1. In this regard it would be most helpful if scholars when dealing with philosophic problems would abandon the practice of citing works only by title, juan, and page, and adopt the practice of citing the pian title as well. When scholarship has sufficiently progressed, paragraph number within the pian would be an ideal form of citation. In this paper I refer to the individual pian of the Xunzi as separate works and within the pian to paragraphs in accordance with my forthcoming complete translation of the works of Xunzi. These paragraphs correspond for the most part with those found in the Song editions, in the Lu Wenchao and Kubo Ai editions, and in that of Wang Xianqian. Liang Qixiong recognizes many more paragraphs. The form of citation is by pian title, pian number in the Yang Liang edition, and by paragraph number, thus Quanxue 1.12.

2. Later Mohist Logic, Hong Kong: 1978Google ScholarPubMed.

3. Sniki kaichü kösnö , 74.12–14Google Scholar; 78.15; 87.1-2, 14.

4. Xianqian, Wang, Xunzi Jijie , 20.24a–28aGoogle Scholar. Cf. Liu Shipei , Liu Xiang “Sun Qing shulu” jiaobu, 4.16ab.

5. Fengsu tongyi [aD. FSTY], 7.2a (SBBY edition).

6. Zhangup Ce [ab. ZGC], 5.38b (SBCK edition).

7. Hanfeizi [ab. HFZ], 16.4a [SBBY]; Kuan, Huan, Yantie lun [ab. YTL], 2.11a, 4.5b–7aGoogle Scholar [SBBY].

8. Botii Wang Znong, Xun Qingzi nianbiao, in Shuxue buyi, (1869), and Hu Yuanyi, Xun Qing biezhuan kaoyi (apparently never published), are quoted in extenso in Wang Xianqian's Prolegomena (Kaoyi).

9. Shi, Hu, Zhongguo zhexue shi dagang , 303fGoogle Scholar; Qichao, Liang, “Xun Qing ji Xunzi” , Gushibian , IV (1933), 104–115Google Scholar; Genze, Lo, “Xun Qing yuli kao” , Gushibian, IV (1933), 123–136Google Scholar; Mu, Qian, “Xun Qing kao” , Gushibian, IV (1933), 115–123Google Scholar, and Xian Qin zhuzi xinian , Hong Kong: 1956, 437–438Google Scholar.

10. Xuan, Znu, “Meng Xun shu shi kao ji Meng Xun liezhuan snuzheng” , Guoli Taiwan shifan daxue gowen yenjiuso jikan , X (1966), 69–216Google Scholar.

11. Daxue, Beijing, Xunzi xinzhu , Beijing:1979, 513–516Google Scholar.

12. Junzhai dushu zhi , ed. Xianqian, Wang, Chang-sha, : 1884, 10.7b–8aGoogle Scholar; Liang Yusheng, Shi Ji zhiyi , 29.29b.

13. Yanshi jiaxun huizhu, 38b.

14. SBBY edition, 20.20a.

15. This edition is described in a traced manuscript version which seems to have been the only one still extant in the 19th century, by Jinwu, Zhang, Airi jinglu cangsnu zhi (1826 edition), 21.1b–2bGoogle Scholar. Qu Yong , Tieqin tongjian lou cangshu mulu(Shumu congbian reprint), 13.2b, gives the date as during the reign of Song Yingzong (1064-1067), but the original postface, as preserved in the Taizhou reprint of 1181, is dated to 1068. Qian Dian's colophon says that the Lü Xiaqing edition was published during the Yuanfeng reign period (1078-1085) which date is accepted by Yang Shaoho , Yingshu yulu (1912 edition), 3.1a.

16. The Taizhou edition was found in Japan by Mori Tachiyuki in the Kanezawa Bunko and reprinted in the Guyi congshu (and again in Sibu congkan). The postface of Tang Yu-zheng is dated to 1181. The circumstances of the discovery and reprinting of the Taizhou edition are discussed in the colophon of Yang Shoujing ., Riben fanshu zhi, 7.18a-19a. The circumstances of its reprinting by Tang are described by Mori in Jingji fangshu zhi (Shumu congbian reprint), 156-158. That the Taizhou edition was the same as the Lü Xiaqing edition is confirmed by Wang Yinglin, Kunxue jiwen, 10.849.

17. The Qian Dian edition does not appear to be extant, but is de-scribed in the colophons of Yong, Qu, Tieqin tongjian lou cangshu mulu 13.2b–3bGoogle Scholar, which quotes Qian Dian's personal colophon dated to 1181. Qian Dian's edition, published in the sixth month, is earlier than the Taizhou reprint of the Lü Xiaqing edition published in the eleventh month. The Qian Dian edition was considered the best and most perfect by Chen Zhensun , Zhiztiai shulu jieti (Shumu xubian reprint), 9.2a.

18. Of the Kaoyi, there are two editions, one edited by Gu Guangqi reprinted in the Congshu jicheng series and another by Miao Quansun published in the Duiliang lou cong-shu, both entitled Xunzi kaoyi . Discrepancies are given at Kaoyi (Miao edition), 9a, The bookstore editions are called the Erzhe and Xishu editions, comprising a total of four different editions. These seem to have been fundamentally the same text since in his variorum, Qian Dian regularly contrasts zhuben, which refers to these four editions, with jianben, the Lü edition.

