1. “In Search of Danyang. I: Historical Geography and Archaeological Sites,” Early China 13 (1988), 116–152 (hereafter. Part I)CrossRefGoogle Scholar. Although there is a good deal more to be said about the Western Zhou period, for a number of reasons the meager evidence available for it is best viewed in light of what can be determined about the early Chunqiu era.

I would like to express my appreciation to Dr. Lu Yun 盧雲 of Fudan University for advice on several issues relating to this paper.

2. I adopt “Dan Valley” here for the combined Dan/Han and Dan/Xi views rather than “Xichuan” (utilized in Part I). This seems more appropriate, since the confluence of the Dan and Han rivers is in Hubei (thus, not in Xichuan county). (The upper Dan area of Shangxian 商縣, Shaanxi [see Part I, 126–127, 136] is not at issue in the present context.) To the discussion of this in Part I (127) might be added the fact that while “Danyang” appears in the Shiji on several occasions (e.g, 40.57/644) in connection with the 312 battle between Chu and Jin, on another (84.5/984) the current redaction has “Dan Xi.” (Citations to the Shiji and its commentaries are to Kametarō, Takikawa 瀧川麁太郞, Shiji kaichū kōshō [Shiji huizhu kaozheng] 史記會注考證 [1932–1943; repr., Taibei: Zhongxin, 1977].Google Scholar) It is true that some editions from the Song on have “Danyang” at this point; see Shiki kaichū kōshōkōho 史記會注考證校補 (1957; repr., Taibei: Guangwen, 1972), 4:84.2/2678Google Scholar. Nevertheless, the fact that the early Tang Sima Zhen 司馬貞 quotes the text here as having “Dan Xi,’ while the later Tang gloss of Zhang Shoujie 張守節 employs “Danyang”, suggests that either Tang redactions differed on the wording at this juncture or an emendation from “Dan Xi“ to “Danyang” was made during Tang times. If the original was in fact “Dan/Xi,” the battle assuredly took place in southwest Henan; and this would bolster my conclusion that the Dan River is the only verifiable pre-Han “Dan” placename.

3. A convenient survey of most of the geographical and chronological schemes arising from these may be found in Mitsuru, Taniguchi 谷口滿, “So to Tanyū tansaku — Kodai Sokoku seiritsuron” 楚都丹陽探索一古代楚國成立試論, Tōyōshi ronshū 東洋史論集 1 (1984.1), 9–11Google Scholar. This does not, however, take into account all of the scenarios which by now have been proposed. Also, since Taniguchi had not seen a 1984 paper by Quan, Shi 石泉 (“Chu du he shi qian Yin” 楚都何時遷郢, Jiang-Han luntan 江漢論埴 1984.4Google Scholar; reprinted in Gudai Jing-Chu dili xintan 古代荆楚地理新探 [Wuhan: Wuhan University Press, 1988], 349–354)Google Scholar, his characterization of the latter's position is not quite accurate.

4. All itinerant hypotheses posit a shift of the Chu center during Western Zhou times. It should be noted that there is no direct evidence that anything but the initial site was known as “Danyang.” Still, it is conventional to employ this term for any pre-Ying locale, and I follow this here.

5. It will be recalled (Part I, 126) that “Southern School” refers to those who place Danyang in southwest Hubei; “Northern Schoor applies to those who argue for northwest Hubei or southwest Henan. With the exception of Wen Bigui 文必貴 (see Part I, 126), most adherents of the Southern School have advocated Zhijiang for Danyang for early Chunqiu times. The archaeological site originally proposed, Jijiahu 季象湖, in present-day Dangyang 當陽, though, has been reinterpreted by many as an early site of Ying; see Part I, 128, 135 and Taniguchi, , “So to Tanyū”, 18–19Google Scholar.

6. Shuye, Tong 童書業, “Chunqiu Chu Yingdu bianyi” 舂秋楚郢都辨疑, Zhongguo gudai dili kaozheng lunwenji 中國击代地理考證論文集 (Shanghai: Zhonghua, 1962), 91–92Google Scholar; Quan, Shi, Gudai Jing-Chu, 349–354Google Scholar. Most other supporters of the Dan Valley as the location of the original Danyang do not address the issue of whether this was the only location; see Part I, 126–127, 136.

7. Tiefu, Gu 顧鐵符, “Chu san yi kao” 楚三邑考, in Chushi yanjiu zhuanji 楚史硏究專輯 (Wuhan: Hubeisheng Chushi yanjiuhui and Wuhan shifan xueyuan xuebao, 1983?), 17–27Google Scholar.

