² I have used the term Shang shu rather than Shu jing because this text is not in the extant Classic. There are many quotations from the Shu in the early philosophic texts which are not in either the modern script (jinben) or ancient script (guben) versions of the text and there may have been different collections of these ancient documents. They were believed to be contemporaneous ancient documents and divided chronologically according to the period from which they were believed to have originated; our text was thus a Yin (i.e. Shang)shu

³ The reference to this and other texts which cite versions of this text are given in the bibliography following the table (p. 539). Kong Anguo in his commentary to the Mozi follows the Guo yu in taking Tang shi as the correct title

⁴ Mozi 2/10a; Mengzi 1A.2 (l/3b). Titles given in the form X/Ya refer to the juan/page number of the Sibu congkan editions published in Shanghai in the 1930s

⁵ See Zhang, Xinzheng, Weishu tongkao (Shanghai: Commercial Press, 1954), 126–98, esp. 183–4Google Scholar

⁶ This occurs in two contexts; See 4/2b–3a, 7/6a

⁷ These references follow the table (pp. 535–9). The Diwang shiji is no longer extant. Thus, these six versions are quotations of the Diwang shiji culled from later sources. Other texts which include references to the story of the drought in language that does not derive directly from the text I am considering include the Zhuangzi 6/26a (bian 17), Guanzi 22/9b (bian 75), the Shizi (Jun zhi). Commercial Press, Shanghai, 1926, 16a, and the Zhushu jinian 1/22b

⁸ The text used for the Tang gao is that of Karlgren, B. in The Book of Documents (Bulletin of the Museum of Far Eastern Antiquities, XXII, 1950)Google Scholar

⁹ These numbers are the section numbers in the translation which follows as well as those in the appended table. See also notes 20 and 21 for more detailed comparison of the two texts

¹⁰ (2)–(3) may be a conflation, see p. 528 below and note 13

¹¹ The Diwang shiji draws from many of the same sources as the author of the Guben Shany shu; thus its author Huangfu Mi was considered one of the possible forgers of the Guben Shang shu. See Zhang Xinzheng, p. 169

¹² Mengzi 1A.4 (l/6b)

¹³ The Xunzi and the Shuo Yuan include (2)–(3), an invocation questioning the rectitude of Tang's government, but not the story of the self-sacrifice. This may be a Confucian transformation of the drought story which appears in the Diwang shiji as a conflation. See below, p. 528 for discussion

¹⁴ See for example (8): the pre-classical particle wei appears in two forms and is omitted in the Lun yu; yu‘I’ (which Qu Wanli uses to evaluate the date of the Jin teng, see p. 531 below) occurs in two forms; gan zhao occurs in some texts but not others; and, most important, the nomenclature for the gods to whom the announcement is addressed is different in every text and given in two further forms in the Baihutong quotations of the Lun yu

¹⁵ I have followed the zheng of Diwang shiji B, the Xunzi, and the Guo yu, but ‘passions not restrained’ is also possible

¹⁶ The six lines of text in (3) are the six faults of which Tang accuses himself in the Lun heng—see (7), but the first two lines can be read as one, and elsewhere the Lun heng refers to ‘five faults’ (18/8a, 15/10a)

¹⁷ Sou shenji, Lun heng, Lushi chunqiu and Guo yu simply refer to Tang praying personally at the Mulberry Grove here, but the first three of these include the preparation and self-sacrifice in (14) which is also the section in which the Mozi refers to the self-sacrifice

¹⁸ I have followed Diwang shiji A, sheng, since this appears to refer to Tang himself; xuan was the colour of the sun-bird and the Shang kings, see p. 529 below

¹⁹ The Baihutong quotes the Lun yu as huang tian shang di (7/6a) and also as huang wang hou di (4/2a), thus the extant Lun yu version is possibly corrupt

²⁰ I have included (11) in the translation because it is quoted in the Lun yu as well as the Mozi. (9) does not occur in any other text and although its form is much like that of a Shang announcement, ‘Now, Heaven has caused a great drought; it is upon myself’, since it only occurs in the Mozi which omits the first part of the text in which there is reference to the drought, I assume it is an insertion. (10) occurs only in the Mozi and the Tang gao and is the only part of the text to introduce a positive virtue (shan) rather than simply faults to be allayed (and the curse lifted)—see p. 530 for discussion of zui

²¹ (13) is a comment by Mozi. (14) is equivalent to (6), but the order has been reversed—see note 17. Although the order of the Mozi text differs from that of the Diwang shiji and it is six or seven centuries earlier, it can with the exception of (11) which I have included in my translation readily be construed as a philosopher's derivation from an original document very similar to Diwang shiji A

²² See Chen, Mengjia, Yinxu buci congshu , Peking, 1956, 96–7Google Scholar

²³ Zhang, Bingquan, Xiaotun di er ben: Yinxu wenzi: bing bian (Tabei: Academia Sinica, 1957), vol. 1, pt. 1, no. 63Google Scholar

