<Previous Section>
<Next Section>

Introduction 1

The text here translated is a heterogeneous collection of ethical, ritual, and anecdotal materials not easily characterized. Its title, "Exoteric commentary on the Han school text of the Classic of Songs," claims a relationship with one of the basic works of the Confucian canon that on examination turns out to be tenuous at best. The place of the Han shih wai chuan in Han dynasty Confucian scholarship can be clarified only by briefly describing the activities of the classical schools of the time.

The Ch`in dynasty (B. C. 221-206) had placed a ban on the private study of the books particularly venerated by the Confucian school. During the disorders and civil war following the downfall of the Ch`in some of these classics disappeared entirely and had to be recovered wholely or in part from the memories of aged scholars. The Shih ching (Classic of Songs) was one so recovered, and in the second century B. C. it existed in four recensions, each known by the name of either the founder of a school of interpretation of the text or by the name of the locality of which the founder was a native. Only the text of the Mao school survives to the present, but the four texts of the Shih ching were probably essentially the same; the schools owed their identity to their divergent interpretations of the text. Of the nearly thirty exegetical works mentioned in the catalogs and histories as belonging to the four schools during the Han dynasty, there survive only two of the Mao school, and the Han shih wai chuan,attributed to Han Ying (fl. B. C. 150) founder of the Han school. There is no reason to doubt the attribution, but, as it will appear, the HSWC is more an anthology than an original composition, and it certainly is not primarily a work of exegesis on the Shih ching. It was a textbook used by Han Ying's school, not to present his interpretations of the Classic (other works performed that function) but to demonstrate the practical use of the Classic: a tag to clinch an argument, a stanza to sum up a philosophical principle, a punning line to delight or confuse. Quotations from the Shih ching had been so used in pre-Han times and continued in use in Han writings. Han Ying provided his disciples with a convenient handbook from which they could study to perfect their technique of the apt quotation.

If my theory of the purpose and use of Han Ying's sole surviving work is correct, it explains though it does not entirely resolve, the difficulty of properly classifying it. Usually it is listed with the commentaries on the Shih ching, for the reason suggested with some annoyance by the Ssŭ-k`u editors, 2 "If you do not put it with the works on the Shih, there is no other place for it." However, a more appropriate category would be that catchall of the Chinese bibliographer, the Section of Philosophical Writings (## ##), where one finds the similar collections Shuo yüan and Hsin hsü. One would hesitate to classify the HSWC and its congeners as primarily literary works, though they are of interest as representing an early phase of the development of the anecdote and story form in the literary language. It is in them that the anecdote begins to appear, occasionally at least, as a story for its own sake, not as in the pre-Han philosophers solely to illustrate a point of doctrine, nor as in the romanticized histories (Kuo yü, Chan-kuo ts`ê) as an episode in a historical context intended to account for the motives of a principal actor. By sometimes calling attention to anachronisms and contradictory versions, I may seem to be implying that I attribute to these stories a historical character, to the ones at least where no obvious contradiction occurs. In fact, I regard all these anecdotes as unhistorical, though I do not deny the possibility that many may be based on actual events and deal with historical persons. It is rather that such stories were preserved not as a record of events, but as themes illustrative of ritually prescribed conduct. As such they could be applied to any person, historical or fictional, whose activity fitted a given role.

The materials for the HSWC were derived for the most part from pre-Han dynasty sources, sometimes rewritten, more often reproduced without significant change. Only a part of the HSWC is anecdotal; rather more than half is made up of philosophical essays plagiarized with a fine indifference to doctrinal consistency from the writings of several different schools. As Hsün-tzŭ is the favored source, the book takes on a character strongly reminiscent of the collection Hsün-tzŭ, where the formula "the Ode says" (##) is of common occurrence. As a whole the HSWC shows some similarity to the more miscellaneous chapters of the Li chi, and to the Ta-Tai li-chi; but the most closely related works are Shuo yüan, Hsin hsü, and Lieh-nü-chuan, in all three of which occur passages borrowed directly from HSWC. They differ chiefly in having their contents classified in chapters devoted to special topics, such classification being Liu Hsiang's contribution to the development of the form.

