Seven
Liu Feng-lu and New Text Confucianism

Liu Feng-lu represents in many ways the intellectual culmination of the Ch'ang-chou New Text school. After Liu, New Text Confucianism transcended its geographical origins in Ch'ang-chou, becoming a powerful current in early-nineteenth-century intellectual circles via the writings of Kung Tzu-chen and Wei Yuan. Both of these men studied in Peking under Liu Feng-lu's direction in the 1820s, when Liu had become a celebrity in the capital.^[1]

Liu Feng-Lu as an Official

A child prodigy who benefited greatly from the cultural resources of his two wealthy and influential lineages, Liu Feng4u early on was considered one of the most promising talents in Ch'ang-chou, rivaled only by the slightly more senior Li Chao-lo. In due course, Liu Feng-lu and Li Chao-lo became regarded as the most exceptional local examination candidates and were known as the "two Shen" (liang Shen; i.e., both had shen as a character in their literary names) of Ch'ang-chou. A product of Ch'ang-chou Han Learning, Li Chao-lo nonetheless admired, as we have seen, the achievements of Chuang Ts'un-yu and had close ties with the Chuang lineage.

Although Liu Feng-lu did not achieve chin-shih status until the capital examination of 1814, his examination essays for the c hü -jen degree in 1805 (which he prepared for the provincial examinations held in the capital rather than his own Chiang-su Province) had already caused a stir. Liu's use of Kung-yang interpretations to explicate the Classics "greatly astonished," said Tai Wang, "those who read over his examination materials," and Liu was treated with the respect due a national literatus. New Text had come to Peking.^[2]

It of course mattered that Feng4u possessed impeccable family and lineage credentials. His grandfathers Liu Lun and Chuang Ts'un-yü had, after all, long since left their marks on the highest levels of the Peking bureaucracy (see chapter 2). Moreover, the Liu and Chuang lineages had for generations propelled their Ch'ang-chou sons through examination success to the upper echelons of the imperial state. After the demise of Ho-shen, Liu Feng4u continued this legacy into the nineteenth century. Appointed to the Hanlin Academy for his high standing on the 1814 palace examinations, Feng-lu quickly achieved prominence in the Ministry of Rites, a standard stepping-stone to high position in the internal politics between the inner court and outer bureaucracy (see chapter 3).^[3]

Liu Feng-lu lived out what appears to have been a conventional career at the Ministry of Rites. Using his classical expertise, for example, Liu assembled proposals and precedents on the proper rituals for the sudden and unusual death of the Chia-ch'ing Emperor (r. 1796-1820), who was apparently struck by lightning. Liu also wrote a favorable opinion on a proposal from the Henan provincial director of education, advocating that T'ang Pin be canonized in the official Confucian temple in Ch'ü-fu, Shan-tung. Then director of the Ministry of Rites, Wang T'ing-chen was pleased with Liu's advocacy of this controversial early Ch'ing Confucian, and the Tao-kuang Emperor (r. 1820-50) accepted the proposal.^[4]

Using his unconventional knowledge of historical precedents, Liu Feng-lu carved out an unusual place in the foreign policy decisions of the Ministry of Rites. His knowledge of the "precedents" in the Spring

and Autumn Annals, for instance, proved invaluable in 1824, when a tribute mission from the recently crowned king of Vietnam and ardent sinophile Minh-mang (r. 1820-41) went awry. The Vietnamese emissary in charge of the mission objected to the language of the official rescript, which had been prepared by the Tao-kuang Emperor for transmission to the Vietnamese king. The emissary was affronted by the reference in the rescript to the Vietnamese as "outer barbarians" (wai-i), a term that derived from the distinction between "inner feudal lords" (nei chu-hsia ) and "outer barbarians" (wai i-jung ) and had been prominent during the Chou dynasty (1122?-221 B.C. ). It also was a term that figured prominently in the Kung-yang interpretation of foreign affairs chronicled in Confucius's Annals.^[5]

The case reveals that not only Westerners but also Asian tribute states were offended by the pejorative connotations of Confucian cul-turalism. A Han Chinese official serving a Manchu emperor (who represented a conquering army of barbarians), Liu Feng-lu had to tread carefully in responding to the Vietnamese objections. Like Chuang Ts'un-yü in the 1758 bannermen riot (see chapter 3), Liu Feng-lu was caught between a classical legacy based on Chinese protonationalism and the present-day political realities of barbarian rule. Ts'un-yü precipitated a riot when he demanded that Manchu and Mongol bannermen meet the same rigorous standards expected of Chinese. Liu Feng-lu was more tactful.

The Vietnamese emissary had proposed to the Ch'ing court that, instead of "outer barbarians," the Vietnamese be named "outer vassals" (wai-fan ), a less ethnocentric nomenclature that had in fact been applied to the Mongols and Tibetans under the institutional mechanism known as the Li-fan yuan (Court of Colonial Affairs—that is, "office for regulating vassals of the state"). The Ch'ing court employed the Li-fan yuan to differentiate ethnic Chinese from Manchus and to regulate certain minorities outside the regular bureaucracy in the northern and northwestern frontiers.

Liu Feng4u played his first classical hand carefully, observing that in the "Offices of Chou" (Chou-kuan, that is, the Choudi ) regions outside the capital region had been divided into nine areas (chiu-fu ). The "barbarian region" (i-fu ) was located seven thousand li (a Chinese "mile" approximately equal to 360 paces or.35 of an English mile) from

the king's area, while the "vassals' region" (fan-fu ) was nine thousand li from the capital region. Liu concluded: "Consequently, vassals were distant and barbarians close." Reference to the Vietnamese as "barbarians" accordingly implied higher respect within the Chou dynasty system of foreign affairs.^[6]

Next, Liu cited an etymological precedent drawn from the Later Han Shuo-wen chieh-tzu dictionary of paleography. Philology would corroborate diplomatic protocol. According to Liu, the ancient Chinese graphs reserved for most non-Chinese groups were composed of a classifying radical (pu-shou ) used for words referring to "creatures" (wu, that is, animals such as snakes, boar, etc. ). This condescending orthography was conspicuously absent in the graph i, used for barbarians, which followed the radicals of "great" (ta ) and "bow" (kung ), terms of respect for the "outer barbarians."

The clinching argument, according to Liu Feng-lu, for maintaining the use of "outer barbarians" for the Vietnamese was the far more recent policy established by the Ch'ien-lung Emperor himself. When i appeared in the various ancient and contemporary books the editors were collecting and collating for inclusion in the Imperial Library, the emperor issued an edict that said it was unnecessary to change the character i for barbarians to another more innocuous character. This edict, moreover, was promulgated at a time when one disrespectful word uttered against a Manchu or Mongol was severely punished.^[7] Liu Feng-lu was discreetly pointing out that in the midst of strenuous efforts to weed out anti-Manchu sentiments in books and manuscripts as well as to trace the origins of the Manchus as a people, the Manchu emperor himself had found nothing disrespectful in the term for "barbarians."

By citing the culturalist policies whereby Manchu rulers "united six racial groups into a single state" (liu-ho i-chia ), Liu Feng-lu brought the full weight of the long-standing late imperial barbarian-Chinese synthesis down on the Vietnamese emissary. Drawing also on the Ch'ien-lung Emperor's edict that initiated research on Manchu origins in 1777, Liu explained that the sage-king Shun (tr. r. 2253-2205?? B.C. ) was called an "Eastern Barbarian" (Tung-i chib jen ) by none other than Mencius. Moreover, Mencius went on to refer to King Wen (tr. r. 11427-1135?), the founder of the Chou dynasty, as a "Western Barbarian" (Hsi-i chih jen ).

The Ch'ien-lung Emperor had used Mencius to play down Han Chinese chauvinism. Liu Feng-lu used Mencius to gainsay the Vietnamese. If sage-kings were "barbarians," what truth could there be to the Vietnamese claim that their status in the Confucian world order was being impugned? According to the imperial account of the incident, "the Vietnamese emissary had nothing he could say in reply and retired." Such issues could not be finessed for much longer, however. The tensions between Chinese and Manchu, Confucian and foreigner, eventually exploded into the Opium War (1839-42)—a battle initiated by Great Britain for full and equal diplomatic relations with China—and the Taiping Rebellion (1850-64), an anti-Manchu peasant rebellion.^[8]

Liu Feng-lu served in the Ministry of Rites for twelve years, becoming well-known for his ability "to decide problematic cases on the basis of classical meanings" (i ching-i chueh i-shih ). Liu also used historical precedents and classical meanings in the Annals to decide an 1817 family succession dispute in An-hui Province and an 1824 burial controversy, applying ritual precedents to determine how the status of wives influenced family succession. In the latter, Liu cited Ho Hsiu's explication of the Kung-yang chuan as the authoritative basis for ascertaining proper burial attire based on the status of the deceased.^[9]

Liu Feng-lu's training in Kung-yang Confucianism under Chuang Ts'un-yü and Chuang Shu-tsu enabled him to tap a Former Han classical tradition that, it appears, was particularly useful in solving problems in ritual protocol and legal decisions. In rediscovering that the links between Han classical studies and ritual and legal precedents could also be applied to contemporary questions, Liu Feng-lu and other New Text scholars were also reconfirming the centrality of classical studies for political discourse.^[10]

Liu Feng-Lu and Han Learning

Unlike his distinguished grandfathers, Liu Feng-lu became more widely recognized as a scholar than as an official. In Peking he came into contact with the leading intellectual figures of the early nineteenth century and in so doing helped to transform the Ch'ang-chou Kung-yang tradition into a national concern.

Liu Feng-lu's friends and acquaintances in Peking included many of

the most prominent Han Learning and k'ao-cheng scholars of the day. In addition to close relations with Ch'ang-chou Han Learning scholars such as Li Chao-lo, Sun Hsing-yen, Yun Ching, and Chang Hui-yen, Feng-lu was also intimate with leading Han Learning patrons such as Yang-chou's Juan Yuan. Although a patron of Han Learning on a grand scale for much of his career, Juan in the 1820s had begun to take a more moderate scholarly position, which encompassed the strengths of both Han and Sung Learning.

