UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Blo gsal grub mtha' MacDonald, Anne Elizabeth 1988

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

Item Metadata


831-UBC_1988_A8 M3.pdf [ 8.85MB ]
JSON: 831-1.0097719.json
JSON-LD: 831-1.0097719-ld.json
RDF/XML (Pretty): 831-1.0097719-rdf.xml
RDF/JSON: 831-1.0097719-rdf.json
Turtle: 831-1.0097719-turtle.txt
N-Triples: 831-1.0097719-rdf-ntriples.txt
Original Record: 831-1.0097719-source.json
Full Text

Full Text

BLO GSAL GRUB MTHA'  <$ov\tc*~S<L # (  by ANNE ELIZABETH MACDONALD B.A., McMaster University, 1979 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T O F T H E REQUIREMENTS FOR T H E D E G R E E O F MASTER O F ARTS in T H E F A C U L T Y O F G R A D U A T E STUDIES D E P A R T M E N T O F RELIGIOUS  STUDIES  We accept this thesis as conforming to the required standard  T H E UNIVERSITY  O F BRITISH COLUMBIA  November ©  1988  Anne Elizabeth MacDonald , 1988  In  presenting  degree freely  this  at the  thesis  in  partial  University  of  British Columbia, I agree  available for reference  copying  of  department  this or  thesis for by  his  or  fulfilment  of  and study. I further  the  scholarly purposes may be her  permission.  Department The University of British Columbia Vancouver, Canada  DE-6 (2/88)  Clc^f-  AD/tfr  that the  representatives.  for  an  It  is  granted  by the  understood  be allowed  advanced  Library shall make  agree that permission for  publication of this thesis for financial gain shall not  Date  requirements  that  it  extensive  head of copying  my or  without my written  ii  Abstract Bio gsal grub  mtha':  Translation and Study of a Fourteenth Century Grub mtha' Text This thesis presents the translation and study of the twelfth section of Bio gsal grub mtha',  an early fourteenth century Tibetan text composed by the  bKa' gdams pa scholar, dBus pa bio gsal. Bio  gsal  grub  mtha'  literature known as literature.  as a whole represents a distinct sort of scholarly  Grub  Tibetan Grub  mtha' mtha'  that finds its roots in Indian  siddhanta  texts set forth, as the name in translation  reveals, the "established tenets" of various Indian, Tibetan, and occasionally Chinese philosophical schools.  The section of Bio gsal grub mtha' translated  here presents the tenets of the Madhyamika school of Tibetan Buddhism in general,  and their  fourteenth  century  bKa' gdams  pa manifestation  in  particular. The central tenet of Madhyarnika philosophy is that all phenomena are empty of self-nature.  Even that which is discovered to be the ultimate -  emptiness (stong pa nyid, Hunyata) - is. also said to be devoid of any real selfnature.  A l l phenomena are dependent-arisings, lacking reality, existing like  dreams and magical illusions.  These assertions are discussed in detail in the  translation and in the second part of the introduction. Of special interest to scholars of both Indian and Tibetan Buddhism, however, is dBus pa bio gsal's^classification of the Madhyamika subschools. The early Tibetan Buddhist scholars took upon themselves the task of categorizing and inventing names for the various Madhyamika "schools", and dBus pa bio gsal's classification represents the development of such thought to the fourteenth century. The introduction elucidates both dBus pa bio gsal's divisions of the Madhyamika sub-schools and elaborates on earlier and later classifications set forth by Tibetan scholars.  The investigation provides  iii insight into both the tenets of the Madhyamika school and the attempts of the Tibetans to arrange the previously unclassified Madhyamika subschools in a manner that would render them more logical and accessible to themselves and to future generations of scholars.  iv  CONTENTS Abstract  ii  Contents  iv  Acknowledgements  v  List of Abbreviations  vi Part I:  Introduction  The Madhyamika School 1.  Inception and Development in India  1  2.  Madhyamika into Tibet  7  3. The bKa' gdams pa Sect  12  4. dBus pa bio gsal  14  Grub mtha' Literature 6. Bio gsal grub mtha'  5.  16  25  7.  Early Categorization of Madhyamika Subschools  28  8.  Development of Madhyamika Classification  36  The Madhyamika View 1. The Middle Way  43  2.  48  The Madhyarnika Object of Negation  3. The Two Truths  53  4.  57  Real and False Conventional Truths  5. Ultimate Truth  59  6.  Dependent-Arising  63  7.  Divisions of the Madhyamikas:  8.  Sauttantika-Madhyamika  Yogacara-Madhyamikas, and 'Jig sden grags sde phyod pa  66  Svatantrika and Prasangika  69  Part II: Translation  Bio gsal grub mtha' Bibliography  XII: Madhyamika  76  164  V  Acknowledgements I wish to express my gratitude to Dr. Shotaro Iida for his instruction, advice, and guidance, as well as for his unflagging support and enthusiasm toward this project. I wish to thank Dr. Ashok Aklujkar for his excellent instruction in Sanskrit. I want to thank Dr. Mosca for the efforts he put forth in his role as Graduate Advisor, the other members of the Dept. of Religious Studies at U.B.C. for their support and valuable instruction in the course of my studies, and the University of British Columbia for its financial support. I owe thanks to Gareth Sparham for the many hours he devoted to explanation of the finer points of Madhyarnika philosophy, and for his proofreading and editing. My thanks to Dr. Ken Bryant for permission to use his Mcintosh Hindustan font. I also wish to express my appreciation to Geshe Tashi Namgyal for the years of patient instruction in Tibetan language and Buddhist philosophy.  I thank my parents for their broadmindedness and  support, and I thank Michael Madrone for his proofreading, encouragement, and patience.  vi ABBREVIATIONS AAA  Abhisamayalamkaraloka  of Haribhadra, ed.  Wogihara (1932). ACS  Acintyastava Sanskrit.  of Nagarjuna, Tib. ed., reconstructed Patel (1932).  ADSPP  AstMaSasahasrika-prajnliparanuta.  AK  Abhidhaimakos'a(karika) of Vasubandhu, Sanskrit ed. with AKBh, cf. Gokhale (1946); Tib. ed. of chap. MIL Fukuhara (1973).  AKBh  Abhidharmakos'abhasya  of Vasubandhu, Skt. ed.  Pradhan (1967). AMN  Aksayamatinirdes'a.  AP  Alambanapanksa of Dignaga, Tib. ed., frag, and reconstructed Sanskrit by S. Yamaguchi (1953) pp. 1-13.  AS  AntaraSloka  BCA  Bodhicaryavatara  of Santideva, Skt. Tib. ed. by V .  Bhattacharya (1960). BCAP  BodhicaryavatarapafTjika  of PrajHakararnati, Skt.  ed. by Vaidya (1960). BCV  Bodhicittavivarana  BriK I  First  of tan trie Nagarjuna.  Bhavanakrama  of Kamalasila, Skt. Tib. ed. by  Tucci (1958). BhKN  Bhavanakrama  BMP  Bodhimargapradlpapahjika  BPP  Bodhipathapradlpa of AtlSa, Tib. ed. by Eimer  of tan trie Nagarjuna. of Atisa.  (1978).  mtha' of dBus pa bio gsal.  BSGT  Bio  C. or [C]  Co ne edition. The numbers in parentheses  gsal grub  immediately after C. indicate the inventory number in Mibu (1959) for the bKa' 'gyur. CKGT  ICang skya grub mtha',  Sarnath ed.  vii CS  Catuhs'ataka  of Aryadeva,  Tib ed., frag, and  reconstituted Skt. by Vaidya (1923). DBh  Das~abhumika,  Skt. ed. by Rahder (1926).  D. or [D]  sDe dge edition. The numbers in parentheses immediately after D. indicate the inventory number in Tohoku.  H. or [H]  lHa sa edition. The numbers in parentheses immediately after H . indicate the inventory number in Takasaki (1965).  JIP  Journal  of Indian  Philosophy.  JNA  Jilana§rimitranibandhavali,  JNA (SSS)  SakarasiddhiSastra  JNA (SSS)  Sakarasamgrahasutra  JSS  Jffanasarasamuccaya  ed. by Thakur (1959).  of Jffimasnrrutra. of Jnlinas'iimitra. of Aryadeva, Tib. ed. with  JSSN. JSSN  Jflanasarasamuccaya-nibandhana  of Bodhibhadra,  Tib. ed. in Mimaki (1976). k  karika  KP  KaSyapaparivarta,  Skt. Tib. Chin. ed. by Stael-  Holstein (1977). Lank  Lankavatarasutra,  LAS  Lokatitastava  ed. by Vaidya (1963).  of Nagajuna, reconstructed Skt. by  Patel (1932), Tib. ed. by LVP.Q913). LRCM  Lam rim chen mo of Tsong kha pa, Peking ed.  LS  Drang nges legs bshad nyings po of Tsong kha pa,  Sarnath ed. LS(R)  dGe bshes Rab brtan's commentary on the LS, see Drang nges mam 'byed...see Tib. Bibl.  LVP  la Vallee Poussin, Louis de  MA  Madhyamakalankara  MAI  Madhyamakaloka  MAM  of Santaraksita. of KamalaSlia.  Madhyamakalamkara-vrtti-  madhyamakapratipadasiddhi MAP  Madhyamakalankarapanjika  of RatnakaraSanti. of Kamalaslla.  viii MAS  Madhyamakarthasamgraha  of the second (?)  Bhavaviveka, Tib. ed. in Ejima (1980). MAU  MadhyamakalamkaropadeSa  MAV  Madhyamakalankaravrtti  MAv  Madhyamakavatara  of Ratnakara^anti. of Santaraksita.  of Candraklrti, Tib. ed. with  MAvBh. MAvBh  Madhyamakavatarabhasya  of Candraklrti, Tib. ed  by L V P (1907-1912). MH  Madhyamakahrdaya  of Bhavaviveka, chap. III.  MM  Mulamadhyamakakarika  of Nagarjuna, Skt. ed. by  De Jong (1977), Tib ed. in LVP's notes (1903-13). MsTH  Manuscripts of Touen-houang.  Mtyut  Mahavyutpatti.  N. or [N]  sNar thang edition. The numbers in parentheses immediately after N. indicate the inventory number in Nagashima (1975) for the bKa. 'gyur and Miby (1967) for the bsTan 'gyur.  NB  Nyayabindu  NBh  Nyayabhasya  NBT  Nyayabindudka  of Dhannaklrti, ed. with NBT. of Paksilasvamin. of Dharmottara, Skt. ed. by  Stcherbatsky (1918), Tib. ed. by Stcherbatsky (1904). NVT  Nyayavarttikatatparyanka  P. or [P]  Peking edition.  of Vacaspatimisra.  The numbers in parentheses after P.  indicate the volume and the reproduction number. PPS  Pitaputrasamagamasutra.  PPU  Prajhliparamitopades'a  PrasP  Prasannapada  of Ratnakarasanti.  of Candraklrti, Skt. ed. by L V P  (1903-13), Tib. ed. by De Jong (1949) and May (1959). PS  Pramanasamuccaya  of Dignaga, Tib. ed. chap. I by  Hattori (1968), chap. II, III, IV, VI (partial) by Kitagawa (1965).  ix PV  Pramanavarttika  of Dharmaklrti, Skt. Tib. ed.  Miyasaka (1971-2). PVV  Pramanavarttikavrtti of Manorathanandin, Skt. ed. by D. Shastri (1968).  PVSPP RA  Paffcavims~atisaliasrika-prajhliparamita. Ratnavali of Nagarjuna, Skt. Tib. ed. by Hahn (1982).  RGSG  Ratnagunasamcayagatha,  Tib. ed. by Obermiller  (1960), Skt. ed. by Yuyama (1976).  SDVP  Satyadvayavibhangakarika Saryadvayavibhangavrtti of Satyadvayavibhahgapahjika  Siddhi  L a Vallee Poussin (1928-9).  SMVK  Sugatamatavibhangakarika of Jitari. Sugatamatavibhangabhasya of Jitari. Samdhinirmocanasutra, Tib. ed. by Lamotte Samadhirajasutra, ed. by Vaidya (1961). Siksasamuccaya of Santideva, ed. by Bendall  SDVK SDVV  SMVBh SNS SR SS  of JHanagarbha. Jfianagarbha. of Santaraksita.  (1935).  (1897-1902). SS  Subhasitasamgraha,  ed. by Bendall (1903) and  (1904). SSS  Sarvasiddhantasarngraha,  ed. by F.W. Thomas and  LVP. SSap SV TA TAv TAvV  TKh TRat  Sunyatasaptati of Nagarjuna. Slokavarttika of Kumarila Bhatta. Tattvaloka-nama-prakarana of Kamalasltla. Tattvavatara of Srigupta. Tattvavataravrtti of Srigupta. ITa ba 'i khyad par of Ye ses sde. Tattvaratnavali of Advayavajra, Skt. ed. by U i (1952).  TS  Tattvasamgraha(karika) TSP.  of Santaraksita, ed. with  Trisvabhavanirdes'a  of Vasubandhu, Skt. Tib. ed. in  Yamaguchi (1972); L V P (1932-3).  TattvasamgrahapaTljika  of Kamalasila, ed. by D.  Shastri (1968).  Vadanyaya  of Dhairnaklrti, ed. by D. Shastri  (1972).  VimSatikaXkarika) of Vasubandhu, see Vs"V. Vims'atikavrtti of Vasubandhu, Skt. ed. by Levi (1925), Tib ed. by G. Sasaki (1922).  Vigrahavyavartani  of Nagarjuna, Skt. ed. by  Johnston and Kundst (1951), Tib. ed. by Tucci (1929).  Vigrahavyavartanivrtti  of Nagarjuna, Skt. Tib. ed.  same as for V V .  YuktisastikaXkarika) of Nagarjuna. Yuktisastikavrtti of Candraklrti.  1 Introduction The  Madhyamika School:  Inception and Development in India  Early in the first millenium C . E . , the person recognized as the 1  founder of the Madhyarnika school of Buddhism composed a series of philosophical treatises that revolutionized Buddhist thought of his day and  left  a profound imprint on that of  the  following  centuries.  According to Tibetan tradition, he was born in South India of Brahmin 2  parents who, not long after his birth, were informed by an astrologer that their son would die at the age of seven.  During the year of his  predicted death they felt themselves incapable of bearing the sight of his lifeless body and so sent him off to travel with a servant.  The  servant and the child wandered awhile and, as luck or legend would have it, the young boy eventually found himself at the gates of the great monastery of Nalanda, where he was welcomed, advised to don the robes of a Buddhist monk, and allowed to secure the rites with which to cheat the Lord of Death.  The predicted time of demise passed without  event, and he commenced the study of Buddhist texts and practices under his preceptor Rahulabhadra, the abbot of Nalanda. As the years passed he became a scholar and teacher of such renown that even the Nagas, the mythical dragon-like beings from the Naga water-realms, came to listen to his discourses.  Reports of his expertise soon reached  the ears of their king, and an invitation to teach in the Naga kingdom was extended to the great scholar.  Inspired by mention of previously  unavailable Buddhist scriptures that were protected in the Naga world, the monk accepted the invitation to the watery kingdom.  He emerged  from it many years later, weighted with the Prajnaparamita, the  1. For the variation in scholarly dating for Nagarjuna's life, see D. Seyfort Ruegg (19,81), p. 4, n. 11. 2. For English translations see N . Roerich (1976); E . Obermiller (1932); D. Chattopadhyaya (1970). For other sources, see R.A.F. Thurman, (1984), p. 32, n. 22.  2  Avatamsaka, the Ratnakuta, and other books of the Mahayana school that had been guarded there from the time of the Buddha.  He became  known as Nagarjuna - "one who has achieved [his goal] with the aid of the dragons".  1  The Chinese Kumarajiva's version of the story is slighty different. It 2  does not include the prediction of an early death and instead reports that Nagarjuna was a bold and passionate youth who, together with two companions,  had versed  himself  in the  art of  magic,  and had  specifically mastered the art of invisibility, so as to gain entry to the royal harem. One night, however, upon the trio's stealthy entry into the harem (their secret having been revealed to the royal guards), the protectors of the harem violently slashed through the air with their swords, mortally wounding Nagarjuna's two invisible, yet vulnerable, cohorts.  Nagarjuna himself narrowly escaped.  Shaken by his brush  with death he reflected, and realized deeply that the origin of suffering is desire.  The experience inspired him to involve himself with the  Buddha's teachings, and soon thereafter he entered the Buddhist Order. Finding his subsequent study of all available Buddhist texts incapable of quenching his deep thirst for wisdom, he began a search for better  1. Thurman (1984), p. 24. Obermiller (1932), p.128, translates B u ston as follows, (In the name) Nagarjuna, Naga (has the following signification): 1. Born from (that ocean) which is the Essence, the Plane of the Absolute (just as the real Naga is bom in the sea), 2. not abiding in the two limits or extreme views of Eternalism and Nihilism (just as the real Naga knows no limits as regards his abode), 3. securing the possession of the treasury of the Jewels of Scripture (just as the Naga possesses immense wealth in gold and jewels), 4. endowed with an insight (that is like fire), burning down and illuminating (akin to the fiery eyes of the Naga). Arjuna has the meaning of "he who has secured power". the teacher is Arjuna since he is:  1.  Accordingly,  The guardian, the ruler of the  kingdom of the Doctrine and 2. The subduer of the hosts of enemies, that is of all the sinful powers of this world.  Being united, these two  component parts form the compound name Nagarjuna. 2. Other sources for Nagarjuna's life, including the following Chinese version, are cited in K . V . Ramanan (1978), p. 336, n. 5.  3  texts - a search that found its consummation in the discovery of the Mahay ana sutras in the Naga kingdom. Although the traditional accounts are numerous and encrusted with the weight of the miraculous, there is general agreement, based on archaeological, epigraphical, and literary evidence, that an historical Nagarjuna did exist. T . V . R . Murti writes, Though the traditions of his life are greatly overlaid with legendary  details,  there  is  no  reason  to  doubt  that  Nagarjuna was a real person. The circumstances of his life are briefly told.  He was, in all probability, a Brahmin  from the South who came to Nalanda and propogated the new PrajM-pararnita teaching.  The legend which credits  him with having brought the Satasahasrika from the abode of the Nagas means that he was the founder of a new and important phase in Buddhism.  All our accounts agree in  connecting his abode with Dhanyakataka or Sriparvata in the South, and of his personal friendship with the king Satavahana (Andhra) for whom he wrote the Suhrllekha. Tradition  places  him  four  hundred  years  after  the  parinirvana of the Lord, whereas the consensus of opinion among European scholars is that he lived about the middle of the second century A . D .  The  1  decorative elaborations of the legends aside, the historical  Nagarjuna remains one of the most influential and profound scholars of the Mahayana.  He is credited with being "one of the first and most  important systematizers of Mahayanist thought".  2  1. T.V.R. Murti (1955), p. 87-88. 2. D.S. Ruegg (1981), p. 7.  Although recensions  4 of the  Rrajiiaparamita-sutras  may have been in existence in a more  primitive form as early as the second century B . C . , it was Nagarjuna 1  who first, in clear and scholarly expositions, revealed their purport.  In  Mula-Madhyamakakaiikas, the Yuktisastika? the Sunyatasaptati, the Vaidalya-sutra, and the Vigrahavyavartani, he systematically explained the fundamental  treatises such  as the  2  famous  4  5  philosophical stance around which, according to him, the Mahayana  (gang zag gi bdag med, existence (chos kyi bdag med,  revolved, i.e., the theory of the emptiness of self  pudgalanairatmya) dharmanaiiatmya).  and of all elements of  One of the schools of thought that Nagarjuna's treatises engendered became known as the Madhyamika.  Madhyamaka, a Sanskrit, word,  has come to mean "that which proclaims the middle". Madhyamika school  Although the  maintained unique views, proclamation of the  middle was not an assertion new to Buddhist thought, for all the schools of Buddhist tenets that had come into existence in the years following the Buddha's sermon at Deer Park, in which he had referred to a middle way, claimed to follow a middle way. position  free  from the two  The Buddha had advocated a  extremes that had been part of the  experience of his life prior to his Enlightenment and that had, in fact, acted as hindrances to his attainment of peace and wisdom. Neither his youth as a prince blessed by the succulent fruits of cyclic existence nor the years he had spent as a self-mortifying ascetic in the Indian forests had brought him final peace, and thus at Deer Park he had taught the  1. Nakamura (1976), p. 75. 2. The Sanskrit text is found in Candraklrti's Prasannapada-Madhyamakavrtti.  Four  manuscripts are extant. 3. Preserved in Tibetan and Chinese.  Translated from Chinese into German by P.  Schaeffer (Materialien zur Kunde des Buddhismus 3, Heidelberg, 1924). 4. The verses of the Sunyatasaptati are found in the bsTan 'gyur in three separate and sometimes differing versions:  the karikas, the verses with a commentary by Nagarjuna, and  the verses of Candraklrti's Vrtti. 5. See K. Bhattacharya (1978) for an English translation of the verses and autocommentary. See also C . Lindtner (1982). For concise summaries of all these texts, see Ruegg (1981), p. 9-23.  5 Eightfold Path, the middle way free from the extremes of the radical lifestyles and views that had once acted as links in the chain that had bound him to samsara. The Madhyamikas considered themselves adherents to the Middle Way not so much because they progressed along a middle path, but rather because they understood the middle way of phenomena.  And in  their proclamation of the middle way they asserted a mode of existence of phenomena that was unique and distinct from that of any other school.  A l l other  coterminous  with  Indian  schools,  Nagarjuna's  including  exegesis  of  the  Buddhist ones  Mahay ana  philosophy,  placed within their systems the assertion of a final reality. Samkhyas declared the ultimate reality to be a dualistic one, VaiSesikas  asserted  atoms,  and  the  Buddhist  The the  Abhidharmikas,  Sarvastivadins, and Sthaviravadins, etc., all posited some sort of final nature that acted as a substratum for existence.  Nagarjuna refrained  from positing any sort of final reality that truly existed, stating that to do so was to fall from the middle way to the extreme of permanence. He repudiated the possibility of the existence of ontological entities and declared  the  emptiness  of  all  things.  1  For Nagarjuna and the  Madhyamikas who were to follow him, nothing at all, anywhere, could exist in reality, independendy, possessed of self-nature, able to stand by itself, solid and unchanging.  With the propogation of Nagarjuna's works, the Madhyarnika view increased  in popularity in India, and soon became  innumerable  heated  debates between  Buddhists  schools and between Buddhists themselves.  and  the focus of the  orthodox  The fact that the Master's  verses could be interpreted in different ways inspired a number of scholars in later years to compose commentaries on the original. Three  1. J.W. de Jong briefly comments on the main Western scholars who have studied Nagarjuna's "emptiness" in his article entitled Emptiness (1972). For a more detailed survey, see de Jong (1974).  6 commentators, , whose expositions were studied and whose arguments 1  have-  resounded  throughout  the  halls  of  Mahayana  Buddhist  monasteries until the present day were Buddhapalita, Bhavaviveka, and Candrakirti. CE.,  2  Buddhapalita, born in South India in approximately 470  authored the MuUa-madhyamika-vrtti.  He was the first main  commentator to employ the logical tool of consequences^, in lieu of syllogisms, in the elucidation of Nagarjuna's treatises.  His utilization  of the prasanga (consequence) method, which uses the opponent's own views to force the opponent to see position,  was  refuted  approximately 500 C . E . then  severely  by 5  the contradictions within his  Bhavaviveka, born in South India in 4  He studied Buddhapalita's commentary and  criticized  Buddhapalita's  methodology  in  his  Prajnapradipa, claiming that the prasanga method was an inadequate 6  tool in the attempt to establish the Madhyamika's position, and insisted that such could only be accomplished with the aid of  syllogisms.  Candrakirti, the third commentator of this group, born in South India around  the  beginning  of  the  seventh  century,  7  in turn refuted  Bhavaviveka's criticism of Buddhapalita to re-establish the validity of  1. The three mentioned here are included because of their impact on later Tibetan scholasticism.  For other commentators (Aryadeva, Sthiramati, etc.) see Ruegg (1981), p.  47-49. 2. Thurman (1984), p. 39. 3. Murti (1955), p. 95 translates prasanga-vakya as reductio ad absurdum. 4. Various of Bhavaviveka's treatises have been examined, for example, by S. Iida (1980), M.D. Eckel (1980), and C. Lindtner (1986). 5. Ruegg (1981), p. 61. 6. The Prajfiapradipa is no longer extant in Sanskrit; it is available in Chinese and Tibetan. Of equal importance are Bhavaviveka's Madhyamakahrdayakarikas (available in Sanskrit and Tibetan) and their commentary, the Tarkajvala (Tibetan only).  This work, as will be  noted later, was one of the earliest siddhanta (tenets) texts; it contains a review and discussion of the main schools as they existed in Bhavaviveka's day.  For the titles of the  chapters of the Madhyamakahrdayakarikas (Sanskrit, Tibetan, and English) and a list of translations of the chapters O^nghsh, German, and Japanese) see Iida (1980), p. 12-17. 7. Ruegg (1981), p. 71.  7 the  prasanga  method  of  reasoning.  His  works  include  Madhyamakavatara and the Prasannapadamulamadhyamakavrtti.  1  the The  tendencies of these authors to employ differing logical techniques to establish their position not only revealed methodological preferences, but  also  indicated differences  in opinion regarding the  status of  conventional "reality"; the split in understanding found fruition in the founding of the two major sub-schools of the Madhyamika. these sub-schools and  received the designations  In Tibet  Svatantrika-Madhyamika  Prasangika-Madhyamika. India in the seventh and eight centuries C E . saw the continued  flourishing of the Madhyamika school and the further delineation of views within it.  Scholar-monks continued to compose philosophical  treatises - the most notewothy  of these scholars being Santideva,  recognized as a Prasangika-Madhyamika, Jnanagarbha, who followed in the tradition established by Bhavaviveka (but established his own school that has been termed by the Tibetans the Yogacara-SvatantrikaMMhyamika), and two scholars in his lineage, Santaraksita and his student Kamalas"ila.  Their works were studied in the huge monastic  complexes of northern and central India.  In time, the fame of the  teachers and teachings within these centres began to spread far beyond the confines of India's borders and, in the same way that a mountain stream's spring trickle soons turns into a rushing torrent of water, Madhyamika Buddhism saw its movement into the northern countries.  Madhyamika into Tibet ...the main Buddhist centres of central India during the eighth  to  twelfth  universities  of  centuries  Nalanda,  were  the  Bodhgaya,  great  monastic  Odantapuri  and  1. L. de la Valle Poussin edited the Tibetan translation of the Madhyamakavatara and Bhasya; see Bibliotheca  Buddhica ix (St. Petersburg, 1907-12).  Museon (1907, 1910, 1911).  He. translated part of it in  He also edited the Sanskrit text of the Prasannapada; see  Bibliotheca Buddhica iv (St. Petersburg, 1903-13).  8 Vikramashila, housing thousands of monks and learned men, those who had taken monastic vows and those who had not, and attracting scholars from every Asian country which had developed an interest in Buddhism.  From the  fourth to the eight centuries Chinese monk scholars were frequent visitors, and from the eighth century to the final eclipse of Buddhism in India at the end of the twelfth the Tibetans were constantly  visiting Nepal and India for  texts, instructions and initiations.  1  Conversely, by the eighth century, Indian Buddhist scholars had also begun to accept invitation to spread Buddhist teachings outside their homeland.  Santaraksita, mentioned above, is recognized as being  "chiefly responsible for the implantation of Buddhism in Tibet", due 2  to, initially, his visits there, according to Tibetan records, in 763 C E . and  his residence  there  inspired construction  from 775-788.  of the  3  It was his influence that  first Buddhist monastery  bSam yas , 4  modelled after the Indian Odantapuri, and that planted the seeds for the transfer and synthesis of traditional academic monastic training.  After  his death, his student KamalaSila was invited to Tibet to further the work begun by Santaraksita.  Kamalas"Ila's "victory" over the Chinese  Hvashang Mahayana in the Great Debate of bSam yas in 792 was 5  representative of the increasing interest in Indian Buddhist philosophy and practice in Tibet and of the influence of Indian Buddhist scholars.  6  1. D. Snellgrove and H . Richardson (1980), p. 72. 2. Ruegg (1981), p. 88. 3. Ibid., p. 89. 4. bSam-yas-mi-'gyur-lhun-gyis-grub-pa'i-gtsug-lag-khang,  built in 775 C E .  One version  of the political and social struggles that backlit the construction of bSam yas during the early diffusion of Buddhism in Tibet is described in the first chapter of Tucci (1980). 5. D. Snellgrove and H. Richardson (1980), p. 79. 6. Tibetan accounts attest to one Great Debate. see Paul Demieville (1954).  For an account based on Chinese sources,  It would seem more likely that there was a series of debates  that reflected both Chinese and Indian Buddhist influences and the differing philosophical  9 By the year 800 C.E., many Indian sutras and sastras had been and were in the process of being translated by highly competent translators. The intensity of the importation of Indian Buddhism into Tibet is reflected in a catalogue of translated works that has been preserved in the Tibetan Canon.  It includes the first "official" translations made during  1  Santaraksita's visit and those made, most probably, up until King Khri gTsug lde brtsan's (Ral pa can) assassination in 838 C E . : ' the list is composed of seven hundred and thirty-six titles of translated works. The  proliferation of the doctrine was  cut short in 838  with the  assassination, and with the usurpation of the throne by gLang dar ma. While  Tibetan accounts  accuse  gLang  dar ma  of  being  solely  responsible for a persecution of the faith and for the decline of Buddhism in Tibet, it would seem more tenable that at that time internecine disputes and border concerns began to take priority over religious issues.  Royal attention focused on the encroaching Chinese  and, with the murder of gLang dar ma in 842, a political dissolution that would last for two centuries began. As Tibet fell into a state of anarchy the connections with the Indian intellectuals were severed.  (Tibetan  historians designate the years from 838 to 842 C E . the end of the "early diffusion" accounts  (snga  exaggerate  dar)  of Buddhism).  Suggesting that the traditional  a limited persecution,  D. Snellgrove  and H .  Richardson write,  Buddhism was now neglected, rather than persecuted, for the continuing civil strife exhausted and impoverished the leading families in Central Tibet, on whom organized religion of any sort always depended for patronage. But in  positions within the Chinese and Indian schools. See also D. Snellgrove (1987), p. 426-436; Yoshiro Imaeda (1975), p. 132-133.  1. The catalogue is titled sTong thang ldan dkar. See D. L. Snellgrove (1987), p. 440. Snellgrove suggests that work on the Mahavyutpatti (Bye brag tu rtogs byed chen po), the dictionary of Sanskrit and related Tibetan terms, was commenced during Khri srong lde brtsen's reign (1987, p. 441). See also M . Lalou:  "Les textes bouddhiques au temps du roi  Khri-Srong-lde brtsan" (Journal Asiatique, 1953, p. 313-353), p. 319.  10 the outlying regions, such as the small principalities of the east where Buddhism was the established religion, and later in the new kingdoms of the west, which bordered on the  Buddhist lands  of  north-western  India, Tibetans  continued and developed their practice of the new faith.  1  According to Tibetan accounts, it was not until 978 C E . that a full 2  restoration of Buddhism in Tibet was instigated by Tibetan teachers and Indian  scholars.  3  In the years that followed, numerous translations of  sutras and tantras, together  with their volumnimous commentaries,  were completed and, in parts of Tibet, the reigning house once again spread the umbrella of royal patronage over the favoured religion. Sixty-four years after the commencement of the Buddhist renaissance, Dipamkara  Srijnana,  a  renowned  Indian  scholar-monk,  more  commonly known as Atisa, accepted the invitation of King 'Od lde, King Byang chub 'od, and King Zhi ba 'od to come to Tibet. Perhaps the greatest stimulus to religious developments in Tibet in the eleventh century was the mission of the great Indian teacher Atiia, who arrived in Gu-ge in 1042 at the age of sixty after repeated invitations from the religious kings of Western Tibet. Indian  monastic  He had studied and taught at the  universities  of  Bodhgaya, Odantapuri,  and Vikramashila, and he was probably the most famous and revered religious teacher in India at the time.  4  Alaka Chattopadhyaya also writes,  1. D. Snellgrove and H. Richardson (1980), p. 112. 2. Ibid. 3. For an account of the commencement of the later diffusion (phyi dar) from a traditional source see G . Tucci (1980), p. 17-18. 4. Ibid.  11 The crowning achievement of the rulers of western Tibet was, of course, the bringing of AtiSa.  This shaped the  subsequent history of the country.  1  R. Sherburne briefly comments on the traditional Tibetan accounts, All Tibetan historians mark his arrival at Tho ling in 1042 as the rebirth of Buddhism in their country - the "Second Spread"  (phyi daf)  of the religion that had first crossed  their mountains in the seventh century.  2  During the first three years of his residence in Tibet Atlia gained both the devotion of powerful members of the ruling family of Western Tibet and the respect of the great translator Rin chen bzang po.  3  With  the latter, he corrected earlier translations and oversaw numerous new translations of Indian texts. Arisa had been invited to Tibet primarily to strengthen traditional monastic values during the Buddhist renaissance there, based on the kings' hope that his influence would discourage the Tibetans'  involvement  in  corrupted Tantric  practices.  Probably  partially to fulfill the kings' wishes, during these first three years Atisa  Bodhi-patha-pradipa, Bodhi-marga-pradipa-panjika,  composed his magnum opus, the its autocornmentary, the  4  together with both of which  explicate the Madhyarnika view and lay strong emphasis on the practice of proper morality. But it would seem that one of the most significant events during these years was his meeting with his future disciple, 'Brom ston pa.  At 'Brom ston pa's request, AnSa postponed his  departure for India in order to undertake a tour of central Tibet, where thousands of monks resided.  In the years that followed, AtiSa provided  1. Alaka Chattopadhyaya (1967), p. 286. 2. Richard Sherburne (1983), p. xi. 3. Rin chen bzang po lived from 958 - 1055 C E . He had been sent by the Buddhist King Y e shes 'od to study in Kashmir. He returned to Tibet and eventually translated 158 texts from Sanskrit into Tibetan. 4. For an English translation of this text, see R. Sherburne (1983).  12 the Buddhist community with valuable teachings  and succeeded in  instituting many of the desired reforms within it.  The bKa' gdams pa Sect Upon Atlsa's death at Nye thang in 1054, 'Brom ston pa proceeded to Rva sgreng to found what would remain the main monastery of his religious  order.  1  His "order" became known as the bKa' gdams pa  2  sect. His intention in founding the order was the same as that which had inspired him to encourage Atis'a to go with him to central Tibet: restore discipline to religious life.  to  In constrast with the free-roving  non-celibate tannic practitioners, the bKa' gdams pa followers were required , to  refrain  from  possession of money.  marriage,  intoxicants,  travel,  and  the  We are reminded that "only in retrospect can  'Brom ston pa's foundation be referred to as a new religious order"  3  since no distinctive religious order had yet existed in Tibet. In addition, AtiSa and 'Brom ston pa's efforts had been largely in reaction to the  1. D. .Snellgrove (1987) p. 481. 2. D. Snellgrove and H . Richardson (1980), p. 131, translate b K a ' gdams pa as 'Bound by Command'; D. Snellgrove (1987), p. 479, translates it as 'bound to-the (Buddha's) word'. Tucci (1980), p. 22-23, writes, When 'Brom ston asked Atisa which was more important and more basic igtso), the text of the scriptures (bka\ Revelation, and bstan bcos, S k t sastra, the books written by Indian masters), or one's teacher's instructions (bla ma'i gdams ngag), Atisa replied that direct instruction from one's teacher is more important.  So it came about that the first two schools of  Buddhism to appear in Tibet, those which trace their origins back to Atisa and Marpa, both bear the name bKa' brgyud...Ths school which Marpa founded still retains the name of bKa'  brgyud, while the followers of  Atisa, the bKa' gdams pa, also called themselves bKa' rgyud bKa' gdams  pa. Thu bkan bio bsang Chos kyi nyi ma , in A . Chattopadhyaya (1967) Appendix A Section 5 p. 385, reports that Atisa and 'Brom ston pa's sect received the appellation b K a ' gdams pa "because it conveys the preaching of the Buddha word for word, without omitting any word". 3. D. Snellgrove (1987), p. 485.  13 loosening  of  religious  discipline  due  to  the  Tibetan  populace's  involvement in the freer and often malpractised Tannic rituals, and 1  had not been explicitly for the purpose of establishing a Buddhist sect. Thus, at the beginning of the bKa' gdams pa order, most of those persons and monasteries that had come under AtiSa's influence were not necessarily cognisant of being "bKa' gdams pa", and nor did they term themselves such; they tended rather to delineate and name their religious "orders" based on their spiritual lineages, i.e., based on the succession of scholars and gurus through whom they had received their knowledge. It was from within the enclaves of this sect that the dGe lugs pa sect,. actually a later continuation of the bKa' gdams pa, emerged in the fifteenth  century.  Thus absorbed by its successor, the bKa' gdams pa Order disappears from the Tibetan scene, and in retrospect tends to be remembered only as a passing phase of Tibetan monasticism.  However, its influence has been far more  widespread than its comparatively short-lived existence might suggest, in that it affirmed the importance of a sound monastic tradition precisely at a time when the importation of Indian Buddhist teachings into Tibet seems to  have  depended  so  much  upon  the  exertions  of  independent scholars, usually non-celibate, who traveled in pursuit mainly of tantric initiations, and the relevant tannic texts and commentaries.  2  1. G . Tucci (1980), p. 35, writes, • While not rejecting the Tantric exercises of the Vajrayana, this school developed them intelligently, opposing the aberrations and heresies into which the followers of Tibetan Buddhism had fallen. 2. D. Snellgrove (1987), p. 486.  14 Before their "reconstitution" the bKa' gdams pas were responsible for 1  the establishment and growth of many monasteries and for the vast production of  translations  and exegetical works.  Still,  the  most  significant contribution that this organized Buddhist order made to Buddhism in Tibet - that led to an overall unity within later Tibetan monasticism - was the acceptance of, and emphasis on, the Indian 2  Buddhist monastic rule (Vinaya). Snellgrove goes on to state, Presumably the label of bKa' gdams pa can only be applied in a restrictive sense to those religious foundations that were consciously modeled on the pattern set by 'Brom ston.  But since all monastic foundation in Tibet, whatever  variations existed in their transmitted teaching traditions, were inevitably based upon the bKa' gdams pa model so far as adherence to any monastic rule was concerned, 'Brom ston pa might well be hailed as the father of Tibetan monasticism.  3  dBus pa bio gsal Thu bkan bio bzang Chos kyi nyi ma (1737-1802) recorded that dBus and gTsang, two of the three main provinces of Tibet, were filled with monasteries of the bKa' gdams pa sect during the years following Atlsa and Brom ston pa's activities.  4  The most renowned of these were Rva-  sgreng and, in the province of gTsang, sNar thang monasteries.  The  latter was founded by gTum ston, a disciple of the famous bKa' gdams  1. Ibid. 2. Ibid. 3. Ibid., p. 507-8.  