Technical Terms

In a text of this sort technical terms offer a special problem. Technical words, even within the limits of usage of a single school at a given period, seldom have a single equivalent translation word. The HSWC is as heterogeneous as its sources, and it is too much to expect to find terms like jên ##, li ##, i ## used with any consistency. Originally I had planned to leave them untranslated. However, some of them occur in contexts where they may adequately be translated by an English word—li, for example, in 1/12 clearly is "etiquette, courtesy"—and it seems misleading not to translate such occurrences, even though translation involves the sacrifice of consistency. Hence the word ## will be found as "etiquette," "ritual," "rites"—in each case with the romanized (li) in parentheses after the translation—and also simply as "li." Jên ## is used with connotations so vague as to defy English rendering, even the most inconsistent, though occasionally I have written "the humane (jên) man," in preference to the awkward circumlocution "the man endowed with jên." The word i ## used of behavior means the act appropriate to the situation and the individual involved; when it is used to denote an abstract virtue, I have left it untranslated. Other terms, such as hsiao ## "filial piety," hsin ## "trustworthiness," chung ## "loyalty," lien ## "integrity" (adj. "scrupulous") are used with a narrower range of meanings that correspond to those of their single English equivalents. Wang ##, except as a title, is used in contrast with pa ## "hegemon," and I have translated as "the True King." The terms for Confucian adepts are awkward to handle. I use "saint" for shêng ##, "sage" for hsien ## (but "worthy" in contrast to ##), "superior man" for chün-tzŭ ##, and "gentleman" for shih ##, except where an emphasis on military virtues makes the archaic value "soldier" more appropriate, or where the context calls for "official."

In its technical use, the word tao ## seems to mean the "True Way," "the Kingly Way" to Han Confucians. It occurs with a metaphysical Taoist use a few times in the text, as in 1/23. Where it is used other than technically, I have translated it variously to fit the context. Yin and yang are too well established to require a note. Anatomical and medical terms occur in 3/9 and 10/9; they are dealt with in the notes to those sections. Ch`i ## seldom occurs simply as "breath," and out of ignorance of its true force I have usually left it untranslated. Hsüeh ## likewise carries more weight than the English "blood." I have translated ## sometimes in accordance with Waley's note on the use of the word in The Analects of Confucius, p. 33, and sometimes as "virtue" when the context has demanded a vaguer word. It must be remembered that by Han times such words had already a long history and were seldom used in a strictly technical or etymological sense.

An expression that occurs very frequently in HSWC is ##. It cannot be taken as referring to HSWC itself in the way ## is sometimes used to introduce a passage in that work, since HSWC is not a "commentary" following a line of the Shih; instead the quotation normally follows the passage which illustrates it. Nor can it be referred to a single specific source. In 1/5, 2/6, 3/5, 22, 32, 7/23, and 10/4, the quotation is from Hsün-tzŭ (where ## is lacking), in 9/24 it is from Han-fei tzŭ; 7/9 is from either YTCC or Han-fei tzŭ; 3/27 is from LSCC. The great majority of occurrences introduce no identifiable quotation, though at least one half are again reproduced in Hsin hsü, SY, or other later compilations, with or without the introductory ##. A study of the passages so introduced for which a source cannot be found in the extant early literature discloses three types: (1) A general disquisition on moral conduct, such as 1/20, 2/30, 4/24, 7/19. These are very similar to 2/6, where Hsün-tzŭ seems to be the source, and may well be quotations from philosophical works now lost. (2) An aphorism (3/11, 5/15, 20, 7/5). These passages are either short or the force of the ## may be restricted to the opening sentence. Of the same sort are those occurrences of the expression within a given paragraph, as 1/1. (3) A story or anecdote that is common to a number of pre-Han works (2/1, 3/10, 4/7, 7/13, 15). I believe the basic and most general rendering of the phrase would be "There is a tradition that . . . ." In some cases a written text is being referred to; but there would be no reason always to expect to find the source in a text, even if all the texts to which a Han writer had access were still available. Oral tradition undoubtedly played an important part in the teachings of the schools attached to the Classics, and any saying or dictum an author considered worth emphasizing might rate a "tradition has it." In translating the expression I have varied the English phraseology to fit the line or passage introduced by it.