Liu Feng-Lu and Juan Yuan

Around 1820 Liu encouraged Juan Yuan—then a major provincial official serving as governor-general of Kuang-chou—to reissue the Shih-san-ching chu-shu (Commentaries and subcommentaries to the Thirteen Classics), a Sung dynasty collectanea (ts'ung-sbu ) that contained the major classical studies of the Han, T'ang, and Sung dynasties. In addition, Liu suggested that Juan compile a comprehensive collection of Ch'ing dynasty contributions to classical scholarship. The result was the 1829 publication in Kuang-chou of the Huang-Ch'ing ching-chieh (Ch'ing exegesis of the Classics). After four years of preparation and editing at the prestigious Hsueh-hai-t'ang Academy, the book was greeted with great acclaim in Han Learning circles in China and sent to Korea, Japan, and later the West. Designed as a continuation to the Sbih-san-ching chu-shu, the Huang-Ch'ing ching-chieh represented a major tribute to the classical research of k'ao-cheng scholars.^[11]

Juan's tribute to Ch'ing scholarship stressed the eighteenth-century Hah Learning movement, which had championed the Later Hah classical "schools system" of Cheng Hsuan and Hsu Shen, among others. But Liu Feng-lu prevailed upon Juan Yuan to include Ch'ing dynasty works dealing with the Former Han New Text "school system" based on Tung Chung-shu (1797-1047 B.C. ) and Ho Hsiu. As a result, a number of works by men connected to the Ch'ang-chou New Text tradition were reprinted in the collection: Chung Ts'un-yü's Correcting Phrases in the Spring and Autumn Annals; three works by K'ung Kuang-sen, including Penetrating the Meanings in the Kung-yang Commentary to the Annals; seven works by Liu Feng-lu, most prominently his Master Ho Hsiu' s Explications of Precedents in the Kung-yang Commentary to the Annals Classic and the famous Evidential Analysis of Master Tso's

Spring and Autumn Annals (Tso-shih ch'un-ch'iu k'ao-cheng); and two works by Ling Shu, a follower of Liu Feng-lu, including his Ritual Explanations in the Kung-yang Commentary (Kung-yang li-shuo ).^[12]

Besides associating with and patronizing scholars who studied the Kung-yang chuan, such as Liu Feng-lu, Juan Yuan himself developed an interest late in life in the Former Han New Text alternative to the Later Han Old Text "schools system." In an introduction to K'ung Kuang-sen's Penetrating the Meanings, Juan traced the transmission of Kung-yang studies and described their vicissitudes before their mid-Ch'ing revival. Although he pointed out four areas where K'ung's interpretation of the Kung-yang chuan differed from Ho Hsiu's (differences Liu Feng-lu would also stress in his own criticism of K'ung's research [see below]), Juan Yuan concluded that the Kung-yang chuan was superior to the Tso chuan in elucidating the organizing principles Confucius had subtly encoded in the Annals.^[13]

In 1793, during his tenure as director of education in Shan-tung Province, Juan Yuan directed the rebuilding of a shrine commemorating the Later Han classicist Cheng Hsuan. This effort epitomized Juan's emphasis on Cheng Hsuan as the key figure in reconstructing the "schools system" of Han Learning. Moreover, as governor of Che-chiang Province in 1801, Juan had established the Ku-ching ching-she (Refined Study for the Glossing of the Classics) Academy in Hang-chou. He noted that the academy was named to honor the Later Han classical studies of both Cheng Hsuan and Hsu Shen, author of the Shuo-wen chieh-tzu dictionary.^[14]

Juan Yuan established the Hsueh-hai-t'ang Academy in Kuang-chou in 1820 along the same Han Learning guidelines used for the academy in Hang-chou. But the new school—unlike the Ku-ching ching-she, which was named after Cheng Hsuan and Hsu Shen—was named in honor of the New Text scholar Ho Hsiu, who had been known honor-rifically as "Hsueh-hai" (Sea of learning) because of his great erudition on the Classics. We can conclude, therefore, that Chuang Ts'un-yü's rediscovery of Ho Hsiu and the Kung-yang tradition was by 1820 more than just a local Ch'ang-chou tradition tied to the Chuang lineage.^[15]

Hsu Jung, one of the eight original directors of the Hsueh-hai-t'ang, caught the symbolic significance of Juan's name for the academy. He composed the following verse in 1824 for the dedication of the academy's new buildings on Yueh-hsiu Hill:

The Green-jade Pavilion is high and out of sight,
Southward from the South Garden only desolate weeds.
Of Clear Spring Refined Study no foundations remain,
Precious Moon Tower stands alone—what sounds will follow?

The roots of a hundred generations lie in past events,
This broad mansion of a thousand rooms holds the mind of the ancients.
We will be able to take the Kung-yang learning of the two Hans,
And swiftly sweep away the superficial to reveal the Old and New.^[16]

Juan Yuan had personal ties with many of the scholars associated with the Ch'ang-chou New Text tradition. Liu Feng-lu's sometime follower, Ling Shu, served late in life as Juan's assistant, also tutoring Juan's sons. Juan's close friend Ch'eng En-tse was a well-known scholar and tutored many of those involved in the New Text tradition. Ch'eng's circle of friends in Peking included Liu Feng-lu, Wei Yuan, and Kung Tzu-chen, all of whose statecraft proposals and political activities were widely admired. Such links between Han Learning and New Text scholarship in the early nineteenth century suggest that we should broaden our understanding of the Old Text-New Text controversy and be prepared to consider the manner in which the Ch'ang-chou New Text initiative entered the k'ao-cheng mainstream.^[17]

K'Ao-Cheng Epistemological Leverage

Standing on the borderline between Han Learning and Sung Learning, the Kung-yang studies produced by Chuang and Liu scholars in Ch'ang-chou assumed national prominence through the preliminary efforts of Chuang Shu-tsu, Chuang Yu-k'e, and Chuang Shou-chia. They united Chuang Ts'un-yü's Kung-yang interests to Han Learning k'ao-cheng studies. Liu Feng-lu brought this ongoing synthesis to virtual completion. Through him, the latest philological discoveries of Han Learning were conjoined to the reconstruction of the Former Han New Text clas-

sical tradition. He also impressed the theoretical perspectives of Kung-yang Confucianism on leading Han Learning scholars, who began to see the import of Chuang Ts'un-yü's efforts to overcome the threatening implications of the debate between Hah and Sung Learning.

Liu Feng-lu realized that the theoretical conclusions reached by Chuang Ts'un-yü and other members of the Ch'ang-chou school needed support from empirically based k'ao-cheng methods of demonstration. Without such support, Kung-yang studies lacked the epistemological leverage to be taken seriously enough to appear alongside mainstream Hah Learning studies in collections such as the Ch'ing Exegesis of the Classics. Juan Yuan, as we saw in chapter 4, declined to include Chuang Ts'un-yü's writings on the Change Classic in the latter collectanea, for example, because they were not informed by k'ao-cheng methods. Evidential research methods were employed so widely in the eighteenth and early nineteenth centuries that the accepted forms of scholarly discourse had been transformed in epistemological terms. Works that lacked systematic methods of demonstration could no longer achieve academic respectability. This scholarly shift was as true for Han Learning as for New Text studies.

As Chuang Shu-tsu and Chuang Yu-k'e had already demonstrated, k'ao-cheng was no one's monopoly. Liu Feng-lu devoted considerable attention to philology and phonology to gain the respect of his Han Learning colleagues. Liu made important contributions to k'ao-cheng research, particularly in the field of reconstructing ancient pronunciation based on classical rhymes (ku-yin ). When such research was allied with New Text studies, the latter acquired the epistemological leverage needed to gain scholarly credibility.^[18]

Liu Feng-Lu and New Text Studies

Liu Feng4u transformed Kung-yang Confucianism from an idiosyncratic theoretical position into a legitimate form of Hah Learning. Using k'ao-cheng to give epistemological legitimacy to the Kung-yang interpretation of the Classics, Liu Feng-lu in effect opened the door for full recognition of the scope of New Text Confucianism (chin-wen-hsueh ). Along with Sung Hsiang-feng—who raised questions concerning the authenticity of the Rituals of Chou in connection with doubts about the legitimacy of the Old Text Classics (see chapter 6)—Feng-lu co-opted

the accruing philological evidence on the provenance of the Old Text portions of the Documents Classic and other Classics for the Kung-yang agenda.^[19]

As the Old Text Classics increasingly became caught up in authenticity debates, the philological evidence unwittingly supplied by Han Learning scholars for more than a century became a powerful weapon in the New Text arsenal. For much of the nineteenth century, the New Text Classics enjoyed an aura of philological authenticity that contrasted sharply with the misgivings that clouded the reputations of the Rituals of Chou, the Tso chuan, and the Old Text Documents. Although research had by the early nineteenth century shown no unanimity on the Five Classics as a whole, the direction in k'ao-cheng research was evident.

In the seventeenth and eighteenth centuries, efforts to recover and reconstruct the Han Learning legacy had taken the form of a "return" (fu-ku ) to the exegetical tradition of the Later Hah "schools system." Continuing this scholarly return to the authenticated sources of the classical canon, followers of the Ch'ang-chou New Text school in the late eighteenth and early nineteenth centuries advanced the frontiers of their knowledge by focusing on the scholarship associated with the "schools system" of the Former Han dynasty as a superior source for recovering the Confucian legacy. Even the more mainstream k'ao-cheng scholars who were not directly associated with the Ch'ang-chou school were moving in this direction.^[20]

The Fu-chien scholar Ch'en Shou-ch'i (1771-1834) is a good example of this more general scholarly shift. A New Text philologist, Ch'en was among the first to recognize the superiority of the New Text version of the authentic chapters of the Documents Classic over the Old Text version. In the early part of the nineteenth century, Ch'en completed a reconstruction of the Former Han New Text commentary to the Documents. Such emendations of the findings of Tuan Yü-ts'ai and other Old Text supporters were continued by Ch'en Shou-ch'i's son Ch'iao-ts'ung. In addition to collating the New Text recension of the Documents, Chen Ch'iao-ts'ung also assembled in another work most of the conclusions reached by k'ao-cheng scholars concerning the Old Text Documents controversy in the preceding two centuries. Like Chuang Shu-tsu and Liu Feng-lu, the Ch'ens rejected the Old Text re-

cension of the Documents in favor of New Text variants dating from the Former Han "schools system."^[21]

Philology became the means to reorder the classical legacy. Convinced of the need for an exact textual understanding of the Confucian Classics in order to reach doctrinal conclusions, k'ao-cheng scholars reversed earlier Sung-Ming Neo-Confucian agendas by making philology, not philosophy, the key to recovering the teachings of antiquity. Liu Feng-lu's genius lay in correctly perceiving the crucial role played by the textual criticism of the Old Text Classics. In fact, he viewed it as the philological premise for the restitution of the long-defunct New Text Classics initiated by Chuang Ts'un-yü when he turned to Kung-yang Confucianism. Kung-yang theory was now grounded on k'ao-cheng discourse. The joining of activist Kung-yang theory and intellectualist Han Learning philology would yield New Text Confucianism.^[22]

As his imperial biographer explained, "Liu Feng-lu's scholarship endeavored to penetrate great meanings without overstressing the parsing of sentences and phrases [in the Classics]." Although a student of Chuang Ts'un-yü, Liu moved away from his teacher's stress on Ho Hsiu back to the Former Han New Text Confucianism of Tung Chung-shu. The unanimity of Han Learning was effectively over. Feng-lu's shift epitomized the retreat from Cheng Hsuan and the classical "schools system" of the Later Han.^[23]

Liu Feng-lu perceived in the New Text concerns of the Former Han a bedrock of Confucian theory that could replace the overly philological concerns of the Later Han Old Text school. The Former Han New Text stress on theory contrasted in a fundamental way with the Later Han Old Text emphasis on philology. This contrast reaffirmed, albeit in new terms, the difference between Sung Learning moral philosophy and Han Learning k'ao-cheng. New Text Confucianism gave Feng-lu and others the textual means to restore the balance between theory and philology.^[24]

Liu Feng-Lu vs. Ch'Ien Ta-Hsin

Liu Feng-lu's remarkable synthesis proceeded simultaneously on a number of fronts. The central axis of his New Text synthesis was Confucius's Spring and Autumn Annals, which he viewed as the key to "seeking out the intent of the sages" (sheng-jen chih chih ). Showing remarkable professional courage and intellectual integrity, he wrote an essay entitled "Discourse on the Annals" ("Ch'un-ch'iu lun") early in the nineteenth century that directly controverted the conclusions on the Spring and Autumn Annals drawn by Ch'ien Ta-hsin. The most distinguished Han Learning scholar of the Ch'ien-lung era, Ch'ien Ta-hsin was also the chief spokesman for the Su-chou Han Learning tradition after the death of Hui Tung. Just as his grandfather had opposed Su-chou and Yang-chou Han Learning currents a generation before, Liu Feng-lu, it appears, was confident enough of his roots in Chuang traditions of classical learning to challenge the giants of the k'ao-cheng movement.

