2 Some of these translations are subject to discussion, e.g. ming gui (percipient ghosts, insight in ghosts, explaining ghosts), shang tong (conform upward, identification with above), and jian ai (universal love, indistinctive care…). For a discussion of the translation of these catchphrases, see also Johnston, Ian, The “Mozi”: A Complete Translation (Hong Kong: Chinese University Press, 2010), xxxiv–lxviGoogle Scholar. For the sake of consistency, I will translate jian ai as “care for all”, “inclusive care” or a combination of both. For other translations of jian ai, see Defoort, Carine, “Are the three ‘Jian ai’ chapters about universal love?”, in Defoort, Carine and Standaert, Nicolas (eds), The “Mozi” as an Evolving Text: Different Voices in Early Chinese Thought (Leiden and Boston: Brill, 2013), 35–67CrossRefGoogle Scholar.

3 See Yi, Yang 杨义, Mozi huanyuan 墨子还原 (Beijing: Zhonghua shuju, 2011), 24Google Scholar. For other Chinese presentations of Mozi's thought or early Mohism as basically a collection of ten central ideas, see e.g. Qichao, Liang 梁启超, Mozi xue'an 墨子学案, in Mozi daquan 墨子大全, ed. Jiyu, Ren 任继愈 and Guangxing, Li 李广星), vol. 26, 1–183 (Beijing: Beijing tushuguan chubanshe, 2004), 15Google Scholar; Kung-chuan, Hsiao (tr. Mote, Frederick), A History of Chinese Political Thought (Taipei: Caves Books, 1980)Google Scholar (this is the translation of Zhongguo zhengzhi sixiang shi, 2 vols, 1945 and 1946); Jiajian, Tan 谭家健, Mozi yanjiu 墨子研究 (Guizhou: Jiaoyu chubanshe, 1995)Google Scholar. In Mozi daquan, vol. 80, 11; Shenglong, Li 李生龍 (ed.), Xinyi Mozi du ben 新譯墨子讀本 (Taipei: Sanmin shuju, 1996), 5Google Scholar; Jiewen, Zheng 郑杰文, Zhongguo Moxue tongshi 中国墨学通史 2 vols (Beijing: Renmin chubanshe, 2006), 4–23Google Scholar. The emergence of this presentation of early Mohism deserves a study of its own.

4 Johnston, “Mozi”, xvii, xxxii. For other Western scholars, see e.g. Graham, Angus C., Disputers of the Tao: Philosophical Argument in Ancient China (Chicago: Open Court, 1989), 35, 226Google Scholar; Schwartz, Benjamin, The World of Thought in Ancient China (Cambridge, MA: Harvard University Press, 1985), 137–8Google Scholar; Cheng, Anne, Histoire de la pensée chinoise (Paris: Éditions du Seuil, 1997), 89Google Scholar; Knoblock, John, Xunzi: A Translation and Study of the Complete Works. 3 vols (Stanford: Stanford University Press, 1988–1994), vol. I, 58Google Scholar.

5 These two characters do not appear in the oldest preserved complete edition of Mozi, nl. in the Daozang (1447); they were added in 1832 by Wang Niansun, whose emendation has been accepted by Sun Yirang and consequently taken for granted by the scholarly community. Wang Zhong may have quoted the full fragment in 1790, but without making any comment. See Wang Niansun 王念孫, Mozi zazhi 墨子雜志, in Mozi daquan, vol. 14, 1–290, quoting p. 207: “The two characters gong and gu were missing from the old edition. I added them now on the basis of the text above and the ‘Fei gong’ chapters”. 舊本脫攻故二字. 今據上文及非攻篇補.

6 Except when specific fragments are commonly recognized by a number (e.g. Lunyu,1.2, Laozi, 24, and Mengzi 3A9), all quotes of Chinese masters refer to Lau, ICS Ancient Chinese Texts Concordance Series: the chapter number is given first, followed by a colon and then the page number and line number separated by a slash. Quotes from dynastic histories are to the Zhonghua shuju edition.

7 For an elaboration of what I mean by motto, see part 3 of this paper.

8 Hanshu 漢書. By Gu, Ban 班固 (Beijing: Zhonghua shuju, 1962, 30)Google Scholar. 1738 mentions a Mozi in 71 pian, as it still is today. See also Maeder, Erik W., “Some observations on the composition of the ‘core chapters’ of the Mozi”, Early China 17, 1992: 27–82CrossRefGoogle Scholar, esp. 29–34; and Graham, Angus C., Later Mohist Logic, Ethics, and Science (Hong Kong: Chinese University Press and London: School of Oriental and African Studies, 1978), 65Google Scholar.

