The Apocryphal Texts of the Han Dynasty and the Old Text/New Text Controversy

The Apocryphal Texts of the Han Dynasty and the Old Text/New Text ControversyAuthor(s): Hans Van EssSource: T'oung Pao, Second Series, Vol. 85, Fasc. 1/3 (1999), pp. 29-64Published by: BRILLStable URL: http://www.jstor.org/stable/4528775 .

Accessed: 29/11/2014 11:02

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

BRILL is collaborating with JSTOR to digitize, preserve and extend access to T'oung Pao.

http://www.jstor.org

This content downloaded from 194.95.59.195 on Sat, 29 Nov 2014 11:02:23 AMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=bap

http://www.jstor.org/stable/4528775?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp




30 HANS VAN ESS

with exhaustively, at least not in the West. A survey of various institutional matters of importance to Han society on which both collections of fragments happen to contain evidence shows that the traditional claim that apocryphal texts belonged to the stream of thought advocated by the New Text school is indeed correct. As these institutional problems also constituted the main topics of contention during the period of reforms at the end of the Former Han dynasty, our conclusion is that, aside from purely philological considerations, it is indeed plausible, following Hsu Shen, to describe the debate in terms of a conflict between two traditions, one associated with a group of literati who used the officially established "new text" version of the Classics as a basis for their arguments, the other with an opposing camp which referred to the allegedly newly discovered manuscripts written in old script.

Defining the term "apocrypha"

Sometime during the Han period a genre of texts called ch'en- wei X,* (or ch'an-wei) ,2 generally translated as "apocryphal scriptures" in English-language works on Han scholarship, began to play an important role in the exegesis of the Chinese Classics. These texts, whose name does not appear in the sources of the Warring States period or of the Early Han,3 were at least in part ascribed to Confucius.4

2 In Western literature the character X is often transcribed ch'an, but all major dictionaries published in China today romanize it ch'en. See e.g. the Han- yii ta tz'u-tien - vol. 11/466; Kuang-yiun W, which is quoted there, gives the spelling X (cf. Karlgren, Grammata Serica Recensa (GSR), 660j (X): *tsidm/ tsiam-, modern spelling: chen). This indicates that the ending should be "en" as opposed to "an". The latter ending also exists for this character, when it is exchangeable with $t ch'an (Chi-yiin : ZM Karlgren, GSR, 609c (s): *klam/ kam/chien), "to regret", "to confess".

3 The bibliographical chapter of the Book of the Han (Han shu, hereafter HS) does not mention even one title of the apocryphal texts known to us today.

4 Bent Nielsen, "The Qian zuo du ,t, a Late Han Dynasty (202 B.C.-AD 220) Study of the Book of Changes, Yi jing A0", unpublished dissertation, Univer- sity of Kopenhagen, 1995, p. 27: "There are no less than 37 instances of 'Confucius said' RL#~EH [in the Qian zuo du]. What follows 'Confucius said', are, of course, apocryphal quotations, and it is impossible to decide whether the entire passage down to the next 'Confucius said' is to be regarded as a statement by Confucius, or if it is supposed to be an initial statement by Confucius, which is commented on by the author of the Qian zuo du."



THE APOCRYPHAL TEXTS OF THE HAN DYNASTY 31

The significance of the apocrypha for the purpose of legitimiz- ing various Chinese dynasties was pointed out in 1966 by Jack Dull in his important dissertation on the historical background of the development of ch'en-wei literature.5 Since then and until very recently, however, not much work has been done in the West on these intriguing and fragmentarily preserved texts.6 The Cam- bridge History of China, Vol. I, for example, mentions them only briefly.

The translation of the term ch'en-wei as "apocryphal texts" is not itself uncontroversial, since a generally accepted common denominator to Chinese and Western apocrypha has yet to be defined. In the European tradition, "apocryphal books" are scriptures that do not belong to the canon of the books of the Bible, but are nevertheless closely associated with it genetically and the- matically. And yet, the Greek word apokryphos means "hidden": hence, apocryphal scriptures originally were texts with a hidden meaning. In some cases the word is simply used for books whose author is not known.7 Only in recent times has it come to be used for scriptures which were at some stage eliminated from the can- ons of the Old or the New Testament, but remained related to them in some way.8

Apocryphal scriptures attached to the Old Testament have been defined by Martin Luther as books that do not belong to the authoritative canon but are nevertheless useful to read.9 As has been established by modern scholarship, when these scriptures appeared, somewhere between the third century B.C. and the first century A.D., a canon of authoritative texts from the

5 Jack Dull, "A Historical Introduction to the Apocryphal (ch'an-wei) Texts of the Han Dynasty", Ph.D. dissertation, University of Washington, 1966.

6 Probably the most important article since Dull's thesis is Anna Seidel's "Imperial Treasures and Taoist Sacraments-Taoist Roots in the Apocrypha", IVIe1langes chinois et bouddhiques 21 (1983), pp. 291-372. In addition there is Nielsen's recent dissertation on the Qian zuo du, cited in note 4 above. Compare also Fabrizio Pregadio, Zhou Yi Cantongqi, Dal Libro dei Mutamenti alElisir d'Oro, Venice 1996. Pregadio regards the Chou-i tsfan-t'uing-chMi as belonging to the context of the apociyphal scriptures.

7 Compare the entry "Apokryphen" in Theodor Klauser, ed., Reallexikon fur Antike und Christentum, vol. 1, Stuttgart 1950, pp. 516-519.

8 Wilhelm Schneemelcher, ed., Neutestamentliche ApokTyphen in deutscher Ubersetzung, fifth edition, Tfubingen 1987, introduction, p. 5.

9 Quoted in Schneemelcher, op. cit., p. 1. See also E. Kautsch, Die Apok?yphen und Pseztdoepigraphen des Alten Testaments, vol. 1, Tuibingen 1900, p. XI.



32 HANS VAN ESS

Jewish tradition had yet to be formed, at least as far as the Hel- lenistic Jews were concerned. The latter used a number of books (such as Tobias, Judith, Baruch, and others) which for the Pales- tinian Jews could not claim authority. Only after their reception by the young Christian community were these texts eliminated from the canon of books used in the synagogues. It is these books which later came to be called apocryphal texts attached to the Old Testament. However, until the fourth century A.D. the word "apocryphal" was, in the Jewish as well as in the Christian tradition, reserved for heretical texts and not used in the modern sense, that is, for particular books of the Old Testament.10

Although Hieronymus (ca. 340-420 A.D.) laid down the difference between libri canonici, the Hebrew collection, and libri eccle- siastici, the apocryphal texts handed down within the Hellenistic Jewish tradition, the Christian tradition has never rejected these books entirely, and at the concilium of Trento in 1546 it eventually included almost all of them into the canon of the Old Testa- ment.11

The situation is somewhat different with the apocryphal literature attached to the New Testament. Scriptures called at the time "apocryphal" were eliminated from the canon as early as the second or third century, and they were thereafter used only by a few groups which needed a basis for heretical theological positions. Later, these scriptures became more of a hagiographic tradition, serving mainly the purposes of edification and entertainment.

Most of the apocryphal texts of the Christian tradition were modeled after the literary genres of the New Testament or written within the same stream of thought as the gospels. Their contents dealt in a popular way with subjects not covered by the Bible itself: the life of the virgin Mary, the childhood and youth of Jesus, the fate of the apostles, and so on. Even though the church may have tried to stop the rapid spread of the apocrypha as early as the sixth century, they nevertheless became one of the most important sources for literature and art in medieval Europe.'2

10 Kautsch, op. cit., p. XII. Schneemelcher, op. cit., p. 6f, stresses the fact that gnostic authors appealed to bibloi ap6kiyphoi, and that the word was assigned a negative sense in the early Christian tradition since the church rejected the gnostic secret sciences.

11 See Leonhard Rost, Einleitung in die alttesttamentlichen Apoktyphen und Pseu- doepigraphen, Heidelberg 1971, p. 15f and pp. 22ff.

12 Compare the entry on "Apokryphen" in Lexikon des Mittelalters, Vol. 1, Munchen and Zurich 1986.




According to Robert Kramers, the Chinese "apocryphal texts" consisted of ch'en X, the term for oracles and predictions, and wei 01, which "indicated a literature containing esoteric explanations of the ching E or Classics. Ching originally meant the warp of a loom, and wei meant its woof'.13 It should be pointed out that, even though these translations are perfectly correct, apparently the terms ch'en and wei were often used interchangeably, and the same is true of several other terms such as tu SE1 (dia- grams) or lu - (records).14 Despite this arbitrariness, out of these four terms only the "esoteric explanations" (wei) are by definition comparable in some way to the apocryphal texts of the Western tradition; the other terms suggest contents that are not found within the framework of Western apocryphal scriptures.

