UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Echo of the master, shadow of the Buddha : the Liezi as a medieval masters text Kreger, Wayne 2016

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

Item Metadata


24-ubc_2016_may_kreger_wayne.pdf [ 1.73MB ]
JSON: 24-1.0228779.json
JSON-LD: 24-1.0228779-ld.json
RDF/XML (Pretty): 24-1.0228779-rdf.xml
RDF/JSON: 24-1.0228779-rdf.json
Turtle: 24-1.0228779-turtle.txt
N-Triples: 24-1.0228779-rdf-ntriples.txt
Original Record: 24-1.0228779-source.json
Full Text

Full Text

ECHO OF THE MASTER, SHADOW OF THE BUDDHA:THE LIEZI 列子 AS A MEDIEVAL MASTERS TEXTbyWAYNE KREGERA THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES(Asian Studies)THE UNIVERSITY OF BRITISH COLUMBIA(Vancouver)April 2016© Wayne Kreger 2016ABSTRACTThe present work examines the Chinese Masters Text (zishu 子書) the Liezi 列子, purported tobe composed in the 5th century BCE, though more likely achieving its current form in the 4th centuryCE. It situates the claims Liezi in the intellectual and spiritual climate of 4th century CE in its role as amember of the Masters Text category, and reads the text's ontological and normative program in light ofthe  flourishing  xuanxue 玄學  and  prajñāpāramitā discourse  of  that  era.  Chapter  One  traces  theevolution of the Masters Text category from the Warring States period, through the Han dynasty, andinto the early medieval period,  relying on the understanding of “Masters Text” offered by WiebkeDenecke in her  Dynamics of Masters Literature (2010). I argue that textual authority accrued by thiscategory serves as a sufficient impetus to create such an inauthentic document in approximately 350CE. Chapter Two reviews the most recent contributions to the debate over the authenticity of the Liezi,and concludes that the text is certainly a 4th century CE compilation, though containing some earliermaterial. Chapter Three is a concise survey of the ontological and normative position of the text, with achapter by chapter analysis of the  Liezi. Chapter Four uses this analysis of the  Liezi to compare thethought therein with contemporary thinkers such as Wang Bi, Guo Xiang, Ruan Ji, and Xi Kang. Iconclude that the Liezi was likely compiled in an effort to argue for the ontological scheme of Wang Biagainst that of Guo Xiang, and that it does not explicitly follow Ruan Ji or Xi Kang in advocating forthe pursuit  or practice of longevity techniques.  Chapter Five compares notions of “Nonbeing” and“emptiness” in the Liezi to Buddhist speculations on “emptiness” unfolding in China up to and duringthe 4th century CE. I conclude that despite frequent speculation on the part of modern and pre-moderncommentators,  there  is  little  conceptual  alignment  between the  Liezi and  the  developing Buddhistschools. iiPREFACEThis dissertation is original, unpublished, independent work by the author, Wayne Kreger. iiiTABLE OF CONTENTSABSTRACT...............................................................................................................................................iiPREFACE.................................................................................................................................................iiiTABLE OF CONTENTS..........................................................................................................................ivACKNOWLEDGEMENTS......................................................................................................................viDEDICATION.........................................................................................................................................viiINTRODUCTION.....................................................................................................................................1A Concise Literature Review................................................................................................................3Structure of this Dissertation.................................................................................................................7Notes on Conventions and Definitions................................................................................................10CHAPTER ONE: THE LIEZI AND MASTERS TEXTS.......................................................................111.1 Masters Texts And Authority.........................................................................................................131.1.1 The Creation of Categories: Sima Tan, Liu Xiang, and Ban Gu...........................................211.1.2 Elements of Masters Texts.....................................................................................................251.2 Masters Texts in the Warring States Period...................................................................................281.3 Masters Texts After the Fall of Qin...............................................................................................381.4 Conclusion.....................................................................................................................................50CHAPTER TWO: DETERMINING THE DATE OF THE LIEZI..........................................................532.1 A Brief Textual History of the Liezi..............................................................................................582.2 The Language and Grammar of the Liezi......................................................................................652.3 Parallels between the Liezi and the Zhuangzi...............................................................................712.4 Parallels between the Liezi and other Works.................................................................................822.5 Traces of South Asian Thought in the Liezi..................................................................................892.6 Conclusion.....................................................................................................................................99CHAPTER THREE: THE CORE TEACHINGS OF THE LIEZI.........................................................1023.1 The Eight Pian of the Liezi..........................................................................................................1033.1.1 “Tian rui” 天瑞 - “The Portents of Heaven”......................................................................1033.1.2 “Huangdi” 黃帝 - “The Yellow Emperor”..........................................................................1073.1.3 “Zhou Mu wang” 周穆王 - “King Mu of Zhou”................................................................1113.1.4 “Zhong ni” 仲尼 - “Confucius”..........................................................................................1163.1.5 “Tang wen” 湯問 - “The Questions of Tang”.....................................................................1203.1.6 “Li ming” 力命 - “Effort and Fate”....................................................................................1253.1.7 “Yang Zhu” 楊朱 - “Yang Zhu”.........................................................................................1293.1.8 “Shuo fu” 說符 - “Explaining Connections”......................................................................1323.2 Conclusion...................................................................................................................................135CHAPTER FOUR: THE LIEZI AND EARLY MEDIEVAL XUANXUE THOUGHT.......................1394.1 Contextualizing Xuanxue............................................................................................................1394.2 Wang Bi, Guo Xiang, and the Liezi.............................................................................................1484.3 The Worthies of the Bamboo Grove and the Liezi......................................................................1624.4 Conclusion...................................................................................................................................177CHAPTER FIVE: THE LIEZI AND BUDDHIST PRAJÑĀPĀRAMITĀ THOUGHT.......................1815.1 The Search for the Buddhism in the Liezi...................................................................................1825.2 Buddhism in Translation up to and in the 4th Century CE.........................................................1885.2.1 The Primacy of the Prajñāpāramitā.....................................................................................193iv5.2.2 The Ontology of Non-dualism Described............................................................................1965.2.3 The Ontology of Non-duality in Practice............................................................................1985.3 The “Six Houses and Seven Schools” of the 4th Century...........................................................2015.3.1 The Xinwu 心無 School (“Emptiness of Mind”)...............................................................2025.3.2 The Jise 卽色 School (“Emptiness is Identical with Matter”)............................................2055.3.3 The Benwu 本無 School (“Original Emptiness”)..............................................................2105.4 The Prajñā Schools and the Liezi Compared..............................................................................2165.5 Conclusion...................................................................................................................................228CONCLUSION......................................................................................................................................231REFERENCES.......................................................................................................................................235Primary Sources................................................................................................................................235Secondary Works in Asian Languages..............................................................................................238Secondary Works in European Languages........................................................................................239APPENDIX: A FINDING LIST FOR THE LIEZI................................................................................252vACKNOWLEDGEMENTSI am grateful to the many individuals and organizations that have helped make this project areality. Generous funding was provided by the Social Sciences and Humanities Research Council ofCanada,  the  Canadian  Chinese  Help  Care  Society,  the  Khyentse  Foundation,  and  the  UniversalBuddhist Temple of Vancouver.Many  friends  and  colleagues  have  made  my  time  pursuing  this  project  more  fruitful  andfulfilling than it would have otherwise been. I am extremely grateful to my friend Matt Hamm, whoread through each version of the present work, all the while offering immensely thoughtful feedbackand support. I am also very thankful to Professor T. H. Barrett, who has generously shared with me hisknowledge about and enthusiasm for the Liezi. I am most indebted to my dissertation committee - ChenJinhua, Bruce Rusk, and Edward Slingerland - and offer my sincere appreciation for their tireless helpand encouragement in this project. Without their deep insights and generous efforts this project wouldbe  gravely  deficient.  I  extend  special  and  sincere  appreciation  to  Professor  Slingerland,  who  hasconsistently  and  kindly  offered  to  me  indispensable  guidance  throughout  my graduate  career.  Ofcourse, all errors and shortcomings in the present work are solely my responsibility.Finally, I wish to thank my family for their unfailing support both within and beyond academia.I am deeply grateful to my Aunt Gloria, my Uncle Greg, and my grandmother Helen Rotzien, whoencouraged me from the very beginning of my studies. I am extremely appreciative to my parents, Garyand Mil Kreger, as well as my siblings Amy, James, and Tami, for their unfailing support and love.Finally, I wish to express my enduring and wholehearted gratitude to my wife Minami and my daughterSenli for all they have done for me. I am so fortunate to have had their care and patience to see methrough to the end of this project and beyond. Everything I may accomplish is due only to their support,and nothing would be worthwhile if not for them.viDEDICATIONTo my family千里之行 始於足下viiINTRODUCTIONThe  Liezi 列子 , cousin to the much more widely read  Laozi 老子  and  Zhuangzi  莊子 , isrelatively under-studied in western sinology. Where the  Laozi and  Zhuangzi are accorded their  duerecognition in most courses dealing with Chinese thought, the  Liezi will scarcely garner a mention.Likewise, an abundance of scholarly and popular publications have exhaustively challenged how weread and understand the  Laozi and  Zhuangzi.  The  Liezi, by comparison, has not received the sameattention. This dissertation will explore the Liezi text in greater depth; this introduction is provided toexplain why the Liezi is deserving of the attention is has heretofore been denied.It is not that the  Liezi has been ignored in the Chinese tradition as it has in recent westernstudies.  During the Tang dynasty it  was elevated by means of a  new title,  being called the “TrueScripture of the Void” (chongxu zhenjing 冲虛真經 ).1 At the same time it, along with the Laozi, theZhuangzi,  and  the  Wenzi 文 子 ,  became  the  basis  for  a  new  textual  curriculum  to  replace  the“Confucian”  classics.2 The  text  is  referenced  with  great  frequency and  presumed  authority  in  theTaiping yulan 太平御覽 on diverse matters. It is the source of common “set phrases” or chengyu 成語in modern Chinese, such as Qiren you tian 杞人憂天 (“A man from Qi worries about the sky [falling]”,Liezi 1.13),3 Yugong yi shan 愚公移山  (“The foolish old man moves a mountain”,  Liezi  5.3), andXiaoer bian ri 小兒辯日 (“Small children debate about the sun”, Liezi 5.7). In the modern period, thecontemporary  spiritual  teacher  Nan  Huaijin  南怀瑾  (1918-2012)  has  published  a  three  volumeinterpretation for a popular audience.41 Schmidt in Schipper and Verellen (2004), p. 62.2 Barrett (1996), p. 68.3 To my knowledge there is no standard method for reference to  Liezi chapters and pericopes. This dissertation uses acombination of two numbers: the first indicates which of the eight chapters is being referenced, and the second thepericope. Because this method is non-standard, I have provided a finding list as an appendix, which will help readersfind equivalent passages in important critical editions and A.C. Graham's English translation. See the Appendix for moreinformation. 4 Published as Liezi Yishuo 列子臆説 in 2011. 1I suggest that the primary reason the Liezi is not a popular subject in western sinology is that itsprovenance is disputed. Indeed, when the Liezi does receive scholarly attention, it is usually only thisquestion of the origin of the text that is pursued. This question is not unique to the modern period, aswe shall  see, but the modern period is unique in that for the past  several decades the question ofprovenance has dominated virtually all discussion. This provenance is rightly disputed. A great deal has been written about the authenticity of theLiezi –  specifically,  whether  it  is  a  legitimate pre-Qin text  or  a  Jin  dynasty compilation  (or  somecombination of the two). Until the publication of A. C. Graham's important essay in 1960,  The Dateand Composition of the Liehtzyy, most western scholars were comfortable with the idea of a pre-QinLiezi.5 Graham's article caused a sea change in western scholarship on the Liezi, so much so that hisconclusion that  the  Liezi was a medieval forgery is  now the assumption that implicitly undergirdswestern sinological explorations of the text.In his research on the  Liezi, Graham makes extensive reference to the relevant scholarship ofYang Bojun 楊伯峻  (1909 - 1992). It is unsurprising, then, that he ultimately concludes that the Lieziis not to be read as an authentic pre-Qin text, for this is the conclusion of Yang's pioneering efforts. It isonly in the past few decades that Chinese scholarship has largely begun to challenge the Liezi sceptics,claiming that the  Liezi is authentically pre-Qin. This important research has mostly been ignored bysinologists in the west.The paucity of scholarly investigation into the Liezi, in comparison to that directed at the Laoziand the  Zhuangzi,  offers  a  great  deal  of  opportunity for  new discoveries  in  the  study of  Chinesethought. Moreover, that which has been written on the Liezi is often contentious and contradictory. Thisdissertation aims to both offer new insights into the text as well as evaluate existing research. It is myhope that such an endeavour will begin to fill the gap in our understanding of this text.5 These positions are explored in detail in Chapter Two of this dissertation. 2Below, I give a brief review of the major scholarly works produced in the last several decadessince Graham's important 1960 article. I will offer my assessment of the strengths of weaknesses ofeach. Following this, I offer an outline of the present dissertation and conventions followed therein.A Concise Literature ReviewAlongside the aforementioned work of Yang Bojun, entitled Liezi jishi 列子集釋,6 an importantChinese study used by Graham is Wang Shumin's  王叔岷  (1914-2008)  Liezi buzheng  列子補正 ,published in 1948 in four volumes. The strength of Wang's meticulous work is his identification ofparallel  passages  that  link  the  Liezi and  other  texts  in  the  extant  record.  His  findings  are  neatlysummarized in Graham (1990b), pp. 225-228. While Wang only comments on sentences and sentencefragments that he has deemed worthy,  Yang's work is a complete critical edition, and notes textualvariations and includes important commentaries. Moreover, Yang is critical of the text, and in both hiscommentary and appended essays presents a strong case for the compilation of the document in theearly medieval period.In the decades following Graham's studies (which will be discussed below), several works inChinese  have  been  published  which  challenge  Yang's  conclusions.  Important  works  include  XiaoDengfu's  蕭登福 Liezi tanwei 列子探微  (1990), Chen Guangzhong's  陳廣忠  three part series ofessays in the periodical  Daojiao wenhua yanjiu 道教文化研究  (1996), and five short chapters byZheng Liangshu 鄭良樹 in a collection of his works entitled Zhuzi zhuzuo niandai kao 諸子著作年代考  (2001). Xiao's book is primarily topical, aimed more at addressing the theoretical framework andphilosophical claims of the Liezi, though it does include insights into the nature of the text. Chen's threepart series, collected under the general title Liezi fei weishu kao 《列子》非偽書考, argues that ZhangZhan could not have been the compiler of the text, that the presence of textual parallels in the text are6 Originally published in 1958 in Shanghai by Longmen lianhe shuju 龍門聯合書局. I have used Zhonghua shuju's 中華書局 2007 reprint of this document as my basic source text.3not  convincing in  establishing a  late  date  for  the  Liezi,  and  that  the language of  the  text  itself  isevidence of the document's early origin. Perhaps because of the succinct nature of Chen's argument,this series of essays is often cited in defense of the antiquity of the Liezi. Most of Zheng's essays aremore focused on particular chapters of the Liezi, and many aim to establish the antiquity of these withreference to their presumed adaptation of parallel sentences in other works (such as the  Zhuangzi).Zheng also includes a chapter on the grammar of the Liezi. The most extensive survey of the textualhistory of the  Liezi to date is Liu Peide's  劉佩德  Liezi xue shi 《列子》學史  (2015), which iscomprehensive in its survey of commentaries to the text. The most extensive recent attempt to advocate for the authenticity of the Liezi is Ma Da's 马达Liezi zhenwei kao bian 《列子》真伪考辨 (2000). The author argues that the text is definitely pre-Qin, and systematically considers most arguments made against this proposition. An entire chapter isdedicated to deconstructing Ma Xulun's twenty arguments, as well as other sceptical essays such asthose offered by Liu Zongyuan and Zhu Xi. Yang Bojun's work on the language of the Liezi, which heclaims  attests  to  its  inauthenticity,  is  also  singled  out  for  special  consideration.7 A.  C.  Graham'sarguably more conclusive complement  to this  linguistic  research receives  no equivalent treatment.8Another  chapter  is  dedicated to  elucidating  the discrepancies  between the  Liezi and Zhang Zhan'scommentary. The bulk of Ma's argument rests on his comparison of Liezi passages to parallels in otherextant writings from both the pre-Qin and post-Qin periods. In each instance, he remarks that the Lieziis indisputably the earlier document.I  will  challenge these claims of  antiquity in  Chapter  Two.  There are  three reasons for thisundertaking. First, it is simply a matter of course that any modern, substantial discussion of the Liezimust at least address the debate, as the bulk of the scholarly work published on the text is focused on7 See both Ma's preview (p. 137) and deeper discussion (pp. 398-417) in Ma (2000). 8 Ma's bibliography, while among the most extensive I have found in the Chinese language, does not mention Graham'swork. 4this aspect. Second, at the time of this writing no summary of recent publications on this question(especially those in Chinese) exists. Though this is not the central aim of this work, I hope that myproject here can at least temporarily fulfill this role, until a dedicated study emerges. Finally, knowingwhen the Liezi was likely compiled can tell us a great deal about what it is saying. Situating the claimsof the text in their cultural context reveal to us the concerns that were driving the compilation of thetext. If we conceive of the claims of the Liezi as answers, knowing more about the situation in which itcame together will begin to disclose what the questions they were intended to answer were. The earliest translation of the  Liezi into English is Lionel Giles' nearly complete edition, firstpublished in 1912, entitled Taoist Teachings from the Book of Lieh Tzü. The translation is often stiltedand lacks the nuance of later scholarly works, but it is certainly suitable for its time. Certain smallpericopes are omitted, as is the entirety of the “Yang Zhu” 楊朱 chapter. In that same year, a translationof the “Yang Zhu” chapter by Anton Forke, entitled  Yang Chu's Garden of Pleasure, was published.Both translations are part of the Wisdom of the East series of translations.The translation by Angus C. Graham has become the standard English version, and as such itdeserves a closer investigation.9 The translation is complete, and well annotated with notes that explainallusions made in the text as well as the translator's own interpretation of more complex philosophicalproblems. Graham, well known as a percipient interpreter of Chinese works of philosophy, rarely offersopportunity for criticism of his work. The latest reprint also offers a new preface, a general reading list,and textual notes that are especially valuable for researchers hoping to match Graham's interpretationwith what they find in the source text. In terms of translations, there is also the 1995 Eva Wong work  Lieh-Tzu: A Taoist Guide toPractical Living from Shambala Publications. Wong is explicit in her introduction in letting her readers9 Though Graham's translation was originally published in 1960, I have made use of the 1990 reprint. 5know that this volume is not a translation,10 though the work seems to be more or less a completeparaphrase of the basic Liezi text. Because of this, and the fact that the book lacks any kind of scholarlyapparatus11 and is uncritical in its approach to the text,12 I will not make reference to Wong's book in mystudy. It is perhaps more useful as a “self-help” style guide to readers uninterested in the historicalcontext or philosophical nuance of the text. In addition to his translation, Graham has also produced scholarly literature on the Liezi. One,mentioned above, is his 1960 The Date and Composition of the Liehtzyy.13 It is a neat summary of thecontroversy over authenticity up to the time of the article's publication, the passages shared between theLiezi and other texts, the linguistic evidence for a late dating of the text, and remarks on the structure ofthe document as a whole. There is also Graham's earlier (1959) short article  The Dialogue betweenYang Ju 楊朱  and Chyntzyy 禽子 , which is narrowly focused on one pericope of the Liezi (7.11). Itoffers a concise summary of the Yangist argument as it is found in this chapter, as well as commentingon the nature of the “Yang Zhu” chapter of the Liezi as a whole. Without question the most important work in English on the Liezi produced since Graham's isJune Won Seo's 徐晙源 doctoral dissertation, “The Liezi 列子: The Vision of the World Interpreted bya Forged Text” (2000). As suggested in the title, Seo agrees with the claim made by Yang and Grahamthat the Liezi is not best read as an authentically pre-Qin text – a claim which is presented alongsideabundant textual evidence. Where Seo's work is truly remarkable is in its reading of the  Liezi as a10 Wong (1995), p. 20. It is, however, explicitly listed as a translation on the Shambala Publications website. 11 Wong's book does not included a bibliography, and there is no information about which edition of the text forms thebasis of her paraphrase. 12 For example, Wong suggests that “Most historians now agree that [Lie Yukou] was born around 400 BCE, about twohundred years after Lao-tzu and Confucius”. (Wong (1995), p. 3). Though this is a logical deduction, based on thebiographical information about Liezi the person, it is far from accurate to suggest that this is a majority view amonghistorians; indeed, a significant number of scholars are reluctant to suggest that Liezi was a person at all. This is coveredespecially well in Seo (2015), pp. 449-450.13 Though I have made use of both the original 1960 printing and the 1990 reprint found in the collection  Studies ofChinese Philosophy and Philosophical Literature,  for reference purposes I will  make reference to the more widelyavailable 1990 version.6medieval text in dialogue with the xuanxue 玄學14 discourse of the Wei-Jin period. The arguments hemakes are crucial to my own discussion of the Liezi and xuanxue in Chapter Four.The only collection of scholarly essays on the Liezi in English is Ronnie Littlejohn and JefferyDippman's  Riding the Wind with Liezi (2011),  which is  comprised of  twelve essays  related to  thetextual history, philosophical content, and practicable elements of the text. Certainly there are withinthis volume very good reflections and investigations into the nature of the text; however, some chaptersdeal only tangentially with the text, and appear to me more as an afterthought to an unrelated bookproject than to a targeted study of the Liezi. Second, almost all essays ignore the important academicwork done on the  Liezi since Graham's study and translation in the 1960's – both the abundance oftextual work done in Chinese, as well as the June Won Seo's groundbreaking dissertation. Despite thesedrawbacks the work is valuable and does offer innovative and thoughtful perspectives on the text.The  best  recent  summary  of  the  Liezi in  English  is  that  by  June  Won  Seo  for  the  DaoCompanion to Daoist Philosophy (2015), which neatly summarizes his dissertation discussed above. Asin that work, Seo comes down firmly on the sceptical side of the argument, suggesting that “[the Liezi]is highly likely a fourth-century forgery concocted by a person within [the Zhang] family.”15Structure of this DissertationThis work will be divided into two parts. In the first part, Chapters One and Two, I explore boththe motivation for the creation of the Liezi in the early medieval period and review evidence both forand against this relatively late date. In the second part, Chapters Three to Five, I interpret the teachingsof the Liezi text in the context of the intellectual climate of the early medieval period. I believe each ofthese chapters  offers  something of  value to  the study of ancient  and medieval  Chinese thought  ingeneral and the study of the Liezi in particular, as will be summarized below.14 The term “xuanxue”, usually translated indefensibly as “Neo-Daoism”, will be discussed at length in Chapter Four. 15 Seo (2015), p. 450.7In Chapter One I will make use of recent publications on the topic of Masters Texts (zishu 子書).I will recruit the arguments found here to lay out a broad history of this textual category, with specialreference to their elevation in status over time. I argue that this elevation in status fostered a reluctanceto produce new Masters Texts. In an era in which great authority was invested in a Masters Text, but inwhich the outright production of a new Masters Text was discouraged, a compiled Masters Text like theLiezi makes a great deal of sense.Chapter Two reviews the best evidence to date on the question of the authenticity of the Liezi. Iwill  review  both  sides  of  the  argument  in  detail  in  an  effort  to  reach  a  tentative  but  reasonableconclusion, in order to expand Liezi research beyond the narrow scope of authentication studies. To doso, we need to situate the text historically. I will ultimately argue the  Liezi is a compilation createdaround the middle of the fourth century of the common era, and speaks to third and fourth centurymatters of a philosophical and religious nature. Though I believe the compiler of this text was workingin the third century, it is likely that earlier material was used and adapted for this purpose. It is likely(and unsurprising) that some of this material can be legitimately traced to the early Warring Statesperiod.  I  remain  unconvinced that  the  main  commentator  of  the  Liezi,  Zhang Zhan  張湛 ,  is  thecompiler of this text.The third chapter of this dissertation is a survey of the contents of the Liezi. I outline both thebroad general themes of the text as well as the narrower intellectual and spiritual claims of individualchapters. This chapter also offers some detail on the literary qualities and style of the text. The chapterends with a classification of the text's chapters into two parts, which I call the “Core Chapters” (i.e.,Chapters One to Five) and the “Appended Chapters” (i.e., Chapters Six to Eight). I argue this divisionmainly based on qualitative differences of the material presented, with some additional reference toquantitative differences and historical textual evidence. 8Chapter Four situates the  Liezi in the context of  xuanxue thought as found in the writings ofWang Bi 王弼 (226 – 249 CE), Guo Xiang 郭象 (died ~310 CE), Ruan Ji 阮籍 (210 – 263 CE), and XiKang 嵇康  (223 – 262 CE). This work builds upon the pioneering efforts of others, especially Seo'sgroundbreaking dissertation. I argue that the Liezi is unique in the manner in which it sheds light onand addresses the intellectual controversies found in xuanxue: as a medieval document masqueradingas  a  much  earlier  one,  it  can  unselfconsciously  and with  presumed  authority  make  claims  wherecommentaries and writings outside the collection of recognized Masters Texts must resort to precedentand careful argumentation. In the fifth chapter I explore the question of the Liezi's relationship to the Buddhist thought ofthe period. I review the most common claims of “Buddhist material” in the  Liezi, and suggest thatwhile  a small  minority are  compelling,  most of these claims are not  convincing.  