History of the Text and Editions

The Han shu "Essay on Literature" 3 lists a HSWC in six chüan and a Han-shih nei-chuan in four chüan. The Sui shu "Essay" 4 has only the HSWC in ten chüan; likewise the "Essays" in the T`ang shu,5Hsin T`ang-shu,6 and Sung shih.7 That the ten-chüan text is essentially a combination of the HSWC and the Han-shih nei-chuan listed in the Han shu is a hypothesis proposed by Yang Shu-ta 8 and further developed in my article "The Han-shih wai-chuan and the San chia shih." This amalgamation must be supposed to have occurred during the biggest lacuna in the record of the text's history: between the Han shu entry, valid for the first century A.D. and the Sui shu entry, based on sixth century catalogues. The Han school of the Shih was no longer active after the third century, 9 but a copy of the Han shih was still known to the compilers of the Sui shu "Essay," which lists numerous other works of the school, now all lost. 10

In the absence of contemporary Han citations from the HSWC by name, there is no certainty that our ten-chüan edition is at all the same work. On the other hand, the numerous parallels in Hsin hsü, SY and LNC, some of them demonstrably filiated with the HSWC version, furnish good grounds for assuming that a part at least of the present text represents the Han dynasty HSWC. Furthermore there is no apparent motive for a hypothetical forger to compile an obscure work belonging to a defunct school of classical exegesis when the resulting book provides no excuse for a new departure in either exegesis or doctrine. The case for the Han-shih nei-chuan is more tenuous. None of the three lines quoted in Po-hu-t`ung as from Han-shih nei-chuan occurs in the present text, and the question must be dealt with chiefly from internal evidence. I refer again to my earlier article for the reasons which make it likely that we now possess a fragmentary text of Han-shih nei-chuan imbedded in the present HSWC.

There are many obviously defective passages in the text of HSWC which is here translated. Not all of these imperfections can be laid to the hypothetical pre-T`ang editor who established the ten-chüan text. There are over two hundred citations of HSWC in the T`ang encyclopedias and collections of excerpts, 11 and in the commentators on the Classics, Histories, and other works, 12 of which nearly ten per cent are not to be found in the present text. There was apparently a certain amount of doubt in the minds of some of those commentators as to just what text they were quoting; Chia Kung-yen ## thrice attributes a line to HSWC in his commentary on I li,13 only to repeat it in his commentary on Chou li14 as from the Han shih shuo. A Han shih chuan is cited four times; 15 none of the lines cited appears in the present text. Chang Shou-chieh ## quotes seven times from HSWC;16 only three of his quotations are in our text, and of them one was used by Hsü Kuang ## in an earlier passage. 17 Of four quotations by Ssŭ-ma Chên ##, one is no longer in the HSWC.18 Li Shan provides the best testimony that one T`ang text at least was essentially the same as the one we have, since only nine 19 of his 115 citations are missing, and of them one 20 is a lexical gloss unlike anything in our text (I suspect they are from one of the other works of the school), whereas the only long quotation turns up twice in the same commentary as being from Han-shih nei-chuan.21 It is worth pointing out also that in his commentary three passages from HSWC are introduced by ## ##. 22

The Sung bibliographers 23 knew the HSWC, and it is quoted 157 times by the TPYL, of which twenty-three quotations are not in the present text.

The oldest published text is the (?) Sung edition reprinted by Mao Chin ## (1599-1659), 24 an edition he believes to be the same as that described by Hung Mai ## (1123-1202), who says, 25 "During the Ch`ing-li ## period (1041-1048) the Chung-chiang tso-chien-chu-pu ## Li Yung-chang ## 26 edited [the HSWC] and ordered workmen to cut [blocks for its publication] in Hang[-chou]. There is also a postface which says, `Over 3,000 characters have been rectified by Minister Wên.' 27 I have a [copy of] this book in my house." Three thousand emendations in a text of approximately 50,000 words—say six per cent —is admittedly high. That there are nothing like so many discrepancies between Mao's text and the other available editions 28 may mean that they too are derived from the one published by Li Yung-chang. No modern catalog lists a Sung edition, and the only Yüan edition seems to be the one from which Ch`ü Chung-jung ## (1769-1842) made notes. 29