Except for seven wei-scriptures attached to the I ching,'5 the Chinese apocrypha were lost at some point during the T'ang and the Sung dynasties. Their disappearance was probably a conse- quence of the edicts of condemnation issued by Emperor Hsiao- wu of the Liu-Sung (r. AD 457-465), by Emperor Wu of the Liang (r. AD 502-520), and by Emperors Wen (r. 581-604) and Yang (r. 605-616) of the Sui.16 Not until the last of these proscrip- tions, however, did the apocrypha cease to play an important role in the thought of the Chinese literati-as is attested by the fact that scholars who had studied these texts could ascend to the

13 See Cambridge History of China, Vol. I, p. 759. 14 See Ch'en P'an W09, "Ch'en-wei shih ming" :O -, Li-shih yui-yen yen-chiu-

so chi-k'an 11, 1943, pp. 297-316. 15 See Nielsen's dissertation cited above, note 4. Nielsen states (p. 20) that

"the major part of the text [of the Qian zuo du] is devoted to studies within the Yi jing tradition, and it incorporates cosmological elements usually associated with the Confucian New Text School". He also draws parallels (p. 21) between the I ching studies of the New Text scholars Ching Fang GFW- and Meng Hsi S:e. There is a serious difficulty here with the tradition of Meng Hsi, who lived at the time of Emperor Hsiuan (r. 73-48 B.C.) and presumably wrote in chin-wen; as Pan Ku VIEW (32-92) says in the Han shu (Peking, Chung-hua shu-chiu ed., 1962) 30/ 1704, it was, at the very least, not in accordance with the ku-wen version of the I-ching. Hsiu Shen, however, identified the Meng tradition as ku-wen (see Nylan, "The Chin wen/Ku wen Controversy", p. 93). There might be a simple solution to this problem: Pan Ku was referring to the historical situation at the time of Emperor Hsuian, whereas Hsui Shen referred to his own times. During the ca. 150 years that had elapsed since Emperor Hsuian when the Shuo-wen WC was published in 100 A.D., the Meng tradition may very well have decided to incorporate knowledge derived from the ku-wen classic.

16 Sui shu M (Peking, Chung-hua shu-chii ed.), 32/941.



34 HANS VAN ESS

highest positions in the Imperial Academy.'7 Emperor Yang finally ordered that all ch'en-wei scriptures be collected and burned, obviously because their use was deemed to be potentially dangerous to the new dynasty. This final proscription seems to have been much more effective than the one issued against the Christian apocrypha by the Roman church about a century earlier: it led to the almost entire disappearance of the ch'en-wei texts. Not until the Ming and Ch'ing dynasties did remaining fragments of the apocrypha come to be collected and edited by several eminent scholars, such as Ma Kuo-han ,9bESR in his Yii-han shan-fang chi i-shu iUE * M or Huang Shih 3 in his Huang- shih i-shu k 'ao lASSXt. It seems that these recompilations were at least partly inspired by the rise of the famous New Text School of the Ch'ing, which based its arguments on accounts of miraculous events that had, according to the apocryphical scriptures, taken place during the lifetime of Confucius. By this means the New Text school wanted to prove its claim that the master had been an "uncrowned king" (su-wang * E), a Messiah who had laid down the laws for the future generations in the canonical books. 18

Robert Kramers has justly pointed out that the analogy between European and Chinese apocrypha is somewhat remote.19 And yet, all qualifications notwithstanding, the similarities that do remain between them are striking. The most important feature

17 See for example the case of Ma Kuang . in Sui shu 75/1717. 18 See the articles by Chou Yfi-t'ung M ; PM, "Wei-shu yu ching chin-ku-wen

hsuieh" Y in Chu Wei-cheng *f, Chou Yii-t'ung ching-hsiueh-shih lun-chu hsiian-chi J (Shanghai 1983), pp. 40-69 (esp. pp. 59-64); by Harada Masami WMIE, "Shinmatsu shisoka no isho kan-Ko Ya-I shingaku gikei ko o chashin to shite" i o) 1 WW.9< *> #x .4 lT U (Late Ch'ing thinkers' views of wei-shu: K'ang Yu-wei's Hsin-hsiieh wei-ching k'ao), in Yasui Kozan !WWt1, ed., Shin'i shiso no s6go teki kenkyut - ,21,O n? ,, pp.

271-299; and by Kondo Mitsuo jgtk , "Shin-cho keigaku to isho" i (Ch'ing classical studies and the wei-shu), in ibid., pp. 249-269. Compare also Feng Yu-lan 8.:M, Chung-kuo che-hsiieh shih J Shanghai 1934, pp. 1011ff, where it is said that the contents of the wei-shu satisfied the search for the founder of a Chinese religion, of which China was believed to be deficient according to K'ang Yu-wei and his followers.

19 Cambridge History of China, Vol. I, p. 759. Those Chinese Buddhist apocrypha which pretend to be translations from Sanskrit sutras, but are in fact autoch- thonous texts written in China by Chinese authors, would seem to be closer to the Western apocrypha. On these texts see Buswell, ed., Chinese Buddhist Apocry- pha, Honolulu 1990.




justifying the comparison of Chinese apocrypha with their West- ern counterparts is that they covered subjects left open by the canon of the Classics. Also in agreement with the definition of Christian apocryphal scriptures is the fact that wei texts often served as a basis for scholarly positions that could not claim to be rooted in any passage from the Classics. Finally, similar to the apocrypha of the Occidental tradition that report miraculous stories about the life of Jesus, Chinese apocrypha contain some hagiographical material on Confucius and a number of other heroes of the Chinese tradition. Given the fact that the usage of the term "apocryphon" changed considerably over the centuries, and that what is meant by it is not the same for different groups of apocryphal scriptures even in the Western tradition, it is certainly permissible to use it in the Chinese context.

But one major difference between apocryphal texts in the West and in the East should not be overlooked. Christian apocrypha resemble closely certain parts of the New Testament. By contrast, a glance at the seven wei attached to the Changes and at the fragments of apocrypha attached to other Classics, which through quotations in commentaries, encyclopedias and other sources also survived Emperor Yang's edict and the disapproval of contemporary and later scholars, reveals that the ch'en-wei texts did not take the Chinese Classics as their formal model. On the contrary, their technique consisted in elaborating on subjects touch- ed upon but not explained by the Classics, much in the same way as a commentary does. For example, an eclipse in the Book of Odes may be taken by the apocrypha as the basis for long dissertations on astronomical, astrological or prognostication topics; similarly, the mention of a certain sacrifice in a text on rites and ceremonies gives the author of an apocryphon the opportunity to put for-ward his own opinion on the subject. The resemblance between the contents of ch'en-wei texts and the opinions on passages in the Classics contained in the commentarial traditions of the Han can be extremely close, to the extent that the most recent and best recompilation of apocryphal quotations in ancient texts contains some fragments whose provenance is obviously not an apocryphal text, but rather an explanatory note (shuo S) extracted from a commentary.20

20 See the critical edition of sur-vivinlg fragments of the Han apocrypha pre- pared byYasui Kozan 2g L[ and Nakamura Chohachi It3A in the 1960's and published in Tokyo between 1971 and 1993 asJitshiu isho shusei , Vol. 1-6 (cited



36 HANS VAN ESS

So, while in our tradition apocrypha and commentaries were very different things, this difference has never been so obvious in China. Given that, as we just saw, Chinese apocryphal texts were close to commentarial traditions, it would seem logical that con- troversies carried out in commentaries should have left traces in apocryphal texts as well. This has been the traditional Chinese claim since at least the time of the compilation of the Sui shu, where it is stated that one group of erudites, namely, the Old Text scholars of the Han Dynasty, did not make use of apocryphal and prognostical scriptures.21 Because of the doubts that have been expressed against this seemingly all-too-easy picture,22 it seems necessary at this point to look first at the commentarial tradition of the Han, before returning to the main question of the impact of the apocrypha on scholarly thinking.

The Debate on the Commentaries to the Classics in Han Times

In Han China, commentaries seem to have been the medium for a scholarly dispute that took place at about the same time as the apocryphal texts rose to prominence, namely, during the last decades of the Former Han and the first century of the Later Han

below as 'Yasui", followed by the volume and page number). For examples of shuo being quoted as apocryphal texts, see vol. 3, p. 59, or vol. 5, p. 37f. I have dealt briefly with the shuo commentaries in my Politik und Gelehrsamkeit in der Zeit der Han-Die Alttext/Neutext-Controverse (Politics and Erudition in the Time of the Han-the Old Text/New Text Controversy), Wiesbaden 1993, p. 56f. These shuo commentaries could in some cases be identical with apocrypha, but as long as there is no evidence that a given shuo is at the same time mentioned as a distinct apocryphon I think we should still separate the two genres. Apocrypha may look like commentaries to us, but because of the supposed authorship of Confucius, in Han times they were clearly not regarded as the same kind of commentaries as those produced by scholars who lived during that dynasty.

21 Sui-shu 32/941: "Only scholars like K'ung An-kuo L Mao-kung -tX, Wang Huang TA and Chia K'uei Wi rejected them [the apocryphal scriptures] ... They used the texts in ancient scripts (ku-wen) which King Kung of Lu *C M. and King Hsien of Ho-chien Mr1si: of the Han had found. Thus they formed their own opinion, which they called "ancient learning" (ku-hsuieh tf). Some of the erudites of that period rejected and denunciated this [learning], with the result that it did not gain acceptance."

22 M. Nylan, "The Chin wen/Ku wen Controversy in Han Times", pp. 109-112. On p. 110 Nylan says: "Commonsense tells us that there should be no exact correlation between the study of ku wen Classics and abhorrence of the apocrypha."




(ca. 50 B.C-AD 100). Traditionally, the Former Han is described as the period of the so-called New Text learning, whereas the Later Han is said to have been the period during which "Old Text" learning was prevailing.23 New Text learning is said to have been based on Classics written down during the Han in the new script that had been introduced by Li Ssu f i, the infamous chancellor of the First Emperor of Ch'in. Old Text opinions, according to this traditional view, go back to versions of the Clas- sics that probably came into being before the Ch'in dynasty, and were found and collected under various circumstances during Western Han times. Despite this all-too-easy scheme, it seems clear that scholars from the second century of the Later Han, such as the great Cheng Hsiuan Wk (AD 127-200), took in fact a synthesizing approach towards the Classics and were for a number of reasons-most often political reasons-not exclusively biased towards one side or the other.24

The apparently opposing scholarly camps provided, in my opinion, classical and orthodox sanction for the differing positions of the two factions which voted for or against political change during the Former Han dynasty. 25 One early source on the political

23 As an example of this opinion, see P'i Hsi-jui's ,2SS Ching-hsiueh li-shih gf -, especially the preface, pp. 1-4. This enormously influential book, which first appeared in 1906, was commented on by Chou Yfi-t'ung in 1928, and has since been republished in 1959 and 1980 by Chung-hua shu-cha in Peking.

