan's ascendancy to the office of Lord Protector is recorded at Zuo zhuan 左傳, “Zhuang Gong” 莊公 (Chun qiu Zuo zhuan zhuzi suoyin 春秋左傳逐字索引, Institute for Chinese Studies Concordance [Hong Kong: Commercial Press, 1995], B3.15.1/50/30)Google Scholar.

23. For an excellent study of the early discourse surrounding Guan Zhong, and the office of Lord Protector, see: Rosen, Sidney, “Changing Conceptions of the Hegemon in Pre-Ch'in China,” in Ancient China: Studies in Early Civilization, ed. Roy, David T. (Hong Kong: Chinese University Press, 1978), 99–114Google Scholar.

24. Shi ji, 46.1880; Zuo zhuan, “Zhuang Gong”莊公 (B3.22.1/56/3–24).

25. Rickett, , Guanzi, Volume 1, p. 60Google Scholar.

26. Kanaya places this chapter within the second stratum of the Guanzi, which he dates to c. 285–235 B.C.E. (Kanaya, , Kanshi no kenkyū, 328–29)Google Scholar. Rickett concurs with this date for the first section of the chapter, but argues that a latter section was of different provenance and date (Rickett, , Guanzi, Volume 1, 284–85)Google Scholar. I would agree that “Da Kuang” appears to be later than preceding chapters of the Guanzi (such as Chapter 1, “Shepherding the People” 牧民) but how much later is an open question. I do not claim that “Da Kuang” represents the original or most “authoritative” section of the Guanzi corpus as a whole. Its abundance of narrative detail, however, make it useful for illustrating certain polemical agendas that help contextualize the seemingly earlier portions of the text, which though terser generally accord with the rhetorical positions of Chapter 18, as I will argue below.

27. The relationship between the Guanzi, Guo yu, and Zuo zhuan accounts of Guan Zhong's early career is a very complex question. The Zuo zhuan treats these events very differently than the Guanzi, excluding many details and rhetorical themes that the latter text accentuates (see Zuo zhuan, “Zhuang Gong” [B3.8.3/44/15-B3.9.5/45/30]). As I will argue below, this is unsurprising, as these two texts were produced by rival groups of intellectuals, each trying to construct these events to particular polemical ends. The Guo yu account is very closely related to the “Xiao kuang” chapter of the Guanzi. This would seem to contradict my last assertion, as the Guo yu and Zuo zhuan were treated by scholars in imperial times as mutually complementary texts. Many interpreters, however, have noted the rhetorical dissonance between the “Qi yu” sections of the Guo yu and the general rhetorical thrust of the Zuo zhuan as a whole. I would agree with David Schaberg that the “Qi yu” sections of the Guo yu were the latest materials compiled into that anthology (Schaberg, David, A Patterned Past: Form and Thought in Early Chinese Historiography [Cambridge: Harvard University Asia Center, 2001], 329n9)CrossRefGoogle Scholar, after the tensions between the different intellectual centers that had produced the Guanzi and Zuo zhuan had subsided.

28. Guanzi, “Da kuang” 大匡 (Guanzi zhuzi suoyin 管子逐字索引, Institute for Chinese Studies Concordance [Hong Kong: Commercial Press, 2001], 7.1(18)/51/16–27)Google Scholar. Translated in Rickett, , Guanzi, Volume 1, 284–85Google Scholar.

29. Guanzi, “Da kuang” (7.1[18]/51/27–29). See translation in Rickett, , Guanzi, Volume 1, 285–86Google Scholar.

30. See: Chang, , Art, Myth, and Ritual, 9–32Google Scholar.

31. The only other place in Warring States literature where we see the state treated in similarly abstract and “transferable” terms is in the complex of legends surrounding Yao 堯 and Shun 舜. In that instance, however, though the “world” was genuinely transferred from one ruler to another, the dynastic title of the ruling house was not. Yao ruled under the state title of Tang 唐, Shun under that of Yu 虞.

32. Guanzi, “Qi fa” 七法 (2.1[6]/14/32–15/3). See translation in Rickett, Guanzi, Volume 1, 131Google Scholar.

33. Guanzi, “Fa fa” 法法 (6.1[16]/47/18, 6.1/49/13). Rickett, , Guanzi, Volume 1, 260, 261, 266Google Scholar.

34. “Shepherding the People” is the most oft-cited chapter of the Guanzi in other Warring States texts. It was thus among the texts most broadly associated with Guang Zhong from early on. See Rickett, , Guanzi, Volume 1, 51–52Google Scholar.

35. Guanzi, “Mu min” 牧民 (1.1[1]/1/7–8). See alternate translation in Rickett, , Guanzi, Volume 1, 52Google Scholar.

36. Compare to the “Hymns” collected in the Shi jing: Waley, Arthur, trans., The Book of Songs, ed. Allen, Joseph R. (New York: Grove Press, 1996), 291–323Google Scholar.

37. Guanzi, “Guo song” (1.1[1]/1/9–12).

38. Guanzi, “Da kuang” (7.1[18]/52/24, 53/29).

39. Guanzi, “Xiao kuang” 小匡 (8.2]20//59/5–11).

40. For clear articulations of this principle, see Lüshi chunqiu, “Zheng ming” 正名 (16.8/98/11–13); “Li su” 離俗 (19.1/119/24–28).