4. Thu bkan bio bzang Chos kyi nyi ma, Grub mtha' thams cad kyi khungs dang 'dod tshul ston pa legs bshad shel gyi me long in A . Chattopadhyaya (1967), Appendix A , Section 5, p. 390.  15 pa scholar Sar-ba-pa, in 1153. sNar thang, as it existed in the first half 1  of the fourteenth century, was the residence of the author of the Bio  gsal grub mtha'  (the  Madhyamika  section of  which  appears in  translation in the latter part of this paper). His full name seems to have  sDe dge rgyal rabs, he is called Bio gsal byang chub ye shes; in the bKa' 'gyur dkar chag he is referred to as dBus pa bio gsal sang rgyas 'bum; in the Co ne bsTan 'gyur dkar chag he is termed dBus pa bio gsal rtod pa'i seng ge. He will been dBus pa bio gsal byang chub ye shes. In the 2  be referred to as dBus pa bio gsal throughout this thesis. He is probably best known for his participation in the compiliation of the first Canon in Tibet, the sNar thang Canon . 3  The circumstances surrounding dBus pa bio gsal's involvement in the establishment of the Canon are rather humourous and were well-known throughout  Tibet.  4  The account of the events that led up to the  gathering of texts for the first Canon that involve our author are as follows:  dBus pa bio gsal had two teachers, bCom ldan rig pa'i ral gri,  a learned scholar at sNar thang who had a legion of disciples, and 'Jam pa'i  dbyangs, a "maha-pandita".  5  'Jam pa'i dbyangs was, at the same  time, a student of bCom ldan rig pa'i ral gri. One night, as a practical joke, 'Jam pa'i dbyangs disguised himself with the mask of a wrathful religious protector, changed his gait (so as to appear like a demon) and, under the pale light of the moon, crept up on bCom ldan rig pa'i ral gri and chased him around the monastery.  bCom ldan rig pa'i ral gri was  so surprised and terrified by 'Jam pa'i dbyangs' appearance that when he realized the true identity of the "demon", he banished 'Jam pa'i dbyangs from sNar thang.  His joke having backfired, 'Jam pa'i  dbyangs travelled to Sa skya. Once established there, he was invited to the court of Buyantu Khan, the fourth Emperor of the Mongol dynasty  1. D. Snellgrove (1987), p. 507.  2. A s documented in Deb ther sngon po and bKa' gdams chos 'byung gsal ba 7 sgron me. For folio no., see K. Mimaki, p. 13, n. 26. 3  . The Cone, Derge and Peking Canons were compiled in later years.  4. George N . Roerich (1976), p. 337. 5. Ibid.  16 of  Yuan, to act as the Emperor's guru.  The banished "demon"  proceeded to the Mongol Court and, once there, in attempts to pardon himself before his sNar thang teacher, sent bCom Idan rig pa'i ral gri presents.  None of the presents were able to appease bCom ldan rig pa'i  ral gri's wrath until a box of Chinese ink, a necessity for book-copying, arrived.  'Jam pa'i dbyangs also sent his former student, dBus pa bio  gsal, a similar gift to sNarthang. 'Jam pa'i dbyangs then requested that all the books of the bKa 'gyur and bsTan 'gyur be collected and copied by the sNar thang scholars, to be stored within the monastery. dBus pa bio gsal, supplied by 'Jam pa'i dbyangs with the necessary tools, with the aid of Lo tsa ba bsod nams 'od zer and rGyan ro byang chub 'bum, applied himself to the task of tracking down the Tibetan texts to include them in the Canon or to copy them so that they could be combined with those already at sNar-thang.  Grub mtha'  Literature  Within the context of this work, however, our attention focuses on another of dBus pa bio gsal's accomplishments : the 1  Grub pa'i mtha'  mam par shad pa'i mdzod, known here, for the sake of convenience, as Bio gsal gmb mtha'. Gmb mtha' works are, as a group, a distinctive sort of literary genre that present the varying views of schools of philosophy.  Their intent is usually to distinguish the various views one  from another.  The Tibetan compound  gmb mtha' (siddhanta  in  Sanskrit), translates as "established conclusion", and, by extension, as "tenet".  Thus literature included under this rubric states, and often  1. dBus pa bio gsal is also credited with the composition of a Chos 'byung and a grammatical treatise.  Neither, however, have survived the passage of time.  It might also  be noted here that both of dBus pa bio gsal's teachers, 'Jam pa'i dbyangs and bCom ldan rig pa'i ral gri (see List of Rare Books) were responsible for composing their own Grub mtha' but that neither book seems to have survived the centuries. In- parts of Bio gsal grub mtha' previous to the Madhyamika section, it appears that dBus pa bio gsal criticizes the opinions of one of his teachers. See Mimaki (1982), p. 15, n. 31.  17 elucidates, the "established conclusions" of varying schools.  dKon  chog 'jigs med dbang po, an eighteenth-century dGe lugs pa author, in his Grub pa'i mtha'i mam par bzhag pa rin po che'i phreng ba defines "grub mtha'", The  etymology  for  "tenet"  (siddhanta) is:  a tenet  [literally, an established conclusion] is a meaning which was made firm, decided upon or established in reliance on scripture and/or reasoning and which will not be forsaken for  something  Commentary  else. on  Dharmamitra's Clear [Maitreya's]  Words, A  "Ornament for  the  Realizations" (Abhisamayalanikarakarikaprajnap&amitopades'as'astratika) conclusion"  [tenet]  signifies  says:  one's  "Established own  established  assertion which is thoroughly borne out by scripture and reasoning.  Because  one  assertion, it is a conclusion.  will  not pass  beyond this  1  Tibetan Grub mtha' literature finds its roots in the earlier Indian dars~ana (philosophical school) tradition and was generally modelled after Bhavaviveka's Madhyamakahrdaya (and autocommentary) and Tarkajvala, Santaraksita's  Tattvasamgraha, and  Kamalaslla's  Tattvasamgraha-pahjika ' all of which review and discuss the doctrines 2  of the main Indian schools.  Of these three, it was Bhavaviveka's work  that acted as the prototype for the works of Tibetan encyclopaedists. In terms of style, however, Tibetan Grub mtha' tend to resemble tannic Aryadeva's  Jnanasarasamuccaya and Jitari's  Sugatamata-  vibhangakarika}. Within Tibet, two types of Grub mtha' existed.  One type focused on  the Indian schools and the other focused on the Tibetan, Bon po and 1. Geshe Lhundrup Sopa and Jeffrey Hopkins (1976), p. 53.  2. The Sarva-dars'ana-samgraha,\he. Sarva-siddhanta-sarngraha,md the Sad-daiSanasamuccaya, etc., were also compiled in Sanskrit at about or before this time. 3. K. Mimaki (1982), p. 2.  18 Chinese schools.  Bio gsal grub mtha' is of the first sort and, following  in the pattern established centuries earlier in India, and in league with the  majority  of  the  Tibetan  Grub  mtha' texts, attends to  the  philosophies of the Indian non-Buddhist schools as well as to those of the four major Buddhist schools . 1  A few of the Tibetan Grub mtha'  focus exclusively on the tenets of the Buddhist schools (see the list below - no. 21, 22, 28, 31).  Unlike the Sa skya pas, bKa' gdams pas,  dGe lugs pas andbKa' rgyud pas, the sNying ma pas and Bon pos do not consider the Madhyamika as the highest school and include the tantric schoolsin their enumeration of Buddhist schools.  They therefore list  nine Buddhist schools. Following Mimaki, the Tibetan Grub mtha' texts are as follows.  2  They are divided according to sect. I. sNying ma pa and other early authors: 1.  Manuscripts of Touen-houang:  ST 260, 607, 666, 692, 693, 694;  F T 116, 121, 812, 814, 815, 817, 819, 820, 837, 842, 2101. 2. Ye shes sde, ITa ba 'i khyad par. 3. dPal brtsegs, ITa ba'i rim pa bshadpa. 4. Nyi ma 'od, ITa ba'i rim pa. 5. Rong zom Chos kyi bzang po, ITa ba'i brjed byang. 6. Rong zom Chos kyi bzang po, Grub mtha'i brjed byang. 7.  Rong zom Chos kyi bzang po, Man ngag lta ba'i phreng ba zhes  bya ba 'i 'grel pa. 8. Klong chen rab 'byams pa, Grub mtha' mdzod. 9. Klong chen rab, 'byams pa, Yid bzhin mdzod (and Rang 'grel).  1. Schools examined in Bio gsal grub mtha' are: Lokayata ('Jig rten rgyang phan pa), Samkhya (Grangs can pa), Saiva (dBang phyug pa), Vaisnava (Khyab 'jug pa), Digambara (gCer bu pa), the eighteen sects of the Lesser Vehicle (sDe pa bco brgyad), the Vaibhasika (Bye brag tu smra pa), Sautrantika (mDo sde pa), Yogacara (Sems tsam pa), and Madhyamika (dBu ma pa). 2. Mimaki (1982), p. 6-8  19 10. 'Ju M i pham rgya mtsho, Yid bzhin mdzod kyi grub mtha' bsdus pa. 11. bDud 'joms rin po che, sNying bstan mam gzhag. II. Sa skya pa: 12. Grags pa rgyal mtshan, rGyud kyi mngon par rtogs pa rin po che'i ljon shing. 13.  Sa skya Pandita Kun dga' rgyal mtshan, gZhung lugs legs par  bshad pa. 14. sTag tshang Lo tsa ba Ses rab rin chen, sTag tshang gmb mtha'. 15.  Pan chen Sakya mchog ldan, dBu ma mam par nges pa'i bang  mdzod lung dang rigs pa 'i rgya mtsho. 16.  Pan chen Sakya mchog ldan, dBu ma'i byung tshul mam par  bshad pa 'i gtam yid bzhin lhun po. 17. Go rams pa bsod nams seng ge, rGal ba thams cad kyi thugs kyi dgongs pa zab mo dbu ma'i de kho na nyid spyi'i ngag gis ston pa nges don rab gsal. III. bKa' gdams pa: 18. dBus pa bio gsal, Bio gsal gmb mtha. IV. dGe lugs pa: 19. Tsong kha pa, Lam rim chen mo.  1  20. Tsong kha pa, Drang nges legs bshad snying po.  1. Translated by Wayman (1978) (Samatha secton only).  Inclusion of this text and some  others in the list reveals that Mimaki's definition of Grub mtha' is very wide. (1982), p. 46, defends inclusion of Drangs nges legs bshad snying po, Parmi les oeuvres de Tsong kha pa, c'est cet ouvrage qui s'aparente le plus a un grub mtha'. exposant  le  II consiste de deux grandes parties, le premiere  philosophic  des Vijnanavadin  en s'appuyant  sur le  Samdhmirmocanasutra et la deuxieme celle des Madhyamika en se fondant sur V Aksayamatinirdes'a. L a partie concemant les Madhyamika a deux subdivisions:  celle des Svatantrika et celle des Prasangika.  de Svatantrika est encore subdivisee en deux parties...  L a subdivision  Mimaki  20 21.  Se ra rJe btsun pa Chos kyi rgyal mtshan, Grub mtha' rnam  gzhag. 22.  dGe 'dun rgya mtsho, Grub mtha' rgya mtshor 'jug pa'i gru  rdzings. 23. Pan chen bSod nams grags pa, Grub mtha'i mam gzhag. 24. 'Jam dbyangs bzhad pa, Grub mtha' chen mo. 25.  Sum pa mkhan po Ye shes dpal 'byor, Grub mtha'i mam bzhag  nyung 'dus. 26. ICang skya II Rol pa'i rdo rje, ICang skya grub mtha'. 27. ICang skya II Rol pa'i rdo rje, Dag yig mkhas pa'i 'byung gnas, chap. 5: Grub mtha'i skor. 28. bsKal bzang lha dbang, Grub mtha'i mam bzhag dge legs 'byung gnas. 29. dKon mchog 'jigs med dbang po, Grub mtha' rin chen phreng ba. 30. Thu'u bkwan 3 Bio bzang po, Grub mtha' shel gyi me long, chap. 1: 'Phags yul du phyi rol pa dang rang sde'i grub mtha' byung tshul: 31.  dGe bshes Ngag dbang nyi ma, Nang pa'i grab mtha' smra ba  bzhi'i 'dod tshul gsal bar bshad pa bio gsar rig pa'i sgo 'byed. V. bKa' rgyud pa: 33.  'Ba' ra ba rGyal mtshan dpal bzang, Grub mtha'i mam bzhag  (and dKa' 'grel).  2  VI. Various: 33. Bu ston Rin chen grub, Bu ston chos 'byung. 34. Bo dong Pan chen Phyogs las rnam rgyal, Encyclopedia Tibetica, Vol. 11. VII. Bon po:  2. Mimaki (1982), includes this text with the next two.  21 35.  Vairocana=Ba gor Rin chen bio gsal, Thag pa rim pa mngon du  bshad pa'i mdo rgyud. 36. Tre ston rGyal mtshan dpal, Bon sgo gsal byed. 37. dPal btsun Nam mkha' bzang po, Theg pa rim pa gsal ba'i sgron ma. 38. Sar rdza bKra shis rgyal mtshan, Theg dgu'i grub mtha' mam.  Occasionally, scholars have questioned the value of the study of Tibetan Gmb mtha' literature. Mimaki poses their dilemma, On entend souvent la critique suivante a propos de l'etude des grub mtha':  dans l'6tude de grub mtha', s'agit-il du  bouddhisme indien ou du bouddhisme tibe'tain?  1  To this criticism which implies that Gmb mtha' literature represents neither, Mimaki responds that it is both. Tibetans wrote  them, the  He states that since the  Gmb mtha' indeed  represent  Tibetan  Buddhism, and since they examine Indian Buddhism, the Gmb mtha' also show Indian Buddhism.  He terms the criticism a "word-game" , 2  and urges scholars to go beyond such attitudes. D. Seyfort Ruegg, in a short article on the Madhyarnika school that also deals with Gmb  mtha' literature, addresses the same, but more  general issue. In order. no doubt to better establish the specificity and identity of Tibetan culture and also of Tibetology as an academic  discipline, a tendency  has recently appeared  among some scholars to discount connections  between  India and Tibet even in the area of Buddhist thought. Now, when we acknowledge the dependence of much of  1. Ibid., p. 3. 2. Ibid.  22  European philosophy on Plato or Aristotle we certainly do not put in question the original contributions made by West-European philosophers starting in medieval times; or when the Arabist notes the links between medieval Islamic and Greek philosophy he does not thereby deny all specificity to Islamic philosophy. It is then suggested here that, by the same token, the study of Buddhism in Tibet and indeed of Tibetan civilization as a whole can lose nothing by fully acknowledging their close ties with the Buddhism of India and with Indian civilization. studies  can  indeed  only  gain  by  being  Tibetan  pursued in  coordination with (but certainly not in subordination to) Indian studies... As for the dBu ma pas, there is no evidence to indicate that they have understood their task to be to set themselves off from their Madhyarnika predecessors in India. contrary, they explain  and  have put  very clearly striven to  into  practice  the  On the  penetrate,  understanding  of  Buddhism achieved by Nagarjuna and his disciples up to Abhayakaragupta  and  Sakya-Pandita;  to  their  interpretations they regularly refer, and also defer in a not uncritical manner.  They thus combine close adherence to  the traditions and likes of thought established by their predecessors in India with the production of very valuable contribution of their own in the area of textual exegesis and philosophical hermeneutics as well as in the domain of philosophical and meditative theory and practice.  1  Indeed, there is great value in the study of Tibetan Grub  mtha'  literature, particularly of a text written as early as that of the bKa' gdams pa dBus pa bio gsal.  In general, the Grub mtha' texts, usually  composed prior to the relatively late printing (seventeenth-eighteenth  l.D.S. Ruegg (1980), p. 279.  23 century) of the Tibetan Canon, tend to contain readings closer to the original texts. They thereby allow for the correction of mistakes within the  Canon.  For example,  Satyadvayavibhangakarika  the  Canon's version  of Jnanagarbha's  lOcd reads,  de bas 'di ni kun rdzob sde / yan dag don yin yang dag min /A Therefore, this is conventional [truth]. It is ultimate [truth] and is not correct.  Bio gsal grub mtha' min. The translation of  has in place of the Canon's verse's the  Bio gsal grub mtha'  yin its  negative,  verse reads:  Therefore, this is conventional [truth]. It is not ultimate [truth] and is not correct. Obviously, in keeping with Madhyamika thought, dBus pa bio gsal's version is correct.  Thus one can be guided to rectify the error in the  Canon. The  Grub mtha'  are also a welcome help for the interpretation of  difficult passages in Indian texts that 1) have commentaries that in themselves are laconic, or that 2) lack commentary altogether.  For  Grub mtha' clarify, and for the the Grub mtha' themselves act as  texts with difficult commentaries, the latter - texts without commentary commentary.  An additional contribution that authors of the offered  toward  the  understanding  of  Indian  Grub mtha' literature Buddhism  was  the  ingenious invention of and the application of rubrics to the Indian philosophical systems, under which the systems could be more clearly defined.  The  terms  Madhyamika,  Yogacara-Madhyamika,  1. SDVP P. [100] (5283) sa 14M-2.  Prasangika-Madhyamika, and  Svatantrika-  Sautrantika-Madhyamika  24  were not names with which, as for as we can determine at present, the Indian Buddhists classified themselves or by which they delineated themselves from each another, but were rather appellations construed by the Tibetan scholars to aid in the understanding of the various doctrinal positions held by the Indian scholar-authors.  M . Eckel, in his  examination of the Indian scholar Jnanagarbha, writes, The intricacies of the relationship between Jnanagarbha and  Dharmakirti  indicate  some  of  the  intellectual  atmosphere in the Indian Buddhist community at a time when Indian Buddhist thought was at the height of its powers.  During these years, traditions of interpretation  developed that in time produced the categories we now use to classify the different "schools" of Buddhist thought. But in the minds of the philosophers themselves the categories of philosophical affiliation do not seem to have been as sharp as they now seem to us, with nearly a thousand years of historical and philosophical study to color  our  understanding.  philosophical  exploration  sometimes  conflict,  in  In  Jnanagarbha's  time  traditional  lines,  crossed  sometimes  in  agreement,  and  sometimes in the complex combination of conflict and agreement that makes it hard to see clearly where one tradition starts and another ends.  1  Thus the Tibetan classifications of Indian schools, while undoubtedly beneficial, need to be approached with an air of cautiousness. the above-mentioned  Due to  "porousness" of the walls between the Indian  schools, the authors of the later classifications of the schools, as will be noted later, were not always in agreement.  1. Malcolm D. Eckel (1986), p. 57-58.  25 Bio gsal grub mtha' Let us turn now to the importance specifically of Bio gsal grub mtha'. It should be noted that on folio lb of Bio gsal grub mtha', the Sanskrit title,"Siddhantavyakhyanakosa"',  is written before the Tibetan, giving  the work the appearance of being a translation of a Sanskrit text.  It is  not a translation; this style that includes a Sanskrit title before the Tibetan title seems to have been in vogue in the fourteenth and fifteenth centuries.  1  The Sanskrit was most likely invoked to lend credence and  a sense of authority to the treatise. Bio gsal grub mtha's importance, in regard to the Tibetan Buddhist tradition, lies mainly in the date of its composition.  A Khu Rinpoche  Shes rab rgya mtsho (1803-1875) includes Bio gsal grub mtha' in his List of Rare Books . 2  The books included in his list are as follows:  3  1. Phya pa Chos seng gi Phyi nang gi grub mtha'i mam bzhag bsdus pa. 2. Bya mChad kha pa'i Grub mtha' mdor bsdus. 3. bCom ldan rigs ral gyi Grub mtha' rgyan gyi me tog. 4. de'i slob ma mChima 'Jam pa'i dbyangs kyi Grub mtha'. 5. De'i slob ma dBu pa bio gsal gyi Grub mtha'. 6. Shes rab rdo rje'i Grub mtha'. 7. Nya dpon Kun dga' dpal pa'i Grub mtha 'i mam bzhag. 8.  Rong ston Sakya rgyal mtshan gyi Nang pa'i grub mtha' smra ba  dang po gsum gyi mam bzhag. 9. lHa lung gi dbang phyug Byang chub rin chen gyi mChad kha pa'i Grub mtha' gees bsdus kyi tik chen. 10. Ra ba stod pa mChog lha 'od zer gyi Grub mtha'.  1. Mimaki (1982), p. 16, includes Saskya Pandita Kun dga' rgyal mtshan, Klong chen rab 'byam pa and sTag tshang lo tsa ba shes rab rin chen as examples of other Tibetan authors who adhered to this style.  2. dPe rgyun dkon pa 'ga' zhig gi tho yig don gnyer yid kyi ku nda bzhad pa 'i zla 'od 'bum gyi snye ma. 3. Mimaki (1982), p. 9.  26 11. Dwags po bKra shis mam rgyal gyi Grub mtha' shan 'byed 'bring  bsdus  gsum.  12. bKra shis 'od zer gyi Grub mtha' mdor  bsdus.  13. dKa' bcu Padma bzang po'i Grub mtha' mdor 15. sTag lo'i Grub mtha' kun shes rtsa  of Rare  bsdus.  'grel  With the exception of Tsa lo'i's Grub mtha' the List  rgyas  and the fifth book in the  Books - dBus pa bio gsal's Grub  mtha' - these books  have not survived the passage of time. However, Bio gsal grub  mtha's  rarity lies not so much in its quantitative scarcity as in its qualitative uniqueness.  dBus pa bio gsal wrote in the first half of the fourteenth  century, nearly four hundred years before the full emergence of this literary genre. The Grub  mtha' works best-known by Western scholars  issue from the later centuries. The pioneer of Grub  mtha'  study was V.P. Wasselief, a Russian  scholar who examined parts of a sixteenth century work in the mid1  ninteenth century.  He was followed by Sarat Chandra Das who, in  1881-82, translated into English and published portions of Grub  mtha'  shel gyi me long by Thu bkan bio bzang chos kyi nyi ma. Perhaps the 2  most well-known of Grub  mtha' English translations are those of dKon  mchog 'Jigs med dbang po's Grub pa'i mtha'i rnam bzhag rin po che'i 3  phreng  ba.  The Madhyamika section was translated by S. Iida in his,  PhD thesis in 1970 and published in 1980 in his book, Reason Emptiness.  and  The Grub mtha' was translated by Herbert V . Guenther and  published in 1971 in Buddhist  Philosophy  in Theory  and Practice.  1. Grub mtha' chen mo by 'Jam dbyangs bzhad pa (1648-1722).  In  See V. Vassilief, Le  bouddhisme, ses dogmes, son. histoire et sa litterature (Paris, 1865). 2. Published inJoumal of the Asiatic Society of Bengal. It was reproduced and edited by Ngawang Geleg Demo and Gene Smith, Collected Works of Thu'u bkwan (Delhi, 1969). Sections of this Grub mtha' were subsequently translated and published by Helmut  Hoffman in Quellen zur Geschichte der Tibetischen Bon-Religion; D. Seyfort Ruegg in Journal of the American Oriental Society, Vol. 83 (1968) p. 73-91; Lu Cheng in Studia Serica Series B, no. 1 (Dhentu, 1942); ; K.K. Mittal in A Tibetan Eye-View of Indian Philosophy (Delhi, 1984). 3.1728-1781.  27 1976 Geshe Sopa and Jeffrey Hopkins published their translation of the same Grub mtha' in Practice and Theory of Tibetan Buddhism. Guenther, in his preface to Buddhist Philosophy in  Theory and  Practice writes (referring to dKon mchog 'Jig med dbang po), As a dGe lugs pa (follower of the "new" tradition started by Tsong kha pa [1357-1419), he strictly adhered to the tradition which recognizes only works of Indian origin as authoritative; consequently Indian,  that  philosophy.  is,  he emphasizes  epistemological  the customary  aspect  of  Buddhist  1  However, although there existed the tradition in Tibet that ascribed authority only to the Indian texts, there existed simultaneously the tendency  amongst  scholars,  once  the  commentary  translation of an original Indian text was commentary  alone.  2  on  a  literal  made, to rely on the  The reliance on the Tibetan commentary over the  original translation led Tibetan authors to glean and cite their Indian sources from the commentaries and, in consequence, to occasionally cite erroneously verses or of parts of verses. Hence, dBus pa bio gsal's distance  in time  from many of  the  later commentators  and his  proximity to the earlier Tibetan tradition place many of his textual citations in better accord with the Indian originals.  Further, with the  establishment of the dGe lugs pa tradition by Tsong kha pa and his disciples  mKhas grub rje and rGyal tshab rje, both of the later  fourteenth-early fifteenth centuries, dGe lugs pa writers of later years made a practice of almost never altering the fundamentals of the founding fathers' philosophical points.  dBus pa bio gsal's treatise then,  written prior to the advent of the dGe lugs pa school, occasionally points to areas in variance with those of the later tradition.  1. H.V. Guenther (1971), p. ix. 2. Mimaki (1982), p. 17.  28  Early Categorization of Madhyamika Subschools One of the major areas within Bio gsal grub mtha' that reveals discrepencies with the works of the later dGe lugs pa school concerns the  subdividing  of the Madhyamika school.  dBus pa bio gsal  subdivides the school as follows, Group I 1) Sautrantika-Madhyamika  Bhavaviveka  2) Yogacara-Madhyamika  Santaraksita,  3) 'Jig rten grags sde spyod pa'i dbu ma pa  Haribhadra -Jnanagarbha, Candrakirti  Group II 1) Svatantrika  Bhavaviveka (Jnanagarbha, Kamalas"ila)  2) Prasangika  Buddhapalita  Note  that he sets  forth  (Candrakirti)  two separate groups,  one consisting  of  Sauttantika-Madhyamika, Yogacara-Madhyarnika, and 'Jigs rten grags sde spyod pa'i dbu ma pa (Madhyamikas who conform to that which is renowned in the world), the other of Svatantrika and Prasangika. Later dGe  lugs pa authors combined the two categorizations:  Sautiantika-Madhyamika  and  they placed  Yogacara-Madhyamika  under  Svatantrika, and equated and dissolved the 'Jigs rten grags sde spyod pa'i dbu ma to the Prasangika division. later categorizations  Following are examples of the  by two sixteenth century and one seventeenth  century dGe lugs pa authors.  dGe 'dun rgya mtsho's (1475-1542)  divisions of the schools is given first.  1  1. Classification from 29.  Grub mtha' rgya mtshor 'jug pa'i gru rdzings. In Mimaki (1982), p.  29  1) Svatantrika - Sautrantika-Madhyamika  Bhavaviveka and followers  - Yogacara-Madhyamika  Santaraksita and followers  2) ' Prasangika  Buddhapalita and followers  Chos kyi rgyal mtshan's divisions are similar.  1  1) Svatantrika - Sautrantika-Madhyamika  Bhavaviveka,  JfTanagarbha  - Yogacara-Madhyamika  Santaraksita, Haribhadra,  KamalaSlla 2) Prasangika  Buddhapalita, Candrakirti,  Santideva  'Jam dbyangs bzhad pa's divisions are more extensive, but continue to follow the dGe lugs pa pattern that makes the Sautrantika-Madhyamika and Yogacara-Madhyamika subschools of the Svatantrika and that does away with the 'Jigs rten grags sde spyod pa'i dbu ma pa classification.  2  1) Svatantrika - Sautrantika-Madhyamika  Bhavaviveka,  Jfianagarbha  - Yogacara-Madhyamika a) - rNam bden dang mthun pa  Santaraksita,  Kamalas'iia,  Aryavimuktisena b) - rNam rdzun dang mthun pa  Haribhadra, Jitari, Kambala  1. Classification from Grub mtha' mam gzhag. In Mimaki (1982), p. 29. 2. Classification from Grub mtha' chen mo. In Mimaki (1982), p. 29.  30  2) Prasangika  i) - Dri bcas dang mthun pa  Jitari  ii) - Dri med dang mthun pa  Kambala  Buddhapalita, Candrakirti, Santideva  dBus pa bio gsal does not anticipate fully the later dGe lugs pa systematization and, in a flourish that at first glance appears to complicate matters, includes various of the Indian scholars in two schools simultaneously.  Bhavaviveka is mentioned as a representative  of the Sautrantika-Madhyamika school in the first grouping and, in the second set of classifications, consisting of the division of Madhyarnika into  Svatantrika and Prasangika, is  set  forth  as  a Svatantrika.  Candrakirti is recorded as being a 'Jig rten grags sde spyod pa'i dbu ma pa and as a Prasangika. And JHanagarbha, curiously, is held to be both a 'Jig rten grags sde spyod pa'i dbu ma pa and a Svatantrika. For example, in folio 100b2 of Bio gsal grub mtha', dBus pa bio gsal writes, The master Jnanagarbha and the master Candrakirti assert conventional [truths] as being the accepted conventions of worldly beings. He then quotes JHanagarbha's Satyavayavibhangakarika (k 21) and Candraklrti's Madhyamakavatara V I 35 and states, Because of this [JHanagarbha and Candrakirti] are known as the 'Jig rten grags sde spyod pa'i dbu ma pas. In folio 98b3 of Bio  gsal  grub  mtha',  as  substantiation  for  JHanagarbha's assertion of Svatantrika views, dBus pa bio gsal quotes from  Jnanagarbha's Satyadvayavibhangakarika  curiousness  of the  assignment  of this  specific  (k 12 and 8). dual  The  affiliation to  JHanagarbha lies in the equation and assimilation in later texts of the 'Jigs rten grags sde spyod pa school to that of the Thai 'gyur (the  31 Prasangika) school and not to the Rang gyud pa (the Svatantrika) school. The mystery of the inclusion of the Indian authors in two separate groups of schools is resolved with examination of the classifications given by authors prior to, contemporary with, and subsequent to the fourteenth century dBus pa bio gsal.  For the earliest Tibetan references  to the Madhyamika school we turn to two of the most ancient Tibetan texts still extant:  the ITa ba'i rim pa bshad pa by dPal brtsegs and the  ITa ba 'i khyad par by Ye shes sde, both of the ninth century. Each of these authors, it can be seen, differentiated Bhavaviveka's school of thought  from Santaraksita's  Sautrantika-Madhyamika  school  school  of thought,  and  and set forth the  the Yogacara-Madhyamika  school as the two divisions of the Madhyamika school.  1  One should  note however, that the Prasangika and Svatantrika schools are not referred to in either treatise. that  also  contain  the  There are other texts of the ninth century appellations  Yogacara-Madhyamika and  Sautrantika-Madhyamika but, except for Nyi ma 'od's ITa ba'i rim pa that mentions the Yogacara-Madhyamika school, the authors of these texts are not known.  2  Later, in the thirteenth century, but earlier that our author, the famous Sa skya pa scholar Sa skya Pandita Kunga rgyal mtshan (1182-1251), in his gZhung lugs legs par bshad pa, classified the schools under two main rubrics that determined the names that appeared below them. The rubrics were 1) kun rdzob kyi sgo nas dbye ba, i.e., division by way of the conventions and 2) don dam gyi sgo nas, i.e., division by way of the ultimate.  Two of the schools under the rubric "division by way of  conventions", the tha snyad kyi mam gzhag nyan thos dang mthun pa'i phyi rol don yod par smra ba'i dbu ma pa the  existence of the exterior  object  (Madhyamikas who assert  in accord with the Sravaka  presentation of conventions) and the tha snyad kyi mam gzhag sems  1. dBus pa bio gsal includes, in his Grub mtha', the verse from Ye shes sde's ITa ba'i  khyad par that mentions the two schools. See Bio gsal grub mtha' 100b5. 2. For references to Nyi ma 'od's ITa ba'i rim pa, see Mimaki (1982), p. 42.  32 tsam pa  dang mthun pa mal  (Madhyamikas who presentation  of  classifications  'byor spyod pa pa'i  are Yogacaras in accord with  conventions)  relate  roughly,  dbu ma pa the mind-only  respectively,  to  the  Sautiimtika-Madhyarnika and Yogacara-Madhyamika of  the Bio gsal grub mtha'. It would also seem that the school termed tha snyad kyi rnam gzhag gang dang yang mi 'gal bar smra ba 'i dbus ma pa (Madhyamikas  who  assert without  contradiction  with  anyone  the  presentation of conventions) would correspond to the'Jig rten grags sde spyod pa'i dbu ma pa classification.  Sa skya pa Pandita Kunga rgyal  mtshan's classification can be presented as follows, 1) (kun rdzob kyi sgo nas dbye ba) - tha snyad kyi rnam gzhag nyan thos dang mthun pa'i phyi rol don yod par smra ba'i dbu ma pa - tha snyad kyi rnam gzhag sems tsam pa dang mthun pa rnal 'byor spyod pa pa'i dbu ma pa - tha snyad rnam gzhag gang dang yang mi 'gal bar smra ba'i dbu ma pa  2) (don dam gyi sgo nas dbye ba) - dbu ma sgyu ma lta bu - Rab tu mi gnas pa a) - Rang rgyud pa b) - Thai 'gyur ba We can note that the names Rang rgyud pa (Svataritrika) and Thai 'gyur ba (Prasangika) appear under the second rubric "division by way of the ultimate". Mimaki draws attention to another point regarding Sa skya Pandita's classification.  He writes,  33 II faut noter qu'il considere les Svatantrika et Prasangika comme sous-ecoles de 1'ecole des  Rab tu mi gnas pa.  L'Scole des sGyu ma lta bu et celle des Rab tu mi gnas pa sont deja connues dans un texte indien, la Tattvaratnavali d'Advayavajra.  L'originalitS de Sa skya Pandita consist a  ranger les ecoles des Svatantrika and des Prasangika sous celle des Rab tu mi gnas pa.  1  Bu ston Rin chen grub (1290-1364), writing only a few years after dBus pa bio gsal, in his Chos 'byung , the introduction to his recension 2  of the sNar thang Canon , classifies the Madhyamikas as follows, 3  dbu ma Thai 'gyur ba =  'Jig rten grags sde spyod pa'i dbu ma pa  5  Buddhapalita, Candrakirti  Sautrantika-Madhyamika  Bhavya  Yogacara-Madhyamika  Jnanagarbha, Kamalas'ila,  1. Mimaki (1982), p. 33.  4  Srigupta,  Santaraksita,  Haribhadra  Compare with the bKa' rgyud pa 'Ba ra ba rgyal mtshan dpal  bzang's arrangement of schools.  He equates the Rab tu mi gnas pa to the Thai 'gyur pa,  and the dBu ma sgyu ma lta bu to the Rang rgyud pa.  2. Abbreviation for bDe bar gshegs pa 'i bstan pa 'i gsal byed chos kyi 'byung gnas gsung rab rin po che 'i mdzod. 3. E Obermiller (1932), Introduction, p. 3. Roerich (1976) translates in the Blue Annals, p. 338, Bu ston Rinpoche brought the (original) copy of the bsTan 'gyur from sNar-thang and excluded from it all duplicate texts, for the original sNar-thang copy contained all the texts available at that time.  He classified the texts which had remained unclassified, as  well as added about a thousand new texts.  This new copy (of the collection) was  deposited at the vihara of Zha lu. 4. E. Obermiller (1932), p. 135. 5. Grags pa rgyal mtshan (Sa skya Pandita'a uncle) appears to have been the first Tibetan scholar to include the 'Jigs rten grags sde [phyod] pa school as a Madhyamaka subdivision. His examination of the schools is recorded by Go rams pa bsod nams seng ge in his rGyal  bu thams cad kyi thugs dyi dgongs pa zab mo dbu ma'i de kho na nyis spyi'i nag gyis ston pa nges don rab gsal. See Mimaki (1982), p. 33.  34 Bu ston Rin chen grub, for the first time within a Tibetan text, equates the Prasangika school to the 'Jig rten grags sde spyod pa school. that  he  terms  Jnanagarbha  a  Yogacara-Madhyamika,  Note in  contradistinction to the majority of later dGe lugs pa writers.  1  The fourteenth century (1310-1391) bKa' rgyud pa author 'Ba' ra ba rgyal mtshan dpal bzang includes the two schools recorded earlier by Sa skya Pandita - the dBu ma sgyu ma lta bu and the Rab tu mi gnes pa - in his Grub classification  mtha'i mam  bzhag and  autocommentary.  2  His  follows.  - Sautrantika-Madhyamika - Yogacara-Madhyamika - sNang ba mi spyod dbu ma [='Jig rten grags sde spyod pa'i dbu ma pa] a) dBu ma sgyu ma lta bu [=Rang rgyud pa] b) Rab tu mi gnas pa [=Thal 'gyur ba] In contrast to Sa skya Pandita, who considered the Svatantrika and Prasangika as subschools of Rab tu mi gnes pa, author 'Ba' ra ba rgyal mtshan dpal bzang equates the Prasangika school with Rab tu mi gnes pa and equates the Svatantrika with dBu ma sgyu ma lta bu. These in turn are considered sub-schools of the 'Jig rten grags sde spyod pa'i dbu ma pa.  With regard to the fact that he too cites Jnanagarbha's  1. Even the later dGe lugs pas disagreed with each other regarding Jnanagarbha's school. D. Lopez (1987), p. 446, n. 16, in accord with the previous quote from M. D. Eckel (1986) states that the "philosophical climate of late Indian Buddhism was substantially more fluid" than the later Tibetan works might imply. 2. Tsong kha pa, in his lHag mthong chen mo, asserts that some of the Tibetans who preceded him classified Madhyamikas dependent on their stands with regard to the ultimate. According to him, the two groups were the sGyu ma rigs grub pa (Establishers of Illusion through Reasoning) and the Rab tu mi gnas par smra ba (Proponents of Thorough NonAbiding). The sGyu ma rigs grub pas asserted that the ultimate truth is a composite of emptiness and appearance, while the Rab tu mi gnas par smra bas asserted that the ultimate truth is the mere elimination of superimpositions with respect to appearance. See D. Lopez (1987), p. 56; K. Mimaki (1982), p. 33 n. 67.  35 Satyadvayavibhanga-karika 21 and Candrakirti's Madhyamakavatara VI 35 to support his classification of 'Jig rten grags sde spyod pa'i dbu ma pa, Mimaki writes, "il est possible que dans sa classification il ait ete influence par dBus pa bio gsal".  1  By the fifteenth century, the Sa skya pa scholar Pan chen Sakya mchog ldan (1428-1507) was in a position to refer to and criticize the tradition of Madhyarnika classification that had arisen in Tibet.  In his  dBu ma mam par nges pa'i chos kyi bang mdzod lung dang rigs pa'i rgya mtsho he presented, as he viewed it, the classification according 2  to the earlier masters (sngon gyi slob dpon mams) and, following that, set forth his own categorization. I. sngon gyi slob dpon rnams (earlier masters) (kun rdzob) - Sautrantika-Madhyamika  Bhavaviveka, Jnanagarbha  - Yogacara-Madhyamika  S a n t a r a k s i t a , Kamalasila  - 'Jig rten grags sde spyod pa'i dbu ma pa  Candrakirti  (don dam) - Svatantrika - Prasangika II. his position - grub mtha' 'og ma -* Madhyamika - Jnanagarbha  Sautrantika  - Santaraksita  Madhyamika  Sautrantika -» Yogacara -* Madhyarnika  - not grub mtha' 'og ma, but 'Jig rten grags pa nyid -•. Madhyarnika  Sakhya  Candrakirti  mchog ldan was of the opinion that the Madhyamikas could be  divided into two main groupings.  However, unlike his predecessors,  1. Mimaki (1982), p. 35.  2. Chapter 2 (kha): dBu ma thai rang gi gyes 'tshams dang grub mtha'i gnas bstan pa. Found in The Complete Works of gSer mdog Pan chen Sakhya mchog ldan.  36 who divided Madhyamikas relative to their views on conventionalities and  ultimates, he divided them with respect to whether not they  achieved the Madhyamika view by way of initially working with the lower tenets (grub mtha' 'og ma), i.e., the tenets of the Sautrantika and Yogacara schools.  Those who achieved the Madhyamika view without  first being involved with the lower tenet systems followed "what is known in the world" ('jig rten grags pa nyid) and thereby moved directly into the Madhyamika view.  According to his classification,  there is the implication that there were scholars whose views related to more  than  one  school.  For example,  Santaraksita,  consistently  considered a Yogacara-Madhyamika, here is seen to have attained the Madhyamika view after having relied on both the Sautrantika and Yogacara  tenet systems.  Probably this  relates  to  the  fact  that  Santaraksita (and Kamalaslla) held views that were in accord with certain views within the Sautrantika school.  1  Other subdivisions of the Madhyamika existed; those presented here 2  are a small, yet representative group that serve to demonstrate the efforts made by scholars, beginning in the early ninth century, to attempt to differentiate the schools that adhered to the Madhyamika doctrine.  Development of Madhyamika Classification It may prove beneficial, at this point, to review briefly the historical development of the Madhyamika school in Tibet. Ye shes sde's ITa ba'i khyad  par,  representative  written of  at  the  beginning  of  the  the period known as the  is  snga dar - the early  propogation of Buddhist thought and technique. entry and assimilation of Buddhism into Tibet.  ninth century,  This period saw the It commenced with the  1. The main view they held in commom with the Sautrantika school was that of the momentariness of things (ksanabhanga, skad cig ma 'jig pa). See Mimaki (1976), p. 74-5. 2. Other classifications may be found in Mimaki (1982), p. 27-37.  37 visits of Santaraksita and Kamalasila to Tibet in the eighth century, and with the exoduses of Tibetans to the great Indian monasteries to seek out teachings.  At this time a preliminary attempt was made to interpret  and systematize the material acquired and translated. The snga dar in 1  Tibet "ended" with the disintegration of the political structure in the mid-ninth century; a new era, known as the phyi dar began, in large part due to the efforts of the western kings in the eleventh century.  D.S  Ruegg  "full  classifies  the  second  propogation  as  the  period  of  assimilation" , encompassing the years from the end of the tenth to the 2  fourteenth centuries.  