Following this,  Icompare the notions of Nonbeing (wu 無 ) and emptiness (xu 虛 ) as they are found in the  Liezi topopular notions of Buddhism present in China during the third and fourth centuries of the common era.I conclude that though both the  Liezi and Buddhist accounts employ similar language their ultimategoals are divergent. I argue that the Liezi is best understood as an alternative to Buddhist thought, ratherthan as strongly influenced by such notions.This dissertation concludes by presenting the claims of the Liezi clearly in their context. It was afabricated  textual  object,  though certainly containing  material  from an earlier  era,  designed  to  beinserted  into  xuanxue debates  over  the  primacy of  Nonbeing  (wu 無 ).  Despite  the  great  deal  ofscholarly emphasis  placed on Buddhist  material  in  the  text,  we discover  that  the  Liezi position  isparallel to Buddhist responses developing at that time, rather than springing from them. Taken togetherwith the positions of Wang Bi, Guo Xiang, and Buddhist  xuanxue participants, the  Liezi presents amore complete picture of the ontological crisis in intellectual circles in early Medieval China.  9Notes on Conventions and DefinitionsFinally,  I  conclude  this  introduction  with  some  words  on  the  citation  conventions  anddefinitions  employed in this  work.  I  have  endeavoured to  cite  as  source texts  reliable  and widelyavailable publications. I have taken as my primary source for the  Liezi Yang Bojun's critical edition.Because there is no standard citation for the Liezi, I have provided an appendix to facilitate readers inlocating particular passages from this dissertation in Graham's translation and two important criticaleditions. All translations, unless otherwise noted, are my own. For the purposes of this dissertation I haveemphasized literal fidelity over elegant rendering in my translation work. As a result some translationsmay seem slightly unnatural, though remain comprehensible. In this project I have placed highest valueon philosophical precision, and in the translations aim to exchange literary sophistication for clarity.Finally,  this  work  will  occasionally  use  the  English  words  “Daoist”  and  “Confucian”  inreference to certain textual lineages and communities. When used, I strive to address the difficulties ofsuch  terminology,  and  employ  them  only  as  a  convenience.  Generally,  when  I  employ  the  term“Daoist”, I refer principally to texts categorized in the Hanshu yiwenzhi 漢書藝文志 under the rubricdaojia 道家; likewise, reference to “Confucian” works are those categorized in the Yiwenzhi as rujia 儒家 . Exceptions and clarifications will be noted as appropriate. I will resist the use of the translation“Neo-Daoism” for xuanxue, and offer reasons for doing so in Chapter Four. 10CHAPTER ONE: THE LIEZI AND MASTERS TEXTSIn the fifth century CE, Liu Xie 劉勰 (465-522) wrote in his Wenxin diaolong 文心雕龍 aboutthe value of the various masters (zhuzi 諸子):若乃湯之問棘1,云蚊睫有雷霆之聲;惠施對梁王,云蝸角有伏尸之戰;列子有移山跨海之談,淮南有傾天折地之說,此踳駁之類也。是以世疾諸子,混同2虛誕。3When Tang questioned Ji4, Ji told him that in the eyelash of a mosquito therewas the sound of a thunderclap; Hui Shi, in responding to the King of Liang,said that on the horns of a snail there were the fallen corpses of a battle.5 TheLiezi has  a  tale  about  moving a  mountain  and stepping across  the  sea,6 theHuainanzi has  talk  about  Heaven  collapsing  and  Earth  breaking7 –  these[writings] are of a contrary and contradictory type. This is why our era deteststhe [writings] of the various masters – they are mixed with and permeated byempty deceptions.Liu's readings of these writings is decidedly negative, though he does praise other works in the samecategory, such as the  Mengzi and the  Xunzi. While his attitude may have been true for his time andplace, it does not hold true for much of the history of the writings of the masters, known to us as“Masters Texts”, or zishu 子書. We know that the writings of the masters were highly regarded, thoughperhaps rarely as highly regarded as the Classics (jing 經), and that these works had invested in them ahigh degree of authority.In this chapter, I intend to address two central questions. First, what is a Masters Text? Second,why does it matter if the Liezi is a Masters Text? To answer the first question, I will critically review1 Reading ce 棘 as ji 革, in accordance with the Liezi text.2 Reading dong 洞 for tong 同.3 Source text from Shih (1983), pp. 190-192.4 Liu here is referring to the first section of the Liezi chapter “Tang wen” 湯問; see Liezi 5:1 and 5:2. 5 This is in reference to a story about Hui Shi 惠施 (370-310 BCE) in the Zhuangzi chapter “Ze yang”, found in the MixedChapters.6 Liu may have in mind here  Liezi 5:3, where a determined man resolves to slowly remove a mountain. Various otherfloating mountains are mentioned in this section. 7 See Huainanzi Tianwen. This may be a reference to Major et. al. (2010), p 115.11the recent effort by Wiebke Denecke to produce a descriptive category for Masters Texts, as found inher The Dynamics of Masters Literature: Early Chinese Thought from Confucius to Han Feizi (2010;hereafter “Dynamics”). I then give a brief account of the creation of the category of Masters Text as itoccurs in the early bibliographical works of the Han period. Following this, I offer a condensed surveyof Masters Texts from the early Warring States period into the Han and Wei-Jin periods, demonstratingtheir rise in prestige, as well as the eventual complications that accompanied their production. Ultimately, I intend to show that Masters Texts had a special kind of intellectual power, but asthat power grew it  began to preclude the production of new Masters Texts. This is  to say that,  asMasters  Texts  were  invested  with  greater  authority,  their  special  status  was  a  barrier  for  furtherproduction.  Masters  Texts  were  hallowed  works  that  came  to  their  readers  from  antiquity;contemporary writers could not claim that degree of prestige for their own thought. One response tothis problem was to reclassify one's work: instead of explicitly producing a Masters Text, one couldproduce a work of a similar nature without claiming that title. Another response to this problem was tomanufacture a Masters Text with a false history and attribution. I will argue in Chapter Two that thissecond alternative was the motivation for the production of the Liezi. Thus is it pertinent that I demonstrate the manner in which the Liezi is embedded in the MastersText tradition, even if I believe its compilation took place centuries after the “golden age” of MastersText production in the Warring States. In doing so, I aim to go beyond the question of whether or notthe Liezi is an authentically pre-Han text (I will argue that it is not), and also address the question as towhy the Liezi was compiled in the early medieval period. In this chapter I suggest that the documentwas  intended  to  adopt  the  authority  invested  in  the  Masters  Text  category  to  lend  credence  to  aparticular philosophical position. Having established the impetus and evidence for the creation of thetext in the early medieval period in my first and second chapter, I will investigate the philosophical12position in its context beginning in Chapter Three. 1.1 Masters Texts And AuthorityDenecke offers a list of five attributes, deduced from her extensive survey of seven MastersTexts from the Warring States period, that are characteristic of the category. This is the goal of herproject in Dynamics, as she sums up in her conclusion: “What happens if we scrape away as much aspossible of the disciplinary and conceptual overlay that has accrued on the surface of Masters Texts, theinterpretive branches of the last half millennium since the Jesuit mission?”8 The five characteristics thatbecome apparent to her offer for us one useful framework by which we can categorize and interpretMasters Texts that goes beyond bibliographic lists.The first characteristic Denecke suggests is the focus of these texts on their respective masterfigures.9 The master10 can be present and explicit, as Mengzi is in the Mengzi text, or a shadowy andimplied teacher, as Laozi is in the Laozi. Though most Masters Texts give us a third person view of theMaster – for example, we are often eavesdropping on Kongzi and his disciples in the  Lunyu – it isimportant to note that the text of the Laozi never directly mentioning Laozi does not exclude it from thecategory of  Masters  Text.  Denecke's  second characteristic  is  related  to  the  first  –  the  presence  ofinterlocutors  with  which  the  masters  can  debate,  teach,  or  otherwise  engage.  Though  she  givesexamples of how this characteristic is expressed in the  Mozi,  Zhuangzi, and other texts, Denecke issilent  regarding  the  Laozi on  this  point  of  interlocutors  –  a  silence  that  will  be  crucial  for  ourinterpretation below.Denecke's  third characteristic  of Masters Texts is  that these works make assertions that areinextricably bound up in their style. She cites the systematic arguments of the Mozi and the apophatic8 Denecke (2010), 326. This is a reiteration of Denecke's program as outlined in the introduction to Dynamics.9 Denecke (2010), 326. The description of the five characteristics that follow this note are all derived from Denecke'sconclusion, specifically the section that comprises pages 326 to 329.10 The vast majority of “masters” found in early and medieval Chinese writing, when identified either as historical orfictitious persons, are male – to the point of exclusivity. For this reason when using a third person pronoun in referenceto a “master” I will generally use “he”; this is a function of the historical circumstances of the composition of these texts.13language of the Laozi (among others) as examples of this – the implication seemingly that the use ofsystematic thought in the  Mozi is tacit endorsement of systematic thinking, and that the paradoxicalclaims  of  the  Laozi betoken  a  more  fundamental  break  with  convention  in  thought.  The  fourthcharacteristic is the presence of “gestures of affiliation” found in Masters Texts that undergird theirclaims, at least internal to their respective traditions. The idea here appears to be the conscious andintentional self-identification with other people that is found in Masters Texts – take, for example, theconspicuous manner in which the  Mengzi and the  Xunzi take  Kongzi as their primary authority, andthereby ground their authority with him. Although Denecke does not emphasize the point here, I thinkit is also relevant to point out that this grounding of authority is not linked only to people, but also totexts – consider again the Mengzi and the Xunzi, and their use of the Shijing 詩經.11Finally, Denecke points to a “deep structure” that fosters a sense of homogeneity in a particularMasters Text. This “deep structure” unites the disparate claims of a text and is the means by which wecan begin to describe the nature of thought found in these individual 'discursive spaces'. One exampleDenecke gives is the “trope of exemplification” in the Hanfeizi – this is the frequent use of anecdotes inpersuasive  writing,  which  Denecke  feels  betrays  Han  Fei's  paranoia  typified  in  his  reference  toprecedent (and consequent mitigation of criticism or punishment).12 This  concept  of  “deep  structure”  is  somewhat  nebulous,  and  so  I  will  turn  to  Denecke'sdiscussion of the Mengzi in an attempt to further clarify the notion. She suggests that an unstated butimportant idea in the text is that of “depth” – in time, in texts, and in the body. 13 In terms of depth oftime, Mengzi is credited with beginning the Confucian lineage in his claiming the role of Kongzi's11 Of course, this question is certainly linked back again to the grounding of authority in a person. If the author of aMasters Text takes the compiler or author of another text as a figure of authority – as many in Han understood Kongzi'srelationship to the Chunqiu – then the authority of the text can in a sense become an extension of the authority of thatperson. This does not weaken the authority of the text, but rather broadens the authority of the person.12 Denecke (2010), 328-329.13 Denecke (2010), 156. 14disciple.14 His  contribution  to  “depth”  in  texts  is  that  of  exegesis,  with  a  more  concentrated  andsustained focus on “classics” than that found in the Lunyu.15 Finally, by locating the boundary betweenthe inner workings of the human heart-mind and their external manifestations, Denecke suggests thatMengzi deepens the understanding of the body.16 These unstated and seemingly unrelated ideas are theaspects of  the “deep structure” which Denecke suggests  exist  in  all  members of  the Masters Textcategory. Taken together, these five characteristics offer us a useful set of criteria for defining MastersTexts outside of bibliographical listings. Having gotten clear about the criteria, the next crucial step istesting its limits. Below I aim to demonstrate that these five criteria are flexible enough to include agreater variety of work than the seven central texts discussed in Dynamics. Though Denecke does notsuggest that these seven texts are the only Masters Texts, if I am to argue that the Liezi is a MastersText it is useful to establish the bounds of such a category.***The imagined boundary that surrounds the category of “Masters Texts” must be permeable. TianXiaofei gives an account of this permeability in her assessment of medieval Masters Texts, noting thatthe category is more than simply a list of books with titles ending with the character zi 子.17 MichaelNylan also stresses the flexibility of titles in her discussion primarily on the meaning of the term jing經,18 noting that “Masters Texts” can often be suffixed with the characters shi 氏 or gong 公.19 If we areto take Denecke's idea of a “discursive space” seriously, then we must consider all texts that fit her14 Ibid.15 Ibid. 16 Ibid., 157. 17 Tian (2006), 466. 18 It is also useful to discuss what qualities separate a Masters Text from what comes to be known as a “Classic” or jing 經(here, not solely those texts which come to be known as the “Confucian” Classics). While some of the basic definingqualities of a Masters Text, as outlined by Denecke, are absent in the Classics, their often less explicit prescription ofnormative behaviour and greater emphasis on technical, historical, or aesthetic knowledge serve to exclude them fromthe category of Masters Texts.19 Nylan in Lagerway and Kalinowski (2009), 732.15descriptive criteria of Masters Texts, regardless of whether or not they categorized as such by laterbibliographers. After all, Denecke herself notes that our earliest bibliographic record, the Yiwenzhi ofthe Hanshu, lists the Lunyu in the jing 經 category, not that of Masters Texts.20 This does not preventher from categorizing the Lunyu as a Masters Text, and designating it the blueprint for later instances ofthe classification. As we shall see in the second part of this chapter, she is likely justified in doing so.The expanding boundary of  “Masters  Texts”,  if  based on Denecke's  concept  of  “discursivespace”, will necessarily include a variety of texts not considered in her Dynamics of Masters Literature,as well as works that lay beyond the historical scope of her book. Here I intend to address the generalquestion  of  archeological  texts,  with  a  focus  on  the  important  works  found  at  Mawangdui  andGuodian.21 Denecke does address at least one of these texts, discussing, for example, the Guodian textcommonly referred to as Xing zi ming chu 性自命出 in her sixth chapter, as well as explaining in herintroduction that the presence of these texts do not complicate her central thesis.22 In this she may becorrect,  though in  a  book review Guo Jue has  demonstrated  that,  for  any study of  early Chinesethought, the Guodian texts now offer us an abundance of insight that cannot not be lightly dismissed.23Guo points out that among the finds recovered at Guodian are preserved versions of the Laozi text, thereceived version of which Denecke takes as a major exemplar of the Masters Text category. I intend tofurther press the point by suggesting that the same “discursive space” Denecke sees operating in thereceived tradition is present in the texts at Guodian that had eventually dropped out of the record.As a representative example and test case of archeological texts that fit well in the classification20 Denecke (2010), 91. Denecke suspects that this was done so as not to denigrate the work (or Kongzi) by associating itwith texts (i.e., Masters Texts) considered lesser than the Classics, or jing 經. 21 For  a  general  summary of  the  texts  found  at  Mawangdui,  see  Yates  (1997)  and  Chang and Feng (1998).  For  anintroduction to the texts at Guodian, see Cook (2012). It is worth noting that the Mawangdui texts fall outside of thehistorical scope of Denecke's project, and this is the likely reason she does not address them. However, because of theirimportance in this chapter and discussion of the Liezi as a whole, I include mention of them here. 22 Denecke (2010), 28.23 Guo (2014), 246-247. 16of “Masters Text” I take the Wuxing 五行, or Five Processes.24 There are three reasons to do so. First,the nature of the discourse is mostly congruent with Denecke's five point description of a Masters Text;in this case, the  Wuxing closely resembles the “expository essay” format she associates most closelywith the  Xunzi.  Second,  I  select  the  Wuxing  because our knowledge of  the document is  relativelycomprehensive when compared to other archeological finds: having recovered the document at bothMawangdui and Guodian offers two versions of the text (and a commentary) to compare, and suggeststhat the Wuxing was read in both the late Warring States and the early Han. Finally, the Wuxing is mostinteresting because it appears to fit into the debates on human nature and counterfeit ru that we knowwere taking place during the Warring States, and may have been directly engaged by the  Xunzi in acritical way in the  Jiebi 解蔽  chapter. Fitting the  Wuxing into the discourse of the Warring Statesdemonstrates that the category of Masters Text, if understood as “discursive space”, is much broaderthan the bibliographic lists from which the term “Masters Text” is derived. I suspect the same holdstrue for the majority of texts found at Guodian, as well as the Huangdi sijing 黃帝四經 25 recoveredfrom Mawangdui.The second and third points endorsing the Wuxing text as relevant to an investigation of MastersTexts are relatively straightforward, but the way in which the Wuxing fit Denecke's five characteristicsof a Masters Text deserves further elaboration. I will take each of Denecke's points, described above,and indicate how each is can also reasonably be seen as characteristic of this archeological text, makinguse of both explicit and implicit claims in her conclusion. The point of this exercise is to test the limitof the Masters Text category, and demonstrate the variety of texts found under that rubric. On Denecke's first point, the  Wuxing appears somewhat lacking in the presence of a master.24 Both  the  Mawangdui  and  Guodian  versions  of  the  Wuxing have  been  treated  with  deep  and  thoughtful  analysis.Important studies include Csikszentmihalyi (2004), Holloway (2008), Meyer (2011), and Cook (2012). 25 I do not share the confidence of Chang and Feng (1998) that  the texts recovered from Mawangdui are indeed theHuangdi sijing listed in the Yiwenzhi bibliography, but choose to employ that designation out of convenience. 17However,  this  is  only  if  we  constrain  ourselves  to  see  Masters  Texts  as  third  party  records  ofinteractions between masters and interlocutors, as we find in the  Lunyu or the  Mengzi. If we insteadrecall that the Laozi takes the form of a collection of mostly first person teachings devoid of a namedspeaker, and see the association with a master (in this case, Laozi) as stemming perhaps from an oraltradition  or  popular  title,  we can  see  that  the  presence  of  a  master  in  a  Masters  Text  is  at  timesimplied.26 The same is likely true of the Wuxing: common attribution of the text is given to Kongzi'sgrandson Master Zisi (子思子), in large part following the condemnation of his teachings as aberrant inthe Xunzi. While this attribution can be plausibly questioned, the fact that the association with a masterfigure exists at all helps the Wuxing to meet the first criteria. Taking the Laozi as our model again, wefind the  Wuxing meeting the second criteria  laid out above:  as discussed above, Denecke's  silenceregarding the dearth of interlocutors in the  Laozi seems to suggest that at the very least this will notexclude it from the category of Masters Texts, and perhaps is simply a shifting of the of interlocutor inthe text to the role of the reader. Regardless of how much we are comfortable speculating about thissilence, what is clear is that a lack of explicit interlocutors for a master, common both to the Laozi andthe Wuxing, is not enough to warrant the rejection of either from the designation of Masters Text.The third characteristic outlined by Denecke was use of stylistic features to make what she calls“intellectual claims”.27 In the case of the Wuxing, the stylistic elements are readily apparent to all whoread it: the text is replete with repetition and sorites (or “chain arguments”). In the set of definitions ofvirtues that opens the text, we find four of these five virtues described in identical terms – the argumenthere is that these four are, in the  Wuxing text, hierarchically equivalent. This argument is not madeexplicitly, but embedded in the format and style of the writing. So too are the use of sorites in the text,26 The accretion of aphorisms that eventually made up what we call the Laozi, and its subsequent association with a personreferred to as Laozi, is a complex matter worth much more deliberation than I can offer in the present project. Theprecise details of this process are not pertinent to our purposes here – what is pertinent is that the process did take place.27 Denecke (2010), 327. I adopt this terminology here and throughout. 18which recall (or perhaps prefigure) the expanding circle of influence we find in the Daxue 大學 chapterof the  Liji 禮記 .  The implicit claim here is that the elements of the sorites chain enters into beingfollows the establishment of that which comes before it – so that the final result of the chain is theinevitable result of the initial element that initiates the chain. Again, the intellectual claim follows style.The fourth mark of a Masters Text as described above is the “gesture of affiliation”. Simply bytaking the Wuxing's reference to poetry found in the Shijing as such gestures can we group it with textssuch as the Mengzi and the Xunzi, in what may be best described as a radial category that takes the ideaof “Kongzi as master” as the central point around which these texts, at varying distances, orbit. Byquoting the Shijing as authoritative, the author of the Wuxing is affiliating with a mode of thought thatvalues not only precedent as set in antiquity, but a very specific precedent captured in the compilationand transmission of the Shijing. The use of  Shijing material signals to readers of the Wuxing that theauthor has invested time in this material, and consequently respect – an investment likely either sharedor desired by the intended reader.By Denecke's own account, the fifth characteristic is the most difficult to uncover: “It can be avertiginous enterprise to search for intersections between local ideas in the text and a deeper structureof the argumentative thrust of the text within the framework of the genre's basic rules, but such analysiscan clarify why these texts mattered to those who wrote and compiled them and why and how theymight  matter  to  us  today.”28 This  undertaking  is  burdensome,  as  Denecke  implies,  and  for  thisdiscussion of the Wuxing, serving as it does only as a prelude to the question of the Liezi, it must besufficient only to point the direction rather than draw the map. In Mark Csikszentmihalyi's cogent andthorough  analysis  of  the  Wuxing in  his  Material  Virtue,  he  suggests  one  primary impetus  for  theproduction and circulation of the text was to combat a growing suspicion of the authenticity of  rupractice – that is,  the increasingly common criticism that  ru practitioners could put on an external28 Denecke (2010), 329.19display  of  sincerity  that  would  mask  ulterior  motives  and  a  potential  inner  insincerity.29Csikszentmihalyi  suggests  that  the  Wuxing text  does  its  part  to  combat  these  criticisms  by layingemphasis on internal motivation as true virtue and the physical manifestations of this true virtue in thehuman body.30 Framed in the style of trope identification, one way we may describe the Wuxing is aspossessing the “trope of seeking authenticity” – here evident in the instructions for generating andrecognizing authentic ru virtues.This brief digression on a text that has little direct bearing on the study of the  Liezi ought toserve as an illustration of how inclusive the category of Masters Text can be. This should not devaluethe category, but instead enrich it, by demonstrating the variety it can contain. Yet the category does notbecome all-inclusive – there remain spaces outside that of the category of Masters Texts. Now that wehave tested some of the boundaries of the idea of Masters Texts we can go on to trace the generaloutline of the term below.***This survey chapter is ultimately not intended to be a thorough or exhaustive evaluation of theindividual texts considered below. Instead, I have two main goals for this chapter, both selected inservice  to  the  project  of  the  paper  as  a  whole:  first,  I  hope  to  trace  clearly  the  contours  of  the“discursive space” of the Masters Text category, as described by Denecke, from its inception in theearly Warring States to the late Jin period, which we will take as our terminus ad quem for the Liezi.Drawing on Denecke's work on Masters Texts, as well as that of Michael Puett,  Tian Xiaofei, andothers, I hope to weave together many threads in order to give a brief but robust account of the kinds ofassertions and claims surrounded the notions of “Masters” and “Masters Texts” up to the production ofthe Liezi. 29 Unable to explore these arguments in great detail here, I refer the reader to Csikszentmihalyi (2004), especially thediscussion of criticisms of the ru that find their origin outside the ru social group in Chapter One, pages 32 ff.30 For the discussion of internal motivations, see Csikszentmihalyi (2004) pages 70 ff., and for a discussion of the physicalmanifestations of virtue, see Csikszentmihalyi (2004), pages 77 ff.20In  my  selection  of  texts  for  this  survey,  I  will  not  confine  myself  to  Han  or  post-Hanbibliographic lists, though these lists will be important in serving as starting points. At each historicalstage investigated I intend to examine not only representative examples of Masters Texts, but also textsthat occupy the same discursive space as Masters Texts while falling outside that bibliographic rubric(if they had indeed been catalogued in bibliographies at all).  These will include encyclopedic texts(such as the  Lüshi Chuqiu 呂氏春秋 ) and texts written in imitation of Masters Texts (such as YangXiong's Fayan 法言), among others to be discussed in this chapter.The second main goal of the present chapter is to evaluate the texts and movements discussedbelow primarily on the ways that they relate to the Liezi. It is my hope that this will lay a groundworkfor later chapters and undergird the discussion of the Liezi 's use of earlier material, and anticipate theways in which this work borrowed from – and was divergent from – earlier thought. My ultimate aim isto explain the role  the  Liezi had as  a  Masters  Texts  produced in the early medieval  period – thisnecessarily requires an examination and explanation of the role of Masters Texts up to and includingthe Wei-Jin period.1.1.1 The Creation of Categories: Sima Tan, Liu Xiang, and Ban GuThe final chapter of the Lie zhuan 列傳 section of the Record of History, entitled Taishigongzixu 太師公自序, includes a list and description the jia 家31 as categorized by Sima Qian's father SimaTan. This is one of the earliest attempts to systematically categorize and classify the various traditionsin the extant record, and the first give us the  jia designation. Sima Tan's list has six categories, eachwith a description of their strengths and weaknesses as he sees them. An exhaustive translation here is31 The term jia here is one worth examining in detail. Commonly in Classical Chinese it means 'family', and it is in thissense, extended and understood very broadly, that it is used here. In the context of its usage here it has often beentranslated as 'school', though this usage has recently been challenged after close scrutiny. For an in depth evaluation ofthe term and its history in Han historical writings, see Nylan and Csikszentmihalyi (2003). For my purposes here I willuse either 'lineage' or 'expert', as the situation requires, or the untranslated term itself, with the understanding that themeaning  and  usage  of  this  term was  never  fixed  and  often  laden  with  many meanings  for  all  the  periods  underconsideration.21not necessary; below I give part of his description of the daojia 道家 or “Daoist” lineage:32道家無為,又曰無不為,其實易行,其辭難知。其術以虛無為本,以因循為用。無成埶,無常形,故能究萬物之情。不為物先,不為物後,故能為萬物主。33The Way Experts34 engage in non-action,35 yet it is said there is nothing that isnot done; the reality is easy to practice, yet the words are difficult to understand.Their methods take emptiness and Nonbeing36 to be the root, and take beingadaptive and following37 to be the function.38  They lack a fixed stance39, andthey lack constant forms40 – and therefore are able to investigate the essence ofthe myriad things. They do not [put themselves] ahead of things, nor do they[put  themselves]  behind things,  and therefore are  able  to  be the lord of  themyriad things.The above quotation, brief as it is, illustrates the beginnings of the classification of Warring32 For a full translation, see Harold Roth and Sarah Queen's rendering in Sources of the Chinese Tradition, (1999), p. 280-281.33 Shiji (1973), vol. 10, p. 3292.34 “Way Experts” here translates daojia, more commonly translated as “Daoists”.35 “Non-action” is wuwei 無為 , a critically important term in texts that would eventually be grouped under the rubric ofdaojia or “Daoist”, such as the Laozi and the Zhuangzi. For a survey of the usage of the term in “Daoist” texts and theHuainanzi, see Liu (1991). For a comprehensive investigation of the term and related concepts as it was understood inthe  Warring  States  period,  see  Slingerland  (2003).  A  survey  of  the  term's  usage  in  the  Han  is  available  inCsikszentmihalyi (2006). 36 “Emptiness” xu 虛  and “Nonbeing” wu 無  are crucial terms that are first explored, albeit vaguely, in the Laozi. Theybecome significant in later periods, especially among the “xuanxue” thinkers of the post-Han period, and as such arecritically important concepts in the Liezi. For more on these ideas and the choice of translations here, see Chapters Fourand Five of this paper.37 “Being adaptive and following” renders  yinxun  因循 . This binome is unknown in both the  Laozi and the  Zhuangzi,which likely predate Sima Tan's composition here, and it is not found in the Liezi text. However, the term is present inboth the Wenzi 文子 as well as the Heshang gong 河上公  commentary to the Laozi. The dating of the Wenzi remainscontroversial, and the Heshang gong commentary is dated to the Latter Han.38 The pairing here of “root” ben 本 and “function” yong 用 is interesting. Typically “root” is paired with “branch” mo 末,while “function” follows “essence”  ti 體 . The root-branch dyad is found in both the  Lunyu (19.12) and the  Mengzi(6B1), but the essence-function dyad is probably first expressed in the work of Wang Bi 王弼, after the fall of the LatterHan and much later  than Sima Tan's  composition. While my translation here follows these standard translations –perhaps somewhat anachronistically – it is done to preserve the sense the parallelism intended in the original ClassicalChinese, as well as reflect the standard pairings as they are found in subsequent writings. 