¹ Sun Tzu: The art of war. Translated and with an introduction by Samuel B. Griffith. (Unesco Collection of Representative Works, Chinese Series.) xvii, 197 pp., front., 2 maps. Oxford: Clarendon Press, 1963.

² Yü Yüeh eq1 alone amongst the eminent scholars of the Ch'ing dynasty, has written notes on the Sun tzu, but these amount to no more than a couple of pages. See Chu tzu p'ing yi pu lu eq2 Peking, 1956, ch. 3.

³ Emended from eq3. Cf. Yang Ping-an eq4 Sun tzu chi chiao eq5, Peking, 1959, p. 14. The reading eq6 seems preferable because this passage deals with types of action appropriate to varying degrees of numerical strength relative to the enemy and there is too little difference between eq7 and eq8.

⁴ The pagination given after quotations from the Chinese text refers to the Shih yi chia chu Sun tzu eq9 Peking, 1962, which is based on the Sung edition in the Shanghai Library, reproduced photographically in a limited edition in 1961. I have used this edition because the text in Sun Hsing-yen's eq10 edition is open to criticism. See below, n. 9.

⁵ For the convenience of the reader, section numbers according to Griffith, preceded by the original Chinese chapter numbers, have been inserted into quotations from Griffith's translation and, in cases where his translation has not been quoted, into the Chinese text.

⁶ See Wang Nien-sun eq11, Tu shu tsa chih eq12, Wan yu wen ku eq13 eq14ed., 3.13.

⁷ Ching chuan shih tz'u eq15, Peking, 1956, p. 134.

⁸ When no pagination follows a translation it is my own.

⁹ Griffith states in his introduction, ‘Sun Hsing-yen was the leading authority on Sun Tzu to emerge during the Ch'ing (Manchu) dynasty; it is his edition (in the preparation of which his friend Wu Jen-chi collaborated) which has been considered standard for almost two hundred years, and which is the basis for the present translation’ (p. 19). It is a pity that Griffith has followed Sun's edition in view of the criticism that has been levelled against it. The editors of the Shih yi chia chu Sun tzu say that as a result of comparing the Sung edition with the Ming.

¹⁰ Duke Chuang, 10.

¹¹ Sun tzu shih chia chu eq16, Ssu pu pei yao eq17 ed., hsü lu, p. 13b.

¹² Wang Nien-sun, Kuang ya shu cheng eq18, Wan yu wen ku ed., p. 215.

¹³ op. cit., p. 497.

¹⁴ e.g. the Lu eq19 version of the ‘Analects of Confucius’reads in eq21 (5/27) and, similarly, the Cheng eq22 version reads eq23 (18/4). See Lu Te-ming, eq24 25, Ching Hen shih wen eq26, Tsung shu chi cheng eq27 ed., p. 1384 and p. 1398.

¹⁵ e.g. in the sentence eq28. See Sun Yi-jang eq29, Ting pen Mo tzu chien ku eq30 1.17b.

¹⁶ Commentary to Ode 247.

¹⁷ The use of captions and recapitulations in ancient texts is not uncommon. A good example is ch. 9 of the Hsün tzu.

¹⁸ This has been noted by Yang Ping-an (op. cit., p. 36).

¹⁹ ‘(1/6) By terrain I mean distances, whether the ground is traversed with ease or difficulty, whether it is open or constricted, and the chances of life or death ’(p. 64).

²⁰ ‘(10/17) Conformation of the ground is of the greatest assistance in battle. Therefore, to estimate the enemy situation and to calculate distances and the degree of difficulty of the terrain so as to control victory are virtues of the superior general’ (p. 127–8).

²¹ There is one feature in Griffith's translation which ought to be mentioned, even though only in passing, and that is his practice of using different renderings for the same Chinese term used with exactly the same meaning, no doubt for purposes of literary elegance. This is unexceptionable if the work translated is one of literary merit and translated for that reason. In a technical work like the Sun tzu which is also extremely cryptic, the closer a translation is to the original the more helpful it will be to the reader who is unable to read the original. This is particularly so in cases of technical terms. Unfortunately, Griffith allows literary elegance to outweigh considerations of meticulous fidelity to the original even in cases where fidelity is absolutely essential. An example in the present connexion will serve to show what I have in mind. In 6/15 (p. 98), the term kua eq32 occurs five times with the same meaning. It is only once translated as ‘few’. It is translated once as ‘ fragile’, once as ‘vulnerable’, and twice as ‘weak’. This is in spite of the fact that juo eq33 is also a technical term in the Sun tzu, and ‘weak’ should properly be reserved for its translation.

²² Chia Lin eq34 is the only one who gives any indication that he thought the term might mean something different. He comments: ‘The best can only be attained by taking the enemy's state intact while, at the same time, preserving my own state intact’. But though Chia puts in the remark about ‘preserving my own state’, he still clings to the idea of ‘taking the enemy's state intact’. Thus he seems to want the term ch'üan to mean both of these things.

²³ I have substituted ‘preserving’ for ‘protecting’ in Griffith's translation.