Continuing a line of interpretation with roots in Sung dynasty skepticism, Ch'ien Ta-hsin had argued that the Annals "had no historiographical message" (wu shu-fa ). Indeed, according to Ch'ien, the Annals contained no secret message of "praise and blame" encoded by Confucius but was simply a straightforward chronicle of events (chih shu ch'i shih ). Ch'ien therefore contended that the Tso chuan was the superior commentary on the Annals because it added flesh to the bones of the events listed in the Annals. The Kung-yang Commentary dealt only with unverifiable and mysterious notions (wu-wang chih shuo ) of little value in classical studies.^[25]

Liu Feng-lu correctly perceived that Ch'ien Ta-hsin's efforts to define the Annals as a mere historical record constituted a throwback to Wang An-shih's attempt centuries earlier to dismiss the Annals as a useless collection of names and dates devoid of any moral message. Defending the Kung-yang Commentary, Liu wrote:

[Ch'ien Ta-hsin's argument] does not prove that Kung-yang [Kao] does not compare to Master Tso, but only that the Annals itself does not compare with Master Tso. Master Tso gives details on affairs, but the Annals stresses meanings, not affairs. Master Tso never mentions precedents, but the Annals is filled with precedents. It does not loosely construct precedents. Instead, by

emphasizing affairs, [the Annals] preserves only one-tenth of the hundreds and thousands [of affairs]. What is not recorded outnumbers what is recorded. Accordingly, the Annals does not loosely construct precedents.^[26]

By accepting the superior historical status of the Tso chuan, Liu Feng-lu placed an enormous conceptual distance between the intent of the Annals as a Classic and the historical reach of the Tso chuan as a commentary. Ch'ien Ta-hsin and his predecessors had subsumed the Annals under the historical minutiae of its Tso chuan. By giving priority to the Annals as a Classic, Liu Feng-lu dismissed efforts to relegate it to the status of mere history. The debate within Han Learning over the provenance of the Classics (ching ) vis-à-vis history (shih ) was beginning to take shape. Liu Feng-lu made Ch'ien Ta-hsin his foil for the Old Text position and than articulated the special transhistorical status of the New Text Classics against it.

The Classics vs. History

The debate over the relationship between the Classics and Dynastic Histories became prominent in the seventeenth and eighteenth centuries. As historical studies became almost as prestigious as classical studies in the late eighteenth century, the conceptual chasm demarcating the universality of the Classics from the particularity of the Histories was called into question. The Documents Classic and Spring and Autumn Annals, for instance, were historical records derived from antiquity that had become Classics. A recurring problem was explaining the distinction between 'Classic' and 'History.'^[27]

A specialist in classical and historical fields of k'ao-cheng research, Ch'ien Ta-hsin went further than most in his claim that, ultimately, there was no essential difference between the Classics and the Histories. This distinction had not existed in antiquity, according to Ch'ien, but had first been used in the bibliographic "four divisions" (ssu-pu, that is, classics, history, philosophy, literature) system of classification instituted after the fall of the Han dynasty. Consequently, Ch'ien rejected the priority accorded the Classics over the Histories in reconstructing the classical tradition. For Ch'ien, the Histories were equally important sources.^[28]

Critical of fruitless speculation and arbitrary praise-and-blame historiography, k'ao-cheng historians such as Ch'ien Ta-hsin and his close confidant and brother-in-law, Wang Ming-sheng, favored application of evidential research to paint the true face of history. Wang Ming-sheng, also a follower of Su-chou Han Learning, maintained: "Discussions of praise and blame are merely empty words. The writing of history is the recording of the facts. Overall the goal is simply to ascertain the truth. Besides the facts, what more can one ask for?" Similarly, Ch'ien Ta-hsin asserted that historical facts themselves would show whom to praise and whom to blame. The process of laying blame, according to Ch'ien, should be analogous to the deliberations involved in deciding court cases. There must be no self-serving use of historical evidence to support unverifiable and mysterious theories of praise and blame drawn from the Annals. Ch'ien and Wang both rejected the a priori assumption that Confucius's Annals was the model for and embodiment of the underlying "meaning" of history.^[29]

The well-known slogan "the Six Classics are all Histories," made famous in the late eighteenth century by the historian Chang Hsueh-ch'eng, should be understood in light of contributions made by k'aobeng historians such as Wang Ming-sheng and Ch'ien Ta-hsin. They set the stage for Chang Hsueh-ch'eng's dramatic conclusion, which historicized classical studies and placed the Classics unequivocally under history's purview. We should add that this demotion of the a priori status of the Annals at the same time threatened Confucius's status as a historian. If the Annals were a useless chronicle, then what of the compiler?^[30]

Chang Hsueh-ch'eng, for example, argued that Confucius had been only the most important of many late Chou dynasty theorists, and Confucianism but one school among others. Confucius's role, he explained, had been limited to transmitting the teachings enunciated by the Duke of Chou centuries before. Chang's discussion of the different roles played by Confucius and the Duke of Chou revived debates concerning the priority of the Rituals of Chou (associated with the Duke of Chou) over the Annals (associated with Confucius) precisely at the time when Chuang Ts'un-yü and his Ch'ang-chou kin were developing their Kung-yang portrait of Confucius as a sage who used history to declare classical truths.

Questions about Confucius's historical pedigree revealed growing dissatisfaction in the eighteenth century with the Sung Learning version of the "orthodox transmission of the Way" (Tao-t'ung). Since the Sung dynasties it had stressed Confucius and Mencius as the core of the classical legacy. According to Chang Hsueh-ch'eng (who spoke unwittingly for the Old Text tradition of the Later Han when he defended the Chou-li), the Duke of Chou, not Confucius, had been the last of the world-ordering sages. While New Text scholars were dismantling the complementary ties between the Annals and the Chou-li by stressing the former, Old Text advocates were rending the same ties in favor of the latter. In effect, the two opposing sides were aiming for a common if still unclear result: the polarization of Han Learning traditions into New Text vs. Old Text.^[31]

The revival of the unorthodox Kung-yang commentary to Confucius's Annals by Ch'ang-chou scholars suggests that opposing views concerning the historical Confucius were taking shape in the late eighteenth and early nineteenth centuries. New Text scholars deemphasized Mencius, a move that indicates their break with the Sung Learning conception of orthodoxy. For Liu Feng-lu and other New Text scholars, Confucius, not the Duke of Chou, had been the central figure.

Liu's attempts to refute Ch'ien Ta-hsin, consequently, represented more than just personal dissatisfaction with Ch'ien's position on the Annals. Feng-lu's criticism symbolized the reemergence of the Han dynasties' Old Text vs. New Text controversy in fully polarized form. It also represented the struggle between historicism and classicism. New Text scholars sought to save the Classics from Old Text historians without recapitulating the Sung Learning orthodoxy already bankrupt in the highest echelons of elite literati circles.

For Liu Feng-lu as for Chuang Ts'un-yü, the Kung-yang chuan was the central text: "I have accordingly recognized for some time that all scholars seek the [wisdom of the] sages. The Way of the sages is complete in the Five Classics, and the Spring and Autumn Annals is the key to the lock." Without understanding the a priori "esoteric principles" (wei-li) informing the Annals, the Five Classics, Liu maintained, would remain inexplicable, no matter how much philological research was done.^[32]

Liu Feng-Lu vs. K'Ung Kuang-Sen

In the second part of his essay "Discourse on the Annals," Liu turned to the Kung-yang studies of K'ung Kuang-sen, a direct descendant of Confucius who had studied briefly in the 1770s under Chuang Ts'un-yü. Although he praised K'ung for recognizing that the Kung-yang Commentary was intimately tied to the Han classical "schools system," Liu Feng-lu believed K'ung had misrepresented the Later Han Kung-yang tradition by not acknowledging Ho Hsiu as the orthodox transmitter of Former Han teachings on the Kung-yang Commentary. He charged that K'ung, by focusing on much later post-Han interpretations of the New Text "schools system," had missed Ho Hsiu's connection with the Former Han New Text tradition. For the Ch'ang-chou New Text scholars, Ho Hsiu represented the only surviving link to Former Han New Text teachings. The latter derived from the "schools system" of the Annals in the Imperial Academy when the erudites Hu-wu Sheng and Tung Chung-shu had championed the Kung-yang chuan.^[33]

By not accepting Ho Hsiu as the legitimate transmitter of the "original meanings" (pen-i ) in the Annals described by the Kung-yang chuan K'ung Kuang-sen had—according to Liu Feng-lu—overlooked the "bequeathed theories of earlier teachers of the Kung-yang and those of [Confucius's] seventy disciples." In particular, Liu asserted that K'ung had misconstrued the san-k'e (three classifications) and chiu-chih (nine points) formulations upon which the Former Han Kung-yang position on the Annals had been based. Rather than basing his theories on the orthodox commentaries—that is, the Kung-yang and Ku-liang—K'ung Kuang-sen had relied on the text of the Classic alone for his interpretation.