39 Reading shi 勢 for shi 埶.40 The phrasing wu cheng shi wu chang xing 無成埶無常形 seems to be unique in the received record, although some nearmatches exist. The nearest parallel I am able to locate is that found in the “Xu shi” 虛實 chapter of the Sunzi bingfa 孫子兵法 (better known as Master Sun's Art of War), in a passage extolling the virtues of water and the strategist's need toemulate it:  故兵無常勢水無常形  “Therefore in war there are no constant conditions; [just as in] water there is noconstant shape”. See source in Liu (1996), p. 5. For the context of the entire passage in English translation, see Ivanhoe(2011), p. 40. The usage in the Sunzi bingfa recalls both the example of water in the Laozi and the metaphorical usage inthe Mengzi. The Huainanzi's usage of the expression chang xing 常形 in the “Bing lue” 兵略 chapter (here found as wuchang xing shi 無常形勢 ) seems to suggest a more than incidental commonality with military terminology. (see He(2006), vol. 3, p. 1051) For the “Bing lue” chapter, see Major et al., (2010).22States modes of thought in the Former Han, a process that reflects the equally polemical attempts seenin the “Tian xia” 天下  chapter of the Zhuangzi and the “Fei shi'er zi” 非十二子 and “Jie bi” 解蔽chapters of the  Xunzi.41 While this classification scheme is based around methods rather than textualcollections, it does offer us the first set of jia, six in total: the yinyang jia 陰陽家 (“Yinyang Experts”),the rujia 儒家 (“Confucian Experts”),42 the mojia 墨家 (“Mohist Experts”), the fajia 法家 (“LawExperts”),43 the mingjia 名家 (“Naming Experts”), and the daojia 道家 (“Way Experts”). These six  jia are included in the ten bibliographic subcategories devised by Liu Xiang andincluded in his Bielu 別錄, the document that formed the basis for the Yiwenzhi 藝文志 chapter of theHanshu attributed to Ban Gu. Here, the six above  jia are grouped with four additional lineages, tocomplete the zhuzi shijia 諸子十家 or “various masters of the ten expert lineages”: the zonghengjia 從橫家 (“Vertical and Horizontal [Alliance] Experts”), the zajia 雜家 (“Miscellaneous Experts”),44 thenongjia 農家 (“Agricultural Experts”), and the xiaoshuojia 小說家 (“Lesser Explanation Experts”).Though the Yiwenzhi classification scheme does not promote the daojia above all others as Sima Tanhad done in his list of six jia, it does place the Masters Text category as a whole in an inferior positionto the Classics. As well, the nine jia worth discussing demonstrated increasing doctrinal divergence in achaotic world:諸子十家,其可觀者九家而已。皆起於王道既微,諸侯力政,時君世主,好惡殊方,是以九家之說蠭出並作,各引一端,崇其所善,以此馳說,取合諸侯。其言雖殊,辟猶水火,相滅亦相生也。仁之與義,敬之與和,相41 These antecedents to Sima Tan's enumeration of the jia are discussed at length in Chapter One of Denecke (2010), p.42ff. 42 Following, for simplicity, one standard rendering of the term ru as “Confucian”. 43 Like “Daoist” and “Confucian”, the rendering  “Legalist” for  fajia is burdened with outdated assumptions and is notemployed here, in favour of the more awkward and context dependant “Law Experts”. For a good discussion of theproblems with the designation “Legalist”, see Goldin (2011). 44 The  designation  “miscellaneous”  for  za is  used  instead  of  the  equally  useful  translations  “mixed”,  “eclectic”,  or“syncretic”. For a good overview of the understanding and translation of za as a category of Masters Text, especially asit applies to the Huananzi, see Major, et al. (2010), pp. 28-29. 23反而皆相成也。45Regarding the various masters and ten jia: of those that ought to be investigated,there are actually only nine  jia.46 In all cases they arose from the Way of theKings having become indistinct, and the feudal lords used force to engage ingovernment.  Present  rulers  and  contemporary  lords  like  or  dislike  differentmethods, and therefore the theories of the nine  jia directly47 out of this werecreated together;  each [ruler]  drew up one end, and honoured that which hefound good;  by this  they spread the theories,  and gathered the feudal  lords.Although their doctrines are different, they regulated one another as though theywere water and fire – mutually extinguishing one another, and then mutuallygenerating  one another.  [Like]  benevolence's  relationship  with  righteousness,[like] respect's relationship with harmony, they mutually opposed one another –yet in all cases they completed one another. Though divergent and corrupt, the  Yiwenzhi recognizes the mutual influence and dependencethese various types of expert learning have upon one another, and stresses that they merely capture partof a more complete picture of proper conduct and reality. This summary of the Masters Texts in theYiwenzi goes on:仲尼有言:「禮失而求諸野。」方今去聖久遠,道術缺廢,無所更索,彼九家者,不猶瘉於野乎?48Zhong Ni49 had a saying: “If ritual is lost, then seek it out in the wilderness”.50Now at this time, our displacement from the Sages is remote and distant, and themethod of the Way51 is defective and cast aside - there is nowhere that they canbe sought again. These nine jia - are they not still better52 than the wilderness?45 Hanshu (1975), vol. 2, p. 1746.46 The implication here is that the tenth kind of jia, the xiaoshuojia, is not of equal value with the other nine. 47 Reading dun 頓 for feng 蠭.48 Hanshu (1975), vol. 2, p. 1746.49 That is, Kongzi or Confucius.50 As far as I can discern this is the first occurrence of this phrase attributed to Kongzi. The Hanshu also has: 夫禮失求之於野,古文不猶愈於野乎? “As for ritual's being lost and seeking it in the wilderness – are the old texts not stillbetter than the wilderness?” (Hanshu (1975), vol. 2, p. 1971). Like Ban Gu's selection from the Yiwenzhi, the point hereis that though deficient, some writings – here the the “Old Texts” discussed later in this chapter – are still better thanlooking for the ritual outside of mainstream scholarly culture.51 I take dao shu 道術 very broadly, so as to interpret in the context of Ban Gu's quoting of Kongzi and avoid anachronisticconnotations of longevity practices. Compare Ban Gu's approach here with that of the opening of the “Tian xia” chapterof the Zhuangzi: 古之所謂道術者,果惡乎在?曰:无乎不在。 :  “That which was called the method of the Way – inthe end, where is it? I say: there is nowhere that it is not present.” (source in Guo (2004), vol. 3, p. 1065). Ban Gu seesthe method of Way as lost, while the author of the “Tian xia” chapter sees it everywhere. They both agree, however, thatthe various divergent schools capture part of the truth – though when ranking the various masters, the  Tianxia authorsuggests Zhuang Zhou was most successful.52 Reading yu 瘉 as yu 愈 based on the parallel passage quoted in note 50 of this chapter, above.24Ban Gu does suggest that the various jia – which, we must keep in mind, appear in the Yiwenziprimarily as an enumeration of Masters Texts, with only a minimal description of their methods – as thebest way to repair the lost Way. This suggestion follows a thorough explanation of their partiality andresulting deficiency, elaborated on earlier in the Yiwenzhi in an explanation of the death of Kongzi andhis disciples: 昔仲尼沒而微言絕,七十子喪而大義乖。  “Formerly, when Zhong Ni perished, subtlewords53 were cut short; after the seventy disciples were lost great righteousness was disordered”.54 ForBan  Gu,  adopting  the  format  of  Liu  Xiang  and Liu  Xin,  the  Masters  Text  class  of  texts  is  bothnecessary and limited. Finally, it is important to note that it is in the  Yiwenzhi bibliography that theLiezi is first documented in eight pian. 1.1.2 Elements of Masters TextsHow does the Liezi fit the five point criteria of a Masters Text offered by Denecke? Below I willdemonstrate the ways in which the  Liezi fits  Denecke's criteria, before proceeding on to a concisesummary of Masters Texts and their reception up to the early medieval period. The question of theLiezi's  inclusion in the category of Masters Texts is relevant because I  believe that it  will  help usexplain why the the text was created. We can trace the trajectory of these texts in history. We see thetexts invested with a greater authority that precludes their continued creation, and understanding thisprocess explains why a new Masters Text, like the  Liezi, would be compiled in the medieval period.The latter part of this chapter will provide evidence for this trajectory, but below I first endeavour toprovide evidence for the inclusion of the Liezi in this textual category. Denecke's characteristics, andhow they apply to the Liezi, are useful for this purpose. 53 “Subtle words” here translates wei yan, a critically important hermeneutic term in Han thought. Its meaning stems fromthe understanding that Kongzi had composed the Chunqiu 春秋, and that his true meanings were, in a sense, in code, inaccordance with his station as scholar, not king. These meanings were decoded later in the three primary commentarialworks on the Chunqiu – the Zuozhuan 左傳, the Gongyang zhuan 公羊傳, and the Guliang zhuan 穀梁傳, as well asfurther interpretations on these texts during the Han period.54 Hanshu (1975), vol. 2, p. 1701.25The first mark identified as distinctive of a Masters Text is the presence of a master. This isperhaps the most unambiguous of the five criteria, and evaluating the  Liezi in light of it is likewiseunambiguous. Like many of the Warring States texts discussed below, such as the  Zhuangzi or theXunzi, the Liezi takes its title from its ostensible intellectual hero, Lie Yukou, and suffixes his surnamewith the honourific “master” or  zi 子 . Like many of the other texts, the content of the  Liezi oftenfollows the titular character's interactions with his interlocutors – though, like the Zhuangzi, the Liezitext also relates many narratives of Kongzi or other characters, without mention of the eponymousmaster. Unlike most Masters Texts, however, Lie Yukou is often not presented as the ultimate authority,and at times takes direction from another master, such as Huzi 壺子.55 The phenomenon of an imperfectmaster is not entirely unique to the Liezi: see, for example, a story about Zhuang Zhou's errors as canbe found in chapter twenty of the Zhuangzi, “Shan mu” or “Mountain Tree”.56 Moreover, even chaptertitles in the Liezi are named after masters – in the text we find the chapters “Zhong Ni” 仲尼  (betterknown as Kongzi) and “Yang Zhu” 楊朱, where these masters serve as central characters (though notalways heroes).Interlocutors for the master are also found in the Liezi, and this satisfies the second of the fiverequirements. These interlocutors, like in other Masters Texts, are thinkers, specialists, or rulers. Theyserve as the foil to the master and his teachings, whether or not that master is Lie Yukou. As we find inthe Zhuangzi, Kongzi often takes instruction from unlikely teachers in the Liezi, like boat handlers andcicada catchers.57 As described above, Lie Yukou can himself act as interlocutor to the master as well. The third characteristic is the presence of intellectual claims made through style. Here the Lieziagain closely resembles the Zhuangzi, in its use of “scenes of instruction”, a format that Denecke sees55 For example, see Liezi 2.13, which is mostly identical to a story of Liezi, Huzi, and the shaman Ji Xian as found in theZhuangzi (see Guo (2004), vol. 1, p. 297ff.). 56 For an in-depth reading of this section, which includes comparative work between English interpretations, see Philip J.Ivanhoe's “Zhuangzi's Conversion Experience” (1991).57 See for example Liezi 2.8 and 2.10, which parallel the Zhuangzi in part, but with some changes in meaning.26as originating in lore and writing surrounding Kongzi, which will be discussed in greater detail below.58As well, we do find occasional echoes of the Laozi's first person perspective, seemingly intended as afirst person master passing on their instruction through the written medium to a reader.59 Finally, theeighth chapter of the  Liezi is comprised of anecdotes, at times relating snippets of narrative withoutexplicit master figures. This use of anecdote recalls similar usage in the Hanfeizi, Lüshi Chunqiu, andHuainanzi, although this chapter lacks the explicit unifying context usually presented by these otherMasters Texts.The  fourth  characteristic  is  the  making  of  “gestures  of  affiliation”  that  Denecke  findsthroughout Masters Texts. Here again the Liezi fits the pattern well. Like the Zhuangzi, the Liezi makesfrequent reference to Lao Dan and uses him as a mouthpiece for his own agenda. This is not the sameas Han Fei's use of the Laozi text in the Jie Lao 解老 and Yu Lao 喻老 chapters of that work – whereHan Fei is commenting on the text of the  Laozi document (as he had it), the  Zhuangzi and the  Liezionly occasionally make reference to that first person text, citing it the same way the Xunzi cites an odeof the Shijing. Both the Zhuangzi and the Liezi appear more concerned with the appropriation of LaoDan's prestige than the faithful interpretation of the text attributed to him. This is merely one exampleof a gesture of affiliation; another more interesting gesture is the use of the term wu 無 or Nonbeing, aterm that in its employment suggests an intentional signal of alignment with the thought expressed inwhat has been deemed xuanxue 玄學 – this point will be taken up in Chapters Three and Four.Finally, the Liezi must possess a “deep structure” that unites the different intellectual claims itmakes. As discussed above, a program of employing anecdotes to distance oneself from controversialclaims is one deep structure that runs through the  Hanfeizi; as I have argued, the “trope of seekingauthenticity” permeates the text of the Wuxing. Denecke also identifies a “trope of interiority” in the58 Denecke (2010), 90-91.59 One example of this style is Liezi 1.2, which is a description of the process of cosmogony without any explicit masterfigures mentioned.27Mengzi,  which  is  meant  to  bolster  his  claims  of  affiliation  with  Kongzi.60 This  deep  structure  isunstated, and links together seemingly unrelated arguments and concerns. In my investigation of themajor themes of the Liezi, I believe we may point to a “trope of non-existence” or “absence” which ismanifest  in  the  Liezi's  frequent  emphasis  on  the  priority  of  Nonbeing,  Lie  Yukou's  valuing  of“emptiness” (xu 虛 ), and the acceptance of the death of all things as inevitable and even desirable.These concepts are explored more fully in Chapters Three, Four, and Five.1.2 Masters Texts in the Warring States PeriodWe now turn to our survey of Masters Texts at the point of their inception in the early WarringStates period; the subsequent part  of this  chapter will  carry on this  survey into the Han and earlymedieval China. The accounts of texts provided below will be brief, as only the elements of the workthat relate to Masters Text in general and the Liezi in particular will be emphasized. For most of theWarring States works I will again draw on the observations and terminology used by Denecke in herDynamics of Masters Literature; in discussing subsequent works I will draw upon other sources, butwith the intention of extending the understanding of the category to periods later than the end of theWarring States. The evaluations of relevant texts will focus only on the characteristics pertinent to theirstatus  as  Masters  Texts  and  their  relationship  to  the  Liezi;  more  nuanced  and  comprehensivediscussions of the arguments and worldviews of these texts are not in short  supply,  and interestedreaders are encouraged to seek this manner of analysis in works appropriate to that aim.61Any work intending to deal with thought in early China must in some way address the person ofKongzi, especially as he has been recorded in the variety of texts that are attributed to him or thedisciples that encountered his teachings first hand. These attributions need not be verifiable in order to60 Denecke (2010), 328.61 Some important classic works that  deal  with these texts in a comprehensive way include Fung Yu-lan's  History ofChinese  Philosophy (1937),  Benjamin  Schwartz's  World  of  Thought  in  Ancient  China (1985),  and  A.C.  Graham'sDisputers of the Tao (1989). An excellent more recent contribution to the large list of survey works on early Chinesethought is Bryan van Norden's Introduction to Classical Chinese Philosophy (2011). 28be valuable; it matters as much what subsequent generations believe Kongzi taught as does it what thehistorical Kongzi may have said. The created persona of Kongzi as teacher and transmitter forms for allsubsequent Masters Texts, without exception, a common point of reference when promoting their ownideals. Whether Kongzi is taken as an exemplar upon which one models oneself, an ultimate authorityto which one adopts as master, an intellectual opponent against which one must inveigh, or a puppetinvoked in parody, his presence, either explicit or implicit, is found in all Masters Texts. Every mastermust deal with, in some way, the original master. Denecke acknowledges that the  Lunyu, as we have it,  probably did not take shape until themiddle of the second century B.C.E., many centuries after the words it purports to record were spoken,though she suggests it preserves part of an older collection of “Confucius lore”.62 Of the variety ofaspects of Masters Text that Denecke traces to the Lunyu the most relevant to a reading of the Liezi isthe creation of the “scene of instruction” – according to Denecke, these are not accurate renderings ofthe master's words and deeds, but likely fictionalized accounts of Kongzi's teaching designed to projecta particular image.63 Regardless of their historicity, these “scenes of instruction” become the prototypefor  later  Masters  Texts,  which  will  either  adopt  and  adapt  the  trope  or  reject  it  in  favour  of  analternative means of asserting authority and conveying claims. The “scene of instruction”, we will see,is the primary rhetorical strategy found in much of the Liezi.The next  major  example  in  Masters  Texts  that  Denecke examines  is  the  Mozi.  Any readerfamiliar  with  the  Mozi knows  that  the  primary  target  of  work  are  the  ru generally  and  Kongzispecifically, criticizing core ru values such as the large scale state funded production of music and theexpression of grief through elaborate, costly, and prolonged funerals for one's parents. In the  Mozi,itself a composite text, entwining together writings from competing Mohist lineages dating from after62 Denecke (2010), p. 90. For a thorough account of the Lunyu in the Han, see “Confucius and the Analects in the Han”,Csikszentmihalyi (2002).63 Denecke (2010), pp. 96-98.29the death of the movement's founder, we find a shift away from what Denecke has described as the“charismatic” master's interaction with disciples and intellectual opponents to a more “depersonalized”form of writing, resembling an essay more than an account of instruction.64 She calls this mode ofdiscourse the “scene of construction”, a kind of “predictable discursive machine”65 that eliminates theneed for a charismatic teacher by using repetition and formal arguments to make assertions. Kongzi's detractors certainly were not limited only to the Mohists, and moreover the thinkerssubsequent to Kongzi were not exclusively antagonistic towards him and his teaching. Indeed, it wouldbe surprising if a single thinker could provoke such sustained animosity as we find in the Mozi – that isto say, attacks on Kongzi were not merely directed toward a dead philosopher, but were also attacks onthe body of followers of Kongzi – both those that presumably knew him in life and those that adoptedhis teachings after his death. In the thinker Meng Ke  孟軻  (372-289 BCE) we find the latter – anitinerant scholar, much in the style of Kongzi, that would take up the  ru mantle in the promotion ofsocial values and the defence of the deceased master against the spreading philosophical diseases ofMohist and Yangist66 thought. Mengzi – the honorific by which Meng Ke would come to be hailed –became the subject of the Masters Text of the same name. This text is similar to the Lunyu, in that itrecords in narrative form Mengzi's travels and travails, in which he sets out to educate state leaders,warning them against their misguided efforts and intentions, or engages in sophisticated philosophicaldebate with his likewise intellectual peers. Where  the  Mengzi differs  significantly  from both  the  Lunyu and  the  Liezi is  in  its  more64 Denecke (2010), pp. 131, 136.65 Denecke (2010), p. 135.66 “Yangist” here refers to the thought of Yang Zhu 楊朱 (4th century BCE?), a thinker of whom we have no real surviving text; his impact on the intellectual climate of the Warring States is known primarily through his detractors, such as Mengzi. According to A. C. Graham, some “Yangist” texts exist, written by the latter day followers of Yang Zhu – examples appear in the Zhuangzi and the Lüshi Chuqiu (see Graham's Disputers of the Tao, p. 55). Those familiar with the Liezi will be aware that the seventh chapter of that text is named for Yang Zhu, and many anecdotes contained withinmake explicit reference to him. The Liezi's relationship to the figure of Yang Zhu will be addressed in detail in Chapter Three of the present work.30sustained nature, either as a result of its composition or redaction. In claiming this I mean to say thatthe episodes67 of the Mengzi, at least internal to their respective chapters, seem to follow a somewhatchronological plan, most episodes building on those that prefigure them; both the Lunyu and Liezi arecomposed primarily of isolated episodes, perhaps related thematically,  but most regularly involvingdrastic changes in scene, time, and persons. While the Mengzi may have in time become acknowledged as the successor text to the Lunyu,during the Warring States (and indeed during the subsequent Han period) it shared that status. In asurvey of  Masters  Texts  it  shared  that  status  most  notably with  the  Xunzi,  a  collection  of  essaysattributed to the Warring States thinker Xun Kuang 荀况  (313-238 BCE). Xunzi, like Mengzi, tookhimself as purveyor of Kongzi's teachings, advocating his own interpretation of the Master's way anddefending Kongzi not only from the intellectual attacks of his detractors but also his wayward andineffectual supposed followers – among which he explicitly named Meng Ke. Stylistically, we see inthe  Xunzi a return to a kind of “first person” account of thought even stronger than the  Mozi. Gone,with  few exceptions,  are  the  narrative  elements  of  the  Lunyu or  the  Mengzi,  replaced  with  whatDenecke calls the “expository essay”,68 trading the “discursive capital”69 found in potentially (or likely)fictionalized  records  of  instruction  with  a  direct  argument,  resting  on  the  strength  of  rhetoricalpersuasion and logic rather than the credibility bundled with “scenes of instruction”.  This style ofMasters Text is  consistent with many writings  that can be traced to  the Han period,  including theLunheng 論衡 of Wang Chong 王充 (27-97 CE) or the Qianfu lun 潛夫論 of Wang Fu 王符 (85-162CE). The Liezi, however, contains little of this expository style.Dropping for now the thread that begins with Kongzi, we may pick up that which begins with67 By “episodes” here I mean to suggest the textual unit known in Classical Chinese as zhang 章. 68 Denecke (2010), p. 180. 69 Denecke (2010), p. 181.31Laozi – that is to say, we may turn our attention to the trajectory of Masters Texts that are to find theiralleged70 point of origin in the apophatic and nebulous sayings of the Laozi or Daodejing.71 While therecent archeological findings at Mawangdui, Guodian and the strips of the Beida collection cast doubton the historicity of the Laozi as a complete text recorded by Lao Dan, especially taken together withthe paucity of reliable bibliographic information from the pre-Han period, what cannot be questioned isthe importance of that collection of material, even when its contents expand or shift. This is not tostress only the importance of the intellectual claims made by the Laozi (in any of its forms) and theirimpact on thought in this or subsequent periods, but is also meant to raise the point that the very idea ofLaozi and the work he authored is equally critical in understanding the  Laozi's place in the scope ofMasters Texts. For, as discussed above, Denecke points out here that Laozi offers no explicit scenes ofinstruction. The text is “depopulated”,72 and what remains are terse statements relating to normativeprograms for the self or society. Without the explicit described presence of a master, as in the Lunyu orthe  Mengzi, the text grounds its authority in the implicit master-author – though, unlike the  Mozi orXunzi, the Laozi does not rest on a foundation of logical persuasion. One may suppose that the Laozipresupposes a community, and this in part may explain its creation and popularity – one wonders howthe text would fare unattached to a particular master figure. By tying the text to the figure of Lao Danthe promoters of the text can appropriate that prestige and combine it with the compelling, but oftenbaffling, language of the text.7370 The question of dating these texts – specifically, the Laozi and the Zhuangzi – is important, but significantly complicatedby their  composite  nature.  For  our  purposes  here,  it  is  sufficient  to  note  that  the  later  (i.e.,  Han  and  post-Han)bibliographic and hagiographic traditions place the Laozi, and its purported author Lao Dan, at the head of the tradition.In the investigation of Masters  Texts,  what  is  believed to be true of the texts is  of equal  importance with what  isverifiable of the texts. 71 For general scholarship, I find both titles to be adequate, though tend to prefer the title  Daodejing in order to clearlydistinguish the work from the person of Lao Dan; however, for the sake of consistency in a discussion of Masters Texts(zishu 子書), the designation Laozi will be preferred here.72 Denecke (2010), p. 213. What Denecke means here is that there are virtually no named interlocutors in the text, asidefrom an occasional first person (expressed as wo 我 or wu 吾), or reference to a generic shengren 聖人.73 For more on the history of the idea of Lao Dan, the person, and how he may have come to be associated with the Laozitext, readers are referred to A. C. Graham's “The Origins of the Legend of Lao Tan” in Studies in Chinese Philosophyand Philosophical Literature (1990), p. 111.32This is not to suggest that the language of the Laozi is devoid of meaning. However, because thewriting is couched in jargon and often opaque, the text is often bent to suit the interpretations of thosewho read it. As it is used in the  Hanfeizi, it becomes a manual for gaining political control; in theHeshang gong commentary we find a program of physical self cultivation; in the hands of the ThreeKingdoms period scholastic  prodigy Wang Bi the reader  is  given an interpretation that  focuses oncosmogony, cosmology, and ontology; and a catechism that elevates Laozi to the status of Lord Laoand takes Heaven as anthropomorphic deity interested in human affairs comprises the Xiang'er 想而commentary.74 This is to say nothing of the glut of modern, sometimes hackneyed, translations. TheMozi and the  Xunzi draw their authority from their lucid arguments, and the  Lunyu and the  Mengzidraw their authority from the charisma of their protagonists, as captured in writing; the  Laozi lacksthese elements, but instead offers powerful language that can be adapted to suit a commentator's agenda– as has been demonstrated above, this is often done with great success. As I will argue in Chapter Two,as similar appropriation of a master's authority is apparent in the creation of the Liezi. Of all the Masters Texts to be reviewed here, none matches the Liezi more closely in style andformat than the Zhuangzi, the purported writings and records of Zhuang Zhou 莊周  (369-286 BCE).Attribution of the text to one source is exceedingly dubious, as has been demonstrated by Graham, Liu,and others,75 yet most scholars do find a degree of coherency between chapters – for example, it isgenerally agreed that the first seven chapters of the work, known as the Inner Chapters (nei pian 內篇),can be attributed to one remarkable intellectual of the Warring States period. This author, and thosewho have had their essays appended to his, blend the format of scenes of instruction with a peculiarmutation  of  the  expository  essay.76 The  Zhuangzi's  scenes  of  instruction  are  a  burlesque:  the74 For a translation of the Hanfeizi commentaries, see Liao (1959); for the Heshang gong commentary, see Erkes (1958);for the Wang Bi version, see Wagner (2003); for the Xiang'er commentary, see Bokenkamp (1997).75 See Graham (1981), Liu (1984), and Roth (1991). 76 See discussion in Denecke (2010), especially pp. 238, 261 ff.33protagonists are now ostensible degenerates, and if Kongzi is permitted to teach, his lessons would beunrecognizable to the sage of the Lunyu. The expository essays are likewise parodies: the “Qi wu lun”齊物論  chapter  mixes  humour  with  argument  in  its  assessment  of  language,  in  which  the  authorlampoons his own thesis and method –  “Now I have said something, but I do not know whether what Ihave said really has a meaning or does not really have a meaning!”77 A third stylistic form that is employed by the Zhuangzi authors is the use of anecdote or parable,taking the narrative format  of  the scene of instruction but  eliminating the explicit  master  figure.78Consider for example the story that opens the  Zhuangzi,  the account of the transformations of thecreature known as either the kun 鯤 fish or the peng 鵬 bird; though this tale is used for instruction thatis delivered to us through a narrator, it  replaces the emphasis on premises and definitions found inearlier  expository formats with rich and colourful imagery.  The master is still  present, but he nowcomes to us as a story teller, not as an element of the narrative. This use of parable is also a crucialelement of the style and strategy employed in the Liezi. The final Masters Text that Denecke discusses at length is the treatise purported to be from thebrush  of  a  member  of  the  aristocracy  of  the  state  of  Han  韓 ,  Hanfeizi.  The  Hanshu  yiwenzhibibliography lists the work as Hanzi 韓子 in fifty-five juan among the fajia or “Law Experts” category.While  the  designation of  fajia is  anachronistic  –  in  the  text  the author  at  no point  expresses  anyscholastic association, or knowledge of the idea of fajia – the collection of thinkers under this rubric isnot entirely without logic. Not all of the works listed here survive, but those that do remain in thereceived record – albeit often in a fragmentary form –  share a commonality in that the Hanfeizi adaptsmajor themes. From the Shangjun (shu) 商君(書) the Hanfeizi author adopts Shang Yang's 商鞅 (390-338 BCE)  emphasis  on  fa 法 ,  or  “laws”;  the  Hanfeizi's  emphasis  on  shu 術  (“techniques”)  has77 今我則已有謂矣,而未知吾所謂之其果有謂乎,其果无謂乎?