It was marked by continued effort on the part of  the Tibetans to penetrate, explain, and organize Buddhist philosophy. This period was followed by the "classical period" that lasted from the 3  fourteenth  to  systematization  the of  sixteenth the  centuries,  religious  in  which  and philosophical  exegesis materials  and were  stressed. It was during this initial diffusion of Buddhism, strongly influenced by Santaraksita and Kamalasila, that Ye shes sde and dPal brtsegs came to include the  name rNal  'byor spyod pa'i dbu ma (Yogacara-  Madhyamaka) in their texts.  During these early years few Prasangika  texts were translated:  out of fifty-one Madhyamika texts translated  from Sanskrit to Tibetan, only five were texts of the Prasangika school.  4  The paucity of translated Prasangika texts resulted in the better  acquaintance of scholars with the works by Bhavaviveka; because of this, the second name, mDo sde (spyod) pa'i dbu ma (SautrantikaMadhyamika) came to be included in the ITa ba'i khyad par and the ITa ba 'i rim pa bsad pa. The use of the terms for the two schools became widespread, and Tibetan  authors,  their  attention  focused  on  the  composition  of  exegetical treatises, began to include the newly-coined designations in 1. D.S Ruegg (1980), p. 278. 2. Ibid. 3. Ibid. 4. Mimaki (1982), p. 45.  Yuktisastika-vrtti.  The only work by Candraklrti that was translated was the  38 their commentaries.  By the eleventh century, the rNying ma pa Rong  zom Chos kyi bzang po had utilized the terms in his commentary on 1  Padmasambhava's eighth century Man ngag lta phneng, an Indian text 2  in which they had not been employed.  Even though the Kashmiri nun  Laksml used the two terms in her Paffcakrama&a-Kramarthaprakas'ika, written at the beginning of the eleventh century (two centuries after Ye shes sde introduced them) it is likely that she learned of them from Tibetan  sources;  the terms Yogacara-Madhyamika and Sautrantika-  Madhyamika had not been employed in works by Indians prior to hers. It seems, based on the information that has been accrued to date, that the Tibetans were the inventors and instigators of the use of the two appellations. At the commencement of the phyi dar, and in the years that followed, Candrakrrti's books and many other Prasangika texts were translated. Due  to an increased familiarity with the Prasangika texts, and an  understanding of their relationship to aspects of Bhavaviveka's works, the names Rang rgyud pa (Svatantrika) and Thai 'gyur pa (Prasangika) came to be coined and used.  Thus we can see that the two groups of  terms, the Sautrantika-Madhyamika and Yogacara-Madhyarruka on the one hand, and the Rang rgyud pa and Thai 'gyur pa on the other, arose from within quite different historical contexts. groups  of  appellations  responded  approaches to Madhyamika thought. positions  taken  by  In addition, the two  respectively  to  two  different  The first grouping related to the  Madhyamikas with  regard  to  the  status  of  conventional phenomena, and the second grouping was made according to whether a Madhyamika employed syllogisms or consequences to bring about an understanding of the ultimate. Hence, until the later dGe lugs  pas  subsumed  the  Sautiantika-Madhyamika and Yogacara-  Madhyamika under Svatantrika, the two sets of terms, as in the case of dBus pa bio gsal, remained differentiated.  1. Rong zom Chos kyi bzang po also authored the first two Grub mtha' of the later  diffusion: the ITa ba'i brjed byang and the Grub mtha' brjed byang. 2. Mimaki (1982), p. 44.  39  It appears that Pa tshab Nyi ma grags (1055-?), the Tibetan scholar who translated a number of Cancfraklrti's works, was the first to use the terms Rang gyud pa and Thai 'gyur pa. As in the case of the terms coined during the early diffusion, the designations Rang gyud pa and Thai 'gyur pa, noted above, were also Tibetan innovations. The fact that the terms were Tibetan products was known and recorded by the Tibetans. This is evident in the writings of Tsong kha pa and of Sakya mchog ldan Tsong kha pa (1357-1419) is credited with having initiated the melding of the two groups. In his Lam rim chen mo, composed in 1402, he employs the terms Sautrantika-Madhyamika, YogacaraMadhyamika, Svatantrika and Prasangika, but does not explicitly bring together the two groups. Neither does the Drang nges legs bshad snying po, written in 1406, explicitly set forth a conmingling of the two sets of classifications. However, since in the latter part of the treatise he subdivides the Madhyamika into Prasangika and Svatantrika, with Svatantrika further subdivided into two - groupings of Madhyamikas according to their adherence to Bhavaviveka's position or to Santaraksita and KamalasTla's position - Drang nges legs bshad snying po is held to be the first work to contain the unification of the two classifications. His classification of the Mahayana Buddhist schools is as follows: 1  2  3  4  1. Lam rim chen mo (kha) 6b6-7: Gangri'ikhrod kyi phyi dar gyi mkhas pa mams dBu ma pa la Thai 'gyur pa dang Rang rgyud pa gnyis kyi tha snyad byed pa ni Tshig gsal dang mthun pas rang bsod mi bsam mo II  "The scholars of the second diffusion of Buddhism  made, for the Madhyamika, the two names Svatantrika and Prasangika, but I do not think they are pure invention, because they accord with the Prasannapada."  2. dBu ma rnam par nges pa'i bang mdzod lung dang rigs pa'i rgyo mtsho (kha) 13b3: Rang rgyud pa dang Thai 'gyur pa shes bos mams kyis tha snyad byas pa mams soil "Svatantrika and Prasangika are names made by the Tibetans." 3. See Mimaki (1982), p. 46, for the folios that record the classification. 4. Mimaki (1982), p. 46.  40  Vijnanavadin Madhyamika 1) Svatantrika Bhavaviveka Santaraksita and KamalaSlla 2) Prasangika It was  this basic plan, with the Yogacara-Madhyamika and  SautTantika-Madhyamika  1  school subsumed under the Svatantrika  division, that the later dGe lugs pa scholars followed and incorporated into their works. Any certainty to the conclusion that Tsong kha pa was the first to combine the two sets of classification is undermined by statements from a Bon po Grub mtha'. Tre ston rGyal mtshan dpal's Bon sgo gsal 2  byed (also known as Bon sgo dkar chag) also reveals a classification of the Buddhist schools that includes the Sautrantika-Madhyamika and Yogacara-Madhyamika as subschools of the Svatantrika. According to historical information on and by the Bon po, it is highly unlikely that Tre ston rGyal mtshan dpal produced his Grub mtha' after Tsong kha pa's lifetime, and it is quite probable that he lived during the latter half of the thirteenth century and the first part of the fourteenth century (possibly a contemporary of dBus pa bio gsal).  In his Grub mtha' he  writes, With regard to the Madhyamikas there are the three:  the  Saurrantika-Madhyarnikas who assert the luminous mind, devoid of the two [i.e., subject and object]; the YogacaraMadhyamikas who assert [the meditation] like the veins and winds; and the Grags ste spyod pa'i dbu ma pas who 1. Tsong kha pa did not employ these exact terms in Drang nges legs bshad snying per, he divided the Svatantrikas according to whether they held Bhavaviveka's or Santaraksita and Kamalailla's position. 2. Mimaki details the problem. See Mimaki (1982), p. 47-51.  41  assert emptiness free from extremes. two  are called  Moreover, the first  Svatantrika-Madhyarnikas and the  last  Prasangika.  1  It would seem that Tre ston rGyal mtshan dpal was not wellacquainted with the Indian Buddhist schools. Sautrantika-Madhyamika  corresponds  better  Madhyarnika, and he seems, in his statement  His description of the to  the  Yogacara-  regarding the Yogacara-  Madhyamika, to confuse the "yoga" of Yogacara with the practice of tantric yoga.  However, leaving his confusion aside , one must search 2  for the possible basis for his statement, "the first two are said to be Svatantrika-Madhyamika and the last is Prasangika".  He could have  turned to Bu ston, dBus pa bio gsal or 'Ba' ra ba to equate Grags ste spyod pa'i dbu ma pa  3  to Prasangika.  Mimaki suggests that if his  source was Bu ston, Tre ston rGyal mtshan dpal might have classified Sautrantika-Madhyamika  and  Yogac^a-Madhyamika  under  Svatantrika simply because Bu ston equated 'Jig rten grags sde spyod pa with Prasangika. Whatever his sources, the existence of the Bon po Grub mtha' suggests that Tsong kha pa's classification had precursors and merits further attention. In conclusion, let it suffice to say that the Grub mtha' literature is an apt testimonial to the efforts made by Tibetan Buddhist scholars to analyze and categorize Indian Buddhism.  In general, Bio gsal grub  mtha' is representative of the attention that was given to classification and elucidation of the Indian schools and, in particular, this Grub mtha' represents such early fourteenth-century attention and effort.  In the  Madhyamika section of Bio gsal grub mtha', by way of terse karikas,  1. The Tibetan is found in Mimaki (1982), p. 48.  dBu ma la gsum / mDo sde spyod pa 'i dbu ma pa ni I sems gal ste gnyis med du 'dod I rNal 'byor spyod pa 'i dbu ma ni / rtsa dung ltar 'dod I Grags ste spyod pa 'i dbu ma ni / stong nyid mtha' bral du 'dod / de yang dang po gyis ni / dBu ma rang rgyud pa I phyi ma ni dBu ma thai 'gyur ba shes smra ba 'o / 2. Here, too, one cannot discount the possibility of textual corruption. 3. 'Jig rten grags sde spyod pa'i dbu ma pa.  42  brief commentary on the karikas, and numerous quotations from Indian Buddhist texts, dBus pa bio gsal presents his analysis of, as it has been regarded in Tibet, the most profound school of Buddhist thought.  43  The  Madhyamika  View:  The Middle Way  In the Exhortation to Katyayana (Ka tya ya na yi gdams nag, Katyayanavavada) the Bhagavan, who knows being and non-being, negated both "it is" and "it is not". M M K X V k 7, as quoted in Bio gsal grub mtha' 103bl  Nagarjuna, writing in the second century C.E., sought to reiterate and clarify what he felt to be the Buddha's understanding of the nature of existence in order to refute  the philosophical positions  that had  developed and become prevalent in the centuries that followed the Buddha's Nirvana. The Madhyarnika school of Buddhism came to base itself on his exegesis of the Tathagata's teachings, and took its name from Nagarjuna's principal work, the Mula-madhyamaka-karikas. The Madhyamikas who appeared subsequent to Nagarjuna returned again and again to his texts to quote, explain, and elaborate on his philosophy. The Madhyarnika is the only Buddhist school to assert the emptiness (stong pa nyid, Sunyata) of all aspects of existence. Later Madhyamikas defined  themselves  propounding  in  their  treatises  as  those  Buddhist tenets  asserted  that  no  dharma), including particles, truly existed.  1  persons  who,  in  phenomena (chos,  They also stated that they  were the only Buddhists who maintained a middle way free from the 1. Geshe Sopa and Jeffry Hopkins (1976), p. 122. This is the definition given in dKon mchog 'jigs med dbang po's (1728-91) Grub pa'i mtha'i mam par bzhag pa rin po che phreng ba. The Tibetan is found in Iida (1980), p. 28:  bden grub kyi chos rdul tsam yang med par khas ten pa'i grub mtha' smra ba'i gang zag de dbu ma pa'i mtshan nyid  44  two extremes of permanence and annihilation, and claimed that they fully  upheld  the  meaning  of  their  name  'Madhyamika",  translates as "belonging to, or pertaining to, the middle".  which  The later  Tibetans, in translating the title of the school founded by Nagarjuna, kept the original meaning and rendered the name of the school the dBu ma, "The Middle", and called adherents to it dBu ma pa, "Persons of the Middle". Although only the Madhyamika school actually has the word "middle" inherent in its name, it is not recognized by the other Buddhist schools as the only school to adhere to middle way.  Jacques May  writes, Now,  the  Middle  Way  constitutes  principle of Buddhism as a whole. conceived  that  all  Buddhists  Madhyamaka or Madhyamika. Indeed,  the  Vaibhasikas,  the  fundamental  So it could be  should  be  called  1  Sautrantikas  and  Yogacaras,  while  not  insisting that they too be termed Madhyamikas, also claim to follow a middle way. Adherence to the middle way basically means not leaning toward or following extremes.  These extremes, however, have received different  interpretations dependent upon the school in which the middle way has been couched, and while each Buddhist school makes the claim to avoidance of the extremes, the "higher" schools insist that, from their point of view, those below have involved themselves with one or both of the extremes. According to later Tibetan interpretation, the Vaibhasikas, who base themselves primarily on the Mahavibhasa, an abhidharma text, assert partless particles of matter and consciousness and the true existence of these, yet claim to avoid the extreme of permanence by refuting what they interpret to constitute a view of their permanence: the existence of  1. Jacques May (1978), p. 233.  45 the principal (pradhana, gtso bo) of the Samkhyas.  1  They further claim  to adhere to the middle through their refutation of the extreme of annihilation, which, in their school, is seen as the rejection of the law of cause and effect of actions. The Sautrantikas, or the other Hinayana school of tenets, emphasize the sutras over the abhidharma texts, and also assert the true existence of external objects.  In accord with the Vaibhasikas they tread the  middle way which avoids the positing of the principal and the rejection of cause and effect. The Yogacarins, belonging to the school of one of the two major divisions of Mahayana, believe that all phenomena are of the nature of the mind.  Within their scheme of the three natures they find their  middle way:  the extreme  of permanence,  instead  of being the  unchanging agent of the Hlnayanists, here is viewed as being the true existence of the first of the three natures, the imaginary (kun  btags,  parikalpita) natures. To believe that objects are not the nature or entity of  consciousness  and to  conceive  of  them  as  separate  from  consciousness is, for the Yogacaras, to slip to the side of permanence. Slippage into a view of annihilation would involve the denigration of the second and third natures - the assertion that the dependent (gzhan dbang, paratantra) natures and the thoroughly established (yong grub, parinispanna) natures do not exist at all. They avoid this extreme by asserting that these two natures ultimately exist, that is, that objects ultimately exist as the nature of consciousness, that consciousnesses truly exist, and that the emptiness of the duality of subject and object truly exists.  2  Thus, by avoiding these two extremes, the Yogacaras claim to follow a middle way free from the extremes of permanence and annihilation.  1. Lopez, (1987), p. 42. 2. Ibid., p. 43.  46 The Madhyamikas, however, find fault with the so-called "middle paths" of the lower schools, charging that none of them are able to maintain a middle stance and do, in fact, fall to extremes.  For the  Madhyarnika, the extremes are not those set forth by the Vaibhasika, Sautrantika  or  Yogacara,  existence of phenomena. for  the  Madhyamika,  and rather involve  extreme  modes  of  The actual mode of existence of phenomena, is  without  self-nature  (rang zhin med  pa,  nihsvabhava), i.e., without true existence (bden par med pa, satya asat). This mode of existence, this "emptiness" (stong pa nyid, $unyata) of phenomena is absolutely necessary  for maintenance of the law of  dependent-arising taught by the Buddha.  Without it, the dependent-  arising of phenomena would be impossible to maintain as a true Buddhist should.  Phenomena with self-nature would be unable to  change, since they would exist in and of themselves, independent of, and different from, all other phenomena:  a frozen mode of existence  that would not allow for the fluidity of dependent-arising. Thus, for the Madhyarnika, the extreme of permanence is not the adherence to a view that asserts a "principal" or to a view of the true existence of imaginary natures but is rather the conception of the ultimate existence of phenomena (don dam par grub pa, paramarthasiddha). The ascription of ultimate existence to phenomena superimposes a mode of existence (gnas lugs) that is not in harmony with the actual nature of things. Anything that ultimately exists would be unable to change and would not be related to anything else. Since this does not accord with the real mode  of existence of phenomena,  a view  that assumes ultimate  existence is termed an extreme, specifically, an extreme of permanence. The Madhyamikas hold that phenomena do not ultimately exist, but in refuting the extreme of permanence, do not negate phenomena entirely. To do so, according to them, would be to fall to the extreme of annihilation. removal  of  This opposite ultimate  extreme  existence  completely eliminate phenomena.  from  would not only involve phenomena,  but  would  the also  There would occur a denigration of  phenomena to the extent that the name of the extreme implies: "annihilation" of phenomena to complete non-existence.  an  47 Thus the Madhyamikas follow a middle path that refutes ultimate existence but affirms the conventional existence (kun rdzob tu yod pa, samvrtisat) of phenomena. Both extremes are avoided with the singular assertion  of  the  dependent-arising  of phenomena,  for  "dependent-  arising" infers and affirms the conventional existence of phenomena and yet negates their ultimate existence.  This affirmation of the  conventional existence of phenomena keeps the Madhyamikas from the extreme of annihilation and, at the same time, because phenomena are understood to exist interdependently, and not inherently, holds them from the extreme  of permanence.  ultimate existence is rejected.  Phenomena are retained; their  Emptiness is simply the lack of ultimate  existence of phenomena. It (emptiness) is designated such to give expression  to the final nature of all phenomena.  The fact that  phenomena are empty allows for dependent-arising. Based on the Madhyamika assertion of the emptiness of phenomena and the statement's positive implication of dependent-arising all the other schools' extremes are seen to be "trifling and fabricated", and the 1  views of the schools themselves are seen to fall to an extreme.  The  Vaibhasika position that allows for the ultimate existence of the partless atoms of matter and consciousness, according to the Madhyamika, falls to the extreme of permanence.  The Sautrantika's assertion of the  ultimate existence of things likewise finds itself mired in a conception of permanence.  The Yogacarins, under the critical eye  of their  Mahayana cohorts, in maintaining that the dependent and thoroughly established natures truly exist also slip to the extreme that has snared the  Hinay artists.  The  oral  commentary  provided by  a modern  Madhyamika further implicates the positions held by the Vaibhasikas, Sautrantikas, and Yogacarins. If a phenomenon is truly existent, it must be unchanging in nature; thus for it to cease, it must become utterly nonexistent.  l . Ibid.  Since  the  Vaibhasikas,  Sautrantikas,  and  48 Yogacarins  assert  that  these impermanent phenomena,  which they hold to be truly existent, disintegrate every moment,  they  thereby come  to  hold the  extreme  of  annihilation.  1  Hence the Madhyamikas contend that only adherents to their school warrant the appellation "Proponents of the middle way".  The views  held by the other schools lead their members to misappropriate the title and ultimately draw them into the philosophical quicksand that those following the Madhyamika middle path so discreetly, by refuting both "it is" (yod pa, sat = permanence) and "it is not" (med pa, asat = annihilation) avoid.  The Madhyamika Object of Negation  Here, the object of negation is just the belief in the truth [of appearances and emptiness] and not the appearances [themselves].  This is because [there is] suffering in that  [belief] and also because it is not necessary or possible [to negate] appearances.  2  The Madhyamika philosopher's inquiry into the nature of phenomena does not come about from mere intellectual speculation or curiosity, but rather serves a soteriological purpose.  1. Ibid., p. 44. 2. Bio gsal grub mtha', 104b2, k XH-14.  Realization of the emptiness  49 of self and phenomena is said to bring about liberation from cyclic existence and, in the same way that the Buddha's inquiry into reality had brought about the achievement of vast intellectual understanding, the attainment of Nirvana, with its concomitant joy and freedom from the suffering of cyclic existence, so is the philosopher/meditator's investigation said to have the power to bring about the highest wisdom and experience.  The main shackle that prevents liberation and keeps the  individual circling endlessly throughout the realms of existence is said to be the conception of true existence of phenomena.  1  The conception  of true existence is said to cause sentient beings to attach a mistaken reality to phenomena which, in turn, drives them to desire or to be repulsed by the various objects of consciousness. to commit actions that samsara.  They are thereby led  bring about suffering and continued bondage in  This conception of true existence is also known as the  "extreme of permanence" or "the extreme of existence" (yod mtha', astyanta), and a synonym for this "extreme of existence" which sheds some light onto the meaning of the term "true existence" is termed the "extreme of superimposition" (sgro 'dogs kyi mtha', aropanta). According to the Madhyamika, on the occasion of the perception of phenomena, a mistaken conception is "superimposed" onto phenomena, thereby "covering" the actual nature of phenomena to masquerade as this actual nature. This mistake is made from the side of the mind and is imposed onto the things of the world, misleading the mind as to the actual status of things. The error is habitual and innate, so much a part of the fabric of the mind that every other doctrinal system (according to the Madhyamika), has included it under the rubric "reality".  That  which is superimposed onto phenomena is the conception that things truly exist, independently. Western scholars examining the Madhyamika presentation of the mode of existence of phenomena have chosen to rely on varying 1.  Hypothetical synonyms  establishment pa,  for true existence (bden par yod pa, satya-sat) are true  (bden par grub pa, satya-siddha), ultimate establishment (don dam par grub  paramartha-siddha), established properly (yang dag par grub pa, samyaksiddhah) and  established through its own entityness (ngo bo nyid gyis grub pa, svabhavata-siddha).  50 Sanskrit and Tibetan terms to describe the object of the Madhyamika negation,  which  superimposition.  negation Richard  is  implicit  Robinson,  in  in a  concepts short  such  presentation  as of  Madhyamika thought, defines the object that is to be negated, svabhava, as, ...a term that means something 1) existing through its own power rather than that of invariant  and  inalienable  another, 2) mark,  and  possessing 3)  having  an an  immutable essence.  1  He goes on to state, Intellectually, svabhavas are false reifications, conceptual figments.  Emotionally, they are the foci of obsessions, the  illusory idols which enslave the passions.  2  In a recent work Lopez, in explaining his choice of English word "entityness" to represent the Madhyamika object of negation writes, "Entityness" is chosen here to suggest something that is capable of independent existence,  something similar to  substance, as described by Wittgenstein in The Tractatus Logico-Philosophicus  2.024,  "Substance is  what exists  independently of what is the case." (Pear and McGuiness translation).  It would be the Madhyamika position that  phenomena have entities, that is, that they exist, but that entityness, some kind of substantial property or absolute object, is a falsity hypostasized by ignorance.  Entityness  seems akin to what Heidegger calls "the thingness of the  1. Richard Robinson (1970), p. 51-52. 2. Ibid., p. 52.  51 thing" in his essay "The Origin of the Work of Art". He writes: This block of granite, for example, is a mere thing. It is hard, heavy, extended, bulky, shapeless, rough, coloured, partly dull, partly shiny. features  in  characteristics.  the  We can take note of all these  stone.  Thus  we  acknowledge  But still, the traits signify  proper to the stone itself.  its  something  They are the properties.  The  thing has them. The thing? What are we minking of when we now have the thing in mind? Obviously a thing is not merely an aggregate of traits, nor an accumulation of properties by which that aggregate arises. everyone  thinks he knows, is  properties have assembled.  A thing, as  that around which the  We speak in this connection of  the core of things. The Greeks are supposed to have called it hupo  keimemnon. For them, this core of the thing is  something lying at the ground of the thing, something always already there. called ta  The characteristics, however, are  sumbebekota, that which has always turned up  already along with the given core and occurs along with it.i  Svabhava literally means "[its] own (sva) existence or being or nature (bhava)". Translated by Western scholars in a variety of ways, it is 2  3  categorically denied by the Madhyamika. The "thingness of the thing" in  this Buddhist system is no more than the result of * mistaken  perception.  It is understood to be a conceptual construction that has as  much reality as the traditional Tibetan Buddhist examples  of non-  existents, for example, the horn on a rabbit's head or a turtle's hairs. A contemporary Tibetan Madhyamika scholar comments: 1. D. Lopez (1987), p. 445, n. 12. 2. William Ames (1982), p. 161. 3. For example:  "own being, aseity" (Ruegg); "self-being" (May); "substantial existence"  (Iida); "substance" (Lindtner); "intrinsic nature" (Ames); "inherent existence" (Hopkins).  52 ...most fundamentally it is that inherent existence is, in fact,  totally  imaginary  and  non-existent.  To  exist  inherently would mean to exist independently of any and all  conditioning factors.  For example,  an inherently  existent table would be one that somehow exists entirely from the side of its own intrinsic, essential nature.  It  would be something that stands out all by itself, as though its  causes,  its  parts  and  its  being  conventionally  apprehended as a table all had nothing to do with its essential being.  It is rather easy to recognize intellectually  that such a mode of existence is a logical impossibility; nevertheless, we instinctively apprehend all phenomena including ourselves- as existing in exactly this impossible manner.  1  It is this hypostasized, fabricated "reality", posited by fundamental ignorance to be the nature of phenomena, that is, as dBus pa bio gsal writes in his karika 14, negated by the Madhyamika. phenomena are not refuted.  Conventional  Their refutation would leave a sheer  nothingness, an annihilation of all.  Thus, in his explanation to his  karika 14, dBus pa bio gsal writes, Therefore,  believing  in the  truth of  appearances  and  emptiness, which is the basis of all suffering, is the object of negation.  2  1. Geshe Rabtan (1983), p. 9.  2. Bio gsal grub mtha', 104b3.  53 The Two Truths Buddhist philosophers, in their attempt to explain and distinguish the plethora of knowables, set forth many varied categorizations of objects of knowledge.  The categorization that is emphasized and employed by  1  the Madhyamika is headed by the rubric "two truths" (bden pa gnyis, satya-dvaya). In an essay where he substantiates his assertion with mention of Nagarjuna's Mulamadhyamikakarika XXIV, Candraklrti's Madhyamakavatara V I , and Santideva's Bodhicaryavatara I X , T.V.R. 2  Murti states, All  Madhyamika  treatises  take  the  two  truths  -  paramarthasatya and samvrtisatya - as vital to the system; some even begin their philosophical disquisitions with the division.  3  dBus pa bio gsal begins the Madhyamika section of Bio gsal gmb mtha' with an explanation of the two truths.  His first karika,  emphasizing the all-inclusiveness of the two truths, reads, The perfect Buddha himself taught objects of knowledge in just two truths. Therefore, objects of knowledge, which are also the five categories  4  are not other than the two  truths.  5  Later, in his sixth karika he emphasizes the fundamental importance of the two truths in the quest for liberation and Buddhahood. 1. Objects of knowledge  (shes bya, jfTeya)  include all objects that are cognized by  consciousness. Since emptiness can be cognized by the highest wisdom consciousness, it too is an object of knowledge. 2. T.V.R. Murti (1973), p. 16-17. 3. Ibid., p. 25, n. 16. 4. For the five categories, see B S G T 96b4. composite factors, and unconditioned phenomena. 5. Bio gsal grub mtha', 96a5, k XII-1.  They are form, mind, mental factors,  54  From knowing or not knowing the two things [two truths] one finds or does not find the state of Omniscience.  1  In a quotation from the Pitaputrasamagama-sutra he reveals the identity of the two truths: conventional (kun rdzob, samvrti) and ultimate (don dam, paramartha). Thus  the  conventional  Tathagatha [truth]  understood  and ultimate  the  two  [truth].  truths,  Objects of  knowledge also are exhausted in these, conventional and ultimate  [truths].  2  The two truths exhaustively include all objects of knowledge.  They  "cover" all objects of knowledge and, including them all, do not allow for a third truth.  Since they are mutually exclusive, no object of  knowledge exists that is either both or neither; hence, the possibility for a third  truth is  nullified.  All objects  of  knowledge,  including  emptiness, and all other possible categorizations of objects are included under the umbrella of the two truths, i.e., are either one or the other. Truths however, exist as they appear.  3  It would be pointless, and  actually quite impossible, to have two different sets of truths, were they to uphold this definition of truths.  Conventional truths appear one way  (truly existent) and exist another (devoid of true existence).  Since  conventional truths do not exist the way they appear they do not fulfill the requirements of "truth".  Only emptiness exists as it appears to the  mind cognizing it and thus  only emptiness qualifies  as  a truth.  Emptiness is, in fact, ultimate truth. The division into truths, then, does not find its base in "truths". dBus pa bio gsal writes,  1. Bio gsal grub mtha', 99b4, k XII-6. 2. Ibid., 96a6. 3. Jeffrey Hopkins (1983), p. 405  55  [The two truths are] distinguished only from [the point of view] of appearances. The union [of the two truths] is the Middle Path.  Appearances  1  (snang ba) are the basis of the division into two truths.  In his elaboration on karika XII-4c-d, where dBus pa bio gsal writes, "The  definitiveness  of  the  number  is  contingent  on  consciousnesses," he clarifies the fact that the two truths 2  the  find  two  then-  bases not in truths, but in appearances, and that they are dependent for their designations on different types of consciousnesses.  Within the  Madhyarnika system of the two truths, clearly one of the truths, from the point of view of the qualifications necessary for a "truth", i.e., existing as it appears, does not represent the actual status of phenomena. dBus pa bio gsal etymologizes the word samvrtisatya (conventional truth) and asserts that it is "the truth which completely conceals".  3  He  goes on to quote Candrakrrti's Madhyamakavatara VI k 28. The error [in the mind] is the one which completely conceals because it obscures the real nature. On account of that whatever is artificial seems to be truth. The Muni said that [krtrima] is. truth for that which completely conceals and the artificial things [he called] samvrti, that which completely conceals as well.  4  Two aspects of samvrti are revealed in the above quotation. Samvrti is the  concealer,  i.e.,  the  ignorant  consciousness  that,  due  to  its  misapprehension of the nature of phenomena, assumes that what is  1 Bio gsal grub mtha', 97b6, k XII-4c-d. 2 Ibid., 98al. 3. Bio gsal grub mtha', 98bl.  His etymology, spelled out, would be as follows:  The  etymology of samvrtti (sic) -satya is a truth (satya=bden pa) which completely (sam=yang  dag) conceals (vrti= sgrib byed). 4. Bio gsal grub mtha', 98bl.  56 empty of true existence is truly-existent.  The objects of knowledge  onto which the concealing consciousness superimposes true existence are also known as concealers.  1  The objects are called concealers  because, when cognized by the ignorant consciousness they do not represent their actual nature. They are, in a way, like costumes worn by actors in a play - hiding the actual identity of the actor while causing the audience to mistake the character played by the actor for the person so costumed.  They are called conventional truths not because they exist  as they appear, i.e., laden with true existence, but because they are truths for an ignorant consciousness, which assumes that all objects exist the way they appear. An ignorant consciousness is not just a consciousness which does not know suchness but a concealer of suchness through actively objective  assenting  existence.  to the false  appearance of  Even in direct sense perception,  forms, sounds, odors, tastes, and tangible objects appear to exist inherently, as if they existed in their own right. All phenomena except emptinesses appear to the  cognizers  that certify them as existing as if they exist inherently.  2  The concealing truths, or, as they are more commonly known conventional truths - are not established by the ignorant consciousness. If they were established "error" (gti mug,  moha)  by the mistaken consciousness, when the  on the mind was removed and replaced with a  consciousness non-mistaken as to the mode of existence of objects, conventional truths would completely disappear, thereby making the middle way as defined by dBus pa bio gsal (the union of the two truths) impossible.  Only the fabricated true existence of conventional truths is  1. The three aspects of samvrti are found in Candraklrti's Prasannapada, where he comments on Mulamadhyamakarika XXIV.8. The third aspect is "conventions of the world" Cj'grtenpa'i tha snyad). See T.V.R. Murti (1973), p. 17; D. Lopez (1987), p. 2067; J. Hopkins (1983), p. 417-8. 2. J. Hopkins (1983), p. 417.  57 deigned existent by the ignorant consciousness; the removal of the conception of true existence that comes about with the cognisance of the lack of true existence of objects does not negate conventional truths. When true existence is refuted, their conventional existence is not eradicated. reality.  Mere conventionalities remain, free of any superimposed  The removal of true existence and the retention of mere  conventionalities make possible the middle way.  Real and False Conventional Truths Conventional truths are further divided into the two categories of real conventional truths (yang dag kun dzob bden pa, tathyasamvrtisatya) and  false  conventional  truths  (log  pa'i  kun  mithyasamvrtisatya). Real and false conventional  rdzob bden pa, truths differ  according to the Madhyarnika subschool that posits them. the  Svatantrika-Madhyamikas, real conventional  truths  According to are divided  dependent upon their ability or inability to function (don phed nus pa, artha-kriya). dBus pa bio gsal describes the Svatanriika-Madhaymika divisions: Among these two, functional appearances are free from imagination, arise from causes and appear in conformity with what can perform a function; those lacking [these qualities] are non-functional appearances.  1  The  traditional  Svatantrika-Madhyamika example  conventional truth is water.  of  a real  Water is drinkable - therefore functionable  - and is suitable for the quenching of thirst - therefore functionable. Water in a mirage serves to illustrate a false conventional truth:  1. Bio gsal grub mtha', 98M.  it  58 appears to be drinkable and suitable for the soothing of thirst but is unable to fulfill either function. The Prasangika-Madhyamikas disagree with the Svatantrika assertion of real and false conventional truths, holding that since all conventional consciousnesses are mistaken with regard to their objects - taking them to exist as they appear - the positing of real contradicts  the  Madhyamika  Prasangika-Madhyamikas  thesis  understand  of all  no  conventional real  truths  existence.  conventional  truths  The to  be  falsities that appear one way but exist another, and claim that it is not possible or logical to make the distinction into real and false.  However,  having stepped into the realm of conventional truth they deign to make a distinction, and base that distinction on the quality of the sense faculties. They do so because it is with this in mind that "the world" i.e., all those who do not know emptiness, makes its distinction between real and false.  The Prasangika-Madhyamikas refrain from asserting real  conventionalities in the way that the Svatantrika-Madhyarnikas do, and instead just "go along with" the world's presentation of such categories. dBus pa bio gsal states, The Prasangikas just go along with worldly beings who believe that which appears to the consciousness of an undamaged sense faculty to be true and that which appears to the consciousness of a damaged faculty to be untrue.  1  He quotes Candraklrti's Madhyamakavatara VI k 25, The world knows  that which is  undamaged sense faculties. of view of the world.  grasped by the  It is truth just from the point  The remainder is considered false,  just from the point of view of the world.  1. Ibid., 98b6. 2. Ibid., 99al.  six  2  59  Ultimate Truth Ultimate truth (don dam bden pa, paramarthasatya) is the truth that appears as it exists.  Only one object of knowledge  requirement, that object being emptiness.  fulfills  this  All phenomena other than  emptiness fall under the heading "conventional truths", for they are cognized by the consciousness that assents to the appearance of true existence.  Emptiness is a truth because its appearance to a wisdom  consciousness is in accord with its mode of existence.  When it is  directly cognized by a wisdom consciousness only emptiness, minus the false appearance of inherent existence, is cognized.  Paramarthasatya is  etymologized by dBus pa bio gsal as "the truth of the most supreme reality" . Emptiness, for him, is quite simply the lack of true existence 1  of an object of knowledge . It is not, as later dGe-lugs-pa scholars would have it, an existent (yod pa, sat) , for it is beyond existence and 2  non-existence.  When the final nature of an object is sought by the  analytical consciousness, absolutely nothing is found. dBus pa bio gsal quotes from an unidentified source,  When one analyses with reasoning anything appearing in such [and such] a way nothing is found. Just what is not found is ultimate [truth] and it is the original state of reality.  3  A Western scholar comments, ...things  dissolve  under ultimately  penetrative  analysis,  which is common knowledge for Centrist [Madhyamika] philosophers.  The smallest subatomic particle disappears  1. See B S G T 98b2. 2. See, for example, 'Jam yang shay ba's inclusion of emptiness under the category of existents in J. Hopkins (1983), p. 405-6, p. 213-19. 3. Bio gsal grub mtha', 97a4.  60 when one tries to pin it down ultimately.  It becomes a  mere probability, impossible to determine whether it is a wave or a particle.  