¹ The Secular Scripture (Cambridge: Harvard University Press, 1976), pp. 6, 7, 17Google Scholar. I am of course familiar with Jaroslav Průšek's “History and Epics in China and the West” (in his Chinese History and Literature [Dordrecht: Reidel, 1970], pp. 17–34; first published in Diogenes [Montreal] 42 [1963]). Průšek offers many sound insights into the relationship between epic and history in Greece, but his assimilation of historiography in China to the mode of lyric poetry is, in my opinion, fundamentally mistaken. Frye provides a far more persuasive intellectual model than Průšek or the theoreticians he relies on. The first chapter of Frye's book is especially brilliant and stimulating.

² The Secular Scripture, p. 8.

³ “Matter” (or matière) is used by scholars of medieval European literature to refer to the entire range of incidents and characters that came to be associated with King Arthur, with Charlemagne, or with ancient Rome (the matière de Bretagne, matière de France, matière de Rome). Wu Tzu-hsü is obviously far less important in Chinese literary tradition than Arthur or Charlemagne were in Europe, but it is extremely useful to have a word that refers specifically to all the stories, anecdotes, ideas, and so on that cluster around certain individuals or events. These systems of“stuff-material” seem to me a prime subject for future research.

Studies of the chanson de gate, the volksepos, and other medieval European epic genres have much to offer the student of early Chinese literature. A good introduction to the recent scholarly work is provided by Fisher, John, ed., The Medieval Literature of Western Europe: A Review of Research, Mainly 1930 to 1960 (New York: New York University Press, 1966).Google Scholar

⁴ Although the range of titles ought to preclude any accusation of cultural chauvinism, of “Orientalist” imposition of non-Chinese standards, let me say explicitly that I believe this question to be fundamentally different from questions like “Why was there no industrial revolution in China?” or even “Why did the Chinese produce no tragic drama?” The category of epic or saga is virtually universal; the others are not.

⁵ Compare the statement of the Chadwicks: “Our belief is that primary hero stories are contemporary, i.e., that the first stories which celebrate a hero's exploits are composed within living memory of the events” (The Growth of Literature, 3 vols. [Cambridge: Cambridge University Press, 1932–40], 1:215)Google Scholar; quoted in Wailes, Stephen, “The Niebelungenlied as Heroic Epic,” in Oinas, Felix, ed., Heroic Epic and Saga (Bloomington: Indiana University Press, 1978), p. 135.Google Scholar

⁶ Tso chuan, Chao 20, 30, 31; Ting 4; Ai 1, 11. See Legge, James, trans., The Chinese Classics, vol. 5, The Ch'un Ts'ew with the Tso Chuen (1872; rpt. London: Henry Frowde, n.d.): 680B–81A, 734B–35B, 737B, 755B–56A, 794B, 825B–26AGoogle Scholar. All material cited from Tso chuan is in these passages.

⁷ Kuo yü (Ts'ung-shu chi-ch'eng ed.), 19.216–19; chüan 20 and 21, passim.

⁸ Lü-shih ch'un-ch'iu (Ssu-pu ts'ung-k'an ed.), 9a–9b; 14.8b–9a, Bb–14b;23.5b–7a.There is a German translation: Wilhelm, R., trans., Frühling und Herbst des Lü Bu We (Jena: E. Diedrichs, 1928). 1928).Google Scholar

⁹ Shih chi, chüan 66, with supplementary information in chüan 31 and 41 (pp. 1461–72 and 1739–46 of the Chung-hua shu-chü ed.). Shih chi 66 is translated by Watson, Burton in his Records of the Historian: Chapters from the Shih Chi of Ssu-ma Ch'ien (New York: Columbia University Press, 1969), pp. 16–29.Google Scholar

¹⁰ Wu Yüeh ch'un-ch'iu, chüan 3–5, with scattered material in chüan 7–9.