24 For an outline of Cheng Hsuian's scholarly position and his views on politics, see my Politik und Celehrsamkeit, part III and especially part IV, p. 287f.

25 See my Politik und Gelehrsamkeit. My arguments there are based partly on Michael Loewe's views as stated in Crisis and Conflict in Han China, London 1974, p. 12f: "Modernists and Reformists differed over the authenticity of certain versions of ancient texts and over the choice of books for inclusion in the 'canon': the Modernists sponsored the cause of the recently discovered copies written in contemporary script (chzin-wen), while the Reformists favoured those that were written in the obsolete seal script (ku wven). Modernists liked for example the Kung-yang commentary to the Spring and Autumn Annals, while the Reformists introduced first the Ku-liang commentary and later the Tso chuan... Modernists saw the growth of a prosperous economy by way of unrestricted ownership of land and the imposition of a state monopoly of the iron and salt mines..." See also p. 165f. Also compare Hans Bielenstein's contribution on Wang Mang in the Cambridge History of China, Vol. I, p. 239. I believe that I have proven, however, that the Wu-ching i-i E WA!% shows clearly that modernist arguments are based on texts like the Tso-chuzan and the Chiou-1i, which have been classified as texts of the Old Text tradition, whereas reformist policies were founded on the New Text Classics-not the other way round as Loewe thought in 1974. The confusion goes back to Fujikawa Masakazu 4)IIiER and Ch'ien Mu 0., who confuse the two



38 HANS VAN ESS

ideas of these factions is the Discussions on Salt and Iron (Yen-t'ieh lun WRA), probably written down as early as the reign of Em- peror Hsiuan of the Han (73-48 B.C.). In his eulogy at the end of the description of the northern rival of the Chinese, the Hsiung- nu Sjty, Pan Ku testifies as well to the existence of two factions. Discussing the opinions of Han officials from the second century B.C. down to the end of the Former Han, he claims:

Reducing it to the main principles, two positions remain: the erudites with their red girdles advocated self-defense and peace, whereas the armoured men advocated attack and war.26

In this chapter Pan Ku seems to be talking solely about military matters. But Michael Loewe has shown that the assumption of a division of Han officials into two camps produces the most plausible picture of the entire political constellation of the era from Emperor Wu's reign (140-88 B.C.) until Wang Mang's usurpation in the first years AD-at least as far as Pan Ku's History of the Former Han is concerned.27 These factions opposed each other over a

schools as well (see for example Fujikawa's Kandai ni okeru reigaku no kenkyui ttf-:ttco40fit, Tokyo 1968, p. 139, and Ch'ien's Liu Hsiang Hsin fu-

tzu nien-p'u CIJR'yf , in Liang Han ching-hsiueh chin ku wen ping-i M -7mtSCIV, Taipei 1971, p. 89, as discussed in my Politik und Gelehrsamkeit, p. 153, n. 4 and p. 212, n. 27). Recently, Loewe has changed his mind: "How far a distinction should be drawn at this time between the two attitudes later to be described as Ku-wen and Chin-wen may perhaps not be known... It remains to be shown whether there was a definite association between those who proposed the abandonment of the shrines and some of the specialist schools of learning." See his "The imperial tombs of the Former Han dynasty and their shrines", T' oung Pao LXXVIII (1992), pp. 302-340, on p. 340.

26 HS 94B/3830. 27 In my articles "Die geheimen Worte des Ssu-ma Chien", Oriens Extremus 36

(1993), pp. 5-28, and "The meaning of Huang-Lao in Shiji and Han shu", Etudes chinoises XII, 2 (1993), pp. 161-177, I have argued that in both the Records of the Grand Historian and the Book of the Han a similar division is found for the second century B.C. Of course, history is always more complicated than the simple bipolar representations of historiographers would allow. It is, however, interesting to note that for the sake of clarity the earliest two dynastic histories written in China evidently worked with the hypothesis of a dualism in Chinese politics. More or less the same can be observed in the "southerners" versus "northern frontier interests" conflict described, according to H. Bielenstein (The Restauration of the Han Dynasty, part IV, Bulletin of the Museum of Far Eastern Antiquities 51 [1979], pp. 1-300), in the History of the Later Han. For another case of bipolar historiographical writing in another Chinese dynasty, we could cite the famous struggle between the Huan 4- and Hsieh : families during the Chin dynasty. Obviously, even though reality may have looked much more complicated than historians




wide range of politico-institutional matters, including the question of whether to take a peaceful or an aggressive stance towards the neighbours of the Chinese empire.28

Finally, about 150 fragments survive from the Wu-ching i-i AWA-A (Different Meanings of the Five Classics), a collection on institutional and intellectual problems which preoccupied Han scholars. This book, dating from about the time of the completion of the Book of the Han, was written by Hsiu Shen 5F-1h (ca. 55- ca. 140), better known today as the author of the Shuo-wen chieh- tzu WC4*, China's first character dictionary. The surviving fragments show the same clear-cut division of opinions on institutional matters that was significant during the Han, as seen in the Discussions on Salt and Iron and the Book of the Han. At the same time, they are our best source for studying the content of the scholarly debates of the day regarding different interpretations of the Classics.

Hsui Shen listed a great number of items in his book, all apparently discussed by adherents of the "new" and the "old" approaches to the Classics. As a separate entry he added his own deliberations concerning each of these problems. A century or so later, Hsiu Shen's conclusions were in turn critically evaluated by Cheng Hsuian in a book entitled Refutations on the Different Mean- ings of the Five Classics (Po [Hsiu Shen] Wu-ching i-i ,K Scrutiny of the fragments reveals a consistent agreement between the arguments labelled by Hsiu Shen as "Old-(Text?)"29 (ku W)

tell, the scheme of pro and contra is common in Chinese historiography-and it may have been at the root of the construction of Han political conflict in terms of the Old Text/New Text controversy as well.

28 See especially Loewe's Crisis and Conflict. 29 As Michael Nylan has pointed out, the Different Meanings of the Five Classics

(hereafter WCII) never speak of "chin-wen 4k" (new-text) or "ku-wen t3." (old- text), but only of "chin" and "kt". There is, in fact, an exception to this rule in WCII, as edited by Ch'en Shou-ch'i IWo, in Huang Chl'ing ching-chieh QM,9 1248-1250, p. 1250/55b, where a "chin-wen Shang-shu Ou-yang" opinion is quoted; but Ch'en Shou-ch'i is probably right when he says that this must be an error because P'ei Sung-chih W (352-451), in his commentary to the San-kuo chih (Peking, Chung-hua shu-chMi ed., 1959, 11/360), is the first to speak of a chin-wen Shang-shu, and that scholars from the Han only spoke of "chin". Although it is not quite clear whether there really is a difference between these terms-Ch'en Shou-ch'i apparently did not think so-for the sake of clarity I will speak only of "new" and "old" in this article. In fact, it seems that Hsfi Shen wanted to imply that one tradition was old and the other new-and it is easy to imagine that an "old" tradition could claim more authority than a new one-not that one used



40 HANS VAN ESS

exegesis, and the persuasions of the political faction of the first century B.C. that opposed any deviation from the strongly au- thoritarian approach to internal and external representation of the dynasty prevailing at the time of Emperors Wu and Hsuian. Because this group favoured the political and bureaucratic system newly created by the Ch'in and continued by the Han, Loewe has called its adherents "Modernists" who were concerned with a choice of particular politics which strengthened the role of the state.30

During the time of Hsuian's successors, Emperors Yuan (r. 48- 33 B.C.) and Ch'eng (33-7 B.C.), protest against these policies, which had drained the imperial coffers and demanded heavy contributions from the emperor's subjects, was uttered more and more openly. The protesters eventually succeeded in implement- ing large-scale institutional changes. Among the demands of this group was a return to what they thought was the old system of the Chou-hence Loewe's term "Reformists" for it. Most reformist statesmen had formerly studied at the Imperial Academy (t'ai- hsueh t&) the Classics in the new character version, which was in fact the version easily accessible to the ordinary scholar studying at Ch'ang-an.31 The political arguments brought forward by this

texts in old characters and the other texts in the new style. In any case, it seems clear that the texts of the "old" traditions were difficult to read, either simply because the whole basic text was written in ancient characters (whatever that means), or because there were many ancient characters in these texts. See Nylan's excellent analysis of this problem in "The Chin wen/Ku wen Controversy in Han Times", pp. 88-97.

30 For Loewe, the "full force of modernist politics" was at play during the reign of Emperor Wu of the Han. See Cambridge Histoiy of China, Vol. 1, pp. 152- 179.

31 Archeological evidence has shown that some versions of Han Classics, such as the I-li ZZ discovered in 1964 at Wu-wei RO, or the so-called Fu-yang Shih- ching g (on which see the articles "Fu-yang Han chien <Shih-ching>"

W1%%ME Wen-wu 1984, 8, pp. 1-13, and "Fu-yang Han chien <Shih-ching> chien-lun" , ibid., pp. 13-21, by Hu P'ing-sheng jMT and Han Tzu-ch'iang "#Z)' were obviously not written either in the "ku" or in the "chin" mode. This does, however, only prove the not very surprising fact that in Han times there existed different copies of these texts and that writing was not yet completely standardized. The establishment of chairs for erudites of the Classics at the Imperial Academy obviously served the purpose of establishing an ortho- doxy. Certainly more important than variants of characters in two manuscripts of the same text was the fact that ku and chin traditions relied on different scriptures, viz. the Tso-chuan vs. Kung-yang, or Chou-li vs. some Li-chi chapters.