Tables and chairs and houses and  human beings come apart, piece by piece, mental event by mental event, and nowhere is anything durable, analysisresistent, to be discovered.  1  Ultimate truth, in turn, does not exist ultimately. The prerequisite for ultimate existence is findability under analysis, i.e., findability by the focused consciousness that searches for the reality of its object.  When  an object is analyzed only a "lack" is found - an emptiness of true existence  in place  of  something  substantially  existent;  a similar  cognisance of lack of reality results from any analysis of emptiness itself.  This inability to withstand analysis deems that ultimate truth is  also not ultimately existent.  In fact, for the Madhyarnika, it is a grave  error to ascribe true existence to emptiness.  To believe emptiness is the  "Absolute" in terms of something that exists truly and independently is to fall from the middle way to the extreme of permanence.  T.V.R. Murti, using the word "Absolute" in a positive and non-mistaken sense, describes the ultimate as follows. The  Real is  the Absolute - at once  transcendent of  empirical determinations and immanent in phenomena as their innermost essence.  2  The Madhyamika emptiness is a self-emptiness that refers specifically to an object's lack of its own true existence. dBus pa bio gsal warns of the  dangers  involved  in  1. R.A.F. Thurman (1984), p. 164. 1. T.V.R. Murti (1955), p.  slipping into  the  view  of  self-existent  61 emptiness, first in his karika XII-13 and then in a quotation from the Ratnavali, [The Buddha] said that those who are attached to nonexistence are incurable.  Therefore, those who  desire  liberation should not take emptiness to be an absolute.  1  Moreover, stupid because he apprehends this [Dharma] wrongly he is stupid and conceited in his learning, and through abandoning [Dharma], destroys himself and goes headfirst into the Avici Hell.  2  Ultimate truth, within the Madhyarnika system, does not exist in a domain  removed  from  the  world  of  phenomena,  nor  does  it  miraculously appear as the "Real Substance" inherently existent in all phenomena.  Relieved  of  ontological  status, it is  the  mode  of  subsistence of objects of knowledge. dBus pa bio gsal quotes tantric Nagarjuna's Bodhicittavivarana on this point. Ultimate truth.  [truth]  is  not perceived  outside  conventional  Emptiness is conventional [truth] and conventional  [truth] is emptiness.  3  Emptiness has been called the highest quality or highest predicate of all objects of knowledge.  4  Madhyamikas assert that conventional and  ultimate truths are not the same or different.  Although they are  different objects of knowledge for different consciousnesses, since the basis of the division into the two truths is appearances, they cannot be 1. Bio gsal grub mtha', 104a2, k XII-13. 2. Ibid., 104a5. 3. Ibid., 97a3. 4. R.Robinson (1970), p. 51.  62 different entities.  Later exegesis within the Tibetan dGe lugs pa  tradition states that the two truths are not separate entities, but one entity within nominal difference.  1  It is asserted that were conventional  and ultimate truths to be completely different entities, the emptiness of an object could not be its final mode of existence, because the emptiness would have to exist totally divorced from the object.  If that were the  case, not even a person who had obtained omniscience could claim to have abandoned the conception of true existence since his realization of emptiness would be separate from objects and would be powerless in overcoming the innate misapprehension of the nature of phenomena.  2  Were the two truths to be the same, they could not exhibit any deviation from each other, and would have to be mirror images of each other. Since conventional truths have aspects such as shape, colour, and tangibility,  ultimate  truth  would  have  to  sometimes  be  round,  sometimes be blue and, if the conventionality were a blanket, would have to be soft to the touch. The Madhyamikas' understanding of the relationship between the two truths inspires them to call this understanding the middle path. dBus pa bio gsal claims to have designated the middle path, which all other Buddhist schools have failed to discover, the "union of appearance and emptiness"; right in the "middle" because of its avoiding the extremes of superimposition (sgro 'dogs pa, samaropa) or permanence (rtag pa, iMvatS) and denigration (skur pa 'debs pa, apavada) or annihilaton (chad pa,  uccheda).  Realization that  conventionalities  lack real  existence avoids the extremes of superimposition and realization of the fact that conventionalities are relieved of true existence, and not their mere  existence,  keeps  denigration.  •1. J . Hopkins (1983), p. 413. 2. Ibid.  the  Madhyamika  from  the  extremes  of  63  Dependent-Arising The Madhyamika's view of the true condition of existence is one that allows for only "mere appearances"; a world of form, feelings, etc., that neither exists nor does not exist.  The true status of phenomena  recognized, the Madhyamika describes  existence with a series of N  metaphors which describe it as having no more reality than a rnirage or a bit of froth.  The three realms of the Buddhist world lose their  substantiality and seem to nearly fade away, remaining, as it were, like a dream. dBus pa bio gsal quotes from Nagarjuna's  Sunyatasaptati k 66,  Composites are similar to a city of Gandharvas, a magical illusion, a rnirage, a hair net, froth, a bubble, a magical creation, a dream and a circle drawn by a firebrand.  1  Jacques May, commenting on another of Nagarjuna's  karikas,  writes,  A dream, a mirage have a very tenuous existence; but they exist somehow; they are there somehow; they are not nothing.  2  The phantom world appears, yet it does not truly exist. Phenomena, for the Madhyamika, are understood to be "dependent-arisings," due to the fact  that  they  themselves,  arise  from  causes  and conditions,  or from truly-existent others.  and not  from  "Causes and conditions"  implies causes and conditions as normally conceived, i.e., the causes and conditions of a rose include the bush, the bush seed, water, light, oxygen, etc.  Within the framework of Madhyamika thought, causes  and conditions also refer to the parts of the observed object itself, for instance, the petals, stamen, pollen of the rose and, for the Prasarigika-  1. Bio gsal grub mtha', 105bl. 2. J. May (1978), p. 236.  64 Madhyamika, to the thought that designates the existence of the object, i.e., the thought "rose" that imputes "rose" onto the arrangement of petals, pistils, etc.  1  Just the word "dependent-arising" brings, quite  automatically, even to the mind untrained in Madhyamika terminology, the idea of interdependence.  dBus pa bio gsal, writing of the middle  way as the union of emptiness and appearance, quotes Nagarjuna's verse of homage to the Buddha in his Vigrahavyavartani that praises his proclamation of the synonymity of emptiness, dependent-arising, and the middle path.  2  All objects of knowledge are empty of self-existence and therefore are not independent.  They are contingent upon something else.  If  objects were not empty and truly existed, independent of everything else, they could not be affected by causes and conditions.  This is the  insight behind the equation of emptiness with dependent-arising.  The  rose actually exists free of the solidity of true existence - it is not independently established but it has true existence mistakenly imposed onto it.  Since it is not there independendy, it can only exist dependent  on something else, i.e., it is a dependent-arising. Everything that is a 3  dependent-arising lacks true existence and is therefore empty.  All  phenomena that are empty do not exist under their own power and are therefore dependent-arisings.  As Streng says,  Considered in the context of emptiness (Sunyata), cooriginating dependently  loses its meaning as the link  between two "things"; rather it becomes the form for expressing the phenomenal "becoming" as the lack of any self-sufficient,  independent reality.  1. J. Hopkins (1983), p. 168.  2. Bio gsal grub mtha', 98a5: gang gis stong dang rten 'byung dag dbu ma 'i lam du don gcig par gsung mchog mtshungs pa med pa yi sangs rgyas de la phag 'tshal lo// 3. Lopez (1987), p. 40. 4. Frederick J. Streng (1967), p. 63.  4  65  The Madhyamikas employ various logical analyses to show that objects of knowledge lack true existence. dependent-arising.  Of these, the "king of reasons" is  It is set out in a syllogism as follows:  do not truly exist because they are dependent-arisings". powerful reason because extremes.  of its  ability to  "Phenomena It is the most  overcome  both of the  In a recent publication R. Thurman writes,  The most powerful reason advocated by the Dialecticist is known as the "reason of relativity", namely, "all thing are empty  of  intrinsic  identifiability;  relativistically originated."  because  they  are  Contemplation of the thesis  cures absolutism, and contemplation of the reason cures nihilism.  1  As noted earlier, the Yogacarins need two separate reasons to refute their extremes of permanence and annihilation:  their claim that  phenomena are not of a separate nature from consciousness holds them from the extreme of permanence, and their assertion that impermanent phenomena truly exist as the same entity of consciousness keeps them from the extreme of annihilation.  The Madhyamikas assert the single  "king of reasonings" - dependent-arising - which perfectly refutes both extremes.  The extreme of permanence is avoided because phenomena  arise dependently, not independently, and the opposite extreme, the extreme of annihilation is avoided because only true existence, not mere  existence,  of  an  object  of  knowledge  is  negated.  Mere  conventionalities, or as dBus pa bio gsal says, "mere appearances" are not non-existent. independent  They  arise in dependence,  devoid only of an  self-nature.  Thus, for the Madhyamika, dependent-arising and emptiness are synonymous.  The two are not synonymous in the same way that  "water" and " H 2 O "  1. Thurman (1984), p. 163.  are synonymous, but are equivalent because an  66 object's being empty of true existence is the meaning of its being a dependent-arising.  This compatibility of emptiness and dependent-  1  arising allows the Madhyamika to set forth cyclic existence and nirvana in a way that accounts for the variety and relationships of things while, at the same time, denying that anything - from persons to even the most minute particles of the universe - truly exists.  Divisions of the Madhyamikas: Sautrantika-Madhymika,  Yogacara-Madhyamika,  and 'Jig sden grags sde phyod pa dbu ma pa The Madhyamika school stood united behind the assertion that all objects of knowledge are empty, and it was this assertion that made it separate from all other Buddhist schools.  Internally, however, from  about the sixth century C.E., there arose disagreements and debates about the proper interpretation of Nagarjuna's profound karikas. As noted earlier, the later Tibetan authors gave the contending groups different names in order to delineate, understand and explain the differences between them.  Beginning with Ye shes sde's 77a ba khyad  par, composed during the sna dar, or early dissemination of Buddhism into Tibet (ninth century C.E.), the schools were classified in terms of their position on the status of conventionalities.  The Madhyamikas who  said conventionalities existed as the nature of the mind were seen to reflect the views of the Yogacara school of Buddhist thought, and were thus  designated  Yogacara-Madhyamikas.  The Madhyamikas  who  asserted that conventionalities were not the nature of the mind but existed separate from the mind were called Sautrantika-Madhyamikas, since their views concurred with those of the Sautrantika school. The now famous terms, Svatantrika and Prasangika, gained currency during the phyi dar, the second dissemination of Buddhism to Tibet  1. Hopkins (1983), p.171.  67 (eleventh century).  These terms divided the Madhyamika according to  the modes used to generate an understanding of emptiness.  Thus  Tibetan texts written early in the second dissemination that set forth the presentations  of  the  Madhyamika  schools  often  divide  their  presentations into two parts, with one part containing the divisions of the Madhyamika with regard to their views on conventionalities, and the other containing divisions "from the ultimate point of view"  1  dBus pa bio gsal follows the pattern established by his predecessors. He commences his presentation of the schools with a karika detailing the Madhyamika views of conventional truth. The  Yogacara-Madhyamikas,  etc.,  [i.e.,  and  the  Sautrantika-Madhyamikas and 'Jig rten grags sde spyod pa'i  dbu  ma  pas]  are  listed  as  those  who  assert  conventional [truth] to be the nature of the mind, the external object, and as things appear, respectively.  2  dBus pa bio gsal utilizes Ye shes sde's terminology and, continuing in the tradition established by Grags pa rgyal mtshan, inserts a third categorization of Madhyamika, the 'Jig rten grags sde spyod pa'i dbu ma pas - the "Madhyamikas who conform to that which is reknowned in the world".  He lists Santaraksita and Haribhadra as Yogacara-  Madhyamikas and, to demonstrate their views on the equivalency of objects  and  consciousness,  Madhyamakalankara.  quotes  from  Santaraksita's  His inclusion of a verse from Nagarjuna's  Yuktisastika shows his acquaintance with Yogacara "source" material. Bhavaviveka is identified by dBus pa bio gsal, in agreement with earlier Tibetans, as a Sautrantika-Madhyarnika, since "his conventional [truth] is concordant with [the theories  of]  the Sautrantikas".  3  The third  division of Madhyamikas is into the one identified by later Tibetans 1. See, for example, the divisions the Madhyamika school according to Grags pa rgyal mtshan in Mimaki (1982), p.32. 2. Bio gsal grub mtha', 100a2, k XII-7. 3. Bio gsal grub mtha', lOObl.  68 (for example,  Bu ston  Rin chen grub) as  synonymous  with the  Prasangika: the 'Jig rten grags sde spyod pa. Included in this division, according to dBus pa bio gsal, are Jnanagarbha and Candrakirti.  With  regard to their views on conventionalities, it can be said that the 'Jig rten grags sde spyod pas refrained from making any independent statement as to the nature of conventionalities, and merely deferred to the world for such statements.  Whatever was accepted in worldly  consensus was suitable to be used as the 'Jig rten grags sde spyod pa's assertion of conventionalities.  In the years following dBus pa bio gsal,  when the Tibetan authors fell away from the habit of delineating of schools according to their views regarding both conventional and ultimate truth, and began an inclusive melding of the sub-schools, the 'Jig rten grags sde spyod pa school was naturally dissolved into the Prasangika.  As noted in the earlier discussion of the Prasangika's  rejection of the division of conventional truth into real and false, and in material that will follow, this same view toward conventional truth is upheld;  the  Prasarigikas merely  accepted by worldly beings. gsal's  that  requires  "go  along  with" the  conventions  But, within a framework like dBus pa bio  groupings  based  on  assertions  from  both  conventional and ultimate points of view, the inclusion of the 'Jig rten grags sde spyod pa view is needed, for it allows for a point of view on conventionalities that is compatible with Prasangika assertions. dBus pa bio gsal finds a basis for the 'Jig rten grags sde spyod pa's assertion of conventionalities  in accord with those of the world in  Candralorti's Madhyamakavatara VI k 3, Since, if one analyzes these things, instead of [arriving at] anything which has a real nature, one ends up with unfindability. analyzable.  1  l . ibid., 100M.  Therefore, worldly transactional truth is not  69 Analysis of conventionalities leads to emptiness and emptiness takes one  beyond conventionalities,  ultimate itself.  Since this cognizance  perception of truly-existent "harms"  the  which obscure  i.e.,  ultimate,  to  the  of emptiness contradicts the  conventionalities,  conventionalities,  the  it  analysis,  takes  away  so to speak, their "truth".  Therefore the 'Jig rten grags sde spyod pas leave conventional truth in the hands of a world that does not know emptiness.  They merely "go  along with" the decisions and assertions of the world.  Svatantrika and Prasangika The second grouping of subschools of the Madhyamika finds its basis in methods used to produce an understanding of emptiness in a person's mind.  The Svatantrika school, starting with Bhavaviveka, holds that a  syllogism must be employed to bring about a correct understanding of emptiness.  D.S. Ruegg writes,  Bhavaviveka took  up a position radically opposed  to  Buddhapalita's on the matter of the logical establishment of the Madhyamika's philosophical position in general and of the negative statements in particular.  In his view  the necessary co-ordination with scripture (agama) of an adequate  logical method  of reasoning  (yukti) requires  more than prasanga arguments because, to establish the Madhyamika's position, there is needed in addition an independent  (svatantra)  inference  (anumana), which can  also be embodied in a proper "syllogism" (prayogavakya). And it is from this characteristic use of a svatantranumana that  Bhavaviveka's  Svatantrika.  1  1. Ruegg (1981), p. 61.  school  has  received  its  name  of  70 The  Prasangika  school,  relying  on  Chandrakrrti's  rebuttal  of  Bhavaviveka's procedure, points out that syllogisms are not necessary for  realizing emptiness,  which can be  understood with prasanga  statements only. Thus, the two schools, when classified by the Tibetans, received their names from the two main types of logical statements that were used to establish an understanding of emptiness. In the BSGT, dBus pa bio gsal sets forth his view of the main tenets held by the two schools, and we can compare the Svatantrika and Prasangika views by examining his statements.  They are as follows:  i  They [Svatantrika] assert 1) that all objects and minds are not necessarily false and mistaken; 2) that there exists a view which is admitted on the conventional level; 3) that on the conventional level there is truth and that truth can be divided into real and false; 4) that the absence of selfnature of an illusion is the ultimate and 5) that the Buddhas have a self-perpetuating wisdom consciousness.  1  Set beside his list of Prasangika assertions, the differences between the two are clear. Thus, [the Prasangikas] assert:  1) that all objects that  appear are false and all consciousnesses are mistaken; 2) that no view exists [even on the conventional level]; 3) that on the conventional level there is no truth nor [division of it] into real and false; 4) that ultimate truth is free from elaborations; 5) that Buddhas do not have selfperpetuating  consciousnesses.  1. Bio gsal grub mtha' 101a5. 2. Ibid., 101b3.  2  \  71 These would appear to be the five main differences that can be contrasted.  However, immediately following this presentation of the  assertions of the Svatantrikas and Prasafigikas is a clarifies the difference. The  karika that further  It reads:  real difference of these modes [of Svatantrika and  Prasangika] is the assertion or non-assertion of reasons from real things and of valid cognizers [that come about from  them].  [The  differences]  that  are  otherwise  expressed by others are cleared away.  1  With this statement it would appear that dBus pa bio gsal concurs with the views of his Madhyamika predecessors difference between the two schools.  in regard to the main  We can note that in a remark  2  following the karika, he refers to the previous distinctions listed, i.e., that the Svatantrikas say that the Buddhas have a self-perpetuating wisdom consciousness while the Prasafigikas do not, etc., as being "slight" (cung zad). The essential difference relates to whether or not a reason that brings about an inference is asserted.  Such a reason is one  of the main elements of a syllogism. Let us examine the basic characteristics of and differences between syllogisms  and consequences according to Tibetan Buddhist writers.  When emptiness is cognized, it is initially cognized inferentially, that is, in reliance on a correct sign.  This inferential cognition comes about  first because emptiness is not an object of knowledge that can be known by  ordinary direct perception (mngon sum,  pratyaksa); emptiness,  unlike tables and houses that are known by direct perception and that are classified as manifest phenomena (mngon gyur, abhimukhi), is, like previous lifetimes  and subtle impermanence, a hidden phenomenon  (lkog gyur, paroksa). Once emptiness is realized inferentially, i.e., by means  of an inferential cognizer (rjes dpag tshad ma, anumana-  1. Bio gsal grub mtha' k XII-11 101b4. 2. B S G T 102a3,  72 pramana) repeated meditation on and the resultant familiarity with this conceptual understanding of emptiness has the power to bring about a transformation of the understanding of emptiness - from one based on thought to one that is based on a direct cognition of emptiness. The cognition which develops and which facilitates direct perception of emptiness is known as a yogic direct perceiver (mal 'byor mngon sum, yogipratyaksa). A correct sign (rtags yang dag, samyag linga) or a correct reason (rgyu mtshan yang dag, samyag nimitta), serves to produce inferential cognition.  It brings  unknown. reasons  about  knowledge  of something  previously  Terms, within the Buddhist schools, are reasons, and these  aid a person in discovering information that is otherwise  hidden. Any thing can stand as a correct reason for something else. The usual Buddhist example of a good sign, or reason, is the smoke on the distant hill.  Although the person perceiving the smoke cannot see  anything but smoke, in reliance on the sign "smoke" (due to its cause and effect relationship with fire) the presence of fire on the hill can be inferred. correct  Thus the knowledge of fire can be achieved in reliance on a sign.  1  In the same way, knowledge of emptiness can be  achieved through reliance on a "good sign". A correct reason, or sign, is one of the parts of a proper syllogism and, depending on it, an inferential cognition can be produced. traditional Nyaya syllogism is made up of five members.  The  Using the  traditional example that uses "smoke" as the sign for "fire", the Nyaya syllogism is as follows: Theory (pratijfia): The mountain possesses fire Cause (hetu): Because there is smoke Example (udaharana): Wherever there's smoke, there's fire, like in a kitchen  1. Correct  signs are of three types:  samakkaryahetu),  correct  nature  correct  signs  effect signs ('bras  rtags yang  dag,  (rang bzhin gyi rtags yang dag,  samyaksvabhavahetu), and correct signs of non-observation (ma migs pa'i rtags yang dag, samyakanapalabdhihetu). The example above is a correct effect sign.  73 Application (upanaya): The mountain is similar [i.e., possesses smoke] Conclusion (nigamama): therefore it is similar [i.e., possesses fire] Buddhist syllogisms are the same as Nyaya syllogisms, but are set forth in three parts instead of five.  Besides the sign, the other parts of a  syllogism are the probandum (bsgrub by a, sadhya), divided into the predicate of the probandum (bsgrub bya'i chos, sadhyadharma) and the subject (chos can, dharrnih). An example of a Buddhist syllogism would be: The  subject,  sound,  is  an  impermanent  phenomenon  because of being a product.  1  The subject of the syllogism, the locus of inference, is "sound".  The  predicate of the probandum is "impermanent phenomenon", and the sign, or reason is "being a product". Given that this example provides the basic framework for a syllogism, many more syllogisms could be constructed.  Syllogisms, however, may be correct or incorrect, and  since inferential cognitions depend on correct syllogisms, it is necessary to fulfill the requirements of correct syllogisms.  These requirements  can be met by the sign; a correct sign creating a correct syllogism must fulfill the three relationships that are designated the "three modes" (tshul gsum, tri rupa). The three modes are 1) being the property of the subject (phyogs chos, paksadharma) 2) the positive inclusion (rjes khyab, anvayavyapti) and 3) the negative exclusion (ldog  khyab,  vyatirekavyapti). The first  means that the sign must be an attribute or quality of the subject. Returning to the example, the sign "producthood" must be an attribute of sound. Since in the Buddhist world sound is a product, the sign is the first mode. For it to be the second mode, i.e., the positive inclusion, the  1. A syllogism of this sort would be directed to certain former Vaislsikas who, according to Buddhists, would be questioning an inner contradiction in the Vaisasika world view which holds sound is a product and is also, as an attribute (guna) of space (akasa), permanent.  74 sign must be pervaded by the predicate of the probandum. The sign in the example is also the second mode, for whatever is a product is pervaded  by being  impermanent.  The third mode,  the negative  exclusion, requires that the negative of the predicate be pervaded by the negative of the reason - more simply, whatever is not the predicate must necessarily not be the reason.  In the syllogism, "The subject, sound, is  an impermanent phenomenon because of being a product", whatever is not the predicate "impermanent" must necessarily not be the sign. Anything that is not impermanent, i.e., permanent or  non-existent,  cannot be a product. As can be seen, the reason in the example fulfills each of the three prerequisites (tshuT), and is thus a correct reason within a correct syllogism. Svatantrika school. accepted  This is the sort of syllogism used by the  According to them, the three modes must be  by themselves and by the opponent  towards whom the  syllogism is directed. Consequences (prasanga), on the other hand, are not of the three modes.  They are constructed so that a consciousness that infers the  implied three modes could be generated, but, in and of themselves, they only restate the opponent's own assertions. manner  that  a  consequence  of  the  These are stated in such a  opponent's  assertion  contradicts other of his assertions become self-evident, unable to respond.  which  leaving him  When the opponent reassesses his views and  resolves the contradictions in his views, his is able to discover and state the syllogism implied by the consequence.  For example, the opponent  who holds that sound is permanent and a product, and who also believes that whatever is a product is impermanent would be met with the consequence,  "It follows  that the subject, sound, is not a product  because of being a permanent phenomenon".  Unlike the Buddhist, he  would accept all three modes.  He would have no response, and would  be  thesis,  left  with  the  unwanted  "sound is  not produced" that  contradicts his own assertion that sound is a product.  Thus the  consequence put out by the Buddhist serves to generate the implied and, according to the Buddhist, correct syllogism: impermanent because of being a product".  "The subject, sound, is  75  This difference that is based on the assertion or non-assertion of syllogisms is posited as the major difference between the Svatantrika school and Prasangika school. Though both schools use the two methods, only the Svatantrikas insist that syllogisms be explicitly stated. The Prasangikas respond that consequences alone are sufficient to break adherence to mistaken views, for once the opponent recognizes his error, it is redundant and of no purpose to explicitly set forth the syllogism.  76  BLO  GSAL  GRUB  MTHA'  (Bio gsal grub mtha' XII: Madhyarnika) [1. General explanation of the two truths] [1.1.  Universality of the two truths] [96a5]  I will now explain the tenets of the Madhyamika  Mahay anists: [the Buddha] taught all phenomena in terms of two truths (bden pa nyis, satya-dvaya). The perfect Buddha (rdzogs pa'i sangs rgyas, sambuddha) himself taught objects of knowledge in just two truths. Therefore, objects of knowledge, which are also the five categories (gzhi lnga, paffca-vastu) are not other than the two truths. [96a6]  (k As  the  Bhagavan  XII-1)  [=Buddha]  said  in  the  Pitaputrasamagama-sutra : 1  Thus  the  conventional  Tathagata [truth]  understood  the  two  truths,  (kun rdzob, samvrti) and ultimate  [truth] (don dam pa, paramartha). Objects of knowledge also are exhausted in these, conventional and ultimate [truths].  Moreover, the Bhagavan thoroughly saw, knew  and realized these in so far as they are empty. Because of  1. Pitaputrasamagama-sutra (D. nga 60M-5, P. zhi 70a4-6). The Sanskrit is found in Bodmcaryavatarapattjika (117.16-18 (=SS 256.4-6): etavac caiva' jffeyam-yad uta samvruh paramarthas' ca / tac ca bhagavata Sunyatah sudrstam suvidsrtam susaksatkrtam / te sarvajfia ity ucyate / tatra samvrtir loka-pracaratas tathagatena drsta / yah pun paramarthah so 'nabhUapyah...'caitat SS sa SS 2  The first sentence of PPS as cited in our text does not have corresponding Sanskrit in the Sanskrit BCAP, and is only found in the Tibetan version of the BCAP.  77 this, he is called "Omniscient" (thams  cad  mkhyen  pa,  Of these [two truths] the Tathagata saw that  sarvajna).  conventional  [truth] is  spyod  pa,  (brjod  du med  activity  ('jig rten  gyi  Ultimate [truth] is inexpressible  loka-pracara). pa,  worldly  anabhilapya).  Also [in the same sutra] : 1  These two truths of the Knower of the World ('jig mkhas  pa,  loka-vidu) , 2  rten  you have seen them yourself,  without having heard [of them] from others. conventional [truth] and ultimate [truth].  These are  A third truth  does not exist. [The Buddha] also said in the Lankavatara-sutra  (X k 299 = II k 185)3:  There is conventional [truth] and ultimate [truth]. A third [truth] arisen from cause does not exist.  Conventional  [truth] is constructed by conceptuality (rtog pas brtags kalpita).  pa,  When this is cut off, it is [ultimate truth which is]  the domain of experience of the Superiors. [1.2. The five categories and the two truths]  1. PPS (D. nga 61M-5, P zhi- 71a7). The Sanskrit is found in BCAP 175.8-11:  satya ime duvi loka-vidunam dista svayam aSrunitva paresam I samvrti ya ca tatha paramartho satyu na sidhyati kim ca trtlyu II 2. 'jig rten mkhas pa, Buddhist hybrid Sanskrit loka-vidu. Sanskrit would be loka-vidvas- (cf.  The exact equivalent in normal  Buddhist Hybrid Sanskrit Grammar, 16.49); but the  usual epithet is loka-vid , Tib. 'jig rten mkhyen pa (cf. Mvyut 8) 3. Lankavatara-sutra X  k 299 (= II k 185) (cf. tr. Izumi  (1927) p. 75, p. 174, Suzuki  (1932) p. 112-113, Yasui (1976) p. 118, p. 270; II k 187 following them):  samvrtih paramarthaS ca trtiyam nasti hetukam / kalpitam samvrtir hy ukta tac-chedad arya-gocarah //  78 [1.2.1. Real conventional truth: form, mind, mental factors and nonassociated compositional factors] [96b4]  Therefore, the five categories (gzhi nga, pafica-vastu) of  knowables do not exist apart from the two truths either. Thus, With regard to the five categories, the first four are, for the most part, real conventional [truths] (yang dag kun rdzob, tathya-samvrti ). 1  (k XH-2ab) [96b4]  Form (gzugs, rupa), mind (sems, citta), mental factors  (sems las byung ba, caitta), and composite factors which are not occurrences of those [three preceding categories] are capable of causal 2  efficiency  (don  byed par  conventional plane;  nus  pas,  arthakriyasamartha) on the  and, on account of that, are real conventional  [truths] (yang dag pa'i kun rdzob, tathya -samvrti). In reality (de kho nar,  tattvatas)  they  exist without entity  (ngo bo nyid med pa,  nihsvabhava). The term "for the most part" means that appearances such as dream objects, hairs appearing in the sky, a double moon, and meaning-generalities  (don spyi, samanyartha) are false conventional  [truths] (log pa'i kun rdzob, mithya-samvrti). [96b6]  For some Yogacara-Madhyamika [logicians] even these 3  [appearances] are real conventional [truths].  1. The Sanskrit term is found in Madhyamakahrdaya  III k 7 and 12; and also in  Abhisamaymamkaraloka 20..22, 79.17, 112.2, 169.19, 548.19, 594.22, 641.19-20, 914.16 (cf. Amano (1965) p. 623 n. 5). 2. Cf. B S G T X 79a3 (Mimaki (1979a) p. 202 and Mimaki (1980) p. 155). 3. A s will be seen later (BSGT 100a3-4), this refers to Santaraksita and Haribhadra.  79  [1.2.2.  Unconditioned phenomena and suchness] The  three  unconditioned  phenomena  ('dus ma byas,  asamskrta) are just mere names (ming  tsam, samjfia-  matra). Suchness (de zhin nyid, tathata) is ultimate 1  [truth].  (k XII-2cd)  [  Unconditioned phenomena]  [97al]  The three [unconditioned phenomena], space (nam  mkha', akas~a) and the two cessations ('gog pa, nirodha) are conceptual constructions expressing, relative to their object of negation (dgag bya, pratisedhya) , that it does not exist. 2  They have no reality and are  exhausted in their mere names (ming tsam nyid, samjna-matra). As it is said in the Mulamadhyamaka-karika (V k 7) : 3  Therefore, space is not a thing (dngos po, bhava) nor is it a non-thing (dngos med, abhava). It is not characterizable (mtshan gzhi, laksya) and does not characterize (mtshan  1. Suchness as the fourth unconditioned phenomenon is asserted especially by the Madhyadesa VaibhaSikas (Yul dbus kyi Bye brag tu smra pa), by the Vijflanavadiri authors such  as Asanga,  Vasubandhu and Sthiramati,  and, among  the Madhyamikas, by  Bhavaviveka. See May (1959) n. 398 and Mimaki (1976) n. 420. 2. For example, when one says that space is only the non-existence of a tangible obstructor, it is the tangible obstructor that is the object of negation (dgag bya, pratisedhya). 3. Mulamadhyamakakarika V k 7 (cf. tr. Schayer (1931) p. 9, Inada (1970) p. 58, Sprung (1979) p. 107:  tasman na bhavo nabhavo na laksyam napi laksanam / akaSam akas'a-sama dhatavah paffca ye 'pare//  80  The five other elements (khams, dhatu) are similar to [the element] space [ultimately]. nyid, laksana).  1  [ Suchness] [97a2] Suchness (de bzhin nyid, tathata) is ultimate truth. It is also stated in the PrajnaparamitS[-sutra] 2  The suchness of conventional [truth] is ultimate [truth].  1. At first glance dBus pa bio gsal's citation of the M M V k 7 here seems a bit strange. This verse of the M M relates to the element space, one of the six elements (dhatu:  earth,  water, fire, air, space and consciousness) and not to space as one of the unconditioned phenomena (asamskrta). In the Abhidharmakos'abhasya, Vasubandhu asks himself if the first is the same as the second, and responds in the negative:  the first forms part of the  category of visibles (rupa-dhatu ) . See AKBh 18. 10-25 (ad AK I k 28); LVP i pp. 49-51. Thus in Abhidharmic philosophy these two "space"s are clearly distinguished.  But in the  philosophy of the Sautrantika and above, unconditioned space is just a name.  Cf. AKBh  92.4-9 (ad A K II k 55d);  BSGT 79b2-4 (text:  Mimaki (1979a) pp. 202-3, tr:  Mimaki  (1980) p. 156 and n. 36).  dBus pa bio gsal is most likely using the quotation within the  context of Madhyamika philosophy; he is stating that the final nature of unconditioned phenomena is like that of conditioned phenomena, i.e., without entity. See also May (1959) n.398. 2. Cf. for ex., Conze E., The Gilgit manuscript of the As&dasasahasrikaprajRaparamita , Chapters 55 to 70 corresponding to the 5th Abhisamaya, Text and English Translation, Serie Orientate Roma paramarthasya  XXVI, Roma, 1962, p. 166. 4-5:  tathata.  (cf. tr. Conze (1975) p. 529).  lokasamvrtes  tathata  saiva  81 The master [Nagarjuna] also taught in the Bodhicittavivarana : 1  Ultimate [truth].  [truth]  2  is  not perceived  outside  conventional  Emptiness (stong nyid, Sunyata) is conventional  [truth] and conventional [truth] is emptiness.  And also:  When  one investigates with reasoning  (rigs pa, yukti)  anything appearing in such [and such] a way, nothing is found.  Just what is not found is ultimate [truth] and it is  the original state of reality (ye nas gnas pa) . 3  From the Satyadvayavibhanga [-karika (k 17ab) by Jnanagarbha] : 4  [The  Buddha]  asserted  the  suchness  of  conventional  [truth] as ultimate [truth].  1. Like all the Tibetan authors, our author dBus pa bio gsal does not distinguish tantric Nagarjuna (modem scholarship dates him 7-8th C.) from Madhyamika Nagarjuna (modem scholarship dates him 2nd-3rd C). For example, ICan skya II Rol pa'i rdo rje enumerates the Bodhicittavivarana among Nagarjuna's other Madhyamika treatises, cf. ICang skya grub mtha' 282, 18. The question of two or three Nagarjunas does not, however, pose a problem for the Tibetans, since for them Nagarjuna is one and the same person who lived for six hundred years. See, for example, ICang skya Grub mtha' 281.5. For the third Nagarjuna, author of MahaprajnaparamitopadeSa , see Lamotte (1970), Traits L HI, p. 1375. 2. Bodhicittavivarana a (P. [61] (2665) gi 54b8); b (P. [103 (5470) gi 224b2). The form of this verse in the Tibetan canon is different from that of this text, though the meaning is the same: kun rdzob las ni logs shig na'/ de nyid nye bar mi dmigs so / 1  kun rdzob stong pa nyid du bshad / stong nyid kho na kun rdzob yin'/l 3  'tha dad par -a  2  dmigs pa ma yin te - a  3  pa - a  * ste - b.  3. Original source not identified. This verse is not in the BCV. 4. Satyadvayavibhangakarika k 17ab (C. 2b2, D. 2b2), also cited in Satyadvayavibhangavrtti (C. 9b4, D. 9b4). Cf. SatyadvayavibhaAgapaSfjika D. 36M-5, N. 26M-2, P. 28b6-7).  (C. 36M-5,  82  [1.3 The two truths and their connection with definitive meaning and meaning requiring interpretation] In the way that a doctor takes care of a patient, so the Omniscient One taught, by means of the two truths, the scriptures (gsung rab, pravacana) that show definite meaning (nges don, nitartha) and meaning requiring interpretation (drang don, neyartha). (k xn-3) [97a5] For example, a wise doctor shows the ways to nurse a patient. In the same manner, the Bhagavan, having seen the character (khams, dhatu) mental dispositions (bsam pa, Maya) and emotional afflictions (nyon mongs, kleia) of trainees (gdul bya, vineya-jana) * taught, as the aforementioned two truths the various sutras (mdo sde) with definite meaning and meaning requiring interpretation. As it is said in the 1  Aksayamati-nirdes'a : 2  A sutra taught for the establishment of conventional [truth] requires interpretation. A sutra taught for the establishment of ultimate truth is definite.  1. Or vaineya-jana, cf. B C A P 175.13. 2. AksayamatinirdeSa  (P. [34] (842) bu 155b7): mdo sde gang dag kun rdzob sgrub bar  bstan pa de dag ni drang ba'i don ces bya 'o// mdo sde gang dag don dam pa sgrub par bstan pa de dag ni nges pa 'i don ces bya 'o// This passage is cited in Indian and Tibetan works such as Madhyamakaloka Lam rim chen mo  kha 4a8-bl, C K G T 313. 18-20.  [P] 161b6-8,  The A M N title that dBus pa bio gsal  presents here, 'Bio gros mi zad pa bstan pas zhus pa' is unusual, and he later, in B S G T 109b6, gives the one that is normally used, 'Bio gros mi zad pas bstan pa'i mdo'. It is known, furthermore, that the name of this sutra was transmitted in at least two different ways: cf. for example, Mvyut n. 1344 (Bio gros mi zad pas bstan pa / Aksayamati-nirdes'a), n. 1400 (Bio gros mi zad pas zhus pa (-pariprccha ). Is the present case a combination of the two titles, perhaps due to some confusion on the part of dBus pa bio gsal?  83 Also in the Samadhiraja-sutra h Whatever the Sugata taught of emptiness one knows as a definitive sort of sutra. Where person (gang zag, pudgala) sentient being  ( sems can, sattva) and man (skyes bu,  purusa) are taught, one knows all these phenomena to require  interpretation.  [97b2] The master [Nagarjuna] also stated, in regard to this same thing, in the Mulamadhyamaka-karika (XXIV k 8) : 2  The Buddhas taught Dharma in dependence on the two truths:  worldly conventional truth ('jig rten kun rdzob  bden pa, loka-samvrti-satya) and ultimate truth. [1.4] Intermediary verse (bar skabs kyi tshigs su bead pa, antarasloka): Similarly, words [directly presenting Buddha's] intention, (dgongs, abhipraya), masking his thought (Idem dgongs), literal (sgra ji bzhin) and non-literal (sgra ji bzhin ma yin); all he said out of mercy is not other than the two truths . 3  (AS XII-1) [2. Detailed explanation of the two truths]  1. SR VII k 5 (cf. tr. Tamura (1975) 1 p. 137):  nltartha-sutranta-vis'esa' janali yathopadista sugatena Sunyatal yasmin punah pudgala sattva puruso neyarthatam janati sarvadharman // 2  1  -sutranta viSesa Gilgit = Dutt (1941)  1  yasmin Gilgit  (Tib.) P. [31] (795) thu 22a5-6. 2. M M XXJV k 8 (cf. tr. May (1959) p. 225, Inada (1970) p. 146, Sprung (1979) p. 230):  dve satye samupaSritya buddhanam dharma-desana I loka-samvrti-satyam ca satyam ca paramarthatah II 3. For an excellent discussion of dgongs pa/dgongs gzhi, see D.S. Ruegg (1985), p. 309-325.  84  [97b3] Explanation of the two truths : The two truths are:  the mistaken objects and nothing  whatsoever presenting itself to the non-mistaken mind. (k XII-4a-c) [97b4]  Samsara and nirvana, appearing in all their diversity to worldly  consciousness that  (tha snyad pa'i bio), are conventional truth.  When just  is analyzed and nothing whatsoever presents itself to a non-  mistaken mind, there is the ultimate truth.  It is also stated in the  Satyadvayavibhanga[-karika (k 3) of Jnanagarbha] : 1  [In various Sutras] the Muni (thub pa) taught the two truths, conventional [truth] and ultimate [truth].  Only that  which corresponds to appearances is conventional [truth]. That which is different [from appearances] is the other (gcig shos, itara) [truth]. The master Candrakirti also says in the Madhyamakavatara (VI k 23) : 2  All things have a dual identity which is established by 3  correct and false sight.  The domain of those who see  correctiy is called the ultimate  4  and [the domain] of those  who see falsely is called conventional truth.  1. SDVK k 3 (C. lb3-4, D.lb3-4), cited also in SDVV (C. 4a2, D. 4a2). 2. Madhyamakavatara VI k 23 (cf tr. LVP (1910) p. 299, Ogawa (1976) p. 80), of which the Sanskrit is found in BCAP 174.26-29: samyag-mrsa-dariana-labdha-bhavam rupa-dvayam bibhrati sarva-bhavah / samyag-drSam yo visayah sa tattvam mrSa-drSam samvrti-satyam uktam II 3. The MAv version (ngo bo gnyis: dual identity) is closer to the Sanskrit than the BSGT version (ngo bo nyid). 4. The MAv version (de nyid: reality) is closer to the Sanskrit than that of the BSGT (don dam).  85  [2.1 Union of the two truths] [The two truths] are distinguished only from [the point of  (snang ba, abhasa). The union (zung du chud pa, yuganaddha) [of the two truths] is the Middle Path (dbu ma'i lam, madhyama pratipad). view] of appearances  (k XII-4c -d) 2  [97b6] appearances.  The basis for the division into these [two] is mere Since the nature of an appearance is ultimate [truth], this  does not contradict [the ultimate]. The divisions are the two truths. The definiteness of the number is contingent on the two consciousnesses. In regard  to  the  significance  Samdhinirmocanasutra  1  Composite elements  of  the  division  since  the  says:  ('du byed khams, samskara dhatu ) 2  and [the] ultimate are [characterized by being] free from being the same or different.  Those who imagine that they  are the same or different have gone wrong and enter into error. Therefore, different.  [conventional and ultimate truths] are not the same or They are, in fact, inexpressible as just that or anything else,  for as it says in the Maitreyapariprccha  :  3  1. Samdhinirmocanasutra III: 7. 2. See May (1959) n. 108 and n. 255. 3. Maitreya-pariprccha (ed. Conze and Iida (1968) p. 236.43-237.2, cf. tr. Conze (1975) p. 647, Hakamaya (1975) p. 200): ...evam veditavyam, na tasmat samskara-nimittad vastunonya nirabhilapya dhatur, napi tasmad ananya mirabhilapya dhatur...iti... ( Tib.) 1. (chap. 83 o the AstadaSasahasrika-prajflaparamita) P. [20] (732) phi 173M-3: ...'du byed kyi mtshan ma di las brjod du med pa'i dbyings gzhan pa yang ma yin la I ...gzhan ma yin pa yang ma y parrigpar bya'o II; 11. (chapter 72 of the PaffcavmSaUsahasrika-prajflaparamita) P. [19] (731) di 248a5-6 = P. [90] (5188) ca 373a7-bl.  86  One should know that the sphere of the inexpressible is not different from the signs of the composite ('du byed kyi mtshan ma, samskara-nimitta) and that it is also not not different. [98a3] Therefore, the union of appearance and emptiness is the Middle Path (dbu ma'i lam, madhyama pratipad), free from the extremes of superimposition (sgro 'dogs pa, samaropa), denigration (skur pa 'debs pa, apavada), permanence (rtag pa, §a§vata) and annihilation (chad pa, uccheda). For as the master [Nagarjuna] says: That which is a dependent-arising we call emptiness. Emptiness is a metaphorical designation. Just that is the middle path . 1  The Maitreya-pariprccha corresponds to chapter 83 of the AstadaSa-sahasrikaprajffaparamita (abr. ADSPP) and to chapter 72 of the Paffcavimsatisahasrikaprajttapararnita (abr. PVSPP).  But these chapters are respectively titled Byang chub sems  dpa 'i slab pa rab tu dbye ba and Byang chub sems dpa 'i bslab pa la rab tu phye ba in the Tibetan Canon; the tide given by our author, and analogous udes, are found withih the works of Tibetan authors (cf. for ex., LS (Drang nges legs bshad snying po) 215.13-14: Byams zhus kyi le'w, CKGT 341.9: Sher phyin Byams zhus kyi le'u). Iida (1966) identified this chapter in the Sanskrit manuscript of the PVSPP.  There is an edition of the Sanskrit  text (Conze and Iida (1968)), an English translation (Conze (1975) p. 644-652) and a Japanese translation (Hakamaya (1975)).  See also - Obermiller (1932) p. 97-98,  Yamaguchi S. (1964a) p. 150-153), -Lamotte (1935) p. 14-16, -Lamotte (1938) tome II, p. 91, -Hakamaya (1975) p. 210-208. 1. M M XXIV k 18 (cf. tr. May (1959) p. 237, Inada (1970) p. 148, Sprung (1979) p. 238, Nagao (1979) p. 31): yah pratllyasamutpadah Sunyatam tarn pracaksmahel sa prajfTaptir upadaya pratipat saiva madhyama II This verse is cited among others in the MadhyamakalaAkaravrtti (D. 71al-2, P. 68b5: for pada c, D. rgyur byas, P. bsgyur byas) by Santaraksita. On examining the variants, it can be seen that the verse cited by our author is more like the MAV verse than that of the M M . It is possible that dBus pa bio gsal cited it from the MAV: a common practice amongst the Tibetan authors.  For analogous stances, see May (1959) p. 237 n. 840.  Wayman in JAOS, vol. 89, p. 145.  See also Alex  87  Homage to the unparalled Buddha who excellently proclaimed that emptiness, dependent-arising and the middle path are synonymous . 1  Similarly, from the Yuganaddhakrama [the fifth and last tan trie Nagarjuna's Pancakrama] :  krama  of  2  Having understood the two [truths], conventional and ultimate, separately from their difference (so so'i char, prthag vibhagatah) union (zung du 'jug pa, yuganaddha) is said to be where they are joined together. In the said :  Satyadvayavibhanga [-karika  (k 36) by Jnanagarbha it is also  3  Since it arises from causes therefore why should it become annihilated? If the [cause] disappears the [effect] disappears: so explain how there is permanence.  [2.2 Etymology of the terms samvrti and  1. Vigrahavyavartani  paramartha]  151.14-15 (verse of homage at the end of the treatise: cf. tr.  Yamaguchi S. (1929) pp. 61-62 (k 72 following him), Bhattachacharya K. (1978) p. 48, Kajiyama (1974) p. 184):  yah funyatam pratltyasamutpadarn madhyamam pratipadam cal ekartham nijagada pranamami tarn apratima-buddham II (Tib.) ed. Tucci (1929) p.75. Cf. Drang nges legs bshad snying po 101.5-7. 2. Paffcakrama-Yuganaddhakrama k 13 (LVP (1896) p.46, cf. tr. Sakai (1956) p. 228):  samvrtirn paramartham ca prthag jffatva vibhagatah/ samnulanam bhaved yatra yuganaddham tad ucyate II (Tib.) D. 56al-2, P. 62b6-7. 3. SVDK k 36 (C. 3a4-5, D. 3a4), cited also in S D W (C. 13b7, D. 13b7).  88 [98b 1] The etymology (nges pa'i tshig, nirukti) of samvrtti (sic) satya is: a truth (satya = bden pa) which completely (sam = yang dag) conceals (vrti= sgrib byed), for it says in the Satyadvayavibhanga [-karika (k 15ab) by Jnanagarbha] . 1  2  3  That by which or in which reality is concealed is asserted as conventional [truth]. Also from the Madhyamakavatara [(VI k 28) by Candrakirti] : 4  The error [in the mind] (gti mug, moha) is the one which completely conceals (kun rdzob, samvrti) because it obscures the real nature. On account of that whatever is artificial (bcos ma, krtrima) seems to be truth. The Muni said that [krtrima ] is truth for that which completely conceals (kun rdzob bden, samvrti-satya) and the artificial things (bcos mar gyur pa'i dngos, padartham krtakam) [he called] samvrti, that which completely conceals as well.  1. The inversed gi gu offorihas been taken to indicate of the vowel r. See the same use of gi gu in the transcribed Sanskrit titles of the Tibetan Canon, for ex., in that of the Sunyatasaptati-vrtti D. (3831) tsa 110a4 (ed. Tokyo, 1977, p. 55) and the Vigrahavyavartanl-vrtti , D. (3832) tsa 121a4 (ed. Tokyo, 1977, p. 61), etc. Cf. Hackin J., Formulaire Sanskrit-tibetain du X siecle , Mission Pelliot en Asie Centrale, Serie petit inoctavo, tome II, Paris, 1924, p. 88. 2. The version in our text is samvrtti and not samvrti. Thefirstis derived from sam +vrt (to become, evolve) and the second from sam +vr (to cover, envelop). The Madhyamika authors, notably Candrakirti, rely more on the second interpretation (cf. Prasannapada 492, 10-12, May (1959) p. 226 n. 777), while the Yogacara authors rely more on the first (cf., Nagao (1978) pp. 39-43, 305-20, Nakamura (1980) p. 250-1). dBus pa bio gsal writes samvrtti, but he bases his entire interpretation on samvrti. Most likely the text is corrupt. 3 SDVK k 15ab (C. 2bl), D. 2bl), cited also in SDW (C. 9a2, D. 9a2). 4. MAv VI k 28 (cf. tr. LVP (1910) p. 303. Ogawa (1976) p. 91). The Sanskrit is found in BCAP 171.6-9: mohah svabhavavaranaddhi samvrtih satyam taya khyati yad eva krtrimam jagada tat samvrtisatyam ity asau munih padartham krtakam ca samvrtim //  89 [98b2]  Furthermore, paramarthasatya is the truth (bden pa,  satya) of the most supreme (dam pa, parama) reality (don, artha). For as it is said in the Lankavatara[-sutra X k 258cd] : 1  [The Yogin] sees in the [consciousness] free from subjectobject the supreme wisdom free of Self . [2.3  Divisions within conventional truth]  [2.3.1. Svatantrika opinion] With regard to conventional truth there are two divisions: [that which is] able to perform a function and [that which is] not able to perform a function. (k XII-5ab) [98b3]  The Svatantrikas assert as real conventions (yang dag  pa'i kun rdzob, tathya-samvrti) appearances that are able to perform a function  (don  byed  nus  pa,  arthakriya-samartha) and as false  conventions (log pa'i kun rdzob, mithya-samvrti) appearances that are unable to perform a function. It is also said in the Satyadvayavibhanga [-karika (k 12) of Jnanagarbha] : 2  1. Lank X k 258cd (cf. tr. Izumi (1927) p. 171, Suzuki (1932) p.247, Yasui (1976) p. 266): jHanam anatmakam Srestham nhabhase na paSyati II (Tib.) P. 184a5: bdag med ye shes mchog yin te I snang ba med tshe mi mthong ngo // As the comparison of the variants shows clearly, dBus pa bio gsal does not cite this directly from Lank. This sentence is cited in many treatises such as Madhyamakalankaravrtti P. 79b3-4, Madhyamakaloka, P. 17la4, Bhavanakrama 1 (Skt.) 210.14: (Tib.) 259.13-14, PrajttaparamitopadeSa. 183M-2; and the comparison of variants allows us to conclude that our author cited it from either the Madhyamakalankaravrtti or the Madhyamakaloka, although the verb at the end of pada d is non-honorific in these two treatises, mthong ngo, instead of gzigs so. One ought especially notice that pada d considerably differs from the Lafik as it is cited in Bhavanakrama I (nirabhasena paSyati / snang ba med pas mthong bar 'gym) and in PPU (theg pa chen pos (D. 161a7 po) mthong bar 'gyur). 2. SDVK k 12 (C. 2a5, D. 2a4-5), cited also in SDW (C. 6b5, D. 6b5).  90 Although real and false conventional [truths] are similar in appearance, they are distinguished by their ability or inability to perform functions. [98b4]  Among these two, functional appearances are free from  conceptuality, arise from causes and appear in conformity with what can  perform a function; those lacking  functional  appearances,  for  Satyadvayavibhangakarika  as  [these qualities]  this  same  are non-  [treatise,  the  (k 8) of Jnanagarbha] goes on to say : 1  One should know that real conventional [truth] is the mere thing (dngos  tsam,  signification  and  conditions].  False  vastu-matra),  free  from imagined  arisen in dependence on [causes and 2  [conventional  truth]  is  completely  imagined (kun brtags, parikalpita). [2.3.2. Prasangika opinion] The  other [Madhyamikas, i.e., Prasangikas say real and  false conventions relative to] non-defective and defective [sense faculties] are from the viewpoint (ngof) of the ordinary folk (byis pa, bala); there is no real and false [for them]. (k XII-5cd) [98b6]  The Prasangikas just go along with worldly beings who  believe that which appears to the consciousness of an undamaged sense faculty to be true and that which appears to the consciousness of a  1. S D V K k 8 (C. 2a2-3, D. 2a2), cited also in S D V V (C. 5b2-6, D. 5b3-6). A comparason of variants indicates that dBus pa bio gsal cited this verse from the S D W instead of the SDVK. 2. The word par in pada b of our text (gang skyes par) renders the reading difficult, while the reading of either the S D V K (skyes pa ste) or of the S D V V (gang skyes te) is easier.  91 damaged  sense  faculty  to  be  untrue . 1  As it is said in the  Madhyamakavatara (VI k 25) : 2  The  world knows that which is grasped by the six  undamaged sense faculties. of view of the world.  It is truth just from the point  The remainder is considered false,  just from the point of view of the world. And also [in the Madhyamakavatara (V k 24)] : 3  Furthermore, the seeing wrong is asserted to be of two sorts [dependent upon whether or not] the faculties are clear or have a defect. sense faculty is  The consciousness of a defective  asserted to be false  relative to the  consciousness of a good sense faculty. [99a3]  Therefore, with regard to things, there are objects, sense  faculties and consciousnesses.  They do not function yet appear to  function, and all three have a similar mode of existence in the state of a dream and also in the state of awakening. Hence, [the Prasangikas] do not assert a distinction between real and false conventional truth. Again, from the same treatise (Madhyamakavatara VI k 53) : 4  This [ultimate truth] is like being awake.  The three  [object, sense faculty and consciousness] exist until one  1. Cf. also Pras P 493.1-4 (tr. May (1959) p. 226-7). 2. MAv VI k 25 (cf. tr. de Vallee Poussin (1910) p. 301, Ogawa (1976) p. 86). The Sanskrit is found in BCAP 171.15-18:  vinopaghatena yad indriyanam sannam api grahyam avaiti lokahl satyam hi tai lokata eva' Sesam vikalpitam lokata eva mithya// 1  evam BCAP, eva LVP (1910) p. 301 n. 1.  3. MAv VI k 24 (cf. tr. LVP (1910) p. 300, Ogawa. (1976) p. 85). 4. MAv VI k 53 (cf.tr. LVP (1910) p. 332, Ogawa (1976) p. 158-9).  92 awakes - and awakening, all three do not exist there. Awakening from the sleep of ignorance is like that. And also [in the Madhyamakavatara (VI k 54)] . 1-  When a consciousness [generated] from a tirriira-afflicted sense faculty sees hairs because of the influence of a visual defect (rap rib,  timira)  2  both [the hairs and vision] are  true, relative to that consciousness; both are false relative to clear vision of the object. [99a5] the  They [Prasangikas] also speak of the conventionality of  Superiors ('phags pa'i  kun  rdzob, arya-samvrti) and of mere  conventionality (kun rdzob tsam po, samvrti-matraf [2.4.  Ultimate truth] Because the elaborations (spros pa, prapaifca) have been pacified, ultimate [truth] is indivisible. (k XII-6ab) [99a6]  Ultimate truth is indivisible since all ideas of extremes  such as existence and non-existence etc. have been pacified. The master  1. MAv VI k 54 (cf.tr. LVP (1910) p. 333, Ogawa (1976) p. 159). 2. Cf. May (1959) n. 779.  3. The term kun rdzob tsam (samvrti-matra) is found in Madhyamakavatarabhasya 108.5-6, 12, 15 (ad MAv VI k 28). However, although the term 'phags pa'i kun rdzob  (arya-  samvrti) is not found as such, it seems that it is based on the MAvBh 108.11-15: de Itar na  re zhig bcom ldan 'das des kun rdzob kyi bden pa dang kun rdzob tsam gsungs pa yin no II de la so so 'i dkye bo mams kyi don dam pa gang yin pa de nyid 'phags pa snang ba dang bcas pa'i spyod yul can mams kyi kun rdzob tsam yin la I ...(cf. tr. LVP (1910) p. 305: "Thus the Bhagavan taught samvrtisatya and samvrtimatra. That which is real (paramartha) for ordinary people is only samvrti (samvrtimatra) for the Aryas who are found in the sphere where there is appearance.")..  93 [Nagarjuna] stated in the Madhyamaka-Sastra (dBu ma 'i mdo, XVIII k 9)1:  Not  known  from  aparapratyaya) , at 2  elaborations,  other peace,  beyond  (gzhan not  las  shes  manifesting  concepts  (mam  nirvikalpa), without diversity, this is  the  min,  from rtog  med,  characteristic  (mtshan nyid , laksana) of the ultimate. Also from the Rajaparikatha Ratnavali(l k 62) : 3  Know therefore that the Buddha's immortal teaching that is beyond existence and non-existence, called profound, is the  gift of  the  Dharma (chos kyi  khud pa, dharma-  yautaka).  1. M M XVIII k 9 (cf. tr. De Jong (1949) p. 29, Inada (1970) p. 115, Sprung (1979) p. 183):  aparapratyayam Santam prapaficair aprapaficitam / nirvikalpam ananartham etat tattvasya laksanam II Note that dBus pa bio gsal, in pada d, changed de nyid (of the MM) into don dam, most likely in order to conform to the context of his text. Furthermore, dBus pa bio gsal used three different appellations to designate the MM:  dBu ma 'i mdo , dBu ma rtsa ba and dBu  ma rtsa ba shes rab. For the frequency of these terms, see the Index of Proper Names, Mimaki (1982), p. 275-280. 2. Aparapratyaya  has been translated literally here ("not known from other"), following the  Tibetan. Cf. PrasP 373.1-2:  tatra nasmin para-pratyayo  'stlty  aparapratyayam  paropadeSagamyam svayam evadhigantayyam ity arthah A (Tibetan: De Jong (1949) p.  104.10-12) de la 'di la gzhan las shes pa yod pa ma yin pas na gzhan las shes min te I gzhan gyis bstan pas rtogs par bya ba ma yin gyi /rang nyid kyis rtog par bya ba yin no zhes bya ba 'i don to II 3. Ratnavalil k 62 (cf. tr. Tucci (1936) p. 332, Uryuzu (1974) p. 243, Hopkins (1975) p. 26):  dharma-yautakam ity asman nasty-astitva-vyatikraman / viddhi gambhiram ity uktam buddhanam Sasanamrtam II (Tib.) C. 118a, D. 109a, N. 126b, P. 132a; ed. Hahn (1982).  94 [99b2]  Anything that is called ultimate [truth] expressible in words (rnam grangs kyi don dam pa, paryaya-paramartha) is the negation of the convention of production, etc. That object of negation is then a conceptual construct, and the negation is therefore also a 1  1. The term appears in Madhyamakarthasamgraha k 4, where Bhavaviveka divides ultimate truth into two:  ultimate truth expressible in words (mam grangs kyi don dam, paryaya-  paramartha) and ultimate truth inexpressible in words (rnam grangs ma yin  don dam,  aparyaya-paramartha): dam pa'i don ni spros bral te / de yang mam pa gnyis su bya / mam grangs kyi ni don dam dang / mam grangs ma yin don dam mo II There is not agreement that the author of the MAS is the same person as the famous Madhyamika dialectitian.  Ejima (1980, p. 18-33) tries to eliminate the MAS from the  works of the famous Bhavaviveka. Lindtner (1981, p, 200, n. 14), while aware of Ejima's discussion, believes it authentic.  It is also necessary to note that among Tibetan authors  there is the opinion that supposes the existence of two Bhavavivekas (cf. for ex. CKGT 283 11-12; Mimaki (1982) n. 67). The term paryaya-paramartha  does not appear in SDVK, SDVP, MA, MAV, or BhK.  It seems therefore that it was not known by Jnanagarbha, Santaraksita or KamalaSlla. However, the equivalent of this term, don dam pa dang mthun pa['i don dam pa] (paramarthanukula  [-paramart/ia]"ultimate truth in conformity with ultimate truth") was  known by these three authors: cf. SDVK k 9ab, SDVV D. 6al-2, SDVP P. 6a4 (ad SDVK k 5) and P. 13vl-4 (ad SDVK k 9ab), BhK I (Skt.) 199.7, (Tib.) 245.26.  Jnanagarbha  explains that the negation of production depends on ultimate truth because it conforms to ultimate truth, but that when one examines it through logic, it is seen only to depend on conventional truth:  cf. S D W D. 6al-2: skye la sogs pa bkag pa yang (SDVK k 9a)// yang  dag par skye ba la sogs par rtog pa 'i dngos po bkag pa 'i gtan tshigs kyis I I yang dag pa dang mthun phyir 'dod  (SDVK k 9b)// don dam yin par kho bo cag 'dod do II gzhan dag  (=Yogacara, cf. SDVP P. 13b3) ni yang dag pa kho nar 'dzin pas /yang zhes bya ba ni bsdu ba 'i don to II de yang rigs pas dpyad na kun rdzob kho na ste I... The counterpart of the term in question, mam grangs ma yin don dam  (aparyaya-  paramartha), mentioned also in the MAS, does not appear either in the SDVK or in Jflanagarbha's SDVV, but does appear in SantarakSita's SDVP (P. 6a5).  Indeed, according  to this, it is the SDVK k 5 that expounds the ultimate truth that does not appear in any consciousness, i.e., aparyaya-paramartha.  It is appropriate, however, to emphasize the fact  that none of the three terms {paryaya-paramartha, paramarthanukula[-paramartha ], aparyaya-paramartha) appear in the MA or the M A V by Santaraksita nor in the MAP by KamalaSila, although one finds in these tracts (ad M A kk 69-78) a fairly discussion on the related question. See also Ejima (1980) p. 29-30.  elaborate  95  conceptual construct; and it is therefore conventional truth. It is said in the  Satyadvayavibhahga  [-karika  (k 10)] ^  How could the negation of that which is essentially a conceptual construct not [itself] be a conceptual construct? On account of that, it [negation of production, etc.] is conventional [truth]. It is not ultimate [truth] and it is not correct. 2  However, since it [negation of production, etc.] is in conformity with ultimate truth (don dam pa dang mthun pa, paramarthahukula)^ it is given that name. As it goes on to say in the same [treatise, the Satyadvayavibhanga-karika (k 9ab)] : 4  Because the negation of production, etc. is in conformity with reality, we assert it as ultimate [truth]. 1. SVDK k 10 (C. 2a3-4, D. 2a3), cited also in S D W (C. 6a3-5, D. 6a4-5) and SDVP (C. 25a3 and 25b2, N. 13b4 and 14a4, P. 13b8-14al-2). SDVK k lOab will be cited again in BSGT 109a3. 2. In pada d, our text reads yang dag don min, whiles the Canon reads yang dag don yin The reading of our text is very important. It allows for the correction of the Canon's reading, which in the past has posed considerable difficulty. 3. The term's equivalent is found, for ex., in BhK I (Skt.) 199.7, (Tib.) 245,26. 4. =SDVK k 9ab (C. 2a3, D. 2a2), cited also in S D W (C. 6al, D. 6al-2) and SDVP (C. 24b5, D. 24b5-6, N. 13a6-7, P. 13b 1-2). The form of this verse as cited in our text differs considerably from that of the Tibetan Canon. But, since ICang skya Rol pa'i rdo rje (for example) cited this verse in exactly the same form (cf. CKGT 353.14-15), it is evident that this verse was transmitted in this form among the Tibetan authors. As to its difference with the Tibetan Canon: the first pada is found as is in the text of the Canon; but our second pada is found as is (...pa'i phyir) only in the SDVP and in a slighdy different form (...phyir 'dod) in the SDVK and the SDVV. Our third pada does not exist anywhere in these texts of the Canon. The problem is fairly complicated, but we could suppose the following possibilities. Pada b of the Canon (SDVK and SDVV) could be made by the contraction of the second and third padas of our verse. Or, it is also possible that dBus pa bio gsal erroneously took as a verse the prose of the SDVV (don dam pa yin par kho bo cag 'dod do: C. 6al-2, D. 6a2) which immediately follows the SDVK k 9b: it has already been noted that dBus pa bio gsal has the tendency to cite the SDVK for the SDW.  96  [2.5. Importance of the divisions of the two truths] From knowing or not knowing the two things [=two truths] one finds or does not find the state of Omniscience. (k XII-6cd) [99b4]  Those who know completely the two truths reach the  state of Omniscience by the force of training in the two collections  (tshogs, sambhara). It is said in the Satyadvayavibhanga [-karika (k 2)1»: Those who know the division of the two truths do not misunderstand the words of the Able One.  accumulated the collections (tshogs,  They, having  sambhara ) in their  entirety , accomplish [their own and other's welfare (bdag 2  dang gzhan gyi don, sva-parartha)^ and go definitely to 4  the perfect other side. [99b6]  They are different to those ignorant [of the two truths].  The  master [Nagarjuna] stated in the Mulamadhyamaka [-karika XXIV k 9]5:  1. SDVK k 2 (C. lb2-3, D. lb2-3), cited also in S D W (C. 3b7, D. 3b5) and SDVP (C. 17al-4, D. 17a2-4, N. 4b6-5al, P. 4a4-6). 2. Cf. SDW[D] 4al: ...bsod nams dang ye shes kyi tshogs ma lus par bsgrubs pa...; SDVPfP] 4a5:  bsod nams dang ye shes kyi tshogs ma lus pa yongs su rdzogs par 'gyur la /.  3. Cf. SDVV[D] 4al:  bdag dang gzhan gyi don phun sum tshogs pa bye brag tu byar med  pa'i pha rol tu nges par shin tu phyin par 'gyur ro gyi don phun sum tshogs pa rab kyi mlhar phyin 'gyur  //; SDVP[P] 4a5-6:  ...bdag  dang  gzhan  par 'gyur te / de ni sangs rgyas nyid du  roll.  4. Cf. SDVP[P] 4a7:  'gro  'am/mi  'gro snyam du the tshom za ba'i rgyu med pa'i  phyir/  'gro ba nyid ces bya ba smos so II  5. MM XXJV k 9 (cf. tr. May (1959) p. 228, Inada (1970) p. 146, Sprung (1979) p. 231: ye 'nayor na vijhnanti vibhagam satyayor dvayoh / te tattvam na vijhnanti gambhlram  buddha-Sasane II  97  Those who do not know the division into the two truths do not understand the profound reality  of the Buddha's  teaching. The master Candrakirti also stated [in the Madhyamakavatara VI kk 7980]: For those who stray from the path of the master Nagarjuna there is no way [to get] peace.  They have lapsed from  conventional [truth] and the truth of reality.  On account  of that lapse, they do not achieve liberation . 1  The truth of convention (tha snyad bden pa, vyavaharasatya) is the method and ultimate truth is the outcome of the method; those who do not know these two divisions, 2  on account of  mistaken conceptions (mam rtog log pa,  mithya-vikalpa), enter into bad paths . 3  [3.  Three sub-schools of the Madhyamika:  the Sautrantrika-  Madhyamika, the Yogacara-Madhyamika, and the 'Jig rten grags sde spyod pa'i dbu ma pa]  1. MAv VI k 79 (cf. tr. LVP (1910) p. 355, Ogawa (1976) p. 199). The Sanskrit is found in SS 396.3-6, differing slightly from the Tibetan version of the MAv: acarya-Nagarjunapada-margad bahirgatanam na Sivabhyupayah I bhrasta hi te samvrti-satya-margat tad-bhramSataS casti na moksa-siddhih // La Vallee Poussin (id. p. 355 n. 1) tries to correct, following the Tibetan version, Sivabhyupayah for Sive 'sty upayah and samvrti-satya-margat for samvrti-tattva-satyat. 2. The demonstrative pronoun is in the singular in the Sanskrit anil also in the Tibetan version of the MAv, but in the plural in our text 3. MAv VI k 80 (cf. tr. LVP (1910) p. 356, Ogawa (1976) p. 199). The Sanskrit is found in SS 396.7-10; that of pada ab in BCAP 179.26-27 (BCA IX k 4): upaya-bhutam vyavahara-satyam upeya-bhutam paramartha-satyam / tayor vibhagam na paraili yo vai mithya-vikalpaih sa kumarga-yatah //  98 [100a2] Explanation of the ways in which the two truths are explained. The Yogacara-Madhyamikas etc. [i.e., and the SautrantikaMadhyamikas and 'Jig rten grags sde spyod pa'i dbu ma pas] are [listed as those who assert] conventional [truth] to be 1) the nature of mind, 2) the [exterior] object and 3) as things appear, [respectively]. (k XH-7) [100a3] assert  The master Santaraksita and the master Haribhadra, etc.  appearances of forms, etc.,matter, mind and mental factors, and  dream objects, etc. [and other analogous illusory consciousnesses] as 1  correct conventional [truth]. As it is said in the Madhyamakalankara (kk  91-92)2; Cause  and effect  are  also  just  consciousness  alone.  Whatever exists in its own right is consciousness. In that they are contingent on "mind only" (sems tsam, citta-matra) exterior things are known to be unreal. Based on  this  principle  (tshul,  naya), relative to it  [their]  complete selflessness (anatman) is known as well. Also, since in the Astasahasrika-[prajnaparamita-]aloka^, conventional [truth] is spoken of as concordant with the "Mind Only" [School], these [masters] are known as Yogacara-Madhyamikas (rNal 'byor spyod pa'i dbu ma pa).  1. Cf. B S G T 96b6. 2. M A k 91-92 (C. 56a5-6, D. 56a6-7, N . 47b4-5, P. 52a3-4):  k 91 is cited in M A V (C.  78b7, D. 78b7, N . 73b7-74al, P. 78b4) and k 92 in M A V (C. 79a5, C . 79a5-6, N . 74a7, P. 79a3-4). 3. Cf. A A A by Haribhadra 884.17-885.7.  See also the parallel passage in the 'Gre7 chung  don gsal by the same author: Tib. Skt. ed. Amano (1975) p. 226-9, Jap. tr. Manao (1972) p. 225-6.  99  [100a5] The master [Nagarjuna] also asserts this since he states in the Yuktisastika (kk 21 and 34ab): Here, nothing is produced and nothing is destroyed. Production and cessation are just consciousness . 1  1. YS k 21 (C. 21a5, D.21a5, N. 21a4-5, P 23b2, MsTH PT no. 796 fol. 2a2-3; cf. tr. Yamaguchi S. (1965) p. 64, Uryuzu (1974) p. 49-50), also cited in YSV (C. 15b2-4, D. 16M-7, N. 19bl-4, P. 18b8-19a4). A Sanskrit sloka of close semblance is found in JNA (SSS) 488.22-23 and JNA (SSS) 545.3^:  dharmo notpadyate kaScin napi kaicin nimdhyate I utpadyante nirudhyante pratyaya eva kevalah // It is important, however, to be aware that it is not exactly the same verse as YS k 21 in the Tibetan Canon, which is in a generally different form from that of our text; notably pada d , which differs completely. Here is the YS k 21 of the Tibetan Canon:  de ltar gang yang skye be med II gang yang 'gag par mi 'gyur ro II skye ba dang ni 'jig pa 'i lam 7/ dgos pa 'i don du bstan pa 'o II 'gang YSV MsTH, ci YS 'gang YSV MsTH, ci YS 'lam YSV 1  2  MsTH, las YS . The comparison of variants indicates that the Yuktisastikavrtd verse corresponds to that of the Manuscripts of Touen-houang, and not to that of the Yuktisastika(karika). This shows that the translation of the Yuktisastikavrtd was done earlier (about the ninth century by Ye shes sde, etc.) than that of the Yuktisastika Qsy Pa tshab Nyi ma grags (1055-?) etc.) which is actually preserved in the Canon. The Manuscripts of Touen-houang of the YS Q7T no. 795 and 796) are therefore of considerable importance: they prove the existence of an older translation of the same text (cf. Catalogue de IDan dkar ma, Lalou (1953) n. 591). This verse 21, accompanied by the following (YS k 34), as in our text, is cited in many other texts, such as the Madhyamakalamkaravrtti (P. 79b5-7) by Santaraksita, the Prajffapaiamitopade&a (P. 161b8-162al) by RamakaraSanti, the MadhyamakMamkara-vtttimadhyamakapratipadasiddhi (P. 199b6-7, cf. 133b2) by RatnakaraSanti and the MadhyamakalamkaropadeSa (P. 260a8-bl) by Rataakaransanti (without YS k 34 in MAM): all these treatises cite it in the same form as that of our text The form of this verse cited by Ratnakarasanti is, on the other hand, slighUy different in Tibetan translation. Cf. -PPU (P. 161b8), MAU (P. 260a8-bl):  dngos mams skye ba yod min zhing // 'gag pa 'ang gang na yod min zhing  7/ shes pa 'di nyid kho na ni II skye shing 'gyur ba yin II  100  1  pa MAU  -MAM (P. 119b6-7), MAM (P. 113b2: pada cd only): 'dir gang cung zad skye med pa II cung zad 'gag pa 'ang yod ma yin // skye be dang ni 'gag pa dag 7/ shes pa 'ba' zhig kho na 'o II 1  ni 113b2  Santaraksita (eighth century), in his MAV, cited this verse in a slightly different form than that of the YS, and it is possible that dBus pa bio gsal cites it from the MAV, Ratnakaras"anti from somewhere else. The remark that Kamalas'Ila (eighth C.) made concerning this verse in his MAP merits attention. He does not consider it belonging to the YS, but rather to the Lank. Cf. MAP P. 138b2: 'phags pa Lang kar gshegs pa las gsungs pa'i khungs shes pa 'ba' zhig kho na 'o Zhes bya ba smos te I. Here the reference is to Lank II k 138 (=X k 85). na hy atrdtpadyate kimcit pratyayair na nirudhyate I utpadyante nirudhyante pratyaya eva kalpitah // Cf. -Lank II k 138 (Tib.: D. 88b7, P. 97b3-4): di la gang yang rkyen mams kyis II skye ba med cing 'gag pa med II btags pa 'i rkyen mams kho na ni II skye zhing 'gag par 'gyur yin no '// 'gyur ba yin no P -Lank X k 85 (Tib.: P. 177a4): 1  di la rkyen gyis mi skye ste II gang yang 'gag par my 'gyur ro II brtags pa yi ni rkyen mams nyid II skye ba dang ni 'gag par 'gyur II. Kamalas'Ila, however, considers the second verse as that of Nagarjuna; thus YS k 34. Cf. MAP P. 138b3-4: de'i phyir 'phags pa Klu sgrub kyi zhal snga nas gsungs pa'i tshigs su bead pa gnyis pa "byung ba che la sogs bshad pa zhes bya ba smos so II.  Ratnakarasanti (eleventh C.) takes these two verses as Nagarjuna's. Cf. -PPU (P. 161b7): Klu sgrub kyi zhal snga nas kyang...; -MAU(P. 260a8)K/u sgrub kyi zhal nas kyang II; MAM (P. 119b6): Klu grub kyi zhal snga nas kyis kyang ji skad du II;... -MAM (P. 133b2): slob dpon Klu sgrub kyi zhal snga nas kyis kyang II... Tsong kha pa addresses this problem in his Drang nges legs bshad snying po, and supports Kamalas'Ila. Cf. LS 123.8-14: 'Grel bshad las tshigs bead dang po Lang gshegs las drangs par bshad do II Sha nti pas 'di gnyi ga Klu sgrub kyi yin par byas nas 'dis sems tsam bstan pa la dBu ma pa dang sems tsam pa gnyis 'dra bar bzhed de / zhi ba 'tshos rgyu tu drangs pa la bsams par mhgon yang der byed pa po so sos mdzad pa rgyun gcig la d pa gzhan yang yod do II "In the commentary (=MAP), [Kamalaslla] explains that the first  101  The great elements ('byung ba che, mahabhuta) and so forth which have been spoken about are included in consciousness . 1  verse is cited from the Lankavatarasutra.  Santi-pa (=Rataakarasanti), in attributing these  two verses to Nagarjuna, states that the two [schools], i.e. the Madhyamika and the Vijrfanavadin are the same, since [the position] "mind only" is set forth by these [verses by Nagarjuna].  It seems that [Santi-pa] had in mind the fact that Santaraksita cited [these two  verses] in succession.  However, in this case, there are other [examples] where [two  verses], composed by different authors, are cited in succession". Bhavaviveka also cites this verse (YS k 21) in his MRP (P 327M-2, cf. Lindtner (1981) p. 171 and p. 203 n. 41) and attributes it to the master (slob dpon, thus Nagarjuna):  'di na 'ga' yang skye ba med // 'gag par gyur par gyur pa cung zad med II skye ba dang ni 'gag pa dag II brtags pa 'i rkyen rnams kho na 'o II Owing to the fact that Kamalasila would have to ignore this passage in order to give his interpretation concerning this verse in the MAP, and that the tendency to attribute it to Nagarjuna seems to be rather strange, the author of the MRP could definitely not be the same person as the famous Madhyamika author.  ICang skya Rol pa'i rdo je (1717-1786)  moreover expresses this opinion. Cf. CKGT 283.11-12: dBu ma rin chen sgron ma ni Legs  ldan chung bar grags pas mdzad pa yin gyi slob dpon 'dis mdzad pa min no II " The MRP was composed by he who was known as the littfe Bhavaviveka, and not by this master (=the famous Madhyamika Bhavaviveka )"; CKGT 17.16-7:  Legs ldan phyi mar grags pa'i dBu  ma rin chen sgron ma las kyang... See Mimaki (1982) n. 67 and n. 441. Here in our text, "some" think that the two verses in question are from the YS of Nagarjuna.  Thus, by these "some" our author probably understood RatnakaraSanti and  others. See also -Katsura (1976) p. 486, -Hayashima (1978) p. 1009, -Kajiyama (1976) p. 132 and n. 31, Matsumoto (1980b) p. 176 n. 4, -Lindtner (1981) p. 203 n. 41. 1. YS k 34ab (C. 21M-5, D. 21b5, N. 21b4 P. 24a3-4, MsTH PT 796 fol. 2b3, cited also in YSV C. 20b2, D. 21b7, N. 23b6, P. 24b4; cf. tr. Yamaguchi S. (1965 p. 76-7, Uryiizu (1974) p. 63). The Sanskrit is found in JNA (SSS) 405.1 and JNA (SSS) 545.7:  mahabhutadi vijftane proktam samavarudhyate I See also the preceding note. In addition to the sources that have been enumerated, the PPU 169b8-170al cites this verse, attributing it to Nagarjuna.  102 [lOObl] For the master Bhavaviveka, conventional [truth] is not [set forth] concordant with the Mind Only [School]. He says in the Madhyamakahrdaya (V k 28cd) : 1  [The Buddha] taught "mind only" in the Sutra in order to negate the creator {byed po, karaka) and the enjoyer (za po, bhoktf).  Therefore, [Bhavaviveka who] asserts that there is an object of the knowing mind and that consciousness comes into being together with the form [of the object] is known as Sautrantika-Madhyamika (mDo sde spyod pa 'i dbu ma pa) since [his] conventional [truth] is concordant with [the theories of] the Sautrantikas. 2  [100b2] The master Jnanagarbha and the master Candraldrti assert conventional [truths] as being the accepted conventions of worldly beings. It is stated in the Satyadvayavibhanga [-karika (k 21) : 3  1. M H V k 28cd (Tib. ed. tr. Yamaguchi S. (1964a) p 12 from the end; (Tib.) D. 21a4, P. 23b5). The Sanskrit manuscript reads: sastr'eva cittamatroktih kartr-bhoktr-nisedhatah // 1  2  sastreva ms. okti ms. It is interesting to note that the reading of the Sanskrit ms. does not accord with that of Tibetan canonical version, notably for pada a. This fact shows without doubt the existence of a Skt. ms. in Tibet other than that used for the Tibetan canonical translation. But it is more interesting to note that Tsong kha pa cites these verses precisely in the form that the actual Sanskrit ms. presents. Cf LS 120.19-20: bstan bcos sems tsam zhes gsungs pa II byed po za po dgag phyir ro II. This fact proves the existence of a Tibetan translation other than that of the Tibetan Canon. Here is a good example that shows that it is necessary to treat the transmission of Sanskrit ms. and Tibetan translations with a great deal of care. 1  2  2. Famous theories of the Sautrantika set forth among others in AP k 6 and PV III k 247. See Mimaki (1976) p. 37-38 and n. 143. 