Moreover, instead of following reliable Later Han dynasty sources (such as Ho Hsiu's surviving writings) on the meaning of san-k'e and chiu-chih, K'ung Kuang-sen accepted unorthodox definitions which Liu Feng-lu thought betrayed the original meanings of these central terms. How ironic that this defense of Confucius by a New Text scholar, Liu Feng-lu, includes an attack on a direct descendant of the sage. Liu charged in effect that K'ung had swallowed an Old Text bill of goods that diminished Confucius's status. Ch'ang-chou New Text scholars now claimed preeminence.^[34]

Differences in interpretation between K'ung Kuang-sen and Liu Feng-lu were substantial and politically significant. For the "three classifications" in the Kung-yang Commentary, K'ung Kuang-sen gave (1) "the Way of heaven" (t'ien-tao ), (2) "the methods of ruling" (wang-fa ), and (3) "human sentiments" (jen-ch'ing ). The "nine points" were organized in threes according to the "three classifications" (see list below). The Way of heaven was defined in naturalistic terms: (1) seasons, (2) the lunar month, and (3) the solar day. Methods of rulership were conceived of as (4) criticism (chi ), (5) blame (pien ), and (6) renunciation (chueh ). Human sentiments were divided into (7) respect (tsun ), (8) intimacy (ch'in ), and (9) uprightness (hsien ). K'ung's interpretation implied an unchanging cosmos within which constant natural, political, and moral phenomena interacted. The vision was at odds with the voluntaristic interpretations of change and reform that Tung Chung-shu and Ho Hsiu had read into the Annals.^[35]

Although rejecting the Tso chuan in favor of the Kung-yang chuan as an acceptable commentary on the Annals, K'ung Kuang-sen relied in part on the Later Han Confucian Sung Chün's (fl. ca. A.D. 25-26) explication of the "nine points" and not on Ho Hsiu's. In addition, K'ung refused to accept the cyclical interpretations of the "three classifications," which both Ho Hsiu and Sung Chün had acknowledged. What disturbed K'ung about Ho Hsiu's version of the san-k'e and chiu-chih was that it contained superstitious prophecy, which Kuang-sen thought anachronistic. Confucius was a teacher not a prophet; he could not have known what the future held. To read into the Annals prophecies regarding the Han dynasty, according to K'ung, "had no basis in fact." K'ung thus refused to follow the Former Han version of Kung-yang Confucianism. He rejected Ho Hsiu's commentary because Ho had been misguided, his views not only speculative but also wildly irrational (tsung-heng i-shuo ). They deserved, K'ung thought, the disrepute they had for centuries received.^[36]

Skepticism concerning the prophetic aspects of New Text classical scholarship had been voiced in Ch'ang-chou by Tsang Lin (1650-1713) in his pioneering efforts to restore Han Learning to classical respectabil-

ity (see chapter 4). The notion that the Annals was a repository of prophecy did not sit well with the empirical bent of the early champions of Ch'ing Han Learning. Similarly, the k'ao-cheng historian Ku Tung-kao (1679-1759), a native of Wu-hsi County in Ch'ang-chou, spent a lifetime reconstructing the historicity of the Annals; he attacked New Text advocates for their defense of superstitions that mainstream Confucians could not comprehend.^[37]

Liu Feng-lu—using evidential research techniques to defuse the criticism of New Text ideas by Han Learning scholars—attacked K'ung Kuang-sen's interpretation of the "three classifications" and "nine points" because it had no basis in the Han classical "schools system" of the Former and Later Han dynasties. By disregarding Tung Chung-shu and Ho Hsiu, K'ung had avoided the inevitable conclusion that Confucius had indeed intended the apocalyptic implications captured by New Text interpretations of the Five Classics. Ho Hsiu's subcommentary on the "three classifications" and "nine points," according to Liu Feng-lu, was the true legacy of the Former Han New Text orthodoxy.

However difficult it might be for Old Text advocates to fathom, the prophecies and prognostications that were part of Tung Chung-shu's portrait of Confucius as sage and "uncrowned king" reemerged in the writings of Sung Hsiang-feng and Liu Feng-lu. Liu was integrating the Former Han vision of Confucius as larger than life, as an "uncrowned king" who had appeared "not because Heaven had wanted to save the Eastern Chou [dynasty] from chaos, but instead to mandate him through the Annals to save ten thousand [succeeding] generations from chaos."^[38]

Ho Hsiu and the "Three Classifications"

Liu Feng-lu maintained that Ho Hsiu had been careful to transmit the interpretations of his predecessors. Ho's definitions were therefore, truer, Liu concluded, to the Former Han New Text tradition than the sources from which K'ung Kuang-sen had drawn. Liu accordingly followed Ho Hsiu on the "three classifications": (1) "the unfolding of three epochs" (chang san-shih ), (2) "going through three periods of uni-

ty" (ts'un san-t'ung ), and (3) "differentiating the outer [barbarians] from the inner [Chinese]" (i wai-nei ). As the following outline suggests, Ho Hsiu's position was more radical in its exposition than K'ung Kuang-sen's because, as Liu Feng-lu pointed out, Ho's interpretation elaborated a notion of cosmic and dynastic change, which Ho's predecessor Sung Chün had also accepted:

As an underlying theory of change and reform the "three classifications" were, in Ho Hsiu's hands, tied very closely to the "nine points" (tsai san-k'e chih nei ). Although Sung Chün had accepted the same version for the san-k'e that Ho Hsiu had, he had not subsumed the "nine points" under them. For Sung, "the nine points were outside the three classifications" (chiu-chih tsai san-k'e chih wai ). Consequently, the latter seemed to have no direct correlation to the former, a point that had permitted K'ung Kuang-sen to reject Ho's and Sung's version of the "three classifications" and yet retain Sung's "nine points" intact (see list).

Sung Chün had given priority to the "unfolding of three epochs" as the "first classification." In contrast, Ho Hsiu made "going through the

three periods of unity" the "first," and relegated the "unfolding of three epochs" to the "second classification." For Ho Hsiu, "going through the three periods of unity" included "three points": (1) "making Chou the new dynasty" (hsin-Chou ), (2) "relegating [the former Shang dynasty] to [the state of] Sung" (Ku-Sung ), and (3) "using the Spring and Autumn Annals as [the basis for] establishing the new king [to come]" (i Ch'un-ch'iu tang hsin-wang ).

Ho Hsiu's "first classification" and its "three points" implied that a mandate (ming ) to rule had been granted to Confucius and prophetically encoded in the Annals "to relegate the Chou dynasty and entrust the kingship to the state of Lu" (Ku-Chou wang-Lu ) as a harbinger of the Han dynasty unification of China. The image of Confucius as an uncrowned king of the state of Lu was clearly magnified by this conception of change and reform in the three epochs (san-shih ) of time. In another work on Ho Hsiu, Liu Feng-lu explained:

"To entrust the kinship to Lu" is what the Annals means by "using the Spring and Autumn Annals as [the basis for] establishing the new king [to come]." The master received the mandate to establish institutions. Thinking that it would be better to put them into effect rather than rely on empty words, [Confucius's wisdom] was broad and deep, discriminating and clear. He cited histories and records, imbuing them with the mind of a king. Mencius said "the Annals represents the affairs of the Son of Heaven." In establishing the models for the new king and awaiting later sages [to put them into effect], why did [Confucius] have to choose Lu? The answer is that he used the written history of Lu to avoid giving the impression that in creating institutions he was usurping the prerogatives of the ruler. Moreover, of what Confucius had known and heard, only the [affairs of] Lu were close at hand. Therefore, he chose Lu as the capital region for enunciating the basis of governance.^[39]

The "unfolding of the three epochs" as the "second classification" in Ho Hsiu's interpretation overlapped with the scale of historical change enunciated in the first. The "three points" belonging to this stage were: (4) "using different terms for what [Confucius personally] witnessed" (so-chien i-tz'u), (5 ) "using different terms for what [Confucius personally] learned about" (so-wen i-tz'u ), and (6) "using different terms for what [Confucius] had learned through transmitted records" (so-ch'uan-wen i-tz'u ). These middle three points established that

Confucius had encoded the Annals to differentiate the epochs of the Chou dynasty.

For the three rulers of Lu during Confucius's own lifetime (so-chien ), the Annals "made its terminology esoteric" (wei ch'i tz'u ). For the four preceding rulers Confucius had heard about (so-wen ), the Annals "expressed sorrow for the calamities that had occurred" (t'ung ch'i kuo ). Finally, for the five earliest rulers of Lu about whom Confucius had only read, the Annals set aside compassion and gave a dispassionate account (sha ch'i en ). Unlike K'ung Kuang-sen, Chuang Ts'un-yü in opening remarks for his Correcting Terms in the Annals, for example, had already stressed the importance of the "three ages" for grasping Confucius's intent:

[Confucius] based himself on [the reign of Duke] Ai [494-468 B.C. ] and recorded events [back to the reign of Duke] Yin [722-712 B.C. ]. With sympathy he pointed to [events deserving] recognition or criticism. His intentions were flexible but firm, his words were [at times] detailed [at other times] sketchy. [Yet] he did not endanger himself with his knowledge. His sense of right did not threaten his superiors. Where [punishment for] a crime had not yet been determined, his terminology served as reference. Rid of chaos and with sprouts of orderly rule, the world gradually reached [the age of] ascending peace. The twelve [reigns of the rulers of Lu] were images [of a pattern]. [The age of] great peace was brought to completion.

The unfolding of the three epochs was expressed through a "politics of language" that also implied a cosmic vision. According to Chuang Ts'un-yü and Liu Feng-lu, that vision represented a historical transition from a time of chaos (luan, that is, what Confucius read about), to a time of "ascending peace" (sheng-p'ing, that is, what Confucius heard about). These two epochs culminated in a time of "great peace" (t'ai-p'ing, that is, what Confucius personally saw). According to Ho Hsiu's version of the "second classification," Confucius had encoded the Annals with a vision of "ascending order" that would yield a "great unification" (ta i-t'ung ).

Tung Chung-shu (upon whom Ho Hsiu's and hence Liu Feng-lu's accounts depended) had already noted that the 242 years (counting to 481 B.C. ) of the Annals were divided into twelve generations and then subdivided into "three groups" (san-teng ). According to the Kung-yang Commentary, Confucius composed the Annals in such a way that particular terminology was applied to each epoch, which demonstrated the historical judgment he had attached to each event. Tung Chung-shu enlarged on the Kung-yang view:

The Annals is divided into twelve generations, which fall into three groups: those that were [personally] witnessed [by Confucius], those that he heard of [from eider contemporaries], and those that he heard of through transmitted records. Three [of these twelve generations] were [personally] witnessed, four were heard of [through oral testimony], and five were heard of through transmitted records. . .. Those that were [personally] witnessed comprise 61 years [541-480 B.C. ); that were heard of, 85 years [626-541]; and that were heard of through transmitted records, 96 years [722-626].

Regarding what he witnessed, Confucius used concealing phraseology; regarding what he heard of, he expressed sorrow for calamities; regarding what he heard through transmitted records, he set his compassion aside [and wrote dispassionately]. This is in accordance with the feelings [appropriate to each situation].

On the cosmological level, Tung's theory of three epochs was em bedded in a wide-reaching notion of cyclical change (pien ), which was caused by the complementary interactions of yin and yang and correspondences with the five evolutive phases (wu-hsing ). Political theory, the world of rulers and officials, was intertwined with cosmology, the workings of heaven. Moral philosophy of pre-Han Con-fucianism was wedded to the cosmological theories of nonorthodox schools prominent in intellectual life before and during the Former Hah dynasty. Tung explained:

The method of the Annals is to cite events of the past in order to explain those of the future. For this reason, when a phenomenon occurs in the world, look to see what comparable events are recorded in the Annals ; find out the essential meaning of its subtleties and mysteries in order to preserve the significance of the event; and comprehend how it is classified in order to see what causes are implied. Changes wrought in Heaven and on earth, and events that affect a dynasty will then all become crystal clear, with nothing left in doubt.