42 HANS VAN ESS

Yu-wei's l (1858-1927) ideas,35 has wondered whether the Wu-ching i-i really can be treated as a source revealing a strong conflict in Han times. He argues that a strict separation of Han scholarship into two camps with distinct features was first suggested as late as 1886 by Liao P'ing in his Chin-ku hsiieh-k'ao ~ *.36 Li identifies several major arguments in Liao P'ing's book: that the opposition between the Old Text and New Text schools was like between fire and water; that the first author to mention a chin-wen and a ku-wen tradition at the same time was Hsui Shen in his Wu-ching i-i; that Hsui Shen drew a strict line between Old Text and New Text arguments, adhering himself to the Old Text school; that this line only broke down with the synthesizing efforts of Cheng Hsiuan at the end of the Later Han; and, finally, that scholars were not divided over philological arguments on the Classics but over different opinions on the ritual system of the Han.37

Li Hsuieh-ch'in criticizes Liao P'ing for committing several mis- takes. He quotes several instances where Hsiu Shen agrees with "new" opinions in his Different Meanings of the Five Classics. He also shows that there are examples of "new" traditions being in agreement with old ones. But we could as well argue, on the basis of the findings of R.A. Miller more than forty years ago, that, notwithstanding the fact that Li Hsfieh-ch'in's observations are in every quoted case perfectly correct, enough evidence remains that in the overwhelming majority of instances Hsui Shen did indeed adhere to "Old" opinions, and that in more than ninety percent of all cases old traditions were in fact in opposition to new opinions.38

` The Ch'ing scholar Chang T'ai-yen tA, K'ang Yu-wei's most important opponent, even said that K'ang plagiarized Liao P'ing's ideas. See his "Ch'ing ku Lung-an fu hsuieh chiao-shou Liao chuin mu chih-ming" L BM2Z, in Chang T'ai-yen ch'ian-chi W it- 5 (Shanghai 1985), p. 264f. Li Hsuieh-ch'in, however, quotes Liao P'ing to the effect that Liao taught K'ang Yu- wei about his ideas (see the article cited note 34, p. 127).

36 Li does not make reference to the account of Sui shu 32/941 on the "ancient learning" of the Han.

37 Ibid., pp. 128-130. The Chin-ku hsiueh-k'ao has been republished in Liao P'ing hsiueh-shu lun-chu hsiuan-chi (1) f Ch'eng-tu 1989, pp. 29-112. For Liao's main arguments see also his Liu-Pien chi , in ibid., pp. 535-626, esp. the first two pien, pp. 545-547.

38 See R.A. Miller, "Problems in the Study of 'Shuo-wen chieh-tzu"', Ph.D. dissertation, Columbia University, 1953, pp. 63ff, and Miller, "The Wu-ching i-i of Hsui Shen", Monumenta Serica 33 (1977-1978), pp. 1-21.




The most important point which Li Hsuteh-ch'in advances against Liao P'ing is that he cannot offer any evidence for his claim that in writing the Different Meanings Hsa Shen wanted to present a complete ritual system, and, furthermore, that Hsiu's eclectic approach is itself an argument against the assumed existence of two different systems.

The difficulties which Li Hsiueh-ch'in brings on himself with this line of argument are plain. He fails to see that, when he presented opinions from "Old" and "New" traditions, Hsiu Shen was not discussing the opinions of his own day, but arguments that were at his time already more than a hundred years old.39 Neither Han society nor its institutions were static: it was almost impossible in Hsui Shen's days to agree in its entirety with an institutional system that was in some respects simply obsolete. Eighty to ninety percent is in this context a surprisingly high rate of agreement: it makes Hsiu Shen the best ku author one could possibly be in his time. To call his approach eclectic because his acceptance of ku opinions involved only ninety percent, not the totality, of 150 cases does not seem to me to be very convincing. Hsiu Shen discussed a wide range of institutional matters that together added up to a whole system. Therefore, an explicit statement of his own that he wanted to present the best system of his choice was not needed: the implication was obvious. To demand such a statement is anachronistic.40

As far as intellectual history is concerned, Michael Nylan has called for a reorientation of the focus of research away from textual and towards institutional and intellectual problems. 41 She

39 As already said, the arguments brought forward in Wu-ching i-i are those of the political discussions of the last fifty years of the Former Han. Hsut Shen is a late ally of the modernists of that time, described by him as ku scholars.

40 Given the fragmentary nature of Wu-ching i-i and the fact that Hsui Shen composed a preface to the Shuo-wen chieh-tzu, it is quite possible that there existed a preface to the Wu-ching i-i as well, but that it was lost later. Thus, it is probable that it will not be possible in the future to get more substantial textual basis for our argument on this point. We can, however, speculate about the words that Hsui Shen might have used in such a preface on the ground of the actual contents of the Different Meanings-and I would venture to say that Liao P'ing was right when he said that Hsfi Shen wanted to present the best system for the Han.

41 M. Nylan, "The Chin zven/Ku wven Controversy in Han Times", in particular p. 134f. Nylan follows Michael Loewe in singling out six "intellectual problems", namely worship of ti W versus t'ien i, belief in expediency vs. the insistence on principle, love of imperial expansion vs. desire for retrenchment, desire to con- trol the economy vs. desire to foster laissez-faire and private initiative, devotion to



44 HANS VAN ESS

has shown that the Han shu and Hou Han shu as well as other Han texts do not appear to uphold a clear-cut division among Han scholars along the lines of their respective positions on Old Text and New Text problems. She has therefore suggested that famil- iarity with the late Ch'ing Old-Text/New-Text controversy may have "coloured the perception of modern historians ..., causing them to read into Han history a comparably virulent debate among Han scholar-officials".42

From Nylan's meticulous research into historical texts it is by now quite clear that Han scholars were not primarily concerned with disputes over chin-wen or ku-wen etymology.43 This in itself is a highly interesting finding, something which until then had not been stated so explicitly, at least not in the West. Nylan is correct when she says that Ch'ing scholarship has obscured the true nature of the conflict and that the whole story of an Old-Text domi- nated period in the Later Han must be wrong. With respect to the Han shu and Hou Han shu it is certainly right to say that almost no complete information on theoretical positions is preserved.44

Yet, I do think that, when looking for consistent political agendas of either chin-wen or ku-wen scholars, the Different Meanings of the Five Classics, which has been quoted only incidentally in recent scholarship, may be of decisive importance. Although the text is now in fragments, most of these are extracted from subcommentaries to the Classics written as early as T'ang times. And, as a comparison with quotations from other sources shows, T'ang scholars seem to have quoted the Different Meanings of the Five Classics very carefully, which makes it a reliable source.

new music vs. traditional music, and the acceptance of esoteric cults vs. more practical measures. Although I would hesitate to call these issues "intellectual problems", and rather see them as political issues, I fully agree with her conclusion on p. 135: "It is on such issues that we must continue to work if Han learning is ever to regain its rightful place in intellectual history as the foundation for the Confucian ideology of imperial China."

42 Ibid., p. 86. 43 Ibid. 44 One of the few exceptions that could be quoted is Chia K'uei's discussion

of the differences between the Kung-yang and the Tso commentaries to the Sp7ing and Autumn Annals which certainly seeks to provide a theoretical basis for two different scholarly approaches to the problem of the position of the emperor and of the dynasty in their relationship to the empire and the bureaucracy. See Hou Han shu (Peking, Chung-hua shu-chui ed., 1965, hereafter HHS) 36/1236ff.




In my view, the dynastic histories-first of all the Book of the Later Han, which was composed much later than the occurrence of the conflict-should not be quoted anymore as primary evidence concerning a scholarly dispute over the Classics which was pursued within the commentarial tradition and which, as Nylan has shown, was not a topic that the dynastic histories focussed on. Clearly, the commentaries themselves, including the Different Meanings of the Five Classics, must be considered as the primary source, and the dynastic histories as only secondary. From the careful division which Hsui Shen provides between "chin" and "ku" opinions, it is indeed possible to reconstruct a "political agenda" of chin and of ku adherents. And, as has been noted above, these agendas are remarkably consistent with those of the reformist and modernist factions of the first century B.C., respectively.45