(Guo (2004), vol. 1, p. 79)78 See discussion in Denecke (2010), pp. 268 ff. 34antecedents in what remains of Shen Buhai's 申不害 (385-337 BCE) writing the Shenzi 申子; the focuson shi 勢 or “power” can be traced to the Shenzi 慎子 of Shen Dao 慎到 (395-315 BCE).79 There is noevidence that any of these thinkers were intent on forming or participating in a scholastic lineage, fajiaor otherwise,  but  in  the light  of  their  common connection to  the  Hanfeizi their  collection  into ananachronistic category is not altogether surprising. It is not only these thinkers that the author of the Hanfeizi – perhaps Han Fei himself – foundintriguing. As indicated previously, the Hanfeizi makes extensive use of the Laozi,80 which is the centralsubject of two Hanfeizi chapters: the jielao 解老 or “Explaining Laozi” chapter and the yulao 喻老 or“Understanding  Laozi”  chapter.81 These  chapters  are  presented  in  a  commentary  style,  with  mostsections making a philosophical claim that is then buttressed with a quotation from the Laozi text. Thisis hardly the only style of the work, however; we find again expository essays, much in the style of theXunzi, grounded in a systematic explication of premises and definitions. As well,  the  Hanfeizi alsomakes use of parable-like narratives, often used as historical example or precedent – this usage isespecially relevant to the study of the  Liezi, for many of these anecdotes are found in the  Liezi text,particularly the eighth chapter “Shuo fu” 說符. An important text that Denecke does not include in her survey of Warring States Masters Textsis the Lüshi Chunqiu 呂氏春秋; neither should its relatively late date of composition (in Warring Statesterms) nor its title lacking the zi 子  suffix disqualify it from a survey of Masters Texts. On the latterpoint, it is salient that the Lüshi Chunqiu does indeed appear alongside all the Masters Texts discussedabove in the  Hanshu yiwenzhi bibliography “Masters” (zi 子 ) section, as a member of the  zajia or79 This list adapted from Denecke (2010), p. 281. 80 See discussion in Denecke (2010), pp. 288 ff. 81 Here I italicize “Laozi” in order to indicate the text rather than the person; it is apparent to me that the author of theHanfeizi was making use of a version of the Laozi text in writing these chapters. This version, as it is quoted, shows ahigh degree of textual congruence with the received and archeological versions of the text. 35“Miscellaneous Experts” classification. Furthermore, its style and content exhibit many similarities tothe aforementioned works. The Lüshi Chunqiu possesses many of the stylistic features of the Hanfeizi –that is, extended rhetorical passages undergirded by anecdote. Differences include the Lüshi Chunqiu'sseasonal manual, which comprises the ji 紀 or “Records” section of the text,82 as well as the addition ofa explanatory post-script (xuyi 序意), an innovation in the category of Masters Texts. Finally, the LüshiChunqiu, like the Hanfeizi, is a rich repository of parallels for the Liezi text.One  feature  of  the  Lüshi  Chunqiu that  distinguishes  it  from the  other  Masters  Texts  is  itscomposition – the Lüshi Chunqiu was a project explicitly mandated for and undertaken by a group ofthinkers, rather than an individual master. This is not to say that other Masters Texts are not the productof many hands and minds; indeed, all the texts discussed by Denecke are in some way composite, if notin  authorship  then  at  least  in  the  sense  of  their  collation  and being  edited  in  the  course  of  theirtransmission.  However,  the  Lüshi  Chunqiu was  from its  inception  was  consciously meant  to  be  acollaborative work, and this sense of collaboration was meant to ensure authority. Information about itscreation and subsequent promulgation is available in the biography of Lü Buwei, the patron of theproject, in the Shiji:當是時,魏有信陵君,楚有春申君,趙有平原君,齊有孟嘗君,皆下士喜賓客以相傾。呂不韋以秦之彊,羞不如,亦招致士,厚遇之,至食客三千人。是時諸侯多辯士,如荀卿之徒,著書布天下。呂不韋乃使其客人人著所聞,集論以為八覽、六論、十二紀,二十餘萬言。以為備天地萬物古今之事,號曰呂氏春秋。布咸陽市門,懸千金其上,延諸侯游士賓客有能增損一字者予千金。83At this time the state of Wei had Lord Xin Ling,84 the state of Chu had LordChun Shen, the state of Zhao had Lord Ping Yuan, and the state of Qi had Lord82 This section of the work is, to a significant extent, parallel to material found in the “Yue ling”  月令  (or “MonthlyCommands”)  chapter of the  Liji 禮記  or  Record of Ritual.  Perhaps not incidentally,  much of this material  is  alsoreproduced later in the Huainanzi chapter “Shi ze xun” 時則訓 (“Seasonal Models”).83 Shiji (1973), vol. 8, p. 2510.84 This and the subsequent three cases are all eminent leaders, comparable in station to Lü Buwei, in states outside Lü'shome state of Qin. 36Meng Chang. In all  cases they had subordinate to themselves scholars85 thatenjoyed acting as guests and outsmarting each other. Lü Buwei, because of thestrength of [his home state of] Qin, was ashamed that it was not like [the otherstates], [so he] also summoned scholars to come [to Qin], and entertaining themgenerously he brought in retainers numbering three thousand people.  At thistime the feudal lords were increasing the amount of disputing scholars, such asthe  disciples  of  Xun  Qing,86 and  their  written  works  spreading  out  underHeaven. So Lü Buwei had his retainers write what they heard, collecting theirdiscussions to make eight “Views”, six “Discussions”, and twelve “Records”;87these were in excess of two hundred thousand words. He took them as completein the matters of Heaven and Earth, the myriad things, and the past and present.He instructed that [the text] be called Mr. Lü's Spring and Autumn Annals. Hedisplayed it above the market gate in Xianyang,88 and suspended above it were athousand pieces  of  gold.  He invited  the  feudal  lords,  roaming scholars,  andguests – those among them able to add or subtract a single character, [to them hewould] give the thousand gold pieces.We are meant to infer that no thinker was able to add or subtract a character – at least to theliking of Lü Buwei. Mark Edward Lewis sees the production of the Lüshi Chunqiu as subverting theauthority of the earlier Masters Texts, which grounded their authority in a single charismatic figure –the  comprehensiveness  and inclusiveness  of  the  Lüshi  Chunqiu generates  an  alternative  source  ofauthority.89 In what Lewis has called the “Encyclopedic Epoch”,90 the charisma of the lone master istraded for the supremacy of the educated think-tank. Yet I would characterize this not as the subvertingof the category of Masters Texts, but instead evolving it; in subverting the authority of the texts whichpreceded the  Lüshi Chunqiu, the patron and his committee may surpass them in prestige yet remianwithin the bounds of the Masters Text style. This is to say that while Masters Texts that appeared priorto the Lüshi Chunqiu my be deemed inferior to it, our category of Masters Text, as defined by Denecke,remains  important  and  essentially  unchallenged.  The  success  of  this  venture,  grounded  in  groupauthorship, is mixed: echoes of this attitude are found resounding in the Han period with the creation of85 The term shi 士 is variously translated in English, and here I chosen “scholar”.86 Xun Qing here is Xun Kuang, or Xunzi.87 These three types of documents comprise the three divisions of the Lüshi Chunqiu. 88 Capital of the Qin state.89 Mark Edward Lewis, Writing and Authority in Early China (1999), p. 308. Michael Puett makes a similarly persuasiveargument in Puett (2007), p. 29ff. 90 Mark Edward Lewis (1999), p. 287.37the Huainanzi 淮南子 , yet we still find texts attributed to lone master-like figures, dutifully read andadded to subsequent bibliographic treatises under the rubric of “Masters Text”.1.3 Masters Texts After the Fall of QinThe texts counted as members of the Masters Text classification find their stylistic origin in theWarring States period, though their designation and classification as Masters Texts is, as best can bededuced from the received tradition, rooted in the Han period (206 BCE-220 CE) as can be found in thework of Sima Qian, Liu Xiang, and Ban Gu, discussed above. However, it would be remiss to implythat,  following the consolidation of empire under the Qin and the subsequent  rise of the Han, theproduction of works in  the Masters Text  category abruptly ended.  Whether  one adopts  the simplecriteria  of  inclusion  in  bibliographic  enumerations  of  Masters  Texts,  with  or  without  the  zi 子designation in the title, or the more nuanced and critical approach of identifying and describing featuresof the category, as Denecke undertakes in Dynamics, one cannot but admit that thinkers in the Han andsubsequent periods laboured as their Warring States counterparts did in contributing to the growingcorpus of Masters Texts. Below I intend to explore only some selected instances of this labour, as hasbeen preserved in the received tradition, taking both those specimens that exhibit typical features of thecategory as well as those that demonstrate innovation in and transformation of the style of MastersTexts. The review here is not exhaustive, but is absolutely necessary if we are to connect the inceptionof the category of Masters Texts during the Warring States period to its state in the early medievalperiod when the Liezi was compiled. By examining some examples of Masters Text in and followingthe Han we can more fully understand why the Liezi was compiled. Embedded in these instances of theMasters  Text  category  are  deliberate  reflections  on  the  notion  of  “Masters  Text”  itself,  whichfurthermore contribute to our understanding of the evolution of this category.Like  the  Lüshi  Chunqiu,  the  Huainanzi is  ostensibly the  product  of  a  wealthy patron  with38intellectual inclinations – in this case, Liu An 劉安  (179-122 BCE), King (wang 王) of the Huainanregion. And like the Lüshi Chunqiu, the  Huainanzi is listed among the “Miscellaneous Experts” as aMasters Text in the Hanshu yiwenzhi bibliography.91 The Shiji account of Lü Buwei's production of theLüshi Chunqiu evokes a sense of competitiveness; the patron vies against his counterparts in otherstates for the prestige associated with intellectual endeavour, and upon the text's completion he displaysit with a public challenge to all that would find it lacking. Liu An's production of the Huainanzi maynot have been motivated by public competition as much as it was a personal message – specifically, tohis nephew and ruler of the Han empire, Wudi (r. 141 – 87 BCE). Liu An would eventually lose his lifeas a result of accusations of intrigue against Wudi, though in reality it remains unclear whether theHuainanzi was a sincere attempt at guidance for the emperor or intended as a statement of intellectualand moral superiority on the part of Liu An. Like  the  Lüshi  Chunqiu before  it,  the  Huainanzi bears  the  marks  of  a  work  that  from itsbeginning was formulated to maximize its authority through comprehensiveness and careful design.There is a clear and regular method in the assigning of titles to chapters and a conspicuous attention toconsistency among the claims and arguments made in the text, encapsulated again in the forms ofexpository  essay  and  anecdote  we  have  seen  throughout  Warring  States.  This  deftness  in  literaryproduction is captured best in the closing “Yao Lüe” chapter of the work, which summarizes the goaland content of the Huainanzi. The chapter begins thus:夫作為書論者,所以紀綱道德,經緯人事,上考之天,下揆之地,中通諸理。雖未能抽引玄妙之中才,繁然足以觀終始矣。總要舉凡,而語不剖判純樸,靡散大宗,懼為人之惽惽然弗能知也,故多為之辭,博為之說。又恐人之離本就末也,故言道而不言事,則無以與世浮沉;言事而不言道,則無以與化遊息。9291 It must be noted that both a  Huainanzi Inner Chapters (Huainanzi nei 淮南子內 ) and a  Huainanzi Outer Chapters(Huainanzi wai 淮南子外) are listed in the yiwenzhi. It is the former that was preserved and serves as the topic here.92  He (2006), vol. 3, p. 1437.39As  for  the  creation  of  this  book  and  discussion:  It  is  a  means  to  giveorganizational structure93 to the Way and Virtue,94 serves as warp and weft forhuman affairs, verifying them above with Heaven, measuring them below withEarth, and in the centre making them consistent with the principle. Although[these writings are] not yet able to pull or draw out the core endowment of DarkMystery,95 they are amply sufficient to observe its ends and beginnings. If [thisbook] assembles the essentials and raises up the ordinary, and [its] words do notcut and discriminate the simple unhewn wood, and do not depart from the GreatAncestor96 -  I97 fear this  will  make people dimly unable to know them. Andtherefore  for  them I  have  made  many  words,  and  for  them I  have  spokenbroadly; yet I still fear people will reject the roots to go to the branches. So if Ispeak of the Way but do not speak of affairs, then they lack the means to floatand sink with the world; if I speak of affairs but do not speak of the Way, thenthey lack the means to wander and rest in transformations.Liu An's purpose in gathering scholars for the creation of a text to instruct the ruler is made clear notonly in the opening of the “Yao lüe” chapter, but is expressed consistently throughout this chapter andthe text as a whole. It is indicative of an emerging trend in period following the Warring States – a selfconscious emulation of those masters and Masters Texts of a bygone era. The six part classification ofjia found in the Shiji had not become current by the time Liu An had finished the Huainanzi, but evenin the early Han period we find texts that emulate the style and methods of certain Masters Texts93 “Organizational  structure” translates  jigang 紀綱 ,  extrapolating from the definition given in Kroll  (2015),  p.  189:“strands and mainstays, skeins and cables, > network, nexus, organization.”94 “Way and Virtue” here translates  dao 道  and  de 德 . These two terms are of course most closely associated with theLaozi, or  Daodejing, literally the  Classic of the Way and Virtue. The  Huainanzi as a whole, and especially the initialchapter “Yuan dao” 原道, draw extensively from the Laozi, and it is the only Masters Text quoted in the Huainanzi withattribution.95 The “Dark Mystery” or xuanmiao 玄妙 is particularly evocative of the close of the first chapter of the received versionof the Laozi: 玄之又玄,衆妙之門: “Darker and darker still, it is the gate of abundant mystery.” (Lou (2009), p. 2). 96 “Simple unhewn wood” is again a tacit reference to the  Laozi, where the idea of  pu 樸  or “unhewn wood” stands inmetaphorically for a person innocent of cultural learning and untrammelled by superfluous desires. See chapters 15, 19,28, 32, 37, and 57 of that text. The “Great Ancestor” (dazong 大宗) reminds one most readily of the Zhuangzi chapter“Da zong shi” 大宗師 , translated into English by Burton Watson as “The Great and Venerable Teacher” and by A. C.Graham as “The Teacher Who is the Ultimate Ancestor”. The expression dazong itself does not occur in the ZhuangziInner Chapters, and only once in the Outer Chapters – 夫明白於天地之德者,此之謂大本大宗,與天和者也: “TheVirtue of brightly illuminating Heaven and Earth, this is called the 'Great Root' and the 'Great Ancestor' - it harmonizeswith Heaven”. (Guo (2004), vol. 2, p. 458). The Zhuangzi is often quoted by the Huainanzi, though without attribution;while much of the former predates the latter, the date of the compilation of the Zhuangzi is still disputed, and may evenbe connected with the compilation of the Huainanzi. See Roth (1991). 97 I have chosen to translate the subject in the first person here. This final chapter of the Huainanzi seems to capture theintentions of Liu An in presenting this work to the throne, and though he did assemble a group of intellectuals to assist inproducing the document, it was his intention that brought about the Huainanzi when it otherwise would not have beencreated. For more on the unique style of the “Yao lüe” chapter in relation to the Huainanzi, as well as its possible oralrecitation, see Queen, Murray, and Meyer in Major et al. (2010), p. 841, 846. 40without explicitly stating an intention to do so.98 In explicitly drawing upon the works of Lao Dan, the composers of the Huainanzi drew on theprestige and cultural capital invested in the Laozi. Yet Lao Dan was not the only master to be emulatedin the early decades of the Han period – the original master, Kongzi, also receives that distinction,especially in the work of Yang Xiong 揚雄  (53 BCE-18 CE). While the idea that Yang's philosophictreatise the Fayan 法言  can be said to be purely an imitation of the Lunyu and collected Kongzi lorecan be rightly challenged,99 it is indisputable that this work makes use of the scenes of instruction tropethat is characteristic of important Masters Texts in general and the Lunyu in particular. Furthermore, itis relevant to our discussion of Masters Texts that Yang Xiong's Fayan is included in the Suishu jingji隋書經籍 Masters Text (zi 子) bibliography,100 compiled in 636 CE, where it is listed as Yangzi Fayan揚子法言, or Master Yang's Model Words.101 Yang Xiong's Fayan is not only salient to our discussion because of its unequivocal membershipin the Masters Text class, serving as it does as another signpost on the long road of Masters Texts98 It needs to be emphasized here that it is quite unlikely that texts like the Huainanzi are in imitation of the category ofMasters Texts as a whole, but likely merely only in imitation of particular members of the category. In the case of theHuainanzi,  likely models are found in the  Zhuangzi,  Hanfeizi,  and  Lüshi Chunqiu,  though the  Laozi is also greatlyimportant as a source text. It is the elements of the category of Masters Texts (as defined by Denecke) present in thesedocuments that serve as the most readily visible template for the  Huainanzi,  and not the characteristics of the  jingcategory of text, for example. It is in this way that the Huainanzi, like other texts discussed below, can be found withinthe trajectory of the Masters Texts category. 99 See Nylan (trans.) Exemplary Figures (2013), p. xii. Nylan notes that while the majority of the Fayan does resemble the Lunyu in style and content, Yang Xiong does innovate on the model. This is true, and seems to be a function of the difference of authorship between the Lunyu and the Fayan: the former is composite text compiled over time, likely representing different (and perhaps competing) factions of Kongzi's disciples, while the latter came from the brush of a single writer. Moreover, Yang Xiong originally came to the capital from his native home in present-day Sichuan on the strength of his prowess in fu 賦 composition – hence, it is unsurprising that his work would be distinguished by a more sophisticated sense of literary style.100 See  Suishu  (1973), vol. 13, p. 998. Yang Xiong is listed as somewhat important in the  Hanshu yiwenzhi, includingreference to his philological work, his poetry, and an entry under the rujia 儒家 Masters section in thirty-eight pian; yet Icannot locate an unambiguous reference to the  Fayan in this bibliography,  though it was extant at  the time of thecompilation of the Hanshu. 101 I note here as well that while the title Fayan is Yang Xiong's own, the designation of “Yangzi” is not an honour Yangwould give himself – at least, not explicitly. This is true also of his  Taixuan jing 太玄經 , which he entitled simplyTaixuan, or Great Mystery; the addition of jing to the title was a later appellation, appended after Yang Xiong's death,granted in order to show the reverence with which the text was read. 41originating in the Warring States period and leading into the Six Dynasties, but also because it is in thetext of the Fayan that we find some of the earliest conscious deliberation on the question of mastersand Masters Texts.  This signals a maturing interest  in the questions around the nature and role ofmasters  and  the  writings  that  are  associated  with  them,  deeper  than  the  simple  refutation  orclassification of others existing in the Masters Text category. Yang goes beyond the terse criticismsXunzi offers of opposing masters at the end of the Warring States period, and instead grants praise orblame more in the style of the Lunyu, in response to unnamed interlocutors. His acclaim of luminariesin the ru tradition, known to him through their transmitted texts, is decisive: 或問「孟子知言之要,知德之奧」。曰:「非茍知之,亦允蹈之。」或曰:「子小諸子,孟子非諸子乎?」曰:「諸子者,以其知異於孔子也。孟子異乎?不異。」或曰:「孫卿非數家之書,侻也;至於子思、孟軻,詭哉!」曰:「吾於孫卿與,102見同門而異戶也,惟聖人為不異。」103Someone  asked,  “[Did]  Mengzi  understand  the  essentials  of  doctrine,  andunderstand the profundities of Virtue?” I [Yang Xiong] replied, “Not only did heunderstand them – indeed, he truly put them into practice (literally: walked thatpath).” Someone said, “You belittle the various masters – was Mengzi not [oneof] the various masters?” I replied, “As for the various masters – one takes theirunderstanding in how they differ from Kongzi: did Mengzi differ? He did notdiffer.” Someone said, “Sun Qing104 was against the writings of the numerouslineages, and this was appropriate; [but] in coming to Zi Si and Meng Ke, hewas wrong!” I replied, “My attitude towards Sun Qing: I see him as being of thesame gate105 but a different door; only the shengren does not diverge.”Of course not every master receives praise from Yang Xiong. Others, known to him throughtheir own Masters texts, undergo measured criticism:102 Here I punctuate the passage differently than both the Wang (1987) edition, p. 499, and Michael Nylan (2013), p. 208,both of which read “吾於孫卿,與見...”. I do so in harmony with another passage in the Fayan, from the “Wen dao”問道 chapter (Wang (1987), vol. 1, p. 114): 曰:「吾於天與,見無為之為矣!」:  “[I] say: As for my attitude towardHeaven, I see it as the acting of wuwei”.103 Wang (1987), vol. 2, p. 498. 104 That is, Xun Kuang or Xunzi. The unnamed interlocutor here likely has in mind the criticism of Zi Si and Mengzi foundin the “Fei shi'er zi”  非十二子  chapter of the Xunzi. Yang seems to suggest that both Mengzi and Xunzi have somelegitimate claim to adherence to the lineage of  Kongzi;  the text  here does not resolve the conflict  inherent in  theantagonism we find in the Xunzi towards Mengzi. 105 That is, Xunzi was also a disciple of Kongzi.42或曰:「莊周有取乎?」曰:「少欲。」「鄒衍有取乎?」曰:「自持。至周罔君臣之義,衍無知於天地之間,雖鄰不覿也。」106Someone  said,  “Does  Zhuang  Zhou  have  that  which  can  be  taken  [asworthwhile]?” I replied, “Reducing desire”.107 “Does Zou Yan108 have that whichcan be taken [as instruction]?” I  replied,  “Self  control.  [Yet]  coming to [thesubject of Zhuang] Zhou, he is wrong on the question of duty between lords andsubjects; and [Zou] Yan lacks knowledge about the space between Heaven andEarth. [So] even if they were next door, I would not visit them.”Yang Xiong's discussions of the various pre-Qin masters also includes Laozi, Hanfeizi, Shen Buhai,Yang Zhu, Mozi, Lü Buwei, and Yanzi, as well as important Han era thinkers such as Lu Jia, DongZhongshu, Liu An, and Sima Qian. There are few intellectuals in the ages before Yang Xiong wrote thatescape his notice, whether it be for praise or blame. In the Fayan we have a calculated and purposefulsurvey of the various masters, and by extension, their texts.109Yang's highly comprehensive and critical model is reflected in the Latter Han period in thewritings of Wang Chong, the Lunheng 論衡, or “Balanced Discussions”.110 This expansive collection ofessays  far exceeds the size of Yang Xiong's  relatively compact  Fayan,  and in style is  much morereminiscent of the essay format of the Xunzi than the Lunyu, though the question and answer format isadopted through out. While not bearing the honourific title zi in its title, the Lunheng, like the Fayan, isincluded in the Suishu jingji bibliography as a Masters Text, in the Miscellaneous Experts category.Like Yang Xiong in the Fayan, Wang Chong evaluates the full gamut of thinkers up to his time,106 Wang (1987), vol. 1, p. 134.107 Zhuang Zhou is Zhuangzi.  The injunction to reduce desire is perhaps more immediately redolent of the  Laozi;  forexample, we find at the close of Chapter 19: 見素抱樸,少私寡欲: “Appearing undyed and embracing an unhewnstate, one reduces self-interest and makes few their desires”. (Lou (2009), p. 45)108 Zou Yan (305-240 BCE) is another master of the Warring States period, associated with what in the  Shiji would beknown as the Yinyang Experts (yinyangjia 陰陽家); he is also associated with the Jixia Academy. A text attributed tohim,  listed  in  the  Hanshu yiwenzhi  as  the  Zouzi  鄒子 , is  lost.  Fragments  attributed  to  Zou Yan,  or  biographicalinformation about him contained in works such as the Shiji, suggest his theories were concerned with and influential incosmological thought related to yinyang and wuxing 五行. 109 Other noteworthy and similar projects include the “Tian xia” chapter of the Zhuangzi and the “Fei shi'er zi” chapter ofthe Xunzi. 110 Denecke also offers a reflection on Wang Chong's perspective on “Classics” and “Masters”; see Denecke (2010), pp. 78ff. 43including the various masters of the Warring States and Former Han, including even Yang Xiong in hisevaluations of their merits and deficiencies. Modern interpreters of the Lunheng have picked up on twoimportant features characteristic of the text that are pertinent in an evaluation of Masters Texts and theirreception  leading  into  the  Three  Kingdoms  Period  and  beyond:  that  of  Wang  Chong's  rhetoricalmethods, and that of his precise classification of texts. Many of the masters that precede Wang Chongare  subjected  to  his  methods  and  classifications,  revealing  an  even  more  systematic  approach  todiscussions of written works in general – and Masters Texts in particular – than in earlier writings suchas the Shiji or Fayan.  The  content  of  the  Lunheng is  relentlessly polemical.  Wang Chong takes  on not  only thesophisticated arguments of the Masters Texts discussed so far – he in fact devotes whole chapters toKongzi,  Hanfeizi,  and Mengzi111 –  but  he also critically investigates popular  belief  as held by thepopulace  at  large  and  preserved  in  the  classics.  It  is  best  not  to  conflate  Wang  Chong'sargumentativeness with a Western, post-Enlightenment style rationalism, however; as modern criticshave pointed out, the Lunheng never explicitly lays out a rationalist agenda, but rather uses all mannerof arguments to combat what Wang sees as the ills of his era.112 In his marshaling of premises hiswriting resembles the  Xunzi and the  Hanfeizi, and in this way he carries on that tradition of MastersTexts.Equally  germane  to  the  subject  at  hand  is  Wang  Chong's  classification  of  texts  into  thecategories. In the “Dui zuo” 對作 chapter of the Lunheng, he makes the following assertions in defenceof his decision to write:111 See the chapters “Wen Kong” 問孔, “Fei Han” 非韓, and “Ci Meng” 刺孟, respectively.112 Mark Csikszentmihalyi, for example, accurately describes Wang Chong's belief system as a “materialistic application ofthe correspondence systems existing in the late Han” (Csikszentmihalyi, 2006, p. 93). Michael Puett astutely notes WangChong's uncritical use of arguments in his criticism of other views; see Puett (2005), p. 273, 280. Despite a lack ofclearly stated criterion of evaluation, the Lunheng does at times display a consistent use of some techniques of enquirythat suggest an implicitly applied gauge of veracity in the thought of Wang Chong; for a valuable evaluation of WangChong's use of the terms wen 問 and nan 難 as critical method in the Lunheng, see McLeod (2007). 44或曰:聖人作,賢者述,以賢而作者,非也。論衡、政務,可謂作者。曰:非作也,亦非述也,論也。論者、述之次也。五經之興,可謂作矣。太史公書、劉子政序、班叔皮傳,可謂述矣。桓山君新論,鄒伯奇檢論,可謂論矣。今觀論衡、政務,桓、鄒之二論也,非所謂作也。113Someone  said,  “The  shengren creates,  the  worthies  transmit,  [but]  to  haveworthies create – this is wrong. The Lunheng and the Zhengwu114 can be calledcreations.”  I say they are not  creations, and they are also not transmissions –they  are  discussions.  Discussions  are  inferior  to  transmissions.  Theestablishment of the Five Classics115 can be called a creation. The Book of theGrand Scribe,  Liu Zizheng's  Xu,  Ban Shupi's  Records,116 these can be calledtransmissions. Huan Shanjun's Xinlun, Zou Boqi's Jianlun,117 these can be calleddiscussions. Now, looking at the Lunheng and the Zhengwu, [they are like] Huanand Zou's two discussions; they are not what are called creations. Wang Chong is setting up new textual distinctions that build upon those laid out by Liu Xiang and BanGu in their respective works. Not only are jing distinguished as the highest form of writing – associatedwith the shengren – but the textual space inferior to  jing is further subdivided and made hierarchicalwith “transmissions” and “discussions”. Masters texts are not specifically categorized, but we can inferthat their subordinate status in relation to  jing is congruent with that which we find in the  yiwenzhi;thus, they are consigned to the classifications of “transmissions” or “discussions”. By calling his worka mere “discussion” Wang Chong hopes to defend himself against critics;118 in doing so he adds furthernuance to the increasingly populated category of Masters Texts. That is to say that while Wang Chong,acting in intellectual self-defense, characterizes his work the Lunheng as a “discussion”, I believe it isquite correct to suggest that this text shares in the same “discursive space” as other Masters Texts thathave preceded it.119113 Huang (2006), vol. 4, p. 1180.114 Another work attributed to Wang Chong, now lost. 115 Wang Chong presumably has in mind here the Shangshu, Shijing, Yijing, Chunqiu, and Liji. 116 The Book of the Grand Scribe is Sima Qian's Shiji. Xu here likely refers to Liu Xiang's Xinxu 新序. Ban Shuqi is BanBiao (3-54 CE), who began the compilation of the Hanshu. 117 Huan Shanjun is Huan Tan 桓譚  (20 BCE-56 CE), a thinker very much admired by Wang Chong. A translation andstudy of the surviving fragments of his  Xinlun are available in Pokora (1975). I can find no reference to Zou Boqibeyond this comment in the Lunheng. 118 Michael Puett convincingly argues this is in fact Wang Chong participating in the tradition of denying sagehood inorder to claim legitimacy, as seen in the Lunyu, Mengzi, and Shiji; see Puett (2007), p. 39.119 As in my discussion of the Huainanzi above, I emphasize that the Lunheng participates in the category of Masters Text45Moving from the Han to the period of disunity, we enter the period in which the Liezi was likelycompiled.  As was true in  the Han,  writers  in  this  period are disinclined to  explicitly suggest  thatcompositions in their own name bear the gravitas of the Masters Texts attributed to the Warring States,though the assumption that their works ought to be read as such was often implied. Yang Xiong wascontent to name his work the  Fayan, and Wang Chong created a new humble category in which toensconce his writings. This was a function of the continued – and perhaps in some cases growing –appreciation of Masters Texts, alongside the well regarded  jing. Important works such as the  Lunyu,Mengzi,  Laozi, and Zhuangzi were being meticulously elaborated upon by the work of commentatorsintent on clarifying the meanings of these works rather than producing new Masters Texts outright. Thecategory of Masters Texts continued to carry a great deal of intellectual weight, but it was progressivelymore problematic to claim the authority of a master in one's original work. It was in this intellectualclimate that texts with suspect textual histories and dubious authenticity emerge, seemingly in answerto a new unstated question: How does one appropriate the intellectual power of a Masters Text withoutcomposing one? The response seems to be the creation of new Masters Texts with an ascribed ancientheritage. This is not to say that any of these texts are fully fabricated – the more plausible suggestion isthat existing material was adapted and revised to address the concerns of the compiler.  I will presentthe case for the Liezi being one of these texts in the subsequent chapter; here, I will review the case fora similar attribution of the texts Kongcongzi 孔叢子 and Kongzi jiayu 孔子家語 to the early medievalthinker Wang Su 王肅 (195-256 CE). The  preliminary  move  in  this  case  is  establishing  the  appropriateness  of  classifying  theKongcongzi and  Kongzi jiayu as Masters Texts. In content this is unproblematic; because the worksshare so many instructional narratives in common with other Masters Texts, such as the  Lunyu, it isnot by its  adhering to the parameters  of the category,  but by its  imitation of particular  works in the category thatpreceded it – especially those adopting the style of what Denecke calls the “expository essay”. 46permissible to categorize them among the masters on that point alone.  