8. Zhengming, Zhang 張正名, “Chudu bian” 楚都辨, Jiang-Han luntan 1982.4: 64–66Google Scholar; Hao, He 何浩, “Chuzu nanqian ji qi Jiang-Han Ju-Zhang jidi di zaoqi jingying” 楚族南遷及其江漢沮漳幾地的早期經營, Jiang-Han luntan 1989.6: 75–76Google Scholar; Guanghao, Wang 王光浩, Chu wenhua yuanliu xinzheng 楚文化源流新證 (Wuhan: Wuhan University Press, 1988), 340–376 (especially 359–360), 457–460, 475Google Scholar. In Part I, I also reviewed an Anhui hypothesis; but given both the inadequacies noted there and all that follows below, this hypothesis may be eliminated from consideration for this era.

9. “Hanzhong” 漢中 has traditionally applied to a stretch of the Han considerably further upstream, but with respect to early Chu history, “Middle Han” is a reasonable designation for the area south of Xiangyang.

10. Shiji, 40.11/632; 14.59/214, to much the same effect: Wen Wang [personal name] li, shi du Ying 文王熊貲立, 始都郢. The “authority” for the traditional reading of this to mean that Ying had not existed before Wen Wang's time may have been Gu, Ban 班固 (Hanshu 漢書, 1:28a.17b/715)Google Scholar, who stated flatly that the move from Danyang to Ying occurred under King Wen. (All references to dynastic histories and their commentaries are to the Yiwen shuju edition [Taibei, 1956].)

11. See, for instance, the Mo, Wang 王誤 version (p. 34) in Shiben bazhong 世本八種 (Taibei: Xi'nan, 1974)Google Scholar. The first scholar to advocate this was the Qing period Song Xiangfeng 宋風; see Guotinglu 過庭錄 (Taibei: Guangwen, 1980), 4.9a–14b/82–94Google Scholar; also in Huang Qingjingjie xubian 皇淸經解績編, vol. 411). Sima Qian utilized a Shiben of some sort (see, e.g., Part I, 143, n. 16), but the history of this text (or texts) is so murky that there is no way to know if the redaction he had at hand included the relevant Shiben passage.

12. Quan, Shi, Gudai Jing-Chu, 352–353Google Scholar; Hao, He, “Chuzu nanqian”, 74Google Scholar.

13. Before 706: Xiangfeng, Song, Guotinglu, 91–92Google Scholar; Tiefu, Gu, “Chu sanyi kao,” 27–38Google Scholar; Hao, He, “Chuzu nanqian”, 71–72Google Scholar. Between 703 and 699: Quan, Shi, Gudai Jing-Chu, 349–354Google Scholar. Between 703 and 701: Shuye, Tong, “Yingdu bianyi,” 92–93Google Scholar. By 701: Guanghao, Wang, Chu wenhua, 375Google Scholar. Taniguchi, (“So to Tanyū”, 19–27)Google Scholar argues that there was a move under King Wu (by 699), but that it was from one Ying (Jijiahu) to another (Jinancheng 紀南城, near Jiangling 江陵). These chronological differences relate closely to differences of opinion on the locations of the two capitals, and result in turn in varying interpretations of the geographical implications of Chu foreign relations.

14. Why this was not detected long ago is a mystery, since the evidence has been readily available since at least the sixth century. It has been cited by both the Southern School Taniguchi (“So to Tanyū”, 12–13) and the Northern School Shi Quan (Gudai Jing-Chu, 351–352). (Their disagreement as to the location of Ying in no way affects this conclusion.)

15. Part I, 118–119.

16. A thorough analysis of these issues may be found in Guanghao, Wang, Chu wenhua, 377–456Google Scholar.

17. They generally do so by reinterpreting the Jijiahu site as an earlier Ying (see n. 5).

18. Part I, 118–119; Quan, Shi, Gudai Jing-Chu, 211–348 (211–257, previously published in Wuhan daxue xuebao 武漢大學學報 1982.1, 2)Google Scholar.

19. At least before the reinterpretation of Jijiahu (see n. 5).

20. Zhang has not discussed the issue of the move to Ying in detail; but he advocates Nanzhang for Danyang (n. 8) and is in sympathy with Shi Quan's Yicheng hypothesis on Ying (“Chudu bian,” 65).

21. This is a pre-Qin term, first appearing in the Zuozhuan (Huan 6: 2.42): Wu bu de zhi yu Handong ye 吾不得志於漢東也.

22. Zuozhuan Huan 11: 2.63–65. All references to the Zuozhuan, as well as to the commentaries of Yu, Du 杜預 and Shin'ichirō, Takezoe 竹添光鴻, are to the latter's Saden kaisen (Zuozhuan huijian) 左傳會塞 (1907; rept., Taibei: Fenghuang, 1977)Google Scholar.

23. That forces from the other states did not participate is inferred since the Zuozhuan states that Chu defeated a Yun force (only). Whether Yun's allies decided not to send forces, were prevented from doing so by a Chu force left at Jiao Ying (see below), or whether their forces simply did not arrive in time (the episode is dated merely to “spring”) is impossible to say.