3. SDVK k 21 (C. 2b4, D. 2b4), cited also in SDW (C. 10a7, D. 10a7). Once more, the comparison of variants shows that pada c of the BSGT is different from the SDVK and the SDW; for pada a, the BSGT and the SDVV present similar readings that diverge from that of the SDVK.  103 Since an identity is a [mere] appearance one cannot get at it through analysis. Under scrutiny it turns into something else and is therefore sublated. Also from the  Madhyamakavatara  (VI k 3) : 1  Since, if one analyses these things, instead of anything which has a real nature, one ends up with unfindability. Therefore, worldly transactional truth is not analyzable. Because of this, [Jnanagarbha and Candraklrti] are known as the 'Jig rten grags sde spyod pa'i dbu ma pa [Madhyamikas who conform to that which is known in the world] . Similarly, in the ITa ba'i khyad par of the great translator (lo tsa ba chen po) Ye shes sde it is said: 2  3  The works of the master Bhavaviveka and the master Santaraksita are called Sautrantika-Madhyamika and Yogacara-Madhyamika respectively.  1. MAv VI k 35 (cf. tr. LVP (1910) p. 313, Ogawa (1976) p. 121).  For pada b, the  translation by Nag tsho Tshul khrims rgayal ba (1011-?) etc. (P. [98] (5261) 'a 230a8) is "de bzhin nyid kyi dngos..." instead of "di nyid bdag can dngos..." as in the translation by Pa tshab Nyi ma grags (1055-?) etc. on which the LVP edition is based. Cf. Ogawa (1976) p. 121 n.3. 2. For this school and the expositions of the Tibetan authors regarding it, see the introduction. 3. TKh (C. 218M, D. 213M, N.244a6, P. 252M-2, MsTH PT no. 814 fol 5a5-bl).  The  citation, however, is not literal (see, for ex., the proper names Bhavya and Santaraksita), and our author seems to have changed the text. MsTH:  See the Canon's version and that of the  a tsa rya Bha byas mdzad pa la ni' mDo sde pa 'i dbu ma zhes btags I a tsa 1  2  4  5  rya Sha nta ta kzhi tas mdzad pa la ni rNal 'byor spyod pa 'i dbu ma zhes btags so II tsa 6  NP MsTH shi  Ba'phyas  2,  7  MsTH  8  7 MsTH 'spyod MsTH 'tsa  l  NP MsTH Shan ta rag 6  MsTH bzad MsTH V MsTH . The variant (mDo sde pa'i dbu ma I mDo sde spyod 1  pa'i dbu ma) indicates without doubt that our author used a text closer to the Canon than to the MsTH.  The Tibetan authors often cite this TKh passage, but never literally (cf. LRCM  kha 6M-4, CKGT 292.6).  Is this due to the fact that the TKh is conserved only in a  pitiable state in the Canon? For detail, see Mimaki (1982), p. 40-1.  104 [4. Ultimate truth where the elaborations are pacified] It is asserted that ultimate [truth] is the pacification of the elaborations  (spros pa, prapafica) of existence and non-  existence. The books (gzung, grantha) that clearly explain this are the Madhyamika-Sastra (dBu ma'i mdo) and others. (k XII-8) [100b6]  It is asserted that ultimate truth is the pacification of all  elaborations of existence and non-existence and that furthermore, the Tathagatha, having understood it, taught it.  As the master [Nagarjuna]  said in the Madhyamika-Sastra (XVIII k 7) : 1  When objects within the domain of mind cease, the objects of expression cease. characteristic  similar  The unoriginated and unceasing is a to  liberation  (mya  ngan 'das,  nirvana). The master Kamalas"ila also stated in the Tattvavatara : 2  It is taught that the self-realized suchness is divorced from labels (tha snyad, vyavahara) and other [manifestations of 3  1. M M XVIII k 7 (cf. tr. L V P (1933) p. 39, De Jong (1949) p.22, Inada (1970) p. 115):  nirvrttam abhidhatavyam nivrlte cilta-gocare I anutpannaniruddha hi nirvanam iva dharmata II The variant in pada d (mtshan nyid /chos nyid) should be noted. 2. T a (D. 273a3-4, P. 312a4). The same verse is also cited in B S G T 103b6. ' 3. The variant in pada d (tha snyad I tha dad) should be noted. In the context of the T A , the reading tha dad is indispensible, because it is this last verse of the treatise that posits the idea that truth is neither different (tha dad) from phenomena, nor identical (tha mi dad) to phenomena.  D i d dBus pa bio gsal replace tha dad with tha snyad in order to be in  conformity with the context of the BSGT?  have tha snyadl  Or did the text of the T A that he saw already  105 conventional  truth],  is  at  peace  and  free  from all  elaborations of being and non-being. [5.  Two sub-schools of the Madhyamika:  the Svatantrika and the  Prasangika] [101a2] Explanation of the difference between the Svatantrika and the Prasangika. [5.1. Svatantrika] The Svatantrikas par,  assert as a distinguishing feature (khyad  viSesana) the negative and affirmative theses which  come about through the three reasons (rtags, hetu), the two valid  cognizers  (tshad  ma,  pramana) and the two  consequences (thai ba, prasanga)} They do not assert that objects and so on are false, etc. [101a3]  (k XII-9)  The Svatantrikas such as the master Bhavaviveka etc.  explain that there are three reasons (rtags, hetu): from cause ('bras bu, karya), from nature (rang bzhin, svabhava), and to the unobserved (ma dmigs pa, anupalabdhi ) ; that there are two valid cognizers (tshad ma, 2  pramana): direct (mngon sum, pratyaksa) and inferential (rjes su dpag pa,  anumaha); and that there are two conseqences (thai 'gyur,  prasanga): one which forces a reason (sgrub byed 'phen pa, sadhana) and  one  establish  which damages emptiness,  [the  non-arising  assertion] (sun and  illusoriness  'byin).  When they  (sgyu ma bzhin,  maya), etc., and remove belief in the truth of subject (yul chen, visayin) and object (yul,  visaya) with these there is ultimately (don dam par,  paramarthatas) the state of emptiness (stong pa nyid, §unyata).  They  distinguish their thesis (dam bca', pratijfia) with this, etc.  1. For all these terms of logic and their references, see, for ex., Mimaki (1976) p. 46-66. 2. For these three logical terms and their references, see for ex., Mimaki (1976) p. 46-66.  106 [101a5] They [Svatantrikas] assert  1  1) that all objects and minds  are not necessarily false and mistaken, 2) that there exists a view which is admitted on the conventional level, 3) that on the conventional level there is truth and that truth can be divided into real and false , 4) that 2  the absence of self nature of an illusion (sgyu ma, maya) is the ultimate and 5) that the Buddhas have a self-perpetuating wisdom consciousness (rang rgyud kyi ye shes). [5.2 Prasangika] As  for  the  others  [Prasafigikas]  the  appearance  of  affirmatives and negatives which come about by way  of  the four reasons and the four valid cognizers just indicates the internal contradiction of the opponants' [thesis]. [That which the  Svatantrikas assert as conventional truth]  does not exist [for the Prasafigikas] because it contradicts the  truth.  It  [conventional truth] is  designation (btags pa,  prajffapti).  exhausted in  its  (k XII-10)  [101a6] When the Prasangikas such as master Buddhapalita, etc., with four reasons (rtags, hetu): someone else  (tha  inference conventionally known  snyad du gzhan la grags pa'i  to  rjes dpag, para-  prasiddhanumana)^; consequence (thai 'gyur, prasanga) expressing a  1. Compare with the Prasangika opinion set forth in B S G T 10Tb3-4. 2. See also B S G T 98b3-4.  3. Cf. -PrasP 34.13-35.1 (ad M M I k 1): kim punar anyatara-prasiddhenapy anumahenasty anumana-badha I asti sa ca svaprasiddhenaiva hetuna, na para-prasiddhena, lokata eva drstatvat I; (Tib.) P. [98] (5260) 'a 12b2-3: yang ci gang yang rung ba la grub pa'i rjes su dpag pa 'i sgo nas kyang rjes su dpag pa 'i gnod pa yod dam zhe na I yod de de yang rang nyid la grub pa'i gtan tshigs nyid kyis yin gyi I gzhan la grub pas ni ma yin te/ 'jig rten nyid du mthong ba'i phyir ro II; -PrasP 35.4-5: ata eva ca kaiScid uktam na paratah prasiddhibalad anumana-badha, parasiddher eva riiraciklrsitatvad iti II; (Tib.) P. [98] (5260) 'a 12b5-6: de nyid kyi phyir 'ga' zhig gis gzhan la grags pa'i dbang gis ni rjes su dpag pa'i gnod pa ma yin te / gzhan la grags pa nyid dgag par 'dod pa'i phyir ro zhes bshad do II.  107 contradiction ('gal ba, virodha) ; reason equal to that which is to be proved (bsgrub bya dang mtshungs pa, sadhyasama) ; consequence through similarity of reason (rgyu mtshan mtsungs pa'i mgo snyoms^); and four valid cognizers (tshad ma, pramana): direct 1  2  4  1. The term ('gal ba brjod pa'i thai 'gym) itself does not seem to appear as is in the PrasP, but it could be based on, for ex., PrasP 100.5-8 (ad M M II k 12). 2. In Indian logic, sadhyasama is considered one of the five fallacious reasons (hefvabftasa). Cf. Nyayasutra 1-2-8: sadhyaviSistah  sadhyatvat sadhyasamah II "The sadhyasama [is a  fallacious reason that] is not different from that which is to be proved, because it is to be proved." The sadhyasama example that Padsilasvamin Vatsyayana gives in NBh 384.3-385.1 is as follows: [Assertion:] Shadow is a substance (dravyam chaya). possesses movement (galimattvat).  [Reason:] Because it  In this syllogism, the reason (or middle term) 'shadow  possesses movement' is itself a statement to be proved:  it therefore cannot be used as a  valid reason to prove the assertion. This is sadhyasama as it is explained in NS and NBh. Among the Madhyamika texts, the term is found in, for ex., M M IV  kk 7-8,  Vigrahavyavartani k 28, Vigrahavyavartanlvrtu' ad V V k 28 and k 69, PrasP 33.6, 58.8, 127.11, 153.5, 222.2, 382.16, 413.2. But the Madhyamika sadhyasama does not seem to be exactly the same as that of the Naiyayika. See Bhattacharya K. (1974). See also Randle H. N. (1930), Indian Logic in the Early Schools, Oxford Univ. Press, p. 15 n. 3 and Matilal B. K. (1974). 3. It seems that the term is not found in the PrasP. Here is an explanation of the term based on dGe bshes Rab brtan's commentary on Tsong kha pa's LS:  if, in ultimate truth, the  visual faculty could see the visible (gzugs), it would follow that the auditory faculty (ma ba) could also see the visible, because they are equal as faculties (dbang po)  which  perceive that which doesn't exist in ultimate truth; that is, the thesis that in ultimate truth the visual faculty sees the visual is (by a parallel reasoning) reduced to absurdity. Thus he proves that the visual faculty does not see the visible in ultimate truth. Cf. LSQ3.) 122a6-b3 = p. 243-4 (ad LS 117.6-7): sbyor ngag ji lta bus 'gog snyam na M1G NI chos can DON DAM PAR GZUGS LA MI LTA STEI dbang po yin pa'i phyir I dper na RNA  BA'I  DBANG PO BZHIN NO ZHES PA DANG I 'di thai ngag byas na sgra 'dzin nyan shes kyis gzugs mthong bar thai I gzugs 'dzin mig shes kyis don dam par gzugs la blta ba 'i phyir zhes pa lta bu rgyu mtshan mtshungs pa'i mgo snyoms kyi thai 'gyur ro/ mgo snyoms tshul ni mig gis don dam par gzugs la lta na gang du yang ma bltos par blta dgos par 'gyur zhing di ltar lta na ma bas kyang gzugs la lta chog par thai bar 'gyur zhes pa'o / (L& in capitals). 4. These four Prasangika reasons are mentioned in, for ex., the gZhung lugs legs par bshad pa by Sa skya Pandita (fol. 149a4-5:  'di la bzhi ste / gzhan la grags pa'i rjes dpag dang I  rgal ba brjod pa'i thai 'gyur dang I rgyun (sic, read rgyu mtshan) mtshungs pa'i mgo snyoms dang I bsgrub bya dang mthungs pa'i ma grub pa'o //...; and Lam rim chen mo by Tsong kha pa, kha 60b5-61a3 (cf. tr. Nagao (1954) p. 233-4, Wayman (1978) p. 287).  108 (mngon sum, pratyaksa), inferential (rjes dpag, anumana), scriptural (lung, agama) and from analogy (nye ba 'jal ba, upamana) establish 1  emptiness and remove belief in the truth of subject and object they do not, thereby, make any particular assertion in their thesis since [for them] even conventionally there is not truth.  A l l refutation and  assertion is made relative to an opponent since their (pha rol po, para) assertions  have  an  internal contradiction.  With  regard to  the  Madhyamikas, just as there is no assertion from their (Prasangikas) own side (rang phyogs, sva-paksa), so also there is no refutation of the other side (gzhan phyogs, para-paksa) either. [101b3]  Thus, [the Prasangikas] assert : 1) that all objects that 2  appear are false and all consciousnesses are mistaken; 2) that no view (lta ba, darSana) exists [even on the conventional level]; 3) that on the conventional level there is no truth nor [division of it] into real and false; 4) that ultimate truth is free from elaborations; 5) the Buddhas do not have self-perpetuating consciousness  (rang rgyud kyi ye shes). If  these [points] were to be explained together with their reasons, [our treatise] would become very large.  [5.3.  Real difference between the Svatantrika and the Prasangika]  The  real difference (khyad par dngos) of these modes  (tshul , naya) [of  Svatantrika and Prasangika] is the  assertion or non-assertion of reasons from real things and of valid cognizers [that come about from them]. differences]  [The  that are otherwise expressed by others are  cleared away.  (k XII-U)  1. Cf. PrasP 75.3-9 (tr. Stcherbatsky (1927) p. 163, Yamaguchi S. (1947) p. 110, Sprung (1979) p. 63-64). 2. Compare with the Svatantrika opinion shown in B S G T 101a5-6.  109  [101b5] Here, the real difference between these two [schools, i.e. Svatantrika and Prasangika] is, while both say all phenomena are essentially empty one does, and the other does not, assert reasons that are based on real things (dngos po stobs zhugs, vastu-bala-pravrtta)  1  and that are established autonomously (rang rgyud, svatantra) and valid cognizers [based on those reasons].  Hence it is a division of [asserting  and not asserting] a reason (gtan tsigs, hetu). Thus it says in the Satyadvayavibhanga [-karika (k 18-19)] : 2  To the extent that the factor of appearance exists [in common] in the consciousnesses of the two debators there is an conception of subject (chos can,  dharmin), and  attribute (chos, dharma). Then an inference comes into being, not when it is otherwise.  So, if the logicians (rigs  pa smra pa, nyaya-vadin)^ use such [inferences] who could refute them? It is also said in the Madhyamakaloka [of KamaSila] : 4  1. See the equivalent of the term in, for ex., NBT 56.20: vastu-bala-pravrttam lingamj; (Tib. 128.13-14) dngos po'i stobs kyis zhugs pa'i rtags (ad NB III s 46). 2. SDVK kk 18-19 (C. 2b3-4, D. 2b2-3), cited also in S D W (C. 9b5-6, D. 9b5-6). These two verses will be cited in BSGT 106al-2. Note that Jnanagarbha's SDVK is cited here as representing the Svatantrika opinion. 3. See the equivalent term in , for, ex., PV I k 212d. 4. MAI (D. 181a6, P. 198a5-6) The form of the passage in the Tibetan Canon differs appreciably from that found in Bio gsal grub mtha'; the second sentence of the latter is not found in the Canon. Here is the Canon's version: kho bo cag kyang tshig tsam gyis chos thams cad ngo bo nyid med par sgrub pa yang ma yin la / thai bar sgrub pa tsam gyis kyang ma yin no II 'o na ci zhe na I yang dag pa 'i tshad ma nyid kyis sgrub bo II. Other such discrepencies with the MAI sometimes occur. For example, a MAI passage cited in the ITa ba 'i khyad par by Ye ses sde with the clear mention of the MAI is not found in the MAI (cf. VTa ba'i khyad par P. 254a2-3, MsTH PT no. 814 fol. 9a4-5). In the case of the ITa ba'i khyad par citation, it could be supposed that Ye Ses sde summarized the MAI passage (D. 159a7-bl, P. 173b3-4, cf. Matsumoto (1981c) p. 111). But in our case could the existence of another Tibetan translation of a different transmission be supposed?  110  We do not prove that all phenomena are empty only by consequences (thai 'gyur, prasanga), but prove [it] as well through the force of correct valid cognizers. Moreover, we assert it just because the Bhagavan has authorized it (rjes su gnangba, anujna) . 1  [102a2] And from the  Madhyamakavatara  (VI k 31 ab) : 2  Since the world's [acceptance of what is true] is not at all authoritative, when reality is at issue, the world does not harm [what the transcendent establishes]. It is also said in the Prasannapada :  3  Since a Madhyamika does not assert the other position (phyogs gzhan, paksantara) it is not proper that he formulates an autonomous inference (rang gi rgyud kyi rjes su dpag pa, svatantranumana).  [102a3] This [explanation] clears away [the opinion of] some who differentiate [Svatantrikas and Prasangikas] with slight distinctions such as Buddha having or not having transcendental wisdom or the existence of truth or lack of it on the conventional level 4  1. Cf. Dictionnaire deTshe ring dbang rgyal, Bacot (1931) 56a3-4: rjes su gnang ba I  anurupa, ajflapta, anujRa. 2. M A v V I k 31ab (cf. tr. L V P (1910) p. 308, Ogawa (1976) p. 109). 3. PrasP 16.2 (cf. tr. Stcherbatsky (1927) p. 95, Yamaguchi S. (1947) p. 22, Sprung (1979) p.  37):  na  ca  Madhyamikasya  svatah svatantram  anumaham  kartum  yuktam  paksantarabhyupagamabhavat /; (Tib.) D. 6a2, P. 6b3-4. Cf. L R C M kha 61b5-6 (tr. Nagao (1954) p. 236 and n. 212, Wayman (1978) p. 289). 4. Cf. B S G T 101a5-6 and 101b4. It has not been determined precisely to whom dBus pa bio gsal refers by "some" (kha cig).  Ill and  those others who say the division into Svatantrika and Prasangika 1  did not exist in India ('phags pa'i  [6.]  yul).  Intermediary verses  2  Where there is an absence of true existence, reasoning reasoning.  from  things  and  [the  it excludes  understanding]  from  If there is true existence how could things be  empty of truth?  Even if things (chos can,  dharmiri)  are mere imputations  the [imputing] mind would be a true subjective element. Therefore, since that which characterizes valid knowledge is not suitable vis-a-vis [the subjective aspect] it is not tenable [that it is true].  1. Here dBus pa bio gsal refutes the opinion of "those others" who do not assert the existence of the distinction between the Svatantrika and the Prasangika in India. One cannot be sure that dBus pa bio gsal knew that these terms were invented by the Tibetans, in contrast to Tsong kha pa and Sakya mchog ldan, for ex., who knew it quite well. Cf. L R C M kha 6b6-7 (cf. tr. Nagao (1954) p . l l l , Wayman (1978) p. 182): Gangs ri'i khrod kyi phyi dar gyi mkhas pa rnams dbu ma pa la thai 'gyur ba dang rang rgyud pa gnyis kyi tha snyad byed pa ni tshig gsal dang mthun pas rang bzod (read bzor following the other editions) mi bsam moll "The scholars of the second diffusion of Buddhism to Tibet created, for the Madhyamika, the appellations of the two [sub-schools], Prasangika and Svatantrika, but I do not think that they are pure invention (rang bzo), because they conform to the Prasannapada"; -BN kha 13b3: Rang rgyud pa dang thai 'gyur ba zhes bod rnams kyis tha snyad byas pa rnams so II "Svatantrika and Prasangika are names fabricated by the Tibetans" (cf. Matsumoto (1981d) p. 423 n. 1). 2. Vetter has clarified the role of antaraSloka and in what capacity it differs from samgrahaSloka (recaptulative or mnemonic verse):  antaraSloka expresses an independent  idea, on which the text's author did not think it necessary to comment, while samgrahaSloka summarizes the preceding prose (cf. Vetter (1966) p.7).  But this difference became less  and less strict, and certain antaraSloka became invested with the role of samgrahaSloka amongst the Indian authors. This could be why the. Tibetan "authors almost uniquely used the antaraSloka, (with one exception: Klong chen rab 'byams pa in his Grub mtha' mdzod.). For detail, see Mimaki (1980a).  112  Since there is no valid knowledge your contradictions are made non-definite.  If [they] don't exist what is disproved  or proved is equally without valid cognition. Where  object  and  subject  are  false  1  and  mistaken  [respectively] it is difficult to establish, even from the point of view of the world, a division of correct and incorrect on the conventional level. A valid cognition would be mistaken on account of an object of a single transcendental valid cognition being false.  [If transcendental wisdom] is not mistaken, since its  object would be true one would have contradicted [the basic tenet that] all things are false. Where there is no [subjective element] to establish what is observed a [subjective possible.  Other  lines  element] observing that is not of  reasoning,  having  become  ineffective [in the light of the above logic] do not establish those [subject or object]. Where the omniscient transcendental wisdom does not exist [even conventionally] what Sangha and Dharma are there?  If on the transactional level even cause and effect  do not exist that is a nihilistic view. [7.  The phenomena existing in conventional truth do not exist in  ultimate truth and, in reality, are beyond existence and non-existence]  1. B y definition a valid cognizer apprehends a real thing. question only has a nominal entity for its object. cognizer.  Now, the consciousness in  It is therefore not able to be a valid  113  [7.1. Explanation following the Sutra] The Bhagavan (bcom ldan 'das) taught existence, nonexistence and  transcendence  of  existence  and non-  existence. (k XII-12ab) [102b2]  The Bhagavan taught that all phenomena exist in  conventional [truth], do not exist in ultimate [truth] and in reality (de kho nar, tattvatas) transcend existence and non-existence.  As it is said  in the Lankavatara [-sutra (X k 429] As  conventions  (samvrtya) all phenomena exist; as  ultimates they do not. That which is in respect to [things] without self-nature (rang bzin med, nihsvabhava) is called sam vrti [concealer]. Also in the TrisamvaranirdeSa [-parivarta] : 2  1. Lank X k 429 (cf. tr. Izumi (1927) p. 184, Suzuki (1932) p. 259, Yasui (1976) p. 284): bhSva vidyante samvrtya paramarthe na bhavakahl  nihsvabhavesu ya bhrantis tat satyam samvrtir bhavet/l (Tib.) P. 190b6. Cited in many texts. For ex., SDVV[D] 9a3; BhK I (Tib.) 249.18-21; MA1[P] 168a3-4. dBus pa bio gsal's version differs considerably from that of the Canon, in particular for pada ab which, one might notice, are identified in dBus pa bio gsal to the corresponding padas of Lank X k 120 cited immediately after in our text (102M-5). 2. TrisamvaranirdeSaparivarta (sDom pa gsum bstan pa'i le'u = chap. 1 of the Ratnakuta: P. [22] (760(1)) tshi 10b5. The Sanskrit is found in PrasP 370.6-8 (ad MM XCffl k 8, cf tr. De Jong (1949) p. 27, Sprung (1979) p. 181): [loko maya sardham vivadati] naham lokena sardham vivadami II yal loke 'sti sammatam tan mamapy asti sammatam / [yal Joke n sammatam mamapi tan nasti sammatam...'].  114 I do not dispute with the world, since what the world asserts as existing, I also assert as existing.  Also in the Lankavatarasutra : Everything exists as conventions; they do not exist as ultimates.  It is taught that ultimately all things are .  entityless.  1  and: Non-existence  :  presupposes  presupposes non-existence.  existence  and  existence  Therefore one does not need  to understand existence; neither does one need to suppose existence.  2  [7.2. Explanation following the Madhyamikas ] The wise, who profess the Middle Way, have clarified this doctrine (tshul, naya).  (k XII-12cd)  [7.2.1. Existence in conventional truth] 1. Lank X k 120a-d (cf. tr. Izumi (1927) p. 160, Suzuki (1932) p. 235, Yasui (1976) p. 251):  sarvam vidyati sarnvrtyam paramarthe na vidyate/ dharmanam nihsvabhavatvam paramarthe 'pi dfSyate II (Tib.) p. 178b2-3. Cited in MA1[P] 167b7-8. 2. Lank X k 501 = III k 83 (cf. tr. Izumi (1927) p. 109, 189, Suzuki (1932) p. 164, Yasui (1976) p. 173, 292):  astitva-purvakam nasti asti nastitva-purvakam / ato nasti na gantavyam astitvam na ca kalpayet// (Tib.) P. 193b3-144a6-7. Cited in MAV[P] 72a6-7 and MA1[P] 165b4. Lank III k 83 and 3 k 501 are exactly identical in Sanskrit, but slightly different in Tibetan. comparison of variants, our text is closer to X k 501.  Regarding the  115  [102b6]  The master [Nagarjuna] put forth this same idea in the  Mulamadhyamaka-karika (XXIV k 10) : 1  Ultimate  [truth]  (dam  pa'i  don, paramartha) is not  understood without reliance on convention (tha snyad, vyavahara); one does not attain nirvana without having understood ultimate [truth]. In the Vigrahavyavartani (k 28cd) : 2  We do not make conversation without asserting language. [103a2]  Similarly,  the  master  Bhavaviveka  said  in the  Madhyamakahrdaya : 3  Without the ladder of correct conventional [truth] it is not possible for the wise man to ascend to the top of the palace of ultimate reality (yang dag, tattva).  1. M M X X I V k 10 (cf. tr. May (1959) p. 229, Inada (1970) p. 146, Sprung (1979) p. 232:  vyavaharam anaSritya paramartho na deSyate I paramartham anagamya nirvanam nadhigamyate II 2. V V k 28cd (cf. tr. Yamaguchi S. (1929) p. 30, Bhattacharya K. (1978) p. 21, Kajiyama (1974) p. 156):  sarnvyavaharam ca vayam nanabhyupagamya kathayamah II (Tib.) ed. Tucci (1929) p. 33. 3. This verse occurs between M H III k 11 and 12. The Skt is found in A A A 169.19-20:  tathya-samvrti-sopkham antarena vipaScitah / tattva-prasada-Sikhardrohanam na hi yujyatel/ Cf. Ejima (1980) p. 271; Iida (1980), p. 67 includes the Sanskrit as:  tattva-prasMa-Sikharorohanam na hi yujyatel tathya-samvrti-sopanam antarena yatas tatah/f  116 The master Candrakirti also said in the Prasannapada : l  Because the Madhyamikas assert that dependent-arising exists  conventionally they are not like Nihilists (Med pa  pa, Nastika).  The master Kamalas'Ila also said in the Tattvaloka (kk 93 and 95): These definitions of a valid cognition have also been taught as conventionalities in order to lead people to reality so accept them  2  If one does not depend on language, even though one makes use of other methods,one cannot distinguish reality [taught] by myself and others.  3  These teach the way things exist conventionally. [7.2.2. Non-existence in ultimate truth] [103a5]  On the other hand, the master [Nagarjuna] has also  stated in the Vigrahavyavartani (k 29, 30 and 63): If I had any thesis then I would have this fault. Because I do not have a thesis I am quite faultless.  4  1. =PrasP 368.14-15 (ad M M XVIII k 7): samvrtya  Madhyamikair astitvenabhyupagaman  na tulyata II; (Tib. ed. De Jong (1949) p. 101.13-15: dbu ma pa dag gis kun rdzob tu yod par  khas blangs pa'i phyir la I de dag gis khas ma blangs pa'i phyir mi mtshungs pa nyid do II, cf. tr. De Jong (1949) p. 25. 2. T A (D. 247b5, P. 279b3-4). 3. T A (D. 247b6, P. 279b5-6) 4. W k 29 (cf. tr. Tucci (1929) p. 34, Yamaguchi S. (1929) p. 31, Bhattacharya K. (1978) p. 23, Kajiyama (1974) p. 157:  117  If  something  were  observable,  if  it  were  provable or  disprovable by things like direct perception [I could be faulted].  Since this does not exist, I cannot be criticized.  Because  there  is  no  object  of  pratisedhya) I do not negate.  negation  (dgag  Therefore, you  1  bya,  liber me  falsely in saying that I negate.  2  [All  these] words show the way things do not exist ultimately.  [7.2.3. The transcendence of existence and non-existence ]  [103bl]  Furthermore,  the  master  [Nagarjuna]  said  in  the  Mulamadhyamaka-karika ( X V k 7 and XXII k 11):  yadi kacana pratijfla syan me tat esa me bhaved dosah I nasti ca mama pratijM tasman naivasti me dosah II Cited in PrasP 16.7-8 (cf. Johnston and Kunst (1951) p. 127 n. 4). (Tib.) ed. Tucci (1929) p. 35. 1. W k 30 (cf. tr. Yamaguchi S. (1929) p. 32-33, Bhattacharya K. (1978), p. 24, Kajiyama (1974) p. 158): yadi kimcid upalabhcyam pravartayeyam nivartayeyam val pratyaksadibhir arthais tad-abhavan me 'nupalambhah II Cited in PrasP 16.9-10 (cf. Johnston and Kunst (1951) p. 128 n. 2).  (Tib.) ed. Tucci (1929)  p. 37. Our text's variant for pada is visibly inferior to that of the Tibetan Canon; therefore, it has been corrected. 2. W k 63 (cf. tr. Bhattacharya K (1978) p. 41, Kajiyama (1974) p. 177-8; k 64 following Tib. ed. Tucci (1929) p. 67 and tr. Yamaguchi S. (1929) p. 54): pratisedhayami naham kimcit pratisedhyam asti na ca kimcit / tasmat pratisedhayaslty adhilaya esa tvaya kriyatell (Tib.) ed. Tucci (1929) p. 67. In this verse, the variants of our text are better than those of the Tibetan Canon:  pada b is much closer to the Sanskrit and in pada d the reading (skur  pa) proposed as improvement by Johnston and Kunst (1951) p. 145 n. 8 is found as proposed.  118  In the  Exhortation to  Katyayana (Ka  tya ya na yi gdams  the Bhagavan, who knows being and non-being, negated both "it is" and "it is not". ngag, Katyayanavavada)  1  One cannot say that he [= Tathagata] is empty, or that he is not empty. One cannot say that he is both, [i.e.,empty and not empty at the same time] or neither [i.e., not empty and not not-empty at the same time]. [These terms] are to be used as designations (gdags pa, prajnapti) 2  And in Nagarjuna's  Acintyastava  (kk 21-22ab):  3  "It is" is an etemalistic view. "It is not" is a nihilistic view. Therefore, you taught the Dharma which is free from the two extremes (mtha' gnyis, anta-dvaya) and said that phenomena are free from the four extremes (mu bzhi, catuskoti).  In the  Ratnavali[l  k  57  of Nagarjuna]:  4  1. M M X V k 7 (cf. tr. Schayer (1931) p. 69, Wogihara (1938) p. 588, Inada (1970) p. 99, Nagao (1967) p. 279):  Katyayanavavade casthi nastlti cobhayam / pratisiddham bhagavata bhavabhava-vibhavina II 2. M M X X I I k 11 (cf. tr. De Jong (1949) p. 80-81, Inada (1970) p. 134, Sprung (1979) p. 201):  Sunyam it na vaktavyam aSunyam id va bhavet I ubhayam nobhayam ceti prajRapty-artham tu kathyate// 3. A C S kk 21-22ab (rest Skt.):  astlti $a£vata-graho naslity uccheda-darSanam / antadvaya-vihinas tad-dharmo 'yam deiitas tvayal catuskoti-vinirmukta dharmas tat kathitas tvayal Cf. M M X V k lOab. 4. R A I k 57 (cf. tr. Tucci (1936) p. 321, Uryuzu (1974) p. 242, Hopkins (1975) p. 25:  nastiko durgatim yati sugatim yati castikah I yathabhuta-parijfianan moksam advaya-niSritah II  119  Nihilists proceed to bad rebirths (ngan 'gro, durgati). Eternalists go to good rebirths (bde 'gro, sugati). Those who do not rely on the two [extremes], because they know reality as it is (yang dag ji bzhin, yathabhuta) become liberated. [103b4] Similarly, the master Aryadeva says in the (k 400): i  CatuhSataka  It is impossible to criticise, even after a long period of time, someone who does not assert of something that it exists, does not exist, or that it both does and does not exist.  (Tib.) C. 118a, D. 109a, N. 126b, P. 131b; ed. Hahn (1982). 1. CS k 400 ( =XVT k 25: cf.tr Vaidya (1923) p. 167). The Skt. is found in PrasP 16.4-5 (ad M M I k 1; cf. Stcherbatsky (1927) p. 95)  sad asat sad-asac celi yasya pakso na vidyate I upalambhaS cirenapi tasya vaktum na Sakyatell The variants in Tibetan are considerable. Here is the Canon's version:  yod dang med dang yod mod ces I gang la phyogs ni yod min pa I de la yun ni ring po na 'ang I klan ka brjod par nus ma yin // Cf. Vaidya (1923) p. 128 and 167; Bhattacharya V (1931) p. 296. The reading in Bio gsal grub mtha' resembles that of the M A k 68 (D. 55bl, P. 51a6; M A V D. 72al, P. 69b8-70al) cited also in the Jffanasarasamuccaya-nibandhana 0?. 51b5, (Tib.) ed. Mimaki (1976) p. 206) by Bodhibhadra:  yod dang med dang yod med ces I khas mi len pa gang yin pa I de la nan tan ldan pas kyang / cir yang klan ka bya mi (ma MAfP]) nus II  120 Also, in the Jnanasarasamuccaya [(k 28) of tantric Aryadeva]:  1  1. Jnanasarasamuccaya k 28 (cf. tr. Yamaguchi (1965) p. 315, Mimaki (1976) p. 189). The Skt. is found in BCAP 174.11-12 (cf. 173.30), Tattvaratnavalik 25, Subhasitasamgraha (1903) p. 389.9-10: na san nasan na sad-asan na capy anubhayatmakam / camskod-vimrmuktam tattvam madhyamika viduh II, and also in Sarvasiddhantasamgraha III k 7: na san nasan na sad-asan na cobhabhyam vilaksanam/ catuskoU-vinirmuktam ta[t]tvam madhyamika viduh// The Tibetan of our text is a bit different from that of the JSS: yod min med min yod med min I gnyis ka 'i bdag nyid kyang min pas I mtha'bzhi las grol dbu ma pa / mkhas pa mams kyi de kho na'o II, and much closer to that of the BCAP (P. 213a8): yod min med min yod med min I gnyis kyi bdag nyid du yang med I mtha' bzhi dag nges grol ba I de nyid dbu mar mkhas mams bzhed II. Following Bodhibhadra's commentary (JSSN P. 51M-3, tr. ed. Mimaki (1976) p. 204-7), the third and fourth of the four extremes of the JSS k 28ab are successively the negation of non-existence/non-non-existence and the negation of existence/non-existence: not the usual order. But the reading of our text takes the normal order of the four extremes: the negation of existence/non-existence for the third extreme and that of non-existence/non-nonexistence for the fourth (cf. also MRP following). It is now known that the SMVK by Jitari is composed of exactly the same verses as the kk 21-28 as the JSS and that the SMVK k 8 is the same verse as the JSS k 28. The SMVK k 8 (P. [101] 5296 ha 64b7-8, P. [103] 5461 gi 209a6-7, P. [146] 5867 nyo 283a8) reads: yod min med min yod med min / gnyis kyi bdag nyid du yang med I mtha' bzhi dag las nges grol ba / dbu ma de nyid mkhas pa 1  'dod //  'ba'i [146]. The same verse is cited twice in the MRP (P. 342b3-4 and 345b3) by the second Bhavaviveka: yod min med min yod med min I gnyi ga min' pa ma yin te / 2  mtha bzhi las ni mam grol ba I de nyid dbu ma pa yis rig I ,}  'med 342  la 342  2  mu 342  3  In the MRP, the verse is attributed to Aryadeva, whereas in the SS it is attributed to Saraha. There is one commentary in the Tibetan non-Canonical works for the JSS by Mi pham contained in Collected Writings of • 'Jam-mgon 'Ju Gangtok, 1975, p. 297-323 (fol. 1-I3b4). See Mimaki (1982).  Mi-pham-rgya-mtsho, vol.11,  121 Those skilled in the Middle Way assert a truth definitely free from the four extremes which does not exist, does not not exist, does not both exist and not exist and is not something that is neither. [103b6] The master Kamalas'Ila also said in the Tattvaloka : The  self-realized  truth  ,  which  is  free  l  from  every  elaboration of being and non-being, etc., and is at peace, is taught as being that from which conventional designations (tha snyad, vyavahara), etc., have been removed. These quotations teach how ultimately (de kho naf) [things are] beyond existence and non-existence.  2  [104al]  All these  [quotations  teach  1) how  things  exist  conventionally, 2) ultimately do not exist and 3) are beyond being and non-being, etc.] as well as the reasons (rtags, linga) to be discussed [below] are to establish what the master [Nagarjuna] set forth [in the first lines of his Mulamadhyamika-karika]:^ [Homage to the of  cessation,  Buddha] who taught dependent-arising, which is free  production,  annihilation, permanence,  coming,  going,  diversity, unity and in which elaboration is stilled. [8. Diverse points of view of the Madhyamikas] 1. T A (D. 273a3-4, P. 312a4). The same verse was already cited in B S G T 101a2. 2. A visible blank here has been filled by the addition of de. 3. M M homage verse (PrasP 11.13-15, cf. tr. Slcherbatsky (1927) p. 91-2, Yamaguchi S (1947) p. 16, Inada (1970) p. 38-9, Sprung (1979) p. 35, Mimaki and May (1979) p. 460:  anirodham anutpadam anucchedam aSaSvatam / anekartham ananartham anagamam anirgamam II yah pratltya-samutpadam prapaficopaSamam s~ivam I deSayamasa sambuddhas tarn vande vadatam varam II  122 [8.1. Criticism of the attachment to emptiness as an absolute] [The  Buddha] said that those attached to non-existence are incurable  (bsgrub tu med pa, asadhya). Therefore, those who desire liberation should not take emptiness as an absolute (mtshan ma, nimitta). (k  XII-13)  [104a3] The Bhagavan said that Madhyamikas who are attached to emptiness (stong pa nyid, §unyata) are incurable (bsgrub tu med pa, asadhaya).  This is spoken of many times.  For example, in the  KaSyapaparivarta of the Ratnakuta Sutra:  1  Better the view of the transitory collection ('jig tshogs la lta  ba,  conceited  satkayadrsti) as solid as Mt. Sumeru than the person's  (mngon pa'i  nga  rgyal  can,  adhimanika) view of the emptiness of all dharmas.  It is  not a view, because views are removable but this [conceit] is incurable (gsor mi rung ba, acikitsya).  2  and, For example, a doctor gives a man a purgative to remove an illness (nad, roga). If, having cleared out the morbid elements  (nad,  dosa)3  [the purgative itself] is not  eliminated the cause of the illness (nad,  nidana) [will  1. Cf. KaSyapaparivarta 64: varam khalu puna KaSyapa Sumeru-matra pudgala-drstir asrit na tv evadhimanikasya S[u]nyata-drstimalina / tat kasmad dheto pudgala-drsti-gataham KaSyapa Sunyata nihsaranam S[u]nyata-drsti puna KaSyapa kena nihsarisyamtih /; (Tib.) srung gang zag tu lta ba ri rab tsam la gans pa bla'i I mngon pa'i nga rgyal can stong pa nyid du lta ba ni de lta ma yin no I de ci'i phyir zhe na / 'Od srung lta bar gyur pa thams cad las 'byung ba ni stong pa nyid yin na I 'Od srung gang stong pa nyid kho nar lta ba de ni gsor mi rung ngo zhes nges bshad do /. Cited in PrasP 248.9-11 (Tib. P. 96a2-3), M A I P. 157b6-7, where there is gang zag instead of 'jig tshogs. 2. Cf. PrasP 248.11 (Tib.) P. 96a3 3. For dosa as a technical term of Indian medicine see, for ex., L'Inde classique 1651, 1657.  123 remain] and there will not be recovery. In the same way, though emptiness is the best thing to get a person out of the slough of [wrong] views ('byin pa, nihsarana), if someone views emptiness as an absolute he is incurable. Thus spoke the Victor (rGyal ba, Jina). 1  [104a5]  And the master [Nagarjuna] also says in the Mulamadhyamaka-karika (XIII k 8): 2  All the Jinas (rGyal ba, Jina) have declared that emptiness gets one out of (nges par 'byin pa, nihsarana) all [false] views, [but] that those who view emptiness [as an absolute] are incurable (bsgrub tu med pa, asadhya). Also in the Ratnavali  (II  k 20):  3  1. Cf. KP 65 (with Tib. and Chin.): yatha hi vaidyo pumsasya dadyadd virecanam roga-vimgrahaya / uccalya dosaS ca na nihsareta tato nidanam ca na copaSanti / ...em eva drstl-gahanaSrtesu ya Sunyata nihsaranam param hi / sasu ....(incomplete)... Cf. PrasP 248.11-249.2 (ad MM XIII k 8): tadyatha KaSyapa glanah purusah syat tasmai vaidyo bhaisajyam dadyat tasya tad bhaisajyam sarvadosan uccarya svayam kostha-gata nihsaret/ tat kim manyase KaSyapa api tu sa purusas tato glanyan mukto bhavet// no hi bhagavan gadhataram tasya pumsasya glanyam bhaved yasya tad bhaisajyam sarvados uccarya kostha-gatam na nihsaretII bhagavan aha/evam eva KaSyapa sarvadrspkrtanam Sunyata nihsaranam / yasya khalu punah Sunyataiva drstis tarn aham acikitsyam iti vada (Tib.) P. 96a3-6. 2. MM XIII k 8 (cf. tr. Schayer (1931) p. 37, Wogihara (1938) p. 573, Inada (1970) p. 93, Sprung (1979) p. 150): Sunyata sarvadrstlnam prokta nihsaranam jinaih / yesam tu Sunyata-drstis tan asadhyan babhasire II 3. RA II k 20 (cf. tr. Tucci (1936) p. 243, Uryuzu (1974) p. 253): aparo 'py asya durjMnan murkhah pandita-manikah / pratiksepa-vinastatma yaty avicim adhomukhah/l Cited in PrasP 496.8-9 (cf. Tucci, ibid. n. 1). (Tib.) C. 120a, D. Illa7-bl, N. 129a, P. 134b; ed Hahn (1982).  124  Moreover, stupid because he apprehends this [Dharma] wrongly he is stupid and conceited in his learning and, through abandoning [Dharma], destroys himself and goes headfirst  (spyi tshugs, adhomukha) into the Avici Hell  (mnarmed pa). Also in the Lokatisastava (k 21):  1  You taught the nectar of emptiness to remove all [wrong ideas (kun rtog, samkalpa). You censure strongly anyone, who becomes attached to that. Therefore, the wise who desire liberation will completely remove the view which regards emptiness as an absolute (mtshan ma, nimitta).  [8.2 The object of negation for the Madhyamika] [104b2]  Examining the Madhyamikas' object of negation (dgag  par bya ba, pratisedhya). Here, the object of negation is just the belief in the truth [of appearances and emptiness] and not the appearances [themselves].  