Furthermore,  the remainingcontent that is not parallel to Masters Texts – that which is either parallel to material in a source of adifferent  style,  or  lacking in  clear  parallels  at  all  –  often  fits  the  criteria  laid  out  by Denecke  inDynamics.120 In both texts the reader is presented with many scenes of instruction of the variety wehave come to know from Masters Texts of the Warring States period. The bibliographic evidence isminimal, but available: though the  Kongcongzi does not appear in the  Hanshu yiwenzhi, the  Kongzijiayu is listed in the table headed by the quintessential Masters Text, the Lunyu. Prior to the period of Wang Su's scholarly activity, which was the first half of the third centuryCE, the dominant school of interpretation for the jing and related writings was that of Zheng Xuan 鄭玄 (127-200 CE).121 Though the two scholars never met, Wang mounted strong and prolonged efforts indisplacing Zheng's school as the authoritative word on scholarship of the classics. A record of theseefforts have been partially preserved in the document Shengzheng lun 聖證論 (Discourse on Evidenceof the Sage), known to us now only through fragments preserved in encyclopedias and commentaries.122The  surviving  fragments  generally  follow a  similar  format,  which  in  style  resembles  a  debate:  aquotation from a text is introduced, and is followed (usually) by Zheng Xuan's interpretation; WangSu's rebuttal follows this, which is subsequently followed by commentary in support of either ZhengXuan or Wang Su; ultimately the matter is decided by a boshi 博士.123 Below is a relatively brief but120 The qualifier here is “often” - the Kongcongzi, for example, has its eleventh chapter an etymological work called theXiao erya 小爾雅, which resembles the Erya 爾雅 more than any Masters Text.121 I note here that Zheng Xuan is generally declared to be an “Old Text” (guwen 古文) scholar. For the sake of brevity andin consideration of space restrictions, I will not go into any detail about the complex nature of the Old Text and NewText debate.  Interested readers are encouraged to seek out important  work on this question, such as Nylan, “'ChinWen/Ku Wen' Controversy in Han Times” (1994) and van Ess, “The Apocryphal Texts of the Han Dynasty and the OldText/New Text Controversy” (1999). The question is important, but not immediately relevant to the discussion at hand.122 A useful collection of fragments from the Shengzheng lun is found in Ma Guohan's Yuhan shanfang ji yishu 玉函山房輯佚書.123 Hucker gives the translation “erudite” for boshi (entry 4647), as well as a detailed explanation of the meaning of thetitle over successive centuries. Here the meaning is probably related to the director of an academy with special trainingin the Classics. 47representative example of the Zheng Xuan and Wang Su portions of the exchange: 禮運。其居人也曰養。鄭元124曰:養當為義字之誤也,下之則為教令居人身為義,孝經說曰:養由人出。王肅曰:下云獲而弗食,食而弗肥。字宜曰養家語曰其居人曰養。125 [From the  Liji]  “Li  yun” chapter:  “In their126 dwelling in  a person,  they arecalled yang”.127 Zheng Xuan said, “Yang should be yi; the character is a mistake.In the text below it says that when what is taught and instructed is embodied in aperson, this is  yi.128 The  Xiaojing says 'yang comes from people'.”129 Wang Susaid, “In the text below it says 'to reap but not eat' and 'to eat but not becomefat'.130 The character is properly read as yang; the [Kongzi] jiayu says, 'dwellingin a person it is called yang'”.The precise details of this debate are not immediately relevant; what is crucial here is Wang Su's use ofthe Kongzi jiayu as an authoritative source to be employed against Zheng Xuan's interpretation of theLiji. Indeed, for much of the Shengzheng lun Wang Su cites the Kongzi jiayu as the final word againstthe majority of Zheng Xuan's assertions. Though the Kongcongzi is never explicitly quoted, there existparallels between that text and Wang Su's arguments in the Shengzheng lun as well.131 Wang  Su,  as  he  is  represented  in  the  Shengzheng  lun,  expects  the  reader  to  accept  these124 Zheng Yuan is a variation of Zheng Xuan's name.125 The source text here is adapted from an unpunctuated facsimile of a printed edition of the Yuhan shanfang ji yishu 玉函山房輯佚書  (n.d.) from the Beijing University Library (北京大學圖書館 ), vol. 72, p. 108, available digitally fromhttps://archive.org/details/02097638.cn.126 The context has dropped out of the quotation; “their” here refers to the rites, li 禮.127 Cf. Liji Liyun in Chen Jinsheng (1996) vol. 3. p. 430. 128 Zheng Xuan probably has in mind something like the following, found below in the same chapter: 故禮義也者,人之大端也,所以講信修睦而固人之肌膚之會,筋骸之束也: “Therefore, as for li and yi, they are the greatest extent ofpeople; they are the means by which speech is made trustworthy and cultivation is made harmonious; and it makes solidthe meeting points of a person's flesh and skin, the binding of muscle and bone. ” (Liji Liyun in Chen Jinsheng (1996),vol. 3, p. 430).129 While this particular phrase does not appear in the  Xiaojing, I suspect Zheng Xuan has in mind something like thefollowing: 用天之道,分地之利,謹身節用以養父母,此庶人之孝也: “Using the Way of Heaven and dividing upthe bounties of Earth, they are cautious of themselves and are restrained in their expenses; by this they nourish ( yang)their father and mother. This is the filial piety of the common person.” (孝經鄭注校證, p. 65). The point Zheng seems tobe making here is that  yang is an action that a person does, not a property that dwells (ju  居 ) in a person; however,because of the fragmentary nature of the Shengzheng lun, it is difficult to be conclusively certain. 130 Cf. Liji Liyun, in Chen JinSheng (1996), vol. 3. p. 432. 131 These parallels have been documented. Kramers (1950) lists all instances in the Shengzheng lun where Wang Su takesthe Kongzi jiayu as an authoritative text (pp. 138 – 154), and Ariel (1989) lists two examples from the Shengzheng lun ofparallels with the Kongcongzi (p. 64). 48references as authoritative; it is crucial to note, however, that both the Kongzi jiayu and the Kongcongziwere transmitted by Wang Su. In his preface to the  Kongzi jiayu Wang Su belittles Zheng Xuan andlaments that his own sincere attempts to promote the Way of Kongzi were not taken seriously. He thenexplains that it was his good fortune that a former student of his, a member of the Kong family line,brought to Wang Su the present manuscript of Kong family teachings. Wang then explains that thedocument resolves important debates – as it happens, the victor in these debates being Wang himself.132It is unsurprising that for many centuries scholars suspected Wang Su of producing the Kongzi jiayu asa means of gaining the upper hand in intellectual disputes. The  Kongcongzi is likewise unknown toscholars before its dissemination by Wang Su, and is first recorded in a bibliography in the Suishu.133 In his study of the text, Kramers ultimately takes the Kongzi jiayu to be authentic, in the sensethat it preserves a great deal of legitimately old material, with occasional interpolations by Wang Su.134This  conclusion  is  challenged  by  Ariel  in  his  study  of  the  Kongcongzi,  on  the  basis  of  textualanachronisms,  in  addition  to  the  highly  suspicious  congruency  of  the  text  and  Wang  Su's  ownexegetical stance.135 Ariel suggests the hypothesis that Wang Su created the Kongcongzi first, drawingon  many  extant  texts  but  adding  his  own  material,  for  the  purpose  of  challenging  the  dominantpositions  of Zheng Xuan;  this  effort  was insufficient  to  meet  his  goals,  so Wang Su subsequentlyproduced in a similar manner the  Kongzi jiayu, of which he emphasized a strong connection to theKong family line.136 Whether  it  is  Kramers'  occasional  interpolation  hypothesis  or  Ariel's  more  radical  editinghypothesis that is closer to the truth, it is noteworthy that both take Wang Su's relationship with the132 For a complete translation of this preface, see Kramers (1950), pp. 91-95. 133 On this point, see Ariel (1989), p. 12.134 Kramers (1950), p. 193 ff. Kramers' opinion remains the same in his summary of the work in Loewe (1993), p. 258.135 Ariel (1989), p. 68-69. I note, as Ariel does, that while the document is certainly suspect and likely edited during WangSu's lifetime, there is  no conclusive evidence that  Wang Su himself was the compiler – only strong circumstantialevidence. 136 Ibid., 69.49Kongcongzi and the Kongzi jiayu to be one of an at least partially dubious nature. In either scenario,Wang Su is suspected of altering materials in order to bolster his own scholarly position, by creating asource with a greater claim to authority than either he or his opponent possesses. In the subsequentchapters of the present work I propose that the Liezi is a similar case.1371.4 ConclusionWriting in the early half of the fourth century, Ge Hong 葛洪 (281-341 CE) offers the followingdescription of Masters Texts in his Baopuzi waipian 抱樸子外篇 chapter “Bai jia” 百家 (“the HundredSchools”): 子書披引玄曠,眇邈泓窈。總不測之源,揚無遺之流。變化不繫於規矩之方圓,旁通不淪於違正之邪徑。風格高嚴,重仞難盡。138Masters Texts unfurl and draw out the profound and vast, the minute and distant,the deep and hidden; they bind together sources that cannot be fathomed, theyraise up all139 streams. Changing and transforming, they are not bound by thesquare or round of the compass or the square; boundless and penetrating, theyare not submerged under unorthodox heretical paths. Their style and qualitiesare high and majestic, their weight and measure are difficult to exhaust. Unlike like the Wenxin diaolong quotation used to open this chapter, we find in Ge Hong's writing adeep appreciation for the profundity and sublime nature of Masters Texts. Where Liu Xie derided thewritings of the various masters as “contrary and contradictory”, the Baopuzi celebrates their freedomfrom such labels. In  writing  about  the  Hundred  Schools,  Ge  Hong  makes  little  mention  of  specific  masters,instead choosing to celebrate them all. In my survey of Masters Texts, I have been selective for the sakeof brevity, and recognize the many relevant works that have been excluded from the discussion here:137 It  is  also worth mentioning,  if  only in  passing,  that  Masters  Texts  were not  the only source of  textual  authorityspuriously created in this period. Indeed, discoveries of new versions of the classics or the presentation of apocryphaltexts in the Han perhaps betoken a similar phenomenon, thought it cannot be pursued in detail here. 138 Yang (1991), vol. 2, p. 442.139 “All” translates wuyi 無遺, literally “without omitting [any]”. 50Ge Hong's Baopuzi, for instance, but also works like Yanzi chunqiu 晏子春秋, Sunzi bingfa 孫子兵法,Guanzi 管子 , Xinyu 新語 , Qianfu lun 潛夫論 , and many others, both extant or lost. I have insteadselected  representative  examples  that  demonstrate  both  the  breadth  of  the  category  and  itstransformation over time. Applying the dual lenses of Denecke's descriptive category and the extantbibliographic records,  a  written form emerges  that  encompasses multiple  rhetorical  techniques  andliterary styles.  Their  goal  is  always  the  same:  the  authors  of  these  works  intend to  influence  andpersuade their audience, whether that audience is of the ruling class or of a more scholastic orientation. We have seen the genesis of the style in the recorded dialogues of Kongzi in the Lunyu, and findthis model elaborated on in the Mengzi and parodied in the Zhuangzi. The Mozi, in responding to theclaims of the ru, offers us a role as disciple rather than spectator, and this format is echoed in the Xunziand the Hanfeizi. In the Laozi we are asked to ponder succinct and cryptic aphorisms, often through theguidance of a later commentary. The Lüshi chunqiu and the Huainanzi reject the single master model infavour of a collection of thinkers, suggesting that theirs is a more comprehensive record of what is andwhat should be. Yang Xiong's Fayan here gives us the first inklings that the Masters Text model shouldbe emulated in the Han, after the age of the Hundred Schools, but it should be done so only in style. Aswell, we see a trend of critical reflection on the category of Masters Texts itself, which will developover time. The Lunheng of Wang Chong demonstrates a growing self consciousness of how writers inthe Han thought of and justified their own writing in the Latter Han period. Finally, in the Kongcongziand Kongzi jiayu, as they are offered to us by Wang Su, provide a model of how one may circumventthe obstacles to Masters Text production we see hinted at in the  Fayan and  Lunheng. These are thetemplates of the Masters Texts that have come down to us, and the brevity of this summary does not dojustice to the variety of their forms and the profundity of their arguments. In this unfolding of style and rhetoric we can find a place for the Liezi. By the time of the Wei-51Jin  period,  Masters  Texts  had  an  undeniable  authority  among  intellectuals.  Their  style  and  theircontents were well known, though their interpretations were contested. To ground one's claims in theauthority of a Masters Text was to ground one's claims in an esteemed source of recognized legitimacy.If one could not find such a source, it would be necessary to create one. 52CHAPTER TWO: DETERMINING THE DATE OF THE LIEZI世俗之人,多尊古而賤今,故為道者,必託之于神農、黃帝而後能入說。1The common people of this time greatly respect the ancient and consider lowly thecontemporary; and so one working towards the  Dao must base [their teachings onthat] of Shen Nong or Huangdi, and only then can they enter into the discussion.In 1960 Angus C. Graham published an article on the authenticity of the Liezi that would serveas a complement to his influential complete English translation of the text. Ultimately, he concludedthat the text as we now have it was not purely a product of the Warring States period, as it is purportedto be by its first commentator Zhang Zhan. Instead, he suggests it was more likely compiled in the thirdor fourth century CE.2 In claiming this, he was consciously rejecting the prevailing Western scholarshipon the question,3 and he instead adds his  voice to  those in  the long standing tradition of  Chinesescholarship that reject the authenticity of the Liezi as a Warring States document. In the five and a halfdecades  since  Angus  C.  Graham first  published  his  “Date  and  Composition  of  the  Liehtzyy”4 thesituation has changed significantly – now Western scholarship on the Liezi takes its production to be inthe  early  medieval  period,  as  established  by  Graham,  as  a  matter  of  fact  without  dispute. 5 Themovement to redeem the  Liezi as a product of the Warring States, the golden age for Masters Texts,now lives on predominately in the Chinese language scholarship. The half century since A. C. Grahamcombined the best scholarship in both traditions has not seen his manner of balanced approach fromeither side of the question.1 He (2006), vol. 3 p. 1355. 2 Graham (1990b), pp. 281-282. 3 Graham does note that opinions on the Liezi among his contemporaries were changing at the time of his writing, thoughit is his essay that ultimately transformed the dominant opinion in Western scholarship. See his brief outline of thepositions of Bernard Karlgren, Derk Bodde, Erik Zürcher, and Herrlee Creel in Graham (1990b), pp. 217-218.4 Though Graham's work was originally published in 1960 in Asia Major, I will generally be citing his republished editionin a collection of his essays, Studies in Chinese Philosophy and Philosophical Literature. In preparing this survey I haveconsulted both the 1960 version and the 1990 (listed as Graham 1990b) reprint. 5 As  a  representative  yet  important  example,  see  Roger  Ames'  justification  for  a  complete  adoption  of  Graham'sconclusions in his introduction to Littlejohn & Dippmann's  Riding the Wind with Liezi (2011), pp. 2-3. While I agreewith the points Graham makes and Ames' adoption of them, part of the purpose of the present chapter is to challengethose claims in light of the important scholarship that has transpired since the publication of Graham's conclusions, ifonly to again demonstrate their value.53The  present  chapter  aims  to  begin  to  remedy  this  problem.  Here,  I  hope  to  build  on  thegroundbreaking work of A. C. Graham in addressing the question of the nature of the Liezi, by weavingtogether the earnest and thoughtful work of the many scholars that have taken on the challenge ofunderstanding this text and its origins. This approach will blend the most insightful arguments bothbefore and after Graham's publication, and will serve as a comprehensive and contemporary guide tothe problem of the composition of the Liezi. Ultimately, in reviewing the evidence, it appears no trulysatisfactory challenge to Graham's late dating of the Liezi has appeared;6 however, the scholarship thathas intervened between his publication and the present does give us a richer and more useful picture ofthe Liezi's origins, and consequently, the dynamic thought it contains. Arriving  at  an  approximate  date  of  composition  for  the  Liezi –  and  indeed,  until  furtherevidence comes to light, only an approximate date can be offered – is not a trivial contribution to thestudy of the text in particular, or to the intellectual history of China in general. For the cultural contextin which a text is compiled serves as one lens the reader may use in developing their understanding ofthe intellectual underpinnings of the work. A reconstruction of the style and content of the MohistCanons gave us insight into the technical language that was lampooned in the Zhuangzi's “Qi wu lun”chapter,7 and the unearthing of the Guodian and Mawangdui Wuxing texts sharpens our insight into thecriticisms Xunzi lays out against Zisi and Mengzi in the “Fei shi'er zi” chapter. In this way, knowingthe norms and debates that were driving discourse in the time the Liezi entered the intellectual arenalets us know more clearly what its compiler hoped to achieve. Building on the previous chapter,  I intend to demonstrate that the  Liezi was created severalcenturies after it was purported to have been recorded, much in the same way Wang Su (or someone in6 In his summary of the textual  history of the  Liezi,  Barrett notes that Graham's work appears either unknown to orignored by Japanese and Chinese scholars  working after  its  publication. See Barrett  in Loewe (1993),  p.  301. Myresearch on the topic confirms that this remains generally the case even twenty years later after Barrett's summary; forone important exception, see Zheng Liangshu's 郑良樹 work examined in this chapter.7 Incidentally, we have A. C. Graham to thank for this pioneering work as well. See his Later Mohist Logic, Ethics, andScience. 54his intellectual circle) likely created the Kongcongzi and the Kongzi jiayu. The motive was, I believe,the same: the  Liezi was compiled in order to appropriate the prestige and intellectual authority of aMasters Text. The content of the Liezi, its cultural context, and the debates in which it was used are thesubjects  of  subsequent  chapters;  first,  we will  get  clear  about  the  evidence  for  the  Liezi's  date  ofcomposition.***Before laying out the evidence on the composition of the Liezi, it is prudent to be precisely clearabout certain terms and their assumed meanings; specifically, one must be exact in their understandingof the words “forgery”, “compile”, and “parallels” in the context of an investigation of authenticity. Inthe scope of the current project, “forgery” means only that the document that we now call the Liezi –that is, the one that generally bears Zhang Zhan's commentary – must have significant discrepancieswith the document listed in the  Hanshu Yiwenzhi as  Liezi in eight  juan, and with the document forwhich Liu Xiang wrote a preface.8 In light of all the evidence that will be presented below, what can besaid for certain is that material in the received Liezi is consequentially different from that that couldhave existed in the Warring States period. While we can identify much of this material with certainty,there remains as well much material that we can merely speculate upon. Finally, I note that I generallychoose to not use the terms “forgery” or “fake” in referring to the  Liezi; aside from the unfortunatepejorative connotations, I believe these negative terms obscure the goal of understanding the Liezi in itscontext, whatever that context may be.We must now more fully address the problem of the terms “compile” and “compiler”, whichuntil now I have been using cautiously and without elaboration. To say that the Liezi was “compiled” isvery much  a  calculated  suggestion;  even  in  light  of  the  evidence  that  suggests  the  Liezi reached8 It is not trivial to note here as well that these two documents – the one listed in the Hanshu Yiwenzhi and the one forwhich Liu Xiang wrote a  preface – may not be the same either.  The question remains open until  further evidencesurfaces.55something closely resembling the received version in the early medieval period, there is also muchevidence that a great deal of the material in the Liezi could be traced legitimately to the Warring Statesperiod. Whether this material could be traced to a Warring States version of the Liezi is unknown, andwill likely remain so unless archaeological discoveries offer new insights. The term “compiler” hererefers to an individual (or perhaps individuals) labouring in the early medieval period, and is usedbecause it is more neutral than “author” or “editor”. “Author” in this context seems to suggest a greatercreative role than I believe should be ascribed to the production of the Liezi. Though there is a greatdeal of creativity to be found in the work,  especially when viewed in contrast  with other relevantworks, the compiler clearly draws on a vast number of existing sources. In contrast, “editor” suggeststo my mind a too passive role in the production of the Liezi. The evidence suggests a compiler that didmore than just collate older material, but indeed injected into the text his own thought. I believe that“compiler” here captures both the collative and creative aspects we can see in the composition of theLiezi.Finally, we must be explicit about what it means to find “parallels to” or “sources for” the Liezi.Much of the authentication research to which the  Liezi has been subjected turns on the textual andconceptual similarities between the text and other works. The texts with which the  Liezi shares mostmaterial are the  Zhuangzi, the  Lüshi Chunqiu, and the  Huainanzi;  however, texts originating in theIndian  subcontinent  may  also  possess  similarities  to  the  Liezi,  and  these  non-trivial  similaritiessignificantly impact our understanding of the work. In an investigation of the evidence, however, it iscrucial to not begin with the assumption that the Liezi is a late document that draws upon earlier ones –for example, it is misguided to assume simply that because the Liezi and the Zhuangzi share importanttextual similarities (“parallels”) the Liezi can be traced to a later period chronologically later than thatin which the  Zhuangzi was compiled. This kind of assertion  only makes sense if supported by othercomplementary evidence. We must also always keep in mind that parallels never betray a direct line of56textual appropriation, and may instead be indicative of a common, now lost, source for both documents– a source that may be written or (perhaps) oral. Indeed, the many instances in which the Liezi containsparallels with multiple sources suggest that this is often more likely the case. Ultimately, it is not thepresence of a single textual parallel that makes the case for the compilation of the Liezi in the medievalperiod – even the striking ones that draw on reliably later sources, such as those that appear to be ofIndian origin – but instead the sum total of the many textual and conceptual parallels in the text. It is furthermore crucial to note here that this project is not an attempt to uncover a Liezi urtext.The  Liezi is  best  described  as  what  Paul  Fischer  has  called  a  “polymorphous  text”;9 there  is  noassumption of a “pure” version of the  Liezi from the brush of Lie Yukou himself. The assumption israther that the text is a collection of textual pericopes from a variety of authors and eras, and the textualunit known as the Liezi is one possible instance of the text. This one instance, however, bears the marksof having been constituted by a compiler (or compilers)10 with an intellectual agenda and a discerniblehistorical context. Authentication, in the sense of determining whether or not the  Liezi is what it ispurported to be, is not the ultimate goal; identification of the historical period in which it reached thestate in which we now have it is the primary aim. While this identification may necessarily entail thedisputation of the text's authenticity, it is undertaken only in order to better understand the content.9 For an excellent survey of the history of “Authentication Studies” (bianwei xue 辨僞學), see Fischer's “AuthenticationStudies Methodology and the Polymorphous Text Paradigm” in Early China, volume 31, 2008-2009, pp. 1 – 43. Manyof his useful typologies established in that article will provide a framework for the discussions below. In the case of theLiezi, the idea of exhibiting a “polymorphous” nature is still tempered by the likelihood of a single compiler, thoughcomparisons of extant editions do suggest editorial revisions.10 Throughout the following text,  the idea that  the compiler of the  Liezi could in fact  be multiple compilers workingtogether should not be neglected, even in cases when “compiler” is used in the singular.572.1 A Brief Textual History of the LieziAs best can be discerned, the content Liezi as we now have it enters history with Zhang Zhansometime in the mid to late fourth century CE.11 This sudden emergence onto the intellectual scene ismade all the more striking by the fact that no documents prior to its emergence explicitly quote theLiezi.12 This is an example of what Fischer has called the “sudden appearance” argument, suggestingthat one means of doubting a text's authenticity is to note the paucity of mention it receives in thecenturies it was purported to have existed.13There is mention of the title of the Liezi prior to the commentary of Zhang Zhan, as well as abrief description of its contents.14 Aside from the brief mention in the Hanshu Yiwenzhi, there does exista report on the Liezi attributed to Liu Xiang, which had originally been preserved as part of his Bielü 別綠, most of which is now lost. Liu's report survives, and is the oldest for the Liezi that has come downto our present time. It is as follows:15所校中書列子五篇,臣向謹與長社尉臣參校讎太常書三篇,太史書四篇,臣向書六篇,臣參書二篇,內外書凡二十篇,以校除復重十二篇,定著八篇。中書多,外書少。章亂布在諸篇中。或字誤,以盡為進,以賢為形,如此者眾。及在新書有棧。校讎從中書已定,皆以殺青,書可繕寫。16In collating the five  juan Liezi from the Palace Library I, your servant Xiang, havewith the Elder Sacrificial Officer Can carefully collated the Taichang [version of the]book in three pian, the Grand Scribe's [version of the] book in four pian, my [versionof  the]  book in  six  pian,  Minister  Can's  [version  of  the]  book in  two  pian.  The[versions of the] book from both inside and outside [the Palace Library] in all weretwenty  pian,  and by collation we removed the redundant  twelve  pian,  fixing andsettling on eight  pian. From inside [the Palace Library] the books were many, from11 Unfortunately, no firm dates for the life of Zhang Zhan are known.12 This does not mean that the Liezi does not share common material with many sources – it certainly does, and much ofthe authenticity argument rests on this fact. But no sources prior to the writing of Zhang Zhan's commentary attributequotations to a document called the Liezi.13 Fischer, 2010, p. 4.14 Lie Yukou, the person, makes several appearances in early documents, but a text bearing his name is unknown before theexamples given above.15 For  the  sake  of  brevity,  I  have  omitted  the  document  and  chapter  titles  that  Liu  includes,  as  well  as  standardpreliminaries and closing remarks.16  Yang (2007) pp. 277-278.58outside [the Palace Library]  the books were few. The  zhang were disordered andspread among the various pian. Some characters were wrong, taking 'jin' (盡) to be'jin' (進), or taking 'xian' (賢) to be 'xing' (形) –  instances like this were plentiful. Inthe  new text  they have  been17 removed.  Upon fixing  the  collated  book from theinternal edition, in all cases they were written on fresh bamboo,18 so that the textcould be copied and written.19Even in  Liu  Xiang's  time  the  text  of  the  Liezi was  in  a  degraded state,  and needed to  bereconstituted. After giving the account of how his text was prepared, Liu Xiang gives his take on thenature of the contents of the Liezi:列子者,鄭人也,與鄭繆公同時,蓋有道者也。其學本於黃帝老子,號曰道家。道家者,秉要執本,清虛無為,及其治身接物,務崇不競,合於六經。而穆王、湯問二篇,迂誕恢詭,非君子之言也。至於力命篇,一推分命;楊子之篇,唯貴放逸,二義乖背,不似一家之書。然各有所明,亦有可觀者。孝景皇帝時貴黃老術,此書頗行於世。及後遺落,散在民間,未有傳者。且多寓言,與莊周相類,故太史公司馬遷不為列傳。20Liezi was a man of Zheng, and was contemporary with Duke Mou of Zheng. Hecould  be  said  to  be  one  who  possessed  the  Way.  His  studies  were  grounded  inHuangdi and Laozi, and he was called a “Dao Expert”. As for Dao Experts: Theygrasp the essential and hold the root, they are clear, empty, and are wuwei; in comingto  governing  the  self  or  interacting  with  things,  they  work  at  esteeming  noncontention, and harmonize with the Six Classics. But as for the two “King Mu” and“Questions of Tang” pian, they are absurd and grandiose, and are not the words of ajunzi.  Arriving at  the “Li  ming” pian,  it  solely projects  the allotment  of fate;  the“Yang Zhu” pian only honours freedom and leisure; the two principles are opposed toeach other, and do not seem to be the text of a single school.21 But each has thatwhich it illuminates, and so there is something worth seeing [in them]. In the time ofFilial Emperor Jing,22 when the techniques of Huang[di] and Lao[zi] were honoured,this text circulated widely. Later it was rejected and left behind, scattered among thepeople,  and  and  there  were  none  that  transmitted  it.  Moreover  there  is  muchallegorical language, in the same category as [that of] Zhuang Zhou; therefore theGrand Scribe Sima Qian did not create a biography [in the Shiji]. Liu Xiang's verdict on the text already implies the heterogeneous nature of the writing, and suggeststhat four of the eight chapters are suspect in authorship and content.  Liu's explanation of why no17 My reading here is influenced by Graham (1990b), p. 224. He notes, however, that his reading is far from certain.18 That is, the green skin of the bamboo had been removed. See Yang (2007), p. 278. 19  My translation here benefits a great deal from Seo's work on this report. See Seo (2000), pp. 18-19. 20 Yang (2007) p. 278.21 Translating jia as “school” rather than “expert” in this case. 22 Reign dates 157 to 141 BCE. 59biography for Lie Yukou exists in the  Shiji is unsatisfying, however – though he correctly notes thesimilarity of the language of the Liezi with that of the Zhuangzi, it is unclear in what way that precludeshaving biographies for both Lie Yukou and Zhuang Zhou. The next, and more important, preface that has come down to us is that of the Liezi's primarycommentator, Zhang Zhan. It is as follows: 湛聞之先父曰:吾先君與劉正輿、傅穎根,皆王氏之甥也,並少游外家。舅始週,始週從兄正宗、輔嗣皆好集文籍,先並得仲宣家書,幾將萬卷。傅氏亦世為學門。三君總角競錄奇書。及長,遭永嘉之亂,與穎根同避難南行,車重各稱力,竝有所載。而寇虜彌盛,前途尚遠。張謂傅曰:「今將不能盡全所載,且共料簡世所希有者,各各保錄,令無遺棄。」穎根於是唯齎其祖玄、父咸子集。先君所錄書中有列子八篇。及至江南,僅有存者。列子唯餘楊朱、說符、目錄三卷。比亂,正輿為揚州刺州,先來過江,复在其家得四卷。尋從輔嗣女壻趙季子家得六卷。參校有無,始得全備。