Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

To rebuild the empire: Lu Chih (754-805) and his response to the mid-Tang predicament Chiu-Drake, Josephine 1992

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1993_spring_phd_chiu-duke_josephine.pdf [ 18.74MB ]
JSON: 831-1.0086392.json
JSON-LD: 831-1.0086392-ld.json
RDF/XML (Pretty): 831-1.0086392-rdf.xml
RDF/JSON: 831-1.0086392-rdf.json
Turtle: 831-1.0086392-turtle.txt
N-Triples: 831-1.0086392-rdf-ntriples.txt
Original Record: 831-1.0086392-source.json
Full Text

Full Text

TO REBUILD THE EMPIRE: LU CHIH (754-805)ANDHIS RESPONSE TO THE MID-TANG PREDICAMENTbyJOSEPHINE CHIU-DUKEB.Ed., National Taiwan Normal University, 1973M.Ed., University of Vermont, 1977M.A., The University of Wisconsin, Madison, 1983A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIESDepartment of Asian StudiesWe accept this thesis a conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAOctober 1992© Josephine Chiu-Duke, 1992In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)Department of As t\a,h Sta4e!, The University of British ColumbiaVancouver, CanadaDate OCi l 3, i ipzDE-6 (2/88)Lu Chih^ iiABSTRACTTO REBUILD THE EMPIRE: LU CHIH (754-805)ANDHIS RESPONSE TO THE MID-T'ANG PREDICAMENTbyJosephine Chiu-DukeUniversity of British ColumbiaOctober 1992Supervisors: Dr. Jo-shui Chen & Dr. Daniel OvermyerThis study examines Lu Chih's efforts to rebuild the Tang empire toward the endof the eighth century, revises the previous views of Lu Chih as either a pure pragmatistor a conservative moralist, and establishes the significance of his political endeavors inthe context of the mid-Tang Confucian revival movement.After a thorough exploration of Lu's life and his family background, this workshows that two complementary principles underlay Lu Chih's approach to government:the principle of righteousness (1) and the principle of expediency (Ch'iian). Lu'sapplication of these principles, especially his interpretation of ch'ilan, is demonstratedby reference to his memorials to emperor Te-tsung and by his political practice. Myanalysis of Lu's application of these principles leads me to conclude that Lu's approachto government, both moralistic and pragmatic, may be characterized as a Confucianpragmatist approach. Relying upon this approach during the earlier stage of his officiallife as a Han-lin scholar, Lu Chih is seen to have been instrumental in the restoration ofdynastic stability.Lu Chih continued to employ his Confucian pragmatist approach in formulating anumber of policies during his tenure as Chief Minister. His earlier advice to the throneas Han-lin scholar is consistently reflected in these policies designed to realize hisvision of an ideal Confucian benevolent government. Although all of his proposedpolicies were intended to meet current needs, their ultimate goal is shown to be theimprovement of the public well-being. Lu's commitment to the public good was suchLu Chih^ iiithat he consciously risked his political life for the sake of his Confucian politicalconvictions.A comparative analysis of Lu Chih's political and social concerns and those of theleading figures of the mid-T'ang Confucian revival reveals many close affinities, andthus establishes Lu Chih's genuine place among them. While he failed to break newconceptual ground for the Confucian revival, his political life alone is seen as abehavior paradigm of the ideal Confucian minister for the mid-Tang Confucianrevivalists, and this is precisely Lu's unique contribution to that most significant mid-T'ang movement.Lu Chih^ ivTable of ContentsAbstract^ iiTable of Contents^ ivAcknowledgement vIntroduction^ 1Chapter One: Lu Chih's Life and Family Background^9Chapter Two: A Confucian Pragmatist Approach 102Chapter Three: Lone Pursuit of an Ideal^ 167Chapter Four: Significance in mid-Tang 262Epilogue:^ 353Bibliography 369Appendix: Glossary^ 392Lu Chih^ vAcknowledgementOne-half page allows only minimal space for expressing my sincere gratitude tothose whom I owe great intellectual debts. I would like to first thank my thesissupervisors: Professor Chen Jo-shui for his penetrating advice and insightful guidanceand Professor Daniel Overmyer for his thoughtful comments and helpful suggestions.I also wish to thank Professor Alexander Woodside for his positive support. Myfurther gratitude goes to Professor Lin Yii-sheng for stimulating my interest in Chineseintellectual history in 1980 and for his continuing encouragement; to Professor YilYing-shih who without knowing me nonetheless kindly answered my letter andsuggested various thesis topics including Lu Chih to me; and to Professor Mao Han-kuang for graciously supplying me much needed information about Lu Chih's familybackground. My friends J. J. Lo, Li Hsiao-ti, Osabe Yoshihiro, Stephen Eskildsen,my dear sisters Li-li and Wei-wei and brother-in-law Tse-hua all helped me to locatevaluable materials. My heart-felt thanks to every one of them. I must thank my latefather Chiu Shao-ying for his unique way of upbringing. I only wish he could havelived to see the completion of this study. Finally, my inexpressible gratitude goes toMichael for his loving care and joyful companionship.Lu Chih^ 1IntroductionLu Chih (754-805), better known by his canonical name of Lu Hsilan-kung, was asignificant figure in mid-Tang history. His service as a Han-lin scholar in the early stageof his political life represents another example of the encroachment of imperial personalattendants or advisors upon the power of Tang Chief Ministers. 1 As a Han-lin scholar, LuChih played an unusually crucial role at a time when the Tang state was in crisis. Hisresponses toward the social and political problems brought to him by emperor Te-tsung(reigned 779-805) made a contribution to the restoration of Tang stability. Lu's closerelationship with the throne and his service in this earlier period of his career led to hisrapid rise to the position of Chief Minister at a relatively young age. During his tenure asChief Minister, he attempted to carry out a series of policies, but his efforts in this stage ofhis career had limited impact. The breakdown of his close relationship with the throne ledto a precipitous fall from power and subsequently to his complete withdrawal from the mid-Tang political stage.Lu Chih's life and his entire political service took shape in the post-An Lu-shanrebellion era. The An Lu-shan rebellion (755-763) was a turning point in the history of theTang dynasty. It led a unified empire into a state of permanent division, with but a shortperiod of restoration, and also left the Tang territory rampant with chronic militarismthroughout the second half of the dynasty. 2 Faced with the catastrophic social and political1 As is well known, before the Han-lin scholars, some eunuchs had already seriouslyinterfered with the regular operation of the bureaucracy. Emperor Su-tsung's (reigned756-62) eunuch Li Fu-kuo is the best example. See Ssu-ma Kuang, Tzu-chih t'ung-chien (TCTC hereafter), Peking: Chung-hua shu-chii edition, 1956, 221: 7073-74 and222: 7115. For the development of eunuch power in the Tang, see Wang Shou-nan,T'ang-tai huan-kuan chiian-shih chih yen-chiu, Taipei: Cheng-chung shu-chii, 1971.2 For the An Lu-shan rebellion, see Pulleyblank, The background of the rebellion of AnLu-shan, London: Greenwood Press reprint, 1982. For post-rebellion regionalmilitarism, see C. A. Peterson, "Court and province in mid-and late T'ang," TheCambridge history of China, vol. 3 Sui and T'ang China, 589-906, Part 1 (TheCambridge history hereafter), Cambridge University Press, 1979: 485-560; also seeIntroduction^ 2changes brought about by the An Lu-shan rebellion, many concerned Tang scholarofficials felt compelled to fmd a way to restore their state to its previous order. Theirundertakings created a fertile soil from which a most important intellectual movement, themid-Tang Confucian revival movement, gradually grew into being.The Ho-pei region, where the An Lu-shan rebellion first started, became semi-independent after the Tang court settled the rebellion through a policy of compromise.When emperor Te-tsung attempted to exert central control over the Ho-pei region soon afterascending the throne in 779, the so-called second Ho-pei rebellion resulted and the Tangcourt was driven into another period of grave crisis. It is in the environment created by thegeneral historical conditions of the post-An Lu-shan rebellion era, and against the specifichistorical background of the second Ho-pei rebellion that we witness Lu Chih's emergenceon the mid-Tang political stage.Lu Chih is not a total stranger to Western students of T'ang dynasty history. As ispointed out elsewhere, some parts of Lu's extant works were translated into French in1735 and later into German early in the third decade of the twentieth century. 3 By 1960, asmall portion of Lu's criticism of the famous two-tax system was also translated forEnglish readers. Despite the fact that E. G. Pulleyblank dealt briefly with Lu's approach togovernment in an article on mid-Tang intellectual activities published that year, it was onlyin 1962 when Denis Twitchett published a long essay exclusively concerned with Lu Chihthat the first systematic study of Lu in English finally became available. 4 No furtherPulleyblank, "The An Lu-shan rebellion and the origins of chronic militarism in lateT'ang China," John Curtis Perry and Bardwell L. Smith, eds., Essays on Tangsociety, Leiden: E. J. Brill, 1976: 33-60.3 Professor Denis Twitchett cites J. B. Du Ha1de, Description de la Chine 1735, pp. 616ff; S. Balazs, "Beitrage zur Wirtschaftsgeschichte der T'ang-Zeit, Part 3," Mitteilungendes Seminars fur Orientalische Sprachen zu Berlin, XXXVI (1933): 1 -41 in his "LuChih (754-805) Imperial adviser and court official," (Lu Chih hereafter) in ArthurWright and Denis Twitchett, ed., Confucian personalities, Stanford University Press,1962: 84 and 336, notes 1 and 2.4 For the English translation of Lu Chih's criticism of the two-tax system, see WilliamTheodore de Bary et al., Sources of Chinese tradition, Columbia University Press,1960: 416-423. For Pulleyblank's essay, see his "Neo-Confucianism and Neo-Lu Chih^ 3research exclusively concerned with Lu Chih has been published in English during theintervening thirty years.Lu Chih's name is not entirely unfamiliar to the general population of educated modernChinese either. For example, after the Nationalist government took over from the Japaneseinvaders in 1945, trials were held in Shanghai of those who had served under the Japanesepuppet government in the occupied territories. At one such trial, Lu Chih's 783 suggestionto emperor Te-tsung -- that the throne should adopt a lenient policy toward the rebelliousgovernors and especially toward their subordinate soldiers -- was specifically cited by amilitary lawyer urging a judge to issue a less severe sentence to a chief collaborator whohad served as leader of the Japanese southeastern regional puppet government in 1938. 