¹¹ For a detailed analysis of the Wu Tzu-hsü story as presented in Wu Yüeh ch'un-ch'iu and Yüeh chüeh shu, see my “The Wu Tzu-hsü Pien-wen and Its Sources: Part I,” Harvard Journal of Asiatic Studies 40, 1 (June 1980): 128–43.Google Scholar

¹² “The Wu Tzu-hsü Pien-wen, Part I,” n. 139.

¹³ Wu Yüeb ch'un-ch'iu (Ssu-pu pei-yao ed.), 3.4a–4b; 4.11b.

¹⁴ Wu Yüeh ch'un-ch'iu, 4. lb; 5.6a.

¹⁵ Yüeh chüeh shu (Ts'ung-shu chi-ch'eng ed.), 14.69.

¹⁶ Wu Yüeh ch'un-ch'iu, 10.9b, 5b–6a.

¹⁷ Some of the instances have already been pointed out. In addition, there are several cases where Wu Yüeh ch'un-ch'iu is so dependent on an earlier source that its often faulty text can be corrected by comparison with that source. Comparison of Wu Yüeh ch'un-ch'iu, 4.11a.3, with T so chuan (Legge), 758.11–12, shows that a passage has fallen out between yü and Ch'u, or between Ch'u and shih—a case of omission by homioteleuton, to use the technical text-critical term (where the copyist leaves out the material between two occurrences of a word in single passage). Garbled passages in 5.6b.9–7a. 13 can be corrected by comparison with Kuo yü, 19.218.10–219.3. Comparison of 3.6b.11 with Shih chi (Chung-hua shu-chü ed.), 31.1463.8, suggests that the second tsu is a duplication. And—as Hulsewè has also observed in T'oung pao 57 (1971): 165—3.3b.3 can be corrected by reference to Shih chi, 66.2173.6.

¹⁸ Wu Yüeh ch'un-ch'iu, 10.5a–5b.

¹⁹ Ch'i-yu, Ch'en, ed., Han Fei-tzu chi shih, 2 vols. (rev. ed., Hong Kong: Chung-hua shu-chü, 1974), 9.521, 554Google Scholar; Liao, W. K., trans., The Complete Works of Han Fei Tzu (London: A. Probsthain, 1954), I: 283, 301–2.Google Scholar

²⁰ de Bary, W. T. et al. , eds. Sources of Chinese Tradition (New York: Columbia University Press, 1960), pp. 270–71, 272–73.Google Scholar

²¹ Early Chinese Literature (New York: Columbia University Press, 1962), p. 188.Google Scholar

²² Kierman, Frank Jr., trans., China in Antiquity (Amherst: University of Massachusetts Press, 1978), p. 364Google Scholar. This is a translation of the 1963 revision of Maspéro's La Chine Antique, first published in 1927.

²³ A late Ch'ing compilation of all known references to works dating from Later Han times.

²⁴ A work in 8 p'ien called Wu Tzu-hsü is listed in the “Miscellaneous Philosophers” (tsa tzu) section of the Han shu catalogue (Peking: Chung-hua shu-chü, 1962), p. 1740Google Scholar. This probably presented teachings attributed to Wu Tzu-hsü but could well have contained anecdotes of some complexity about his life as background for his ideas. In Hou Han yi-wen chih, we find in the “Miscellaneous Histories” (tsa shih) section an epitome of Chinese history called Ku chin chu which we are told recounted episodes like the one in which the evil Ch'in eunuch Chao Kao pointed at a deer and called it a horse, and in the “Historical Excerpta” (shih ch'ao) section there is listed a book called Tung li, which was a general history of China and included even trivial matters (Shih-yüan ts'ung-shu ed., 2.17a–18b, 26a–27a). Each of these could easily have contained material on Wu Tzu-hsü. Hou Han yi-wen chih also lists many biographies (2.63b ff.). Although all the titles listed have Han-time subjects, the genre must have been in existence for some time, and hence it is at least conceivable that there had been a biography of W u Tzu-hsü. (On early biographies, see also China in Antiquity, pp. 361–62). Finally, there are works concerning a particular locality or region, such as “Stories of the Elders of P'ei-kuo” (P'ei-kuo ch'i-chiu chuan) and “Accounts of Lu-kuo's Ancient Worthies” (Lu-kuo hsien-hsien chuan), which appear to have combined notices of famous or exemplary native sons with accounts of local traditions and customs (2.66b ff., esp. 67a. 10–11, 68b.2–3, 69b.2). If there were works of this sort for the Soochow or Hangchow areas, they could well have included material about Wu Tzuhsü. Indeed, such works could be among the ancestors of Wu Yüeh ch'un-ch'iu.