41. Rosen, Sidney, “In Search of the Historical Kuan Chung,” The Journal of Asian Studies 35.3 (05 1976), 437CrossRefGoogle Scholar.

42. Lewis, Mark Edward, Writing and Authority in Early China (Albany: State University of New York Press, 1999), 55, 58Google Scholar.

43. See, for example: Queen, , “Inventories of the Past: Rethinking the ‘School’ Affiliation of the Huainanzi,” Asia Major 3.14.1 (2001), 51–72Google Scholar; Csikszentmihalyi, Mark and Nylan, Michael, “Constructing Lineages and Inventing Traditions through Exemplary Figures in Early China,” T'oung Pao 89, no. 1–3 (2003), 59–99CrossRefGoogle Scholar.

44. Eno, Robert, The Confucian Creation of Heaven (Albany: State University of New York, 1990), 54Google Scholar.

45. Lewis, , Writing and Authority, 62Google Scholar. Lewis describes the development of an “enunciatory practice” built around the “teaching scene” in the Analects on page 57.

46. On this score it must be noted that Lewis' chronology is very idiosyncratic—most of the scholars quoted above would not agree with his placing of the diaologues of the Guanzi and the Yanzi chunqiu exclusively in the 3rd century B.C.E.

47. Brooks, E. Bruce and Brooks, A. Taeko, The Original Analects: Sayings of Confucius and His Successors (New York: Columbia University Press, 1998), 227–32Google Scholar. Brooks and Brooks provide charts of perceived parallels/oppositions between passages in the earliest strata of the Guanzi and individual sections of the Analects that, though they take some effort to hunt down, are often quite striking and persuasive. I would note, though, that though I agree that the Analects is an “accretional” text, I do not concur with their long chronology. I would surmise that the text had stabilized into varying forms resembling the current transmitted text by the mid-fourth century B.C.E. at the latest. The parallels the Brookses point out thus open the possibility that the earliest strata of the Guanzi predate the establishment of Jixia.

48. Lun yu, “Ji shi” 季氏 (Lun yu zhuzi suoyin 論語逐字索引, Institute for Chinese Studies Concordance [Hong Kong: Commercial Press, 1992],16.1/45/7–15)Google Scholar.

49. Rickett, , Guanzi, Volume 1, 11–12, 51–52Google Scholar.

50. Lewis, , Writing and Authority, 57Google Scholar.

51. Though different models of the “accretional” theory of the Analects abound, most interpreters would agree that the process by which the text formed began in the third generation of transmission at the latest (i.e., among the disciples of Confucius' disciples). Compare Brooks and Brooks, The Original Analects, 13–38Google Scholar; Lau, D.C., trans., The Analects (London: Penguin, 1979), 227–33Google Scholar; Zoeren, Steven Van, Poetry and Personality: Reading, Exegesis, and Hermeneutics in Traditional China (Stanford: Stanford University Press, 1991), 25–28, 259n13Google Scholar.

52. Lewis, , Writing and Authority, 57Google Scholar.

53. In the following pages I will deal exclusively with the Yanzi chunqiu's discussion of Confucius, as this is most relevant to the question of the Qi model of intellectual participation. The text's treatment of Tian Wuyu, Tian clan patriarch during the career of Yan Ying, provides further evidence of its being a patronage text of the Tian rul- ers, however. Yan Ying delivers several “predictions” of the Tian ascendance to the Qi throne. See: Yanzi chunqiu, “Nei pian wen, shang” 內篇問上 (Yanzi chunqiu zhuzi suoyin 晏子春秋逐字索引, Institute for Chinese Studies Concordance [Hong Kong: Commercial Press, 1994], 3.8/24/19–20)Google Scholar, “pian wen, xia 內篇問下”(4.17/35/20–36/1), “Wai pian chong er yi zhe 外篇重而異者” (7.1/64/1–11, 7.15/66/6–21), all of which would have served to legitimize Tian claims of rulership when most or all of the Yanzi chunqiu was likely authored.

Much of this material appears in parallel form in the Zuo zhuan, “Zhao Gong” 昭公 (B10.3.3/323/23–29; B10.26.11/393/15–394/1). As the Zuo was a text associated with the Zou-Lu fellowship, the parallelism between it and the Yanzi chunqiu raises questions about the role of Tian patronage comparable to those raised by the sharing of material between the Guo yu and the Guanzi. If there was hostility between the Qi and Lu groups, why would texts associated with either locale mirror one another? One answer, as I will argue below, is that tensions between these groups eased over time. But the picture the evidence presents is somewhat more complicated. Though there was genuine hostility between Qi and Lu intellectual centers it was clearly never absolute, and the boundaries between these two groups was obviously always quite porous.

In his excellent study of the Zuo zhuan and Guo yu, David Schaberg skillfully demonstrates that these texts are “traces of a historiographical practice that flourished, before, during, and perhaps after the fourth century B.C.E., almost certainly among thinkers who considered themselves followers of Confucius” (A Patterned Past, 8). I completely agree with Schaberg's identification of this “practice of historical discourse” (APatterned Past, 10). The only emendation I would offer to his picture, on the evidence of the Guanzi and the Yanzi chunqiu, is that the discursive community in which this practice evolved and was maintained must be understood to have extended across the fluid boundary between Qi and Lu groups (and perhaps beyond). In other words, not all of the “historiographers” that developed the discourse Schaberg delineates “considered themselves followers of Confucius.” Though most of the material that made its way into the Zuo zhuan and Guo yu originated with self-identified “followers of Confucius,” some of it was generated by intellectuals who shared the “Confucians'” avocation of “historiographer” (i.e., they were interested in the past and employed a common set of conventional techniques in its representation and interpretation) but not their veneration of Confucius himself.