9 For these and some other hypotheses on the threefold nature of the core chapters, see Karen Desmet, “All good things come in threes: a textual analysis of the three-fold structure of the Mohist ethical ‘core chapters’”, PhD diss., Katholieke Universiteit Leuven, 2007. The “evolution” scenario has lately become more popular than the “three sects” hypothesis. See e.g. Carine Defoort and Nicolas Standaert (eds), The “Mozi” as an Evolving Text: Different Voices in Early Chinese Thought, “Introduction: Different voices in the Mozi: studies of an evolving text”, 1–34, and Chris Fraser, “Mohism”, in Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Summer 2010 Edition).

10 See Defoort and Standaert, The “Mozi” as an Evolving Text, “Introduction”.

11 For the view that the Dialogue chapters are more original than the Core chapters, see e.g. Jiewen, Zheng 郑杰文, Zhongguo Moxue tongshi 中国墨学通史. 2 vols (Beijing: Renmin chubanshe, 2006), 4, 46Google Scholar. For a detailed rejection of this view, see Sixin, Ding, “A study on the dating of the Mo Zi Dialogues and the Mohist view of ghosts and spirits”, Contemporary Chinese Thought 42/4 (Summer 2011), 39–87CrossRefGoogle Scholar, esp. 48–50.

12 See Fraser, “Mohism”, see “Supplement to Mohism: text and authorship”; Ding, “A study on the dating of the Mo Zi Dialogues and the Mohist view of ghosts and spirits”, 73; A. Brooks, Taeko, “The Mician ethical chapters”, Warring States Papers 1, 2010, 100–18Google Scholar, esp. 115. Yoshinaga Shinjirō 吉永慎二郎, Sengoku shisōshi kenkyū: Juka to Bokka no shisōshiteki kōshō 戦国思想史研究－儒家と墨家の思想史的交渉 (Kyōto: Hōyū Shoten, 2004), 96 believes that the last Dialogue chapters can be dated to 338 bce.

13 Term suggested by Mark Csikszentmihalyi and Michael Nylan, “Constructing lineages and inventing traditions through exemplary figures in early China”, T'oung Pao 89, 2003, 69–99, esp. 62, n. 6.

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

15 See Defoort and Standaert, The “Mozi” as an Evolving Text: Different Voices in Early Chinese Thought, “Introduction”. A. Taeko Brooks, “The Mician ethical chapters”, 115, believes that “the triplet chapters had been rounded off, and the Mician doctrines officially fixed at ten” somewhat earlier than 262 bce.

16 Qingyuan, Lin 林清源, Jiandu boshu biaoti geshi yanjiu 簡牘帛書標題格式研究 (Taipei: Yiwen yinshuguan, 2004), 7Google Scholar; on titles, see also Jiaxi, Yu 余嘉锡, Muluxue fahui: Han “Gushu tongli” 目录学发微.含“古书通例” (Beijing: Zhongguo renmin daxue chubanshe, 2004), 200–04Google Scholar.

17 See Lin Qingyuan, Jiandu boshu biaoti geshi yanjiu, 7–9, 48–50. Three other types of titles are created by: (1) expressing the concrete object discussed in the texts; (2) repeating the first keywords of the text; and (3) quoting the first unit of a series of fragments.

18 Sun Yirang attributed the redaction of the Mozi to the imperial Han librarian Liu Xiang (77–76 bce), who was responsible for the order of the chapters and sections in various works, such as Xunzi, Guanzi and Zhan guo ce. There is, however, little proof that Liu Xiang carried out this redaction or gave the titles to the chapters of the Mozi. See Sun Yirang, Mozi jiangu, 653. Many scholars seem to follow this view. See for example Zheng Jiewen, Zhongguo Moxue tongshi, 202, 252, 289; and Graham, Angus C., Divisions in Early Mohism Reflected in the Core Chapters of “Mo-tzu” (Singapore: National University of Singapore, Institute of East Asian Philosophies, 1985), 17Google Scholar.

19 Boltz, William G., “The composite nature of early Chinese texts”, in Kern, Martin (ed.), Text and Ritual in Early China (Seattle: University of Washington Press, 2005) 50–78Google Scholar, at p. 59, and Maeder, “Some observations on the composition of the ‘core chapters’ of the Mozi”, 81–2.

20 This has been argued by Nicolas Standaert, “Problems with titles: naive questions starting from Mozi's idea of heaven”, paper presented at the Seventeenth WSWG Conference, Leiden, September 2003, 4–5. For the “Jian ai” triplet, see part 3 of this paper and Carine Defoort, “Are the three ‘Jian ai’ chapters about universal love?”, 36–9.

21 Mozi's answer in the “Lu Wen” fragment consists of no more than 78 characters.

22 The expression jie zang 節葬 does not occur in the “Jie zang” triplet, but the expression jie sang 節喪 occurs once (in ch. 25).

23 This may be corroborated by the relatively late date that various Mozi scholars attribute to these two triplets in comparison to the others. See e.g. Zheng Jiewen, Zhongguo Moxue tongshi, 126.

24 The suggestion that the attribution of the ten theses gradually took shape along with the editing of the Core chapters has been implied by Taeko Brooks, “The Mician ethical chapters” and Fraser, Chris, “Doctrinal developments in MZ 14–16”, Warring States Papers 1, 2010, 132–6Google Scholar.