24. Other campaigns dated to specific years occurred in 706 (Zuozhuan Huan 6: 2.41–45)Google Scholar, 704 (Huan 8: 2.53–55), 703 (Huan 9: 2.57–58), 700 (Huan 12: 2.69–70), 699 (Huan 13: 2.71–73), and 690 (Zhuang 4: 3.7–9). A fifth-century (478) flashback discourse passage in the Zuozhuan (Ai 17: 30.36), alludes (without dates) to several military and diplomatic events not otherwise recorded. The 706, 704, and 690 incidents appear in the Shiji (40.10–11/632), albeit in quite different form.

25. This approach is central to Shi Quart's and Tanigucht's analyses of the 699 campaign (see n. 14). Shi also utilizes it in attempting to establish the implications of the 703 campaign; Gudai Jing-Chu, 105–126 (originally in Jiang-Han luntan 1980.3Google Scholar). The most intensive use, in connection with the 701 incident, has been by Tong Shuye; “Yingdu bianyi”, 93. However, he follows traditional views on some locations, and even though he questions others, he does not attempt to clarify them.

26. It must be recognized that since ancient rimes both natural and human processes (especially drainage projects, beginning in earnest in Ming times) have altered conditions in the lower Han valley. These changes are of considerable relevance to the feasibility of marches through certain segments of Handong and virtually all of Hannan. (The main issues are the location and extent of the Yunmeng marsh 雲夢澤 and changes in the courses of the Han, its tributaries and distributaries.) However, these matters are so complex and unsettled that they cannot be dealt with in the present context A survey of some of the pertinent issues is available in Quan, Shi, Gudai Jing-Chu (29–35)Google Scholar and “XianQin zhi Hanchu ‘Yunmeng’ diwang tanyuan—gu Yunmengze guzhi xintan zhi yi” 先秦至漢初「雲夢」地望源一古雲夢澤故址新探之一, in Chu wenhua xintan 楚文化新探 (Wuhan: Hubei renmin, 1981), 91–101Google Scholar. It would be awkward to qualify county names at every turn by adjectives such as “ancient” and “modern.” I have done so only where it seems necessary for clarity; otherwise, I trust the context will suffice. Unless otherwise noted, locations are given in terms of contemporary county names, even if these do not preserve early (e.g., Han through Tang) nomenclature.

27. As noted above, in traditional scholarship, with the relatively late exception of Song Xiangfeng, the assumption was that the capital at this juncture was a Danyang in southwest Hubei. Although Song argued for Ying as the capital at this time, he located it in the Yangzi Plain (Jiangling).

28. The principles enunciated in Part I (esp. pp. 117–119), although not repeatedly identified, will be employed. In order to conserve space, opinions expressed in Qing and post-Qing commentaries and compendia are not given in the footnotes unless they were the first to propose an opinion, were particularly influential in preserving one, or serve to illustrate current attitudes. Such works which have been consulted but which do not appear in the footnotes include, Gu Donggao 顧棟高, Chunqiu dashibiao 春秋大事表 (Huang Qing jingjie xubian ed.); Gu Zuyu 顧祖禹, Dushi fangyu jiyao 讚史方輿紀要 (Guoxue jiben congshu 國學基本叢書 ed.; Taibei: n.p., 1968Google Scholar); Jiang Yong 江永, Chunqiu dili kaoshi 春秋地理考釋 (Huang Qing jingjie ed.); Liang Lüsheng 梁履繩, Zuotong bushi 左通補釋 (Huang Qing jingjie xubian ed.); Shizong 世宗 (Yongzheng 雍正), Chunqiu jieyi 春秋解疑 (n.p: n.d); Bojun, Yang 楊伯俊, Chunqiu Zuozhuan zhu 春秋左傳注, 4 vols. (Beijing: Zhonghua, 1981)Google Scholar; Zongchui, Zhang 張宗炊, comp. Hubei tongzhi 湖北通志 (1921)Google Scholar.

29. Daiyan, Liang 梁戴言 (early Tang), Shidaozhi 十道志 (in Mo, Wang 王謨, Han-Tang diiishu chao 漢唐地理書鈔 (1898; repr., Beijing: Zhonghua, 1961], 1.22b-23a/277–278Google Scholar; Tai, Li 李泰 (618–653), Kuadizhi 括地志 (642), ed. Cijun, He 賀次君 (Beijing: Zhonghua, 1980), 218Google Scholar.

30. You, Du 杜佑 (735–812), Tongdian 通典 (Siku quanshu ed.), 605:183.11a/513Google Scholar; and Jifu, Li 李吉甫 (758–814), Yuanhe junxian tuzhi 元和郡縣圔志, 2 vols. (Beijing: Zhonghua, 1983), vol. 2, 27/652Google Scholar. Du Yu (222–284) had provided no location for Pusao; he simply defined it as a Yun town (yi 邑).