5Interest in Lu Chih has continued among Chinese scholars. A number of short articlesregarding his political views and his financial policies have been published since the late1950s in Taiwan and in China after 1980. Two book-length studies of Lu's life and workswere also published in 1975 and 1978 in Taipei. 6Legalism in Tang intellectual life, 755-805," in Arthur F. Wright, ed., The Confucianpersuasion, Stanford University Press, 1960: 93-95. Also see Twitchett, "Lu Chih,"1962: 84-122. Although this study will revise to some extent Twitchett's previousfindings on Lu Chih's life, it has benefited greatly from his pioneer essay on Lu Chih,and his contributions will be in evidence throughout.5 Thus reports a classified document (trial case number 013.11 — 2110) on file in theBureau of Historiography of the National Defense Ministry in Taipei. The name of thedefendant is Liang Hung-chih, and the lawyer's name is Chu Hung-ju. On March 28,1938, Japanese invaders made Liang Hung-chih their formal leader of the third puppetgovernment at Nanking with jurisdiction over Kiangsu, Chekiang, and Anhwei. SeeImmanuel C. Y. Hsii, The rise of modern China, reprinted edition, Taipei: Hung-ch'iaoshu-tien, 1978: 686. I am deeply grateful to my friend, Ms. J. J. Lo, for sending methis information. Ms. Lo came across this modern reference to Lu Chih while doingresearch for her Ph.D. thesis from Oxford University on the problem of the loyalty ofChinese intellectual during the Sino-Japanese War period. Due to the nature of thisinformation, no photo-copy machine was available for duplication. Ms. Lo copied thislawyer's defense brief verbatim and sent it to me in a personal communication.6 In addition to these short Chinese articles, I have also found one article in Japanesewhich deals with some limited aspects of Lu Chih's life. All these sources will beconsulted in the course of this study. The only lengthy studies of Lu Chih's life andworks are: Hsieh Wu -hsiung, Lu Hsiian -kung Chih yen - lun chi ch'i wen -hsiian, Taipei:Chia-hsin shui-ni kung-ssu, 1975, and Liu Chao-jen, "Lu Hsiian-kung yen-chiu,"Shih-chien hsiieh -pao, vol. 9, 1978: 97-125 and vol. 10, 1979: 1-42.Introduction^ 4While those short Chinese articles undoubtedly enhance our knowledge in relation tosome aspects of Lu Chih's approach to government, the limited scope of these essaysprohibits an overall investigation of their research subject. The two lengthy studies of LuChih are not completely satisfactory either. Although presenting much interestinginformation, they tend merely to juxtapose and categorize their research materials and donot undertake a critical analysis of Lu's responses to mid-Tang historical conditions.Moreover, they also fail to reconstruct an accurate account of Lu's earlier career and hisfamily background.By comparison, Twitchett's long 1962 essay on Lu Chih stands out as a major criticalstudy. It improves our understanding of Lu Chih's official life by providing an analysis ofhis responses to mid-Tang historical situations during his two important periods of serviceat the court. It further leads us to see why a close relationship between Lu Chih andemperor Te-tsung deteriorated the way it did. More importantly, it calls our attention to thefact that Lu Chih was not a conservative Confucian, a view expressed in Pulleyblank's1960 article, but a professional court official whose primary concern was to solve practicalproblems confronting the Tang state during the late eighth century. The importantcontribution of Twitchett's critical study of Lu Chih has been widely recognized, and hisessay was translated into Chinese and published in Taipei in 1973. 7There are still many unresolved problems, however. First of all, since Twitchett'sstudy of Lu Chih is also limited in scope, some aspects of Lu's life and his immediatefamily remain unclear to us. Secondly, because Twitchett's essay deals with less than halfof Lu Chih's responses to mid-T'ang historical conditions, the contributions Lu made tothe court during his service as a Han-lin scholar, and his efforts as Chief Minister to carry7 See Chung-yang yen-chiu-yiian Chung-mei jen-wen lee-hsiieh he-tso wei-yiian-hui, tr.and ed., Chung-kuo li-shih jen-wu lun-chi, Taipei: Chung-shan hsiieh-shu wen-huatung-shih-hui, 1973: 104-161.Lu Chih^ 5out a series of policies aimed at rebuilding the Tang empire have still not been given athorough examination.Thirdly, the breakdown of a relationship as complex as the one between Lu Chih andemperor Te-tsung would seem to have involved more factors than the three givenprominence by Twitchett: Lu Chih's outspoken personality, his change of role, and Te-tsung's autocratic rule.Lastly, the position Twitchett maintained thirty years ago that Lu Chih was primarily apragmatist whose orthodox Confucian beliefs were of merely secondary importance seemsto have run into a new challenge. This is because David McMullen's recently publishedbook, in which Lu is only a marginal figure, treats him as a conservative Confucianstatesman devoid of any appreciation of the need to employ expediency in government. 8McMullen's view obviously contradicts Twitchett's position, so much so that the exactnature of Lu Chih's approach to government seems persistently to elude our understanding.Furthermore, since the exact nature of Lu Chih's approach to government seemselusive, and since a detailed and thorough examination of Lu's responses to mid-Tanghistorical conditions is still not available, Lu's political convictions and his politicalcommitment remain unknown to us. As a result, Lu Chih's significance in the mid-Tangis totally undefined.Without satisfactory solutions to the above problems, Lu Chih's political endeavorswill remain a tantalizing puzzle to us, and our understanding of a segment of mid-Tanghistory will continue to be less than adequately clear. From this perspective, it appearsimperative to conduct a detailed and thorough study of Lu Chih's life and his responses tomid-Tang historical conditions. Such a study is important because it will not only provideanswers to previously unresolved problems, but also engage in a new undertaking to8 See David McMullen, State and scholars inT'ang China, Cambridge University Press,1988: 239.Introduction^ 6retrieve Lu Chih's political convictions and his political ideals, and above all, establish Lu'ssignificance in the mid-Tang.Thus, the following related questions are the main focus of this present study. First,what exactly were Lu Chih's responses to mid-Tang historical conditions, and how did heanalyze the needs of his time? In other words, what was Lu's approach to government,and what were the fundamental characteristics of such an approach? Was Lu's approachpragmatic, or was it moralistic, or was it perhaps both moralistic and pragmatic? Second,what basic political convictions and ideals are disclosed in Lu's approach to government;were they related to his political downfall? What are the distinct features of his politicalconvictions and ideals? Do these features qualify him to be considered a conservativeConfucian, or a practical statesman, or both, or neither? Third and last, on what basis canLu Chih's significance in the mid-Tang be established?To provide appropriate answers to these questions, we shall analyze every one of LuChih's extant memorials, both those presented to the throne during his service as a Han-linscholar in the earlier stage of his official life and those presented as Chief Minister in thelater stage. His extant memorials will serve throughout as our fundamental sources.The first chapter clarifies some confusions regarding Lu Chih's earlier career, someaspects of his personality and his immediate family. A clear understanding of Lu's life andfamily background will enable us to examine in a better light his responses to the social andpolitical problems confronting emperor Te-tsung's court.The second chapter attempts both to grasp the exact nature of Lu Chih's approach togovernment, and to delineate how Lu relied upon his approach to assist emperor Te-tsungto resolve the social and political crisis that occurred during his tenure as a Han-lin scholar.The suggestions which Lu proposed to the throne in this period will be the center of ourexamination.Twitchett's position that Lu Chih was primarily a pragmatist and McMullen's view thatLu was a conservative Confucian who refused to apply expediency in government leave usLu Chih^ 7with two paradigms of polarization. In order to provide a more appropriate picture of LuChih and his approach to government, however, it has seemed more fruitful in our re-examination of Lu's approach to transcend any polarization paradigm, and rely upon LuChih's own expressed views to lead us to their most logical conclusions.The third chapter highlights Lu Chih's political convictions and his political ideals. Byinvestigating the policies Lu formulated during his tenure as Chief Minister, we attempt tofind out whether or not a consistency existed between Lu's approach to government in theearlier stage of his official life and that of the later stage. If Lu continued to apply the sameapproach to government throughout his tenure as Chief Minister from 792 to 794, this willmake it possible for us to demonstrate his political convictions and ideals. Once wedemonstrate how Lu Chih's core convictions and political ideals are embodied in hispolitical behavior, this will allow us to see whether he was merely a court official whoseconcerns revolved simply around doing his professional job and preserving his politicalpower, or whether he represented something different in the sense that he aspired to gobeyond his prescribed duty and fulfill some higher goal. In this manner we will be able toproduce a more accurate portrait of Lu Chih.In the meantime, since the political convictions and ideals upon which he based hisactivities during his service as Chief Minister consistently reflect his earlier position, thisfact will offer some clues to explain why in the higher position of Chief