45 I do not follow Nylan in her argument that the "occasional use [of the "tags" chin or ku in Wu-ching yi-yi] suggests that chin wen/ku wen differences hardly preoccupied the authors' minds" (Nylan, "The Chin wen/Ku wen controversy", p. 123). These "tags" are very regularly attached to the opinions quoted, and where they are not, this only proves that Hsui Shen could assume that every reader of his book would in his time know that Tso-chuan or Chou-li were texts attached to a "ku" tradition, whereas the Ta Tai Li-chi * was a "chin" text, as were also the Ou-yang Shang-shu SkM.C and most other books circulating in the scholarly community of the time. The Wu-ching i-i fragments were collected mainly from the subcommentaries to the Thirteen Classics and from certain encyclopedias-it is very likely that the authors of these works were not completely consistent in quoting their sources. Some "ku" or "chin" might thus have been omitted by them and others added. The arrangement of the text, which usually mentions the "chin" traditions first and the "ku" traditions second-and which in almost every instance has preserved a "chin" and a "ku" explanation-provides another argument which makes it quite obvious that a "ku"/"chin" conflict stood at the centre of the authors' interest. Compare R.A. Miller, "The Wu-ching i-i of Hsfi Shen", especially on pp. 12-19. Miller shows that there are some cases in Wu- ching i-i in which an Old Text tradition is aligned to a new one against another "new" interpretation like, for example, Tso-chuan and Kung-yang against Ku-liang (two cases in WCII, Miller, "The Wu-ching i-i", p. 13f). But these few deviations do not contradict the otherwise homogeneous picture. The deviations seem to re- flect the complicated situation resulting from the fact that, as a result of the Shih- ch'ui HU discussions (on which see Tjan Tjoe Som, Po Hu T'ung, The Comprehensive Discussions in the White Tiger Hall, vol. 1, Leiden 1949, pp. 128ff) held in 51 B.C., prior to the "discovery" of the Tso-chuan, the court had established the Ku-liang chuan in addition to the Kung-yang. As the Ku-liang chuan was a competitor to the Kung-yang it is only natural that there must have existed divergent opinions within these "chin" traditions. When the Tso-chuan was found its proponents had in some instances to decide on the question with which one of these texts to align. It has become clear, however, from the tables provided by Miller, that for



46 HANS VAN ESS

To sum up, my argument, which is based mainly on Hsfi Shen's Wu-ching i-i, assumes-in agreement with the traditional argument-a major split along chin and ku lines within Han scholarship (as opposed to philosophy or historiography). Scholarship was important for the system of recruitment of officials, whereas philosophy was not written for such purposes. Thus, it is certainly correct not to believe in a major split as far as thinkers like Wang Ch'ung T-t (ca. AD 27-97) or Yang Hsiung %X (53 B.C.-AD 18) are concerned.46 In my view there existed a conflict among classical scholars, however, a conflict which, while also philologically important, was modeled around a pre-existing political debate. The so-called "old" opinions supported the policies of Emperors Wu and Hsiuan, whereas the "new" ones supported those of the reformist movement that began under Emperor Yiian. Wang Mang, the fact that he is usually called a proponent of the "Old Text School" notwithstanding, took a synthesizing approach using "old" and "new" opinions according to his own interests.47 The scholarly exegesis provided the appropriate language in which arguments and personal opinions on policies could be expressed.

Because of the fragmentary nature of our principal sources, it is admittedly difficult to settle the matter definitively. Yet, despite the fact that the Old Text/New Text Controversy has been one of the best studied topics in recent Chinese intellectual history, the new and better editions of some texts which have recently appeared allow a fresh approach to the sources.48 The purpose of

Hsui Shen the main conflict raged between opinions based on the Tso-chuan and the Kung-yang.

46 Feng Yu-lan certainly misunderstood the controversy when he drew heavily on the writings of these philosophers for his account.

47 It seems that Wang Mang used "chin" scholarship for the reform of the political and administrative system, and "ku" for his own legitimation. See my Politik und Gelehrsamkeit, p. 285. I am not concerned with the scholars of the second century AD here because, as the political situation had totally changed by that time, the conflict did not go on in its original form any more.

48 For instance, with the help of the new Hsin-pien chu-tzu chi-ch'eng *#i3g+tR edition of the Comprehensive Discussions at the White Tiger Hall (Po-hu t'ung-i bS,AIR), Peking 1994, it should be easy to substantiate my impression that this text assembles almost exclusively opinions labelled "chin" in the Wu- ching i-i and that it should not be considered a melange of "chin" and "ku" ideas. Compare Tjan Tjoe Som, Po Hu T'ung, Vol. 1, Leiden 1949, p. 166f. See also my Politik und Gelehrsamkeit, p. 82. Hihara Toshikuni H# IIJf, "Byakko tsfigi kenkyu shoron: tokuni reisei o chuishin toshite" tz- <IM1$ 144m' UV, in




the present article and the choice of its topic for investigation must be seen in this context: the apocryphal texts contain abun- dant materials likely to shed new light on the nature of Han scholarship.49

Institutional topics in the apocryphal texts and in the Wu-ching i-i

As has been recalled in the first part of this article, there has been considerable dispute over the Old Text/New Text controversy; and, as the attitude of the Han classicists towards the apocryphal scriptures has constituted one of the major criteria for most Chinese authorities writing on that controversy, the apocrypha should logically contain materials supporting one or the other view. On the basis of materials provided by the Histo?y of the Later Han, it has been traditionally assumed that ku-wen scholars rejected these texts, whereas chin-wen scholars adhered to them. As even the existence of the whole controversy in Han scholarship has been questioned, this traditional argument has naturally come under attack as well. That too many different subjects are dealt with in the apocryphal scriptures is the main argument for

Nihon Chi-goku gakkai ho 14 (1962), pp. 63-78, quoted in Nylan, "The Chin wen/ Kit wen Controversy", p. 118, states that opinions from both schools were used in the compilation of Po-hu t'ung-i. It is true that there are some opinions remind- ing us of "ku" positions in it; the overwhelming majority are, however, "chin". See on this problem Chung Chao-p'eng 0*, Ch'en-wei lun-liie 20o:, Shenyang 1992, pp. 140-146. The problem with Hihara's argument against the traditional description of the Po-hu t'ung-i as a "new text" document is the same as that with Li Hsfieh-ch'in's discussion of Hsui Shen: a few exceptions to the rule are taken to refute an otherwise coherent thesis. In this context see also Benjamin Elman, Classicism, Politics anid Kinship: The Ch' ang-choou School of New Text Confucianism in Late Imzperial Chzina, Berkeley 1990, p. 190, where he quotes Chuang Shu-tsu 4aDlL (1751-1816) as having shown that "the Tso chuan's interpretations of the Annals had not been discussed at the White Tiger Hall conference". Given the Chuang family's bias towards the Kung-yang commentary we should, of course, be aware that Ch'ing dynasty discussions influenced Chuang Shu-tsu's research. As far as I can see, however, Chuang was correct in his analysis.

49 Only in 1993 has the last of the six volumes of Yasui Kozan's and Naka- mura Chohachi's fine edition of apoclyphal texts been issued in its final version (see note 16). Needless to say, this new edition has greatly enhanced our knowledge of the contents of the apocryphal scriptures of China, and enormously improved our textual basis compared to the editions used by previous scholars writing on Han scholarship.



48 HANS VAN ESS

the claim of some modern scholars that there can be no straightforward "correlation between the study of the ku wen Classics and abhorrence of the apocrypha".50 Therefore, an analysis of the doctrinal positions held in the apocryphal texts should be a good way of investigating whether there was indeed no link between the ch'en-wei texts and the controversy, or whether there is some truth in the traditional claim to the contrary. To date, the arguments advanced pro and contra have, as stated above, been generally based on secondary sources, viz. the dynastic histories. But the value of the latter to decide our problem does not seem very high as long as we do not know how we are to read the relevant statements on the controversy. When he wrote that Chia K'uei (AD 30-101) read texts belonging to the Old Text tradition as well as apocryphal scriptures, for example, did Fan Yeh 7 (398- 445), the author of the Book of the Later Han, think that this was the normal attitude of an Old Text scholar? This would immedi- ately contradict the existence of an exclusive connection between chin-wen and the ch'en-wei literature. Or did he include the story precisely because Chia K'uei was one of the first exceptions to a rule that was previously valid? Pending more detailed knowledge of Fan's historiographical techniques, it is impossible to decide on such a question. This means that neither the traditional Ch'ing-or earlier-reading that identified an Old Text/New Text controversy in the dynastic histories, nor M. Nylan's reading a rebours, can provide a definitive solution. Additional evidence is needed to get closer to the actual contents of the discussions that took place among Han scholars. What follows is therefore devoted to comparing data culled from the apocryphal fragments with the positions that were held, according to Hsui Shen's Differ- ent Meanings of the Five Classics, by scholars consistently adhering either to "old" or to "new" opinions.

As it happens, a first step in this direction has already been taken by Liu Tsung-li, who looked for similarities between subjects dealt with in the apocrypha and in three texts said to be chin-wen, viz. the Shang-shu ta-chuan (Mi.1 , the Kung-yang chuan and the

50 M. Nylan, "The Chin wen/Ku wen controversy", pp. 108-112, rejects the traditional claim for a connection between the chin-wen tradition and apocryphal texts on the ground of contradictory materials in the dynastic histories. I have argued similarly in Politik und Gelehrsamkeit, p. 76f, and in "The Old Text/New Text Controversy: Has the 20th Century Got it Wrong?", T'oung Pao LXXX (1994), pp. 146-170, on p. 158.




Ch'un-ch'iu fan-lu 7W , and by Chung Chao-p'eng in his recent book on apocryphal texts.51 Although Liu's article does provide evidence showing that there should be some connection between chin-wen scholarship and the apocryphal scriptures, his sample is too limited to allow any definitive statement. One may justifiably raise the question whether, had he also looked for parallels between ch'en-wei scriptures and texts of the ku-wen tradition, for example the Tso-chuan or the Chou-li, he would not have found counter-evidence establishing the same connection as with chin-wen scholarship. In sum, faced as we are with Nylan's challenge to the traditional approach, an examination of La's thesis based on further source materials seems in order. And no text is more suitable for this than Hsui's Different Meanings, since it contains opinions classified as "new" as well as their "old" op- posites.

Sadly, most of the subjects discussed in the apocrypha are not related to the concerns of the commentaries to the Classics written during the Han.52 Still, there remain about ten topics which are discussed both in the wei-shu and in the Different Meanings (from here on WCII). We will thus deal with that small proportion of both groups of texts representing their intersection. This con- sists mainly of the fragments concerned with institutional topics.