31. Shan, Wang 王淡 (fl. 17th c), Chunqiu zhuanshuo huizuan 春秋傳說會篸, 61:6.2bGoogle Scholar, and Shiqi, Gao 高士奇 (1645–1704), Chunqiu diming kaolüe , 13.14aGoogle Scholar. (References for both works are to the Siku quanshu zhenben eds.)

32. Shan, Wang, Chunqiu huizuan, 61:6.3aGoogle Scholar; Shiqi, Gao, Diming kaolüe, 8.7a–bGoogle Scholar.

33. For Tang views, see Jifu, Li, Yuanhe tuzhi, 2:27/649Google Scholar; You, Du, Tongdian, 605:183.10b/513Google Scholar; “Menghuitu” 盟會圖, an appendage to Yu's, DuChunqiu skill 舂秋釋例 (Siku quanshu ed.), 7.63ab/185Google Scholar. For Song views, see Shi, Yue 樂史 (930–1007), Taiping huanyuji 太平寰宇記 (Siku quanshu ed.), 131.53/281Google Scholar; Mi, Luo 羅泌 (fl. 12th c), Lushi 路史 (Siku quanshu zhenben ed.), 104:26.6bGoogle Scholar; Qiao, Zheng 鄭樵 (1104–1162), Tongzhi 通志 (Siku quanshu ed), 26.26aGoogle Scholar. Down to the end of the Qing I have detected only two exceptions to this identification: Qinhan, Shen 沈韓 (1775–1860), Chunqiu Zuoshizhuan diming buzhu 春秋左氏傳地名補注 (HuangQing jingjie xubian ed.), 1.14aGoogle Scholar; and Takezoe, , Saden kaisen, 2.63Google Scholar. Otherwise, it was the standard view and is followed in Qixiang, Tan 譚其鼷, Zhongguo lishi dituji 中國歷史地圖集, 8 vols. (Shanghai: Ditu, 1982), 2:29Google Scholar. Thorough consideration of the location of Yun would have to take into account other references to it. Among these, one in 506 (Zuozhuan Ding 4: 27.38), ranks with the present one in importance; see Quan, Shi, Gudai Jing-Chu, 355–416Google Scholar. Also excluded from consideration below is a view which assigns Yun to Xishui 稀水 county. Pan, Chen 陳樂, Chunqiu dashibiao lieguo juexing ji cunmie biao xuanyi 春秋大事表歹爵姓及存滅讓異 [sub-commentary on Gu Donggao, see n. 28], 7 vols. (Taibei: Academia Sinica, 1969) 3:237aGoogle Scholar, offers sufficient indictment of this, although more could be said on the matter. Ignored generally, too, are what would appear to be attempts to reconcile earlier views resulting in a Yun of considerable size; see, for instance, Shiqi, Gao, Diming kaolüe, 14.14abGoogle Scholar.

34. Yu, Du, in Takezoe, Saden kuisen, 2.63Google Scholar.

35. Guowei, Wang 王國維, Shuijingzhu jiao 水經注校, ed. Yingguang, Yuan 袁英光 and Yinsheng, Liu 劉寅生, (Shanghai: Renmin, 1984) 28/914Google Scholar.

36. Du Yu's identification was accepted in the commentary (zhubu 注補)of Liu Zhao 劉昭 (fl. 510) to Sima Biao's 司馬彪 Zu Hanshu (see Xu Hanshu bazhi 續漢書八志 [Jiangning, 1869], 22.4bGoogle Scholar), and in the Jinshu 晉書 (1:15a.23b/351), which was compiled at about the time the Anlu identification for Yun was being proposed. Li Xian 李賢 (651–684), in his commentary (zhu 注） to the Hou 後 Hanshu (3:22.23b, 24a/1277) equivocated, appending Du Yu's gloss on Yun to the Yuncheng in Yundu entry, but mentioning the 701 B.C. incident under the notice of Yunxiang in Jingling.

37. A third factor could have been the placing of Pusao in Yincheng, near Anlu. However, this seems rather improbable since this appeared later than the Anlu proposal on Yun. This could be deceptive, but the evidence for Pusao being in Yin geh eng is so tenuous (see below) that it is hard to conceive that Yun would have been plotted on the basis of it. If, though, this factor was involved, it would only serve to strengthen the conclusions reached in this discussion.

38. See Mi, Luo, Lushi, 104:26.6bGoogle Scholar.

39. A Yuncheng in Anlu appears in Daoyuan's, Li sixth-century Shuijingzhu; 31/1008Google Scholar.

40. Du's mention of the Yundu Yuncheng dates to the third century, and Li's Jingling Yunxiang can be traced to the first century; see Hanshu 1:28a.23b/718Google Scholar.