Minister Lu failedto achieve the success he accomplished during his earlier service as a Han-lin scholar. Itwill also shed more light on the final breakdown of his close relationship with the throne.Having presented a more complete portrait of Lu Chih, our last major chapter exploresLu's significance in the mid-T'ang. To fulfill this task, Lu Chih's approach to governmentwill be compared with that of several notable court officials before and during his time.This will help to demonstrate by exact examples whether or not Lu Chih displayed someunique qualities as a court official.Introduction^ 8Moreover, since the most important intellectual movement during Lu's time was themid-Tang Confucian revival movement, we shall also conduct a comparison between Lu'sapproach to government and that of the leading figures of the mid-Tang Confucian revival.Finding the similarities and differences between Lu's political convictions and ideals andthose of the leading Confucian revivalists will not only help to identify the intellectual campLu belongs in, but more importantly, to discover whether or not we can claim a place forhim in the most important intellectual movement of his time. The results of this inquiry willultimately reveal Lu Chih's significance in the mid-Tang.In order to arrive step by step at the final point where Lu Chih's significance graduallyunfolds, we have to start from the beginning, that is, with a thorough examination of hislife and family background.Lu Chih^ 9nChapter One: Lu Chih's Life and Family BackgroundLu Chih was born in the thirteenth year of the Tien-pao era, the second reign period ofemperor Hsiian-tsung (754 A.D.). This was a time when the glory of the T'ang empirehad begun to decline despite the existence of a seemingly stable and prosperous facade.One year after his birth the calamitous An Lu-shan rebellion broke out; it destroyed theunity of the empire and consigned this golden age of the Tang to the status of a bygonememory.Although the Tang court managed to suppress the An Lu-shan rebellion in 763, it wonits "victory" only by a compromise settlement. Forced to adopt a compromise policy, theTang court not only pardoned the ex-rebel generals, but also allowed them to be themilitary governors of Ho-pei where the rebellion initially started. In so doing, the courtsowed the seeds of a second Ho-pei rebellion in the early 780s when Lu Chih began hisservice in the central government. 1Lu Chih's private life is largely an enigma. This is due to an insufficiency of existingtraditional sources. Nevertheless, a close examination of his collected work -- despite theirincompleteness and possible bias due to the personal preferences of their chief compiler --will help us reach a new understanding of some aspects of his personality while at the sametime elucidating his political life. 21 On the compromise settlement see C. A. Peterson, "Court and province in mid-and lateT'ang," in The Cambridge history, 1979, p. 484.2 A note on sources:On the matter of the insufficiency of source materials, see Denis Twitchett, "Lu Chih,"1962: 84-87. Although Twitchett does not consider Lu Chih's biography in the NewTang History useful, as I shall demonstrate later, it is the New Tang History, not theother extant sources, which actually provides a very important clue for us to reconstruct LuChih's political life. As mentioned already, there were studies which had discussed andtranslated some parts of Lu Chih's extant works into English, French and German beforeTwitchett's 1962 study. Meanwhile, since 1962, quite a few short Chinese articles aboutLu Chih's economic and political thought and two lengthy studies of Lu's life and worksalso became available. Moreover, a chronicle study of Lu Chih's life (nien-p'u), was alsoLife and Family Background^10published in Taipei by Yen I-p'ing in 1975. See Yen I-p'ing, Lu Hsiian-kung nien-p'u,Taipei: I-wen, 1975. In addition, as Twitchett mentions, two useful chronicle studies ofLu Chih's life already existed before Yen I-p'ing's book. One is the "Lu Hsilan-kungnien-p'u" composed by the nineteenth-century scholar Ting Yen, it is included in the 1768Han-yilan chi chu (Collected works of Lu Chih); the other is the "Nien-p'u chi-liieh"appended to the Nien Keng-yao edition of Lu Chih's works reprinted in the Ssu-pu pei-yao(SPPY hereafter) edition. See Twitchett, "Lu Chih," 1962, p. 337 note 9. Another usefulchronicle study was also composed by a nineteenth-century scholar Yang Hsi-min and isentitled Lu Hsiian-kung nien-p'u. It is in the Shih-wu chia nien-p'u ts'ung-shu collection.Compared with the other two Ch'ing dynasty nien-p'u, Yang Hsi-min's study providesmore detailed information on Lu Chih's life. However, all these studies, including Yen I-p'ing's most recent nien-p'u, contain misleading information about Lu Chih's immediatefamily and his career development and should be used with caution.Twitchett gives a detailed description of the various extant editions of the collectedworks of Lu Chih. However, as he says, there is a discrepancy concerning the totalnumber of chapters contained in Lu Chih's extant works. Ch'Uan Te-yii, Lu Chih'scontemporary and probably also a friend, in his Preface (see Ch'ilan T'ang-wen, CTWhereafter, Peking: Chung-hua shu-chii reprint, 1982: 493: 11-15) to the Han-yiian chi(Collected works of Lu Chih), describes Lu Chih's collected works as having twenty-fourchapters altogether. He writes that they contained ten chapters of edicts, seven of privatememorials (tsou-ts'ao), and seven of official memorials (chung-shu tsou-i). The NewTang History, I-wen chih section (Ou-yang Hsiu, et al., ed. Hsin T'ang-shu, HTShereafter, Peking: Chung-hua shu-chii edition, 1975, 60: 1616) tells us that there were tenchapters in the Han-yiian chi, and also twelve chapters in the Lu Chih lun-i piao-shu chi.Obviously, these two volumes together would make twenty-two chapters. The Sungdynasty Ch'ung-wen tsung-mu (compiled between 1034 and 1038) (Kuo-hsiieh chi-pents'ung-shu edition, ch. 5, pp. 377-78) says there were two chapters in the Lu Chih chih-chiand ten chapters in the Han-yilan chi. The much later Chih-chai shu-lu chieh-t'i (probablycompiled in the mid-thirteenth century; Kuo-hsiieh chi-pen ts'ung-shu edition, ch. 22, p.601) says that there were twenty chapters in the Lu Hsiian-kung tsou-i, which is alsoentitled Pang-tzu chi; but on another page (ch. 16, p. 448) it says that there were tenchapters in the Han-yiian chi and twelve chapters in the Pang-tzu chi; it combines these twotogether under the title of Lu Hsiian-kung chi. At the same time, the Chiln-chai tu-shu chih(1151; Kuo-hsiieh chi-pen ts'ung-shu edition, ch. 4A: 363-64) says that there were twelvechapters in the Lu Chih tsou-i. It then mentions that previously there were five chapters inLu Chih's Pang-tzu chi, three chapters in the 1-lun chi, and ten chapters in the Han-yiianchi. The author of the Tu-shu chih suspects that all these previous works were put togetheraround 1090; the title of the works then became Lu Chih tsou-i.Following the above information, we must agree with Twitchett that the records of LuChih's works have become a "bibliographical muddle." However, the Ssu-k'u ch'iian-shutsung-mu t'i-yao (chi-pu, pieh-chi lei, 3) tells us that the information in the Chiin-chai tu-shu chih does not match most of the historical bibliographical descriptions. The T'i-yaosays that according to other Sung bibliographical records ever since the Southern Sung allof Lu Chih's works had been put together and placed under the general title of Han-yaanchi, a book containing twenty-two chapters. The T'i-yao further informs us that the authorof Chiin-chai tu-shu chih must have seen an incomplete edition of Lu Chih's works andthus only listed twelve chapters in Lu Chih tsou-i. Because the Sung scholars also refer toLu Chih's works as the Lu Hsilan-kung tsou-i, the T'i-yao says, many Ch'ing editions ofLu Chih's works follow this title. Based upon the explanation given by the Ti-yao, itseems clear to us that at least by Southern Sung times there were basically only twenty-twochapters in Lu Chih's collected works -- the Han-yiian chi. It is obvious that this Han-yiian chi contained ten chapters of the pre-Southern Sung Han-yi,ian chi and twelvechapters of Lu Chih's other memorials or edicts with a title unclear to us. This explainsLu Chih^ 11why almost all the modern editions of Lu Chih's works have twenty-two chapters (ten ofedicts, six of private memorials and six of official memorials). The Ssu-pu ts'ung-kan(SPTK hereafter) and SPPY editions apparently also follow the Southern Sungarrangement into twenty-two chapters.Although there are two twenty-four-chapter editions in existence now -- one a late Mingedition by Tang P'in-yin and Ma Yuan, the other a 1768 edition with extensivecommentary by Chang P'ei-fang -- as Twitchett points out, the contents of these twoeditions are identical with the twenty-two chapter Sung edition. When we compare theseMing and Ch'ing editions with the twenty two chapter editions, we discover that theireditors completely regrouped Lu Chih's writings into chronological order; there are no newdiscoveries. Twitchett's comment made almost thirty years ago that the best availableedition of Lu Chih's works at that time was the 1768 edition by Chang P'ei-fang stillapplies to the current situation. This is because, like the SPTK and the SPPY editions, this1768 edition was also compiled on the basis of a Southern Sung edition. However, unlikethe SPTK and the SPPY editions which do not give us the exact date of the Southern Sungedition used by them, the 1768 edition used an edition compiled by a Southern Sungofficial named Lang Yeh (not Lang Hua as the editor of the 1768 edition wrote andTwitchett followed in his article). Lang Yeh presented his edition of Lu Chih's works tothe throne in 1132 (1132 edition hereafter). This 1132 edition, entitled Chu Lu Hsiian-kung tsou-i, includes fifteen chapters of Lu Chih's memorials and Lang Yeh's owncommentary. It was reprinted in the Shih-wan-chiian lou ts'ung-shu in 1878 and isreprinted again in the Pai-pu ts'ung-shu chi-ch'eng collection. Because the 1768 editionprovides the date of the Southern Sung edition (the 1132 edition) which it followed, andmoreover, because the 1768 edition preserves not only Sung scholars' comments (thoughnot Lang Yeh's commentary) about Lu Chih's works, but also contains its editor's owncommentaries, this edition seems to be preferable. This is why, in this study, I haveprincipally relied