By T`ang times isolated passages in the HSWC had attracted the attention of commentators on early philosophical and historical texts. 30 Wang Ying-lin ## deals incidentally with a few lines, 31 but it was not until early in the Ch`ing dynasty that a critical edition of HSWC was published with a full-dress commentary. Two Ch`ing scholars brought out their HSWCcommentaries in successive years, each apparently unaware that the other was engaged in the same task. The earlier of the two, by Chao Huai-yü (1747-1823), 32 bears a preface dated the fifth moon, 1790, and was printed by the author in his own establishment, the I-yu-shêng-chai ##. 33 He mentioned in his collation notes eight editions which he made use of, including a Yüan edition (cited once); 34 in addition he gleaned many citations from the encyclopedias and the T`ang commentators. On the basis of such citations and parallel passages in pre-Han and Han works he did not hesitate to make extensive emendations in the HSWC text. It is easy to share Liu Shih-p`ei's enthusiasm for the result, 35 as many difficult passages are simplified by his changes or additions; however, it is questionable how much reliance can be put on the encyclopedias and commentators, given their predilection for abridgment and paraphrase.

Chou T`ing-ts`ai, whose HSWC chiao-chu appeared with a preface dated the sixth moon, 1791, using essentially the same materials, produced a more conservative text. Chou usually limited himself to suggesting in his notes emendations he thought desirable. It is interesting that the two scholars, working independently, are more frequently than not in agreement about doubtful points. It is not surprising, however, that Wu T`ang in bringing out a combined edition of the two commentaries, reproduced Chou's text, relegating Chao's variants to the notes. 36

Under the ambitious title of HSWC su-chêng Ch`ên Shih-k`o published in 1818 an edition (provenance not specified) of the HSWC in which are included most of the parallel passages, whether earlier or later than Han Ying. These are printed in full after each paragraph of HSWC. The usefulness of this compilation is somewhat impaired by the fact that his HSWC text is defective, while the parallels contain many misprints. Despite the title, no commentary is supplied. This work has been the point of departure for locating most of the parallels cited in the notes to my translation; it is not there acknowledged, as the same parallels are nearly always indicated in Chou, CHy, or Chao; it was simply easier to locate them from a page of text than from a single line or phrase.

In the appended bibliography will be found listed the notes by Sun I-jang and Yü Yüeh on the HSWC. They are useful, but deal with only a dozen or so passages. So far the best single work devoted exclusively to the HSWC is Chao's Shan-i's HSWCpu-chêng, to which I am greatly indebted, especially for the rich documentation which it brings together for the consideration of doubtful passages. Sources are given with reasonably exact references (not, of course, page references), and citations are usually in full.

The Kuan-hai-t`ang shu mu ## 1.6b lists a MS. HSWC k`ao ## in ten chüan by Okamoto Hōkō ##, with appendix ## and collation notes ##, each 1 vol. This work is also mentioned by Chao, Introduction, p. 3, as a work he was unable to get; I have never seen a copy.

A new edition of HSWC has been prepared by Professor Hsü Wei-yü ## of National Tsing Hua University, but the MS. still awaits publication. I regret not having had the use of it in preparing my translation, though Professor Hsü kindly offered to put at my disposal his materials, unfortunately just before my departure from Peiping.


1. In a separate article, "The Han-shih wai-chuan and the San chia shih," HJAS 11 (1948) .241-310, I have already given a general account of the HSWC, its sources and imitations, along with a theory about the nature of the work and its relation to the Han School of the Shih. There is no point in repeating the arguments and data presented at length in that article, and I shall confine myself in this introduction to briefly identifying the text and stating my practice in dealing with certain technical terms. Some remarks on editions and the history of the text are appended. For abbreviations and editions of texts cited see the Bibliography.

2. Ssŭ-k`u ch`üan-shu tsung-mu, 16.11a (Ta Tung Shu-chü ed.); see the Appendix for a complete translation.

3. Han shu 30.4b.

4. Sui shu 32.12a.

5. T`ang shu 46.6a.

6. Hsin T`ang-shu 57.4a.

7. Sung shih 202.10b.

8. Han shu pu-chu pu-chêng ## 1.28 (Commercial Press, 1925).

9. At least none of the works belonging to that school in the Sui shu "Essay" are attributed to authors living after the San-kuo period.

10. Sui shu, loc. cit.

11. Pei-t`ang shu-ch`ao ( Shih-nan 558-638) 15 times; I-wên lei-chü (Ou-yang Hsün 557-641) 28 times; Ch`ün-shu chih-yao (Wei Chêng 580-643) 19 times; Ch`u hsüeh chi (Hsü Chien 659-729) 15 times; Po-shih lui-t`ieh shih-lei-chi (Po Chü-i 772846) 13 times.