Why would politically opposed groups of historiographers, despite their shared avocation, adopt common accounts of past events? They would do so if each group perceived the account to yield meaning from the past that served their own interpretive goals, even if that meaning was different in either case. Thus a “prediction” about the rise of the Tian clan could simultaneously serve the goals of Qi historiographers to legitimize the political claims of their patrons and the goals of Lu historiographers to “extend their control over the material of history” and “accomplish a figurative subordination of one sort of power—the ruling lines' reproduction of themselves over generations—to the very different intellectual power of vision (Schaberg, , A Patterned Past, 282, 283)Google Scholar.”

54. For example: Ruigeng, Chen 陳瑞庚, Yanzi chunqiu kaobian 晏子春秋考辨 (Taibei: Changan chubanshe, 1980)Google Scholar.

55. Yuqian, Pian 駢宇騫, Yinque shan zhujian “Yanzi chunqiu”jiaoshi 銀雀山竹簡 “晏 子春秋” 校釋 (Taibei: Wanjuan lou, 2000)Google Scholar.

56. Walker, Richard, “Some Notes on the Yen tzu ch'un ch'iu.” Journal of the American Oriental Society 73 (1953), 156–63CrossRefGoogle Scholar. Zheng Liangshu proposes a more diffuse provenance for the Yanzi chunqiu, though he does assert that the extant text developed from a “core” of materials recording Yan Ying's words and deeds that had been produced in Qi. This core was then elaborated upon and emended by disparate later groups, but the text as we currently have it was largely formed, according to Zheng, by the late Warring States. See: Liangshu, Zheng 鄭良樹, “Lun Yanzi chunqiu de bianxie ji chengshu guocheng (xia)” 論〈晏子春秋〉 的編寫及成書過程 (下), Guanzi xuekan 管子學刊 2000.2, 38Google Scholar.

57. The Yanzi chunqiu was a paradigmatic example of what William Boltz describes as “the composite nature of early Chinese texts” (in Kern, , ed., Text and Ritual in Early China, 50–78)Google Scholar, except that it was not produced, reformed, and transmitted within the “master-text-disciple” framework described by Mark Edward Lewis, but within the context of Qi state patronage.

58. Gengsheng, Wang 王更生, Yanzi chunqiu jinzhu jinyi 晏子今註今譯 (Taibei: Commercial Press, 1987), XGoogle Scholar.

59 . Zongyuan, Liu, Liu Hedong ji 柳河東集 (Shanghai: Commercial Press, 1958), 56–57Google Scholar.

60. Yanzi chunqiu, “Wai pian bu he jing shu zhe” 外篇不合經術者 (8.3/71/7–10).

61. Yanzi chunqiu, “Nei pian za, shang” 內篇雜上 (5.21/46/21–27, accepting the proposed emendation at note 4.)

62. This is the import of the line: “In great matters one should not transgress, in lesser matters there may be give and take.” Confucius' disciples fail to understand that in ritual the dignity of the sovereign is paramount, and that all the particular minutiae of ritual may be bent to preserve it.

63. Yanzi chunqiu, “Wai pian bu he jing shu zhe”外篇不合經術者, (8.4/71/22–23).

64. Yanzi chunqiu, “Nei pian za, xia” 內篇雜下 (6.13/54/15–24).

65. Shi ji, 46.1885.

66. The stridently polemical nature of the anecdote is reflected in the Suo yin commentary, which expresses disbelief that even so blameworthy a figure as Tian Chang could behave so bestially (Shi ji, 46.1885n1).

67. A full description of the vessel may be found in Doty, Paul Darrel, “The Bronze Inscriptions of Ch'i: An Interpretation.” (Ph.D. Dissertation, University of Washington, 1982), 616–17Google Scholar. The tureen and its inscription have been the object of extensive philological and interpretive study. For a review of this scholarship, see Doty, , “The Bronze Inscriptions of Ch'i,” 617–37Google Scholar. See also: Puett, Michael, “Sages, Ministers and Rebels: Narratives from Early China Concerning the Initial Creation of the State,” Harvard Journal of Asiatic Studies 58.2 (12 1998), 450n64CrossRefGoogle Scholar.

68. Wei's given name is recorded in the Shi ji as Yinqi 因齊. The second character in the name “Yinzi” appears on the bronze vessel as a 次 atop a 月. Modern scholars generally agree that this graph was a phonological equivalent of the 齊 given by Sima Qian as the second character in King Wei's given name during the Warring States (See Doty, , “The Bronze Inscriptions of Ch'i,” 618n2)Google Scholar, thus we may be confident that the dedicator of this vessel was King Wei.

69. Yachu, Zhang 張亞初 et al., eds., Yin Zhou jinwen jicheng yinde 殷周金文集成引得 (Beijing: Zhonghua shuju, 2000), 99 (9.4649)Google Scholar. I gratefully acknowledge the assistance of Constance Cook in translating and interpreting this inscription, though any errors are exclusively my own. I have followed the proposed conversions into modern logographic equivalents given by the editors of the Yin Zhou jinwen jicheng yinde in transcribing the inscription for this essay.