The hsia and ti sacrifices53

Because the Book of the Later Han assigns great importance to the apocryphal texts with respect to that period, a first question to ask is: are these texts quoted in WCII, and if so, to which traditions do they belong? The answer is straightforward: WCII in its present form does not contain a single explicit mention of the apocrypha.54 They are, however, quoted three times by Cheng

51 Lu Tsung-li MgJ', "Wei-shu ya hsi-Han chin-wen ching-hsfieh" 4*SYQNMA :tES in Yasui Kozan, ed., Shin'i shisd no sdgo teki kenkyui , Tokyo 1984, pp. 395-426; Chung Chao-p'eng, Ch'en-wei lun-lute (see above, note 48).

52 On the overall contents of the apocryphal texts, see Anna Seidel's article and Jack Dull's dissertation mentioned above, notes 5 and 6.

53 On this and the following section, compare also Yasui Kozan, Isho no seiritsu to sono tenkai t e R, Tokyo 1979, pp. 46-57.

54 Since we know already that, generally speaking, Hsu Shen preferred "ku" over "chin" opinions (see R.A. Miller's study quoted above, note 38), this fact may be read as an argument in favor of the traditional opinion that Old Text scholars did not make use of the apocrypha.



50 HANS VAN ESS

Hsiuan in his rebuttals (po ,) of Hsut Shen's opinions.55 The first instance of a ch'en-wei quotation by Cheng Hsuian is from a prognostication text concerning the rites, the Li-ch'en X-,, with the following short sentence: "The Yin a sacrifice [takes place] at an interval of five years." Cheng himself adds, by way of explanation: "The Yin sacrifice is also called a ti ifir sacrifice."56

The ti sacrifice, usually discussed together with its counterpart, the hsia sacrifice, was a special service for the ancestors of a dynasty whose ancestral tablets had been moved out of the temple of the dynastic founder after five generations. The ceremony of moving out the tablets-called mieh %, lit. "to destroy"-was introduced because it was impossible to make frequent offerings to all the rulers of a long-lasting dynasty. In order not to be deemed unfilial to their ancestors, rulers had to provide sacrificial offerings to the tablets that had been moved out of the temple at least at some points during their reign. A discussion of the correct intervals at which the hsia and ti sacrifices had to be performed can be found in the Book of the Han and the Book of the Later Han, as well as in the chapters on imperial sacrifices in the later stan- dard histories and in all the major encyclopedias.57 The subject was obviously of major concern to all ritual specialists, from the Han down to the end of the Chinese empire in 1911.

The problem first arose during the first century B.C., when several statesmen, led by the famous reformist official Wei Hsiuan- ch'eng : contended that only five ancestors-the four emperors directly preceding the current ruler plus the founder of the dynasty-should receive regular sacrifices,58 while all others should be assembled only once every five years for a hsia and a ti sacrifice. A similar statement is also found in Yasui's collection of

55 For Cheng Hsuan's scholarship and the ch'en-wei, see Lu K'ai [91, Cheng Hsiian chih ch'en-wei hsiueh Wt5Xg, Taipei 1974.

56 WCII, recompilation of Ch'en Shou-ch'i 1 in Huang Ch' ing ching-chieh 0*AI9 1248/38b; Yasui 3/82. Yasui thinks that the second sentence belongs to the quotation from the Li-ch'en. This is not very probable, because Cheng Hsuian quotes this text in order to prove his own opinion concerning the correct interval of the ti sacrifice: he has therefore to explain that the Li-ch' en is also discussing the same sacrifice.

57 In the Book of the Han this discussion is inserted into the biography of Wei Hsuian-ch'eng ; (HS 73/3108ff). In the Book of the Later Han the subject is discussed in the biography of the ritual specialist Chang Ch'un W*i- (HHS 35/ 1193) and in the ninth of Ssu-ma Piao's . treatises (HHS ,i 9/193ff).

58 HS 73/3129.





52 HANS VAN ESS

The solution proposed by the Li chi-ming-cheng, namely, a peri- odicity of three and five years respectively for the hsia and ti sacrifices, is exactly what came to be adopted by Emperor Kuang-wu, the founder of the Later Han, in accordance with the request of the late Han ritual specialist Chang Ch'un WM (who had studied a "new"5 tradition of the rites). The text rejects the "old" claim that the ti sacrifice was only to be performed at the completion of the mourning period for a ruler, but accepts the Kung-yang ("new") claim of a performance every five years, with the slight variation that a hsia sacrifice was added every three years.63 Thus, the Li-ch'en quoted by Cheng Hsiuan and Yasui's apocryphal text on the same subject are, if not in exact agreement with the "new" side, at least considerably closer to it than to the "old" side.

Ritual structures: the Hall of Light, the Circular Moat, and the Terrace of the Spirits

1. The Hall of Light The second of the three apocryphal texts quoted by Cheng

Hsiuan is the Hsiao-ching yiian-shen-ch'i : which according to Cheng was the basis for an opinion expressed by a certain Shun-yii Teng WY-TE, who lived during Wang Mang's reign. Shun- yui Teng's opinion belongs to the "new" explanations concerning the correct site of the Hall of Light (Ming-t'ang PMJ). The quotation goes:

The Hall of Light of the Chou was at the yang a side of the capital (kuo gi), more than three but less than seven miles away, in the position of the cycli- cal signs ch'en R and chi E.64

2.6, that here what was performed was a ti sacrifice (Shih-san ching chu-shu +tEMigE, Chung-hua shu-chiu ed., 1980, p. 1838A). The Sung scholar Yeh Meng- te 3* (1077-1148) writes in his Ch' un-ch' iu chuan *pMk (Ssu-k'u ch'uian-shu ed.), 11/9a, that ti was the term for the sacrifice of the son of heaven, whereas the same ceremony was called hsia with regard to the feudal lords. This opinion is, however, not found in the old commentaries.

63 This variation was most certainly meant as a compromise between the conflicting extreme positions, one of which initially did not want to "destroy" the tablet of any emperor, while the other considered that it would be enough to give regular offerings to five ancestors. To this opinion compare the "new" scholar Ho Hsiu's fJPc subcommentary to Kung-yang, Wen 2.6, Shih-san ching chu- shu p. 2267A.

64 WCII 1249/la-2a. Compare Yasui 5/35, quoting the T'aip'ing yii-lan I3Tit 533 and Yiu-hai iffr 95 as well as the chapter on rites and ceremonies in

the Chiu T'ang shu "*. See also Yasui 4B/24.





54 HANS VAN ESS

2. The Circular Moat and the Terrace of the Spirits What is certainly the most famous story about a so-called "Old

Text" scholar rejecting apocryphal texts can be found in Fan Yeh's Book of the Later Han. According to this story, Huan T'an I0 M turned down Emperor Kuang-wu's request to define the correct location for the Terrace of the Spirits (ling-t'ai E )-another ritual structure whose location, purpose and appearance were highly controversial-with the help of prognostication texts (ch'en 2). He excused himself on the grounds that he did not read prognostication texts and was therefore unable to provide an answer.72

We have just seen that both the apocryphal texts and the "new" tradition proposed to establish the site of the Hall of Light based on an analogy with the firmament. During the Han this method seems to have been the actual practice for building the Hall of Light, as well as all other ritual structures: at least this is what excavations carried out in recent years suggest.73 We know that both Wang Mang and Emperor Kuang-wu built these structures in the ch'en-chi RE location, south-south-east of the capital, just as the "new" traditions and the apocrypha, based on a comparison between heaven and earth, demanded. Huan T'an's opposition to defining the correct location of ritual structures on the basis of prognostication texts meant that he also had to reject the "new" tradition.

Thus, it seems possible that it was this kind of correlative thinking that Huan T'an was objecting to, and not apocryphal scriptures per se. But there are other possible explanations for his answer to Kuang-wu. In WCII the "old" view concerning the Ter- race of the Spirits is that it has to be situated within the temple of ancestors and surrounded by a moat filled with water, called the Circular Moat (Pi-yung W) :74

The explanation attached to the Tso-shih: The Son of Heaven's Terrace of the Spirits is situated in the temple of ancestors (t'ai-miao 7JZT). It is en-

72 HHS 28A/961. 73 On the ritual structures of the Han and Wang Mang in Lo-yang and

Ch'ang-an, respectively, see Wang Zhongshu, Han Civilization, New Haven 1982, pp. 10, 30 and 38-40, and figure 37.

74 The two characters differ according to interpretation. "Pi" is sometimes written C. "Yung" is often written with the "earth" radical, as !, and in several instances the character a, or one of its derivatives like E or 9, is used.




circled (yung *) by the Lake of the Spirits (Ling-chao s This is called the Circular Moat.75

According to this account, neither the Ling-t'ai nor the Pi-yung were considered as independent structures: they had to be lo- cated inside the temple of the imperial ancestors. As mentioned above, the "old" tradition also held that "Hall of Light" was the name for the temple of King Wen. If we put together all the relevant "old" explanations, the result is that all three ritual structures would have to be integrated into a single building.76 This, of course, is in striking contradiction with actual Han (and Wang Mang) practice and with the "new" opinion, which proposed three separate buildings: the Hall of Light where the Son of Heaven promulgated his ordinances, the Terrace of the Spirits where he watched the ethers, and the Circular Moat which was his academy.77

The ch'en-wei texts do not contain any evidence supporting the "old" opinion. On the contrary, they are consistent with the "new" explanation of the Han-shih O#J tradition, according to which the Pi-yung is a building where the Son of Heaven "honours and serves the Thrice Elderly and the Five times Venerable (san-lao wu- keng-t)"78

"New" opinions are also close to what the apocryphal texts have to say on the Terrace of the Spirits. In WCII the following Kung- yang explanation is quoted:

The Son of Heaven has three terraces whereas the feudal lords have [only] two terraces. The Son of Heaven has a Terrace of the Spirits in order to watch the firmament; he has a Terrace of the Seasons in order to watch the four seasons; he has a Garden Terrace in order to watch birds, beasts, fish and turtles.