41. See Shuijingzhu 31/1006 ffGoogle Scholar.

42. Li Xian pointed to Tianmen (the Tang Fuzhou 復州) in one context (Hou Hanshu zhu, 2:75.1a/869Google Scholar), and Zhongxiang (at the time, Changshou ) in another (1:11.1b/178), making no attempt to reconcile the two. This confusion was perpetuated by the early ninth-century Du You; Tongdian, 605:183.9b, 10aGoogle Scholar. In more recent times, Jianjiang in Hannan has also been proposed as the locale of Jingling; see Shoujing, Yang 楊守敬, Lidai yuditu 歷代輿地圖 (1906; repr., Taibei: Lianjing, 1975), 2 (Former Han), Map 49Google Scholar; see, also, Qixiang, Tan, Dituji, 2:49Google Scholar. If, however, Shi Quan's hypothesis on Yundu (see below) is valid. Jingling had to be north of it, in Zhongxiang (Gudai Jing-Chu, 127–155; previously, in Jiang-Han kaogu 1980.1Google Scholar). Moreover, Yang Shoujing equivocated, elsewhere accepting the Zhongxiang view; see Shoujing, Yang and Huizhen, Xiong 貪旨會貞, Shuijingzhu shu 水經注蔬, 18 vols. (1879–1915; facsimile repr. of ms., Taibei: Zhonghua, 1971), 13:28/3523Google Scholar.

43. You, Du, Tongdian, 605:183.9b, 10aGoogle Scholar; Zhang Shoujie 張守節 (fl. 736) Shiji zhengyi 正義 (in Shiji), 73.3/914Google Scholar. Li Jifu was the first to suggest that the capital was in Tianmen; Yuanhe tuzhi, 1:21/536, 538Google Scholar. The Tianmen view on Jingling was perpetuated in the Song: Shi, Yue, Taiping huanyuji, 144.8b/372, 12a/374Google Scholar; Xiangzhi, Wang 王象之 (fl. 1196), Yudi jisheng 輿地已勝, 10 vols. (1849 ed.; citations are to vol. 5), 84.7bGoogle Scholar, and has been followed almost universally since. See, for instance, Pan, Chen, Dashibiao xuanyi, 236b, 237aGoogle Scholar. A Zhanguo campaign of 278 B.C. (Shiji, 73.3/914Google Scholar) in which Jingling figured may well have influenced the placing of Jingling in Tianmen, but this is based on the Southern School assumption concerning the location of Ying, and also involves other problems; see Quan, Shi, Gudai Jing-Chu, 130, 385–386, 392–394Google Scholar.

44. This is suggested by the following inconsistency in Li Jifu's glosses: If Jingling extended from Zhongxiang to Tianmen, the intervening Jingshan should also have been within its sphere. Yet, he assigned Jingshan to Yundu; see n. 54.

45. Xian, Li, Hou Hanshu zhu, 1:11.1b/178, 3:22.24a/1277Google Scholar. On the precise location, see Quan, Shi, Gudai Jing-Chu, 160Google Scholar.

46. Jifu, Li, Yuanhe tuzhi, 1:21/536, 538Google Scholar.

47. An Anlu xian has existed in Eastern Handong since Han times, within Anlu jun 郡 from the mid-fifth centuries through the Tang, and within De'an fu 德安府 from the Southern Song through the Qing. Those surveying the commentary tradition, however, should note that the Qing Anlu fu had its seat at Zhongxiang in Western Handong.

48. Hanshu, 1:28a.23b/718Google Scholar.

49. Daoyuan's, Li gloss (Shuijingzhu, 28/914) readsGoogle Scholar, “… there is a sizable old [e.g., ruined] city, Jingling, which was the [capital of the] ancient Yun state. This is what was governed by Xin 辛, the [Chu] administrator (gong 公) of Yun, which was [later] known as Yunxiang.” The Chu unit alluded to here was identified as a town (yi) in the Hanshu (1:28a.23b/718)Google Scholar; that it functioned as a xian (e.g., Zuozhuan Zhao 14: 23.31Google Scholar) was derived by Du Yu from the title of its administrator; in Takezoe, Saden kaisen, 3.90Google Scholar. No source before Li Daoyuan, however, equated this with the Yun state; see Hanshu, ibid; Sima Biao (240–305; Xu Hanshu, 22.4aGoogle Scholar); and Hou Hanshu (3:22.23b/1277)Google Scholar.

50. Du Yu was silent on the location of the xian. Some might interpret this to mean that he considered it as equivalent to the state. However, he usually did not provide a gloss when he had no basis on which to do so (see my “Notes on the Reliability and Objectivity of the Tu Yü [Du Yu] Commentary on the Tso Chuan [Zuozhuan], “Journal of the American Oriental Society 101.2 [1981], 207–212Google Scholar). Hence, his silence on the location of the xian does not serve as evidence that he took it and the state to be the same. On the nonequivalence of the two, also see Quan, Shi, Gudai Jing-Chu, 146Google Scholar.