70. Lewis, Mark Edward, Sanctioned Violence in Early China (Albany: SUNY Press, 1990), 308n60Google Scholar.

71. Puett, , “Sages, Ministers and Rebels,” 450Google Scholar.

72. Duke Huan's claim to being the consolidator of Tian sovereignty over Qi is not unimpeachable, but the historical record leaves little doubt that he contrived to be perceived as such. The Shi ji (46.1886) records that the title “Marquis of Qi” had in fact been granted by the Zhou king to Tian He 田和, great-grandson of Tian Chang, in 386 B.C.E. He died two years thereafter, however, and Duke Kang 康公 (acc. 404 B.C.E.), the last Lü duke of Qi, did not die until 379 B.C.E. The Shi ji continues to record years subsequent to 386 B.C.E. as falling within the reign of Duke Kang, thus it is not clear that a universally acknowledged transfer of sovereignty from Lü to Tian took place under Tian He.

Though Duke Kang was exiled to a reduced fief on the seashore, he continued to carry on the ancestral cult of his clan and thus presumably retained the state title of Qi. Upon Tian He's death, leadership of the state passed to Tian Yan 田剡 (r. 383–375 B.C.E.), who is recorded in the Bamboo Annals as “Marquis Tian” 田候. The Annals (an official chronicle of the state of Wei, and thus a reliable witness of Warring States political perceptions) records the year of Tian Yan's accession as “the 22nd year of Duke Kang (Shi ji, 46.1887n2),” thus increasing the likelihood that Tian claims to titular sovereignty over Qi had not yet been firmly established. It was during Tian Yan's reign that Duke Kang finally died, evidently without heir. Whether Tian Yan made any formal claims to the Qi title at that point can not be known, however, because Duke Huan overthrew Tian Yan in 375, killing both him and his young son Tian Xi 喜.

To all appearances, Duke Huan waged a successful campaign to expunge Tian Yan's reign from the annals of Qi, (as evinced by Yan's absence from the chronology of the Shi ji) and to present himself as the rightful heir of Tian He, known posthumously as the “Grand Duke” (as evinced by the Shi ji's record that Duke Huan succeeded Tian He [46.1887], when in fact he did not come to the throne until ten years after Tian He's death). It is for Duke Huan's reign that both the Shi ji and the Bamboo Annals begin to chronicle events in Qi using a Tian reign calendar. Thus, whatever moves might have been made in that direction by his predecessors, Duke Huan (motivated in part, no doubt, by the need to legitimize his own usurpation of the throne from Tian Yan) succeeded either in firmly establishing the exclusive Tian claim to the title of Qi or altering the historical record to reflect his having done so.

73. Puett, , “Sages, Ministers, and Rebels,” 450Google Scholar.

74. If Yinzi had already assumed the title “King” at the point this vessel was cast, the choice to address his vassals as “Lords of the Land” would be a natural one: a rising tide lifts all boats, and Yinzi's assumption of the title “King” would necessitate the elevation of his immediate subordinates to “Lords of the Land.” Even if this were not the case, however, there is evidence that Qi rulers customarily addressed at least some of their vassals as hou 侯. The Guanzi and other texts report the existence of “Lords within the Passes” 關內之侯, i.e., local rulers who retained the subregnal rank of “Marquis” granted by the Zhou kings but who were vassals of Qi (see Rickett, , Guanzi, Volume 1, 297n89)Google Scholar.

75. On this score Yinzi is carrying out “the great plan” of his father in a very literal sense. Existing bronze vessels cast by Duke Huan for use in sacrificing to his “mother,” the Filial Great Consort 孝大妃, wife of Tian He (see Yachu, Zhanget al., Yin Zhou jinwen jicheng yinde, 74 [8.4145], 99 [9.4646, 9.4647]Google Scholar; Doty, , “The Bronze Inscriptions of Ch'i,” 590–611)Google Scholar bear inscriptions manifesting the same formulaic pattern as the Chenhou Yinzi dun. The “Lords of the Land” are summoned to court and present bronze vessels for use in casting the inscribed implements. The symbolic import of these vessels is clear: any vassal supplying bronze for the casting of these vessels gave sanction to Duke Huan's claim to be the rightful heir to Tian He, as this was the unambiguous significance of Huan's assumption of the prerogative of sacrificing to Tian He's wife (or, presuming the woman in question actually was Duke Huan's mother, her posthumous elevation to the rank of “Grand Consort”). The rhetorical strategies deployed by King Wei in the Chenhou Yinzi dun are thus all elaborations upon forms of propaganda pioneered in the ritual program of his father, Duke Huan.