Cosmological and dynastic change, in Tung Chung-shu's political theory, brought with it the need for institutional change. For each of the three epochs there was an appropriate institutional framework. Institutions like dynasties changed cyclically according to their correspondences to the five phases and yin-yang. The "three unities" (san-t'ung ) evolved through time and space, bringing along institutional change in their wake:

Ancient kings, after receiving [Heaven's] mandate, which made them kings [of new dynasties), changed the institutions, titles, and beginning of the year [that had been in force]. Having determined the color for clothing, they announced at the suburban sacrifices [the accession of their dynasty] to Heaven and earth and the multitude of spirits. They offered sacrifices to their

distant and nearer ancestors, and then proclaimed [the accession of their dynasty] throughout the empire. This [proclamation] was received in their ancestral temples by the feudal lords, who then announced it to the spirits of mountains and streams. Thus there was a single rule [for all] to respond to. . .. This was the way in which Heaven's sequences were made clear.

Third and last of the "three classifications" formulated by both Ho Hsiu and Sung Chün was "differentiating outer barbarians from inner Chinese" (i wai-nei ). But Ho Hsiu, unlike Sung Chün, set clear boundaries for the content of the last classification through the final three "points": (7) "treat the [Chinese] states as inner" (nei ch'i kuo), (8) "treat the various feudal lords as outer" (wai chu-hsia ), and (9) "treat the various feudal lords as inner and the barbarians as outer" (nei chu-hsia erh wai i-jung ).^[40]

The Annals accordingly harbored a cultural vision of concentrically arranged internal and external groups of peoples, which carried over into the tribute system that was at the heart of the Confucian system of foreign affairs. Inner feudal states of the Chou dynasty had priority over their surrounding tribes and peripheral barbarians. This New Text view of foreign affairs had served as the underlying framework for Liu Feng-lu's resolution of the diplomatic conflict with the Vietnamese emissary, summarized above.

According to Liu, the Annals implied that political status was culturally, not racially, defined. The transformation from barbarian "outsider" status to Confucian "insider" status was an ongoing process of cultural assimilation based on Chinese models of imperial benevolence. Compared with the hard line taken by Northern and especially Southern Sung Confucians on relations with "outer barbarians," Liu Feng-lu's position as an official of a conquest dynasty was more accommodating.^[41]

In the conclusion of his essay, "Discourse on the Annals," Liu Feng-lu pointed out that K'ung Kuang-sen's definitions for the "three classifications" and "nine points" lacked the dynamic vision of historical events encoded by Confucius in the Annals. By missing these elements in the

Annals, K'ung Kuang-sen had taken sides with the Old Text school despite his work on the Kung-yang chuan. Liu concluded: "Without the 'three classifications' and 'nine points,' there is no Kung-yang [Commentary]. Without the Kung-yang, there is no Spring and Autumn Annals. What 'esoteric words' are there then?"^[42]

Liu Feng-Lu on the Analects

In his refutation of Ch'ien Ta-hsin and correction of K'ung Kuang-sen, Liu Feng-lu was motivated by a desire to restore a sense of grandeur to Confucius. At the same time, Liu saw in the Kung-yang view of the Annals a historical framework that would be useful in rehabilitating a holistic vision of the Classics that had been largely lost as a result of the cumulative onslaught of Han Learning. To drive home the Kung-yang interpretation of the Annals, Liu Feng-lu, like his cousin Sung Hsiang-feng, turned to Confucius's Analects to confirm the New Text portrait of Confucius as "uncrowned king."

In an 1802 work entitled Lun-yü shu-Ho (Discourse on Ho [Hsiu's commentary on the] Analects), Liu attempted to reconstruct Ho Hsiu's lost commentary on the Analects. While working on Ho's explication of the Kung-yang Commentary, Feng-lu had recognized the need to reconstruct the lost commentaries Ho Hsiu had prepared for the other Classics. Ho's biography in the History of the Later Han Dynasty (Hou Han-shu), Liu noted, made clear that Ho Hsiu "had meticulously studied the Six Classics, and contemporary Confucians could not compare to him." Liu realized that Ho's explication of the Kung-yang commentary on the Annals was but one piece in a much larger classical framework. Feng-lu thought Ho's notes for the Analects would reveal further aspects of the New Text classical legacy.^[43]

In many ways, Liu was following the lead of Sung and Ming dynasty Confucians, who had made study of the Four Books (Analects, Men-cius, Great Learning, and Doctrine of the Mean) a necessary complement, if not prerequisite, to understanding the nearly impenetrable Five Classics. Consequently, by using the Analects as a "summary of the great meanings in the Six Classics," Liu Feng-lu was applying a Sung Learning tradition to the purposes of New Text Confucianism. The

"esoteric words" (wei-yen) of Confucius would be confirmed by one of the Four Books.^[44]

Confucius's Sense of Moral Crisis

In Liu Feng-lu's hands, episodes in the Analects took on esoteric and prophetic meanings that Old Text scholars considered unacceptable. By linking the Annals and the Analects, Liu Feng-lu, Sung Hsiang-feng, and their follower Tai Wang could now interpret Confucius's life in light of Tung Chung-shu's and Ho Hsiu's New Text portrayal of him as an uncrowned king. For example, Liu interpreted a passage in the Analects (dealing with Confucius's sense of cultural crisis in the sixth century B.C. ) by using one of the Kung-yang chuan's "three classifications"—namely the theory of the "unfolding of three epochs" (chang san-shih ). The Analects stated:

The Master said, "Use your ears widely but leave out what is doubtful; repeat the rest with caution and you will make few mistakes. Use your eyes widely and leave out what is hazardous; put the rest into practice with caution and you will have few regrets. When in your speech you make few mistakes and in your action you have few regrets, an official career will follow as a matter of course."

This passage discussing "ears," "eyes," and "speech" conveniently permitted Liu Feng-lu to branch off into discussion of Confucius's his-toriographical designs on the history of Lu and the Chou dynasty based on the "three points": (1) what Confucius had personally seen, (2) what he had personally heard, and (3) what he had read about. The three epochs were confirmed by the Analects, Liu thought.^[45]

Elsewhere, Liu Feng-lu used a passage in the Analects to illustrate Confucius's notion of dynastic change by referring to the New Text doctrine of "using the Annals as the basis for the establishment of the new king to come." The passage in the Analects read:

Tzu-chang asked, "Can ten generations hence be known?" The Master said, "The Yin [Shang dynasty] built on the rites of the Hsia. What was added and what was omitted can be known. The Chou built on the rites of Yin. What was added and what was omitted can be known. Should there be a successor to the Chou, even a hundred generations can be known."

Liu interpreted Confucius's reply as confirmation of the vision in the Annals of change according to historical circumstances (see chapter 8) whereby "kingship would be entrusted to the state of Lu" (wang-Lu ), an obilique reference to a new dynasty that would succeed the Chou. The succeeding dynasty would build on the rites of the Chou, just as the Chou had relied on its predecessor. Again, the Analects disclosed the prophetic nature of Confucius's classical vision.^[46]

One of the key doctrines in the Analects, accepted by all succeeding Confucians, was the "rectification of names" (cheng-ming ). According to Confucius, the rectification of names referred to a social order in which human behavior corresponded to clearly defined names of social and political functions. Social order demanded a one-to-one correspondence between orderly language and orderly behavior. Confucius had said: "To govern [cheng] is to correct [cheng]. If you set an example by being correct, who would dare to remain incorrect?" "To govern" and "to correct" were cognates demonstrating the moral prerequisites of political and social life.^[47]

For Liu Feng-lu, the rectification of names was best exemplified in the Annals. As a vision of society, the Annals was encoded with a program of "praise and blame" that criticized the breakdown of the social order and directed attention to the hypocrisy of the feudal rulers, whose ideals were no longer grounded in actual practice. Confucius's histo-riographical principles in the Annals, according to Feng-lu, were a detailed demonstration of the "rectification of names" in the Analects. The centrality of ritual in social and political life was a major theme in both the Kung-yang chuan and the Analects, Liu added.^[48]

Visions of Epochal Change

As an analog in the Analects to the Kung-yang doctrine of "going through three periods of unity," Liu cited the following passage: "The Master said, 'The Chou is resplendent in culture, having before it the example of the two previous dynasties. I am for Chou.'" According to this passage, the Hsia, Shang, and Chou dynasties were models for dynastic change. Each represented a "period of unity" that gave way to its successor. Each successor, however, was indebted to its predecessor for its "cultural substance" (wen-chih ). Confucius had enunciated in

both the Analects and the Annals a theme of epochal change based on dynastic cycles (hsun-huan ) that served, according to Liu Feng-lu, as models for ten thousand generations.^[49]

Liu then applied this vision of epochal change to an elaboration of Confucius's position as an "uncrowned king." Once again, the Analects was cited to confirm the Kung-yang's apocalyptic vision:

The border official of I request an audience, saying, "I have never been denied an audience by a gentleman who has come to this place." The followers presented him. When he came out, he said, "What worry have you, gentlemen, about the loss of office? The Empire has long been without the Way. Heaven is about to use your Master as the wooden tongue for a bell."

For Liu Feng-lu, the Analects recorded an episode confirming that Confucius "knew that he was about to receive the mandate [to rouse the empire] and accordingly compiled the Annals to provide lessons for ten thousand generations." The border official at I had recognized Confucius as the "uncrowned king."^[50]

This anticipation of Confucius's future in the Analects, Liu thought, was even more explicit in the famous passage affirming Confucius's sense of mission:

When under siege in K'uang, the Master said, "With King Wen [founder of the Chou dynasty] dead, is not culture invested here in me? If Heaven intends culture to be destroyed, those who come after me will not be able to have any part of it. If Heaven does not intend this culture to be destroyed, then what can the men of K'uang do to me?

In this remarkable passage, Confucius acknowledges his historical mission as the successor to King Wen. The passage dovetailed with the Kung-yang claim that Confucius had communicated his cultural mission to his closest disciples. The Classics, then, were the textual vehicle for doctrines transmitted to posterity by disciples who had been orally initiated in the "esoteric words" and "great meanings" by Confucius himself. In other words, the Analects represented an oral record of esoteric teachings. Liu concluded: "Whenever the Analects and Annals mutually correspond to each other, these are all oral transmissions of the 'esoteric words' from the sage [to his disciples]. These were not included in the bamboo and silk versions [of the Annals]." Liu claimed

that Ho Hsiu and Tung Chung-shu, by building on the Kung-yang chuan, were direct inheritors of this esoteric tradition.^[51]

We have seen that the case for Confucius's encoding the Annals with his esoteric doctrines of praise and blame was argued during the Han dynasties in light of the abrupt ending to the chronicles of Lu in 481 B.C. with the capture of the lin. Tung Chung-shu interpreted the capture of this mythical "unicorn" as the prophetic sign that Confucius had received the mandate of heaven to rule. According to the Kung-yang account echoed later by Ho Hsiu and repeated in the Household Sayings of Confucius, the sage had himself read into the capture that all hope for the Chou kings was lost. Its mandate to rule had ended:

A wagoner of Master Shu-sun named Tzu-ch'u-shang was gathering wood in Ta-yeh, when he came upon a lin. Because its left foreleg was broken, he brought it back home with him. Ch'u-shang, thinking the discovery inauspicious, threw the lin away outside the suburbs. He sent a messenger to inform Confucius saying: "There is a fallow-deer and it is horned. What can it be?" Confucius went to see it and said: "It is a lin . Why has it come? Why has it come?" He turned his sleeve and wiped his face. His tears wet his robe. When Shu-sun heard what it was, he had it brought to him. Tzu-kung [one of Confucius's disciples] asked the master why he was crying. Confucius replied: "The lin comes only when there is an enlightened monarch [on the throne]. Now it has come when it is not time for it to appear, and it has been injured. This is why I am so afflicted."