This is because [there is] suffering in that  1. Lokatisastava k 21 (cf. tr. LVP (1913) p. 14). The Skt. is found in B C A P 174.8-9 and 197.27-8:  sarva-samkalpa-hanaya Sunyatamrta-deSana I yasya tasyam api grahas tvayasav avasaditah /  125 [belief] and also because it is not necessary or possible [to negate] appearances. (k XII-14) [104b3] Since it is not necessary (dgos pa, prayojana) to negate mere appearances, or even possible to negate them, mere appearances are not the object of negation here. Therefore, believing in the truth of appearances and emptiness, which is the basis of all suffering, is the object of negation. As said in Jnanagarbha's Satyadvayavibhanga [karika (k 28)]:i We do not negate the appearance: negate anything that is experienced. Also in the  Madhyamakalankara [-karika  it is not logical to  (k 78) by Santaraksita] : 2  We do not negate as a self, what is, in essence, an appearance. Therefore the presentation of what is to be proved (bsgrub pa, sadhya) and a proof (sgrub pa, sadhana) is not disturbed. [104b5] The master Kamalas"ila also said in the  Tattvaloka : 3  Since the identities of conventional realities such as perceptibles, etc., (mthong ba, drfya) which have a conventional nature are not negated, direct perception (mngon sum, pratyaksa) is not excluded; the imaginary constructions (brtags pa, kalpita) are negated.  1. S D V K k 28 (C. 3 a l , D. 2b7), cilcd also in S D W (C. 12a6, D. 12a6). 2. M A k 78 (C. 55b6, D. 55b6, N. 47a4, P. 51b4), also cited in M A V (C. 75a2, D. 75a2-3, N. 69a7, P. 73b6-7). Our text has bsgrub pa and sgrub pa instead of sgrub pa and bsgrub bya like the M A : the order is inverted, but the meaning remains the same. 3. T A (D, 262a5 P. 297b5).  126 [104b6] Sriguptapstates:  Similarly, the  Madhyamaka-Tattvatara  [by  With the correct view (yang dag lta ba, samyag-drsti), even the most subtle things [i.e. partless particles] are not taken as objective support. So they say there is no selfnature, but not because of rejecting appearances. The master Santideva also said [in the  Bodhicaryavatara  (IX k 26)] : 2  Here, [that which] is seen, known or heard should not be negated. Here, the idea [of these] as true which is the cause of suffering should be removed. [105a2] The manner of negating [the belief in truly existent things is] also explained by the master [Nagarjuna] in the Vigrahavyavartani (k 23) : 3  Like a [magical] creation (sprul pa, nirmitaka) obstructing [another magical] creation; like an illusory man (sgyu 1. One of the antaraSloka of the Tattvavataravrtti (D. 41b3-4, P. 46b5).  Srigupta's  Tattvavatara is not in the bsTan 'gyur, but is completely conserved in the Tattvavataravrtti, the autocommentary.  There is a Japanese translation of the first, taken from the second, in  Ejima (1980) p. 219-21. 2. B C A IX k 26 (cf. tr. L V P (1907b) p. 116-7):  yatha drstam Srutam jfTatam naiveha pratisidhyate / satyatah kalpana tv atra duhkha-hetur nivaryate// 3. W k 23 (cf. tr. Yamaguchi S. (1929) p. 25, Bhattacharya K. (1978) p. 18, Kajiyama (1974) p. 152):  nhmitako nirmitakam mayapurusah svamayaya srstam I pratisedhayeta yadvat pratisedho 'yam tathaiva syat/l (Tib.) cf. ed. Tucci (1929) p. 27. There are variants. Here is the Tibetan Canon's version:  sprul pa yis ni sprul pa dang / sgyu ma yi ni skyes bu yis I sgyu mas phyung la 'gog byed kar I 'gog pa de yang de bzhin 'gyur II Similar verses:  M M X V I I kk 31-32 (cf Yamaguchi S. (1929) p. 65-6 n. 10 and Johnston  and Kunst (1951) p. 123 n. 1.  127 ma'i skyes bu, maya-purusa) obstructing another magical illusion, so is negation said [to function]. [8.3 Reasoning that negates the belief] Since what proves the two truths is the reasoning (rigs pa, yukti) that stops belief [in true existence], it is explained next. Since the variety of [things] are just appearances and are also empty of truth, they are like a dream and a magical illusion. [If they] do not appear or are true [they] are not a magical illusion. (k XII-15) [105a3]  The variety  of  [things] appearing as  apprehended  (gzung ba, grahya) and apprehender ('dzin pa, grahaka) are nothing more than mere appearance to that which is conscious of each of them, and since when they are analyzed they are also empty of truth, it is thus established that they appear like a dream and a magical illusion etc. and are without self-nature.  If they had a nature of truth, did not appear or  had a true self-nature they would not be like a magical illusion, since a magical illusion [merely] appears in consciousness and is devoid of real entity. [105a5] The Bhagavan also said: Form (gzugs, rupa) is like a dream and a magical illusion. Feeling  (tshor ba,  magical illusion.  1  1. Original source not identified.  vedana) is also like a dream and a  128 Also in the Lankavatarasutra (X k 130 = II k 148) : 1  The three existences are like imaginary falling hairs (skra shad, keSonduka), are a misconception, like mirage water (smig rgyu'i chu, maricyudaka), are like a dream (rmi lam,  svapna) and a magical illusion (sgyu ma, maya):  those who meditate thus will be liberated. [105a6]  The  master  [Nagarjuna]  also  stated  in  the  Mulamadhyamaka-karika (VII k 34) : 2  Like a dream, a magical illusion, a city of Gandharvas (dri za'igrong  khyer,  gandharavanagara):  it is taught that  production  (skye,  utpada), abiding (gnas, sthana) and  disintegration ('jig pa, bhanga) are like that. It is also stated in the Sunyatasaptati (k 66) : 3  1. Lank X k 130 ( = II k 148) (cf. tr. Izumi (1927) 55,161, Suzuki (1932) p. 83, Yasui (1976) 85, 252; 2 k 150 following them): keSonduka-prakhyam idam maricy-udaka-vibhramam' / tribhavam svapna-mayabham vibhavento vimucyate// 2  1  -vibhramat 2  2  -akhyam 2  (Tib.) P. 179al (=102b2). Lank X k 130 and II k 148 are almost identical in S k t , but are considerably different in Tibetan.  The comparison, of variants reveals that dBus pa bio gsal  consulted Lank X k 130. Compare with the Tibetan version Q?.) of the Lank II k 148: srid gsum 'di ni rmi lam dang / skra shad 'dzings dang sgyu ma 'dra I smig rgyu chu bzhin nor ba ni I rnam par bsgoms na mam par grol II 2. M M VTI k 34 (cf. tr. May (1959) p. 141, Yamaguchi S. (1949) p. 121-122, Inada (1970) p. 70): yatha maya yatha svapno gandharvanagaram yatha/ tathotpadas taiha sthanam tatha bhanga udahrtam II 3. SSap k 66 (C. 26b5, D. 26b4, N . 26a7-bl, P. 30a4-5; cf. Tib. ed. tr. Yamaguchi S. (1972), I p. 40; tr. Uryuzu (1974), p. 129); Lind'tner (1982), p. 65.  129 Composites ('du byed, samskara) are similar to a city of Gandharvas, a magical illusion, a mirage, imaginary falling hairs, froth (lbu pa, phena), a bubble (chu bur, budbuda), a magical creation (sprul pa, nirmita) a dream and a circle drawn by a firebrand (mgal me'i 'khor lo, alatacakra).  [105b2] These [citations] establish that conventional [truth] appears and is devoid of self-nature. [8.4 * The Madhyamika subject] [105b2]  [Objection:]  Since commonly appearing subjects (mthun par snang ba 'i chos can) do not exist [for the Madhyarnika] it is impossible [to present] a reason that establishes the two truths. [Response:] Just as mere sound free of specifications (khyad par, viiesana) is established, so mere appearance bereft of parts is not not established for both [debators]. [105b3] For example, for the two [debators of which one as Vaisesika] asserts that sound (sgra, §abda) is a quality of space [and the other as Buddhist asserts that it] is a secondary element, mere sound is established for both as a common appearance (mthun snang). Similarly, [for the Madhyamika] here also just the mere appearance bereft of true and false parts is established for both debators (rgol ba, vadiri) and is the subject (chos can dharmin); therefore there is no fault. If one says this is not true, one denies one's own experience and thus becomes even more stupid than a Lokayata ('Jig rten rgyang phan pa); one would also not be able to negate the wrong ideas of the opponents (pha rol po, para) by [exposing] the [internal] contradictions (nang 'gal) inherent in their assertion, due to the fact that the basis (gzhi, Mraya) of the specification would not be is established.  130 [105b5] It is also said by the master Kamalaslla [in the Tattvaloka] : 1  There would ensue the fault that in consequence a basis for the refutation of a permanent functioning thing and for a refutation of a quality (yon  tan, guna)  [substantially  different from a dravya], etc., would not be established if that [basis] were not present here.  There would be no  valid cognition and then in consequence the convention of it not being [permanent or substantially different] would not come about relative to this [basis].  And why wouldn't  this fault [of not being able to establish a given base as devoid of a wrongly-imputed specification] about even if there were a valid cognition?  also come  [It would since  you deny a basis common to both debators].  [105b6]  Therefore, the Madhyamikas, who teach all phenomena  to be devoid of an identity, assert as the subject a mere appearance which is known [by all], from the wise to the cowherds (gnag rdzi, gopala). For as stated in the Satyadvagavibhanga [-karika (kk 18-19) by JHanagarbha] : 2  To the extent that the appearance factor exists in the consciousnesses of the two  [debators],  to that extent,  depending on that, there is an understanding of subject and attribute, etc., [in common for the two]. inference. logicians  Otherwise, there is not. (rigs pa  smra, ba,  Then, there is an Therefore, if the  nyayavadin) use  such  [inferences], who could refute them?  1. T A (D. 260b7-261al, P. 296a3-5). 2. S D V K kk 18-19 (C. 2b3-4, D. 2b2-3), also cited in S D W (C. 9b5-6, D. 9b5-6). verses were already cited in B S G T 101b6-102al.  These  131  Also from the  Madhyamakalankara  [kk 76-7) by Santaraksita] : 1  Leaving aside particular subjects which crop up in doctrine, all these things which prove and are to be proved apply to things known [by all] from the wise to women and children. If that weren't the case, how could you respond with the answer "the basis is not established," (Mrayasiddha) etc., [when someone says the hare's horn is sharp because its a horn]. and also in the  Tattvaloka  [by Kamalaslla] : 2  Moreover, the actual thing is properly thought to be the actual thing which is the subject wherever it's from the point of view of being something. and: Since the experts of logic (rigs pa shes pa, yukti-jffa) always make as the subject something that is in conformity with the property, the wise [i.e., who understand Sunyata inferentially] speak of their own subject. The subject is something without self-being which is in conformity with the non-existent nature which is to be established in respect of it. Therefore, it is not [true] that the [subject of the thesis] is not established (ma grub pa, asiddha)3  1. M A kk 76-77 (C. 55M-5, D. 55b5-6, N. 47a3-4, P. 51b3-4), also cited in M A V (C. 75al2, D. 75al-2, N. 69a6-7, P. 73b5-6). Cf. M A P P. 130b2-6. 2. T A (D. 2 6 0 M , P. 295b2-3). 3. T A (D. 260b2-3, P. 295b4-5). The variant of our text for pada b of the first verse ought be noted.  132 [8.5. The object to be proved for the Madhyamika]: The author of the MadhyamakaSastra (Nagarjuna) and his commentators clearly assert that the mere negation of the object of negation, explained earlier, is the object to be proved, by the reasons that will be explained. (k [106a6] negation  of  XIM7)  [Nagarjuna and his successors] assert that the  above  mentioned  imaginary true,  the mere  permanent or  produced etc., is the object to be proved by the reasons that will be explained. As the master [Nagarjuna] said [in the Lokapariksa\ : l  Here, one asserts the negation of existence but [in doing so] does not establish non-existence. [106bl]  The  master  Bhavaviveka  also  said  [in  the  Madhyamakahrdaya (V k 106)] : 2  Therefore, even though the object of inference is not established  as  real;  [inference]  removes  what  is  not  concordant with (mi mthun phyogs, vipaksa) knowledge of reality. The master Jnanagarbha also [said in the Satyadvayavibhangakarika (kk  29-30)]3  :  The Lokapariksa ('Jig rten brtag pa) by Nagarjuna. Bhavaviveka cites the verse in PrajfTapradipa and Avalokitavrata specifies the title of the work in his Prajnapradlpatlka. Cf. -PrajlTapradipa (P. [95] (5253) tsha 114b4). 1.  question in his  2. M H V k 106 (D. 24a3, P. 26b7-8; cf. Tib. ed. tr. Yamaguchi S. (1964a) p. 38 from the end). The Skt. ms. reads:  ato 'numana-visayam na ta(()tvam pratipadyate I ta(t)tva-jHana-vipakso yas tasya lena nirakriya// 3. S D V K kk 29-30 (C. 3al-2, D. 2b7-3al), also cited in S D W (C. 12a7 and 12M-2, D. 12a7 and 12b2).  The comparison of variants demonstrates again that dBus pa bio gsal  133  The aspects of production, etc., which do not appear, but which others imagine as real, are negated. So it is right to negate just such imaginaries. This nonimaginary refutation hurts self alone. The master Kamalas'Ila also [said in the  Tattvaloka]:  In order to prove in detail the selflessness of all phenomena, we assert here again in a proper manner (Ji ltar rigs pas, yathayogam) the non-mistaken reasons. 1  and: Having got to the relationship [between the hetu and sadhya ] from those, through language [or conventions] and eliminated the imaginary entity as real, what else is there for fools to argue about? 2  [9. Reasonings that prove that phenomena do not have self-nature]  3  seems to have consulted the S D W to cite the S D V K .  Note that his text conserves the  ancient form (stsogs for sags) in S D V K k 29c. 1. T A (D. 261b6-7, P. 297a5-6). 2. T A (D. 262a6-7, P. 297b8). 3. Rubric 9. refers to the famous "four reasons". The term (mu bzhi skye 'gog la sogs pa'i gtan tshigs bzhi) appears in the M A S k 6 (P. 381a2, cf. Ejima (1980) p. 19,22 and Lindtner (1981) p. 200 n. 14 p. 206) of the second (?) Bhavaviveka. In Atisa's works they are clearly presented as 'four great reasons' (gtan tshigs chen po bzhi:  B M P P. 322a7-8, 323b3, 324al)  i.e., in the B P P (193-208) and in his pafTjika ( B M P P. 322a7-324al).  The 'four great  reasons are: 'mu bzhi skye ba 'gog pa'i gtan tshigs, catuskoty-utpada-pratisedha-hetu (= rubric 9.3), rdo rje gzegs pa'i gtan tshigs, vajra-kana-hetu (9.2), 'gcig dang du ma dang 2  bral ba'i gtan tshigs, ekaneka-viyoga-hctu (9.1) and *rten cing 'brel bar 'byung ba'i gtan tshigs, pratltyasamutpada-hetu (9.4). (1981) p. 205-11.  See Ejima (1980) p. 18-23, 30-32, 240-6; Lindtner  134 [9.1.  Reason:  dharmas are empty of being singularities or  pluralities] [106b4]  Now the explanation of reasonings that prove [that  phenomena are empty]. Since in their diversity things lack self-nature, singular or plural, ultimately, they are without a true self-nature, like [objects in] a dream.  (k XII-18)  [106b5] [Minor:]  These appearances of form and so forth lack  self- nature, singular or plural. [Major:]  That which lacks self-nature, singular or plural,  in reality, does not ultimately have true self-nature.  For  example, like a dream [object]. [Conclusion:  These things do not ultimately have true  self-nature] [This syllogism is based on the non-perception of the pervasion (khyab par byed pa mi dmigs pa, vyapakanupalabdhi). That a valid reason 1  must first have the mode of being a property of the subject (phyogs chos,  paksa-dharma) was established by the expert of logic (Rigs pa  mkhyen pa = Dharmakirti). As it is stated in the Pramanavarttika (IV k 22)2  :  Given that it [the subject, for example, sound] precedes the pervasion, even without it [the pervasion], just through  1. It is necessary to understand here the following pervasion (khyab pa vyapti): ultimately real nature is pervaded by having a singular or plural nature.  having an  That which  pervades (khyab par byed pa, vyapaka) is having a singular or plural nature. It is not found here; hence, what is pervaded by it is also not found. 2. P V I V k 22:  vyapti-purve vinapy asmat krtakah Sabda IdrSah I sarve 'nitya Hi prokte 'py arOwt tan-naSadhir bhavet II  135 saying,  'Sound is  a product; all  impermanent' that [sound] is  things like that  implicidy  understood  are as  disintegrating. Also in the Vadanyaya: Here,  [there  is]  no  determined  order  [for  the  premises] because there is no difference between the regard to  what one  wants  two two  [types of  premises] in  to  establish.  Haying established first that the subject exists,  then the pervasion is established. For example, [Minor:] Sound exists or is created. [Major:] Everything like that [i.e. that exists or is created] is impermanent, like a pot and so forth  1  and: On this [point] there is not even a slight determination, and in a particular demonstration, one  should formulate  the members [of the syllogism in any order].  2  1. V N 8.3-5 (cf. ed. Sankrtyayana (1935-6) 6.2-5): atrapi na kaScit kramaniyamah; istarthasiddher ubhayatraviiesat I dharmini' prak sallvam prasadhya paScad api vyaptih prasadhy eva I yathasan Sabdah krtako va, yaS caivam sa sarvo 'nityah, yatha--ghatadir iti I 1  yasmad dharmini Shastri; (Tib.) C 326a2-3, D. 327al-3, N. 384a7-b2, P. 365a6-8. 2. V N  103.7-8 (cf. ed. Sankrtyayana (1935-6) 108.5-6):  na hy  atra kaScit samayah  pratyayanaviSese'py evam evavayavah' prayoktavya iti I ' evavayavah Sankrtyayana, eva Sabdah (?) Shastri; (Tib.) C. 345b7-346al, D. 347b3-4, N. 408a4, P. 390a5-6.  136 [Being] one and [being] many is mutually  exclusive.  Since [being one and being many] is related in existence [being one] is negated by [something being] many [i.e., by its having parts].  Just that  [negation of being one]  establishes the negation of the plurality.  (k XII-19)  [107a4] Since being many is mutually exclusive with being one, something's appearance as many [i.e., having parts] negates its being one.  This is because the mutual exclusivity [between an hypothetical  one and many] is established.  Since being one and many are related in  existence, through the negation of one, many is negated, since the relation [between many and the hypothetical one it is built up on] is negated. [107a5]  Moreover, since these [objects] which appear in their  variety within a bifurcation into subject and object (gzung 'dzin, grahya-grahaka) have spatial and temporal parts, they do not exist as one.  Hence, they are also not established as a plurality which is a  composite.  Therefore, this reason is not unproven.  [107a6] It exists in the concordant example (mthun pa'i phyogs, sapaksa) and not in the disconcordant example (mi mthun pa'i phyogs, vipaksa); therefore, it does not prove the opposite ('gal ba, viruddha) and is not non-ascertained (ma nges pa, anaikantika). This [reasoning] shows emptiness (stong pa nyid, Sunyata) as one of the three doors of liberation (mam par thai pa'i sgo, vimoksamukha) [and consists in] 1  negating [thing's self-nature] after having analyzed the nature.  It is  1. For the three doors of liberation, see, for ex., the VimalaklrtinirdeSa (Tib. P. [34] (843) bu 228al-3) and AKBh 450.6-8 (ad A K VIII k 25) cf. tr. L V P (1971) viii p. 187): nirmalas  tu vimoksamukha-trayam (AK VIII k 25a -b) / anasravas tv ete trayah samadhyas trini vimoksamukhany ucyante I Sunyata vimoksamukham apranihitam ammittam vimoksamuk iti /; (Tib.) P. ngu 87b8-88al: de dag dri med rnam thar sgo gsum mo I ting nge 'dzin gsum po de dag dri ma med pa ni thar pa'i sgo yin pa'i phyir mam par thar pa'i sgo stong pa nyid dang I mam par thar pa 'i sgo smon pa med pa dang / mtshan ma med pa ste / mam par thar 1  pa'i sgo gsum zhes bya'o /. For other references, see L V P (1971) viii p. 187 n. 2.  137 stated in the Madhyamakalankara (k  [Assertion:]  These things talked about by ourselves and  others [i.e. opponants] do not have self-nature. [Reason:]  Because in reality they are not  singular or  plural. [Example:] Like a reflection (gzugs bmyan, pratibimba).  [9.2.  The reason which destroys like a diamond bit:  the impossibility  of production from the four causes]  That which is not produced from the four extremes is not produced.  This is a diamond bit (rdo  rje  gzegs ma,  vajrakana) which destroys the rock of the Substantialists (dNgos por smra pa, Bhavavadiri).  (k XII-20)  [107b2] [Major:] Whatever is free from being produced from self, other [things], both [self and other], and causelessly, in reality is not produced, like space. 1. M A k 1 (C. 53al, D. 53al-2, N . 44a5-6, P. 48b8), cited also in M A V (C. 56b6-7, D. 56b7, N. 48a6-7, P. 52b5-6). The Sanskrit is found in B C A P 173.17-8:  nihsvabhava ami bhavas tattvatah sva-paroditah / ekaneka-svabhavena viyogat pratibimba-vat / It is worth noting that Se ra rJe btsun pa Chos kyi rgyal mtshan (1469-1546) cites this verse  in his bsTan bcos mngon par rtogs pa 'i rgyan 'grel pa dang bcas pa 'i rnam bshad mam pa gnyis kyi dka' ba'i gnad gsal bar byed pa legs bshad skal bzang gin dbang gi rol mtsho (abridged title: sKabs dang po'i spyi don, Ser byes, 25a7-bl), but with the variant bden par med de instead of rang bzhin med de for pada c. This must be connected with the Tibetan interpretation  of  the  conventional truth :  Svatantrikas  and the  Prasafigikas concerning  the acceptance  of  the Svatantrikas assert as conventional truth that which is established  by its own nature - inherently-existent (rang bzhin gyis grub pa), -although they do not assert it as truly established (bden par grub pa); the Prasafigikas do not assert either on the plane of conventional truth. See the discussion, for ex., in L S .  138 [Minor:]  Appearances as things are also free from being  produced from self, other, both and causelessly. [Conclusion:  Appearances, in reality, are not produced.]  [This syllogism is based on] reasoning to what is not found based on the vyapaka (khyab par byed pa mi dmigs pa, vyapakanupalabdhi)} [107b3]  Among those [four extremes], production from self:  c  If the effect is not different from the cause [the effect] is established [as the cause]; therefore, production would be senseless.  Also, if [the effect] is not established it is not  logical that the effect is produced, because the cause would not exist.  2  [107b3]  Production from other and production from both.  If  cause  [the  is]  destroyed  [at  the  moment  of  the  production of the effect the destroyed cause] would not be able [ to produce an effect].  And even if [the cause is] not  destroyed [at the moment of production of the effect, it would follow that the cause exists] at the same time [as the effect].  Therefore, there is no cause so it is not [possible]  that the cause exists. both?  What is produced from other and  3  [107b4]  Causeless production  1. It is necessary to understand here the following pervasion: pervaded by a production from the four extremes. from the four extremes:  production in reality is  That which pervades is a production  it is that which is not perceived in these appearances. Hence, the  non-perception of that which pervades. 2. A verse from B S G T , chapter I V (Grangs can pa, Samkhya) 22a2. 3. According to the context, this must also be a verse from BSGT.  139 If there is no cause [it follows that] what is produced would happen a certain time and not at another [dependent on cause].  Even sharp thorns and so forth are produced  from their own causes.  [Your] example [to show causeless  production] is not established.  1  I have explained [all this] earlier in the refutation of the Sarnkhya and other [schools]; therefore, this reason [of the second syllogism, i.e., non-production from the four extremes] is not unproven.  This [reason]  exists in concordant examples (mthun pa'i phyogs, sapaksa) such as space, a reflection, etc., and is absent from discordant examples (mi mthun pa'i phyogs, vipaksa); therefore, it does not prove the opposite ('gal  ba,  viruddha) and also is not non-ascertained (ma nges pa,  anaikantika). [107b6]  With this reason, called "The Diamond Bit" (rdo rje  gzegs ma, vajra-kana) the rock mountain of the Substantialists (dNgos por smra ba, Bhavavadin) is entirely destroyed.  As KamalaSila stated  [in the Tattvaloka] : 2  It is the subtle diamond (rdo rje phra mo, suksma-vajra) since it entirely destroys the rock of things that ourselves and others discuss and since it is not even turned back by all four extremes.  [108al]  This [reasoning] shows [the meditation of] signlessness  (mtshan ma med ba, animitta) as one of the three doors of liberation and  consists  in negating  [the  self-nature of  things]  after having  analyzed the cause [of things]. The master [Nagarjuna] also said in the Mulamadhyamika-karika (1 k l) : 3  1. A verse from B S G T , chapter III ('Jig rten rgyang phan pa, Lokayata) 12b3. 2. T A (D. 2 6 7 M , P. 304b5-6). 3. M M I k 1 (cf. tr. Stcherbatsky (1972) p. 71, 93, Yamaguchi S. (1947) p. 18, Inada (1970) p. 39, Sprung (1979) p. 36):  140  There is never production anywhere, of any thing, from self, from other, from both, or without cause.  [9.3. Reason: the impossibility of the production of an effect]  These external and internal things, in reality, do not arise, since an existent or non-existent result is devoid of coming into being, like space.  (k XII-21)  [108a3] [Major:]  The existent or non-existent effect, empty of  coming into being, in reality, does not arise, like space. [Minor:]  These appearances as external and internal  things are also empty of coming into being as existent or non-existent results. [Conclusion:  These  appearances,  in reality, are not  produced.] [This syllogism is based on] the reasoning to what is not found based on the vyapaka (khyab par byed pa mi dmigs pa, vyapakahupalabdhi). [108a4]  1  Thus, an existent effect does not arise, because it is  already established.  A non-existent effect does not arise either, because  it is a non-thing.  Something non-established becoming established  does not arise either since there does not exist another [category] over and above existent effect and non-existent effect; therefore, this reason na svato napi parato na dvabhyam napy ahetutah / utpanna jatu vidyante bhavah kva cana ke canal 1. It is necessary to understand here the following pervasion: pervaded by the coming into being of an effect being of an effect: that which pervades.  production in reality is  That which pervades is the coming into  it is not perceived in these appearances; hence the non-perception of In other words, the coming into being associated with an effect is  vyapaka, or pervader of coming into being.  This is one reason to the non-findability of  what is pervaded by the absence of what pervades.  141 is not unproved.  Since the pervasion (khyab pa, vyapti) is established  as above [where examples are in the concordant example and not in the discordant example], it does not prove the opposite ('gal ba, virodha) nor is it non-ascertained (ma nges pa, anaikantika). [108a5] This [reasoning] shows [the meditation of] wishlessness (smon pa med pa, apranihita) [as one of the three] doors of liberation and  consists in negating  analyzed the effects  [the self-nature  [of things].  of things] after having  It is also said by the master  [Nagarjuna] in the Sunyatasaptati (k 4ab) : 1  Existence is not produced, because it exists [already]. Non-existence is not produced, because it is non-existent.  [9.4. Reason: dependent-arising]  Wherever has an immutable self-nature could not possibly be a dependent-arising.  A l l things do not have self-nature,  because they are dependent-arisings, like [for example] a reflection. 1. SSap k 4ab (C. 2 4 b l , D. 24b 1, N. 24a6, P. 27a7; cf. tr. Yamaguchi S. (1972) p. 15, Uryuzu (1974) p. 92). Whereas SSap k 66 cited earlier in B S G T 105M-2 corresponds well enough to its canonical version, SSap k 4ab cited here differs considerably, and corresponds more to that cited in the commentaries, although even there it does not correspond literally.  First,  the canonical version of the SSap k 4ab:  yod phyir yod pa skye min te / med phyir med pa' skye ma yin /. 'pas N P Next the version in the commentaries:  a) "auto-commentary" (C. tsa 109b6, D. (3831) tsa  110b3, N . (3222) tsa 117a5, P. [95] (5231) tsa 126b5); commentary by Candraklrti that cites three times the verse in question: b) C. ya 281a3, D. (3867) ya 285a4, N. (3259) ya 315a7, P. [99] (5268) ya 325b6, c) C. 292b6, D. 296a7, N . 328a6, P. 338al, d) C . 316b6, D. 321a6-7, N. 357b3, P. 364b6:  yod pa yod phyir mi skye la' / mcd pa med phyir me skye zhing'/ 2  'skye ma yin a), skye ba min b) c),  2  / a) [NP]  'phyir mi skye la b)c), pa'i phyir ma yin a)  142 (k XII-22)  [108bl] [Major:]  Whatever arises in dependence, in reality, does  not have self-nature, like a reflection. [Minor:]  These appearances as things which are the basis  of the debate (rtsod pa'i gzhir gyur pa, vivadaspadibhuta) arise in dependence. [Conclusion:  These appearances do not have self-nature]  [This syllogism is based on] reasoning to what is not found based on the vyapaka  (khyab par  viruddhopalabdhi). [108b2]  byed pa  'gal  ba  dmigs pa, vyapaka-  1  Similarly, transmigratory existence ('khor ba, samsara)  is a dependent-arising of ignorance (ma rig pa, avidya), for it is stated in the Prajnaparamita-sutra  (Yum) '2  The Sugata (bDe bar gshegs pa) said that however many sentient beings there are - inferior, middling and supreme, all originate from ignorance.  Liberation (mya ngan las 'das pa, nirvana) is a dependent-arising of the perfection  of  wisdom  (shes rap kyi  pha rol  tu phyin pa,  prajnaparamita): It is also stated in the same [Sutra]!: 1. It is necessary to understand here the following pervasion: having self-nature in reality is pervaded by not being produced by conditions.  That which pervades is not being produced  by conditions, and the opposite of that which pervades is therefore production by conditions: it is just that which is perceived in these appearances; hence the perception of the opposite of that which pervades. 2. R G S G X X V f f l k 5ab (cf. tr. Conze (1973) p. 61):  yavanta sattva mrdu-madhyam'-ukrsta loke sarve avidya-prabhava sugatena uktah / 3 R G S G XXV111 k 6ab (cf. tr. Conze (1973) p. 61):  yavanti jffana-naya-dvara upaya-mulah sarve ti prajfla- vara-paramila-prasutah /  143  As many roots, methods, modes and doors of wisdom as there are, all have arisen from the perfection of wisdom.  [108b3]  The  [positive]  pervasion  1  [of the syllogism] is  established on a double moon (zla gnyis, dvi-candra), reflection (gzugs brnyan, pratibimba) etc.  [The negative pervasion of the syllogism, i.e.,]  the negation of dependent-arising pervades the three characteristics of 2  a self-nature and is also established on space.  The master (Nagarjuna)  stated [in the Mulamadhyamakakarika (XV k 2cd and 8cd)]:  Self-nature (rang bzhin, svabhava) is not fabricated (bcos min, akrtrima) and does not depend on other [things]. It 3  is never possible for self-nature to change.  4  [108b5] of  By this [reasoning], [we] prove that all the phenomena  transmigratory  existence  and liberation  nirvana) do not have self-nature.  ('khor  'das, samsara-  Moreover, it is stated in the  Anavataptanagarajapariprccha-[mahayana-] sutra : 5  According to the reading of our text and of the Canon the compound 'upaya-mula' is taken as a dvamdva (thabs dang rtsa ba), contrary to Conze's translation:  " A s many roots of  skilful devices as there are,...". But the reading of the M s T H (thabs kyi rtsa ba:: recension A after Yuyama (1976) p. 110 and 185) justifies his translation. 1. The positive pervasion of the syllogism: that which is produced by conditions does not have self-nature.  The negative pervasion of the syllogism: that which has self-nature is not  produced by conditions. 2. For the three characteristics of self-nature, see the verses of the M M (XV* k 2cd and 8cd) that are be cited immediately below. 3. M M X V k 2cd (cf. tr. Schayer (1931) p. 62, Wogihara (1938), p. 584, Inada (1970) p. 98, Nagao (1967) p. 274, Sprung (1979) p. 154):  akrtrimah svabhavo hi nirapeksah paratra call 4. M M X V k 8cd (cf. tr. Schayer (1931) p. 72, Wogihara (1938) p. 589, Inada (1970) p. 99, Nagao (1967) p. 181, Sprung (1979) p. 160):  prakrter anyathabhavo na hi jatupapadyate II 5. Anavataptanagarajapariprccha (D (156) pha 230b2-3, P. [33] (823) pu 238a6):  144 Whatever is produced by conditions is [in fact] not produced, and does not have a product's self-nature. Whatever depends on conditions is said to be empty; whoever understands emptiness is cautious (bag yod, 1  apramatta).  The master [Nagarjuna] also said in the k 16)2;  Mulamadhyamaka-karika  (VII  rkyen las skyes pa gang yin de ma skyes It de la skye ba ngo bo nyid kyis 1  med It rkyen la rag las gan yin stong par gsungs It stong nyid gang shes de ni bag yod pa'o It 1  de D, te P  2  // D, / P  It is evident that our author did not cite this verse directly from the Canon. The Sanskrit is found in a , PrasP 239.10-13 (ad M M XIII k 2), b. id. 491.11-14 (ad M M X X I V k 7), c. id. 500.7-10 (ad M M X X I V k 14), d. id. 504.1-4 (ad M M X X I V k 18), e. B C A P 172.11-14 (ad B C A I X k 2), /. SS 395.22-396.1:  yah pratyayair jayati sa hy ajatd ni tasya utpada-svabhavd 'sti / yah pratyayadhinu sa Sunya ukto yah Sunyatam janatf so 'pramattah II 'ajata f no a 'utpadu svabhavo abde, utpada svabhavo c, 2  'utpada evasya bhavet svabhavat f 'Sunyu be "jaftati f This verse is cited also in MavBh 229.2-5, MAT [P] 163a3-4 and L R C M kha 40al-2 (cf.tr. Nagao (1954) p. 184, Wayman (1978) p. 245), 107b6-7 (cf. tr. Nagao p. 346, Wayman p. 380). 1. On the text and interpretation of the second pada, see L V P (1932-3a) p. 74, 93. See also May (1959) p. 224 and n. 770. 2. M M VII k 16 (cf. tr. May (1959) p. 124-5, Yamaguchi S. (1949) II p. 90, Inada (1970) p. 67:  pratltya yad yad bhavati tat tac chantam svabhavatah / tasmad utpadyamanam ca Santam utpattir eva ca I Except for pada b, the Tibetan  version of the Canon is considerably different from our  text; it follows:  rten cing 'byung ba gang yin pa I de ni ngo bo nyid kyis zhi I de'i phyir skye bzhin nyid dang nil skye ba yang ni zhi ba nyid I  145  Whatever arises  in dependence  is  naturally at peace.  Consequently, produced and producer are also at peace.  [10. Emptiness itself is not truly existent]  [108b6] Negating the true existence of emptiness:  Since the negated refutation is not true, the negation also has no truth.  Hence, all phenomena are pacified of the  elaborations of existing and not existing.  (k XII-23)  [109al] Ultimately, the emptiness and non-production etc., which are a negation of true existence, are also not true, because the object of negation which is empty is not true. Furthermore, a negation lacking an object [negated] is not logical.  The master [Nagarjuna] also stated in  the Mulamadhyamaka-karika (XVIII k 7cd)  1:  The non-void does not exist at all so how could there be a void?  [109a3]  Also in [Jnanagarbha's] Satyadvayavibhanga [-karika  (kk 9cd-10ab)]2  :  < » s  Since the object of negation does not exist, it is clear that in reality, the negation does not exist.  How could a  1. M M XIII k 7cd (cf. tr. Schayer (1931) p. 36, Wogihara (1938) p. 572, Inada (1970) p. 93, Sprung (1979) p. 149):  na kimcid asty a&unyam ca kutah Sunyam bhavisyatil 2. SDVK kk 9cd-10ab (C. 2a3-4, D. 2a2-3), cited also in S D W (C. 6a2-4, D. 6a2-4). k lOab was already cited in BSGT 99b3. Contrary to other examples concerning this text, the comparison of variants indicates that our text here is closer to the SDVK that the S D W .  146 negation  of  something  in  essence  a  conceptual  construction not [itself] be a conceptual construction?  [Santaraksita] also said in the Madhyamakalankara  (kk 7lab and  72ab)i:  Since  production and so  forth does not  production and so forth is impossible.  exist, non-  Where there is no  object the affixing of a negation is not proper.  [109a4] Therefore, all phenomena pacified of all elaborations of existence, non-existence, both, and neither is the ultimate [truth]. was explained earlier.  [11.]  This  2  Intermediary verse:  Where there is non-existent production, there is neither abiding  (gnas pa,  sthana),  negation,  permanence, going, coming, one nor many.  [12.  Refutation of objections re:  annihilation, (AS XII-9)  the Madhyarnika thesis that  phenomena do not have self-nature]  [109a5]  These [reasons] prove that the ultimate [truth] is the  absence of self-nature.  Now, in order to remove objections (rtsod pa,  vivada) to the mode [of proving] absence of self-nature [we present first the opponent's objections and then our refutations].  1. MA kk 71ab and 72ab (C. 55b2-3, D. 55b2-3, N. 46b7-47al, P. 51a8-bl), cited also in MAV (C. 73a4-5, D. 73a5, N. 67a4-5, P. 71b3-4. 2. Cf. BSGT 99a5-b2.  147 [12.1. Objections based on Agama and the Madhyamika response] [12.1.1. General objection] [12.1.l.a. Objection]  [Objection: established  The entitylessness of all phenomena is not in  Agama, since  a)  no  one  accepts  [all  statements to be true], b) they are irrelevant, c) or are understood by way of a hidden intention.  (k XII-24)  [109a6] Some of our own schools (rang gi sde pa, sva-yuthya) use the following objection:  1  first of all, that all phenomena lack self-  nature is not established in Agama, since a) no one accepts them to be true; b) the statements merely accord with the beliefs [of certain misguided  trainees]  and  are  irrevelant  as  [statements  of]  fact.  Moreover, since c) when a certain Sutra says : 2  All  phenomena  are  entityless,  non-produced,  pacified  from the beginning and are naturally in a state of nirvana, it should be understood as requiring interpretation (drang ba'i don, neyartha) so the teaching of the Bhagavan is doctrine understood in terms of various hidden intentions (dgongs pa, abhipraya). [12.1.l.b. Madhyamika response]  [109b2]  [Response:] In order to answer [these objections]:  1. Principally the Yogacara. 2. SNS VII 9 (cf. tr. Lamotte (1935) p. 195).  An almost identical passage is cited in BSGT  llla5 with the designation of Samdmnirmocanasutra. For similar passages in other Sutras, see Lamotte (1935) p. 193, n. 2.  Futhermore, in the MAI by KamalasTla (P. 144al-2), the  same passage is cited with the designation "mdo kha cig las" as in our text. And seen that our author, accords considerable importance to KamalasTla's philosophy (BSGT  113al;  BSGT 126a6-bl) and that the structure of our text from rubric 12 (BSGT 109a5 and following) seems to be inspired by the MAI (cf. synopsis of the MAM in Ejima (1980) p. 228-30), it is possible that dBus pa bio gsal cites the passage in question from the MAl, and not directly from the SNS.  148  These foolish reasons are not established because a) the wise accept [such Agama to be true] b) a trustworthy person's statements (yid ches kyi tshig, apta-vada) are not deceptive [with regard to the meaning] (slu ba med pa, avisamvada) c) and they [the Words that explain the absence of self-nature] teach the ultimate [truth] itself. (k XII-25) [109b3] The first reason which you have stated is unproved (ma grub pa, asiddha) because the superior individuals (skyes bu dam pa, saj-jana) wise in ideas assert that the Buddha's word is good in the beginning, middle and end, like fine gold that can stand up to a triple analysis.  1  [109b4]  [Your]  second reason is  also  unproved,  since a  trustworthy person's statements are not deceptive with regard to the meaning.  As the expert of logic (Rigs pa rnkhyen pa = Dharmaldrti)  said [in the Pramanavarttika (I k 216)] : 2  Since a trustworthy person's statement (yid ches tshig, apta-vada) is universal to the point that it does not change [experience] even if its object if a hidden phenomenon and there is not a category [to classify it], this mind (a trustworthy person's statement) is declared inference (rjes su dpag pa, anumana) [by Dignaga ]. 3  1. Three analyses, i.e., burning (bsregs pa, tapa), cutting (bead pa, cheda) and rubbing (bdar ba, nikasa).  