25 The expression fei ming occurs twice, jie zang once and shang tong once (or twice). For more details, see Defoort and Standaert, The “Mozi” as an Evolving Text: Different Voices in Early Chinese Thought, “Introduction”, Table 4. Jian ai occurs eight (or nine) times in the rest of the Mozi, aside from the “Lu wen” fragment.

26 尊天事鬼 is used in 4: 5/1, 9: 12/22, 26: 43/11, 35: 59/7–8, 48: 107/27, 48: 111/7, 49: 111/23, and 49: 114/9. This expression 事鬼神 occurs alone in three more instances but never in the extant “Ming gui” chapter.

27 The Daozang edition has jong jie 用節. Apparently Bi Yuan corrected this to jie yong but without leaving any comment. See Bi Yuan 畢沅, Mozi zhu 墨子注, in Mozi daquan, vol. 11, 325. The field follows him: see e.g. Sun Yirang, Mozi jiangu, 454; Yujiang, Wu 吳毓江, Mozi jiaozhu 墨子校注. 2 vols (Beijing: Zhonghua shuju, 2006)Google Scholar, 704, 716 n. 46; and Li Shenglong, Xinyi Mozi du ben, 437.

28 Chris Fraser has also suggested that some later Mohist authors may have developed slogans for key positions presented in earlier chapters. See Fraser, Chris, “Thematic relationships in Mozi 8–13”, Warring States Papers 1, 2010, 137–42Google Scholar, quoting 138.

29 In the “Tian zhi” triplet 尊天 and 事鬼 occur once, together in the cluster in ch. 26 (quoted in Table 3).

30 The character ai does not appear in the Daozang edition but was added in 1783 by Bi Yuan, Mozi, 36: “Originally this character was lost. I added it on the basis of the meaning”. 舊本脫此字. 以意增. It deserves notice, however, that the Qunshu zhiyao 群書治要 (from 631) quotes ch. 4 (p. 450) with the expression jian ai. The indirect evidence of the Mozi in this and other early sources needs further investigation.

31 Here and elsewhere in the Mozi, this expression is clustered with “benefit each other 交相利”. This combination first occurs in “Inclusive care, middle”.

32 E.g. Xunzi about Mozi being “blinded by utility” 蔽於用 or “neglecting gradations and rankings” 僈差等, are not mottos but rather critical labels. Nor are the expressions “right wrong” (shi fei 是非) or “hard white” (jian bai 堅白), which are often associated with (later) Mohism by other masters, as in Zhuangzi, 33.

33 “Guang(ze)” 廣(澤) I,1. See also Weisong, Shui 水渭松, Shizi duben 尸子讀本 (Taipei: San min shuju, 1997)Google Scholar, 111, 114. Since the point is that all of these concepts amount to the same, I have translated 公 as “public good” and not as “duke”.

34 Hence I did not include Hanfeizi's complaint that “The learned men of today, on counselling the lord of men, all persuade him to discard the profit-seeking mind and follow the way of mutual love”, 今學者之說人主也，皆去求利之心，出相愛之道, which has pernicious results for the state. See ch. 46 “Liu fan” 六反, Liao, W.K., The Complete Works of Han Fei Tzu. 2 vols (London: Arthur Probsthain, 1959), vol. II, 239–40Google Scholar.

35 Or perhaps 11 times: in one occurrence (ch. 4 “Fa yi”) ai was added by Bi Yuan (see note above). Aside from the one occurrence in ch. 16 (“Jian ai, xia”) and one in the “Lu wen” fragment, the expression occurs four times in ch. 28 “Tian zhi, xia”, twice in ch. 44 “Da qu”, and twice in ch. 46 “Geng Zhu”.

36 For jin ai (3 times), see Graham, Later Mohist Logic, Ethics, and Science, 448–9, 250–1; for zhou ai (once), see Johnston, The “Mozi”, 632. Jian ai occurs only twice in the Mohist canon in rather incomprehensible fragments. See Johnston, The “Mozi”, 592, 616.

37 Translators tend to add the word “love” to their translation after 兼 (jian) – sometimes in brackets – where there is no 愛 (ai) in the original text. See, e.g., Johnston, The “Mozi”, 139, 143, 145; and Yi-pao, Mei (trans.), The Ethical and Political Works of Motse: Translated from the Original Chinese Text (London: Arthur Probsthain, 1929)Google Scholar, 166, 170, 172.

38 Yoshinaga, Sengoku shisōshi kenkyū, 77–90, 96–108 believes that Mozi thought in terms of “care” (ai), and that only by the last quarter of the fourth century bcejian ai had become a slogan or current expression among later Mohists and their critics (such as Mencius).