Liu Feng-lu found confirmation of this New Text tradition in the Analects as well: "The Master said, 'The phoenix [feng-wu] does not appear, nor does the river offer up its Chart [that is, the Ho-t'u]. I am done for.'"

Because the phoenix and River Chart were auspicious omens, Liu explained that Confucius's sense of hopelessness during the declining years of the Chou was a major element in both the Annals and Analects. Liu concluded:

This saying [in the Analects ] probably occurred after the capture of the lin [recorded in the Annals ] and [before] Confucius's death. This is proof that Heaven was announcing to Confucius that he would soon perish. The Chou house would soon vanish. Sages would no longer appear. That is why Confucius said, "Why has it [the lin ] come?" And also why he said, "My way has come to an end."

The chronology of the Analects and the completion of the Annals in

481 B.C. yielded overlapping evidence that late in life Confucius had recognized in natural omens the end of an era.^[52]

Ho Hsiu's versions of the Analects and Annals revealed the apocalyptic context that informed both records of the late Chou period. For Liu Feng-lu, both the Kung-yang chuan and the sayings recorded by Confucius's immediate disciples supplied evidence to gainsay the Old Text version of Confucius's life and his historical intent. The Analects belonged to the New Text tradition, Liu contended, not the Old Text: "The Analects represents a general statement of the great meanings in the Six Classics. It clarifies the subtle words of the Spring and Autumn Annals. That is why those who worked on the Old Text [versions] such as [K'ung] An-kuo [156-747 B.C. ] and [Cheng] K'ang-ch'eng [i.e., Hsuan] could not master [the Analects ]."^[53]

The portrait of Confucius as a heroic sage and "uncrowned king" had reappeared in Confucian rhetoric and classical theory. Liu Feng-lu had successfully made the transition from his grandfather's diffuse commitment to the Kung-yang commentary to New Text Confucianism.

Tai Wang and Ch'En Li on Confucius

Both Tai Wang and Ch'en Li, followers of New Text Confucianism in the mid-nineteenth century, lent support to the reappraisal of Confucius's role in forging the classical vision of state and society. They accepted as accurate Liu Feng-lu's portrait of the Master. They compared the Analects and Annals topically and chronologically to add further support for the New Text position.

Tai Wang is perhaps better known for his efforts to promote the teachings of the seventeenth-century northern Yen-Li school (after Yen Yuan [1635-1704] and Li Kung [1659-1733]). But in naming his studio "Hall of the Fabulous Unicorn" (Che-lin-t'ang ), Tai disclosed the personal importance of the Annals' prophetic elements—particularly the capture of the lin. The studio name was also the title of Tai Wang's collected writings. In following Liu Feng-lu's and Sung Hsiang-feng's efforts to tie together the Annals and Analects, Tai concentrated on the capture of the lin as evidence for the New Text portrayal of Confucius.^[54]

To demonstrate that Confucius had received heaven's mandate, Tai cited the celebrated passage in the Analects: "The Master said, 'At fifteen I set my heart on learning; at thirty I took my stand; at forty I came to be free from doubts; at fifty I understood Heaven's mandate; at sixty my ear was attuned; at seventy I followed my heart's desire without overstepping the line.'" Tai Wang concluded: "As an uncrowned king, Confucius created law-models and the Five Classics to continue the law-models of Chou for another hundred generations. This is what Heaven had mandated."

Tai added that the reference in the Arialeers to Confucius (551-479 B.C. ) at age seventy corresponded to the date (481 B.C. ) in the Annals when the fabulous lin appeared and signaled to the Master that the Chou dynasty was coming to an end. Confucius died two years later in 479 B.C. at age seventy-two. Confucius's claim that "at seventy I followed my heart's desire without overstepping the line" was, according to Tai Wang, full of pathos and vision. A time of chaos was reaching its climax; a time of peace was at hand.^[55]

By agreeing with Mencius's assessment that Confucius had compiled the Annals to bring order to a time of chaos, Tai Wang linked Men-cius's views to the Kung-yang Commentary. Tai maintained that the latter was the only commentary on the Annals that conferred the appropriate importance to the chronicles of the state of Lu from 722 to 481 B.C. Mencius viewed the Spring and Autumn Annals with a seriousness generally accepted by later Confucian scholar-statesmen:

The world fell into decay, and the Tao became obscure. Perverse speech and oppressive deeds again arose. There were instances of ministers who murdered their rulers, and of sons who murdered their fathers. Confucius was fearful and compiled the Spring and Autumn Annals . The Annals represents the affairs of the Son of Heaven [that is, the Chou dynasty king as opposed to the rulers of each feudal state]. Therefore, Confucius said: "It will be on account of the Annals that people will know me. It will be due to the Annals that people will condemn me."

Like Liu Feng-lu and Sung Hsiang-feng, Tai used the "three classifications" and "nine points" as frameworks for reevaluating the Analects from the Kung-yang point of view. The Analects became a window onto a New Text landscape.^[56]

Tai interpreted Confucius's frequent mention of the early Chou dynasty and the legacy of the Duke of Chou as evidence that the Master's vision for the future was drawn from the ideals of the past. Using the state of Lu as the focus for the chronicles in the Annals, Confucius, according to Tai Wang, was "entrusting the kingship to Lu" (wang-Lu): "Lu encompassed the rituals and music of the four dynasties [Hsia, Shang, Chou, and Lu]. Through the single transformation [from Chou to Lu], the Way can be achieved. The Annals accepts Lu as the ordering realm. Consequently, Confucius did not want to distance Lu [in his account]."

Confucius endeavored to make Lu the locus for understanding the principles of kingship and the restoration of ideals enunciated by the Duke of Chou. Both Liu Feng-lu and Tai Wang regarded the Annals as a framework for "institutionalizing laws of a new king while awaiting the appearance of later sages."^[57]

The chronicles in the Annals, by delineating the decline of the Chou dynasty and prophesizing the rise of the Han restoration of the classical unity (ta i-t'ung ), furnished a framework for political development through the "three unifications" (san-t'ung ) and cosmologicai change through the "three epochs" (san-shih ). Tai added:

Confucius said if the Way of the Hsia dynasty had not been lost, then the moral power of the Shang dynasty would not have arisen. If the moral power of the Shang dynasty had not been lost, then the moral power of the Chou dynasty could not have arisen. If the moral power of the Chou dynasty had not been lost, then the Annals would not have appeared. Once the Annals appeared, gentlemen realized that the Way of the Chou had been lost.

To outline a time of chaos, Confucius had compiled the Annals as a critique of the contemporary world. To prophesize a time of "great peace" (t'ai-p'ing ), Confucius had ended the Annals with the capture of the lin. According to Tai Wang, Confucius "had relied on historical records to compile the Annals as the proper guide of kingship."^[58]

Although a follower of Liu Feng-lu on the Kung-yang chuan, Ch'en Li was also closely associated with the Tso chuan specialist Liu Wench'i in Yang-chou. The latter's stress on the Tso chuan became a lifelong commitment and a prominent feature of the Lius' subsequent schol-

arly traditions (see chapter 1). In his 1843 preface to Ch'en Li's collected works, Liu Wen-ch'i praised Ch'en's Kung-yang research because it drew heavily on Cheng Hsuan's Later Han classical studies. An Old Text advocate, Liu Wen-ch'i regarded Ch'en Li as superior to both K'ung Kuang-sen and Liu Feng-lu, who had blindly followed Ho Hsiu's superstitions instead of Cheng Hsuan's more reliable commentary.^[59]

At first sight, Liu Wen-ch'i's claims would seem to place Ch'en Li more in the Han Learning tradition of Yang-chou than in the Ch'ang-chou New Text revival. If Wen-ch'i were right, then by relying on Cheng Hsuan for his Kung-yang studies Ch'en Li was in effect giving an Old Text gloss to a New Text commentary. Moreover, Ch'en had studied under the Han Learning philologist Tuan Yü-ts'ai, whose preference for Old Text recensions of the Five Classics—particularly the authentic portions of the Documents Classic—had been criticized by Ch'en Shou-ch'i and Liu Feng-lu. Both scholars had favored New Text recensions.

Upon careful examination, however, Liu Wen-ch'i's efforts to add an Old Text flavor to Ch'en Li's research on the Kung-yang Commentary were overstated. In fact, Liu Feng-lu's influence on Ch'en's scholarship went deeper than Liu Wen-ch'i supposed. Ch'en included an essay in his collected writings that explicitly stated his acceptance of Liu Feng-lu's efforts to reestablish the Kung-yang Commentary as the proper window through which to view the Annals. Entitled "The Annals' Theory of Entrusting the Kingship to Lu" (Ch'un-ch'iu wang-Lu shuo ), Ch'en's essay came out in support of the New Text claim that the Annals had been compiled in order to establish the state of Lu as the precursor of the Han unification.^[60]

By affirming Confucius as an "uncrowned king" whose Annals laid out the "law-models" (fa ) for the "way of kingship" (wang-tao ), Ch'en Li unequivocally accepted the Kung-yang vision of the Annals as first articulated by Chuang Ts'un-yü and then fleshed out by Liu Feng-lu. In Ch'en's mind, "there were no doubts concerning the New Text theory in the Annals of entrusting the kingship to Lu." On all of these points, Ch'en Li was affirming a key element in Ho Hsiu's explication of the Kung-yang chuan and rejecting Cheng Hsuan's criticism of Ho's posi-

tion. The capture of the lin in 481 B.C. represented heaven's mandate and the Annals' calling for a new dynasty.^[61]

Following Liu Feng-lu's position, Ch'en Li rejected Old Text efforts to reduce the Annals to "historical records of the state of Lu" (Lu shih-chi ). The Annals was more than a chronicle of events. Its "extremely unorthodox theories" (fei-ch'ang k'e-kuai chih lun ), according to Ch'en, were a sign of the encoded design of the work. Consequently, Ch'en explicitly rejected the Old Text dismissal of Ho Hsiu's commentary. For example, in a work entitled Kung-yang wen-ta (Questions and answers on the Kung-yang Commentary ), Liu Wen-ch'i's maternal uncle and mentor Ling Shu ridiculed Ho Hsiu's views as "the empty words of Han Confucians." Ling Shu, like Liu Wen-ch'i, applied Old Text criteria to the Kung-yang tradition. Ch'en Li gainsaid this position and argued instead that the Annals "illuminated the esoteric words and great meanings of Confucius's seventy disciples."^[62]