upon this 1768 edition. However, I have also compared the 1768 editionwith the SPTK and the SPPY editions in case differences appear in the text. In addition,for the sake of comparison, I have also consulted the 1132 edition. The 1768 edition, hardto acquire formerly, is readily available now. In 1982 Taipei's Shih-chie shu-chUpublished a reprint of this 1768 edition under the title of Han-yaan chi chu (HYCChereafter) edited by Yang Chia-lo as the sixth volume of the Chung-kuo wen-hsiieh ming-chu collection.Lu Chih's work on medicine, Lu-shih chi-yen-fang, his encyclopedic work Pei-chiiwen-ye , and his fifteen chapters of Pieh-chi on literary works were all lost after the Sungdynasty. See Twitchett, "Lu Chih," 1962: 86 and 337 note 14. As Twitchett mentions,HTS 1-wen chih classes Pei-chii wen-yen with Tu Yu's T'ung-tien and other Hui-yao. Italso says that it contained twenty chapters; see HTS, 59: 1563. The Sung dynasty Ch'ung-wen tsung-mu (Kuo-hsfieh chi-pen ts'ung-shu edition, ch. 3: 178) and the Chiin-chai tu-shu chill (Kuo-hstieh chi-pen ts'ung-shu edition, Hou-chih section, ch. 2: 852) agree withthis information. The Chiin-chai tu-shu chih further mentions that it had more than 450sections and was similar to the Po-shih liu-t'ie (compiled by Po Chii-i), but with moreliterary polish. The Wen-hsien t'ung-k'ao (228: 1828) quotes this passage from Chiin-chaitu-shu chih. However, both Sung shih 1-wen chih and Yii-hai say that it contained thirtychapters instead of twenty chapters. See Sung shih, Peking: Chung-hua shu-chU edition,207: 5293; Wang Ying-lin, ed., Yii-hai, Taipei: Hua-wen shu-chU, 1967 reprint, ch. 201:22. Yii-hai informs us that Pei-chii wen-yen put passages of a similar nature from theclassics and historical works into different categories; altogether it contained 452 sections.Except for the difference in numbers of chapters, all the above sources agree that Pei-chiiwen-yen is a sort of encyclopedic work. It was probably an administrative encyclopedia,as Twitchett points out. For Lu-shih chi-yen-fang, see HTS, 59: 1572; Ch'Uan Te-yU'sPreface, in CTW, 493: 15; Shun-tsung shih-lu (SL hereafter), Ts'ung-shu chi-ch'engedition, Commercial Press, 1936, 4: 16; Wang Ch'in-jo,Ts'e-fu yiian-kuei (hereafterLife and Family Background^12Lu Chih's memorials to emperor Te-tsung (reign 779-805) during the two mostimportant stages of his career clearly reveal his "conscious responses to the situations inTFYK), Peking: Chung-hua shu-chii reprint, 1960: 859 19. HTS says this work onmedicine contained fifteen chapters, but all the sources agree that there were fifty chaptersin Lu-shih chi-yen-fang. It is very likely that fifty chapters is the correct number. For LuChih's Pieh-chi on literary work, see Ch'ilan Te-yii's Preface, CTW, 493: 15.According to the HTS 1-wen chih, the chief compiler of Lu Chih's works, the Han-yiian chi, was Wei Ch'u-hou (773-828). Twitchett suggests that since Wei Ch'u-hou wasvery likely a maternal relative of Lu Chih, his compilation of Lu Chih's collected worksmight have been biased though he does not specify biased in what manner. Twitchett'ssuggestion no doubt deserves our attention, but we should also realize that it was quitecommon for a relative or a friend to do such a compilation in Tang China. It does notfollow that their compilation should thus be considered biased. We do not know exactlywhen Wei compiled the Han-yiian chi, but it may very well have been after 806 and before818. Wei Ch'u-hou was appointed to compile the Te-tsung shih-lu around 808 when P'eiChi was Chief Minister. Since he had access to all sorts of historical documents, he couldhave compiled the Han-yilan chi without much difficulty. During this time he also workedon the Shun-tsung shih-lu until he was demoted to another position in 812. Han Yii (768-824), one of the most famous prose writers of the Tang dynasty, took over thecompilation in 813. Han Yu completed his version of the Shun-tsung shih-lu in 815. E.G. Pulleyblank has suggested that Han Yii's version of the Shun-tsung shih-lu very likelybecame the text used by Chliian Te-yii to write his preface to Lu Chih's collected works,sometime before Ch'ilan died in 818. Pulleyblank has further suggested that the presenttext of the Shun-tsung shih-lu is Wei Ch'u-hou's and not Han Yii's version of the work.Whenever material in the present text of the Shun-tsung shih-lu about Lu Chili differs fromTe-yii's preface, we often find Ch'ilan Te-yii's preface agrees with the materialcontained in the Old T'ang History. According to Pulleyblank this is because the OldT'ang History probably also used Han Yii's version of the Shun-tsung shih-lu as the basictext for Lu Chih's biography. In the early 1980s, Chinese scholars began to debate theauthorship of the present Shun-tsung shih-lu. Some scholars' arguments supportPulleyblank's point of view, that is, Wei Ch'u-hou should be the compiler of the extantShun-tsung shih-lu. Scholars holding a different point of view, however, argue that it wasnot possible for Wei Ch'u-hou's version of the Shun-tsung shih-lu to be circulated once itwas replaced by Han Yil's text of the Shun-tsung shih-lu. Although I find the lattergroup's argument less convincing than that of their opponents, the controversy obviouslyremains unsettled. Nevertheless, precisely because it is unsettled, the material about LuChih contained in the extant Shun-tsung shih-lu, Chitin Te-yii's Preface, and the OldT'ang History thus becomes a more useful cross-reference for any re-evaluation of LuChih's life than we previously thought . On Wei Ch'u-hou's possible connection with LuChih, see Twitchett, "Lu Chih," 1962: 85 and 337 note 10 and 11; on the authors ofdifferent versions of the Shun-tsung shih-lu and the relationship between the Old T'angHistory and Ch'iian Te-yii's preface, see E. G. Pulleyblank, "The Shun-tsung shih-lu,"Bulletin of the School of Oriental and African Studies (hereafter BSOAS), 19:2 (1957):336-44, esp. 340; also see Chang Kuo-kuang, "Chin-pen <Shun-tsung shih-lu> fei HanYu so tso pien," Wen-hsaeh p'ing-lun ts'ung-k'an, vol. 7 (10/1980): 328-340. Forscholars holding opposite opinions, see Chiang Fan, "Chin-pen <Shun-tsung shih-lu> tso-che k'ao-pien," Wen-hsiieh ping-lun ts'ung-k'an, vol. 16 (10/1982): 321-336; Ch'ii Lin-tung, "Kuan-yii <Shun-tsung shih-lu> to chi-ke wen-t'i," Pei-ching shih-fan ta-hsiiehhsiieh-pao, 1 (1982): 45-53.Lu Chih^ 13which he finds himself."3 From them we can detect not only his ideas about how to solvethe problems currently confronting the court of emperor Te-tsung, but also his "attitudes"and "propensities of feelings" toward particular events and circumstances.4 Understandingthese personality traits, Lu Chih will no longer appear merely as a "depersonalized politicalfigure,"5 but rather more fully as a political figure with some vivid inner landscapes.Since many early aspects of Lu Chih's official career have not been dealt withaccurately,6 nor has the information about Lu Chih's immediate family ancestors beentouched upon at all, it is also imperative to establish a more valid account of Lu Chih'spolitical life and his family background. The examination of family background is essentialnot only because it generally shapes a person's character, but also because it played aparticularly crucial role in determining one's social standing and upward social mobilityduring the T'ang. In order to determine the position of Lu Chih's family in the T'angsocial hierarchy, we begin with a discussion of the Lu lineage.Lu lineage in the Tang social hierarchyIt is well understood that early and middle Tang society, like its predecessors, the Sixdynasties and the Sui, was marked by clearly observed social segregation and stratification.An enormous gap certainly existed between commoners (shu-min) and the elite class (shihta-fu),7 but within the elite class there were different categorizations to define the place of a3 Benjamin Schwartz, "The Intellectual history of China: preliminary reflections," John K.Fairbank ed., Chinese thought and institutions, University of Chicago Press, 1957: 16.4 Ibid: 17.5 Twitchett, "Lu Chih," 1962: 86.6 As we shall demonstrate, none of the extant studies of Lu Chih, probably due to thenature and scope of their researches, provide correct information about Lu Chih's earlycareer.7 Denis Twitchett, "The composition of the Tang ruling class: New evidence fromTunhuang (The Tang ruling class hereafter)," Arthur F. Wright and Denis Twitchetteds., Perspectives on the Tang (Perspectives hereafter), Yale University Press, 1973:49.Life and Family Background^14lineage or a clang in that social hierarchy. 9 Generally speaking, most members of theTang elite class belonged to the category of illustrious provincial lineages, but at the top ofthe elite class, a very small number of lineages constituted what some modern historianscall the " aristocratic families." What made these families aristocratic was their "hereditaryhigh social status, independent of full court control." 10 They were a super elite withnational recognition. 11 Most of these aristocratic lineages began to emerge from the earlyfourth century on when north China was taken over by non-Chinese rulers.Education and culture, economic wealth, local power based upon accumulated landedproperty and clan solidarity, the practice of marriage exclusivity, and ingrained socialrespect for birth all amounted to valuable assets for the rise of these aristocratic lineages.Nevertheless, ownership of local property seems to have constituted the chief power basefor their survival during dynastic changes. 