76. Here again Yinzi is following a model laid down in the bronze vessels cast by his father (see previous note).

77. The Yellow Thearch is credited in many texts with having invented the state as an institution (see: Puett, Michael, The Ambivalence of Creation: Debates Concerning Innovation and Artifice in Early China. [Stanford: Stanford University Press, 2001], 115–116, 133)Google Scholar, thus he is a natural model for anyone promoting the ultimate value of the state. The definitive merit of the Lords Protector was to “sustain the perishing and revive the extinct” 存亡繼絕 (see, for example: Guanzi, “Zhong kuang” 中匡[8.1(19)/58/29/, “Ba yan” 霸言 [9.2(23)/69/24], “Xiao wen” 小問 [16.3(51)/118/22]; Lüshi chunqiu, “Shen ying” 審應 [18/108/24]), i.e., to bolster states that were in danger of collapse and to revive states whose sacrifices had been cut off. Thus they likewise stood as fitting symbols of the sanctity of the state. Moreover, when we survey the way that Duke Huan's and Duke Wen's respective legends evolved in early literature, we see convergence on this point of “putting the state first,” even before personal honor. One famous story has Duke Wen accepting the service of Eunuch Pi 寺人披 who had twice been dispatched to assassinate him, once managing to cut the sleeve of the Duke's robe. In persuading Duke Wen to grant him an interview (which Pi seeks at the risk of his life so as to warn Duke Wen of a pending attack upon the palace by rebel vassals) Eunuch Pi invokes the example of Duke Huan's employment of Guan Zhong even after Guan had shot him in the belt-buckle. See: Zuo zhuan, “Xi Gong” 僖公 (B5.24.1/102/21–28); Feizi, Han 韓非子, “Nan san 難三” (Han Feizi zhuzi suoyin 韓非子逐字索引, Institute for Chinese Studies Concordance [Hong Kong: Commercial Press, 2000], 38/121/23–122/5)Google Scholar.

78. There is one other way in which the Chenhou Yinzi dun may provide evidence of policies reflective of the principles articulated in the Guanzi that, though somewhat speculative, bears mention. Ordinarily the phrase translated above as “bronze vessels” 吉金 would refer to vessels captured from defeated enemies in war; otherwise they would not be transferable from vassal to lord in the manner depicted in this inscription (see: Cook, Constance, “Wealth and the Western Zhou,” Bulletin of the School of Oriental and African Studies, University of London, 60.2 [1997], 266–67)CrossRefGoogle Scholar. In this case however, the vessels being donated may in fact have been the actual ancestral ritual vessels of the vassals in question. If this were so, the Chenhou Yinzi dun inscription marks a ceremony whereby the “Lords of the Land” who participated, by surrendering the necessary para-phernalia of their respective ancestral cults, were transformed from hereditary vassals of the Qi king into bureaucratic officials of the Qi state. Thus this ceremony served as a mechanism by which to realize the “mathematically” routinized administrative plan of the Guanzi. Though this would entail a departure from tradition, we have seen that such departures were commonplace in the Tian Qi domain.

79. For an excellent collection of primary source material on Jixia, see: Bingnan, Zhang 張秉楠, Jixia gouchen 稷下鉤沉 (Shanghai: Shanghai guji, 1991)Google Scholar.

80. The precise date of the establishment of Jixia is unknown. Xu Gan 徐幹 (170–217 C.E.) reports that it already existed in the reign of Duke Huan 桓公 (Tian Wu 田午, r. 374–357 B.C.E.), but this is the sole testimony of such a date and it is rather late (lun, Zhong 中論, “Wang guo” 亡國 [Han-Wei congshu 漢魏叢書 (Changchun: Jilin daxue, 1992)], 579b)Google Scholar. Qian Mu 錢穆 (1895–1990) provides a thorough review of the evidence relating to the chronology of Jixia. He lists figures associated with it from the reign of King Wei 威王 (356–320 B.C.E.) to the final destruction of Qi in 221 B.C.E. (Mu, Qian, Xian Qin zhuzi xinian 先秦諸子繫年 [Taibei: Dongda tushu gongsi, 1990], 227–35Google Scholar). Lin Li'e argues that the inception of Jixia should be placed in the reign of Huan, Duke (LinLi'e, Xian Qin Qixue kao Li'e, Xian Qin Qixue kao, 140)Google Scholar. Wang Gesen and Tang Zhiqing chart the history of Jixia into three periods: 1) A formative period during the reigns of Duke Huan and King Wei (374–320 B.C.E.); 2) A “golden age” in which its most famous denizens (Song Xing, Shen Dao, etc.) were resident, during the reigns of Kings Xuan and Min (319–284 B.C.E.); 3) A period of revival following the cataclysm that brought down King Min, during the reigns of King Xiang 襄王 and Tian Jian 田建 (278–221 B.C.E.). It was during this latter period that Xun Kuang 荀況 (Xunzi 荀子, ca. 310–ca. 217 B.C.E.) was the eldest scholar and thrice served as libationer during communal sacrifices (Gesen, Wang and Zhiqing, Tang, Qi guo shi, 509–12)Google Scholar. A detailed study of the history of Jixia may be found in Sato, Masayuki, The Confucian Quest for Order: The Origin and Formation of the Political Thought of Xunzi (Leiden: Brill, 2003), 72–84Google Scholar.

81. Qian Mu gives approximate dates of 350–275 B.C.E. for Pian, Tian, Dao, Shen, Yu, Jie (Xian Qin zhuzi xinian, 618)Google Scholar.