This is a frequent example: see, for ex.,  MAI [P] 160a7, JSS k 31 = TS k  3587 cited also in TSP 15.23-4. 2. PV I k 216:  apta-vadavisamvada-samanyad  anumanata /  buddher agatyabhigita parokse 'py asya gocare II 3. P W 324.12: ...acarya-Dignagabhihita... Here it the PS II k 5 (cf. tr. Kitagawa (1965) p.  92), cf. PS (K) P. 4b2: yid ches tshig ni mi bslu ba I spyi las rjes su dpag pa nyid /. The Sanskrit, is found in NVT 205.17 (cf. Randle (1926) Fragment E) and SV Sabdapariccheda  149  [109b5] The third reason is also unproved because the Scriptures (gsung  rab, pravacana)  ultimate [truth].  which teach selflessness and so forth teach  As it is asid in the  Sarvabuddhavisayavatara-  Jnahalokalamkara-sutra : 1  Anything definitive in meaning (nges pa'i don, nitartha) is the ultimate [truth].  And also in the  Aksayamatinirdes'a : 2  k 23cd: apta-vakyavisamvada-samanyad anumanata // 1. Saivabuddhavisayavataya-Jna^alokMamka^a-na^a-mahayMasuua (D. (100) ga 297b2, P. [28] (768) khu 325b2).  This passage is cited, for ex., in MA1[P] 162a6-7 and L R C M kha  5a2 (cf. tr. Nagao (1954) p. 107, Wayman (1978) p. 179).  The comparison of the variants  indicates that our author has probably cited this passge from the M A I , and not direcdy from the Canon. It may be added that Tsong kha pa cites this passage in the same form as in our text. See also the following note. 2. Cf. A M N (P. [34] (842) bu 156a5-7):  mdo sde gang dag stong pa nyid dang / mtshan ma med pa dang / smpn pa med pa dang I mngon par 'du mi byed pa dang / ma skyes pa dang / ma byung ba dang I dngos po med pa dang I bdag med pa dang I sems can med pa dang I srog med pa dang / gang zag med pa dang / bdag po med pa dang I rnam par thar pa 'i sgo 'i bar du bstan pa de dag ni nges pa 'i don ces bya ste I. Cf. PrasP 43.4-9:  uktam  caryaksaysmatisutre... / yavad  ye  sutrantah  Sunyatanimittapranihitanabhisamska^ajatanutpMabhava nirjivanihpudgalasvamika-vimoksamukha nirdistah ta ucyante mtarthah/. Cf. also L R C M kha 5a2 (tr. Nagao 107, Wayman 179). As can be clearly seen, our author does not cite the passage direcdy from the Canon. It is quite probable that he was inspired by the M A I (D. 149b5-6, P. 162a7):  skye ba med pa  la sogs pa yang 'phags pa bio gros mi zad pas bstan pa las I nges pa 'i don to zhes bstan te /. Tsong kha pa cites the same passage from the M A I in his L R C M (kha 5a2; tr. Nagao 107, Wayman 179). But it may be noted that in the L R C M the citation from the M A I goes from 'de lta bas na' ( L R C M 4b8; M A I D 149b4, P. 162a5) until 'nges so' ( L R C M 5a3; M A I D. 149b6, P. 162a8), and not to 'don yin no' ( L R C M 5al) as Nagao (id. p. 107 and Wayman  150  [Sutras which teach] non-production and so forth are also definitive in meaning.  The master Kamalas"ila stated the same [in the Madhyamakaloka] : 1  We do not cite as authoritative the words of the Bhagavan in order to prove [the absence of self-nature] to the nonBuddhists (mu stegs can, ththika) who do not follow the scriptures of the Tathagata.  However, we present it as a  scriptural authority [for our view] to those who accept Agama, when analyzing what the Agama means.  [12.1.2. Objections of the Vaibhasika and the Yogacara] [12.1.2.a. Objections]  Some  (=Vaibhasikas)  [object]:  [the  theory  of  the  Madhyamikas concerning] the lack of self-nature is not [correct] because [the Bhagavan] set forth sources (skye mched,  ayatana) such as forms (gsugs, rupa). Others  (=Yogacara) say [that absence of self-nature] contradicts the Sutra which teaches consciousness as ultimate [truth], [k XII-26]  [110a3] The Vaibhasikas assert that it is not correct to say that all  phenomena do not inherently exist, because  Bhagavan's  statement  about  sources  such  as  it contradicts the forms.  As  the  Abhidharmasutra (Chos mngon pa'i mdo) says : 2  (id. p. 179) thought 1. M A I (D. 148b5-6, P. 161a3-4).  2. Cf. A K B h 24.3-5 (ad A K I k 35, cf. tr. L V P (1971) i p. 65): uktam ca Sutre caksur bhikso adhyatmikam ayatanam catvari mahabhutany upadaya rupa-prasado rupy anidarSa  151  What is the eye? O Bhiksu, it is made from the four great elements, invisible and undemonstrable yet obstructive.  [110a4] Similarly, the Yogacara-Vijnanavadins [assert] that it is not [true] that all phenomena lack inherent existence, because many  Lankavatarasutra,  Sutras of definitive meaning, such as the  Samdhinirmocanasutra,  the  Dasabhumika, etc. say that consciousness free of subject-object duality (gzung ba dang 'dzin pa gnyis med pa'i shes pa, grahya-grahakadvaya-jnana) ultimately exist. Therefore, there is a hidden intention (dgongs pa can, abhprayika) behind [the Buddha] saying in the vast basket (shin tu rgyaspa'i sde snod) that all the  1  phenomena lack inherent existence..  [12.1.2.b. Madhyamika response]  [110a6] These objections are not correct, because,  [The Bhagavan] spoke of sources such as forms relative to other [beings' awareness], and not in reality.  [He also]  declared "mind only" in order that [beings] might enter into the middle way  (dbu ma'i lam, madhyama pratipad). (k XII-27)  [llObl] There is no fault [in our thesis].  In order to lead beings,  who, due to the power of predispositions are mistaken with regard to subject and object duality [the Bhagavan] spoke about sources such as sapratigham...; (Tib. P. gu 46b5-6): mdo las kyang dge slong mig ni nang gi skye mched yin te I 'byung ba chen po bzhi dag rgyur byas pa gzugs dang gzugs can bstan du med la thogs dang bcas pa stel 1. See also B S G T X I 83a4-5.  The Mahayana Sutras in general and the Prajffaparamita in  particular would probably be meant by 'Sin tu rgyas pa'i sde snod" (vaipulyapitaka).  152 forms to conform with these [beings] and not as definitive meanings. As the master Vasubandu has stated [in the Vimgatika (k 8)] : 1  If sources such as forms have been spoken about, it was with a hidden intention (dgongs pa'i dbang gis, abhiprayavaSat), for trainees ('dul ba yi skye bo, vineyajana), just as  [he has spoken about] a living creature that comes into being spontaneously (brdzus te byung ba'i sems can, upapaduka-saitva).  [110b2] Moreover, the extensive Lankavatarasutra (X k 153c) that:  teaching  in  the  2  The object does not exist; only mind [exists]. and in the Da^abhumikasutra^ that: These three realms are only mind. is in order [that beings] will, having turned from the belief in real outer 1. VS k 8 (cf. tr. Levi (1925) p. 49, Frauwallner (1969) p. 371, Kajuyama (1976a) p. 13-14): rupady-ayatanastitvam tad-vineyajanam prati / abhipraya-vaSad uktam upapaduka-sattva-vat// (Tib.) D. 3 b l ; cf. (Tib./Chin.) ed Sasaki G . (1022) p. 4. Verse already cited in B S G T 83a4. For spontaneous generation, see Levi (1925) p. 49 n. 1. 2. Lank X k 153d (cf. tr. Izumi (1927) p. 163, Suzuki (1932) p. 238, Yasui (1976) p. 255): nasty artham cittam eva tu //. (Tib.) P. 180a8: don ni sems ni 'ba' zhig go II. Cited in M A V P. 61b8 and MAI P. 157a5. Whereas the Tibetan form of this verse of our text is rather different from that of the Canon, it is exactly the same as that of M A V and MAI:  it is quite probable that dBus pa bio gsal cites this verse from these treatises.  3. D B h 49 E: citta-matram idam . yad idam traidhatukam /.  Cf. Vs"V 3.2-3 (cf. tr. Levi  (1925) p. 43): citta-matram bho jinaputra yad uta traidhatukam iti Sutrat /; Siddhi p. 420. The same passage was already cited in B S G T X I 82b3.  153 objects, enter the Middle Path. It is not to show that consciousness free from the two [agent and object] as ultimate [truth]. For as it is said in the Lankavatarasutra (X kk 256-8) : 1  By relying on "mind-only" [the there are the outer objects.  yogi] does not suppose  When based in reality,  [he]  will pass beyond even "mind-only". Having gone beyond "mind-only", [he]  must go beyond  [consciousness] without an appearance [of duality].  2  subject-object  The yogi who abides [in the consciousness] free  of appearance where [it, i.e.,  this consciousness without'  1. Lank X kk 256-258 (cf. tr. Izumi (1927) p. 171, Suzuki (1932) p. 246-7, Yasui (1976) p. 266):  cittamatram samaruhya bahyam artham na kalpayet/ tathatalambane sthitva cittamatram atikramet I cittamatram atikramya nirabhasam atikramet I nirabhasa-sthito yogi mahayanam sa paSyati I anabhoga-gatih Santa pranidhanair viSodhita/ jfianam anatmakam Srestham nirabhase na paSyati// (Tib.) P. 184a3-5.  Cited in treatises such as M A V P. 79M-4, M A I P. 171a 2-4, B h K I  (Skt.) 210.9-14: (Tib.) 259.3-14, B h K N D. 4al-2, P. 48a8-bl (only the first two verses), P P U P. 183a8-b2, T K h (P. [145] (5847) 253a5-7=PT n 814 fol. 6b3-5: K 256-7 only cited). Verse 258cd was already cited in B S G T 98b3.  dBus pa bio gsal does not cite these  verses directly from the Lank, and most likely cites them from the M A V or the M A l . The Tibetan version for these three verses reads:  sems tsam la ni gnas nas ni / phyi rol don la mi brtag go I yang dag dmigs la gnas nas ni / sems tsam las ni 'da' bar bya I sems tsam las ni 'das nas ni / snang ba med las 'da' bas bya / mal 'byor snang ba med gnas na I theg pa chen po mi mthong ngo I lhun gyis grub rtogs zhi ba ste I smon lam dag gis mam par sbyangs I bdag med ye shes mchog yin te / snang ba med tshe mi mthong no I 2. Translated following Kamalaslla's interpretation.  Cf. B h K I 211.10-2:  evam cittamatram atikramya tad api dvaya-nirabhasam yaj jhanam tad atikramet I.  •  .  .  154 appearance of the two does not appear] beholds the Great 1  Vehicle. The  spontaneous  state (lhun  gyis  grub,  pacified and purified through prayer.  anabhoga)  is  He does not see in  the [consciousness] free from subject-object the supreme wisdom free of Self  2  Also, in the [Sarvabuddhavisayavatara]  Jnanalokalamkara  [-sutra] : 3  The Buddhas know the Dharma without ever finding any object or mind.  Homage to [the Buddhas] who have no  objective support.  [12.1.3.  Yogacara objection:  interpretive  the Madhyamika's thesis is only an  teaching]  [12.1.3.a. Objection:]  The  hidden  inherent  intention  existence  of  behind  [teaching]  all phenomena  entitylessnesses (ngo bo nyid  med,  the lack of is  the three  nihsvabhava).  (k XII 1. Translated following Kamalasila's interpretation. Cf. B h K I 211.19-20: tatha  cadvaya-jhaha-nirabhase  sthitatvat  jHane  yada  sthito  yogi  tada  paramatattve  mahayanam sa paSyati /.  2. The two last verses already cited in B S G T 98b3. 3. Sarvabuddhavisayavata^a-JMnalokalamkaya-na^a-mahaymasutra  (D. (100) ga 300b2, P.  [28] (768) khu 328b6-7): sangs rgyas mams kyis' lan 'ga' yang / sems dang chos kun gtan mi dmigs  I chos mams thams cad kun mkhyen pa I mi rten khyod la phyag ' kyi P  1  'tsal lo I  ma D  This verse is cited, for ex., in M A I [P] 171a7-8. The comparison of variants shows that our author probably cited it from the MAT, and not from the Canon.  155 - 28ab)  [11 lal] As it is said in the Samdhinirmocanasutra : 1  My hidden intention, in teaching the entitylessness of all phenomena  is  the  three an  entitylessnesses ; i.e., [that  inherently  2  imaginaries]  lack  existent  defining  characteristic  (mtshan nyid ngo bo nyid med pa, laksana-  nihsvabhava); [that dependent-natures] lack an inherently existing production (skye ba ngo bo nyid med pa, utpattinihsvabhava), and [that the thoroughly established nature] lacks an ultimate inherent existence (don dam pa ngo bo nyid med pa, paramartha-nihsvabhava).  The entitylessness of all dharmas is thus [i.e. a teaching requiring interpretation].  Thus object the Yogacara.  [12.1.3.b. Madhyarnika response]  This  [teaching, i.e., the triple entitylessness] was also  taught to [enable beings] to enter into the Madhyamika path free from the [two] extremes.  [llla2]  (k XII-28cd)  To the extent that a thoroughly imagined character (kun  btags pa'i mtshan nyid, parikalpita-laksana) such as permanence, etc., is not established of conventional [things] which are like magical 1. SNS VII III (cf. tr. Lamotte (1935) p. 193):  Don dam yang dag 'phags ngas chos mams kyi ngo bo nyid mam pa gsum po 'di lta ste / mtshan nyid ngo bo nyid med pa nyid dang / skye ba ngo bo nyid med pa nyid dang / don dam pa ngo bo nyid med pa nyid las dgangs nas chos thams cad ngo bo nyid med pa 'o zhes bstan to /. The Canon's text is slightly different than our text. Cf. SNS VII 8. 2. On the triple entitylessness, see for ex. Siddhi p. 556.  156  illusions, there is the entitylessness of character (mtshan nyid ngo bo nyid med pa, laksana-nihsvabhava). When a dependent-arising's [other-powered nature] (gzhan gyi dbang, paratantra) is analyzed, [it is understood as being] non-produced; therefore there is the entitylessness of production (sieve ba ngo bo nyid med pa, utpatti-nihsvabhava). All phenomena are established ultimately lacking an inherendy existing nature - this is thoroughly established entitylessness (don dam pa ngo bo nyid med pa, paramartha-nihsvabhava).  [Illa4] Thus, since by teaching the hidden intention of the three entitylessnesses, it teaches the Middle Path free from the two extremes through revealing , the Samdhinirmocanasutra is established as a text of definitive meaning. As it is said in the same Samdhinirmocanasutra : 1  Moreover, Paramarthasamudgata (Don dam yang dag 'phags), even in referring to ultimate entitylessness, delineated by the selflessness of all phenomena, I explain that all dharmas are not produced, not destroyed, pacified from the beginning, and by nature completely in [a state of] nirvana. Similarly, in the  Lankavatarasutra:  When one investigates with intelligence (bio, buddhi), there is neither dependent nature (gzhan dbang, paratantra), thoroughly imagined nature (brtags pa, kalpita), nor thoroughly established nature (grub pa'i dngos po, nispanno bhavah): so how could one imagine with intelligence? 2  1. SNS VII 9 (cf. tr. Lamotte (1935) p. 195).  Part of this passage was cited with the  information 'mdo kha cig las' in B S G T 109b 1. 2. Lank II k 196=X k 374 (cf. tr. Izumi (1927) p. 76, 180, Suzuki (1932) p. 114, Yasui (1976) p. 119, 278: 2 k 198 following them):  157 There is no self-being, consciousness (rnam rig, vijffapti), entity, nor basis of all.  Foolish (byis pa,  bala), cadaver-  like nit-pickers imagine these [things].  1  [lllbl]  These briefly  refute that the  Madhyamika thesis]  contradicts Agama.  [12.2  Objections based on reasoning and the Madhyamika response]  [lllbl]  To refute the contradiction [of the Madhyamika thesis]  with reasoning,  [12.2.1. The Madhyamika thesis is in contradiction with perception] [12.2. La. Objection] buddhya vivecyamanam tu na tantram napi kalpitam / 1  nispanno nasti vai bhavah katham buddhya vikalpyate / ' hiX (Tib.) P. 119a4=l88b2-3.  Cited in M A V  (C. 68a6, D. 68a6-7, N . 61b3, P. 65bl).  According to the comparison of variants, it can be seen that our text is closer to the Lank II k 196 than X k 374, and still closer to the M A V (particularly M A V [CD]). 1. Lank III k 48=X k 91 (cf. tr. Izumi (1927) p. 96, 158, Suzuki (1932) p. 145, Yasui (1976) p. 151, 248):  na svabhavo na vijffaptir na vastu na ca alayah / balair vikalpita hy ete Sava-bhutaih kutarkikaih / (Tib.) P. 134b3-4=177a7-8 . 2  Cited in M A v B h  3  160.13-6, M A V  4  (C. 68a6, D. 68a7, N.  61b3-4, P. 65M-2). Our citation does not correspond precisely to any of these versions:  'rang bzhin med cing mam rig med / dngos po med cing kun gzhi med / 'di dag ro bzhin byis pa yis / log pa 'i rtog pa can gyis brtags / 2  rang bzhin med cing mam rig med I kun gzhi med cing dngos med na/  byis pa ngan pa 'i rtog ge pa I ro dang 'dra bas 'di dag brtags I 'rang bzhin med cing mam rig med I kun gzhi med cing dngos med na I byis pa ngan pa rtog ge pa I ro dang 'dra bas 'di dag btags I 'rang bzhin med cing mam rig med I dngos po med cing kun gzhi med / ro mtshungs byis pa 'i rtog ge pa / ngan pa mams kyis 'di dag brtags I  158  [Objection:]  Absence of self-nature is not realized by  direct perception (mngon sum, pratyaksa) because it is an object-possessor [which knows] something.  Since direct  perception is devoid of an identity, how could it [direct perception] realize that [absence of self-nature]?  (k  XII-29)  [lllb2]  The selflessness of all phenomena is not realized by  direct perception because it [direct perception] is an object-possessor [which knows] something. Were it not to be so, since what appears to it would be destroyed, it could not be direct perception. an  identity,  [your  ultimate  truth]  is  incapable  Being devoid of of  generating  a  consciousness in which its identity is revealed [and hence not anything] because the ability to perform a function is the definition of a thing. [Illb3]  Also, if even direct perception itself is devoid of an  identity, how could it [cause one to] realize the absence of self-nature of all phenomena?  If it is not without identity, all phenomena would  not lack self-nature.  [ 12.2.l.b.  Madhyamika response]  [Response:]  The  familiarized  himself  yogi with  (rnal  'byor pa)  reality  phenomena are without self-nature.  realizes  who  has  that  all  Also, as to the second  [point raised by the opponents] the doctrinal system [of the Madhyamikas who operate] by way of the two truths is not able to be confused.  [11 lb5]  (k XII-30)  The first [point] expressed [by the opponants] as a fault  is not correct because it is known through the yogic direct perception (rnal 'byorpa'i mngon sum, yogi-pratyaksa), arisen from meditation,  159 that all phenomena are selfless.  Even if [the opponants] say this [yogic  perception] does not exist, it is not logical, because there would not be the valid cognition that harms (gnod par byed pa'i tshad ma, badhakapramana). It is also not [logical to say that yogic perception does not exist] by [the following reason: if it existed, the  distinction between]  our own and other's object would disappear; there would be [the fault] of non-ascertainment (ma nges pa, anaikantika). [11 lb6]  It is also not logical [for the opponants] to say: "Direct  perception has an object which is a thing" because the direct perception which sees this world has as object [things] which are false [such as] forms and so forth and because the wisdom consciousness of the Buddha and other [saints] does not see any real entity of things. As it is stated in the Dharmasamgiti}:  When no phenomenon is seen, it is the supreme seeing.  [112a2]  There is also not the fault [that the opponants have  raised] in saying: "because even direct perception itself is entityless", because on the conventional [plane] such a direct perception is arisen from real meditative stabilization and realizes the entitylessness of all phenomena; on the ultimate [plane] it is without entity.  Therefore,  there is not any criticism that harms this doctrinal system of the [Madhyamika] sage [who uses] the method of the two truths.  [12.2.2.  The Madhyamika thesis contradicts inference]  [12.2.2.a. Objection]  1. Dharmasamgiti (D. (238) zha 68b6, P. [36] (904) wu 74b2-4): bcom ldan 'das chos thams cad ma mthong ba ni yang dag pa mthong ba 'o/. mi P par P. The Sanskrit is found in SS 264.1-2: adarSanam bhagavan sarvadharmanam darSanam samyag-darsanam its. Cited for ex., in M A V P. 78al-2. Cf. BhK I (Skt.) 212.2-3: tatha coktam sutre I 'katamam paramartha-darSanam / sarvadharmanam adarSanam' iti /; (Tib.) 261.7-9: de skad mdo las kyang gsungs te I don dam pa mthong ba gang zhe na I chos thams cad mthong ba med pa gang yin pa 'o zhes 'byung ngo I 1  2  1  2  160  [Objection:]  Since the example (dpe,  drstanta) and the  subject are not established, inference does not operate.  If  these were established, it would not be [true] that all phenomena are without entity.  [112a4]  (k XII-31)  Inference does not operate to prove either all things or  emptiness because the bases of these - the example, the subject and the reason itself, etc. are not established [for the Madhyamika]; moreover, [for them] they are not produced. If they were established, it would be a  contradiction with the  statement  (dam  bcas, pratijna) that the  entityness of all phenomena is not established.  [12.2.2.b.  [Response:]  Madhyamika response]  This [objection] is without connection because on the real  (de nyid du, tattvatas) [plane] [the opponents] prove the evident (grub pa  bsgrub, siddha-sadhana), and on the conventional (tha snyad du,  vyavaharatas) [plane] it is not [true] that [ the example and other elements of logic] are not established.  Therefore, there is no criticism  against us.  (k XII-32)  [112a6] This objection is without connection with the reasoning. Thus, on the real (de kho nar, tattvatas) plane, the example and other [elements  of  logic]  are  not  established  [for  the  Madhyamika].  Therefore, if [the opponents] prove thus by consequences, they prove the evident, because we do not assert that the example, etc. are established on the real [plane]. opponents]  Because of that, this [objection of the  is not a consequence  (thai ba,  prasanga) because the  consequence reduces [the defendents] to what they do not assert. [112bl]  But if [the opponents] prove that the example, etc. are not  established on the real [plane], it is also [the fault] of non-ascertainment  161 (ma nges pa, anaikantika): thus even this is not an objection. If [the opponents] say that the example, etc. are not established on the conventional [plane] it is [the fault] of non-established reason (ma grub pa, asiddha). If they say that it is not produced even on the conventional [plane] like the horns of a rabbit (ri bong rwa, $a$avisaha) it would be the view of annihilation (chad par lta ba, ucchedadrsti): on the conventional [plane] convention such as actions (las, karman) and effects ('bras bu, phala) indeed exist. Therefore, one makes conventions such as inference, based on the subject etc., which are renowned in the world in order to introduce [beings] to reality. As the expert in logic (Rigs pa mkhyen pa = Dharmaklrti) said [in the Pramanavarttika (I k 85-6] : 1  The establishment of the attribute and the subject of the attribute (chos dang chos can, dharma-dharmin) such that the difference and non-difference [between them] are without connection with the meaning of reality, are known in the world. Similarly, based on this [etablishment and that which follows], the wise make the establishment of that which is to be proved (bsgrub bya sgrub pa, sadhya-sadhana) for the sake of introducing [beings] to ultimate [truth]. Therefore, [the opponants] do not find occasion to reach we Madhyamikas by this manner of criticism. 1. PV I kk 85-6 (kk 87-8 following Frauwallner (1930a), cf. tr. Frauwallner (1932) p. 268, Ota (1982) p. 13):  dharma-dharmi-vyavasthanam  bhedo 'bhedaS ca yadrSah /  asamiksita-tattvartho yatha lokepratlyate / tarn tathaiva samaSritya sadhya-sadhana-samsthitih I paramarthavataraya  vidvadbhir  avakalpyate I  Dhaimaklrti here clearly explains that all the logical rules operate on the conventional plane, and are not applicable to ultimate truth. These verses are cited in MAI [P] 187a8-bl.  162 [13] Intermediary verses: Ultimate [truth] is the object of correct consciousness (rig shes) but beyond the object of the mind grasping the extremes. The subject, although like an illusion, is the object of correct consciousness but is not real. (AS XII-10) 1  Appearance and emptiness are not contradictory: the correct consciousness (rig shes) which understands emptiness does not harm the valid cognizer in which appear forms and so forth. Therefore, there is no contradiction. (AS XII-11) The negation of production occurs on the ultimate [plane] but not on [the plane of] mere appearance. Consequently, there is not falling to the extremes of the view of permanence and of annihilation. (AS XII-12) Regarding this point, one knows in detail the objections and responses to objections from the treatises of the wise Kamalas'iia (Pa dma'i ngang tshul) which clearly show the position of the Madhyamikas free from extremes. (AS XII-13) 2  1. Should rig shes be corrected with rigs shes (logical consciousness)? 2. The 'treatises of the wise Kamalaslla' should be understood principally as the M A I and the T A ; they have been cited abundantly in our text. The that is a sort of summary of the MAI is included here.  Sarvadharmanihsvabhavasiddhi  163  [14. Conclusion and Vow]  [113al]  By that [shown above] the doctrinal position of the  "Propounders of lack of self-nature of all phenomena" on the [five] categories of objects of knowledge has been explained.  The theory of  the [school] on the path (lam, marga) and the effect ('bras bu, phala) will be explained [in chapter XIII ]. 1  May I, through having explained to beings the stainless doctrinal  positions  showing  the  selflessness  of  all  phenomena attain [the state of] omniscience, as king of the Dharma, the most excellent of those who show reality again to the world.  (k XII-33)  End of chapter XII, the Explanation of the Madhyamika Position, from Grub pa'i mtha' mam par bshad pa.  1. BSGT 113a3-126b2.  164  BIBLIOGRAPHY  Ames, William L . (1982), "The Notion of Svabhava in the Thought of Chandraklrti".  Journal of Indian Philosophy 10, pp. 161-177.  Amano, H . (1965), "Haribhadra no Nitaisetsu" (Haribhadra's Doctrine of Two Truths). Journal of Indian and Buddhist Studies 13-2, pp. 176-81. (1975), A Study on the Abhisamaya-alamkara-karika-Sastra-vrtti. Tokyo. Bhattacharya, Kamaleswar.  1974, "A Note on the Interpretation of the term  sadhyasama in Madhyamaka texts". Jounal of Indian Philosophy 2, pp. 225-230. (1978), The Dialectical Method of Nagarjuna. Delhi: Motilal Banarsidass. Bhattacharya, V . (1931), The Catuhsataka of Aryadeva, Part II. Calcutta. Chattopadhyaya, Alaka. (1967), Atisa and Tibet Delhi: Motilal Barnarsidass. Chattopadhyaya, D. (1970), Taranatha's History of Buddhism in India. Simla: Indian  Institute of Advanced Study.  Conze, Edward. (1967), Buddhist Thought in India. Ann Arbor: University of Michigan Press. (1973) , The Perfection of Wisdom in Eight Thousand Lines & its Verse Summary. Bolinas. (1975), The Large Sutra on Perfect Wisdom with the divisions of the Abhsamayalankara. Berkeley: University of California Press. Conze, E . and Iida, S. (1968), "Maitreya's Questions" in the Prajftaparamita". Melanges d'indianisme a la memoire de Louis Renou, pp. 229-242. de Jong, J.W. (1949), Cinq chapitres de la Prasannapada. Paris: Guethner. (1972a), "Emptiness".  Journal of Indian Philosophy!, pp. 7-15.  (1972b), "The Problem of the Absolute in the Madhyamaka School". Journal of Indian Philosophy 2, pp. 1-6. (1974) , "A Brief History of Buddhist Studies in Europe and America". The Eastern Buddhist, NS 7 no. 1 p. 55-106; no. 2 p. 49-82.  165 (1979), Buddhist Studies. Ed. by Gregory Schopen. Berkeley: Asian Humanities Press. Demieville, Paul. (1952), Le Concile de Lhasa. Paris: Impr. Nationale de France. Eckel, M . David. (1980), "A Question of Nihilism: Bhavaviveka's Response to the Fundamental Problems of Madhyamika Philosophy'. PhD. dissertation, Harvard University. (1986) , Jnanagarbha's Commentary on the Distinction Between the Two Truths. Albany: State University of the New York Press. Ejima, Yasunori. (1980), Chukan Shiso no Tenkai. Tokyo. Frauwaller, E . (1930a, 1932, 1933, 1935)  "Beitrage zur Apohalehre".  Wiener Zeitschrift fur die Kunde des Morgenlandes (abbr. WZKM) 37, pp. 259-83; 39, pp. 247-85; 40, pp. 51-94; 42, pp. 93-102. (1969), Die Philosophic des Buddhismus. Berlin. Guenther, Herbert V . (1972), Buddhist Philosophy in Theory and Practice. Baltimore:  Penguin Books.  Hahn, M . (1982) Nagaruna's Ratnavali. Bonn. Hakamaya, N. (1975), Miroku Shomon Sho Wayaku. Komazawa Daigaku Bukkyogakubu Ronshu no. 6, pp. 210-190. Hayashima, O.  (1978), "Ratnakarasanti no Chudo Shiso—  Madhyamakalamkaropadesa ni okeru".  Journal of Indian and Buddhist  Studies n. 26-2, pp. 1012-1008. Hikata, Ryush5. Fukuoka: Hopkins, Jeffrey.  (1958),  Suvikrantavikrami-pariprccha  Prajnaparamita-sutra.  Kyushu University. (1983), Meditation on Emptiness. London: Wisdom  Publications. (1975), The Precious Garland and the Songs of the Four Mindfulness. London. (1987) , Emptiness Yoga. Ithaca, New York: Snow Lion Publications. Iida, Shotaro.  (1966), "Agama (Scripture) and Yukti (Reason) in  Bhavaviveka"  Kanakura Festschrift Kyoto: Heirakuji-Shoten, pp.79-  96. (1973), "The Nature of Samvrti and the Relationship of Paramartha to It in Svatantrika-Madhyamika".  The Problem of the Two Truths in  166 Buddhism and Vedanta. Edited by Mervyn Sprung. Dordrecht, Holland:  D. Reidel Publishing.  (1980), Reason and Emptiness: A Study in Logic and Mysticism. Tokyo: Hokuseido Press. Imaeda,Yoshiro.  (1975), "Documents tibetains de Touen-houang concernant  le concil du Tibet". Journal Asiatique, pp. 132-133. Inada, Kenneth. (1970), Nagarjuna. Tokyo: Hokuseido Press. Ingalls, Daniel H.H.  (1951), Materials for the Study of Navya-Nyaya Logic.  Harvard Oriental Series, Vol. 40.  Cambridge, Massachusetts: Harvard  University Press. Izumi, H . (1927), Hoyaku Bonbun Nyuryogukyo (Jap. trans, of the Lankavatara-sutra).  Kyoto.  Jha, Gangantha. (1967), Tarkabhasa or Exposition of Reasoning. Poona. Johnston, E . H . and Kunst, A . (1951), "The Vigrahavyavartani of Nagarjuna". Melanges chinois et bouddhiques 9, p. 99-152. Joshi, Lalmani. (1967), Studies in the Buddhistic Culture of India. Delhi: Motilal Banarsidass. Kajiyama, Yuichi.  (1974), Ryuju Ronshu. (Nagarjuna's Works). Daijo  Butten (Mahayana Buddhist Texts), no. 15. Tokyo, 1974. (1976), "Later Madhyamikas on Epistemology and Meditation: Theory and Practice". Mahayana Buddhist Meditation. Ed. Minora Kiyota.  Honolulu: University Press of Hawaii.  (1976a), VimSatika. in Nagao, G.M. (1976), pp. 5-30. Katsura, S.  (1969), "Dharmakirti ni okeru Jikoninshiki no Riron"  (Dharmakirti's Doctrine of Self-awareness). Nanto Bukkyo, no. 23, pp. 1-44. (1976), "A Synopsis of the Prajnaparamitopadesa of RatnakaraSanti". Journal of Indian and Buddhist Studies 15-1, pp. 484-87. Kitigawa, H . (1965), Indo Koten Ronrigatu no Kenkyu (A Study of Indian Classical Logic). Tokyo. Klein, Anne. (1986), Knowledge and Liberation. New York: Snow Lion. Lamotte, Etienne.  (1935), Samdhinirmocana-sutra. Paris: Adrien  Maisonneuve. (1938), La somme du Grand Vehicule d'Asanga, 4 vol, tome I, Louvain.  167 (1949-80), Le Traite de la Grande Vertu de Sagesse de Nagarjuna, 5 tomes (I. II. 1949; ffl. 1970; IV. 1976; V . 1980). Louvain: Institut Orientaliste. (1962), L'enseignement de Vimalakirti. Louvain: Institut Orientaliste. la Vallee Poussin, Louis de. (1896), Pancakrama. Grand-Louvain. (1907, 1910, 1911)  Madhyamakavatara, Museon 8 (1907), pp. 249-317;  11 (1910) p. 271-358; 12 (1911) p. 235-328. (1907b), Bodhicaryavatara, Introduction a la pratique de Futurs Bouddhas. Paris. (1913), "Les quatres odes de Nagarjuna". Museon 14, p. 1-18. (1932-3), "Le petit traite de Vasubandhu-Nagarjuna sur les trois natures". Melanges Chinois et bouddiques. Bruxelles. (1971), L'Abhidharmakoia  de Vasubandhu. Bruxelles: Institut Beige des  Hautes Etudes Chinoises. Lati Rinbochay. (1980), Mind in Tibetan Buddhism. Trans, by Elizabeth Napper. Ithaca, New York:  Snow Lion Publications.  Levi, Sylvain. (1925), Vijnatimatratasiddhi. Paris. Lindtner, Christian. (1981), "Atisa's Introduction to the Two Truths, and Its Sources".  Journal of Indian Philosophy 9, pp. 161-214.  (1982), Nagarjuniana: Studies in the Writings and Philosophy of Nagarjuna. Copenhagen: Akademisk Forlag. (1986), "Bhavya's Critique of Yogacara in the Madhyamakaratrtapradipa, Chapter IV". Buddhist Logic and Epistemology. Ed. by B.K. Matilal and R.D Evans. Holland: D. Reidel Publishing Company, pp. 239-263. Lopez, Donald S. (1987), A Study of Svatantrika. Ithaca: Snow Lion Publications. Mano, R. (1972), Genkanshbgonron no Kenkyu. (A Study of the Abhisamayalankara).  Tokyo.  Matilal, Bimal Krishna. (1973), "A Critique of the Madhyamika Position". Problem of the Two Truths in Buddhism and Vedanta. Ed. by Mervyn Sprung. Dordrecht, Holland: D. Reidel Publishing, pp. 54-64. (1974), "A Note on the Nyaya Fallacy sadhyasama and Petitio Principii". Journal of Indian Philosophy!, pp. 211-224.  168 Matsuo, G. (1948), Indo Ronrigaku no Kozo (On the Structure of Indian Logic). Osaka. Matsumoto, S.  (1980), "Ratnakarasanti no Chukanha Hihan"  (RatnakaraSanti's Criticisms of Madhyamika).  Toyo Gakujutsu  Kenkyu\9-2, pp. 152-180. (1981), Tson kha pa no Chukan Shiso ni tsuite" (The Madhyamika Doctrine of Tsong khapa).  Toyo Gakuho 62, pp. 390-427.  (1981) , "ITa bahi dkyad par ni okeru Chukan Rikai ni tsuite" (The Interpretation of the Madhyamika Doctrine in the ITa ba'i khyad par). Sotoshu Kenkyuin Kenkyusei KenkyuKiyo 13, pp. 93-124. May, Jaques. (1959), Prasannapada Madhyamaka vrtti, douze chapitres traduits du Sanskrit et du tibetain. Paris: Adrien Maisonneuve. (1978) ,  "On Madhyamika Philosophy".  Journal of Indian Philosophy 6,  pp. 233-241. Mimaki, Katsumi. (1976), "La refutation bouddhique de la permanence des choses (sthirasiddidusana) et la preuve de la momentaneite des choses (ksanabhangasiddhi)".  Publications de l'lnstitut de Civilisation  Indienne, Fascicule 41, Paris. (1979) , "Le chapitre du Bio gsal grub mtha' sur les Sautrantika". Zinbun 15, pp. 55-112. (1980) , "Sur le role de l'Antarasloka ou du Sangrahasloka". Melanges offerts a Mgr. Etienne Lamotte, pp. 233-44. (1982) , Bio Gsal Gmb Mtha'. Kyoto: Zin bun Kagaku Kenkyusyo. Mimaki, K. and May, J. (1979), "Chudo". Hobogirin V Fascicule, pp. 456470. Mittal, K. (1984), A Tibetan Eye-view of Indian Philosophy. New Delhi: Munshiram Manoharlal Publishers Pvt. Ltd. Moriyama, S.  (1981), "Kamalaslla no Sarvadharmanihsvabhavasiddhi no  Wayaku Kenkyu".  Bukkyo Daigaku Daigakuin KenkyuKiyo 9, pp. 60-  100. Murti, T.V.R.  (1955), The Central Philosophy of Buddhism. London:  George Allen and Unwin.  169 (1973) , "Samvrti and Paramartha in Madhyamika and Advaita Vedanta". The Problem of Two Truths in Buddhism and Vedanta. Ed. by Mervyn Sprung. Nagao, G.M.  Dordrecht, Holland: D. Reidel Publishing. (1954), Chibetto Bukkyo Kenkyu. (A Study of Tibetan  Buddhism).  Tokyo: Iwanami Shoten.  (1954a), "An Interpretation of the Term "Samvrti", (Convention) in Buddhism".  Silver Jubilee Volume of the Zinbun Kagaku Kenkyusho.  Kyoto University, pp. 550-561. (1967), "Daijo Butten" (Mahayana Buddhist Works).  Sekai no  Meichol,  Tokyo. (1974) , "Yuimakyo" (Vimalaklrti-nirdes"a).  Daijobutten 7, Tokyo.  (1978) , Chukan to Yuishiki (Madhyamika and Yogacara).  Tokyo:  Iwanami. (1979) , "From Madhyamika to Yogacara:  An analysis of M M K , XXIV.18  and M V , 1.1-2". Journal of the International Association of Buddhist Studies 2, pp. 29-43. Nakamura, Hajime.  (1976), "A Survey of Mahayana Buddhism with  Bibliographical Notes".  Journalof Intercultural Studies 3, pp. 60-145.  (1980) , "Indian Buddhism".. Kansai U . of Foreign Studies, Intercultural Research Monograph 9, Hirakata. Obermiller, E . (1932), History of Buddhism by Bu-ston. Heidelberg: Heft. Ogawa, I. (1976), Kusho Shiso no Kenkyu (A Study of Sunyata Thought). Kyoto. Ota, S.  (1982), "A Translation of Pramanavarttika I and Svavrtti". Saga-  Ryukokutankidaigaku Kiyo 28, pp. 1-22. Rabten, Geshe. (1983), Echoes of Voidness. Ed. by Stephen Batchelor. London:  Wisdom Publications.  RahderJ. (1926), Datra et Bodhisattvaghmi. Paris. Ramanan, K. Venkata. (1978), Nagarjuna's Philosophy. Delhi: Motilal Banarsidass. Robinson, Richard H . (1967), Early Madhyamika in India and China. Wisconsin:  University of Wisconsin Press.  (1970), The Buddhist Religion. Belmont, California: Dickenson Publishing Co.  170 Roerich, George N. (1976), The Blue Annals. Delhi: Motilal Banarsidass. Ruegg, David S. (1980), "On the Reception and Early History of the dBu-ma (Madhyamaka) in Tibet".  Tibetan Studies in Honour of Hugh  Richardson. Ed. by Michael Aris and Aung San Sun Kyi. Warminster, England: Aris and Phillips Ltd., pp. 277-279. (1981), The Literature of the Madhyamaka School of Philosophy in India. Wiesbaden:  Otto Harrassowitz.  (1985), "Purport, Implicature and Presupposition:  Sanskrit Abhipraya and  Tibetan DGONS PA/DGONS GSI as Hermeneutical Concepts". Journal of Indian Philosophy 13, pp.309-325. Sakai, S. (1956), Tibeto Mikkyo Kyori no Kenkyu . Tokyo. Sasaki, G. (1922), Yuishiki Nijuron no Taiyaku Kenkyu (A Comparative Study of the Vimsatika). Kyoto. Schayer, Stanislaw.  (1931), Ausgewahlte Kapitel aus der Prasannapada.  Cracovie. Sherburne, Richard (trans.).  (1983), A Lamp for the Path and Commentary.  London: Allen and Unwin. Snellgrove, D.S. (1987), Indo-Tibetan Buddhism. Boston: Shambala. Snellgrove, D. S. and Richardson, Hugh. (1980), A Cultural History of Tibet Boulder: Prajna Press. Sopa, Geshe Lhundrup and Hopkins, Jeffrey.  (1976), Practice and Theory of  Tibetan Buddhism. New York: Grove Press. Sprung, Mervyn.  (1979), Lucid Exposition of the Middle Way. Boulder:  Prajna Press. Stcherbatsky, Theodore.  (1927), The Conception of Buddhist Nirvana.  Leningrad: Office of the Academy of Sciences of the U.S.S.R. (reprint Delhi: Motilal Barnasidass, 1977). Stein, R. A . (1972), Tibetan Civilization. Stanford, California: Stanford University Press. Streng, Frederick J. (1967), Emptiness: A Study in Religious Meaning. Nashville:  Abingdon Press.  (1973), "The Significance of Pratityasamutpada for Understanding the Relationship between Samvrti and Paramarthasatya in Nagarjuna". The  171 Problem of the Two Truths in Buddhism and Vedanta. Ed. by Mervyn Sprung.  Dordrecht, Holland: D Reidel Publishing Co.  Suzuki, Daisetz T. (1932), The Lankavatara Sutra. London: Routledge and Kegan Paul. Tamura, Ch.  (1975), "Sanmaiokyo"(Samadhi-raja-sutra). Daijo Butten lu-  l l , Tokyo. Tauscher, H . (1981), Chandrakirti Madhyamakavatarah und Madhyamakavatarabhasyam (Kapitel VI, Vers 166-226),Wein. Thurman, Robert A . (1976), The Holy Teaching of Vimalaklrti. Pennsylvania:  The Pennsylvania State University Press.  (1984), Tsong Khapa's Speech of Gold in the "Essence of True Eloquence". Princeton: Princeton University Press. Tucci, Giuseppe.  (1929), "Pre-Dinnaga Buddhist texts on logic from Chinese  sources". Gaekwad's Oriental Series 49. Boroda. (1934-36), "Ratnavall of Nagarjuna". Journal of the Royal Asiatic Society. 1934, pp. 307-325; 1936, pp. 237-252; and pp. 423-435. (1949), Tibetan Painted Scrolls. Rome. (1980), The Religions of Tibet. California:  University of California Press.  Uryuzu, R. cf. Kajiyama (1974). Vaidya, P.L. (1923), Etudes sur Aryadeva et sa CatuhSataka. Paris. Vetter, T. (1966), Dharmakirti's Pramanavini&cayah. Vienna: Osterreichische Akademie der Wissenchaften. Wayman, Alex. (1978), Calming the Mind and Discerning the Real. New York:  Columbia University Press.  Wogihara, U . (1938), Wogihara Unrai Bunshu (Collected Works of Wogihara).  Tokyo.  Yamaguchi, S. (1929), "Traite de Nagarjuna, Pour ecarter les vaines discussion [Vigrahavyavartani]".  Journal Asiatique, pp. 1-86.  (1947-9), Gesshd-zo Churon-shaku (Candraklrti's Prasannapada).  Tokyo.  (1964) , Bukkyo ni okeru Mu to U tono Tairon (A Buddhist Controversy concerning Existence and Non-existence). Tokyo. (1965) , Chukan Bukkyo Ronko (Studies on Madhyamika). Tokyo. (1972), Yamaguchi Susumu Bukkyogaku Bunshu (Collected Works on Buddhism by Susmu Yamaguchi). Tokyo.  172 Yasui, Kosai. (1970), Chiikan Shiso no Kenkyu (A Study of Madhyarnika). Kyoto. (1976), Bonbun Wayaku Nyuryogakyo (A Jap. trans, of the Sanskrit Lankavatara-sutra).  Kyoto.  Yoshimura, S. (1974), Indo Daijo Bukkyo Shiso Kenkyu (A Study of Indian Mahayana Buddhism).  Kyoto.  Yuyama, A . (1976), PrajM-paramita-ratna-guna-samcaya-gatha. University Press.  Cambridge  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items