12 They enjoyed high esteem and great influence8 I am fully aware of the difference between a "clan" and a "lineage" defined byanthropologists such as Maurice Freedman and Hugh Baker. However, scholars of theTang dynasty have pointed out that most prominent medieval lineages were so looselyknit that "clan" and "lineage" are thus used interchangeably for the sake of convenience.In this study, I shall refer to Lu Chih's kinship group and subgroups as "Lu lineage" or"Lu clan" despite the fact that we can trace the common ancestor of the major Lusubgroups. Hugh D. R. Baker, Chinese family and kinship, Columbia UniversityPress, 1979: 49 and 68. Maurice Freedman, ed., Family and kinship in ChineseSociety, Stanford University Press, 1970: 13-14; Patricia Buckley Ebrey, Thearistocratic families of early China: A case study of the Po -ling Ts'ui family,Cambridge University Press, 1978: 22; David Johnson, "The last years of a great clan:The Li family of Chao Chiin in late T'ang and early Sung," Harvard Journal of AsiaticStudies (HJAS hereafter), 37 / 1 (June 1977) : 98; Chen Jo-shui, "The dawn of Neo-Confucianism: Liu Tsung-yiian and the intellectual changes in Tang China," PhDdissertation Yale University, 1987: 35, and note 5. I am grateful to Professor Chen Jo-shui for giving me a copy of his thesis which is now forthcoming from CambridgeUniversity Press.9 Twitchett, "The T'ang ruling class," Perspectives, 1973: 50.10 In her perceptive study of the aristocratic families of early imperial China, PatriciaEbrey has shown the distinctive characteristics of these aristocratic families in earlyimperial China and carefully provided the above quoted definition. See Ebrey, Thearistocratic families, 1978: 2 and 10.11 Twitchett, "The Tang ruling class," 1973: 47-8.12 Ebrey has demonstrated that there were complex factors at work in the determination ofstatus; throughout time, different factors weighed differently as various dynastieschanged. However, she has particularly emphasized that the maintenance of a localgeographical base was crucial to the survival of these aristocratic families. See Ebrey,The aristocratic families, 1978: 117 and 28-32. On the other hand, some scholarsLu Chih^ 15in society. Their enormous social prestige, moreover, almost always guaranteed them ahigh rank in the nine-tiered system of recruitment and thus allowed members of these greatfamilies to dominate for generations the top positions in government during the Sixdynasties. 13 Their participation in government in turn perpetuated their overall socio-economic power. Consequently, these prominent lineages survived into the Tang despitethe rise and fall of dynasties.These aristocratic families were referred to by various terms during the Tang: chiu -tsu(old clans), shih-tsu (scholar official clans), chu-hsing (famous names), kuei-tsu (nobleclans), or ming - tsu (illustrious clans). 14 Liu Fang, a mid-eighth century historian andgenealogist once wrote a list in which he singled out four regionally based groups ofaristocratic lineages as the most eminent families in Tang China. 15 Each aristocraticbelieve that education and culture were the most important factors in the earliest stage ofthe formation of powerful families at the close of the Western Han dynasty (202 B. C.-A. D. 9) and in the early period of the Eastern Han (25-220 A. D.). Once havingmonopolized education and culture, members of these powerful families naturallycontrolled access to officialdom and economic wealth. In the process, it becomesdifficult to distinguish which factor is the cause and which the effect. As a result, thesefactors mutually influenced each other and constituted a social cycle. See Yii Ying-shih, "Tung-Han cheng-ch'iian chih chien-li yii shih-tsu ta-hsing chih kuan-hsi," in hisChung -kuo chih -shih chieh-ts'eng shih lun (ku -tai 'nen), Taipei: Lien -ching, 1980:113-5; Etienne Balazs, "Significant aspects of Chinese society," Arthur F. Wright, ed.,Chinese civilization and bureaucracy, Yale University Press, 1964: 6-7. For socialeconomic and political factors, see Yang Lien-sheng, "Tung Han to hao-tsu," Ch'ing-hua hsaeh -pao, 11: 4 (1936): 1007-63, Wang I-t'ung, Wu -ch'ao men - ti, 2 vols.,Taipei reprint, 1973, vol. 1, chs. 3 and 5; Mao Han-kuang, Liang -Chin Nan-Pei-Ch'aoshih - tsu cheng -chih chih yen -chiu, 2 vols., Taipei, 1966, vol. 1, ch. 3, esp.: 63-4.,and ch. 7: 230-248; Twitchett, "The T'ang ruling class," 1973: 49.13 Mao Han-kuang, Ibid., vol. 1, ch. 4; Wang I-t'ung, Ibid., vol., 1, ch. 3; Ebrey, Thearistocratic families., 6; Miyazaki Ichisada, Kyahin kanjinhO no kenkya, Kyoto, 1956,ch. 3, esp.: 247.14 Chen Jo-shui, "The dawn of Neo-Confucianism," 1987: 3; Ch'en Yin-k'o, T'ang -taicheng-chih -shih shu -lun kao (Shu -lun kao hereafter), Chung-yang yen-chiu-yiian li-shih yii-yen yen-chiu-so chuan-kan, Taipei reprint, n. d., pp. 56-59, also reprinted inCh'en Yin -k'o hsien -sheng ch'aan -chi, Taipei: Chiu-ssu ch'u-pan, vol. 1, 1977: 225-228; Ebrey, The aristocratic families 1978: 3.15 See HTS, 199: 5676-80; CTW: 372: 7a-1 lb. Liu Fang's classification of aristocraticlineages is an important evidence of what was believed in his own days and has beenwidely cited by scholars of Tang history. See Twitchett, "The Tang ruling class,"1973: 50-1; Ebrey, The aristocratic families., 1978: 10-11; Chen Jo-shui, "The dawn ofNeo-Confucianism," 1987: 4-8.Life and Family Background^16lineage in these four groups could trace its family roots back to the days of the Sixdynasties, but the degree of their power and privilege varied considerably.Geographically speaking, two of these four regional groups were in the north, theShan-tung (east of the mountains, modern Hopei, Honan and central Shansi) aristocraticfamilies, the most prestigious group of lineages among all the four elite groups, and the socalled "Kuan-lung block" to which the T'ang ruling house belonged. The "Kuan-lunggroup" was formed by an alliance of the Kuan-chung (modern southern Shansi and modernShensi) and the Tai-pei (modern northern Shansi) aristocratic families. The other twogroups held local power in southeast China and were defined as the southern aristocracy.While one of these two southern aristocratic groups consisted of emigre clans, the otherconstituted the native aristocracy. Lu Chih's family belonged to the Wu chiin (Wucommandery) Lu lineage, one of the the four most prestigious native aristocraticrepresentatives in southeast China. 16These four aristocratic groups were the cream of T'ang society, according to LiuFang's thesis. They not only commanded the highest respect from society, but also feltthemselves deserving of such respect. However, in the T'ang era, comparatively speaking,due to the suffering and losses inflicted on them by repeated military uprisings in the sixthcentury, the power and prestige of the southern aristocracy could not rival that of thenorthern aristocratic groups. In addition, the fact that Ch'ang-an was now the center ofcultural and political activities naturally put the southern aristocratic families in an inferiorposition to compete with the northern aristocracy. Nevertheless, according to Liu Fang,this did not alter their superior social position in the Tang elite class as a whole. In short,Lu Chih's lineage in the southeastern region, though occupying a less prominent positionthan the northern aristocratic families, actually ranked in the highest level of Tang socialstructure.16 For details of these four groups, see references given in the above note.Lu Chih^ 17Although the presence of powerful aristocratic clans was one of the main characteristicsof T'ang social structure, ironically, it was precisely during this era that thebureaucratization of the aristocracy began to take place through the famous examinationsystem (K'o-chii chih-tu). Since it did not bring about genuine "inter-class" social mobilityin T'ang China, the impact of the examination system should not be exaggerated. Theprocess of bureaucratization was not completely carried out until near the very end of thedynasty. In the process of being bureaucratized, however, members of the aristocraticclans always had a better chance to enter and to advance in the Tang bureaucracy. 17 Inother words, no matter how the Tang aristocracy was transformed, before the T'angdynasty ended a person's standing in social and political life was always closely related tohis family background. Bearing this in mind, we shall first highlight the general history ofthe Lu lineage and then focus on Lu Chih's family background.A profile of the Lu lineageAs mentioned above, the Lu lineage of Wu chiin was regarded as a member of thesuper elite class by the mid-T'ang genealogist Liu Fang. Wu than included the southeastand the northeast areas of modern Kiangsu and Chekiang provinces. Situated at the lowerregion of the Yangtze River, it constituted one of the most important economic areas of theTang state, especially after the An Lu-shan rebellion. During the Tang, Wu chiin wasalso known as Su-chou . Though the T'ang court decided to use Su-chou to replace the old17 To understand how the examination system worked in the T'ang and why its impact onTang society should not be overrated, see Denis Twitchett, The birth of the Chinesemeritocracy: bureaucrats and examinations in T'ang China, A lecture delivered to theChina Society in London on December 17, 1974 and printed by Bendles (Torquay)Ltd.. For the transformation of the T'ang aristocracy, see Denis Twitchett, ed., TheCambridge history, 1979: 21-22 , and Ebrey', The aristocratic families, 1978, esp. herconcluding remarks: 118-9.Life and Family Background^18Han dynasty name of Wu chiin in 758, 18 both names were used interchangeablythroughout the later half of the Tang. Altogether there were seven counties included in Su-chou or Wu chiin under Tang rule. 19Both the New T'ang History genealogy (HTS tsai-hsiang shih-hsi piao) and a prefacewritten in 812 to the re-compiled genealogy of the forty-nine branches by a descendant ofthe twenty-third generation of a major subgroup, one Lu Shu, 20 agree that the person whoinitiated the history of the Lu lineage in Wu county (Wu hsien) of Wu chiin was a certainLu Lie.21 Although we do not know when this actually took place, one thing seems to becertain: before the Eastern Han (25-220) dynasty emerged, the Lu lineage was alreadyfirmly established in Wu hsien of Wu chiin. The History of the Eastern Han (Hou Hanshu) unmistakably refers the Lus of Wu chiin as "a prominent surname for generations"(shih wei tsu-hsing). 