82. Shi ji, 46.1895. A collective biography of some of the figures named here appears in Shi ji 74, which reports that “Zou Yan to the [other] Venerables 先生 of Jixia in Qi, such as Chunyu Kun, Shen Dao, Huan Yuan, Jiezi, Tian Pian, and Zou Shi 鄒奭, each wrote prose discussing the affairs of order and chaos, to attract the rulers of the age. There are too many to discuss.” (Shi ji, 74.2346)

83. An extant fragment of Liu Xiang's Bie lu 別錄 is the oldest supplementary source and records that, “Qi had a Ji Gate, it was a gate of the [capital] city wall. The knights who discussed and persuaded gathered beneath the Ji [Gate].” (Shi ji, 46.1895n6) This fragment is preserved in both the Shi ji jijie 集解 and the Shi ji suoyin 索引. Nathan Sivin states that we can be confident, on Liu Xiang's testimony, that Jixia was divided between residences internal to the city walls and “lecture halls” external to the city walls (Sivin, Nathan, “The Myth of the Naturalists,” Medicine, Philosophy, and Religion in Ancient China: Researches and Reflections [Hampshire: Variorum, 1995], 26)Google Scholar, but he relies on a Bie lu quote embedded in a Taiping huanyu ji 太平寰宇記 passage cited in Qian Mu's autocommentary (Mu, Qian, Xian Qin zhuzi xinian, 232)Google Scholar. The Taiping huanyu is the only source in which this Liu Xiang passage has a range beyond what is cited above, thus it is not clear that anything after “a gate of the city wall” was in fact Liu Xiang's original testimony and not the words of the Taiping huanyu compilers. Our best and earliest testimony about Jixia thus places it within the city walls of Linzi, and informs us that it took its name from the Ji Gate. Li'e, Lin (Xian Qin Qixue kao, 232–39)Google Scholar gives a thorough overview of primary sources and secondary scholarship on the name and location of Jixia.

84. Qian Mu provides a table listing seventeen figures whose names are associated with Jixia in early sources. Of these he counts only fourteen as having been likely to have enjoyed the patronage of Jixia (Mu, Qian, Xian Qin zhuzi xinian, 233–35)Google Scholar. Zhang Bingnan compiles material for 19 figures associated with Jixia (Bingnan, Zhang, Jixia gouchen, 17–162)Google Scholar. Sun Yikai 孫以楷 expands that list to 29 names (Yikai, Sun, “Jixia renwu kaobian” 稷下人物考辨, Qi Lu xuekan 齊魯學刊 1983.2, 37–41)Google Scholar, on somewhat more liberal criteria. A comparative chart of all figures associated with Jixia by various sources and scholars is provided by Li’e, Lin (Xian Qin Qixue kao, 146–47)Google Scholar.

85. Sivin, , “The Myth of the Naturalists,” 19–24Google Scholar.

86. Sivin, , “The Myth of the Naturalists,” 28Google Scholar.

87. The chief exemplars are the retinues of the so-called “Four Lords” 四君 of the Warring States, each of which is given a separate biography in the Shi ji (grouped together as Chapters 75–78, 75.2351–78.2399).

88. Sivin, , “The Myth of the Naturalists,” 21Google Scholar.

89. Take, for example, the retinue of Huang Xie, Lord of Chunshen. We know of two prominent Masters that enjoyed his patronage: Yu Qing 虞卿 (c. 305–c. 235 B.C.E.) and Xunzi 荀子 (Xun Kuang 況, ca. 310–215 B.C.E.). But we also know of Tang Ju 唐且, who was a freelance persuader and diplomat, Han Ming 汗明, an impoverished knight who managed to join the ranks of Huang's retainers, Li Yuan 李園, whose main service to Huang was procuring his younger sister as a concubine, and Zhu Ying 朱英, whose main service was offering to assassinate Li Yuan (see Zhanguo ce, “Chu san 楚三” [188/96/17], “Chu si” 楚四 [199/100/27, 200/101/16, 200/102/4, 201/102/23]; Shi ji, 74.2344–45,48,78.2395–96). We can be fairly sure that intellectuals are overrepresented in this sample, as Huang Xie reportedly had three thousand retainers, and one thousand of them were not likely to have been Masters and/or their disciples. If the ratios between records of Chunshen's court and those of Jixia matched, we should know the names of 28 (2 × 14) non-intellectual retainers who lodged at the latter locale. This is perhaps too much detail to expect from the sources. But even so, it strains credibility to imagine that we would know of no non-intellectual Jixia clients if the social composition of that patronage community had in any way resembled that of the Lord of Chunshen's.

90. Moreover, just as intellectuals loom large in the records of Jixia, Jixia itself looms large in the historical testimony concerning the individuals identified as Jixia knights. For example, for those Jixia figures that authored (attributively or no) texts which survived into the Han 漢 (206 B.C.E.–220 C.E.), the Han shu yiwenzhi 漢書藝文志 in three instances records their affiliation with Jixia as a salient aspect of their biography. Thus we read that Tian Pian “traveled to Jixia,” Zou Yan “lived at Jixia,” and Xunzi “was the Libationer 祭酒 at Jixia.” Yan Shigu 顏師古 (581–645 C.E.), in commentary to the entry for Yin Wen 尹文 (c. 350–c. 285 B.C.E.) cites Liu Xiang 劉向 (77–6 B.C.E.) as recording that both Yin Wen and Song Xing 宋鈃 (c. 360–c. 290 B.C.E.) had “traveled to Jixia (Ban Gu 班固 [32–92 C.E.], Han shu 漢書 [Beijing: Zhonghua shuju, 1962], 30.1725, 30, 33, 37)Google Scholar.” In the Shi ji we read that Zou Yan enjoyed the patronage of many powerful figures, and we have seen above that Xunzi spent considerable time in the retinue of the Lord of Chunshen. If affiliation with Jixia were truly fungible with membership in any other guest-client retinue, why would it be considered particularly worthy of mention in the Han shu yiwenzhi?