19. Discrepancies are noted in the Kaoyi (Miao edition), 9a, which suggests that all editions at his disposal read “fifty” in the Liu Xiang Preface.

20. Besides the three main Song editions and the four quoted in the Kaoyi to the Qian Dian edition, there is the Zhuantu huzhu Xunzi edition of whien a number of exemplars exist, including one at the East Asiatic Library of the University of California, Berkeley. Tnis work was a popular edition published by a Jianyang bookshop witn minor explanatory notes in the huzhu format by a contemporary scholar. It was published in a series with several other philosophic works and with several of the classics. These were undertaken in the late Song and early Yuan period. A colophon to the Laozi in this series is dated to the Qingding reign period (1260-1263). The work is thoroughly described by Xinyuan, Lu, Yigu tang xuba (1892 edition)Google Scholar, 9.2a–3b and by Pang Zongznou , Baoli tang Song ben shulu (1963 reprint), tzu lab. It is to be identified with the Fujian edition quoted by Wang Yinglin and is the basis of the Ming dynasty Shidetang edition. Cf. Bing, Ding, Shanben shushi cangshuzhi (1901 edition)Google Scholar, 15.2b. Other Song editions of less importance are described by Tingznou, Ruan, “Xunzi snulu” , Guoli Taiwan shifan daxue guowen yenjiuso jikan 5 (1961), 409–411Google Scholar.

21. At 5.4b-5b, in an appendix to Yuhai.

22. Tang Yuzheng in his Postface accepted the figure of 137. Chao Gongwu, Junzhai dustiu zhi 10.8a, noted that Xunzi must have been approaching a hundred when he went to Lanling.

23. Siku quansnu zongmu 19l.5b-7b.

24. Guoen, You, “Xun Qing kao” , Gushibian 4 (1933), 95Google Scholar.

25. Xian Qin zhuzi xinian, 437-438.

26. Reversals of tnis sort occur elsewhere in the Shi Ji and in fact there is another example of the identical change. The Pei Yin commentary, Shi Ji 1.4-5, quotes Xu Guang quoting Mozi giving the figure “15.” Sima Zhen quotes Mozi reading “50” and he notes the “popular editions” write “15” wnich is incorrect since, as ne points out, the text says that he is not quick in his wits, implying old age, so that “15” is obviously wrong. That both the Shi Ji and Preface read “50” might be taken to imply two independent reversals of the characters, an argument against the proposed corruption. It is, however, quite probable that the corruption of one text through transposition is reponsible for the change in the other. This has occurred in connection with the relative ages of Fuzi and Confucius. The Shi Ji (67.34) says that Fuzi was 30 years younger than Confucius while the Kongzi jiayu ([SBBY edition] 9.2a) says that Fuzi was 49 years younger than Confucius. The Soyin commentary notes this difference saying that the text of the Shi Ji has “30 years” whereas the Jiayu reads “49 years” (Shixue congshu edition, 18.2a). In consequence of the Soyin argument, the figure in the original text of the Shi Ji was later changed to read “49 years younger” like the Jiayu [so SBBY 67.12a and Bona pen edition 67.15b]. Some editions, then noting that the Soyin commentary was inconsistent with the text changed the Soyin commentary to read: “The Jiayu has '30 years younger than Confucius' wnich is not the same as the Shi Ji text which reads ‘49 years’.” [Also in the SBBY and the Bona pen editions.] Editors of the Jiayu, observing that the Shi Ji read “30 years” corrected the Jiayu to read “30 years” [e.g., Kambun Taikei edition]. The Mao Jin Jiayu reads “40 years” perhaps through carelessness combining the errors. This whole episode, as Liang Yusneng rightly observes, Shi Ji zniyi 28.14b-15b, is utterly nonsensical, out it does show that when texts conflicted, changes of the most incredible sort were possible.

27. Shumin, Wang, “Meng Xun liezhuan xin jiaozhu” , Kong Meng xuebao 13 (1967), 82Google Scholar.

28. There is a certain affinity in the use of contemporary and historical examples and in the combinations of stock examples between the Jiebi and the Mozi 1.8a, Lüsni cnunqiu 2.8a and Hanfeizi 9.3b and 10.2a which may be useful in dating all four.

29. Weiyu, Xu, Lüshi chunqiu jishi Beijing: 1955, 883, 1018, 894Google Scholar.

30. Conveniently summarized by Duyvendak, J. J. L., “The Chronology of Hsuntzu,” T'oung Pao 26 (1929), 91CrossRefGoogle Scholar.

31. Yantie lun 4.5b; Shi Ji 87.14.

32. Yaoji jieti ji qi dufa , Shanghai: 1925, 74Google Scholar, Liang nonetheless believes that Xunzi did live to see Si, Li become Cnancellor, Zhuzi kaoshi , Taipei: 1966, 60Google Scholar.

33. Yantie lun 9.8b quotes and paraphrases a passage concerning Zou Y an immediately preceding the biography of Xunzi.

34. Lu Wenchao rejects this reading and prefers to emend the text. Liu Shipei emends to Lu Shen . Hao Yixing accepts the reading Chunshen.

35. Guoxiang, Du, “Zong Xunzi de Chengxiang pian kan tade fashu sixiang” , in Du Guoxiang wenji , Beijing: 1962, 184–190Google Scholar.