Liu Feng-lu's influence on Yang-chou scholars such as Juan Yuan, Ling Shu, and Ch'en Li demonstrates the transmittal of the Ch'ang-chou New Text school to the home ground of Han Learning. Although Liu Wen-ch'i and others sought to downplay the extent of the Ch'ang-chou New Text influence in Yang-chou, it was clear that the Ch'ang-chou scholars had to be taken seriously by the Old Text camp. Subjected to Han learning influence from Yang-chou and Su-chou for much of the eighteenth century, Ch'ang-chou had by the nineteenth century turned the tables: New Text ideas were penetrating Su-chou through Sung Hsiang-feng and Yang-chou via Ch'en Li.^[63]

Liu Feng-LU on the Tso Chuan

Liu Feng-lu's efforts to restore Confucius to his proper place in the New Text scheme of things were subsequently complemented by his influential assault on the Tso chuan. In his controversial but highly acclaimed Tso-shih ch'un-ch'iu k'ao-cheng (Evidential analysis of Master Tso's Spring and Autumn Annals), published in 1805, Liu Feng-lu made extensive use of k'ao-cheng techniques to prove that the so-called Tso Commentary originally had not been a commentary on Confucius's Spring and Autumn Annals. Liu marshaled his evidence into a carefully

arranged series of arguments designed to gradually expose Liu Hsin (45 B.C.—A.D. 23). Liu argued that during Wang Mang's interregnum (A.D. 9—23) Liu Hsin manipulated the Rituals of Chou and another text known simply as Master Tso's Spring and Autumn Annals (Tso-shih ch'un-ch'iu ) in order to discredit the New Text school of the Former Han dynasty and to support Wang Mang's usurpation of power, thereby undercutting the credibility of the Old Text position.^[64]

The Problem of Tso Ch'Iu-Ming

We have noted that attacks on the orthodoxy of the Tso chuan were not uncommon. Many had disagreed with Tu Yü (222-284) and Liu Chih-chi (661-721), who during the post-Han and early T'ang periods asserted that the author of the Tso chuan, Tso Ch'iu-ming, was a direct disciple of Confucius. Because of Tso's status as Confucius's disciple, Tu Yü and Liu Chih-chi (following Liu Hsin's lead) preferred the Tso chuan over the two other commentaries dating from the Former Han dynasty. T'ang commentators such as Tan Chu (725-70) and Chao K'uang (fl. ca. 8th century A.D. ) had disputed such stress on the Tso chuan, rejecting the claim that Tso Ch'iu-ming had been its author.

Despite additional criticism during the Sung and Ming dynasties, however, the Tso chuan remained not only an orthodox commentary to the Annals but also the preferred Old Text commentary for interpreting the Annals and Confucius's Analects. Perhaps the most vitriolic criticism of the Tso chuan came from the late Ming Confucian Hao Ching, who was the first to suggest that the original text of the extant commentary had not been assembled to accompany and explicate Confucius's Annals. Rather, he contended, the Tso chuan had originally been a totally separate historical record, which was later "discovered" and reshaped into a commentary to the Annals by Liu Hsin and his followers.^[65]

The Uses of K'Ao'Cheng

Liu Feng-lu's Evidential Analysis of Master Tso's Spring and Autumn Annals picked up where earlier criticism of the Tso chuan had left off.

Unlike his predecessors, however, Liu supported his attack with empirical philological documentation, thus placing for the first time this longstanding debate directly in the k'ao-cheng mainstream. Ironically, Liu's models for argumentation were Yen Jo-chü's (1636-1704) influential Evidential Analysis of the Old Text Documents (Shang-shu ku-wen shu-cheng ) and Hui Tung's Examination of Old Text Documents Classic (Ku-wen Shang-shu k'ao ), both of which in the seventeenth and eighteenth centuries provoked heated discussion among Confucian scholars—including, as we have seen, Chuang Ts'un-yü. Liu Feng-lu's work precipitated the same sort of debate in the nineteenth century.^[66]

Liu Feng-lu's use of empirical philological criteria to demonstrate the unreliability of the Tso chuan reveals the deep imprint left by Han Learning and its research agenda on New Text studies. Liu realized that if New Text studies lacked systematic methods of demonstration, it would not receive scholarly recognition—perhaps fearing the obscurity endured by his grandfather Chuang Ts'un-yü. The latter's scholarship, for instance, lacked the rigorous epistemological leverage that Yen Jo-chü, Hui Tung, and Tai Chen had all championed in their evidential research. Moreover, Liu had learned his research lessons from his uncle Chuang Shu-tsu very well indeed. Liu's exposé forced many scholars to reassess the reliability of the Tso chuan as the orthodox commentary on the Spring and Autumn Annals.

Liu's Evidential Analysis of Master Tso's Spring and Autumn Annals caused a major stir in k'ao-cheng circles. For example, Ch'en Li (1810-82—not to be confused with Ch'en Li from Yang-chou discussed above), a leading Cantonese Han Learning scholar, rejected Liu Feng-lu's criticism of the Tso chuan and Liu's reliance on Ho Hsiu's interpretations. Ch'en's attack indicated, however, that in the late nineteenth century Liu Feng-lu's thesis was still considered important enough, even in Kuang-chou, to refute.

Included in Juan Yuan's Ch'ing Exegesis of the Classics, the Tso-shih ch'un-ch'iu k'ao-cheng would eventually influence K'ang Yu-wei directly: he read the Ch'ing Exegesis in the 1880s. In addition, Liu Feng4u effectively opened the door to a series of later studies by Wei Yuan, Liao P'ing (1852-1932), and K'ang Yu-wei, among others, which would link Liu Hsin to the forging of other Old Text versions of the Classics, including the Old Text Documents and the Rituals of Chou. The rise of New Text scholarship in Kuang-chou via Liao P'ing and K'ang Yu-wei in the 1890s, therefore, cannot be adequately explained without taking

into consideration Liu Feng-lu's research and writings on the Tso chuan.^[67]

Liu Hsin as Master Forger

Liu Feng-lu was the first to argue and then demonstrate that Liu Hsin had manipulated a separate text entitled the Tso-shih ch'un-ch'iu (Master Tso's Spring and Autumn Annals ) (mentioned by Ssu-ma Ch'ien in his Records of the Grand Historian [Shih-ch i ] ) with the intention of overturning the Kung-yang Commentary. Feng-lu asserted that the Tso chuan was originally not a commentary to Confucius's Annals at all. Hao Ching had already said as much in the seventeenth century. But Liu pushed Hao's position further, accusing Liu Hsin of deliberately extracting the Tso chuan from a stylistically and grammatically similar "sister text" known as the Kuo-yü (Discourses of the States) and unscrupulously using it as an orthodox commentary to subvert the New Text Classics of the Former Han dynasty. When added to Yen Jo-chii's and Hui Tung's research on the Old Text Documents—a topic Liu Feng-lu himself focused on in his other research—Liu had added still more doubt concerning the troubling origins of the Old Text versions of the Classics accepted by Han Learning advocates.^[68]

Liu Feng-lu boldly declared and then proved what had before only been suspected: Liu Hsin stood accused of "forgery" (tso-wei ) and of "overturning the Five Classics and causing [succeeding] scholars to be deluded." Liu Feng-lu's tour de force proved that Liu Hsin had subverted the "Kung-yang tradition of meanings and precedents" in favor of the Old Text Classics. Feng-lu wrote:

When I was twelve, I read Master Tso's Spring and Autumn Annals and was suspicious of the correctness of its historiographical methods [shu-fa ] because it left out many of the "great meanings" [in the Annals ]. I went on to read Kung-yang [Kao's] and Master Tung [Chung-shu's] accounts and realized that with regard to the Annals it was not a record of events. Nor was it necessary to take Master Tso's [commentary] into account in order to understand [the Annals ].

Liu added that he had exchanged views with his cousin Sung Hsiang-feng on the matter. After Liu answered Sung's query concerning the

superiority of the Kung-yang Commentary to both the Ku-liang and Tso commentaries, Sung is said to have declared: "You not only are good at ordering the Kung-yang, but in addition you can be regarded as a loyal servant of Master Tso." By rejecting the link between the Tso chuan and the Annals, Liu Feng-lu restored the former to its proper status as an independent historical account of the Warring States period.^[69]

To verify Liu Hsin's manipulation of the Tso chuan for Wang Mang's political purposes, Liu Feng-lu presented a case-by-case discussion of the interpolations in the Tso chuan that revealed Liu Hsin as a forger. Often the text suddenly stated "the gentleman says" (chün-tzu yueh ). For Liu Feng-lu, this repeated and forced pattern revealed precisely where Liu Hsin and his followers introduced interpolations to give the text the appearance of a commentary. Feng-lu added that Chu Hsi had also suspected this forced pattern, concluding that this may have contributed to Chu's low esteem for the Annals .^[70]

New Text as Confucian Orthodoxy

It is interesting that throughout his account, Liu Feng-lu cited Chu Hsi's earlier doubts about the Annals. Yen Jo-chü and other k'ao-cheng scholars frequently referred to Chu Hsi in order to give their textual criticisms orthodox support. Like them, Liu Feng-lu was drawing on the main voice of the Sung Learning orthodoxy to strengthen his side in the New Text-Old Text controversy. According to Liu, the Tso chuan represented a straightforward historical record that missed the moral import of the chronicles included in the Annals.

In contrast to the Tso account, the Kung-yang Commentary enunciated principles of praise and blame and affirmed proper ritual. Liu Feng-lu contended that these marked the moral vision bequeathed by Confucius to posterity. It was as if Liu Feng-lu saw in New Text Con-fucianism the moral strength that would overcome the corrosive effects of Old Text Han Learning and piecemeal philological studies, which by dismissing the latter as a Buddho-Taoist smokescreen had taken their toll on the moral philosophy of Sung Learning.^[71]

For collateral evidence, Liu Feng-lu relied principally on accounts included in Ssu-ma Ch'ien's Records of the Grand Historian. Despite some later interpolations, the Shih-chi was largely a Former Han dynasty work not subject to the Old Text cloud of forgeries that plagued Later Han and T'ang dynasty classical studies. Many scholars linked Ssu-ma Ch'ien to the Former Han New Text orthodoxy because of his close contacts with his friend and colleague Tung Chung-shu. Moreover, Ssu-ma Ch'ien regarded his own historical account as a sequel to Confucius's Annals. In the late nineteenth and early twentieth centuries, a tug of war broke out over Ssu-ma Ch'ien, with both New Text and Old Text advocates claiming him as a partisan of their polarized positions.