23I heard my late father say: “My late father, along with Liu Zhengyu and Fu Yinggen,were all nephews of the Wang clan, and while small they played with the family ontheir grandmother's side. Their maternal uncle was [Wang] Shizhou, and Shizhou'scousins Zhengliu and Fusi24 all enjoyed gathering written works; previously they hadtogether obtained Zhongxuan's family texts, totaling nearly ten thousand juan. The Fuclan was also taken by that generation to be scholars. The three of them had sincechildhood  competed  to  copy  unusual  texts;  having  grown,  they  encountered  thedisorder of the Yongjia25 period, and along with Yinggen [they] escaped disaster andwent south. Their carts were heavy and each was filled to the maximum of which itcould be loaded. Moreover, robbers were everywhere, and the road before them waslong. Zhang [my father] told Fu: 'Now we will not be able to save everything we havecarried, so together let us assess which writings are rare in our times, and each of uskeep a record of them to ensure they are not lost.' Yinggen then contributed only hisgrandfather  Xuan  and  father  Xianzi's  collections.  Among  what  my  late  fatherrecorded was the Liezi in eight pian. Upon reaching the southern side of the Yangziriver there was barely anything that survived. Of the Liezi there only remained threejuan: the “Yang Zhu”, “Shuo fu”, and the table of contents in three juan. At the timeof this chaos, [Liu] Zhengyu was made  cezhou26 of Yangzhou, and before comingacross the river returned to his home and [my father] obtained four juan. Seeking outFusi's  son-in-law Zhao Jizi's  home [my father]  obtained six  juan.  Consulting andcollating what  was there and what  was not,  he for the first  time had a  complete23  Yang (2007), pp. 278-279. 24 Wang Fusi, better known as Wang Bi 王弼 (226 – 249 CE).25 Referring most likely to the attack on Luoyang and capture of Jin Emperor Huai in 311 CE by the Wuhu 五胡. 26 The term is not present in Hucker's dictionary as it is written here, but I suspect it is closely related to entry 7567, ceshi刺史, “regional inspector”.60[version].”After attempting to establish the authenticity and history of the text, Zhang Zhan gives his estimation ofthe contents of the Liezi:其書大略明羣有以至虛為宗,萬品以終滅為驗;神惠以凝寂常全,想念以著物自喪;生覺與化夢等情,巨細不限一域;窮達無假智力,治身貴於肆任;順性則所之皆適,水火可蹈;忘壞則無幽不照。此其旨也。然所明往往與佛經相參,大歸同於老莊。屬辭引類特與莊子相似。莊子、慎到、韓非、尸子、淮南子、玄示、旨歸多稱其言,遂注之云爾。27The  general  outline  of  what  this  text  clarifies  is:  all  phenomena28 take  perfectemptiness29 to be the ancestor, and the myriad kinds take final extinction to be theverification  [of  this];  spirit-like  kindness  is  by  concentrated  quietude  [made]enduringly  complete,  and  thought  and  reflection  are  by contact  with  things  self-destroying; both living in wakefulness and transformation in dreams are equivalentsituations, and great and small are not limited to one domain; failure and success lackthe false [origins] of wisdom and power, and in governing the self there is value inbeing unrestrained by responsibility;30 if one follows one's nature, then wherever onegoes  will  be  appropriate,  and  water  and  fire  can  be  tread  upon;  if  one  forgetscorruptions, then there is no mystery that will not be illuminated. This is the purpose[of the text]. Thus that which it clarifies is often similar to Buddhist sūtras, [but] itlargely accords with [the thought of] Lao[zi] and Zhuang[zi].  The style and termsused  are  especially  similar  to  the  Zhuangzi.  Zhuangzi,  Shen  Dao,  Hanfei,  Shizi,Huainanzi, the  Xuanshi, and the  Zhigui31 often approvingly cite its statements, and27  Yang (2007), pp. 279-280.28 The term qunyou 羣有 here is translated in accordance with its meaning in Buddhist texts. That Zhang would use thisterm suggests he, like the compiler of the Liezi, may have had exposure to the ideas of Buddhism. 29 Zhang's description here may be influenced by a description of Lie Yukou's thought in the Shizi, which also associatesLiezi with the idea of “emptiness” or xu 虛 (see Paul Fischer's translation in Fischer (2012), p. 101. ).30 The two concepts suggested here – the capricious natures of success and failure and the ideal of unrestrained indulgence– clearly indicate the “Li ming” and “Yang Zhu” chapters of the Liezi, without explicitly stating so.31 In his annotation of Zhang's preface, Yang Bojun (Yang (2007), p. 280) suggests that the  Xuanshi may be either theYugui taozhong xuanshijing 玉龜胎中玄示經 or simply a text known as Xuanshi; he explains that the Zhigui is the Handynasty Laozi commentary attributed to Yan Zun 嚴遵  (also known as Yan Junping). Regrettably the Xuanshi is nowlost, though it may be the same as the Xuanshijing in ten juan mentioned as a text of the Dao Experts in the Neipian ofthe  Baopuzi (Wang (2002), p.  334), and the same document appears to have been quoted in the  Taiping yulan in the'Nourishing Life' (yangsheng 養生) section. It reads, in part: 夫形體者,特生之具也,非所以生生也。生生乃以素樸為體,以氣為元,以神為形,此乃生之宮庭也。以無為育其神,舒釋玄妙之門,往來無形之間,休息於無鄰,此所謂得玄明之生源。As for the form of the body, it is only an instrument of life; it is not the means by which life is generated. Lifebeing generated then takes the simple and unhewn as the body, takes qi to be the origin, takes spirit to be theform; this then is the generation of the 'palace'. Using wuwei to nourish the spirit, [one is] unhurried and easyat the gate of profound mystery; [one] comes and goes in the space that lacks form; [one is] at rest and reposein that without a counterpart – this is what is called “obtaining the source of the generating of dark andbright”. (Source text from the Hebei Jiaoyu Chubanshe 河北教育出版社 edition of the Taiping yulan vol. 6,p. 236).61therefore [I made] annotations. There are,  in  terms of  the question of the authenticity of  the text,  two major  points  worthconsideration.  The first is the transmission and reconstitution account offered by Zhang, for whichthere is no corroborating evidence or reports. A modicum of authority is present in the inclusion ofWang Bi as part of the community involved in the preservation of the Liezi text, though I am unawareof any instances in which Wang explicitly mentions or quotes the text. The tracing of the transmissionof the text through Zhang Zhan's paternal lineage suggests to Graham that the true compiler may havebeen Zhang's father or grandfather.32 Zhang himself seems to acknowledge that the text was not knownin his time outside the transmission within the family, and his preface may be an attempt to explain therelatively sudden emergence of a complete Liezi. The second significant point to address is Zhang's acknowledgment of textual parallels betweenthe Liezi and many other better known and more widely disseminated documents. Though he does notindicate any particular Buddhist documents by title, he does recognize a conceptual overlap that existsbetween the  Liezi and Buddhist thought. While this latter overlap is interesting enough to suggest tointerested readers in the latter part of the fourth century CE, when Zhang Zhan was writing his prefaceand commentary, textual parallels with Warring States texts ought not to be of significant note. Readerswere surely aware of this phenomenon as it existed among texts like the Lüshi Chunqiu, the Zhuangzi,and the  Hanfeizi, to give only a few examples. Zhang's point here seems to be a desire to make theargument that these texts were instead quoting from and drawing upon the thought of the Liezi ratherthan being drawn upon by it,  as Lie Yukou was assumed to have lived and taught earlier than theauthors enumerated. This appears to me to be a self-conscious and preemptive attempt to defend theIt is clear from the use of concepts like supu and wuwei that this work likely bore the marks of what Denecke has called“gestures of affiliation”, particularly in common with the Laozi and the Zhuangzi. The discussion of “generating life”(shengsheng 生生) may have much in common with Liezi 1:1 and 1:3, though without more a more complete context itis difficult to be certain on this point.32 Graham (1990b), p. 282.62authenticity of the text to anticipated sceptics, though I admit that this interpretation is speculative.33Zhang's commentary, which complements his preface, has been transmitted with the text of theLiezi almost without exception since its appearance on the intellectual stage in the late fourth century.34Though critics like A. C. Graham have noted that Zhang's comprehension of the source text is less thanperfect,35 his rather philosophical interpretation demonstrates a highly sophisticated literary sense andenviable access to both ancient and contemporary documents of historical, intellectual, and philologicalsignificance.36 A  more  robust  investigation  of  Zhang  Zhan's  commentary,  as  it  relates  to  thephilosophical dimensions of the Liezi text, must be delayed until a later time. For the remainder of thisinvestigation of authenticity, Zhang Zhan's commentary will be referenced in the context of verificationof the text as we have it.The second major commentary available for the Liezi text is that of Lu Chongxuan 盧重玄 (fl.mid 8th century), which is included in the Zhonghua Shuju edition of Yang Bojun as the jie 解 or sub-commentary. Barrett dates the work to the period of approximately 739-742 CE, and suggests that Lu'sgreater degree of philosophical interpretation of the  Liezi content devalues this document's use as aninterpretative  tool.37 Lu's  commentary  likewise  does  more  to  emphasize  the  presence  of  Buddhistelements in the text than Zhang's, which, according to Barrett,  served to make the document more33 My opinion here is certainly influenced by a similar suspicion held by Graham: “It is therefore likely that the book waswritten inside Zhang Zhan's family, perhaps by his grandfather Yi (fl. 307)...or by his father Kuang, on whose authorityZhan presents his very questionable account of the book's transmission...Moreover, [Zhang Zhan] calls our attention tothe very points which an accomplice in forgery would wish us to notice.” (Graham, 1990, p. 282).34 See Barrett in Littlejohn & Dippmann (2011), p. 15. Barrett has written two excellent and concise textual histories of theLiezi, both of which were indispensable in my investigation of the text and its commentaries. His account in Loewe(1993)  offers  a  cogent  summary  of  the  best  scholarship  on  the  Liezi,  as  well  as  a  good  account  of  the  majorcommentaries  and  editions.  The article  cited  here  from Littlejohn & Dippman's  Riding the  Wind with  Liezi:  NewPerspectives on the Daoist Classic is less encyclopedic in nature, but presents a fuller account of the “life” of the text itits first millennium. Readers concerned with the details of the transmission and reception of the Liezi before the modernperiod, especially on the question of extant editions, are encouraged to consult these works. 35 Graham (1990b), p. 282.36 On Zhang's erudition, see Barrett in Littlejohn & Dippmann (2011), p. 16.37 Barrett in Loewe (1993), p. 302.63palatable to Tang religious sensibilities.38If Barrett's estimation for the date of composition for Lu Chongxuan's commentary is accurate,then this second major commentary only slightly precedes the first major extant work to cast doubt onthe authenticity of the Liezi – the Bian Liezi 辨列子 of Liu Zongyuan 柳宗元 (773 – 819 CE).39 It isthis work that most scholars engaging in the debate on the Liezi's authenticity first address. Liu's majorcriticism is grounded in the assertion found in Liu Xiang's preface that Lie Yukou is a contemporary ofDuke Mou of Zheng.40 This historical claim is suspicious to Liu, as it contains an historical confusion:the Liezi records a discussion between Zi Chan 子產 (fl. mid 5th century BCE) and Deng Xi 鄧析 (fl.mid 5th century BCE),41 two thinkers that were alive and active after the life of Duke Mou of Zheng, thesupposed chronological contemporary of Lie Yukou. The implication is that it is unreasonable to expectLie Yukou to have recorded a conversation between two philosophers of a later age. While this argument lacks the textual and philological rigour of subsequent investigations onthe  Liezi,  it  serves  as  a  point  of  departure  for  the  subsequent  discussions  of  the  question.  ChenGuangzhong 陳廣忠 dismisses the notion that there is a textual conflict at all, surmising that this veryquestion  had  been  resolved  one  hundred and fifty  years  earlier  than  Liu  had  posed it,  by ChengXuanying 成玄英 (fl. mid 7th century CE) in his commentary to the Zhuangzi.42 Cheng lists Lie Yukou'scontemporary in a similar way as Liu Xiang, but substitutes the character  xu 繻  for Liu's  mou 繆 ;43Chen surmises the problem is a copying error rather than a historical one.44 Furthermore, I suggest thatcomparing the  Liezi text to external documents such as Liu Xiang's preface is ultimately inferior to38 Barrett in Littlejohn & Dippmann, p. 18.39 The text for this document is available in Yang (2007), p. 287ff.40 See the translation of Liu Xiang's preface above.41 Liezi 6:4.42 Chen, “Liezi sanbian – Liezi fei weishu kao zhi er” 列子三辨–列子非偽書考之二 in Daojiao wenhua yanjiu 道教文化研究, vol. 10 (1996), pp. 279-281.43 Guo (2004), p. 19 (note 9). 44 Chen (1996), p. 281.64examining the text itself.452.2 The Language and Grammar of the LieziA fruitful  method  of  investigating  the  Liezi with  the  intention  of  determining  its  date  ofcompilation is the close scrutiny of the language used. This method has been applied with great successby Yang Bojun and A.  C. Graham, and below I will  summarize their  findings.  Both arrive at  theconclusion  that  the  evidence  overwhelmingly suggests  that  the  Liezi was  compiled  in  the  Wei-Jinperiod.Yang Bojun makes a concise summary of his five major grammatical features that serve asevidence that the Liezi is a text other than what Zhang's preface purports it to be.46 His first point is theconstruction of a phrase in Liezi 3:7: shushi nian lai 數十年來 is a way of expressing time unknown inthe pre-Han period, and Yang offers several examples from before the Qin (including the Mengzi, ZuoZhuan, and Mozi) of the preferred construction of that time.47Second,  Yang  addresses  the  use  of  the  word  wu 舞  as  it  appears  in  the  Liezi.  He  citesspecifically an incident recorded in  Liezi 4:12:  鄧析顧其徒而笑曰:為若舞,彼來者奚若?“Dengxi looked back at his disciples and said, laughing, 'How would you like it if [I] had the personthat is coming act like a fool?'”48 Yang states that in his research he is not able to find an occurrence ofwu in this sense before the Han period, suggesting that this particular pericope is of a relatively laterdate.49 45 It ought to be noted that Seo considers Liu Xiang's preface itself to be a forgery, which, if true, renders it completelyuseless in understanding the textual history of the  Liezi. See Seo (2000), p. 34 and Seo (2015), pp. 450-451, whichrejects the authenticity of the report as a forgery based on stylistic reasons. 46 This list is found in Yang (2007), pp. 346-347.47 Yang (2007), pp. 327-328. The edition of Yang's work used here erroneously cites the “Tian rui” chapter as the sourcefor this phrase; it is properly found in the “Zhou Mu wang” chapter.48 I have translated here in accordance with Yang's observation. Graham translates  wu as “dance” in his rendering. SeeGraham (1990a), p. 84.49 Yang (2007), p. 331.65In his third point Yang addresses the use of dou 都 as an adverb in the Liezi text, of which helists four occurrences: Liezi 2:3, Liezi 3:7,50 Liezi 6:8, and Liezi 7:9. According to Yang, it is unknownfor  dou to be used this way before the Qin, and it is somewhat rare during the Han.51 This usage iscommon in  the  Wei-Jin  and following periods,  however,  and Yang lists  many examples  from theShishuo xinyu 世說新語, a text of that period.52Next, in his fourth point, Yang turns his attention to a particular use of the expression suoyi 所以 as found in Liezi 8:25. The owner of a lost sheep explains giving up his search: 吾不知所之,所以反也  “ I  did  not  know where  it  went,  therefore  I  went  back”.  As Yang notes,  the  usage  here  issuspiciously  similar  to  the  modern  usage  of  the  phrase  suoyi,  and  he  lists  more  likely  pre-Qinalternatives, such as shiyi 是以, shigu 是故, or gu 故.53 His many examples of the usual pre-Qin usageof suoyi serve to illustrate the anachronism of its usage here in the Liezi. It ought to be noted, however,that many instances of the pre-Qin usage of suoyi do occur in the Liezi text, and this is most likely ananachronism that slipped into the text rather than a characteristic feature of the document.Finally, for his fifth point Yang investigates the curious usage of the expression buru 不如  inLiezi 8:30: 田氏視之,乃歎曰:「天之於民厚矣!殖五穀,生魚鳥以為之用。」衆客和之如響。鮑氏之子年十二,預於次,進曰:「不如君言。天地萬物與我並生,類也。」54Mr. Tian looked up and giving a sigh he said, “How generous is Heaven toward thepeople; it causes the five grains to propagate and generates fish and birds in order for[the people] to use.” The group of guests that were with him echoed [his statement].[But] Mr. Bao's twelve-year-old son, seated on the farthest rank, came forward to say,50 Yang leaves out a passage from Liezi 4:6, presumably because it is nearly identical to this one, including its usage ofdou.51 Yang (2007), p. 332.52 Yang (2007), pp. 333-335.53 Yang (2007), p. 336.54 Yang (2007), p. 269.66“It  is  not as you say.  Heaven and Earth,  the myriad things,  and us are generatedtogether, and [all] are of a single type.”Yang explains that ru has two usages in ancient Chinese: without a negative such as bu 不, ruhas a meaning something like xiang 像 (“to resemble”); with a negative, ru has a meaning similar to ji及 (“to come”).55 He supplements this claim with copious examples from early documents. It is only inthe Han period, according to Yang, that  buru takes on the new meaning as found in the  Liezi text –rendering 不如君言  as 不像您所說的  (“It is not as you have said”).56 These five linguistic points,taken together, are fundamental to Yang's view of the Liezi as a later document. Graham builds on Yang's linguistic observations, but notes that aside from the usage of dou都they rely on examples found infrequently in the text.57 His own research offers ten types of word usagethat are indicative of a later date for the Liezi; most cases are drawn from portions of the text that lackparallels known in other documents. Despite suggesting that Yang's examples often draw on a limitedrange of examples, some of Graham's insights are equally dependent on a narrow sample set. However,the majority of Graham's examples are supplemented with a great deal of textual evidence. Below Ioffer a concise summary of his grammatical conclusions; all are examples of word usage that differsfrom the pre-Han usage a reader would expect if the Liezi were in fact a Warring States document. (1) wu 吾, wo 我, and post-pronoun zhi 之 – Graham notes that in pre-Han language wu is neverthe object of a sentence, only a subject or marker of the possessive. In the Liezi, he counts as many asten instances of a later usage in which wu is indeed the object of a verb. Moreover, he notes that zhirarely follows pronouns before the Qin period, yet he sees forty-three instances of this usage in theLiezi.58 He helpfully offers a textual example that captures both usages from Liezi 4:8:55 Yang (2007), p. 342.56 Yang (2007), p. 346.57 Graham (1990b), p. 248.58 Graham (1990b), pp. 249-250. 67視人如豕,視吾如人。處吾之家,如逆旅之舍;觀吾之鄉,如戎蠻之國。59I see others as though they were pigs, and I see myself as another person. Dwelling inmy home, it is as though I am lodging in an inn; observing my village, it is as thoughI am in a barbaric land.60In the example above, the phrase shi wu ru ren 視吾如人 places wu in the object position.61 Thephrase chu wu zhi jia 處吾之家 includes the particle zhi, normally redundant if wu is understood as apossessive. (2) ke 可 – Graham tells us that the usage of ke in pre-Han times indicated the verb following itwas passive, with only rare exceptions; it is a later linguistic development that allows the dropping ofthe yi 以 in keyi 可以 with active verbs.62 From Liezi 7:15 Graham offers the following:雖全生身63,不可有其身;雖不去物,不可有其物。64Even if one is complete in giving life to the body, they cannot possess their body;even if they do not send things away, they cannot possess their things.(3)  fu  弗  –  In the Liezi,  fu is often substituted for  bu 不 , ignoring its more specific pre-Hanmeaning. It is described by Graham as being true of sections of text both with and without parallels inthe received record.65 He considers the following from Liezi 5:2:離朱子羽方晝拭眥揚眉而望之,弗見其形66[As for] Li Zhu and Zi Yu, during the day they wiped their eyes and raised theirbrows to look for them, [but] did not see their form. Graham astutely notes that sections with parallels occasionally substitute fu in place of bu, as it59 Yang (2007), p. 129.60 My own translation, though it is influenced by and does not differ substantially from that offered by Graham in Graham(1990b), p. 250, or Graham (1990a), p. 82. 61 While this particular instance may be forced to preserve parallelism, it is only one of many examples in the Liezi. 62 Graham (1990b), pp. 251-252.63 Inserting shen 身 in accordance with Yang Bojun's note (Yang (2007), p. 235). The Dunhuang (S.0777) version of thetext omits this character. 64 Yang (2007), p. 235. 65 Graham (1990b), pp. 252-253.66 Yang (2007), p. 157. 68exists in other texts such  as the Zhuangzi. (4)  wang 亡  –  here Graham offers a particularly detailed history of the usage of  wang inreference to its usage with wu 無. The noteworthy point is that in pre-Han usage wang rarely takes anobject, and in the  Liezi wang is often found where one would find  wu more natural.67 Consider thefollowing example, supplied by Graham from Liezi 6:8:信命者,亡壽夭;信理者,亡是非;信心者,亡逆順;信性者,亡安危。68[As for] one that trusts fate, there is no long life or early death; for one that trustsprinciple,  there  is  no  is  or  not  is;  for  one  that  trusts  the  heart-mind,  there  is  noperversity  or  conformity;  for  one  that  trusts  their  nature,  there  is  no  security  ordanger.(5) dou 都  – Graham's discussion of dou does not differ in any appreciable way from that ofYang, but rather confirms those findings.(6)  yan 焉  –  The fusion particle  yan experienced a “decay” in its meaning in later centuries,according to Graham.69 Originally substituted for yuzhi 於之, though perhaps not exactly equivalent toit, and Graham finds that in most of its usage in the Liezi one is in a difficult position if they desire toargue for that usage. He cites multiple examples from Liezi 3:1, of which I will provide only one:意迷精喪,請化人求還。化人移之,王若殞虛焉。70[King Mu's] thoughts were perplexed and his refined essence was lost, [so] he askedthe magician to find [a means to] return [home]. The magician moved him, and theKing felt as though he were falling through the void.(7) xiang 相 – Graham notes the shift in the meaning of xiang from an indicator of reciprocityto a “pronomial adverb”.71 He suggests that reading reciprocity into its usage in as many as fourteen67 Graham (1990b), pp. 254-256.68 Yang (2007), p. 207. 69 Graham (1990b), p. 258.70 Yang (2007), p. 93. 71 Graham (1990b), p. 259-260. 69cases in the  Liezi is problematic, offering two examples. One of which, from Liezi 5:3, contains thephrase 雜然相許 “[They] all agreed to it”, and neatly makes this case.72(8)  qie 且  –  Graham suggests that in the Han period  qie has a change in meaning, beingtransformed from merely a particle that indicated futurity to taking on a mildly imperative flavour – hecontrasts  the  meaning  of  “about  to”  with  the  later  “let  us  for  the  moment”.73 This  is  clearlydemonstrated in Graham's example from the close of Liezi 7:3: 且趣當生奚遑死後 “For now hurry toyour current life – why be concerned with what happens after you die?”.74(9)  zhu 著  –  This verb takes on a verbal suffix meaning during the Han, as described byGraham.75 It is used in this capacity only a few times in the Liezi, but in striking and unambiguous way.Graham cites Liezi 5:2:而五山之根無所連箸76,常隨潮波上下往還,不得蹔峙焉。77But the roots of the mountains lack that to which they are connected to – they arealways following the rising, falling, moving to and fro of the tides and waves, andcannot for a moment be stable.(10) pronoun inversion – Graham notes that pronouns precede verbs in negative sentences inwhich they are the object, and that an exception to this regularity, coming into vogue during the latterpart of the Han, had it so that this was occasionally true for affirmative sentences.78 He offers manyexcellent examples beginning with Zhang Heng 張衡  (78-139 CE), but cites only one passage of theLiezi that adopts this stylistic feature, albeit with frequency.79 This is Liezi 6:2, which adopts the pattern72 Yang (2007), p. 159.73 Graham (1990b), p. 260-261.74 My reading here, especially in terms of the character huang 遑, is strongly influenced by Graham. See Graham (1990a),p. 141. 75 Graham (1990b), p. 262.76 Note that here the Liezi is using the graphic variant zhu 箸 for zhu 著; the meaning is the same.77 Yang (2007), p. 152.78 Graham (1990b), p. 263.79 Graham (1990b), p. 263-264.70朕與子並世也而人子達 “You and I are of the same generation, yet others have you succeed”.80 In apre-Han text,  one would expect  ren da zi 人達子  rather than  ren zi da  人子達 .  This particulargrammatical feature is not common in the Liezi, but its conspicuous presence is noteworthy.2.3 Parallels between the Liezi and the ZhuangziA separate but equally important means of investigating the  Liezi's content is to examine theways in which material  contained therein parallels  that of other texts,  and the ways in which it  ismodified. As Zhang Zhan points out in his preface, the Liezi in many instances relates tales or makesarguments  very  much  redolent  of  other  texts  such  as  the  Zhuangzi,  Huainanzi,  and  Hanfeizi,  toenumerate only a few. While Zhang suggests that these texts are quoting the Liezi, others, such as A. C.Graham, are convinced it is indeed the Liezi compiler that is making use of these texts, long after theircompletion. Below, I shall examine the most striking examples of this phenomenon.It must be furthermore noted that all parallels in the Liezi are not equal. Zheng Liangshu, in hisessay Cong chongwen de guanxi lun Liezi Huangdi de liuchuan 從重文的關係論列子皇帝的流傳,offers three types of parallel that exist: (1) “Complete repetitions” (wanquan chongfu 完全重複 ),which are nearly identical across instances; (2) “Revised extracts” (zhailu gaixie 摘錄改寫), whichare  clearly  the  same  passages  but  with  details  changed,  added,  or  removed;  and  (3)  “Minimalsimilarities” (jixiao bufen xiangtong 極小部分相同), which are textually quite different, but point tothe same notion.81 Zheng uses these typologies strictly in reference to the Liezi and the Zhuangzi, but Ibelieve they can be applied broadly without modification. It is worth noting that he reads all instancesas the Zhuangzi deriving its text from the Liezi, a position that I will investigate below.As Liu  Xiang explains  in  his  preface to  the text,  the  Liezi has  much in common with the80 This pattern is repeated multiple times without grammatical variation, and thus subsequent iterations are not includedhere. 81 Zheng (2001), pp. 90ff. 71Zhuangzi in terms of content and style. This is especially true of the second chapter “Huangdi”, inwhich about half of the pericopes can be found in the Zhuangzi.82 Ronnie Littlejohn, building on thegrammatical work undertaken by A. C. Graham, has suggested that many of the pericopes that exist inthis chapter without clear parallels in the Zhuangzi or other known texts may in fact be remnants of the“lost” Zhuangzi chapters, sections of the fifty-two chapter Zhuangzi that were excised by Guo Xiang inhis redaction of the text.83 The idea here is intriguing, and with some caveats the theory seems entirelyplausible. There are several pericopes of the “Huangdi” chapter that are unambiguously similar to theZhuangzi – these are  Liezi  2:4, 2:5, 2:8, 2:9, 2:10, 2:13, 2:14, 2:15, 2:16, 2:1984,  2:20.85 While thecontent here is never identical between the two versions as found in the  Liezi and the  Zhuangzi, areasonable reader will recognize the similarities. This leaves approximately half the chapter as possiblesources for insight into the “lost” Zhuangzi chapters. As Littlejohn notes, Graham connects Liezi 2:11and  2:12  to  the  “lost”  Zhuangzi based  on  the  evidence  of  early  commentaries86 and  linguisticevidence.87 The evidence for these thirteen (of twenty-one)88 pericopes suggests that at least half of thesecond chapter of the Liezi is drawn from an early version of the Zhuangzi, or shares a common sourcewith it. The evidence for the remaining pericopes is somewhat more circumstantial.  Littlejohn does82 Or, as Zheng Liangshu has it, the Zhuangzi borrows about 70% of the material from the Liezi “Huangdi” chapter. SeeZheng (2001), p. 111.83 Littlejohn in Littlejohn and Dippmann (2011), pp. 31-48.84 While Littlejohn only marks this pericope as “reminiscent” of the Zhuangzi, and suggests that it may in fact be anotherversion of the “Qi wu lun” monkey-trainer story found in the “lost” Zhuangzi, I see no reason not to list it as parallel tothe “Qi wu lun” version, even with a significantly different conclusion. See Littlejohn (2011), p. 38. Zheng Liangshureads it as a “revised extract”, although he believes it is the Zhuangzi quoting the original Liezi. Zheng (2001), p. 101. 85 A list of parallels in the Zhuangzi can be found in the Littlejohn (2011), p. 37-38.86 For example, Liu Xiaobiao's commentary to the Shishuo xinyu cites this passage as coming from the Zhuangzi, not theLiezi. See Mather (2002), p. 54 and Xu (1984), pp. 58-59.87 Littlejohn (2011), p. 38.88 Littlejohn actually divides the chapter into twenty-two pericopes, seeing Liezi 2:21 as two distinct pericopes, but I willtake a different approach below. 72offer a plausible suggestion in reference to Liezi 2:6 and 2:7, noting that their use of “skill stories” arereminiscent of similar tales in the Zhuangzi.89 I believe the case for 2:7 is even stronger than Littlejohnsuggests; though the narrative itself is somewhat unique in the received corpus, at least one line isclearly shared with the Zhuangzi. Below is the relevant portion of Liezi 2:7: 夫食虎者,不敢以生物與之,為其殺之之怒也;不敢以全物與之,為其碎之之怒也。時其飢飽,達其怒心。虎之與人異類,而媚養己者,順也;故其殺之,逆也。90As for one who feeds tigers, he does not dare to do it by means of giving them a liveanimal, because of their anger in killing it; he does not dare to do it by means ofgiving them a whole animal, because of their anger in tearing it apart. [He is] timelyin their hunger and fullness, successful in [knowing] their angry heart-mind. Tigersare a  different  species  than humans,  but  they love those that  raise  them, because[those that raise them] accord [with the tigers' heart-minds]; so [if] they kill them, [itis because] they have gone against [the tiger's heart-minds]. Though set in a different context, consider the following excerpt from the Zhuangzi “Ren jianshi” chapter, nearly identical in language and meaning to that of the Liezi: 汝不知夫養虎者乎?不敢以生物與之,為其殺之之怒也;不敢以全物與之,為其決之之怒也。時其飢飽,達其怒心。虎之與人異類而媚養己者,順也;故其殺者,逆也。91Do you not know of one who raises tigers? He does not dare to do it by means ofgiving them a live animal, because of their anger in killing it; he does not dare to do itby means of giving them a whole animal, because of their anger in opening it up. [Heis] timely in their  hunger and fullness,  successful in [knowing] their  angry heart-mind. Tigers are a different sort than humans, but they love those that raise them,because [those that raise them] accord [with the tigers'  heart-minds]; and so thosewho are killed are the ones that go against it.While the immediate context of the two quotations is different,  their  content and usage areclearly  related.  Zheng  Liangshu  also  reads  these  passages  as  related,  though  suggesting  that  the89 Littlejohn (2011), p. 37. The quintessential  Inner Chapters skill story is that of Butcher Ding, as found in the “Yangsheng zhu” chapter. 90 Yang (2007), p. 58.91 Guo (2004), p. 167. 