91. Shi ji, 75.2358.

92. Sivin, , “The Myth of the Naturalists,” 20Google Scholar.

93. This is exemplified dramatically by the fate of the sons of Tian Wen, who even though he had accrued enough power as patron to establish himself as the ruler of an independent state, met destruction on their father's death (Shi ji, 75.2358). Sima Qian blames their demise on their own squabbling over the succession to their father's throne, but it is hard to imagine that the weakening of patron-client bonds did not play some role.

94. Zhanguo ce, “Qi si 齊四” (139/69/24–30).

95. For the “dog thief” and “cock-crow knight” see Shi ji, 75.2355.

96. Sivin, , “The Myth of the Naturalists,” 24Google Scholar.

97. As one anecdote about the Lord of Chunshen in the Zhanguo ce puts it, retainers were chiefly of value to their patrons as “pawns” in a contest for political supremacy (Zhanguo ce 188/96/15–21).

98. In the retinue of the Lord of Mengchang, retainers received different board according to rank, “Senior Retainers ate meat, Regular Retainers ate fish, Junior Retainers ate vegetables” 上客食肉, 中客食魚, 下客食菜 (Qingchang, Zhang 張清常, Zhangguo ce jianzhu 戰國策箋注 [Tianjin: Nankai University Press, 1994], 264n3Google Scholar. This is recorded in a fragment from the Lie shi zhuan 列士傳 cited in a commentary by Wu Shidao 吳師道 [1283–1344]).

99. This is exemplified by the story of the Lord of Mengchang's treatment of Feng Xuan 馮諼, an impoverished knight who haggled for a series of improvements to his emolument. Though Mengchang magnanimously granted each request, he personally had to approve each incremental increase. (Zhanguo ce, “Qi si 齊四” [133/64/18–65/30]. The tale also appears in Shi ji, 75.2359–62, where Feng's given name is recorded as Huan 驩).

100. When the knight Han Ming 汗明 especially impressed the Lord of Chunshen, “He summoned his Staff Secretary and had Venerable Han recorded in the roll of retainers, to be given audience every five days” 召門吏為汗先生著客籍, 五日一見 (Zhangguo ce, “Chu si 楚四”[198/101/4]).

101. Testimony differs as to the precise rank given Jixia's senior clients. Shi ji 46.2346 gives the title as “Senior Grand Master,” Shi ji 74.2347 gives the title as “Adjunct Grand Master 列大夫.” For the distinction between the two ranks, see: Hucker, Charles, A Dictionary of Official Titles in Imperial China. (Stanford: Stanford University Press, 1985), 311, 413Google Scholar. Hucker only records 列大夫 as having been an official rank beginning in Han times, so it is possible that the phrase at Shi ji 74 means generically “arrayed Grand Masters,” while Shi ji 46 specifies their rank as “Senior Grand Masters.”

102. The two rivers that bounded Qufu, Confucius' native city. For example: Zixian, Xiao 蕭子顯 Nan Qi shu 南齊書 (Beijing: Zhonghua shuju, 1972), 686Google Scholar.

103. The village from which emerged the “Old Text” Rites, which precipitated a schism among the latter-day disciples of Confucius. For example: Defen, Linghu 令狐 德棻, Zhou shu 周書 (Beijing: Zhonghua shuju, 1971), 742Google Scholar.

104. Sima Wengong wenji 司馬文公文集, “Juan shisi 卷十四.” The text is transcribed in Li'e, Lin, Xian Qin Qixue kao, 148Google Scholar.

105. Ni Yue was evidently a well-known logician during the late fourth/early third century B.C.E. He is also mentioned in Lüshi chunqiu, “Jun shou 君守” (17.2/100/26–101/2). Qian Mu puts him before Gongsun Long 公孫龍 (c. 320–250 B.C.E.) and speculates that Ni Yue inspired his famous “white horse” thesis (Xian Qin zhuzi xinian, 402–4).

106. Han Feizi, “Wai chu shuo zuo (xia)” 外儲說左下 (32/82/7–8).

107. Lu Zhonglian (c. 300–c. 245) was a famous persuader of the late Warring States. He has his own biography in the Shi ji (83.2459–79) and figures in many anecdotes in the Zhanguo ce. The Han shu yiwenzhi records a Lu Zhonglianzi 14 pian among the “Ru jia1” 儒家 texts in the Han imperial library (Han shu, 30.1726).

108. Little is known of Tian Ba other than what is recorded here. Lin Li'e argues that he was ranked among the Grand Masters at Jixia (Xian Qin Qixue kao, 192).

109. The identity of this location is not clear; later commentators identify it as a site in Qi.

110. Reconstructed in: Tingzhuo, Ruan 阮廷焯, Xian Qin zhuzi kaoyi 先秦諸子考佚 (Taibei: Dingwen shuju, 1980), 86–88Google Scholar.