The feudal lords should have a Terrace of the Seasons and a Garden Ter- race. They are lowly and [therefore] not allowed to watch the firmament, so they do not have a Terrace of the Spirits.

Likewise, we find in the apocrypha a statement to the effect that

75 WCII 1249/52a-53a. The poem on the Terrace of the Spirits in the Book of Odes, "Ling-t'ai" (Mao 242), is the locus classicus for the terms Terrace of the Spirits and Lake of the Spirits (J. Legge, The Chinese Classics, III, The Book of Odes, p. 456; B. Karlgren, The Book of Songs, Stockholm 1950, p. 196).

76 Presumably the already existing temple of the ancestors. The conclusion should be that the "old" tradition opposed the building of new ritual structures. Compare my Politik und Gelehrsamkeit, chapter 3, pt. 8.

77 WCII 1249/52a-53 quotes a Kung-yang and a Han-shih Oli explanation. 78 WCII 1249/52a-b and Yasui 5/36 and 69.



56 HANS VAN ESS

"The Son of Heavens's Terrace of the Spirits serves to watch the border between heaven and man",79 and that "The Son of Heaven has a Terrace of the Spirits in order to watch heaven and earth, the feudal lords have a Terrace of the Seasons in order to watch the four seasons".80

By contrast, the following "ku" Mao-shih MA explanation gives the "old" opinion:

The Terrace of the Spirits is not used for watching and overseeing. "Spirit" means essence (ching g). The essence and the light (ming P) of the deities are called "spirit" (shen 1#).81

It is clear that this "old" explanation runs counter to the apocrypha, whereas the "new"9 one agrees with them.

Let us note, before proceeding further, that the third apocryphal text quoted by Cheng Hsuian, a scripture called Shang-shu k'ao-ling-yao f which deals mainly with astronomy,82 cannot be used for our investigation of the relationship between apocryphal scriptures and the Old Text/New Text controversy. This is because, while Cheng Hsuian's rebuttal of Hsfi Shen's views has been preserved, the latter have been lost together with the original chin and ku opinions. We shall therefore have to return to our comparison of quotations from WCII and from apocryphal texts.

The bureaucracy

Among the institutional matters treated in WCII as problems over which ku and chin traditions were divided, is the question of the duties that the Three Excellencies (san-kung-2) had to ful- fill and of the position that they occupied in the bureaucratic hierarchy.83 According to the Ou-yang Shang-shu kJ;@ f(l and the Hsia-hou Shang-shu AI1 fkJ traditions, both of which were "new", san-kung was a designation for the highest official positions in the hierarchy, namely the Minister over the Masses (ssu-t'u HJ), the Minister of War (ssu-ma !,J), and the Minister of Works (ssu- k'ung p]lt).84 By contrast, an "old" Chou-li tradition states that the

79 Yasui 3/50; 5/36. 80 Yasui 3/83. 81 WCII, ibid. 82 WCII 1250/47a. 83 WCII 1250/14a. 84 This opinion was "proven" by means of three passages in the Shang-shu ta-

chuan quoted in Ch'en Shou-ch'i's commentary to WCII 1250/14b.




Three Excellencies were the Grand Tutor (t'ai-shih t&gfi), the Grand Teacher (t'ai-fu Jci-4), and the Grand Conservator (t'ai- pao t&fS); they had no special governmental task, but "sat and discussed the right way".85 This account left the highest actual power with the Chancellor (ch'eng-hsiang Ri) and the Grandee Secretary (yii-shih ta-fu MkAX) of the Ch'in and Han system.

As is well known, Wang Mang reformed the Han titles, adopt- ing the opinion characterized as "new" in the WCII. His reform survived the downfall of his regime and was taken on by Kuang- wu, the founder of the Later Han. And yet, in the Book of the Han Pan Ku sticks to the "old" tradition, mentioning the "new" tradition only briefly.86

At least one quotation from the apocryphal text Shang-shu hsing-te-fang f1Jl shows conformity between the "new" opinion and the apocrypha:

I P was Minister of War (ssu-ma), Hsieh A was Minister over the Masses (ssu- Cu) and Yui r% was Minister of Works (ssu-k'ung). The Three Excellencies are the symbol for three [different] capacities.87

Another fragment runs: "The Three Excellencies are the Minister over the Masses, the Minister of War and the Minister of Works";88 but it cannot be accepted as apocryphal since it quotes in fact from a certain Shang-shu shuo, possibly the name for a commentarial tradition.89 Still, in this context there is conformity between the "new" traditions and what is left of the apocrypha.

Miraculous births

Stories that recount the miraculous births of the ancestors of several ancient dynasties are particularly frequent in apocryphal scriptures.90 The same topic is discussed briefly in WCII in the following terms: 91

The traditions (shuo) to the Ch'i W, Lu t and Han -W Shi '

as well as to the Kung-yang say that all the sages were without fathers. Their mothers per- ceived heaven and subsequently gave birth [to them].

The Tso-shih t tradition: "All sages had fathers".

85 Chou-li (Shi-san ching chu-shu ed.) 905B. 86 HS 19A/722. 87 Yasui 2/63. 88 Yasui 2/70, quoting a Shang-shu wei. 89 Compare note 61. 90 Yasui 2/64, 81, 90; 3/24, 25; 4a/12, 29; 4B/11. 91 WCII 1250/12a-b.



58 HANS VAN ESS

Again, the "new" traditions are in agreement with the apocryphal texts. Interestingly, in the second century B.C. Ssu-ma Ch'ien used both sorts of stories in his Records of the Grand Histo- rian.92 This may suggest that he was not yet concerned with the problem of the "old" and "new" traditions.

The size of the empire

The following apocryphal fragment is also in agreement with the "new" tradition:

Under the Chou there were one thousand and eight hundred feudal lords [whose territories] were spread over [an area] of five thousand miles (li f) 9

In WCII the "new" Ou-yang and Hsia-hou explanations likewise state that the Middle States (chung-kuo rPPS) covered a territory of five thousand miles square.94 This is exactly the "Confucian" argument employed by the erudites and the "worthies" in the Dis- cussions on Salt and Iron and directed against the policy of expansion of Emperors Wu and Hsiuan.95 The expansionists-in WCII justified by the "old" Shang-shu explanation-calculated a total of ten thousand miles square. Nowhere in Yasui's collection do we find the notion that the size of the Chinese empire ever reached such a figure.

The addressees of the offerings at the state-altars

An important question that occupied the minds of the specialists concerned with the state cults was how to determine the appropriate addressees of the sacrifices at the altars of earth and millet. The "new" Hsiao-ching explanation states that: "She ftL, the altar of the earth, is the ruler of the earth (she che, t'u-ti chih chu

Millet is the chief of the Five Grains... Therefore millet is established as the addressee of the sacrifices"; and that: "The altar of the earth is the earth god, millet is the god of grain." The differing "old" opinion is based on a Tso-chuan expla-

92 While in the Basic Annals (chapters 3 and 4) he reports the miraculous stories, in the table of descendants of the early emperors and dynasties (chapter 13) he shows that the same ancestors whose mothers are said in the Basic Annals to have become pregnant miraculously did in fact have fathers.

93 Yasui 5/112. 94 WCII 1250/3b. 95 Yen-t'ieh lun, chapter 16.




nation that says: "At the altar of the earth only Kou-lung '-iJn and Hou-chi J5.g get offerings."96

The great number of fragments preserved on this subject in Yasui's collection suggest that the apocryphal scriptures must have dealt extensively with the question of the sacrifices at the altars of earth and millet. By contrast, WCIIs "old" version substi- tuting deified heroes of the past for the nature deities of the "new"5 explanation is nowhere to be found in these fragments. Again, the apocryphal texts seem to vote clearly and unanimously for the "new" tradition.97

The treatment of the descendants of two preceding dynasties

We find in a fragment from the Hsiao-ching yiian-shen-ch'i (the second of the apocryphal texts quoted by Cheng Hsiuan), and in another one from a certain Hsiao-ching wei :*9w, a discussion of the correct treatment that should be reserved to the descendants of the two dynasties preceding the present ruling house. Accord- ing to these texts, dukedoms should be established for the descendants of both dynasties.98 In WCII, "new" Kung-yang explanations as well as the chapter "Chiao-t'e-sheng" 14t in the Book of Rites are quoted as sharing this opinion, whereas the "old" Tso- chuan tradition adds the "Three Respectable" (san-k'olI) to the descendants of the two previous dynasties, thus reaching the canonical number of five descendants who are to get special treatment from the ruling dynasty.99 This "old" opinion, not attested in the apocryphal texts, obviously reflects the practice of Wang Mang.100

Fragments of dubious origin

Two further fragments showing perfect agreement between "new opinions and apocrypha-one on the frequency of audi-

96 WCII 1248/28a-29b. Kou-lung was the son of Kung-kung #TI, a minister of emperor Yao *. Hou-chi was the founder of the house of Chou. Compare Shih- chi 5Re (Chung-hua shu-chiu ed., 1959) 1/20, 28 and 4/lllf. See also my Politik und Gelehrsamkeit, pp. 184ff.