39 Shinjirō, Yoshinaga 吉永慎二郎, “Jian ai shi shenme: Jian ai gainian de xingcheng he fazhan” 兼爱是什么 – 兼爱概念的形成和发展, Ha'erbin shizhuan xuebao 哈尔滨师专学报, 1999/4, 31–4Google Scholar, quoting 31 and Yoshinaga, Sengoku shisōshi kenkyū, 74–84.

40 For a fuller discussion, see Defoort, “Are the three ‘Jian ai’ chapters about universal love?” Most scholars assume that the three “Jian ai” chapters all make the same point. A few scholars have traced an evolution in them, but often different from what I have argued. Some attribute a full-blown defence of “inclusive care” to the first “Jian ai” chapter, which was watered down or compromised in the following two chapters. See e.g. Weixiang, Ding 丁为祥, “Mojia jian ai guan de yanbian” 墨家兼爱观的演变, Shaanxi shifan daxue xuebao 陕西示范大学学报 1999/4, 70–6Google Scholar and Brooks, “The Mician ethical chapters”, 111. Yoshinaga, “Jian ai shi shenme” traces an evolution from a moral stance to an increasingly political and utilitarian view. Kazutaka, Sakai 酒井和孝, “Makki Boku no ken'ai shisō: Ryoshi Shunjū o baikai toshite” 末期墨の兼愛思想 ——「呂氏春秋」を媒介として, Tetsugaku 32, 1980, 101–14Google Scholar, esp. 101–5 traces an evolution from “care” towards “benefit”.

41 Sato, Masayuki, “The idea to rule the world: the Mohist impact of Jian 兼 in the Xunzi”, Oriens Extremus 48, 2009, 21–54Google Scholar, quoting 31, my italics. I believe that the date remains unsure and there may not have been a common origin. Sato's argument relies on the claim that the Shanghai manuscript (by the contemporary editor), titled “Guishen zhi ming”, belongs to the Mohist school. I am not convinced by this claim nor by the implications derived from it.

42 “Care” does not seem to have been of any concern in the redaction of the Defence chapters (chs 52–71). I follow the dating of these chapters (mid-fourth–third centuries) as proposed by Fraser, “Mohism” (see “Supplement to Mohism: text and authorship”). But their exact chronology remains obscure and is not crucial for my argument. Fraser dates the Dialogues and the canon to c. late fourth and third century bce, and the Opening chapters to the c. third century bce. Taeko Brooks, “The Mician ethical chapters”, 107, 117, tentatively calls the latter “singlets” and dates them, in reverse order, from 270 to 250 bce.

43 For the Core chapters, see “Shang xian, zhong” (ch. 9), “Shang tong, xia” (ch. 13), “Feigong, zhong” (ch. 18), “Jie yong, zhong” (ch. 21), and “Fei ming, shang” (ch. 35). For the “Fa yi” (ch. 4), see also Standaert, “Heaven as a standard”. On the basis of its content, Standaert speculates that it might be situated after “Tian zhi, shang + xia” and before “Tian zhi, zhong”.

44 For evolutions from the Core chapters to the Dialogues, see Chris Fraser, “The ethics of the Mohist Dialogues”, in The “Mozi” as an Evolving Text: Different Voices in Early Chinese Thought. For the canon, see Graham, Later Mohist Logic, Ethics, and Science, 246–52, 270. I find most of these fragments too corrupt to be confidently interpreted.

45 It must be said that in Yilin 意林 (from 786), ch. 14 is quoted with two mentions of jian ai. Indirect evidence of Mozi deserves further study.

46 兼相愛 occurs 13 [+1] times in the book (three times in chapter 14, five times in chapter 15, twice [+1] in chapter 16, once in chapter 26, and twice in chapter 35); 兼而愛 occurs five times (twice in chapter 4, once in chapter 9, once in chapter 26, and once in chapter 28); 兼天下而愛 occurs twice (in chapter 27). See also Sato, “The idea to rule the world”, 38–40.

47 See e.g. Csikszentmihalyi and Nylan, “Constructing lineages and inventing traditions through exemplary figures in early China”, 62–4.

48 Chapter 39 is the only remaining of the “Fei Ru” duplet. Its position as the eleventh dogma has been considered, but I believe there are good reasons to consider it separately from the Core chapters and the Dialogues, two groups of chapters with which it shares characteristics. The second part of ch. 39 is probably from the third century ce. See e.g. Karen Desmet, “All good things come in threes”, 224–43.

49 Aside from this chapter and some fragments in the Dialogues, the Mozi never explicitly criticizes the Ru. See Robins, Dan, “The Moists and the gentlemen of the world”, Journal of Chinese Philosophy 35/3, (September 2008), 385–402CrossRefGoogle Scholar, esp. 392; and Fraser, “The ethics of the Mohist Dialogues”.