73Zhuangzi version is a “revised extract” of the Liezi passage.92 If the passage is indeed derived from the“lost” Zhuangzi, then perhaps this passage existed in multiple forms in the fifty-two chapter version ofthe text.For  Littlejohn,  in  the face of  this  mounting evidence for  the  passages  mentioned above asextracts from the fifty-two chapter Zhuangzi, the remaining pericopes are swept up in the momentum ofthe argument as circumstantially likely remnants of the “lost” Zhuangzi found in the Liezi.93 I believethis is a tenable position, barring explicitly conflicting evidence. However, there do exist other avenuesin the received corpus through which some of these pericopes may have been transmitted; that is to say,while we lack direct evidence that these passages were taken from the “lost” Zhuangzi, they do exhibitparallels with other received texts. Below I will consider two cases.As explained above, on the strength of Graham's evidence Littlejohn has conjectured that Liezi2:11 may in fact be derived from the “lost” Zhuangzi. What his review of the evidence does not includeis that this pericope exhibits non-trivial parallels with content found both in the Lüshi Chunqiu and theextant Zhuangzi “Zhi bei you” chapter. Below I have translated Liezi 2:11 in its entirety.海上之人有好漚鳥者,每旦之海上,從漚鳥游,漚鳥之至者百住而不止。其父曰,「吾聞漚鳥皆從汝游,汝取來,吾玩之。」明日之海上,漚鳥舞而不下也。故曰,至言去言,至為無為。齊智之所知,則淺矣。94There was a man by the sea that loved seagulls. Everyday he went to the sea theyfollowed him where he wandered, and the seagulls that came to him were hundredswithout end. His father said, “I hear that all the seagulls follow you wandering – bringthem so that I can play with them.” The next day at the seaside, the seagulls flitted[above] but would not descend. So it is said: Perfect speech does without speech,perfect action is wuwei;95 if it is ordinary wisdom that one knows, then it is shallow.While this may be adapted from a “lost” Zhuangzi passage, consider the Lüshi Chunqiu passage92 Zheng (2001), p. 103. 93 Littlejohn (2011), pp. 39-40.94 Yang (2007), p. 67-68. 95 Literally, “perfect action is lacking action”. Presumably the seagull lover enjoyed his time amongst the seagulls in awuwei manner before being consciously tasked with the objective of bringing one for his father. 74below, from the opening of the “Jing yu” chapter: 海上之人有好蜻者,每居海上,從蜻游,蜻之至者百數而不止,前後左右盡蜻也,終日玩之而不去。其父告之曰:「聞蜻皆從女居,取而來,吾將玩之。」明日之海上,而蜻無至者矣。96Among the people by the sea there was one that loved dragonflies. Every time that hedwelt by the sea they followed him where he wandered, and the dragonflies that cameto him were hundreds without end. On all sides he was completely surrounded bydragonflies, and he would play with them to the end of the day without leaving. Hisfather told him, “I hear the dragonflies all follow you where you dwell, take one andbring it to me – I want to play with it.” The next day he was at the seaside but therewere no dragonflies that would come to him. Apart from the most obvious difference – the switch from seagulls to dragonflies – the LüshiChunqiu narrative is remarkably similar to that of the Liezi. The Liezi tale ends with a comment thatsummarizes for the reader the meaning of the passage; this comment is notable in that it is similar to anaphorism found among pre-Han and Han texts,  including the  Zhuangzi, Huainanzi,  and  the  LüshiChunqiu,97 where in the latter it serves as a comment to a different narrative in the same “Jing yu”chapter. Below I offer the Zhuangzi parallel, which serves to bring the “Zhi bei you” chapter to a close,as it is the most directly parallel to the Liezi version.至言去言,至為去為。齊知之所知,則淺矣。98Perfect speech does without speech, perfect action does without action. If ordinaryknowing is what one knows, then it is shallow. Aside from the conspicuous change of qu wei (“banishing action”) to the more familiar “gestureof affiliation” of wuwei in the Liezi version, this comment is nearly identical. While Liezi 2:11 may betraceable to the fifty-two chapter Zhuangzi, it is equally plausible that it is a composite pericope craftedfrom selective quotation of the Lüshi Chunqiu and the extant Zhuangzi.96 Xu (2009), p. 481-482.97 It is worth noting the Liezi chapter “Shuo fu” also contains a similar line, which is virtually identical to the Huainanzi orLüshi Chunqiu. See Liezi 8:12.98 Guo (2004), p. 765. 75Liezi 2:17 lacks any obvious content parallels to the Zhuangzi, and although Littlejohn describesit as potentially derived from the “lost”  Zhuangzi, the evidence is hardly more than circumstantial.99While the first half of the section lacks any clear antecedents or parallels in the received corpus, andthus may very well be from an older version of the Zhuangzi, the second half of the pericope bears anundeniable resemblance to the Huainanzi “Yuan dao” chapter.100 Both are notable for their quotation ofthe Laozi, which is characteristic of the “Yuan dao” chapter of the Huainanzi. First, the Liezi version:粥子曰:「欲剛,必以柔守之;欲彊,必以弱保之。積於柔必剛,積於弱必彊。觀其所積,以知禍福之鄉。彊勝不若己,至於若己者剛;柔勝出於己者,其力不可量。」老聃曰:「兵彊則滅,木彊則折。柔弱者生之徒,堅彊者死之徒。」101Yuzi102 said, “Desiring to be firm, you must use softness to guard it; desiring to bestrong, you must use weakness to protect it. The piling up of softness will becomefirm, the piling up of weakness will become strong. Observe their accumulation, andby that know the places of good and bad fortune. The strong overcome those whichare not equal to them, [but] when arriving at one equal to them they are [matched in]firmness;103 the weak overcome what surpasses them, and their strength cannot bemeasured.” Lao Dan said, “If a weapon is strong, then it will be destroyed. If a tree isstrong, then it will break. The soft and weak are the servants of life, the hard andstrong are the servants of death.”104Compare this to the Huainanzi version: 是故欲剛者必以柔守之,欲強者必以弱保之。積於柔則剛,積於弱則強,觀其所積,以知禍福之鄉。強勝不若己者,至於若己者而同;柔勝出於己者,其力不可量。故兵強則滅,木強則折,革固則裂,齒堅於舌而先之敝。是故柔弱者生之榦也,而堅強者死之徒也。105Therefore  desiring  to  be  firm,  one  must  use  softness  to  guard  it;  desiring  to  be99 See his chart of Liezi “Huangdi” pericopes in Littlejohn (2011) pp. 37-38; on page 40 he argues they are likely of thesame origin as nearby pericopes, though he is also cautious in this claim.  100 This parallel also exists in the Wenzi “Dao yuan” 文子道原 chapter; however, due to the difficulties surrounding thattext, I will focus on the more reliably datable Huainanzi. 101 Yang (2007), pp. 82-83. 102 That is, Yu Xiong 鬻熊 (11th century BCE?). No such attribution is made in the Huainanzi version.103 As Zhang Zhan comments: 必有折者 “There must be one that breaks”. See Yang (2007), p. 83.104 See Laozi 76. The contents here are certainly not identical, but bear a strong resemblance. 105 He (2006), p. 49-50. 76strong,106 one must use weakness to protect it. The piling up of softness will becomefirm, the piling up of weakness will become strong. Observe their accumulation, andby that know the places of good and bad fortune. The strong overcome those whichare not equal to them, [but] coming to their equal they are then the same; the softovercome  those  which  surpasses  them,  and  their  strength  cannot  be  measured.Therefore weapons that are strong are destroyed, trees that are strong are broken,leather that is solid will tear; teeth are harder than the tongue and will be worn outfirst. Therefore the soft and weak are the main part107 of life, and the hard and strongare the servants of death.108Littlejohn makes much of the notion that the close of the “Huangdi” chapter is not derived fromthe fifty-two chapter Zhuangzi.109 He is quite adamant in this conclusion, but regrettably offers nothingin the way of concrete evidence,  only the assertion that “[t]here really can be little doubt that theaddition of the final text bead in this chapter is from the editor and not from the source he copiedthroughout the rest of the chapter”.110 I note here that neither Yang Bojun nor A. C. Graham divide thetext in this manner – both, in fact, make it plainly obvious that they view this as a continuation of thepassage.111 Furthermore,  I  cannot  identify any element  of  style  that  marks  it  as  different  from thepericopes  that  precede  it.  One  may  only  speculate  as  to  Littlejohn's  reasons  here.  Below  I  havetranslated the passage in question, which follows Hui Ang's assertion that he knows a way (dao 道) thatis greater than that of courage (yong 勇) or strength (li 力):惠盎對曰:「孔墨是已。孔丘墨翟無地而為君,無官而為長;天下丈夫女子莫不延頸舉踵而願安利之。今大王,萬乘之主也;誠有其志,則四竟112之內,皆得其利矣。其賢於孔墨也遠矣。」宋王無以應。惠盎趨而出。宋王謂左右曰:「辯矣,客之以說服寡人也!」113106 I note here that I have rendered both qiang 彊 and qiang 強 as “strong” in the English translation. Though the meaningsare similar enough that I do not feel compelled to differentiate them in translation, it is worthwhile to indicate that thepassages differ in this respect. 107 Literally, “trunk”. Compare to gan 幹 in Kroll (2015), p. 128.108 While this also contains material parallel to  Laozi 76, it is not explicitly attributed to the  Laozi or Lao Dan in theHuainanzi. This is noteworthy, as the Laozi is the one pre-Qin text that the Huainanzi compilers frequently do attributequotations. Quotation without attribution of this sort in the Huainanzi is expected of the Zhuangzi.109 Littlejohn (2011) p. 40. As he indicates, this is the text portion that begins 惠盎對曰孔墨是已... See Yang (2007), p. 88.110 Littlejohn (2011) p. 48, note 4.111 Yang (2007), p. 88; Graham (1990a), p. 57.112 Reading jing 境 for jing 竟, following Yang Bojun's suggestion (Yang (2007), p. 89). 113 Yang (2007), pp. 88-89. 77Hui Ang responded, “Kong[zi] and Mo[zi] are like this. Kong Qiu and Mo Di lackedterritory and yet acted as lords, lacked office and yet acted as leaders. Among all themen and women of the world, none did not crane their necks and lift up their heels,wishing to give them peace and benefit. Now as for the great king [i.e., you],114 youare the lord of ten thousand chariots. If you sincerely had this will, then those withinthe  four  boundaries  [of  your  state]  would  all  obtain  benefit.  The  worthiness  ofKong[zi] and Mo[zi] – [yours would be] beyond that.” The king of Song could notrespond. Hui Ang hastened to leave.  The king of Song said to those around him,“Such eloquence of disputation115 – he used speech to subdue me!”Perhaps it is the apparent praise for Kongzi and Mozi that serve to distinguish this passage fromthe rest of the “Huangdi” chapter or Zhuangzi material in general; however, a close inspection suggestsit is not the content of their philosophies that is praised, but rather that despite their lack of political ormilitary  power  they  were  able  to  earn  the  veneration  of  the  masses  through  their  application  ofintellectual or moral prowess. Without an articulated argument on the matter one may only speculatewhat element of Liezi 2:21 seems to distinguish it from the rest of the chapter. Unless that argument isforthcoming I am not aware of any reason to set that passage apart.Littlejohn makes the assertion that examining Liezi material that resembles that which we findin the Zhuangzi in terms of “strata” is a misguided approach.116 He is clear in indicating that “strata”here indicates the various lineages of Zhuangzi material identified by Graham – specifically, “ZhuangZhou”, the “School of Zhuangzi”, the “Primitivists”, the “Syncretists”, and the “Yangists”, interspersedwith various “mutilated” passages.117 Littlejohn's research suggests that the  Zhuangzi material in theLiezi follows no discernible pattern, borrowing indiscriminately. If one hopes to find only one textuallineage represented in the “Huangdi” chapter they will be disappointed, for as Littlejohn accuratelydemonstrates both the “Zhuang Zhou” material (i.e., the Inner Chapters) and the “School of Zhuangzi”(or “Zhuangzi Disciples”) are found therein.118 114 Here Hui Ang is addressing Kang of Song, as the first half of this pericope explains.115 I recognize “eloquence of disputation” may be overtranslating bian 辯 somewhat; however, I believe it best captures theessence of the king's remark.116 Littlejohn (2011), p. 41. 117 Graham (1981), pp. 27-33. 118 Littlejohn (2011), pp. 41-42.78Despite this fact, I believe there is indeed an interesting pattern that emerges from the  Liezi'sselective use of the  Zhuangzi material. To see it we must expand our perspective in two importantways. First, it is useful to make use of not only Graham's classification of the Zhuangzi chapters, but toalso  utilize  Liu  Xiaogan's  excellent  and  independent  study  of  the  same  material,  which  reachessomewhat similar conclusions in categorizing the chapters, at least in terms of defining the bounds ofthe various strata of the text. That these two detailed studies arrive at such similar results gives usgreater confidence in identifying lineages or “strata” in the Zhuangzi text, which in turn enhances ourability to identify these lineages in the Liezi text. The second important step in understanding the Liezi'sselective use of the Zhuangzi entails expanding our scope beyond the “Huangdi” chapter to include theentirety of the  Liezi text. In doing so we find that the  Liezi uses the “Zhuang Zhou” and “School ofZhuangzi” material exclusively, with only two important exceptions addressed below.First, I note that Liezi 4:16 begins with an explicit quotation of Guan Yin which is, aside fromminor  differences,  identical  to  a  similar  quotation  of  the  same attribution  found in the  “Tian  xia”chapter of the Zhuangzi. Both Graham and Liu identify this chapter as markedly distinct from either the“Zhuang Zhou” or “School of Zhuangzi” material, respectively labeling it “Syncretist”119 or “Huang-Lao”120.  However,  as the parallel in this case is merely a short  quotation attributed to Guan Yin,  Isubmit that it is likely both documents are quoting a common source attributed to Guan Yin, and that inthis case there is no reason to believe the Liezi is using the “Tian xia” material as a source. The second instance of the Liezi quoting outside the “Zhuang Zhou” or “School of Zhuangzi”lineage is Liezi 8:7, which relates a narrative about Lie Yukou paralleled in the Zhuangzi “Rang wang”chapter.  Graham  identifies  “Rang  wang”  as  “Yangist”121 and  Liu  identifies  it  as  “Anarchist”122.119 Graham (1981), pp. 257-258; 274-285 (translation). 120 Liu (1994), pp. 121-134. 121 Graham (1981), pp. 221-223, 224ff. 122 Liu (1994), pp. 134-147. 79However, there exist two good reasons to believe that the parallel here does not complicate the notionthat the Liezi uses Zhuangzi material selectively. First, as Seo notes, the Liezi compiler seems to havetaken pains to include every known mention of Lie Yukou in the written record in the composition ofthe Liezi.123 The usage of the “Rang wang” material here may simply be an example of inclusion basedon  the  topic  (i.e.,  Lie  Yukou)  rather  than  the  content.  However,  we need  not  rely solely on  thatconjecture. It is furthermore prudent to demonstrate that Liezi 8:7 is paralleled in many sources asidefrom the Zhuangzi “Rang wang” chapter, which include the Lüshi Chunqiu and the Xinxu 新序 of LiuXiang.124 Thus, it is probable that either of these sources (or another, now unknown source) may haveserved as the origin for this particular passage. I propose that  Liezi 8:7, as with Liezi 4:16, offers noserious challenge to the proposition that the Liezi text borrows selectively from the Zhuangzi, utilizingmaterial from the “Zhuang Zhou” and “School of Zhuangzi” chapters.125Having made a digression to address these two exceptions, I return now to my contention thatthe Liezi otherwise draws exclusively on “Zhuang Zhou” and “School of Zhuangzi” material. If true, itmay offer  insight  into  the  process  of  the  composition  of  the  Liezi text.  If  one  concludes  that  thecompiler had access to either Guo Xiang's  thirty-three chapter redaction of the  Zhuangzi,  or,  as isimplied by the notion that the Liezi uses content from the “lost” Zhuangzi, an earlier fifty-two chapterversion of that work, then it appears that the compiler is demonstrating a preference for the “ZhuangZhou” and “School of Zhuangzi” lineages of thought in relation to the “Yangist”, “Primitivist”, or123 Seo (2000), pp. 38ff. 124 This particular episode is also recorded in the  Gaoshi zhuan of Huangfu Mi (215-282 CE), but as this document'stextual history is complicated I have not considered it as a potential source for the Liezi here.125 It is to be noted that this is strictly true for Liu Xiaogan's classification of the Zhuangzi material; however, while somematerial in the Liezi borrows from what Graham has called the “mutilated” chapters, none of it is classified by him asbelonging to  any other  specific  lineage.  These  pericopes  are  Liezi  2:14  (matched  to  Zhuangzi “Lie  Yukou”),  2:15(matched to Zhuangzi “Yu yan”), and 6:3 (matched only in part to Zhuangzi “Xunwu Gui”). As Graham does not includeany of these three narratives in his translation of the  Zhuangzi, and therefore does not explicitly classify them, it isdifficult to determine exactly how he would have understood them in relation to the other Zhuangzi material. He doeshowever note that these three chapters contain elements that often intersect with material in the Inner Chapters.80“Syncretist” lineages.126 This is significant, for if we now know something about the characteristics ofthe  lineages  with  which  the  Liezi compiler  found  commonality  we  gain  further  insight  into  thecompiler's general worldview. Finally, a challenge to the theory that the Liezi quotes Zhuangzi material and is therefore later isoffered by Zheng Liangshu. His work on the question, cited above in reference to his typologies ofparallels, gives twenty examples of textual convergences. In all cases, he asserts that it is the Zhuangzithat quotes the Liezi, most frequently the “Huangdi” chapter. He suggests, for example, that in caseswhere overlap occurs, the Liezi generally employs more obscure characters than does the Zhuangzi –such a phenomenon indicates to Zheng that the author of the Zhuangzi is in his redaction of the textsimplifying the language.127 This argument ignores the complex textual history of the Zhuangzi, as wellas turns on the assumption that in copying from the Zhuangzi the compiler of the Liezi could not haveintroduced character variants (which could certainly be done in compiling the document, especially ifthe compiler was interested in having the document appear older than it was). Furthermore, Zhengsuggests that both the addition of detail in the  Zhuangzi version of the story of the “spirit man” thatlives on Mount Gushe (cf.  Liezi 2.2)128 and the removal of detail from the  Zhuangzi monkey trainerstory (for example, his being a native of Song – cf. Liezi 2.19)129 attest to the ingenuity displayed by theauthor of the  Zhuangzi in his appropriation of  Liezi  material. I am unconvinced that the addition orremoval of information in this manner demonstrates the chronological priority of either the Zhuangzi orthe Liezi, and thus reject this particular line of reasoning. 126 I must also suggest the admittedly speculative possibility that the Liezi compiler had access to a version of the Zhuangzithat only contained “Zhuang Zhou”, “School of Zhuangzi”, and “lost” Zhuangzi material. While this would render thequestion of the compiler's attitude toward other  Zhuangzi lineages open and perhaps unanswerable, it would grant afascinating perspective on the reception history of the Zhuangzi.127 Zheng (2001), p. 91, 93. 128 Zheng (2001), p. 99. 129 Zheng (2001), p. 101. 812.4 Parallels between the Liezi and other WorksThere are other works that exhibit significant parallels to the Liezi which by virtue of enteringthe written record at a time demonstrably subsequent to the pre-Qin period suggest it to be most likely apost-Han document. Below I will address the most commonly discussed instances of this phenomenon.As has been demonstrated by Ma Xulun  馬敍倫  (1885-1970), Chen Wenbo  陳文波 , A.C.Graham, and Junwon Seo, the  Liezi “Zhou Mu wang” chapter (specifically  Liezi 3:1) has significantparallels to the Mutianzi zhuan.130 The text was recovered from a tomb around 281 CE; the tomb itselfis thought to be from roughly 350 BCE.131 The parallels are less substantial than those found in theZhuangzi, in the sense that only fragments are mirrored in the  Liezi, not the majority or entirety ofpericopes. Despite this, the parallels are striking, and are often integral parts of the debate over theLiezi.  Aside  from pure  textual  parallels,  Rémi  Mathieu  has  also catalogued many of  the  thematicparallels that are evident in both the Liezi and Mutianzi zhuan accounts of King Mu, which include butare not limited to: the meeting with Xiwangmu 西王母, the voyage to Kunlun 崑崙, and the visitationto Huangdi's Palace.132 Given that these are themes common to tales of King Mu, it is suggestive thatthe Liezi compiler seems to have opted for making use of the Mutianzi zhuan material directly.Mathieu further demonstrates that it is perhaps only the first four of the six juan of the Mutianzizhuan that ought to be accepted as authentic, and that the latter two are likely interpolations; 133 this issignificant to the present investigation, as the Liezi appears to only draw from the first four juan. BelowI will compare examples from both texts as found in the summary prepared by Chen Wenbo:134130 Ma is reprinted in Yang (2007), p. 301; see especially his point five (p. 302). Chen is also reprinted in Yang (2007), p.318; see his helpful chart of comparisons on p. 320. For A. C. Graham see Graham (1990) pp. 242-245, and for Seo seeSeo (2000), pp. 73-76.131 Mathieu in Loewe (1993), p. 342. 132 See a complete list of parallels between the  Mutianzi zhuan and many other texts, including the  Liezi, as found inMathieu (1978), pp. 194-196.133 Mathieu in Loewe (1993), p. 343.134 Yang (2007), p. 320.82王乃歎曰:於乎!予一人不盈于德,而諧於樂,後世其追數吾過乎!135The king then sighed, “Alas! I, the king, have not delighted in virtue but [instead]given in to pleasure – later generations will remember this as my error!”Compare the Mutianzi zhuan version, as found in the first juan:天子曰:於乎!予一人不盈于德,而辨於樂,後世亦追數吾過乎!The Son of Heaven said, “Alas! I, the king, have not delighted in virtue but [instead]been  discriminating  in  favour  of  pleasure  –  later  generations  will  indeed  oftenremember this as my error!”136Such a parallel is demonstrative of the pattern as a whole. Defenders of the authenticity of theLiezi acknowledge the similarities between it and the Mutianzi zhuan, but insist that these similaritiesare not proof of the later compilation of the text. The issue is addressed directly by Chen Guangzhongin  his  widely  cited  tripartite  proof  for  the  authenticity  of  the  Liezi.137 Chen  Guangzhong  firstacknowledges the points of similarity made by Ma Xulun and Chen Wenbo in their respective works,but is quick to add that literature relating to the westward journeys of King Mu were not unknownbefore the unearthing of the  Mutianzi zhuan.138 Other known sources, such as the  Shiji, offer similarinsights. He goes on to enumerate reasons to suppose that the Liezi compiler did not borrow from theMutianzi  zhuan,  all  of  which  are  unconvincing,  and do not  address  the  grammatical  and  stylisticelements that are virtually identical between the two texts.139 Chen Guangzhong does point out theimportant fact that significant parts of Liezi 3:1 have no parallel in any part of the Mutianzi zhuan,140but this is not the claim of the critics cited: the claim is simply that fragments seem to be borrowed, notthe entirety of the text. Chen's argument deteriorates further when he cites many examples external to135 Both this and the following quotations follow Chen in Yang (2007), p. 320.136 This is merely one of the four major examples of parallels; space considerations preclude examining more here indepth. The interested reader is referred to the more comprehensive work by Graham (1990b), pp. 242-245 and Seo(2000), pp. 73-76. 137 Chen in Daojiao wenhua yanjiu 道教文化研究, vol. 10, pp. 267-299.138 Chen (1996b), pp. 284-285.139 I note here that Ma Da's critque of Chen Wenbo's work on the Mutianzi zhuan as it relates to the Liezi makes similarassertions about the priority of the Liezi with no evidence. See Ma (2000), pp. 134-135.140 Chen (1996b), p. 286.83Liezi 3:1, or even external to the “Zhou Mu wang” chapter (e.g., from the “Tang wen” chapter).141 Thissidesteps the issue at hand – no critic, to my knowledge, has ever claimed that other portions of theLiezi text derive from the Mutianzi zhuan outside of Liezi 3:1. On the whole, Chen's argument on thispoint is weak, and it does appear that the parallels in the  Mutianzi zhuan remain some of the bestevidence for a later date for the Liezi. If asked to speculate as for the reasons for its inclusion, I wouldsuggest that the compiler was perhaps creating a new version of the King Mu narrative, and includedsome material known to be from the Warring States period as a gambit for authenticity.The final pericope of the “Tang wen” chapter has also often served as a point of contention forthose debating the authenticity of the Liezi. Below is a complete translation of Liezi 5:17.周穆王大征西戎,西戎獻錕鋙之劍,火浣之布。其劍長尺有咫,練鋼赤刃,用之切玉如切泥焉。火浣之布,浣之必投於火;布則火色,垢則布色;出火而振之,皓然疑乎雪。皇子以為無此物,傳之者妄。蕭叔曰:「皇子果於自信,果於誣理哉!」142King Mu of Zhou undertook a great expedition against the Western Rong,143 and theWestern Rong offered him the Kunwu sword and cloth that was washed in fire. Thesword had a length of a chi and a zhi,144 [and was made of] refined steel with a redblade. Using it to cut jade was like cutting mud. As for the fire-washed cloth, onecould only wash it by tossing it in fire; the cloth was coloured like fire, the dirt wascoloured like the cloth. [After] taking it out of the fire and shaking it, it was so whiteas to be mistaken for [the colour of] snow. The prince145 believes that these things didnot exist, that those that transmitted [these ideas] were reckless. Xiaoshu146 said, “Theprince trusted himself as expected, and as expected his understanding was mistaken!”This particular passage is noteworthy because it is mirrored very closely in the Kongcongzi, which asdiscussed in Chapter One of the present work, was likely compiled by Wang Su (or someone in his141 Chen (1996b), p. 287.142 Yang (2007), pp. 189-190. 143 Non-Han peoples  living in  the North-West.  The presence of  King Mu and his  westward journey again makes anappearance in the  Liezi; it should be noted, however, that there are no significant links to the  Mutianzi zhuan in thispassage. 144 About forty centimetres.145 A. C. Graham suggests that huangzi 皇子 could also be interpreted as a personal name. See Graham (1990), p. 117, aswell as the discussion below. The term does not seem to have been used in this way before the Han dynasty.146 No commentary offers a clue to who Xiaoshu is, and his comment is not present in any parallel accounts. Perhaps this isa reference to Xiaoshu Daxin 蕭叔大心, of the Zuo zhuan; however, it is unclear as to why he is quoted here.84circle) at the close of the Han period.147 The material is also paralleled less closely in Zhang Hua's 張華(232-300 CE)  Bowuzhi 博物志 .148 This is critical, for if the particular formulation of this narrativecannot be found earlier than Wang Su or Zhang Hua version, it lends credence to the supposition thatmaterial in the Liezi must be post-Han.Most  critics  operate  under  the  assumption  that  the  Liezi compiler  here  had  Cao  Pi  曹丕(reigned as Emperor Wen of Wei  魏文帝 , r. 220-226) in mind as the misguided prince (huangzi).149According to sources such as the  Baopuzi, Cao Pi had suggested that such items as the jade-cuttingsword and the cloth washed in fire could not exist, but was disgraced and ashamed when such itemswere presented to him.150 While this speculation does rest on the assumption that the Liezi is a post-Hanwork, if accurate it is a potentially intriguing insight into the worldview of the Liezi's compiler. As with the Mutianzi zhuan parallels discussed above, Chen Guangzhong has offered a counterexplanation for this material in the Liezi. He makes much of the fact that the Bowuzhi account of thefire-washed cloth attributes the record of this to the Zhoushu 周書, though no such record exists in thereceived version.151 Chen offers many other examples of a fire-washed cloth in the received record, butall of them certainly post-date that which we find the Kongcongzi.152 Ultimately he concluded that the“prince” referred to is the son of King Mu of Zhou, and not Cao Pi.153 This argument is weak when wecompare it with the evidence from the other sources Chen cites.154 For example, he offers the followingaccount from the Inner Chapters of the Baopuzi, as is mentioned above:155147 For this story, see the translation in Ariel (1996), pp. 31-32.148 Fan (1980), p. 26.149 See Graham (1990), p. 117, and Seo (2000), p. 71.150 Seo (2000), p. 71. 151 Chen (1996b), p. 282.152 Chen (1996b), pp. 282-283.153 Chen (1996b), p. 284.154 Ma Da also addresses the parallels between the Liezi and the Bowuzhi, listing ten with varying levels of divergence.Though he does address most of these cases, albeit unconvincingly, he does not offer evidence specific to the story of theblade and cloth offered above. See Ma (2000), pp. 295, 298-299. 155 Chen (1996b), p. 283.85魏文帝窮覽洽聞,自呼於物無所不經,謂天下無切玉之刀,火浣之布,及著典論,嘗據言此事。其閒未期,二物畢至。帝乃歎息,遽毀斯論。156Emperor Wen of Wei read exhaustively and was widely informed, and boasted ofhimself that among things there were none that he had not surveyed. He said thatunder Heaven there was no blade that cut jade or cloth washed in fire. The Dianlun157was written and it relied on his word on this matter. [But] before a short time hadpassed the two things were brought [to him]. The emperor sighed, and subsequentlydenigrated the Dianlun. The Baopuzi was composed in the mid fourth century CE, roughly the same time many criticsof the Liezi's authenticity suppose that the Liezi text was compiled. A textual tradition explicitly linkingCao Pi to a public doubt as to the veracity of claims of a fire-washed cloth was present in the Baopuziand  other  works,158 while  the  competing  hypothesis  ascribing  this  doubt  to  King  Mu's  son  lackscorroborating evidence. What  textual  sources  do exist  strongly suggest  that  the  Liezi account  wasinfluenced by other accounts found in the fourth century CE.One textual parallel that exists in the Liezi that has received virtually no discussion in the manydebates that have taken place is that which exists between Liezi 8:28 and a passage from the mostly lostChang yan 昌言 of Zhongchang Tong 仲長統 (180-220 CE). The passage is preserved in the “Zhi li”至理 chapter of the Inner Chapters of the Baopuzi.159 The Liezi version is as follows:昔人言有知不死之道者,燕君使人受之,不捷,而言者死。燕君甚怒,其使者將加誅焉。幸臣諫曰:「人所憂者莫急乎死,己所重者莫過乎生。彼自喪其生,安能令君不死也?」乃不誅。160Formerly, there was a man that said he knew the way to never die. The Lord of Yanordered someone to get it, [but] he was not swift and the one who said it died. TheLord of  Yan was very angry at  the one he had ordered [to get  the method],  andplanned to have him executed. A favourite minister remonstrated, “Among the thingspeople worry about, there is nothing more worrisome than death; among the thingsthey take as important, there is nothing that exceeds [in importance] than their own156 Wang (2002), pp. 15-16.157 A now mostly lost literary treatise attributed to Cao Pi.158 A similar version of events is presented in the Soushenji 搜神記, for example. See Wang (1979), pp. 165-166.159 Wang (2002), p. 115. 160 Yang (2007), p. 268. 86lives. He [that claimed to know how not to die] himself lost his life – how could he beable to have you not die?” And so [the man Lord Yan had sent] was not executed.The rest of the Liezi 8:28 offers two comments on this particular narrative, one of which also closelymirrors content from the Kongcongzi. But the above passage is found in nearly the same form with anidentical meaning in the  Baopuzi,  attributed to Zhongchang Tong and his  Chang yan.  