51. Shi, Yue, Taiping huanyuji, 131.5a/281Google Scholar; Gongyue, Cheng, Chunqiu fenji, 31.13b/334Google Scholar.

52. Da Qing yitongzhi 大淸一統志, 11 vols. (Jiaqing rev. ed.; repr. Taibei: Shangwu, 1966), 7:338.20b/4367Google Scholar; Shoujing, Yang and Huizhen, Xiong, Shuijingzhu shu, 13: 28/3526abGoogle Scholar.

53. Involved is an appeal to a statement in Xu Shen's 許慎 Shuowen jiezi 說文解字 (under the entry yun 鄭)to the effect that Yun had been located south of the Han River. Quan, Shi (Gudai Jing-Chu, 165)Google Scholar points to Yang Shoujing and Xiong Huizhen as offering this argument, citing a P.R.C. 1957 photo reprint of their Shuijingzhu shu. I have not had access to this edition. The one available to me (n. 42) lacks the reference, but a lacuna exists where the comment should be (13:28/3526b). As Shi points out, however, reading this passage in such a way as to defend the idea that Yun was in Hannan is entirely unwarranted, for it is appended to a reference to Hanzhong (in the upper Han valley). Zhenxing, Li 李振興, Shuowen dili tukao 說文地理圖考 (Ph.D. diss: Taiwan U., 1972), 216, concursGoogle Scholar.

54. This stems from Jifu, Li (Yuanhe tuzhi, 1:21/538)Google Scholar; but this is illogical in his own scheme, since he claimed that the intervening Tianmen was in Jingling (n. 43).

55. Xiangzhi, Wang, Yudi jisheng, 84.3bGoogle Scholar.

56. Takezoe, , Saden kaisen, 2.63Google Scholar, advocated it; and it was noted in passing in local gazetteers of the Qing and later. It has been adopted in Qixiang, Tan, Dituji, 2:22, 49Google Scholar, but not as the location of Yun.

57. The full argument (as well as quotations of the relevant Hanshu and Hou Hanshu passages) appears in Gudai Jing-Chu, 158–173 (for an earlier version see Jiang-Han xuebao 1961.1Google Scholar).

58. Shuijingzhu, 28/916–917; see Quan, Shi, Gudai Jing-Chu, 162–166Google Scholar.

59. The wording of Du Yu's gloss (“Yun guo mi Jiangxia Yundu xian dongnan you Yuncheng” 郿國在江夏雲杜縣東南有鄭城) is such that Li Jifu's position that Yundu extended from Jingshan into Mianyang would make it possible that the Yun capital was in the intervening Tianmen (even though this would not be the case in his own scheme; see n. 45). However, given the lack of geographical sense in his glosses, along with the limited geographical extent of early Chunqiu states, it is safest to presume that the Yun capital was in Jingshan, as the Song Wang Xiangzhi (Yudi jisheng, 84.3b) maintained.

60. The only other suggestion for Pusao that I have encountered, Zaoyang in Northern Handong, is by the highly idiosyncratic Lü Tiaoyang 呂調陽 (1516–1580); see his Qunjing shidi 群經釋地 (Guanxianglu congshu 觀象廬叢書), 25:3.20bGoogle Scholar. No serious consideration need be given to this.

61. The Anlu identification appeared in the first half of the seventh century, while the “Pusao in Yingcheng” idea appeared in the early ninth century. Unfortunately, neither Du You nor Li Jifu indicated whether this latter judgment on Pusao was made in their own time or whether it derived from some earlier work. While there is no reference to Pusao in the extant Shidaozhi and Kuadizhi, the fact that we have only fragments of these means that it is possible that the proposals on Yun and Pusao stemmed from the same source. If so, they would have been linked from the outset.

62. The association of Danyang with Zhijiang was put forth in the Latter Han, and that with Zigui in the sixth century; see Part I 125; 145, nn. 30 and 32. The relevant Tang analysts (Tai, Li, Kuadizhi, 196Google Scholar; You, Du, Tongdian, 605:183.8b/512Google Scholar; Jifu, Li, Yuanhe tuzhi, 2: Yiwen 逸文 1/1056Google Scholar) all assumed one or the other of these Yangzi Plain locales for Danyang.

63. The allusion to gazetteers was made by Gao Shiqi; Diming kaolüe, 13.14aGoogle Scholar. (Wang Shan simply made unsupported assertions; Chunqiu huizuan, 61:6.2bGoogle Scholar). Since most pre-Ming works took no stand, though, there is no reason to place confidence in such sources, though they are still accepted in Qixiang, Tan, Dituji, 2:29Google Scholar. Qinhan, Shen (Diming buzhu, 1.14a)Google Scholar did question these conclusions, reservations that were repeated by Wenqi, Liu 劉文 (1789–1854), Chunqiu Zuoshizhuan jiuzhushu zheng 春秋左氏傳舊注疏證, 2 vols. (Hong Kong: Taiping, 1966), 1:112Google Scholar.