50 Yoshinaga, Sengoku shisōshi kenkyū, 107–08.

51 See e.g. Gu Jiegang 顧頡剛, “Cong Lüshi chunqiu tuice Laozi zhi cheng shu niandai” 從呂氏春秋推測老子之成書年代, Gushibian, IV, 462–520, esp. 493–4; Kwong-loi, Shun, Mencius and Early Chinese Thought (Stanford: Stanford University Press, 1997), 44–7Google Scholar; Defoort, Carine, “The profit that does not profit: paradoxes with lì in early Chinese texts”, Asia Major 21/1, 2008, 153–81Google Scholar, esp. 174–5.

52 The Qing scholar, Wang Zhong 汪中 (1745–94) pointed out that Mencius was unfair and tried to inflate the difference between him and Mozi. See his Shu xue 述學, “Mozi xu”, quoted by Hsiao, A History of Chinese Political Thought, 231.

53 See e.g. Mencius, 6A14, 7A9, 7A46 and 7B1.

54 Sato, “The idea to rule the world”, 42; and Knoblock, Xunzi, vol. II, 127.

55 Ch. 25 is considered a miscellaneous chapter. Knoblock, Xunzi, vol. I, 31–2 dates it after 238 bce, when Xunzi was demoralized and dismissed from office after his patron's assassination. The influence of the Mozi on Xunzi has been generally attested, beginning in the Shiji, in which it is said that the latter “enhanced the behaviour and practices, successes and disruptions of the Ru, Mo and Daode” 推儒墨道德之行事興壞 to compose his own writing (Shiji 74: 2348). For Xunzi's treatment of jian ai and the “technique of inclusiveness” 兼術, see also Sato, “The idea to rule the world”, 47–8.

56 Knoblock, Xunzi, vol. III, 170.

57 Sato, “The idea to rule the world”, 42, his italics.

58 Hsiao Kung-chuan, A History of Chinese Political Thought, 234, n. 41 pointed out (in 1945) that Xunzi “never touches on Mo Tzu's universal love, in this he is greatly at variance with Mencius”.

59 For some general criticism of Mo, see e.g. chs “Xiu shen”, “Ru xiao” and “Cheng xiang”. See Dubs, Homer H., The Works of Hsuntze (Taipei: Confucius Publishing Co., 1983), 128, 154–5Google Scholar; and Knoblock, Xunzi, vol. III, 177. For Xunzi's criticism of Mozi, see Zheng Jiewen, Zhongguo Moxue tongshi, 129–40.

60 In “Ru xiao”, the vulgar Ru are said to resemble Mozi; see Dubs, The Works of Hsuntze, 154–5.

61 See ch. “Yue lun”; see Knoblock, Xunzi, vol. III, 80–84.

62 See ch. “Wang ba”; see Knoblock, Xunzi, vol. II, 158.

63 They consist of three characters and are not commonly used. Therefore, I do not count them as Mohist mottos.

64 Knoblock, Xunzi, vol. III, 22. The same view was expressed by Xunzi's earliest commentator Yang Liang (ninth century) and followed by Sun Yirang, Mozi jiangu, 747 and Tan Jiajian, Mozi yanjiu, 52. For Knoblock's translations of the preceding quotes, see Knoblock, Xunzi, vol. I, 233 (ch. 6), vol. II, 128–9, 130 (ch. 10); vol. III, 102 (ch. 21).

65 Knoblock, Xunzi, vol. III, 303, n. 76. He refers to, respectively, shang tong and jian ai. For a similar interpretation of this line, see Xianqian, Wang 王先謙 (ed.), Xunzi jijie 荀子集解, 2 vols (Beijing: Zhonghua shuju, 1996), 319Google Scholar.

66 See e.g. Tan Jiajian, Mozi yanjiu, 49, using only Mencius and some Xunzi fragments without jian ai to argue that the greatest opposition that the Mohists “encountered was from the Confucian lineage. The target that they first of all violently attacked as inclusive care”. For his interpretation of Xunzi fragments (from chapters 6 and 17 mentioned above), see his Mozi yanjiu, 51–2.

67 See ch. 50; Liao, The Complete Works of Han Fei Tzu, vol. II, 298–310. Hanfeizi never exclusively associates Mo with jian ai. His criticism of Mozi (or the Mohists) concerns rather their interest in useless technical skills, their useless political advice, their simple rhetoric, etc.

68 E.g. Fengsu tongyi, “Meng Ke” 孟柯 and Fayan, “Wuzi” 吾子.

69 Mair, Victor, Wandering on the Way. Early Taoist Tales and Parables of Chuang Tzu (New York: Bantam Books, 1994), 307Google Scholar; Graham, Angus C., Chuang-tzu: The Inner Chapters (London: George Allen and Unwin, 1986), 240Google Scholar. Graham (p. 234) identifies this chapter as Yangist and dates it in the Qin–Han interregnum. McCraw, David, Stratifying “Zhuangzi”: Rhyme and other Quantitative Evidence (Taipei: Academia Sinica, 2010), 26–32Google Scholar agrees with this date.

70 Most Zhuangzi chapters, except chs. 29, 32 and 33, criticize Mo along with other masters, for their futile debates and aspirations.