In his admiring biography of Confucius, Ssu-ma Ch'ien traced the roots of the historiographical tradition directly back to judgments in the Annals:

Then, utilizing historical records, [Confucius] compiled the Annals , going back to Duke Yin [ruler of Lu, 722-712 B.C. ] and coming down to the fourteenth year of Duke Ai [481 B.C. ], [a total of] twelve dukes. [In the Annals ], he took [the state of] Lu as his standard, kept close to the Chou [dynasty], dealt with Yin [that is, the Shang dynasty] as an ancient time, and propagated [models from] the Three Dynasties. His style was concise, but his meaning rich. Thus, when the rulers of Wu and Yueh [improperly] styled themselves as "Kings," the Annals reproved them by giving them [their proper title of] "Viscount." And at the meeting of Chien-tu [in 632 B.C. ], when the Chou King had actually been ordered to attend [by Duke Wen of Chin], the Annals avoided mentioning this fact by saying: "The celestial king went hunting at Ho-yang." [Confucius] offered examples of this sort to serve as rules for his own age. And if later there are kings who will draw on the intent behind the praise and blame, so that the meaning of the Annals becomes widely known, then rebellious subjects and criminals in the world will be seized with terror.^[72]

Liu Feng-lu relied on the Shih-chi especially to demonstrate that the claims linking Tso Ch'iu-ming to the Tso chuan were patently false. Here, Feng-lu used Ssu-ma Ch'ien to prove the falseness of Liu Hsin's claim that Tso Ch'iu-ming had based his commentary directly on the Master's teachings. Because the Shih-chi did not mention Tso as one of Confucius's disciples—mentioning instead only a work entitled Master Tso's Spring and Autumn Annals, with no indication that it was a commentary to Confucius's Annals—Liu Feng-lu made this "argument

based on silence" a major pillar of his k'ao-cheng inquiry. Later Han accounts in the History of the Later Hah Dynasty and elsewhere were suspicious sources because, Liu said, they had unquestioningly accepted Old Text claims for the legitimacy of the Tso chuan fabricated after Ssu-ma Ch'ien by Liu Hsin.^[73]

According to twentieth-century accounts prepared by distinguished classical scholars such as Liang Ch'i-ch'ao, Ku Chieh-kang, Chang Hsi-t'ang, and Ch'ien Hsuan-t'ung, Liu Feng-lu's Evidential Analysis of Master Tso's Spring and Autumn Annals stood alone alongside Yen Jo-chü's Evidential Analysis of the Old Text Documents Classic as seminal k'ao-cheng studies that prepared the way for total reconsidera-tion of the historical status of the classical legacy after the 1911 Revolution. With the aid of their hindsight, we can properly evaluate how the doubting of antiquity in the twentieth century drew on earlier evidential research. We should add, however, that this modern result was not what Liu Feng-lu, any more than Yen Jo-chü, had intended.

Liu Feng-lu like his grandfather Chuang Ts'un-yü, could see that the fundamentalist thrust behind the Han Learning "return to the ancients" (fu-ku ) threatened to demolish the Sung Learning orthodoxy without providing an alternative moral order and certainty. Although Liu Feng-lu's assault on the Old Text Han Learning position helped to undermine the entire Confucian legacy, his intent was actually the opposite. Liu thought that New Text teachings, drawn from Chuang Ts'un-yü's turn to Kung-yang Confucianism, would fill the vacuum left by eighteenth-century k'ao-cheng. The first of all Confucianisms, the Former Hah New Text orthodoxy, would (Liu hoped) replace Later Han Old Text Han Learning.^[74]

Liu Feng-Lu on Ho Hsiu

After using k'ao-cheng techniques to discredit the Old Text position on the Annals, Liu Feng-lu then proceeded to reconstruct the New Text theoretical position on the Annals and Kung-yang Commentary. With more precision and sophistication than that demonstrated by his predecessors, Liu rebuilt the Kung-yang "schools system" of Ho Hsiu, the

only remaining reliable witness to Kung-yang interpretation from the Later Han. Liu's reconstruction took several forms. Reviewing the published versions of the Ho Hsiu vs. Cheng Hsuan debates of the second century A.D., Liu analyzed Cheng Hsuan's role in the eventual loss of the New Text tradition.

To champion Ho Hsiu meant in effect to attack Cheng Hsuan, the patron saint of Ch'ing dynasty Han Learning. Liu Feng-lu accordingly prepared prefaces for Ho Hsiu's critiques of the Ku-liang and Tso commentaries, to which Cheng Hsuan had written famous rejoinders. Liu's purpose was straightforward: "To spread the theories enunciated by Master Ho [Hsiu] in his Disabling Diseases of Ku-liang [Ku-liang fei-chi] and to counter the rebuttal prepared by Master Cheng [Hsuan]." In his defense of Ho Hsiu's Incurability of Master Tso (Tso-shih kao-huang ), Feng-lu stressed that his aim was not to refute the historical accuracy or usefulness of the Tso chuan. Rather, as he wrote, "I want to restore to the Annals what is its due, and restore to Master Tso's account what it is due." By reopening the Ho Hsiu vs. Cheng Hsuan debate, Liu was unequivocally reasserting the priority of New Text over Old Text for the Han Learning movement.^[75]

In the first of a series of works on Ho Hsiu's Explication of the Kung-yang Commentary, Liu Feng-lu attempted to show why Ho Hsiu's views were superior to Cheng Hsuan's. Liu concluded:

[Cheng] K'ang-ch'eng [i.e., Hsuan] worked on all three commentaries [on the Annals] simultaneously. Therefore, he was not well-versed in the Classic itself. There is only one extant section one can cite from his Expose the Defender [that is, "expose" Ho Hsiu's defense of the Kung-yang ], but it draws mainly from Master Tso. That Cheng's research on Mister Tung [Chung-shu] and Hu-wu Sheng was not very penetrating can be easily ascertained [from this one section].

As a contemporary scholar who "cherished the past and sought the truth" (hao-ku ch'iu-shih ), Liu Feng-lu made clear that his dismissal of Cheng Hsuan did not overturn Han Learning but improved it. The Former Hah New Text tradition, both because of its antiquity and its stress on moral theory, was superior to Cheng Hsuan's classical philology of the Later Hah.^[76]

After summarily dismissing the patron saint of Han Learning, Liu Feng4u turned in 1805 to a complete articulation of Ho Hsiu's position

in Liu's provocative Master Ho [Hsiu's] Explication of the Precedents in the Kung-yang Commentary to the Annals. Liu saw his own efforts in light of the Hah Learning agenda for research. What Hui Tung had accomplished for the reconstruction of Hah dynasty I-ching studies (see chapter 4), Liu hoped to achieve for Hah dynasty studies of the Spring and Autumn Annals, which meant restoring Ho Hsiu to his proper place of eminence in the "schools system" of Later Han classical studies.

Liu's stress on Ho Hsiu and the Former Hah transmission of Tung Chung-shu's New Text teachings on the Annals was exactly parallel, Liu thought, to the success of Hui Tung and Chang Hui-yen in rescuing the Change Classic from its later misrepresentations, which they accomplished by stressing the I -hsueh of Yü Fan. In fact, Chang Hui-yen's research on the Change induced Liu to carry out research on the Yü Fan recension of the Change Classic as well. The Tung Chung-shu and Ho Hsiu axis for New Text Confucianism, Liu asserted, was also central for understanding the I-ching. Moreover, Liu argued the New Text doctrines presented in the Annals were a prerequisite for comprehending fully the Change Classic: "Consequently, before one can discuss the Change, one must first understand the Annals."

The Change encompassed ceremonies and rituals; the Annals used its mastery of ceremony and ritual from the "three dynasties" [Hsia, Shang, and Chou—that is, the "three unifications"] to clarify requirements of statecraft. Like his grandfather Chuang Ts'un-yü, Liu Feng-lu believed the Annals and Change were interlocking classics that opened up the thought world of the ancient sage-kings.^[77]

Liu Feng-lu's use of the term shih-li (explication of precedents) in the title of his reconstruction of Ho Hsiu's position reveals the continuing impact of Tu Yü's precedent-based approach for interpreting the Annals, first seen in Tu's Ch'un-ch'iu shih-li. Chuang Ts'un-y ü also grappled, as we have seen, with this Old Text tradition, bringing censure of Tu Yü's precedent-based analysis into the eighteenth century. Chuang made "precedents" (li ) less important than "key points" (yao- chih ) in his attempt to promote Kung-yang Confucianism. Although Liu Feng-lu consciously mimicked Tu Yü's title in that of his own study, it was clear nonetheless that "Ho Hsiu's explication of precedents"

would, in Liu's account, replace the list of precedents prepared by Tu Yü. Liu put new wine in an old bottle. The "precedents-based" tradition was adapted to a New Text agenda.

Accordingly, Liu's analysis equated "precedents" (li ) with the "three classifications" and "nine points" that Ho Hsiu had used to elucidate the Kung-yang Commentary. Feng-lu, following Chuang Ts'un-yü, streamlined Tu Yü's tedious list of historical precedents culled from the Tso chuan by drawing on New Text theory rather than Old Text his-toriography. Liu's work represented in part an interesting compromise between Tu Yü's precedent-based interpretation and its rejection by Confucians such as Yao Chi-heng (1647-1715?), who had questioned the appropriateness of the "precedent tradition." Among the thirty precedents assembled by Liu Feng-lu, the "three classifications" took priority. The content of precedents had changed from Old Text his-toriography based on the Tso chuan to New Text "great principles" drawn from the Kung-yang Commentary.^[78]

On the "three periods of unity" (san-t'ung ), for instance, Liu moved back and forth from the Annals to the Change Classic to develop his notion of a precedent. His discussion affirmed epochal change and reform of political institutions. The sages, Liu maintained, had accented statecraft concerns in their bequeathed teachings. As precedents, these concerns, Liu felt, were principally articulated in the Annals and Change classics. Political change (pien ) was inherent in the transition from the Shang to Chou dynasties. Moreover, the Annals represented the third stage of that unity through the state of Lu, the harbinger for the Han unification.

As Lu's "uncrowned king," Confucius was the mouthpiece for this third stage. A "precedent" in Tu Yü's terms had implied a vision of stasis. The present was held up for comparison with the unchanging ideals of antiquity. In Liu Feng-lu's hands, this vision of antiquity as the repository for inert precedents was radically transformed. Instead, precedents became part of a vision of future change and reform. The past was held up to the present as a model of foresight and creative potential. The present became the locus for epochal change.^[79]

An additional theme that Liu explored in his reconstruction of Ho Hsiu's Kung-yang teachings was the relation between the Annals and

laws (fa ). Since the Han dynasties both Confucius and the Duke of Chou had at times been perceived as lawgivers. The New Text portrait of Confucius as judge in fact complemented the Old Text portrayal in the Rituals of Chou of the Duke of Chou and his model authoritarian government—a portrayal dating from the early Chou dynasty. It is generally recognized that the Rituals of Chou, because of its textual links to pre-Han Legalist strains in administrative theory and practice, emphasized the use of laws as a system of rewards and punishments to control wayward elements in society.

The apocrypha and New Text image of Confucius as the "creator of laws," however, seems out of place, given our usual understanding of the Confucian stress on morality and ritual. Although Mencius said that Confucius had served as police commissioner in his native Lu, we normally assume that Confucius regarded law as at best a necessary evil. This conventional perspective underestimates, however, the decisive part that Legalism played in the formation of imperial Confucianism during the Former Han dynasty. The Kung-yang Commentary in particular paid special attention to the presence in the Spring and Autumn Annals of a systematic terminological and stylistic framework for making legal judgments. In chapter 8 we shall examine how Liu Feng-lu's revival of New Text Confucianism brought with it recognition of the role of the Confucian Classics in legal discourse during the Han dynasties.^[80]