64. Duanlin, Ma 馬林 (ca. 1254–ca. 1323), Wenxian tongkao 文獻通考 (Siku quanshu ed.), 263.31b–32aGoogle Scholar.

65. Shiyuan, Chen 陳士元 (fl. 1544), Xing Xi 姓摑 (repr., Hubei conqshu 湖北叢書 [1891], vol. 8), 7.4aGoogle Scholar.

66. Chen Shiyuan attributed this to the Hou Hanshu. No such reference appears, though, in the extant redaction. That the idea was expressed, albeit without attribution, as early as the Song (Mi, Luo, Lushi, 103:25.11aGoogle Scholar; Qiao, Zheng, Tongzhi, 26.32a/289Google Scholar) makes Chen's assertion feasible. On the other hand, neither Li Daoyuan nor any Tang writer mentions it.

67. Gongyue, Cheng (Chunqiu fenji, 31.24ab/340)Google Scholar quoted Li Daoyuan as mentioning a Zhenxiang in Zhonglu 中魔 county. There is no question about the position of Zhonglu, which was established in Jin 晉 times; see Shuijingzhu, 28/904–905, and Qixiang, TanDituji, 2:22Google Scholar.

68. As far back as Ming times (Yongle dadian 永樂大典 ed. [Reprint of Sibu congkan ed.; Taibei: n.p., 1971], 8:11, 137.20b)Google Scholar, all editions of the Shuijingzhu seem to have had Lingxiang 鈴鄉 rather than Zhenxiang.

69. This is patently obvious if the march set forth from either the Dan Valley or the Middle Han area. It also applies, though, to the Yangzi Plain, since the march (northward) surely would have been along the west bank of the Han.

70. Qiao, Zheng, Tongzhi, 26.32a/289Google Scholar, and Mi, Luo, Lushi, 103:25.11aGoogle Scholar, respectively.

71. From the Southern School perspective, east or southeast of Chu would imply a site in Hannan, but this should have been no serious obstacle. First, as early as the Song, Zhou was assigned to that area (see below). Second, while a Zhen in that area would require that the Chu force visited it before proceeding to Pusao, Qing analysts implied as much in their own location of Zhen.

72. Both Luo Mi and Zheng Qiao had asserted that the Zuozhuan itself gave this location for Zhen. It actually does no such thing.

73. This derives from Du Yu's allusion to a Zhoulingcheng in Huarong 華容 county; Takezoe, , Saden kaisen, 2.63Google Scholar; Chunqiu shili, 7.4b/156Google Scholar. Liu Wenqi alluded to an unidentified “tujing” 圖經 as the source of assigning Zhou to Jianli; Chunqiu jiuzhushu, 112. (The tujing was an early geographical genre.) The reference does not seem to appear in the tujing reproduced in Wang Mo's Dilishu chao, and I thus can trace this back only as far as the Song; see Gongyue, Cheng, Chunqiu fenji, 31.24b/340Google Scholar. Tan Qi-xiang's map does not plot Zhou; Dituji, 2:29Google Scholar.

74. The Zhoulingcheng in Jianli was presumably the seat of a county by that name established in Han times; see Hanshu 3:28a.21a/717Google Scholar. However, even if this has a connection with the ancient state of Zhou, it is quite possible that tt had moved subsequent to the campaign of 701 B.C. The only other reference to Zhou seems to be from the later Warring States; Shiji, 40.79/649Google Scholar. Whether or not it is reasonable that it lay along the Yangzi at that time depends on several questions (related to the location of Ying) that cannot be addressed here.

75. Quan, Shi, Gudai Jing-Chu, 241Google Scholar, Map, (= Wuhan daxue xuebao 1982.2: 54Google Scholar), 483, nn. 9, 10.

76. Hanshu (1:28a.14a/713)Google Scholar and Du Yu (Takezoe, , Saden kaisen, 2.41Google Scholar) both located it to Suixian. There are, of course, differences of opinion on the precise location of the capital of the ancient state (see Quan, Shi, Gudai Jing-Chu, 92–94Google Scholar), but this is not of concern here.