12. K`ung Ying-ta ## (574-648) in his commentary on Li chi (4 times), Tso chuan (2 times); Yen Shih-ku ## (581-645) in his commentary on Han shu (3 times); Li Hsien ## (651-684) in his commentary on Hou-Han shu (4 times); and especially Li Shan ## (?-689) in his commentary on Wên hsüan (115 times).

13. I li 2.3a, 4.5a, 18.4a.

14. Chou li 41.7b.

15. Ibid. 14.6b, 32.7b; Kung-yang chuan 12.9b (Ho Hsiu); Han shu 29.11a (Yen Shih-ku).

16. Shih chi 2.5b, 28.5a-b, 42.12a, 44.20b, 67.14b, 83.2b-3a, 117.9b.

17. Ibid. 5.33a quotes from HSWC, likewise Chang Shou-chieh's commentary on 44.20b, but not the same line.

18. Ibid. 46.11a, 55.7a, 117.7b. 126.12a is missing.

19. *Wên hsüan 4.2b (35.19a), 9.27a, 13.20a (31.7b), 26.13b, 29.1b (29.20b, 36.3b), 29.10b, 40.34a, 55.22b, 59.27b.

20. Ibid. 16.26b.

21. Ibid. 12.28b, 19.18a.

22. Ibid. 3.1b, 17.19a, 60.14a.

23. Ch`ung-wên tsung-mu ## (Wang Yao-ch`ên ## 1001-1056); Yüehya-t`ang ts`ung-shu vol. 166, 1.9b; Chün-chai tu-shu-chih ## (Ch`ao Kung-wu ##); Chih-chai shu-lu chieh-t`i ## (Ch`ên Chên-sun ##, fl. 1235); Wang Ying-lin, K`un hsüeh chi-wên.

24. Vols. 3-4 of Chin-tai pi-shu ##; cf. his postface ##.

25. *Jung-chai sui-pi II 8.4a.

26. I have not been able to identify him further.

27. ##. There was a Wên Yen-po ## (1006-1097) who was a Minister of State.

28. A Yüan text reprinted by Shên Pien-chih ## in the Yeh-chu-chai ## with a preface by Ch`ien Wei-shan ## dated 1356. For this edition see Bibliography under HSWC, B. Of the four Ming editions listed in the Catalogue of the Kiangsu Province Sinological Library ## 3.23b, the Chint`ung-ts`ao-t`ang ## is reprinted in the Ku-ching-chieh hui-han and in the Han-Wei ts`ung-shu (Commercial Press photographic reprint of the original Ming edition).

29. Ch`ü Chung-jung's notes are printed in the Yao-pu ts`ang-shu t`i-shih hsü-lu 1.2a-4a ## (cf. Chao, Introduction 1-2).

30. Yang Liang's commentary on Hsün-tzŭ (Preface dated 818), and see note 11 above.

31. *K`un-hsüeh chi-wên 2.15b, 3.4b and passim.

32. Cf. Hummel, Eminent Chinese of the Ch`ing Period 71-72. Abbreviated as CHy.

33. The original print is quite rare; however, there is a reprint in the Lung-ch`i-chingshê ts`ung-shu; cf. Bibliography s. v. Chao Huai-yü.

34. Cf. Chao, Introduction, p. 1.

35. Tso-an chi 1.14b-15b ## vol. 27 in Liu Shên-shu hsien-shêng i-shu ## ##. Mo Yu-chih ## (1811-1871) also rates CHy's edition above that of Chou T`ing-ts`ai; cf. Lü-t`ing chih chien ch`uan-pen shu-mu 2.6b ## ##, Hsi-lêng yin-shê ed. ##. Cf. Chao, Introduction, p. 1.

36. It is this combined edition, generally acknowledged to be the best available, which is the basis for my translation. Throughout, I have had access to the original as printed in the Wang-san-i-chai, but folio and line references in the Finding List are to the re-edition easily accessible in the Chi-fu ts`ung-shu.

<Previous Section>
<Next Section>
IATHPublished by The Institute for Advanced Technology in the Humanities, © Copyright 2003 by Anne Kinney and the University of Virginia