²⁵ On the Wu Tzu-hsü cult, see my “The Wu Tzu-hsü Pien-wen and Its Sources: Part II,” Harvard Journal of Asiatic Studies 40, 2 (December 1980): 472–98.Google Scholar

²⁶ Leiden: E. J. Brill, 1968.

²⁷ The Local Cultures, pp. 438–46.

²⁸ The Local Cultures, pp. 446–48.

²⁹ The Local Cultures, p. 452.

³⁰ The Local Cultures, pp. 346–47.

³¹ The Local Cultures, pp. 393–94.

³² The evidence for this conclusion is presented in my “The Wu Tzu-hsü Pien-wen, Part II.”

³³ Maspéro's general position is given in China in Antiquity, pp. 357–65. It was first proposed in a paper on what he called “Le roman de Sou Ts'in,” Études Asiatiques publées à l' occasion du vingt-cinquième anniversaire de l'École Française d'Extrême-Orient (Paris, 1925), 2: 127–41Google Scholar. This was refined in a lecture given in 1929 called “Le roman historique dans la littéature Chinoise de l'Antiquité,” reprinted in his Mélanges posthumes (Paris: Presses universitaires de France, 1950; rpt. 1967), 3: 55–62.Google Scholar

³⁴ China in Antiquity, p. 357. Neither “romance” nor “novel” is a good translation of roman in this context, so I have left it untranslated.

³⁵ Mélanges posthumes, 3: 61.

³⁶ “Le Roman de Sou Ts'in,” p. 135.

³⁷ China in Antiquity, p. 360.

³⁸ China in Antiquity, p. 364.

³⁹ Chan-Kuo Ts'e (London: Oxford University Press, 1971)Google Scholar; Intrigues (Ann Arbor: University of Michigan Press, 1964).Google Scholar

⁴⁰ Chan-Kuo Ts'e, p. 15.

⁴¹ China in Antiquity, p. 363.

⁴² His only detailed work concerns the Shih chi biography of Su Ch'in, and there he reasons from anachronisms, for which one can imagine a multitude of causes other than origin in a roman. He does not make it clear if his belief that other such works existed is also based on the presence of anachronisms or on other criteria.

⁴³ “Narratives in Tso chuan,” Harvard Journal of Asiatic Studies 37, 2 (1977): 323–52.CrossRefGoogle Scholar

⁴⁴ “Narratives,” 350–51.

⁴⁵ “Narratives,” 352.

⁴⁶ Chan-Kuo Ts'e, pp. 15 et seq.

⁴⁷ “Narratives,” 343, 351.

⁴⁸ Hawkes, David, trans., Ch'u Tz'u: The Songs of the South (Boston: Beacon Press, 1962), p. 75Google Scholar. There are numerous other allusions to Wu Tzu-hsü in the “Nine Declarations,” e.g., pp. 64–65 and 80. Hawkes dates the “Nine Declarations” to around 250 b.c. (p. 59).

⁴⁹ I am indebted to Michael Dalby's “Revenge and the Law in Traditional China,” forthcoming in the American Journal of Legal History, for this reference. The Kung-yang chuan is traditionally ascribed to Kung-yang Kao, who is said to have lived in the late fifth century b.c., but it was not put into written form until the middle of the second century b.c. (Kanseki Kaidai, Koson, Kacsura [Tokyo: Meiji shoin, 1922], pp. 28–29Google Scholar; Legge, The Ch'un Ts'ew with the Tso Churn [cited n. 6], pp. 36–37). Ku-liang chuan, whose origins are even cloudier than Kung-yang chuan's, but obviously is closely related to it, contains a similar passage.