71 As is the expression ming gui 明鬼 (the triplet's title), the translation of you gui 右鬼 is contested. Major, John, Queen, Sarah, Meyer, Andrew Seth and Roth, Harold D., The “Huainanzi”: Liu An, King of Huainan; A Guide to the Theory and Practice of Government in Early China (New York: Columbia University Press, 2010), 501Google Scholar translate 右 as “esteem”, as do the Chinese commentaries; Graham, Disputers of the Tao, 380 translates it as “assist”. In the Lun-heng: Part I Philosophical Essays of Wang Ch'ung; Part II Miscellaneous Essays of Wang Ch'ung, New York: Paragon Book Gallery, 1962, 162 (ch. 20)Google Scholar Alfred Forke translates 右鬼 as “the help of the spirits”; on p. 461 (ch. 83) he translates the same expression as “honour the ghosts”, quoting Wang Chong: “If ghosts were not the quintessence of the dead persons, they would have no knowledge if one honoured them” 使鬼非死人之精也，右之未可知. I have chosen the literal translation of “supporting ghosts”, which resembles “serving ghosts” 事鬼 of the “Lu wen” fragment and may have amounted to esteeming, honouring and assisting them.

72 Perhaps Ban Gu, like most other Han scholars mentioned in part 6 of this paper, does not reject jian ai per se, but only the Mohist extension of the idea: “They extend the idea of inclusive care to the point that they don't know how to distinguish between kin and stranger” 推兼愛之意而不知别親疏. See Hanshu 漢書, by Gu, Ban 班固 (Beijing: Zhonghua shuju, 1962), 30.1738Google Scholar.

73 Mair, Wandering on the Way, 336; Graham, Chuang-tzu: The Inner Chapters, 276. Neither considers these two slogans as book titles. Guying, Chen 陳鼓應 (ed. Zhuangzi) jinzhu jinyi 莊子今注今譯 (Hong Kong: Zhonghua shuju, 1995), 869Google Scholar considers “Fei yue” as a title.

74 泛 (literally “overflow”), also written 氾 and 汎, is a not very current object quantifier, functioning like jian 兼. See Harbsmeier, Christoph, Aspects of Classical Chinese Syntax (Malmo: Curzon Press, 1981), 55Google Scholar.

75 See e.g. Mair, Wandering on the Way, 338. This chapter is generally dated just before or during the Han and ascribed to the final editor of the book, who was clearly not Zhuang Zhou. Graham, Chuang-tzu: The Inner Chapters, 257, dates it pre-Han; Roth, Harold, “Who compiled the Chuang Tzu?”, Chinese Texts and Philosophical Contexts, ed. Rosemont, H. Jr (La Salle: Open Court, 1991), 79–128Google Scholar, esp. 121–3 dates it to the former Han.

76 In this chapter, the author not only praises Mozi as one with “a true love of the world”, and “a knight with abilities”, but he also takes part in the shi fei evaluation of others. Chs 29 and 32 (Mair, Wandering on the Way, 306 and 326) are also somewhat atypical, with their rather positive view on Mozi (but less so than ch. 33). For the treatment of Mo in the remainder of the Zhuangzi, see e.g. ch. 2 (Mair 15), ch. 8 (Mair 76), ch. 10 (Mair 87), ch. 11 (Mair 94), ch. 12 (Mair 117), ch. 14 (Mair 141), ch. 22 (Mair 222), ch 24 (Mair 243, 247), ch. 29 (Mair 307).

77 “Roots” 本 are important in the Mozi, as opposed to “twigs” 末, but they are not spoken of as being “strengthened” 彊. The expression “Moderate expenses and elevate the roots/basics” 節用尚本 is attributed to the literati (wen zhe 文者) in e.g. Yantielun, 2 “Li geng” 力耕.

78 For these opponents' views in “Jian ai, zhong” and “Jian ai, xia”, see Defoort, “Are the three ‘Jian ai’ chapters about universal love?” For other masters finding “inclusive care” too demanding, see e.g. Zhuangzi, 33 (Mair 338) and Lun heng, 83 (Forke, Lun-heng: Part I, 461). Wang Chong sees it as the cause of the dismay of Mohism (墨法廢者 […] 墨之法議難從也), but for him the difficulty lies mainly in the contradiction between honouring ghosts and yet not burying them well.

79 “Probably Mo Di was a minister in Song, he was good at defense and practiced ‘moderating expenses’. Some say he lived in the same time as Kongzi; others say he was later”. 蓋墨翟，宋之大夫，善守禦，為節用.或曰並孔子時，或曰在其後. (74: 2350).

80 See Huainanzi 16; Major et al., The “Huainanzi”, 637. This motto is also attributed once to Mozi in the Shizi, preserved in reconstructed fragments in Beitang shuchao and Yiwen leiqu, both in “Le bu” 樂部 about Mozi playing the flute despite his rejection of music. See Yi, Li 李轶 and Shoukui, Li 李守奎, Shizi 尸子 (Harbin: Heilongjiang Renmin chubanshe, 2003), 145–6Google Scholar. The same story occurs in the Lüshi chunqiu 15/7.5 “Gui yin” 貴因.