97 Yasui 5/38, 112. We need to be careful about similar fragments quoted in Yasui 3/94 because no sources earlier than Ming collections are quoted there.

98 Yasui 5/34 (from T'ai-p'ing yi-lan 198), 75. 99 WCII 1250/4b. 100 See Politik und Gelehrsamkeit, 111.2.2.



60 HANS VAN ESS

ences the feudal lords have to pay to their king, and one on the nine clans (chiu-tsu )ij&) of the emperor-will not be discussed here, as Yasui's sources are dubious.101

There are also two items under the heading "Shang-shu wei" (f. K* in which the quotations conform to an "old" explanation in WCII.102 One concerns the five appellations (hao V,) for Heaven, the other the sacrifice to the Six Venerable (liu-tsung At), a service mentioned in the Book of Documents. Evidence that these fragments indeed go back to apocryphal texts is scarce, however: Yasui states that both are quoted as Shang-shu wei in the subcommentary to the Ta-tsung-po {-/if section of the Rites of the Chou; but neither in the Chung-hua shu-chui edition of the Thirteen Classics nor in Sun I-jang's F excellent Chou-li cheng-i tOIL is such a quotation to be found. On the other hand, both "apocryphal" fragments of Yasui's recompilation are quoted verbatim in the same subcommentary, section Ta-tsung-po, as Shang-shu shuo fri} Q. The context makes it clear that in both cases these shuo are "old" Documents explanations contained in the Different Meanings of the Five Classics. According to Hsfi Shen's custom a "new"9 explanation is given in both cases before the "old" one. I consider it almost certain that these fragments are not genuinely apocryphal, but go back instead to the Different Meanings.103

1O0 Yasui 5/115; WCII 1249/54b-55b; andYasui 2/68 and WCII 1250/1Ob-11a. In the first case Yasui quotes Ho Hsiu's %2Jfc Kung-yang chieh-ku , as his authority and acknowledges that the title "wei" (apocryphal) is not used in this instance; in the second case he quotes the chapter "Chung-hui chih kao" 11Nl. of the Book of Documents. In both cases the reason for integrating these fragments into the collection is that Ming or Ch'ing scholars declare them as "wei"; they do not, however, give the sources for this judgement.

102 Yasui 2/68 (twice) and WCII 1248/5b-6a and 19b. 103 The oldest source to include these fragments in a collection of apocryphal

texts is the Ku wei shu W*& dating from the late Ming (ed. Shou-shan ke ts'ung- shu !--,7 3/la-b). As Sun Chueh * the compiler of Ku wei shu, did not include fragments without a specific title into his collection-unlike Yasui, who has a section of fragments simply called "wei" attached to each classic after the fragments which are known under specific titles,-he seems to have assigned these two fragments himself to the apocryphal text Shang-shu ti-ming yen MASWOM. I have not found any statement concerning the criteria that might have lead Sun Chueh to believe that these quotations-which apparently did not have titles in his sources-might belong to this particular apocryphal text. Ma Kuo-han and Huang Shih, on whose recompilations Yasui's collection draws heavily, most probably followed Sun Chueh when they included these quotations into -their recompilations as well, without questioning his approach.




Conclusion

Our survey provides a surprisingly straightforward answer to the question of the relationship between the Han commentarial traditions and the apocryphal scriptures. Setting aside the fragments of dubious origin in Yasui Kozan's recompilation of those scriptures, all the other quotations agree with their "new" counterparts in the only source listing topics of "chin" and "ku" scholarship in the Han, viz. the Different Meanings of the Five Classics; the question of the Hall of Light is the only exception to this pat- tern.104 Of course, since the sources used for this comparison are both late recompilations of primary texts which have been long lost, some uncertainties will remain. Still, the evidence on the relevance of commentarial writing in Han times provided by these texts seems to me very strong.

The investigation carried out above yields two important re- sults. First, it seems clear that there was a connection between the "new" tradition of Han times and the apocryphal texts. Clearly, these texts served as an authoritative foundation for the opinions of "chin" scholars when they could not base their ideas on the Classics themselves because they did not cover the subjects at hand.

Second, and probably more important, our survey strongly suggests that the Han scholarly community was indeed divided over institutional problems that were associated with the opposition between a "new" and an "old" tradition. Hsiu Shen at least described the political conflicts that took place during the last fifty years of the Former Han in terms of a controversy between those two classical traditions. It goes without saying that the labels "new" and "old" do not mean that one tradition was more recent than the other-and if that was the case, in most instances "new" opinions seem in fact to have been older than the "old" ones: "old" meant the system of the Ch'in and the Han, whereas "new"

104 The strange exception of the Hall of Light can perhaps be explained by the fact that this topic became an extremely important one at the end of the Later Han. The ritual structures at Lo-yang had most likely been destroyed by the troops of general Tung Cho 1* in the year 190 A.D. Upon the beginning of the rebuilding discussions might have started about the correct architectural design and the ideological outlook of this building. The original positions discussed in the Old Text/New Text controversy had at this time undergone great changes, which may account for some confusion with regard to this problem.



62 HANS VAN ESS

meant the system favored by those who advocated a return to the old system of the Chou. The primary meaning of Hsfi Shen's two labels was that what was "old" was related to the texts written in old characters, and what was "new" to the texts written in new script. But certainly his intention was also to imply that "old" meant "better", and possibly that "new" was "worse". The reason is that the man who had done the most to promote many "new" (chin 4) ideas and who had established a "new" (hsin Vr) system was the usurpator, Wang Mang. When the erudites at the beginning of the first reign of the Later Han successfully prevented the establishment of a chair for the Tso-chuan at the Imperial Acad- emy, they implicitly voted for continuing the system established by Wang Mang, although he himself had promoted the Tso- chuan. It was against this system and these people that Hsfi Shen's Different Meanings of the Five Classics was directed.

Of course, this does not mean that at the end of the Former Han two monolithic factions stood opposed at the imperial court. One would surely expect that there were officials who, while being inclined towards the modernists rather than the reformists, nevertheless supported certain positions of the latter-and vice versa: we should rather think of "ku" and "chin" as two poles between which a great variety of opinions was possible, and to re- member that it was only for the sake of clarity that the historians and philologists of the Han reduced these opinions to two oppo- sites. Indeed, one finds examples of erudites and politicians who clearly belonged to Loewe's modernists but in some instances held views which are classified as "new" in WCII, or reformists who supported "old" opinions. It is also possible that some scholars took a liking to deciphering ancient characters but did not agree with the "old" solutions to institutional problems that were based on the very texts they were reading. Only in a theoretical model would be such deviations from the rule impossible: the reality is always more complex.'05 Whereas in the Shuo-wen chieh- tzu the term ku certainly relates exclusively to ancient characters, in his Different Meanings of the Five Classics Hsui Shen seems to have

105 For an example of a "new" scholar supporting an "old" view (in this case, that high offices should be hereditary), see the case of Wei Hsuian-ch'eng as interpreted by Michael Friedrich, "Die Ahnen und das Ich", in Helwig Schmidt- Glintzer, ed., Das andere China. Festschrift fur Wolfgang Bauer zum 65. Geburtstag, Wolfenbuittel 1995, pp. 405-434, on p. 433, note 93. Another example is Yang Hsiung, who clearly belonged to Loewe's reformists but was interested in etymology. See my Politik und Gelehrsamkeit p. 290f.




been interested in the contents of the books and in their political meaning: the tags "chin" and "ku" relate primarily to institutional matters and to politics, not to etymology.

This hypothesis has the advantage of explaining why, starting with the second half of the first century, we find in the Book of the Later Han examples of "ku" scholars who studied apocryphal texts. In theory, such a situation would be in contradiction with the notion that "ku" scholars were by definition opposed to apocrypha; that it did occur is partly due to the fact that, at the time of such scholars as Chia K'uei, an "Old Text" scholar who quoted ch'en-wei scriptures,'06 or Li Yu 2W, a "new" Kung-yang erudite who was opposed to using them for decisions on classical matters,107 the conflict had already gone through Wang Mang's at- tempt at synthesis: the interregnum had changed the original nature of the scholarly dispute regarding many topics.

The Ch'ing recompilation of fragmentary texts helps to en- large our limited knowledge of topics about which secondary sources like the Book of the Han and the Book of the Later Han remain largely silent. That the dynastic histories do not provide a complete account of a battle between Old Text and New Text opponents should not seduce us into the conclusion that there was no such controversy whatsoever among Han commentators. The Ch'ing scholars did use the Han tradition for their own political purposes, as exegetes have always done in China; but they did not allow themselves to invent a conflict that had in fact never taken place.108

If Li Hsuieh-ch'in is indeed right in saying that Liao P'ing was the first thinker to rediscover the Old Text/New Text Contro- versy of the Han, then any challenge to Liao's argument ought to base itself on the texts he actually employed for his argument- in this case, the Different Meanings of the Five Classics-and not on the dynastic histories. He and later scholars quoted the dynastic histories only for the supporting evidence they contained for the claim that the conflict described in the Different Meanings had indeed taken place.

It is most probably because he was blinded by the Old Text/ New Text controversy taking place in his own time, in which he

106 HHS 36/1237. 107 HHS 79B/2582. 108 They merely stressed other topics than their Han predecessors, falsely

contending that they had been of primary concern for them.



64 HANS VAN ESS

was himself involved, that Liao P'ing failed to understand the historical relevance of the Han controversy, namely, that it was a mirror of contemporary politics. But it is his undeniable contribution to have recognized that, contrary to the concerns of evi- dential research scholars, the controversy had its focus not on philology, but on institutional matters.