77. Du Yu (Takezoe, , Saden kaisen, 2.63Google Scholar; Chunqiu shili 7.4b/156Google Scholar) placed it in the Zaoyang county of his day. The particular site he pointed to would be in the south of present-day Tanghe 唐河 county, near the Hubei border; see Faren, Cheng, Diming tukao, 122Google Scholar. Gongyue, Cheng (Chunqiu fenji, 31.23b/339Google Scholar) of the Song dynasty and the vast majority of subsequent writers have followed this view. (This Liao should not be confused with one mentioned in 622 B.C. [Zuozhuan Wen 5: 8.34Google Scholar; Shiji 40.17/634Google Scholar]; see Takezoe 2.63, Wenqi, Liu, Chunqiu jiuzhushu, 112.Google Scholar)

78. Du Yu offered no guidance on the location of Jiao. The Tang Lin Bao 林寶 (fl. 9th c.) suggested that Jiao was in Handong (Yuanhe xingzuan 元和姓幕, [Siku quanshu ed.], 7.128/669Google Scholar), an idea repeated in the Song by Qiao, Zheng (Tongzhi, 26.329/289)Google Scholar and Mi, Luo (Lushi, 103:25.11aGoogle Scholar). This was an extrapolation from the events of 701 B.C. and the 478 flashback (see n. 24), placing Jiao closer to Sui, Liao, and Tang 唐. But it ignores clear implications of the account of the 700 B.C. campaign (see n. 24.)

In the early Qing period, Wang Shan proposed a site in Yun 鄙 county well up the Han River, near the Shaanxi border; Chunqiu huizuan, 61:6.2bGoogle Scholar. This subsequently became the standard opinion on the matter; see Qixiang, Tan, Dituji, 2:29Google Scholar. However, on the basis of the events of 701, Jiao could hardly have been that remote from Handong. Recently, Shi Quan, addressing the incident of 700 B.C., has suggested a site considerably further downstream in Jun 均 county; Gudai Jing-Chu, 354, n. 6. Both of these locations are fundamentally conjectural, but the mention in this episode of the Peng River, which is identifiable with the modern Nanhe 南河 (see Du Yu, in Takezoe, Saden kaisen, 2.69Google Scholar; Daoyuan, Li, Shuijingzhu, 29/938Google Scholar) indicates clearly that Jiao was west of the Han.

79. Takezoe utilizes this point in arguing that Yun was in Jingshan rather than Anlu, even though he accepts the traditional proposals on Pusao, Er and Zhen; Saden kaisen, 2.63Google Scholar. It must be acknowledged that this point would have no force if Yun were simply unrealistic in its expectation that assistance could reach Pusao before the arrival of the Chu force. This, however, seems highly improbable.

80. A Zuozhuan flashback discourse passage (Zhao 23: 25.12) can be taken to imply that there had been a Ying as early as the time of Ruo Ao 若敖 (r. 790–764). Tong Shu-ye (“Yingdu bianyi,” 92) and Taniguchi (“So to Tanyū, 18–19) take this to be a different Ying from the one which became the capital under King Wu or King Wen, but no one has considered the possibility that it was Jiao Ying.

81. Xu Shen's Shuowen definition of a state's jiao, “100 li from [the capital] of a state”, would surely have suggested to them that the Chu frontier extended no more than forty-eight kilometers (thirty miles) from the walls of its capital, hence well within the Yangzi Plain.

82. All three points are made by Gao Shiqi; Diming kaolüe, 8.7abGoogle Scholar.

83. In the context of the Southern School perspectives on Danyang and Ying, this location for JiaoYing results in a Chu state of considerable size. Depending on the location of the capital (i.e., in Jiangling or Dangyang), its territory would have extended northward as the crow flies about one hundred kilometers (sixty miles) or overland between 125 and 175 kilometers (seventy-five to 105 miles). I doubt that it was this large at this stage in its development.

84. The only requirement would be that JiaoYing be closer to Yicheng than Zhongxiang is. The latter locale would still be feasible from a strategic standpoint, but it would again imply a Chu state of considerable size. (The seats of Yicheng and Zhongxiang are about one-hundred kilometers, or sixty miles, distant overland, only slightly less than the distance between Jiangling and Zhongxiang.) By the same token, as long as the “jiao” of JiaoYing had anything to do with the Chu frontier, the Dan Valley would be excluded from consideration. A Chu force posted at a JiaoYing anywhere in that vicinity could not have prevented the allies from reaching Yun regardless of its location.

85. From this perspective, the earliest hint for Zhen (“southeast of Chu”, see n. 70) is not entirely inconsistent with the Southern Handong (Yingcheng) position suggested in Qing times. (It would, however, have had to derive from a pre-Tang source, since once the view that Chu was centered in the Yangzi Plain became entrenched, “southeast of Chu” would point to Hannan.) As for Er, from the perspective of the scenario proposed here, the inherently weak foundation for placing it in Eastern Handong (see n. 66) becomes even more suspect, and a site in the vicinity of Zhen would be more reasonable. It should also be noted that the Zuozhuan account does not require that the Chu force was heading directly toward Er. It is instead possible either that a representative from Er was to journey to Zhen or that a convocation was to be held at a place convenient to alt parties.