22Early in 634, the Lu family already appeared in an imperially approved list of "notableclans" (wang-shih).23 Even in the late tenth century it was still recognized as one of the18 Fan Ch'eng-ta, Wu Chan-chih, 6 vols.,Ts'ung-shu chi-ch'eng ch'u-pien edition,Shanghai: Commercial Press reprint, 1939, vol. 1, ch. 1: 2.19 Li Chi-fu, Yiian-ho chiin-hsien t'u-chih (YHCHTC hereafter), ten vols., Kuo-hsiiehchi-pen ts'ung-shu edition, Taipei: Commercial Press reprint, 1968, vol. 6, ch. 25: 66020 This preface to the recompilation of the genealogy of forty-nine Lu branches (812Preface hereafter) is preserved in a Ch'ing dynasty Lu clan genealogy. See Lu-shihshih-p'u, 24 vols., compiled by Lu I and Lu Sheng-wu in 1745 (Ch'ien-lung shan-te-t'ang edition), Columbia University Rare Books Collection, microfilm # 0876. Theauthor of this preface, Lu Shu, was the Surveillance Commissioner (Kuan-ch'a shih)of the Fukien region in 812. This agrees with the information about him contained inHTS. See HTS, 73: 2972. This 812 preface is also preserved in another Ch'ingdynasty Lu clan genealogy, see Lu-shih tsung-p'u, compiled by Lu Chen-chih, 4 vols.,Columbia University Rare Books Collection, microfilm # 548; also see Twitchett, "LuChih," 1962: 87 and 338 note 22. I rely upon Hucker's translations (where available)of all official titles in this study. See Charles 0. Hucker, A dictionary of official titlesin imperial China, Stanford University Press, 1985.21 HTS, 73: 2965 and 812 Preface.22 Hou-Han shu, Peking: Chung-hua shu-chii edition, 81: 2682.23 This list was discovered at Tunhuang and later reprinted by Hsii Kuo-lin in hisTunhuang shih-shih hsieh-ching t7-chi yii Tunhuang tsa-lu, 2 vols., 1937, vol. 2:154b; also see Twitchett, "Lu Chih," 1962: 87 & Ikeda On, "TOdai no gumbo hy6",TOO gakuhe), 42: 3 (1959): 61.Lu Chih^ 19most illustrious "four surnames" in the southeast region. 24 Apparently, the Lu family musthave developed into a very large kinship organization throughout the centuries before theT'ang.According to the New T'ang History genealogy, up to the Eastern Chin dynasty (317-420) there existed at least eight branch groups within the Lu clan. 25 By the mid-T'ang, theclan became so large that it actually expanded to forty-nine branches which wereacknowledged by the imperial ruling house. Lu Shu's 812 Preface stated plainly that eachsubgroup of the Lu clan should establish its separate genealogy, otherwise the lineagehistory might be lost due to demographic expansion. 26Despite this proliferation of Lu subgroups, only three branches were actually mentionedin the New T'ang History, and that was because they had produced six chief ministersduring the T'ang. 27 These three subgroups included the Tan-t'u branch, the T'ai-wei (theDefender in Chief) branch and the Shih-lang (Vice Director) branch which Lu Chih'simmediate family belonged to. 28 While Tan-t'u was the name of a county (modern Chen-hang in Kiangsu), the other two clan names referred to the official titles assumed by theirbranch ancestors.In the Pre-Tang era, among these three subgroups only the activities of the T'ai-weibranch have been continuously documented in dynastic histories, 29 and only the Tai-wei24 Twitchett, Ibid. & Ikeda On, Ibid: 79.25 See HTS, 73: 2965-2968.26 See the 812 Preface.27 See HTS, 73: 2965-2979.28 Ibid.29 Records about less recognizable branches also exist in dynastic histories both beforeand during the Tang. Examples can be found in Lu Kao's biography in Liang shu,and in Lu Chih's biography in Chiu T'ang-shu and HTS. This Lu Chih was a famousscholar of the Spring and autumn annals. His original name was Lu Ch'un, but he laterchanged his name to Chih to avoid a taboo in the name of emperor Hsien-tsung. I shallhereafter refer to him by his original name of Ch'un to avoid confusion with Lu Chih,our protagonist. Liang shu, Chung-hua shu-chii edition, 1973, 26: 398-99; Liu Hsii,et al. Chiu T'ang-shu (CTS hereafter), Peking: Chung-hua shu-chii edition, 1975: 189:4977-8; HTS, 168: 5127-8. For Lu Ch'un's classical scholarship , see Chang Chitin,"T'an Chao Lu san-chia Ch'un-chiu chih shuo," in Ch'ien Mu hsien-sheng pa-shih suichi-nien lun-wen-chi, Hongkong: Hsin ya yen-chiu-suo, 1974: 149-59; E. G.Life and Family Background^20branch, with its regional base in Wu hsien of Wu chiin, was considered one of the fourmost prominent aristocratic lineages in the southeast area. This is largely due to itsenormous social prestige and its members' powerful positions in govemment. 3°Compared with the Tai-wei branch, information about members of the Tan-fu and Shih-lang branches is quite limited prior to the T'ang.31 Nonetheless, it was precisely in theTang that members of the Tan-t'u and Shih-lang families began to make names for theirlineages.To be sure, the social esteem and powerful influence accumulated for centuries alongwith its geographically concentrated local base sustained the Lu lineage during the militaryrebellions and political chaos arising toward the end of Six Dynasties. But a large questionstill remains: how did it maintain power and continue its reputation for being a member ofthe national elite in the T'ang, an era when the social and political stage was mainlydominated by the northern aristocracy, and when the imperial rulers attempted to build astrong and effective bureaucracy? To answer these questions, we shall examine howmembers of the three major Lu branches emerged in T'ang history. Only then can we findout where Lu Chih's immediate family stood in the lineage.Pulleyblank, "Neo-Confucianism and Neo-Legalism," Arthur F. Wright, ed., TheConfucian persuasion, Stanford University Press, 1960: 89-91; David McMullen, Stateand scholars in Tang China, 1988: 101-103; Inaba Ichirii, "ChiItili ni okeru shinjugakuundo no ichi kosatsu," Chagoku chaseishi kenkya, 1970: 390-396.30 For the development of the T'ai-wei branch in the pre-Tang era, see Chin shu, Peking:Chung-hua shu-chii edition, 54: 1467-68; Sung shu, Peking: Chung-hua shu-chiiedition, 53: 1510; Ho Ch'i-min, "Chung-ku nan-fang men-ti," in his Chung-ku men-tilun-chi, Taipei: Hsiieh-sheng shu-chii, 1978, ch. 4; Mao Han-kuang, Liang Chin Nan-pei-ch'ao shih-tsu, 1966, vol. 1: 140-46; vol. 2: 408, 440, 460, 463, 470, 472, 482-83, 494, 502-05, 507, 509-10.31 Perhaps, the size of these two branches made a difference. According to the availablesources, both before and during the Tang, the size of these two subgroups bore nocomparison with that of the Tai-wei branch. See HTS, 73: 2966-80; Lin Pao, Yuan-ho hsing-tsuan (hereafter YliFIT), Ssu-k'u ch'iian-shu chen-pen pieh-chi edition, ch.10, la-4a.Lu Chih^ 21The Tan-t'u branchThe person who carried the line of the Tan-t'u subgroup into the Tang was Lu Te-ming (his real name was Yiian-lang, ca. 560-630). Prior to the T'ang, Te-ming's eruditionhad already won him respect and office in both the Ch'en and Sui dynasties. Whenemperor T'ang T'ai-tsung, then still Prince of Ch'in, recruited Te-ming to be anAcademician of the Institute of Education (or Academy of Literary Study, Chin fu wen-hsiieh-kuan hsiieh-shih) in his palace, Lu Te-ming not only had achieved a reputation as arenowned classical scholar, but was also known for his loyalty in not serving the rebelgeneral Wang Shih-ch'ung. 32 In 624, Lu Te-ming greatly enhanced his reputation as aConfucian scholar in a debate on Confucianism, Taoism and Buddhism. 33 He wassubsequently promoted to the rank of Erudite of the National University (Kuo-tzu po-shih),and was ennobled as Baron of Wu-hsien (Wu-hsien nan).34As far as we know, Lu Te-ming was one of only two members of the entire Lu lineagewho served in the governments of the first two Tang emperors. 35 The obvious point isthat Lu Te-ming's rise in Tang officialdom relied not so much upon his family backgroundas on his wide prestige as a learned scholar. This, of course, is not to deny the importance32 For Te-ming's life and his refusal to serve Wang Shih-ch'ung, see CTS, 189: 4944-5;HTS, 198: 5639-40. As for the Wen-hsiieh-kuan, see Fukusawa Sokichi,"Bungakukan gakushi ni tsuite," Kumamoto daigaku kyoiku gakubu kiy6, 1 (1953):35-41.33 For Lu Te-ming's scholarship, see David McMullen, State and Scholars in T'angChina, 1988: 33 & 72. For the practice of three teachings debate in the Tang, see LoHsiang-lin, "Tang-tai san-chiao chiang-lun k'ao," in his T'ang-tai wen-hua-shih,Taipei: Commercial Press, 1955: 159-76.34 See CTS, 189: 4945 & HTS, 198: 5639-40. For the conferment of noble status, seeCTS, 43: 1821; Tu Yu, T'ung-tien, Peking: Chung-hua shu-chii reprint (1984), ch. 19:110; Wang Shou-nan, Sui T'ang shih, Taipei: San-min shu-chii, 1986: 412-4; Charles0. Hucker, 1985: 35.35 Mao Han-kuang thought Lu Te-ming was the only member who served in the earlybureaucracy. See his "Wu chfin Lu-shih," in Tao Hsi-sheng hsien-sheng chiu-chihjung-ch'ing chi-nien lun-wen-chi, Taipei: Shih-huo, 1989: 61. However, there wasanother Lu Shih-chi who also served during Tai-tsung's reign. See CTS, 188: 4932;HTS, 195: 5584.Life and Family Background^22of family background. After all, very few common people of undistinguished origin couldafford the time and money to be a scholar. 36Although Lu Te-ming himself never seems to have acquired substantial political power,one of his two sons, Lu Tun-hsin, rose to be Chief Minister from 665 to 666 underemperor Kao-tsung (reign 649-83).37 Like his father, Lu Tun-hsin was also ennobled asBaron of Chia-hsing county. 38 Though information regarding his administration or hisability as an official hardly exists, the fact that he was promoted to be Chief Minister in 665nevertheless calls for attention.It is known that by the late months of 660, the court of emperor Kao-tsung was firmlycontrolled by his favorite consort, empress Wu. 39 In spite of her recent victory inliquidating all her enemies in the top echelons of government, empress Wu still tried toamass all the support she could within the bureaucracy in order to pave the way for a futuretakeover of the imperial throne 4 0 To promote someone from a southern aristocratic lineagewhose immediate family had become more recognizable under Tang rule would, on theone hand, maintain the previous policy of balancing the power of regional aristocraticgroups and, on the other hand, cultivate her own power base. 41 From this point of view, itis likely that the elevation of Lu Tun-hsin to the chief ministership was engineered byempress Wu in order to further consolidate her power.The descendants of the Tan-t'u branch continued to serve in the Tang bureaucracy tillthe end of the dynasty. Most of them occupied either high or middle rank positions.42 In36 See Twitchett, Chinese meritocracy, 1974: 24.37 See Lao Ke & Chao Yiieh, T'ang Shang-shu-sheng lang-kuan shih-chu t'i-ming k'ao(Lang-kuan shih-chu hereafter ), Yiieh-he ching-she ts'ung-shu, Kyoto: Chung-wench'u-pan-she reprint, 1978: 64-5; HTS, 61: 1643.38 HTS, ibid.39 Denis Twitchett and Howard Wechsler, ""Kao-tsung (reign 649-83) and the empressWu: the inheritor and the usurper," in The Cambridge history, 1979: 251-55; also seeR. W. L. Guisso, Wu Tse-t'ien and the politics of legitimation in T'ang China,Western Washington University, 1978, ch. 3.40 Ibid.41 Denis Twitchett, "T'ang ruling class," Perspectives, 1973: 51.42 HTS, 73: 2967; also see Mao Han-kuang, "Wu chiin Lu-shih," 1989: 61.Lu Chih^ 23all, according to the New T'ang History genealogy, there were twenty-one male membersin the Tan-t'u branch during the Tang era. Fifteen of that number served in thegovernment; most of their appointments seem to be far away from Wu chiin andconcentrated in the metropolitan area. 43 As their active participation in the bureaucracyapparently provided them with more assets to sustain the elite status of their branch in thesociety, it simultaneously increased their dependence upon the Tang state.The Tai-wei branchAs stated previously, the Tai-wei branch was the largest subgroup within the Lulineage. During the Tang period alone, it produced one hundred and fourteen maleoffspring of whom eighty-one served in the government. Moreover, three descendantsalso rose to be chief ministers." A funeral inscription says that the earliest member of thisbranch in the Tang was Lu Shan-jen who served as a local official in the early Tang. 45We do not know how Lu Shan-jen joined the bureaucracy, but his eldest son, Lu Chien-chih, was famous for calligraphy, and this talent definitely earned him access to theofficialdom. 