81 The two Ru are Zengzi and Kongzi. I follow Major et al., The “Huainanzi”, 651 (and Liu Wendian) in correcting the second Zengzi to Kongzi. For chs 11 and 21 respectively, see Major et al., The “Huainanzi”, 408 and 864.

82 See 18/7.3; Knoblock, John and Riegel, Jeffrey, The Annals of Lü Buwei: A Complete Translation and Study (Stanford: Stanford University Press, 2000), 466Google Scholar.

83 His evaluation is not totally consistent. Briefly, he admires or sympathizes with Mozi's personality, his hard work, his endurance of political frustration and lack of public support, his worries about the world, and his lack of noble ancestors. Wang Chong also criticizes the Mohist view on the dead, on spirits, and on simple burials.

84 There is a debate in e.g. ch. 67 薄葬篇 “Simple burials” and ch. 83 案書篇 “Critic on books” in, respectively, Forke, Part II, Miscellaneous Essays of Wang Ch'ung, 369–75 and Part I, Philosophical Essays of Wang Ch'ung, 461.

85 This emendation was first suggested by Sun Yirang, Mozi jiangu, 745. More drastic is simply to quote the “corrected” text, as does Tan Jiajian, Mozi yanjiu, 55.

86 Both steps are, understandably, inspired by the idea that Mozi stands for jian ai. See e.g. Li Shenglong, Xinyi Mozi du ben, 5; see also Knoblock and Riegel, The Annals of Lü Buwei, 433, but they do not consequently interpret jian as jian ai, so that they end up attributing the ideal of “wholeness” to Mo Di.

87 Shui Weisong, Shizi duben, 112, n. 1. He also translates the fragment accordingly (p. 114). See also Li Shenglong, Xinyi Mozi du ben, 5.

88 For Later Mohists the last expression clearly referred to “loving all others”, as opposed to “horse riding”, which means riding at least one horse: “if one does not love all around, then on the basis of this, one does not love others. But for horse riding, one does not have to ride horses all around to be horse riding”. 不周愛, 因為不愛人矣. 乘馬不待周乘馬,然後為乘馬也. See Graham, Later Mohist Logic, Ethics and Science, 491 (NO 17); and Johnston, The “Mozi”, 632–3 (45.8).

89 Liao, The Complete Works of Han Fei Tzu, vol. II, 280.

90 There are various translations of this fragment. I have mainly followed Chen Guying, Zhuangzi jinzhu jinyi, 347–9, who follows earlier scholars in his reading of 物 (p. 348, n. 5) and in reading 幾乎 as “dangerous” (p. 348, n. 6). For other translations, see also Mair, Wandering on the Way, 125–6; Watson, The Complete Works of Chuang Tzu, 149; Ware, J.R., The Sayings of Chuang Tzu (Taipei: Confucius Publ. Co., 1983), 156Google Scholar. McCraw, Stratifying “Zhuangzi”, 37, wrongly attributes this statement to Laozi. This chapter is generally considered relatively late (Qin–Han). See Graham, Chuang-tzu: The Inner Chapters, 257 and McCraw, Stratifying “Zhuangzi”, 26–32.

91 On the conversations between Laozi and Confucius in the Zhuangzi, see McCraw, Stratifying “Zhuangzi”, 35–7.

92 See also Rickett, W.A., Guanzi: Political, Economic, and Philosophical Essays from Early China, 2 vols (Princeton: Princeton University Press, 1985 and 1998), vol. I, 110Google Scholar. He dates “Li zheng” to the mid-third century bce (p. 100). Brooks, A. Taeko, “Mwòdž 14–16 兼愛 ‘Universal Love’”, Warring States Papers 1, 2010, 129–31Google Scholar, 131 dates it slightly later than 311 bce: “Such was the world's final argument against Universal Love. The danger of Mician antiwar propaganda was precisely that it might succeed in one's own state”.

93 See also Liao, The Complete Works of Han Fei Tzu, vol. II, 230; Jue, Zhang 张觉, Hanfeizi quanyi 韩非子全译, 2 vols (Guiyang: Guizhou renmin chubanshe, 1995), 626–7Google Scholar.

94 See also Liao, The Complete Works of Han Fei Tzu, vol. II, 239–40.

95 A criticism of “care” as policy is dominant (although not exclusive) in the Hanfeizi. In ch. 46, he does not have the current view of early Mohism in mind because he also ascribes to them the rejection of “a profit-seeking heart”. Of course, there may have been thinkers influenced by Mohism who rejected “benefitting oneself”. For such a possibility, see Defoort, “Mohist and Yangist blood in Confucian flesh: the middle position of the Guodian text Tang Yu zhi dao”, Bulletin of the Museum of Far Eastern Antiquities 76, 2004, 44–70, esp. 60–61.