²⁴ Frank, H. Chalfont and Roswell, H. Britton, The Couling-Chalfont collection of inscribed oracle bones, Shanghai: 1935, no. 1811Google Scholar

²⁵ See my ‘Sons of suns: myth and totemism in early China’, BSOAS, XLIV, 2, 1981, 290–326Google Scholar

²⁶ See Lao Gan , ‘Shi zi de jiegou ji shiguan de yuanshi zhiwu’, Dalu zazhi , 14. 3 (1957), 1–4

²⁷ Chen Mengjia ‘Shangdai de shenhua yu wushu,’, , Yanjing Xuebao , 20 (1936), 485–576; Edward, H. Schafer, ‘Ritual Exposure in Ancient China’, HJAS, 14, 1957, 130–84Google Scholar

²⁸ Zhang Bingquan, Bing Bian 11–20 are a set of five almost identical plastrons in which series of divinations are made about the king's toothache and a military campaign. The divinations are formally integrated on the plastron. Although most scholars have assumed that such pairing of questions is accidental, it is possible that the king's tooth ached (i.e. he was cursed) because of an ill-advised military campaign. For a translation of this series of inscriptions, see David N. Keightley, Sources of Shang history (Berkeley, 1978), 77–80. This idea was suggested to me by a paper delivered by David Nivison at SOAS in November 1978 in which he suggested that the divinations concerning weather and the king's illness in Bing Bian 334–5 were related. If this approach is valid, it provides a Shang precedent for our text and the Jin teng. In any case, the spirits' curses could fall upon the king's person just as they fell upon his lands and bis activities and the king sought to forestall such curses and gain favour by his offerings

²⁹ For discussion of the term yi ren, see Hu Houxuan , ‘Shi yu yi ren’, Lishi yanjiu , 1 (1957), and ‘Chong lun yu yi ren- “wenti”’, Guwenzi yanjiu , 6 (1981), 15–33

³⁰ See ‘Sons of suns’, 305–6

³¹ Frank H. Chalfont, The Couling-Chalfont collection (Ku), no. 1957

³² Chengzuo, Shang, Yin qi yi can (Peking: 1933), 525Google Scholar

³³ David Keightley has speculated that by the late Shang, misfortunes were increasingly interpreted as the result of human failings, i.e. ritual failings (‘Shang divination: the magicoreligious legacy’, a paper prepared for the Workshop on Classical Chinese Thought, held at Harvard, 2–13 August 1976, 15). It is also possible that moral failings had come to play a role, but there is no evidence of this

³⁴ SeeAllan, S., The heir and the sage: dynastic legend in early China (San Francisco, 1981), p. 8, n. 9, for the dating of these texts to the early Western ZhouGoogle Scholar

³⁵ Bernard, Karlgren, The Book of Documents, p. 37, v. 9Google Scholar

³⁶ See Da Gao, vv. 2–6. (See also Luo Gao, ibid., 51–2, w. 1–5, for auspicious oracles in the time of Cheng Wang)

³⁷ Chang Tsung-tung (Der Kult der Shang Dynastie im Spiegel der Orakelinschriften, Wiesbaden, 1971, 259) and David Keightley (‘The religious commitment: Shang theology and the genesis of Chinese political culture’, History of Religions, 17, 3/4, 1978. p. 220 and n. 40) have argued that the mandate-of-Heaven theory has its roots in Shang theology. Although I agree in a general sense, I do not agree with their suggestion that Di could sponsor an enemy attack in Shang inscriptions. He could cause them to be cursed in warfare, just as he could cause a plague of locusts, but I do not know of any inscriptions which could be interpreted as favouring an enemy. But once another king could divine independently, he could claim Di's favour. In his eyes, the defeat of the Shang king was not simply a curse, but a sign that he had been favoured

³⁸ Wanli, Qu, Shang shu jinzhu jinyi (Taibei: 1973), 84Google Scholar

³⁹ This translation is based on Karlgren's, p. 36, v. 5

⁴⁰ For discussion of this metal coffer and the practice sealing written spirit communication in ritual containers, see D. Harper, ‘A Chinese demonography’, paper presented to the Workshop on Chinese Divination and Portents, Berkeley, June 1983, 4–5

⁴¹ See Qiu Xigui , ‘Du “Anyang chutude niu jiagu ji qi keci”’ Kaogu, 1972.5, 43–4

⁴² See The heir and the sage, 85–7

⁴³ Guben zhushu jinian jijiao dingbu, 16

⁴⁴ see The heir and the sage, 85–7

⁴⁵ Shi ji (Peking: Zhonghua shuju, 1959), Lu shi jia, zhuan 35, pp.1515–16. (This story does not appear in the Ben ji.)