⁵⁰ Yüan, Juan, ed., Shih-san ching chu-shu (1815;rpt. Taipei: Yi-wen Publishing, 1965), 7: 25. 15a–16a.Google Scholar

⁵¹ 3.4b–5a. The quoted passage is at 5a.2.

⁵² Tso chuan (Legge), pp. 681A, 734B–35B; Lü-shih ch'un-ch' iu (cited n.8), 14.8b–9a; Shih chi (cited n. 9), 66.2173–74.

⁵³ Chan-Kuo Ts'e (Crump), p. 104.

⁵⁴ The-Complete Works of Han Fei Tzu (cited n. 19), 1: 229. Wu Yüeh ch'un-ch'iu also has a version of this episode, naming Chao Pass (3.3b.4–6).

⁵⁵ Chan-Kuo Ts'e (Crump), pp. 535; 145 and elsewhere; and 546–7.

⁵⁶ D. C. Lau, trans., Mencius IB. 3.

⁵⁷ Kuo yü(citedn.7), 20.230.U;Han Fei-tzu chi shih (cited n. 19), 5.308, 6.394, 403; The Complete Works of Han Fei Tzu, 1: 159, 211, 218. It is characteristic that one of the allusions in Han Fei-tzu (Liao 1.159) contains a reference to another incident—King Kou-chien consulting the Ta-P'eng tortoise oracle—that is mentioned nowhere else. Note that Shih chi, 41.1740, recounts Yüeh's offer to submit to Wu, but says nothing about Kou-chien's exile in Wu. Indeed, it seems to imply that the offer was not accepted.

⁵⁸ Chan-Kuo Ts'e (Crump), p. 135.

⁵⁹ The Complete Works of Han Fei Tzu, 1: 283, 301–2; Han Fei-tzu chi shih, 9.521, 554.

⁶⁰ Such works could also have been put into written form at an early date, considering the intensely literary bent of early Chinese high culture. But narratives of the sort I have in mind would have been rather impressive literary achievements, and this makes the seeming absence of references to such works in Han and pre-Han texts rather hard to explain. The nonexistence of such references, however, is a point on which a good deal of further research needs to be done. And, in any case, I believe my argument holds regardless of whether these hypothetical narratives were written or oral.

⁶¹ I am departing from the usual definition of “epic,” but the characteristics I imagine these early works to have had—length, narrative coherence, stress on heroic action—seem to me to be at the heart of what we call epics. Many medievalists, such as Jackson, W. T. H., call works such as the Chanson de Rolandand the Niebelungenliedepics. (See his The Literature of the Middle Ages [New York: Columbia University Press, 1960], p. 175.)Google Scholar

⁶² See Kierman, Frank, “Phases and Modes of Combat in Early China,” in Chinese Ways in Warfare, eds. Kierman, Frank and Fairbank, John (Cambridge: Harvard University Press, 1974), pp. 62–63CrossRefGoogle Scholar. Note also his remarks on the prevalence of warfare in Tso chuan, pp. 28–29. A good impression of the pervasiveness of warfare in Ch'un-ch'iu and Warring States times is provided by Hsu, Cho-yun, Ancient China in Transition (Stanford: Stanford University Press, 1965), chap. 3, esp. tables 5 and 6Google Scholar. See also the remarks on the military in this period in Creel, H. G., The Origins of Statecraft in China, I (Chicago: University of Chicago Press, 1970), chap. 10.Google Scholar

⁶³ D. C. Lau, trans., Mencius IIIB.9.

⁶⁴ Shih chi (cited n. 9), 130.3319–20. Compare Watson, Burton, Ssu-ma Ch'ien, Grand Historian of China (New York: Columbia University Press, 1958), p. 57.Google Scholar

⁶⁵ Sources of Chinese Tradition (cited n. 20), p. 257, quoting Han shu (Chung-hau shu-chu ed.), 56.2523.