96 See Rickett, Guanzi, vol. I, 140–41. He dates “Ban fa” to the late fourth/early third century bce, and “Ban fa jie” in the Han (pp. 136–7).

97 See Li Yi and Li Shoukui, Shizi, 71 (preserved in Taiping yulan, juan 81), discussing Emperor Shun.

98 “Shen ying”, see Knoblock and Riegel, The Annals of Lü Buwei, 442 (18/1.5).

99 Hanshu 30.1738 attributes the now lost 隨曹子 in six chapters to a disciple of Mozi. The quote contained in the Tang anthology 意林 Yi lin (compiled by Ma Zong 馬總 in 786) describes the sage's behaviour as “inclusively caring for all people” 兼愛萬民 because he does not give priority to his own posterity: “he does not cut off non-kin, and does not allow his own posterity to inherit” 有疏而無絕，有后而無遺 (quoted in Tan Jiajian, Mozi yanjiu, 54).

100 “Jun zheng”, see Yates, Robin, Five Lost Classics. Tao, Huang-Lao, and Yin-Yang in Han China (New York: Ballantine Books, 1997), 65Google Scholar adding that this is “one of the fundamental concepts of Mohism” (p. 226, n. 64).

101 See Chengyuan, Ma 馬承源 (ed.), Shanghai bowuguan cang Zhanguo Chu zhushu 上海博物館藏戰國楚竹書. 8 vols (Shanghai: Shanghai guji chubanshe, 2001–2008), vol. 5, 251Google Scholar.

102 See e.g. “Yi tong” and “Daoshu”. For the latter, see Svarverud, Rune, Methods of the Way. Early Chinese Ethical Thought (Leiden: Brill, 1998, 176, n. 77)Google Scholar, who calls jian ai “one of the basic principles of the Mohist ethical principle”. On Mohist influence on the Xinshu, see also pp. 169–70, 199–215 and 282–9.

103 “Jian er ai zhi” in “Qian guai”. See Wenjin, Wang 王文錦 (ed.), Da Dai liji jiegu 大戴禮記解詁 (Beijing: Zhongghua, 1998), 157Google Scholar.

104 In ch. 5, “Dao de”.

105 “Shencha minghao”. Yu, Su 蘇輿 (ed.), Chunqiu fanlu yi zheng 春秋繁露義證 (Beijing: Zhonghua shuju, 1996), 289Google Scholar suggests the emendation of 兼愛 to 兼受.

106 “Taixuanli.” See Nylan, Michael, The Canon of Supreme Mystery (New York: SUNY, 1993), 430–31Google Scholar. Nylan points to the “typically Han confusion of ‘graded’ and ‘ungraded’ love (ren and ai)” (p. 615, n. 38).

107 In ch. 22, “Jiu bian”.

108 In ch. 1, “Ren ben”.

109 Hanshu, 58: 2616. See also Wallacker, Benjamin, “Han Confucianism and Confucius in the Han”, Ancient China: Studies in Early Civilization, ed. Roy, David and Tsien, Tsuen-hsuin (Hong Kong: The Chinese University Press), 1978, 215–28Google Scholar, esp. 221–2.

110 For other associations with 仁, see e.g. Xinshu “Daoshu” and Taixuanjing “Taixuanli”. For 兼愛無私, see e.g. the sources quoted above from “Jingfa”, Xinshu, Wenzi and, very closely, “Cao Mo zhi chen” (兼愛萬民，而亡有私也).

111 See Wallacker, “Han Confucianism and Confucius in the Han”, 222, following Shigemasa, Fukui 福井重雅, Kandai Jukyō no shiteki kenkyū: Jukyō no kangakuka o meguru teisetsu no saikentō 漢代儒教の史的研究: 儒教の官學化をめぐる定說の再檢討 (Tokyo: Kyūko Shoin, 2005)Google Scholar.

112 Of course, a link with the ten theses was one of my own selection criteria, but I have a strong impression that even a broader search (without this last criterion) would have brought out roughly the same scope of mottos.

113 Many Chinese scholars do this. See e.g. Zheng Jiewen, Zhongguo Moxue tongshi, vol. I, 1–23, 41–57; Li Shenglong, Xinyi Mozi du ben, 2–3.

114 Nylan, Michael, “A problematic model: The Han ‘orthodox synthesis’. Then and now”, in Kai-wing, Chow, On-cho, Ng and Henderson, John (eds), Imagining Boundaries. Changing Confucian Doctrines, Texts, and Hermeneutics (New York: SUNY Press, 1999), 17–56Google Scholar, quoting p. 31.

115 Nylan, Michael, “Kongzi and Mozi, the Classicists (Ru 儒) and the Mohists (Mo 墨) in Classical-era thinking”, Oriens Extremus 48, 2009, 1–20Google Scholar, pp. 4–7 list 27 instances in which Mozi and Kongzi are mentioned together.