Incidentally,though the two passages diverge in the style and wording of their  conclusions, both also offer thesuggestion  that  the  practitioner  of  immortality  that  died  did  not  necessarily pass  away due  to  histechnique being wrong – it was just that he could not perform the technique correctly. If this narrativedoes belong to the writings of Zhongchang Tong as Ge Hong suggests, then it is another instance ofpost-Warring States material to be found in the Liezi.161 Finally, it is valuable to dedicate some space to a brief evaluation of the work of Ma Da on thequestion of textual parallels between the  Liezi and the received corpus. Ma offers an extensive anderudite selection of parallels between the Liezi and texts from the Warring States up into the Jin period,and in every case suggests that it is the  Liezi  that has priority. His estimation mirrors that of ZhangZhan, in that he asserts that in all cases of textual parallels the Liezi serves as the original. Despite theundeniable breadth of scholarship offered, I ultimately believe that Ma's reasoning is faulty. That is tosay, while his work is exceptional and exhaustive in its presentation of parallels the arguments intendedto convince the reader of the Liezi's authenticity regularly lack merit when compared to those offeredby Yang or Graham. A citation by citation discussion of Ma's book would be an enormous undertaking,and so I hope to offer some representative examples in place of such a project. Because the thought ofRuan Ji makes up a significant portion of Chapter Four of this dissertation, I elect to focus on Ma's161 Arthur Frederick Wright, in a note to his translation of Etienne Balazs' essays, has also suggested a link between thewritings of Bao Jingyan 鮑敬言 (early third century?) preserved in the Baopuzi waipian and Liezi 7.8 in the “Yang Zhu”chapter, which both praise the virtues of a state without the “Way of Lord and Minsiter” ( junchen zhi dao 君臣之道).While this characterization is accurate, the Bao Jingyan passage is descriptive of a utopian ideal in the past, while theLiezi passage is more prescriptive in nature. See Wright's note in Balazs (1964), p. 244n22. 87appraisal of these works in relation to the Liezi.162 Ma offers three examples of parallels between the works of Ruan Ji and the Liezi.163 His firstnotes that the Ruan Ji uses the expression dahe 大壑 (meaning “great ocean”) in his Dongping Fu 東平賦; Liezi 5.2 also makes use of the expression dahe with the same meaning, and thus Ma concludes thatRuan Ji has found this expression in the  Liezi.164 The claim is spurious on at least two counts:  noevidence is offered that the compiler of the  Liezi could not have been influenced by the writings ofRuan Ji, and more importantly, Ma does not recognize that expression is also accessible to Ruan Jithrough its presence in the Zhuangzi, the Chuci, and the Shanhaijing.Ma's second example quotes from Ruan Ji's  Zouji yi Cao Shuang 奏記詣曹爽 , which statesthat: 昔榮期帶素,仲尼不易其三樂 . “Formerly there was Rong [Qi]qi dressed plainly – Zhong Ni[Kongzi] did not change [his mind about] his three joys.”165 I offer a translation of the Liezi pericope inChapter Four – here, it is sufficient to recount that the story is that Kongzi happens upon a shabbilydressed and destitute Rong Qiqi who nonetheless assures his interlocuter that he has three joys (beinghuman, being male, and being old). Ma acknowledges that a very similar text appears in the Huainanzi,Xinxu, and Shuoyuan, but all neglect the important details included in both the Liezi and Ruan Ji's work– namely, the plainness of Rong Qiqi's attire, his three joys, and Kongzi's praise. Such discrepancies,according to Ma, rule out these documents as a source for Ruan Ji, leaving only the Liezi. He ignoresthe fact that the same account, with all  these elements and very little variation, is available in theKongzi jiayu, which is at least contemporary with Ruan Ji. Certainly an authentic Liezi could serve as acommon source for both the Kongzi jiayu and Ruan Ji, but Ma's evidence here is at best circumstantialand suffers by not accounting for alternate hypotheses.162 Jean Lévi offers a review of Ma Da's material on Xi Kang. See Lévi (2014), pp. 171-172. 163 Ma (2000), pp. 289-290.164 Ma (2000), p. 289.165 Source text from Ma (2000), p. 289.88Finally, in his third point, Ma makes reference to a story in Liezi 8.25 in which Yang Zhu helpsa neighbour recover a lost sheep, only to be confounded by too many forks in the road.166 He links thisto the twentieth167 poem of Ruan Ji's Yonghuai shi 詠懷詩 collection, which does appear to referencethe event: 楊朱泣歧路 “Yang Zhu wept at the fork in the path”. This is then taken as evidence that theYang Zhu story from the Liezi serves as the basis for Ruan Ji's poem. Were the Liezi's textual history anuncomplicated  one  this  conclusion  would  be  much  more  plausible;  as  it  is,  the  assertion  isunconvincing in light of abundant evidence to the contrary. Based on the example offered by Ma in thisinstance we cannot draw a positive or negative conclusion, and as such the evidence is again lacking.The textual examples provided in Ma Da's volume are of this style. If we follow his reasoning,it  appears  that  the  Liezi text  serves  as  the  basis  for  many later  works.  But  he  offers  no  concreteevidence that this must be the direction of influence, and his many examples serve equally well insupporting the alternative hypothesis – that the Liezi likely draws on many pre and post Han sources.Balanced against other forms of evidence, the argument for an early-date Liezi falls apart.2.5 Traces of South Asian Thought in the LieziStaunch defenders of the legitimacy of the Liezi as a Warring States text are able to explain thepresence of seemingly borrowed content by suggesting that these texts had quoted the Liezi, much asZhang Zhan had suggested in his introduction. This is on its face entirely plausible – the parallels thatexist between the Liezi and the Zhuangzi, Lüshi Chunqiu, Hanfeizi, Mutianzi zhuan,  Kongcongzi, andChang yan could have originally been content derived from an eight  juan Liezi. This is only merelyplausible, however; that it is to say, it is not probable when considered in light of other evidence. It ispossible, perhaps likely, that these anecdotes and aphorisms were in the intellectual ether, not clearlybelonging to the Liezi or another text, but moving fluidly between texts and speakers. Content external166 Ma (2000), p. 290.167 Or twenty-third, by Ma's reckoning. I have followed the order as laid out in Holzman (1976) and Chen (1987).89to the Liezi but available in received sources makes up approximately one quarter of the text, accordingto  Graham,168 and  yet  because  of  the  generally  polymorphous  nature  of  texts  in  the  period  underconsideration the evidence found in the textual parallels discussed thus far only strongly suggests thelate compilation of the text, rather than conclusively proves it. However, the Liezi clearly also makesuse of content found in documents that can be traced back to the Indian subcontinent. This evidence, inconjunction with the grammatical evidence and parallels discussed above, should settle the matter.Zhang Zhan notes  in  his  preface  the  influence  of  Buddhist  thought,  though he  neglects  tomention any specific texts with which a reader could compare the Liezi, leaving us only to speculate asto how knowledgeable he was on this subject, or exactly which facets of Buddhist thought he had inmind. In the final section of this chapter I will discuss only a few clear textual parallels that have beendocumented in the course of debates as to the authenticity of the  Liezi,  with a view to discuss thepossible intellectual impact of Buddhist thought on this text in Chapter Five of the present work. Thatis to say, the material presented here is done so to contribute to the identification question; readersshould expect a more thorough conceptual analysis of this and other Buddhist material in the fifthchapter. Like Zhang Zhan's somewhat vague and perhaps circumspect indication of Buddhist thought inthe Liezi, I have found that (to my dismay) many writers addressing the content of the Liezi do not givethe important question of Buddhist thought in the Liezi adequate consideration, but merely mention it inpassing as an afterthought or curiosity.169With the above source of consternation in mind, I choose to begin with the most commonlycited example of a Buddhist interpolation into the  Liezi: the story of the automaton as found in the“Tang wen” chapter (Liezi 5:13). Almost invariably this is the pericope that is cited when the question168 Graham (1990b), p. 225.169 A. C. Graham's 1960 translation of the Liezi is perhaps the most frustrating example, if only because it is outstanding innearly every other way. This otherwise masterful examination of the text makes only occasional reference to Buddhismand Buddhist thought, all the more perplexing as Graham was undoubtedly aware of the important work done on thequestion, such as that of Chen Dan examined below.90of Buddhist thought in the Liezi is addressed, and perhaps with good reason – the conceptual parallelsare undeniable. Here I will relate both the Liezi version as well as that as found in the Shengjing 生經.In both cases I will highlight only the most relevant aspects, in the interest of brevity. First, a selectionfrom the Liezi 5:13. My translation below begins in media res, as King Mu of Zhou has just had theartisan Master Yan show him his greatest creation:王薦之,曰:「若與偕來者何人邪?」對曰:「臣之所造能倡者。」穆王驚視之,趣步俯仰,信人也。巧夫顉其頤,則歌合律;捧其手,則舞應節。千變萬化,惟意所適。王以為實人也,與盛姬內御並觀之。技將終,倡者瞬其目而招王之左右侍妾。王大怒,立欲誅偃師。偃師大懾,立剖散倡者以示王,皆傅會革、木、膠、漆、白、黑、丹、青之所為。王諦料之,內則肝、膽、心、肺、脾、腎、腸、胃,外則筋骨、支節、皮毛、齒髮,皆假物也,而無不畢具者。合會復如初見。王試廢其心,則口不能言;廢其肝,則目不能視;廢其腎,則足不能步。穆王始悅而歎曰:「人之巧乃可與造化者同功乎?」170The King gave him a straw mat and said, “Who is this other person that has comewith  you?”  He replied,  “This  is  what  I  have  made that  can  entertain.”  King Mulooked at it in shock, hastily stepping to look at it from bottom to top, believing it washuman.  The  craftsman  nodded171 the  chin  [of  the  artificial  man],  and  it  sang  aharmony;  he  held  the  hand,  and  it  danced  responding  in  time.  A  thousandtransformations and ten thousand changes – [one need]  only have the idea and itwould do it. The King took it to be a real person, and he observed it riding in thechariot alongside Sheng Ji.172 The show was about to end and the entertainer winkedits eye and beckoned to the concubines surrounding the King. The King was greatlyangered, and on the spot he desired to punish Master Yan. Master Yan was greatlyafraid, and immediately opened and took apart the entertainer to show the King; itwas completely made of leather, wood, glue, lacquer, and coloured white, black, red,and blue. The King closely examined it: on the inside there was a liver, gall bladder,heart, lungs, spleen, kidney, intestines, and stomach; on the outside there was muscleand bone, fingers and joints, skin and hair, teeth and hair on the head. They were allfalse things, but none among them were not whole and complete. He had it broughtback together again, as he had seen it before. The King tried to take out its heart-mind, and its mouth was not able to speak; he took out his liver, and his eyes couldnot see; he took out his kidneys, and his feet could not walk. King Mu was finally170 Yang (2007), pp. 179-180.171 The character  qin 顉  here means something like “nod” or “bow”, and is probably used in the causative sense. YangBojun helpfully notes variants among the Liezi editions (see Yang (2007), p. 179). Possible substitutions are han 頷 orzhen 鎮; the former offers little change in meaning, while the latter means “press”, and seems to be the choice Grahammakes in his translation (see Graham (1990a), p. 110). I follow Yang, but note that the choice does not impact the overallmeaning or usefulness of the passage.172 A concubine of King Mu of Zhou.91relaxed and sighed, saying, “The skill of humans – can it in fact accomplish the samethings as as that which makes and transforms [i.e., the “Creator” or zaohuazhe 造化者]?”173The  King  packs  up  the  artificial  person  and  the  remainder  of  the  section  is  a  commentcomparing the skill of others to the skill of creating this automaton. Let us compare this story with thatwhich we find in the Shengjing, which was translated by Dunhuang 敦煌 born translator Dharmarakṣa(Zhu Fahu  竺法護 , ~230-316 CE) in the Western Jin period, around 285 CE. The relevant portionfollows:應時國王, 喜諸技術,即以材木,作機關木人,形貌端正, 生人無異,衣服顏色,黠慧無比, 能工歌舞,舉動如人,辭言:『我子生若干年, 國中恭敬,多所餽遺。』國王聞之,命使作伎, 王及夫人,升閣而觀。作伎歌舞若干方便, 跪拜進止,勝於生人。王及夫人,歡喜無量。 便角䁯174眼,色視夫人。王遙見之, 心懷忿怒,促敕侍者:『斬其頭來。 何以䁯眼視吾夫人?謂有惡意,色視不疑。』其父啼泣, 淚出五行,長跪請命:『吾有一子,甚重愛之, 坐起進退,以解憂思,愚意不及,有是失耳。 假使殺者,我共當死,唯以加哀, 原其罪舋。』時王恚甚,不肯聽之。復白王言: 『若不活者,願自手殺,勿使餘人。』王便可之。 則拔一肩榍*,機關解落,碎散在地。 王乃驚愕:『吾身云何瞋於材木?此人工巧, 天下無雙,作此機關,三百六十節,勝於生人!』175In response to the current King, who rejoiced in technique and skill, he [“Prince ofSkill”]  then  made a  mechanical  wooden person from timber,  [with]  its  form andappearance correct and from a living person lacking any differences; it was clothedand coloured [correctly], [and] its shrewd intelligence was without comparison. It wasable to sing and dance, and move like a person. [The maker] said, “My son has livedfor several years, within the state he is respected and has received many gifts andhonours.” The King heard this, and ordered him to demonstrate his talent. The Kingand Queen ascended a tower to watch. He demonstrated his talent by singing anddancing in several ways, kneeling and bowing, advancing and stopping, better than aliving  person.  The King and Queen watched and enjoyed without  limit.  Then helooked back and winked his eye,176 lustfully looking at the Queen. The King saw thisfrom afar, and in his heart-mind he felt anger, he urged and ordered those servinghim, “Cut off his head! Why do you wink looking at my Queen? I say there is a173 “That which makes and transforms” (zaohuazhe 造化者) is perhaps a gesture of affiliation, and is found with frequencyin the Huainanzi. Its locus classicus appears to be the Zhuangzi “Da zong shi” chapter.174 CBETA has this as [目*翕]; I have made the change to 䁯 both here and in its one subsequent appearance. 175 From CBETA T03no154, with only very minor changes as noted above.176 The exact translation here is tentative, but captures the basic gist of the passage.92wicked intention here, and do not doubt that glance was lustful.” His father howledand cried, the tears coming out in five streams. He knelt and implored, “I have asingle son, and many greatly love him; sitting and rising, advancing or withdrawing,by this I resolve sorrowful thoughts – simply by ignorantly thinking this would nothappen there was this mistake. If you kill him I am to be killed along with him. Onlyby  your  pity  could  he  be  forgiven  of  this  crime.”  At  this  time  the  King's  rageincreased and he was not willing to listen to him. Again explaining to the king hesaid, “If he cannot live, I request he die by my own hand, and that you not commandanother [to do it].” The King agreed. Then he pulled out a single wedge177 from theshoulder, and the machine fell apart, going to pieces upon the ground. The King wasthen alarmed: “How can I speak of being angry at bits of wood? This artisan – underHeaven he has no match. In making this machine, the three hundred and sixty jointsare better than a living person!”Though the parallels here are not as grammatically striking as those between the Liezi and theZhuangzi examples cited above, the basic narrative is certainly the same. In both versions an artisandemonstrates for a ruler his mechanical person, and in both versions this mechanical person makes anunwanted advance towards the ruler's consort, provoking his anger. In each case the exchange endswith the ruler's anger being abated by his realization that the mechanical person is not legitimatelythreatening, and instead he praises the artisan as being without equal.  The resemblances are muchgreater than the differences, though some exist. For example, in the Liezi version King Mu intends tohave Master Yan punished, while the Sheng jing version suggests that the mechanical person will bearthe brunt of the punishment. One may only speculate as to how the compiler of the Liezi was exposedto this story, but in light of the other evidence it is plausible that he had either read or heard recited thistale, and adapted it for his own purposes.In his Liezi Yang Zhu pian weishu xinzheng 列子楊朱篇偽書新證178 Chen Dan 陳旦 arguesthat material found in the “Yang Zhu” chapter of the Liezi resembles Buddhist material closely enoughthat it is reasonable to suppose that it may have served as a source for that document. A close reading ofmaterial suggests that the evidence is not as conclusive as that found in the Sheng jing narrative, but is177 Reading xie 楔 for xie 榍. 178 Included in Yang (2007), pp. 311-318.93perhaps further corroborating evidence to the idea that the  Liezi was compiled at least in part fromBuddhist materials. Chen Dan quotes from Liezi 7:1, 7:2, and 7:3. Below I translate 7:3, the most relevant of thepassages quoted, in its entirety.楊朱曰:「萬物所異者生也,所同者死也。生則有賢愚、貴賤,是所異也;死則有臭腐、消滅,是所同也。雖然,賢愚、貴賤非所能也,臭腐、消滅亦非所能也。故生非所生,死非所死;賢非所賢,愚非所愚,貴非所貴,賤非所賤。然而萬物齊生齊死,齊賢齊愚,齊貴齊賤。十年亦死,百年亦死。仁聖亦死,凶愚亦死。生則堯舜,死則腐骨;生則桀紂,死則腐骨。腐骨一矣,孰知其異?且趣當生,奚遑死後?」179Yang Zhu said, “As for the myriad things what they differ in is their life, and whatthey are the same in is their death. In life they are talented or stupid, honoured orlowly, and this is how they differ; in death they are stinking and rotting, decaying anddecomposing,  and  this  is  how  they  are  the  same.  However,  [as  for]  talent  andstupidity, honour and lowliness, it is not of their own capabilities; [as for] stinkingand rotting, decaying and decomposing, is also not of their own capabilities. So: lifeis not what generates, dying is not what kills, talent is not what grants talent, stupidityis not what grants stupidity, honour is not what grants honour, lowliness is not whatgrants  lowliness.  However,  the  myriad things  are  equally alive and equally dead,equally talented and equally stupid, equally honoured and equally lowly.180 [To live]ten years and then die or a hundred years and then die – the benevolent and sagelysurely die, [and] the unlucky and stupid surely die. If alive one is Yao or Shun, thenwhen dead they are rotting bones; if alive one is Jie or Zhou, then when dead they arerotting bones. Rotting bones are as one, who knows of their differences? For nowenjoy your current life – why be concerned with what happens after you die?”Chen Dan sees in this passage conceptual parallels to material found in the Chang ahan jing 長阿含經, (Dīrgha Āgama or Long Discourses), specifically the Shamen guo jing 沙門果經.181 The textis attested in Pāḷi as the Sāmaññaphala Sutta and Sanskrit as the Śrāmaṇyaphala Sūtra. The entirety ofthe Dīrgha Āgama was translated into Classical Chinese in 412 CE by Buddhayaśas 佛陀耶舍 (fl. early5th century CE) and Zhu Fonian 竺佛念 (fl. early 5th century CE). A date even in the early fifth century179 Yang (2007), p. 221.180 Graham reads these three sentences as a “Daoist” interpolation similar to the content of the “Li ming” chapter. SeeGraham (1990a), p. 141. 181 Chen in Yang (2007), p. 315-316.94seems to put it too late to have an influence on the Liezi's composition; however, Chen Dan also makesreference to an earlier translation by Zhu Tanwulan 竺曇無蘭 (fl. late 4th century CE), transmitted asJizhi guo jing 寂志果經 ,182 which could have translated between 381 and 395 CE.183 One furthertranslation not mentioned by Chen is the unofficially titled Wugen xin 無根信, perhaps from as early as384 CE.184 Even these earlier  dates are rather late if  the conventional  Liezi sceptic's  chronology isaccepted,  conceiving  its  compilation  as  having  taken  place  in  the  mid  to  late  4 th century.  Chen'shypothesis would seem much more tenable if one were to suppose an even earlier translation of thisdocument, perhaps now lost to us.185 Given the number of received translations, this is not implausible.In either case, it is worthwhile to assess the conceptual parallels with a version that does exist in thereceived record. In his analysis, Chen Dan draws on the English translations made by T. W. Rhys Davids, 186though Rhys Davids' translation is of the Pāḷi text, not Classical Chinese. The Shamen guo jing itself isa record of King Ajātaśatru (Ashishi wang 阿闍世王) questioning various non-Buddhist philosophersabout the rewards inherent in the life of renunciation, ultimately arriving at a satisfactory answer in hisconversation with the Buddha. In the course of his investigation, he rejects the advice of the othervarious philosophers, who teach forms of materialism or fatalism. Because it is the words of theseheterodox teachers that Chen views as influential to the Liezi, it is perhaps important to note that in thiscase it seems that the Liezi bears the influence of “Buddhist texts” rather than “Buddhist thought”. The following translation is of the  Shamen guo jing response of Ajita Keśakambalin (Ayituo182 Chen in Yang (2007), p. 312.183 MacQueen (1988), p. 17.184 This document is found in T.124. See MacQueen (1988), pp. 17-18.185 For example, the important translator of Buddhist works Lokakṣema (Zhi Loujiachen 支婁迦讖, 2nd century CE) wasknown to have translated a text with the same subject, King Ajātaśatru, and similar (but not identical) content. This text,the Ashishewang jing阿闍世王經, was translated in the late Han dynasty. Regrettably, there is insufficient space here topursue the matter further.186 Chen mistakenly takes Max Müller as the translator of this volume; rather, he was the editor of the Sacred Books of theEast series to which Rhys Davids' translation belongs. See Chen in Yang (2007), p. 313.95Chisheqinpoluo 阿夷陀翅舍欽婆羅, ~6th century BCE)187 to the King's question. Ajita Keśakambalin isgenerally characterized as a proponent of materialism and non-existence after death.188 Chen Dan hasargued that his presentation of death here may contain some roots for the words of Yang Zhu translatedabove. 『受四大人取命終者,地大還歸地,水還歸水,火還歸火,風還歸風,皆悉壞敗,諸根歸空。若人死時,床輿舉身置於塚間,火燒其骨如鴿色,或變為灰土,若愚、若智取命終者,皆悉壞敗,為斷滅法。』189When a person that has received the Four Great Elements190 obtains the end of theirlife, the Earth element returns to earth, the Water [element] returns to water, the Fire[element] returns to fire, the Wind [element] returns to wind. Everything is destroyed,the sense faculties191 return to the void.192 If someone has come to their hour of death,a cart193 carries the body as it is put in the graveyard, fire scorches their bones [until]they are the colour of a pigeon194 or they have changed into ash. Whether stupid orwise [one] obtains the end of their life, and everything is destroyed, eradicating allthings.195Particularly relevant to the discussion of the Liezi passage is the emphasis on the impartiality ofthe dying process – as in Yang Zhu's declaration, both those possessing superior mental faculties (xian賢 in the Liezi and zhi 智 in the Shamen guo jing) and those with defective mental faculties (yu 愚 inboth texts) are fated to the same end. This answer is unsatisfactory to the King, who ultimately finds187 This  is  the  transliteration  used  in  the  present  Shamen  guo  jing text.  Another  common  transliteration  is  AqiduoChisheqinpoluo 阿耆多翅舍欽婆羅.188 MacQueen (1988), pp. 152-153, makes clear that the configuration of Ajita Keśakambalin's views are consistent in thePāḷi version of the text, as well as the  Shamen guo jing version; in other Classical Chinese translations, this view isvariously attributed to other philosophers, greatly complicating the question. Because Chen Dan has used the Shamenguo jing version in his analysis I have done so as well.189 CBETA T01no1, juan 17.190 Si da 四大 is a technical term, and comprises the four elements (earth, water, fire, and wind) described below.191 “Sense faculties” here translates zhugen 諸根. Commonly they include the five basic sense organs/sensations of the ears(hearing), eyes (seeing), nose (smell), mouth (taste), and the skin (touch). 192 “Void” here translates  kong 空 . I have avoided the more standard “emptiness” for  kong, as that term is laden withmeaning in Buddhist thought, much of which likely does not apply to the teachings of Ajita Keśakambalin here - “void”is intended to be more philosophically neutral, with the intention of leaning towards the materialistic implications of theargument.193 Literally “bed-cart”, presumably some manner of cart on which a corpse lay in a reclined position. 194 That is, white or pale in colour, bleached by their exposure after death. 195 “All things” here translates fa 法, another crucially important technical term in Buddhist thought (see note the note onkong above). 96the Buddha's teachings to be superior. Yang Zhu's teachings on the question are not challenged directlyin the above passage from the Liezi, but as both Liu Xiang and Zhang Zhan suggest in their prefaces,the entire “Yang Zhu” chapter appears to be heterodox in nature. While the argument made by ChenDan is not as conclusive as that made in reference to the Sheng jing material, it still may be indicativeof the Fojing 佛經 material Zhang Zhan mentions in his preface to the Liezi. In addition to the sources discussed above, there is the suggestion that some material originatingfrom the  Indian  subcontinent  found  in  the  Liezi is  not  derived  from  explicitly  Buddhist  texts.196Lorenzan and Maeth have suggested that at least one section of Liezi 7.2 may in fact be a stanza of theŚatakatraya of poet Bhartṛhari.197 Below is the relevant portion of  Liezi 7.2, a typical passage of the“Yang Zhu” chapter.楊朱曰:「百年,壽之大齊。得百年者千無一焉。設有一者,孩抱以逮昏老,幾居其半矣。夜眠之所弭,晝覺之所遺,又幾居其半矣。痛疾哀苦,亡失憂懼,又幾居其半矣。量十數年之中,逌然而自得亡介焉之慮者,亦亡一時之中爾。則人之生也奚為哉?奚樂哉?」198Yang Zhu said, “One hundred years is the extent of a life, [but] those who obtain onehundred years are not one in a thousand. If there is one [who does live this long], theyspend almost of this time in childhood or old age. Nights extinguished in sleep, [and]days lost while awake – [in this they] spend almost a further half [of their life]. Pain,illness, grief, suffering, loss, failure, worry, and fear – [in this they] spend almost afurther half [of their life]. Of the dozen years in the middle, [those] that are free and atease, lacking even a small199 worry – indeed there is not even an hour among them.Then as for a person's life – what are they to do? Where is joy?”Lorenzen and Maeth describe their fortuitous discovery of a very similar sentiment in the work ofBhartṛhari:The span of man's life is a measured hundred years; / Yet half is lost to night / And of196 Whether a text was “Buddhist” likely mattered very little to the early medieval compiler of the Liezi, but would perhapsbe relevant to medieval Buddhists in China.197 See Lorenzen and Maeth (1979). Bhartṛhari may also be romanized “Bhartrihari”.198 Yang (2007), p. 219. 199 Reading jie 介 as wei 微, in accordance with the shiwen 釋文 commentary of Yin Jingshun 殷敬順 and Chen Jingyuan陳景元. See Yang (2007) p. 219.97his waking time, / A portion each claim callow youth and hoary age; / His prime isspent in servitude, suffering / The anguish of estrangement and disease. / Where domen find happiness / In life less certain and more transient than the waves?200The parallels here are even more striking than those noted by Chen Dan above, though not asimpressive as  those found between the  Liezi  and the  Shengjing.  Most  of  the core components  arepresent, though arranged and expressed differently. Lorenzen and Maeth note, however, that their thesis– that the Liezi compiler adapted this material from the work of  Bhartṛhari – is complicated by the factthat  Bhartṛhari  is  commonly believed to  have  worked in  the  mid-seventh  century.201 They offer  anumber of solutions to this problem. One is that the earliest report of Bhartṛhari, by the travelling monkYi Jing 義淨 (635-713 CE) is in error, either conflating this poet with a grammarian of the same name,or simply wrong about the dates of his life.202 They suggest a date of approximately 450 CE for thedeath of Bhartṛhari, which does not resolve the chronology problem.203 A more plausible solution is thatthe poems of Bhartṛhari are not the compositions of a single individual, but instead are a compilation ofworks from various poets on similar themes.204 If this were the case, then the poem in question mayhave found its way to the  Liezi compiler before even  Bhartṛhari.  This is entirely speculative, andconsidering  the  absence  of  the  Śatakatraya in  Classical  Chinese,205 the  question  of  the  similarityapparent here is difficult to resolve.Moreover,  there  may  exist  in  the  Chinese  tradition  plausible  antecedents  to  the  notionsexpressed  in  the  “Yang  Zhu”  chapter  above.  While  they  do  not  match  the  account  found  in  theŚatakatraya as closely, they do reflect musings on this notion indigenous to China. In the  Baopuzineipian we read this reflection on the brevity of life:200 Translation from Bhartrihari: Poems by Barbara Stoler Miller (1967), p. 147, which includes the romanized Sanskrit.For a more recent translation, see  Satakakāvyas of  Bhartṛhari by Rabindra Kumar Panda (2006),  pp. 72-73, whichincludes the Sanskrit in the original Devanagari script.201 Lorenzen and Maeth (1979), p. 699.202 Lorenzen and Maeth (1979), p. 699.203 Lorenzen and Maeth (1979), p. 705.204 Lorenzen and Maeth (1979), p. 699.205 Lorenzen and Maeth (1979), p. 705.98幼弱則未有所知,衰邁則歡樂並廢,童蒙昏耄,除數十年,而險隘憂病,相尋代有,居世之年,略消其半,計定得百年者,喜笑平和,則不過五六十年,咄嗟滅盡,哀憂昏耄,六七千日耳,顧眄已盡矣,況於全百年者,萬未有一乎?206When young one does not yet know [things], [and] with gradual decline happinessand joy both disappear. [So] ignorant youth and the darkening of age do away withseveral  decades,  and  peril,  danger,  grief  and  disease  take  turns  succeeding  oneanother.  Of [one's]  years dwelling in  the world,  [these]  take about  half.  One thatanticipates living a hundred years is happy and peaceful, yet in not [succeeding inliving] more than fifty or sixty years they are, alas, extinguished. In mourning, grief,and the darkening of age they have merely six or seven thousand days – with but aglance  they are  gone.  How much  more  so  the  case,  since  those  that  have  a  fullhundred years are not one in ten thousand?We also read in the “Dao zhi” chapter of the Zhuangzi, in which the eponymous bandit chastisesKongzi:人上壽百歲,中壽八十,下壽六十,除病瘦死喪憂患,其中開口而笑者,一月之中不過四五日而已矣。207For people, the uppermost limit is a hundred years of age. Those in the middle haveeighty years,  and at  the  least  they have  sixty years.  Excluding [time]  spent  withmalady, death, and anxiety, instances of opening one's mouth and laughing simply donot exceed more than four or five times in a month.Though we may not take these two examples as textual sources for the “Yang Zhu” chapter ofthe Liezi, they may serve alongside the work attributed to Bhartṛhari as a textual antecedent. I believean elaboration on these themes in China, taking place alongside similar developments in India, is themost plausible standpoint in this case. 2.6 ConclusionBecause  of  the  polymorphous  nature  of  the  Liezi,  the  authenticity  question  may  never  beentirely resolved. Even when we can reliably trace a passage to a source demonstrably post-Qin, thatevidence is only conclusive for that particular pericope. Much material can be traced to earlier sources,206 Wang (2002), p. 253.207 Guo (2004), p. 1000.99and these lend some credibility to the claim that a great deal of the Liezi narratives and arguments datefrom the early Warring States, as Zhang Zhan would have readers of his preface believe. But although the varied components of the text cannot all be reliably traced, I believe that theLiezi's compilation, as we now find it in the received record, can be discerned with a good degree ofconfidence.  Taking  the  various  lines  of  evidence,  be  they  grammatical  or  otherwise  linguistic,  aspresented by Yang Bojun and A. C. Graham, or be they in the analysis of parallel passages, as they existbetween the Liezi and the Zhuangzi, Mutianzi zhuan, or texts of India