46 However, the Tai-wei branch does not seem to have restored theirpowerful pre-T'ang position in the government until Chien Chih's nephew, Lu Yuan-fang(639-701), rose to be Chief Minister during empress Wu's rule (690-705). 47Lu Yuan-fang entered the bureaucracy through the doctoral examination system. Hepassed the ming-ching (enlightening the classics) examination and special decree43 HTS, Ibid.44 HTS, 73: 2968-78.45 This tomb inscription was written during empress Wu's reign (690-705) by the famousscholar official Chang Yiieh to commemorate the Chief Minister Lu Yuan-fang, whodied in 701. See CTW, ch. 231, 17. Mao Han-kuang does not include Lu Shan-jen inthe first generation of the T'ai-wei branch members who served in Tang officialdom.See his "Wu chiin Lu-shih," 1989: 57.46 CTS, 88: 2875; HTS, 73: 2968 & 116: 4235.47 CTS, 88: 2875; HTS,116: 4235; CTW, ch. 231: 17.Life and Family Background^24examinations probably around 659. 48 Another early member who also entered the civilservice by means of the doctoral examination was Lu Yii-ch'ing. He was Yuan-tang'suncle, and obtained the chin-shih (advanced scholar) degree during emperor Kao-tsung'stime.49The careers of Lu Yilan-fang and Lu Yii-ch'ing shared a similar pattern. To our bestknowledge, they were the first two T'ai-wei Lus to pass the doctoral examinations in orderto acquire a nominal official status. This not only proves that their scholarship and literaryability reached a certain recognizable standard, but also implies that their decisions to takethe examinations could have been influenced by Kao-tsung's new emphasis on theimportance of the examination system in 659. 50 To be sure, the examination system wasnot the only means to obtain nominal official status, 51 but it gradually became a necessarychannel for officials who had ambitions for the highest positions in the court. 52 From thispoint of view then, it was by no means a coincidence that both Lu Yuan-fang and Lu YU-ch'ing chose to join the bureaucracy through the doctoral examinations. They must haverealized that the examination system was the surest way to enter and advance in thebureaucracy. More importantly, they must have also believed that it was the safest channelto preserve the traditional power position of their families.48 CTS, ibid.; HTS, ibid.; CTW, ibid.; Hsti Sung, Teng-k'o chi-k'ao, Peking: Chung-huashu-chii edition, three vols., 1984, vol. 3, 27: 1106. According to the Teng-k'o chi-k'ao, in addition to the ming-ching examination, there were eight special subjectsdecree examinations held in 659. Since Lu Yiian-fang was then about twenty yearsold, it is very likely that he took the ming-ching examination and these special subjectsdecree examinations in that year.49 HTS, 116: 4239, also see Teng-k'o chi-k'ao, vol. 3, 27: 1031.50 Kao-tsung's emphasis on the examination system is demonstrated by his ordering thecompilation of the Hsing-shih lu [Record of surnames and lineages] in which thecriteria for ranking families was strictly based upon office and personal achievementduring the reigning dynasty. For the compilation of Hsing-shih lu and the problemsentailed, see Denis Twitchett, "Tang ruling class," Perspectives, 1973: 62-4.51 Denis Twitchett, Chinese meritocracy, 1974: 8-12; also see his introduction in TheCambridge history, 1979: 21.52 Twitchett, Chinese meritocracy, 1974: 23.Lu Chih^ 25The second similarity of Lu Yuan-fang and Lu Yii-ch'ing's careers is that theirpromotions to high ranking posts were all conferred by empress Wu as rewards for theiraccomplishment of missions in the border regions. 53 This again demonstrates that in orderto consolidate her own power, empress Wu apparently preferred to promote members whowere not likely to get involved in factional intrigues54 and whose immediate families hadjust begun to participate in the government.In fact, the power and prestige of the Tai-wei branch was reassured by Lu Yuan-fangand Lu Yii-ch'ing's performance in government and was continued by that of theirdescendants.55 As far as we know, at least six members of the Tai-wei branch occupieddifferent official posts during Hsiian-tsung's rule (712-756). Such achievement must havehelped them to establish a powerful position in the bureaucracy. Ironically, thebureaucratic success of the Tai-wei Lus loosened their ties with their local property in Wuchitin. Thus, either in the late eighth or the ninth century, the descendants of Lu Yiian-fangbegan gradually to sell off their local property. 56This sale of local property might very well be a logical result of their success in theTang bureaucracy. Living a bureaucratic life at court and in different parts of the Tangempire must have made it difficult for the Tai-wei Lus to care for their local property inWu chiin. Besides, in the late ninth century, Lu Yiian-fang's seventh generation grandsonis said to have owned at least several hundred Chinese acres (mou) of farm land and thirty53 For Lu Yuan-fang, see CTS, 88: 2875; HTS, 116: 4235; Lu Yii-ch'ing, see HTS, 116:4239. Only the HTS biography mentions Lu Yii-ch'ing's achievement in this matter.Since this event took place at the beginning of 697, the pacification of the northwestborder area most certainly must refer to the invasion of Ling-chou (modern Ninghsia)by the Turks in 697. See Ssu-ma Kuang, Tzu-chih t'ung-chien (TCTC hereafter),Peking: Chung-hua shu-chii edition 1956, 206: 6512-6.m Indeed, Lu Yuan-fang was demoted from his position as Chief Minister when empressWu thought he was involved in a factional intrigue, but he was soon reappointed to thesame position after she learned of his innocence. See CTS, Ibid.; HTS,.Ibid.; also seeMao Han-kuang, "Wu chiin Lu-shih," 1989: 58.55 For Lu Yllan-fang's sons, see CTS, 88: 2876-77; HTS, 116: 4236-8. For Lu Yii-ch'ing's descendents, see HTS, 116: 4239-40; CTS, 145: 3937-38. For the Yinprivilege, see Denis Twitchett, Chinese meritocracy, 1974, p.9.56 Ta-T'ang chuan-tsai, Shou-shan-ke ts'ung-shu edition: 1; also see HTS, 196: 5613.Life and Family Background^26houses in the southeast, even though he might indeed have considered himselfimpoverished.57 This being the case, their property in the eighth century must have beenmuch larger than is implied in the above account. Such large property holdings naturallybecame a burden for the Lus if they wanted to succeed in a government which increasinglytended toward the bureaucratization of pretentious aristocratic families.Of course, once the Tai-wei Lus lost their local property due to practical difficulties,they could always purchase land near the capital such as in the popular Lo-yang region.This was exactly what many other aristocratic families did during the Tang. In so doing,these aristocratic families became more and more centralized as a metropolitan elite andgradually lost their local ties. 58 As a result, the power foundation of their aristocraticcachet was also in danger. In short, from the point of view of aristocratic families as awhole, the Tai-wei Lus' property sale testifies to the gradually accelerating trend ofbureaucratic transformation of the aristocracy.To be sure, there were members of the Tai-wei branch who stayed in Wu chiin andgained fame through scholarship or a Taoistic life style without entering the ranks ofofficialdom. A case in point is the famous late Tang scholar Lu Kuei-meng, the seventhgeneration grandson of Lu Yiian-fang. 59 On the other hand, Lu Kuei-meng's uncle, LuHsi-sheng, and his descendents never seemed to have ceased pursuing officialappointments. As a matter of fact, Lu Hsi-sheng was even appointed to the position ofChief Minister in 895 though he only served for a short while. 60 Compared with thebureaucratic path taken by the majority of the Tai-wei Lus, or for that matter the Tan-t'u57 This can be found in the biography of Lu Kuei-meng. See HTS, 196: 5613. Also seeWang Tang, T'ang Yii -lin, Shanghai: Ku-chi ch'u-pan-she edition 1978, 4: 147.58 On the matter of centralization of the Tang aristocracy, see Mao Han-kuang, "Ts'ungshih-tsu chi-kuan ch'ien-i k'an T'ang-tai shih-tsu chih chung-yang-hua," in Li-shih yii-yen yen -chiu -so chi-k'an, 52 (1981): 421-510, esp. see his conclusion; also see Ebrey,The aristocratic families, 1978: 28 -32; Twitchett, The Cambridge history, 1979: 22.59 HTS, 196: 5613.60 HTS, 116: 4238, 63: 1750-1751, 73: 2974.Lu Chih^ 27Lus, Lu Kuei-meng's case hardly represented the general trend of Lu clan developmentduring the Tang.A new perspective on the Shih-lang branch andLu Chih's immediate familyThe general trend of active government service applied equally to the evolution of theShih-lang branch. Like the other two major subgroups, the great majority of the Shih-langbranch members also participated in the bureaucracy. Unlike the Tai-wei and Tan-fu Luswhose sub-choronym was always represented by Wu hsien of Wu chiin, the geographicbase of the Shih-lang branch was in Chia-hsing county of Wu chan. 61 It is not clear whenthe Shih-lang group became established in Chia-hsing 62 but, like the Tan-fu branch, itcould not match the high prestige and power generally enjoyed by the Tai-wei brancheither before or during the Tang.According to the New T'ang History genealogy, the person who initiated the Shih-langbranch was Lu Kuan. The only information we have about him is that he occupied the postof Vice Director of the Secretariat (Chung-shu shih-lang) in the Eastern Chin dynasty. 