111. A coalition of the armies of Zhao 趙, Hann 韓, Wei 魏, and Yan 燕, led by the Yan general Yue Yi 樂毅 (fl. c. 300–279 B.C.E.) overran Qi in 284 B.C.E., causing the flight and ultimately the death of King Min. Qi would only be restored to a semblance of its former territorial extent in 279 B.C.E. (Wang and Tang, Qi guo shi, 465–70, 511).

112. For the place of Mencius in the history of the Lu intellectual community see: Lau, D.C., trans., Mencius (London: Penguin Books, 1970), 7–12, 205–19Google Scholar.

113. Mengzi, “Liang Huiwang, xia” 梁惠王下 (2.4/9/3–20); Lau, , Mencius, 63–64Google Scholar. The parallel text is found at Yanzi chunqiu, “Nei pian wen, xia” 內篇問下 (4.1/31/15–25). One suspects subtle but deliberate irony: Mencius is using the very words of the king's own patronage text against him.

114. The range of this “Qi romance” narrative extends across Mengzi 1.7–4.14 (Mengzi zhuzi suoyin, 3–24; Lau, , Mencius, 54–94)Google Scholar.

115. Mengzi, “Liang Huiwang, shang” 梁惠王上 (1.7/3/20–24).

116. Mengzi, “Gongsun Chou, xia” 公孫丑下 (4.2/20/9).

117. Mengzi, “Gongsun Chou, xia”公孫丑下 (4.10/23/20–25).

118. This is a matter of some controversy. Mu, Qian (Xian Qin zhuzi xinian, 235–37)Google Scholar asserts that Mencius never lodged at Jixia; other interpreters argue the opposite (see Li'e, Lin, Xian Qin Qixue kao, 207–10)Google Scholar.

119. Csikszentmihalyi, Mark, “Confucius,” in The Rivers of Paradise: Moses, Buddha, Confucius, Jesus, and Muhammed as Religious Founders, eds. Freedman, David Noel and McClymond, Michael J. (Grand Rapids: William B. Eerdman's Publishing Company, 2001), 267, 268Google Scholar.

120. For example, Mencius' refusal of a gift of gold from the King of Qi even though he had accepted gifts of gold from other rulers made at ritually appropriate occasions (Mengzi, “Gongsun Chou, xia” 公孫丑下 [4.3/20/14–20]).

121. Lüshi chunqiu, “Gong ming” 功名 (2.4/10/ 11–13).

122. This is not to suggest that all of these groups embraced one and the same model of group identity and intellectual participation in public life, quite the contrary. All groups were free to and did innovate and elaborate upon the basic conventions of the “Master-disciple” fellowship and applied them to different ends. Mozi's latter-day disciples, for example, attempted to organize themselves into a paramilitary hierarchy that extended beyond the immediate Master-disciple group; thus they might be viewed as promoting a new model of intellectual participation distinct from either that of Qi or Lu (see Schwartz, Benjamin, The World of Thought in Ancient China [Cambridge: Harvard University Press, 1985] 138, 161)Google Scholar.

123. Yingjie, Qu 屈英傑, Xian Qin ducheng fuyuan yanjiu 先秦都城復原研究 (Harbin: Heilongjiang Renmin, 1991), 250Google Scholar.

124. Hung, Wu 巫鴻, “Rethinking Warring States Cities: An Historical and Methodological Proposal.” Journal of East Asian Archaelogy 3.1–2 (2001), 248–49Google Scholar.

125. For a thorough reconstruction of Xunzi's biography see: Knoblock, John, Xunzi: A Translation and Study of the Complete Works, Volume I (Stanford: Stanford University Press, 1988) 3–35Google Scholar.

126. Mengzi, “Gongsun Chou, shang”公孫丑上 (3.1/14/6); Lau, , Mencius, 74Google Scholar.

127. Mengzi, “Gongsun Chou, shang”公孫丑上 (3.1/14/24–26, 29).

128. This is exemplified by the story of the hermit, Chen Zhongzi 陳仲子, in Mengzi, “Teng Wengong, xia” 滕文公下 (6.10/35/12–25).

129. Mu, Qian, Xian Qin zhuzi xinian, 462–64Google Scholar.

130. Tsuen-hsuin, Tsien 錢存訓, Written on Bamboo and Silk: The Beginnings of Chinese Books and Inscriptions (Chicago: Chicago University Press, 2004) 94–106Google Scholar.

131. Knoblock, , Xunzi, 6Google Scholar.

132. For Gongsun Longzi, see: Hansen, Chad, Language and Logic in Early China (Ann Arbor: University of Michigan Press, 1983)Google Scholar.

133. Graham, A. C., Later Mohist Logic, Ethics, and Science (Hong Kong: The Chinese University Press, 1978)Google Scholar.

134. Guanzi, “Dizi zhi” 弟子職 (19.2 (59)/137/8–138/14). Translated in: Rickett, Allyn, Guanzi: Political Economic, and Philosophical Essays from Early China, A Study and Translation, Volume II (Princeton: Princeton University Press, 1998) 283–91Google Scholar.

135. For example, the abovementioned Tian Pian and Tian Ba.

136. Graham, A. C., “The Background of the Mencian Theory of Human Nature,” in Studies in Chinese Philosophy (Albany: State University of New York Press, 1990) 7–66Google Scholar.

137. Sivin, , “The Myth of the Naturalists,” 28Google Scholar.