63The New Tang History genealogy further states that Lu Kuan's seventh generationgrandson, Lu Hsiin, acquired an important position in the southern Ch'en dynasty (557-61 See the biographies of Lu Yuan-fang, Lu Hsiang-hsien, and Lu Kuei-meng of the Tai-wei branch, Lu Te-ming of the Tan-t'u, and Lu Chih and Lu I of the Shih-lang branch.CTS, 88: 2875-6; HTS, 116: 4235-7 & 196: 5612; CTS, 189: 4944-5; HTS, 198:5639-40; CTS, 139: 3791 & HTS, 157: 4911; CTS, 179: 4668 & HTS, 183: 5383.62 As noted above, Lu Tun-hsin, Lu Te-ming's son, was ennobled as Baron of Chia-hsing(HTS, 198: 5640) during Kao-tsung's reign. Both Twitchett and Yen I-p'ing havefollowed the Ch'ing dynasty nien-p'u in writing that Lu Tun-hsin was Lu Chih 'sgreat-grandfather, and thus implying that the Shih-lang branch probably had becomeestablished in Chia-hsing at least by Tun-hsin's time. See Denis Twitchett, "Lu Chih,"1962: 88; Yen I-p'ing, Lu Hsiian-kung nien-p'u, 1975: 1. However, we shall see thatLu Tun-hsin was not Lu Chih's great-grandfather; at least, no T'ang source makes sucha statement. The 1745 genealogy entitled Lu-shih shih-p'u which I have cited in thisstudy does not make any such connection either.63 HTS, 73: 2978.Life and Family Background^28589).64 From the Ch'en dynasty to the early Tang, the Shih-lang Lus do not seem to haveacquired any position in the bureaucracy. It is only during the Kai-ytian era (713-741) thatwe find Lu Hsiin's ninth generation grandson, Lu Ch'i-wang, serving as Director of thePalace Library (Pi-shu-chien) with a rank of 3b.65 In general, this position was onlyconferred upon officials who had outstanding literary ability. 66 This makes it seem verypossible that Lu Ch'i-wang, like members in the Tan-t'u and Tai-wei branches, reliedupon his literary knowledge to enter the bureaucracy.The New Tang History genealogy records that Lu Ch'i-wang had eight sons, and allof them received official appointment. The same source states that one of his sons, Lu Pa,was Lu Chih's father, thus making Lu Ch'i-wang Lu Chih's grandfather. 67 Thisgenealogical attribution for the Shih-lang branch has, however, created a certain amount ofconfusion. In the following discussion, we shall partially reconstruct the Shih-lang branchin order to obtain a valid picture of Lu Chih's immediate family.In the first place, instead of eight sons as listed in the New T'ang History genealogy,Lu Ch'i-wang had only six sons. The eldest son named in the New T'ang Historygenealogy was Lu Mi. He actually belonged to a Lu clan in Honan which was of foreignorigin. This explains why Lu Mi was not included as Lu Ch'i-wang's son in the Yuan-hohsing-tsuan. 68 The two Lu Ch'an listed in the New Tang History genealogy should64 Ibid .65 YHHT records that Lu Ch'i-wang occupied this position during the K'ai-yiian period.See YHHT, ch. 10, 4a. This would seem to invalidate Mao Han-kuang's speculationthat Lu Ch'i-wang served during empress Wu's reign. See his "Wu chiin Lu-shih,"1989: 60. Some sources say that Lu Ch'i-wang's position was Vice Director instead ofDirector of the Palace Library (Pi-shu shao-chien). See YHHT, ch. 10, 4a and DenisTwitchett, "Lu Chih," 1962: 88. However, according to a Tang tomb inscriptionpreserved in a Sung dynasty source, Ch'i-wang's position was Pi-shu-chien. See Chi-ku lu-mu, Ou-yang Fei compiled, in Miao Ch'iian-sun, ed., Yiin tzu-tsai k'an ts'ung-shu edition, ch. 8, and Po-k'e ts'ung-pien, Ts'ung-shu chi-ch'eng edition, ch. 14: 388-9. Also see Ts'en Chung-mien, Yiian-ho hsing-tsuan ssu-chiao chi (hereafterabbreviated as YHHTSCC), 3 vols., Shanghai: Commercial Press, 1948, vol. 10: 901.66 See Mao Han-kuang, "Wu chiin Lu-shih," 1989: 60.67 HTS, 73: 2979.68 On Lu Mi's origin, see CTW, 684: 29b; YHHT, 10: 5a.Lu Chih^ 29actually refer to one and the same person, because two different characters for Ch'an wereused interchangeably during the Tang.69 From an essay written by a contemporary, weknow the seniority of Ch'i-wang's six sons should very likely be listed in the followingorder: Lu Wei, Lu Feng (or Lu Li), Lu Pa, Lu Jun (or Lu Chien), Lu Huai, and LuCh'an.70 Meanwhile, contrary to the New T'ang History genealogy, it seems that amongLu Ch'i-wang's six sons only Lu Feng's (or Lu Li's), not Lu Ch'an's, family lineextended into the last years of the Tang. 71Secondly, Ch'i-wang's son Lu Pa was not Lu Chih's father. According to a T'angsource, the name of Lu Chih's father was Lu K'an. 72 Although, the name provided byCh'iian Te-yii (759-818) in his Preface was Lu K'an-ju,73 it is quite certain that these twonames referred to the same person and that Lu Chih's father was Lu K'an or Lu Ktan-ju. 74The question that still needs to be answered is: were Lu Pa and Lu K'an the same person ornot?69 Ts'en Chung-mien, Lang-kuan shih-chu t7-ming hsin k'ao-ting, 1984: 79.70 The person who mentioned that "Lu Feng (or Lu Li) and his younger brothers Lu Pa,Lu Jun and Lu Huai . . ." was Fu Tsai. Moreover, we find that Lu Ch'an must be theyoungest son since he still served in the government toward the end of Yiian-ho era(805-20). See Ts'en Chung-mien, YHHTSCC, ch. 10: 902. Although we have noinformation to confirm the age difference between Lu Wei and Lu Feng, it seems verylikely that Lu Wei was the eldest brother. For while Lu Wei was highly regarded bythe notable ku-wen (ancient prose) writer Hsiao Ying-shih (706-58), Lu Feng (or LuLi) was a close friend of Hsiao Ying-shih's son. See CTW, 691: 8a. Of course, it isalso possible that Lu Wei could very well have been the second eldest brother. Withoutfurther information, we can only speculate on their order of seniority. Fu Tsai led areclusive life on Lu mountain in Kiangsi until around 797. He was then given a post inprovincial government by Li Sun, a notable financial official in emperor Hsien-tsung'sreign. See CTW, 690: la, and 688: la. For Li Sun, see CTS, 123: 3522 and HTS,149: 4805. We shall soon explain why Lu Feng and Lu Jun had other names.71 The biography of Lu I says that Lu Feng was Lu I's great grandfather. This contradictswith the HTS genealogy. See CTS, 179: 4668; HTS, 73: 2978-79. Since no otherTang sources offer any information on this matter, and since the HTS genealogy oftencontains mistakes as is and will be shown here, I shall accept the biographicalinformation contained in the CTS as valid for the time being.72 YHHT, ch. 10: 4a.73 See Ch'ilan Te-yii's Preface, "Tang tseng ping-pu shang-shu Hsiian-kung Lu ChihHan-yiian-chi hsii," in CTW, 493: 11 a. Also see the one contained in HYCC, 1768edition,! -4.Lu Chih's biography in the CTS also says that his father's name was Lu K'an. SeeCTS, 139: 3791 & Ts'en Chung-mien, YHHT ssu-chiao-chi, ch. 10: 902.Life and Family Background^30The answer is negative. Chitin Te-yil seems to have been Lu Pa's close friend. Heonce composed a rhyme-prose (fu) to see Lu Pa off for Ching-chou (modern Hupei). 75 IfLu Pa were indeed Lu Chih's father, or if Lu Pa and Lu K'an were the same person, itwould have appeared very unreasonable for Ch'iian Te-yii not to have mentioned any ofthese "facts" in his Preface to Lu's extant works. Furthermore, since Lu Chih's father diedlong before he reached adulthood, and since Ch'iian Te-yii was actually five years youngerthan Lu Chih, it would have been virtually impossible for Ch'iian Te-yii to have befriendedLu Pa if Lu Pa were Lu Chih's father. Evidently, Lu Pa and Lu K'an were two differentpeople; Lu Pa was Lu Ch'i-wang's son and definitely not Lu Chih's father.Thirdly, since Lu Pa was not Lu Chih's father, and Lu Ch'i-wang in turn could nothave been his grandfather, Lu Chih definitely belonged to another family under the Shih-lang branch. That is, we still need to find out who Lu Chih's grandfather really was andwhat the actual connection between Lu Chih and Lu Ch'i-wang's family was.As listed in the Yiian-ho hsing-tsuan, Lu Chih's grandfather was Lu Ch'i-cheng. Acousin to Lu Ch'i-wang, Ch'i-cheng served as a district Magistrate, probably during theK'ai-yiian era when Lu Ch'i-wang was a Director in the Palace Library. 76 It is said that LuCh'i-cheng had only one son whose name, as mentioned above, was Lu K'an or Lu K'an-ju. Like his father, Lu K'an's only official position was also as Magistrate, either of Li-yang or Li-shui county (near modern Nanking). 77 In addition to Lu Chih, the Yilan-hohsing-tsuan says that Lu K'an had two other sons, Lu Shang and Lu Keng. 78 Thiscontradicts the New T'ang History genealogy which lists Lu Wei as Lu Shang's father andsays nothing about Lu Keng. 79 Since the New T'ang History genealogy often contains75 CTW, 483: 2b.76 YHHT, 10: 4a. Also see Ts'en Chung-mien, YHHTSCC, 10: 901-2.77 YHHT mentions Li-shui while CTS and Ch'ilan Te-yii's Preface agree that Li-yang wasthe county where Lu K'an served. See YHHT, ibid.; CTS, 139: 3791; CTW, 493:11 a.78 YHHT, ibid.79 HTS, 73: 2980.Lu Chih^ 31mistakes80 as has already been shown by its incorrect identification of Lu Pa as Lu Chih'sfather, it seems more reasonable to believe that Lu Shang and Lu Keng were Lu Chih'sbrothers, even though we have no way to prove it. In any event, as far as we know, theirfamily line seems to have been transmitted only by Lu Chih's son Lu Chien-li. Aside fromknowing that Lu Chien-li entered the officialdom through the chin-shih examination noearlier than 816, his life remains a complete blank to us. 81With this revision of our perspective on the Shih-lang branch, especially with ourknowledge of the connection between Lu Chih's immediate family and Lu Ch'i-wang'sfamily, we shall now discuss some characteristics of Lu Ch'i-wang's family in order toacquire more substantial understanding of Lu Chih's own family background.The main characteristics of the Shih-lang branchThe first distinct feature of the Shih-lang branch, shared actually by members of theTan-t'u and Tai-wei branches as well, is their literary and cultivated background. One oftheir contemporaries once described Lu Ch'i-wang's sons in this way: "Lu Feng and hisyounger brothers Lu Pa, Lu Jun, and Lu Huai were all famous for their literary abilities andvirtuous conduct (wen-hsing)." 82 Despite the fact that such praise may often be subjective,it reflects the image of the Lu brothers among their contemporaries. In fact, the circle of theLu brothers' friends included many illustrious literary figures of the time.80 On this matter, see Chou I-liang, ed., Hsin T'ang shu tsai-hsiang shih-hsi piao yin-te,in Harvard-Yenching Index series, no. 16, Taipei reprint, 1966, introduction: i-xvii.81 When the Chief Minister Ts'ui Chitin (772-832) administered the 816 doctoralexamination, he did not pass Lu Chien-li despite the fact that he himself had ac