Tibetan Citations Of Bhartrharis Verses And Problem Of His Date

Author(s): Hajime Nakamura
Publisher: Hajime Nakamura
Catalog link: https://jainqq.org/explore/269718/1

Page #1

TIBETAN CITATIONS OF BHARTRHARI'S VERSES AND THE PROBLEM OF HIS DATE.
by.
Hajime NAKAMURA
University of Tokyo
About twenty years ago, the illustrious scholar to whom we are dedicating this commemoration volume pointed out the fact that some verses of a Vedāntin called Bhadrahari(?X is mentioned in the Tibetan version of some philosophical works of later Mahāyāna Buddhism. The author of the present article wants to elaborate on the problem, gathering Tibetan verses ascribed to him, and collating some of them with Sanskrit originals which have been identified by the author.
In the first place the following verse is cited in the Tibetan version of the Mahāyāna-sütrālamkāra-lika, a commentary by No-bo-ñid mcd-pa (=Asvabhäva) upon the Mahāyānasüträlankara : - (fr. 1.) Bha dri ha ri(s) ji skad du/
यथाह महरिः nam mkhah sa rlun ñi ma dah I : : 8H) 6 2. rgya mtsho glin dan phyogs rnams ni
JT25: Natt nan gi ses pahi no bo yi //
अन्तःकरण्तावस्य cha las phyi rol Ita bur hhug/
भाग बटेरिव २ ces bśad pa lta bu yin no // hdzin pahi don las bzun bahi don tha mi dad mthon baho // shes bya ba ni gzun ba dan hdzin pahi rnam par ses pa kho nar zad kyi rnam par ses pa la ma rtogs pa byis pas kun brtags pa mig la sogs pa hdsin paḥi don ni yod pa ma yin no shes bya bar te / dehi phyir don du snan ba thams cad sems kho na yin par Itahidon sems las gshan par ni mi blta ho // dehi tshe ni hdiņi gzun ba la rnam par gyen pa spans la ḥdzin paủi rnam par rtog pa hbaḥ shig lus te/de ni bzod pahi gnas skabs yin no/
“As was declared by Bhadrahari...!
Ether, carth, wind, the sun, the sea, continents and directions ... they exist like external things, having been manifested from the in
nermost, proper portions of the interior intelligence (=adhyatmajnana)."
1 Prof. Susumu Yamaguchi, " ELES F#" VIII, Kyoto 114 (1936), pp. 152-153. : His cominent upon Dharmaparyesti 5. Bstan-hsyur of the Peking edition, XLY, 90 a, 8-b, 1.

Page #2

BHARTRHARI'S VERSES
123
Then follows Asvabhāva's annotation, in which he says, “So everythings which has been manifested as an object is regarded as nothing but mind, and objects are not regarded as separate from mind." So we can assuredly conclude that Asvabhāva himself recognized in Bhadrahari's assertion something similar to the thought of Buddhist Idealism (vijñaptimätrată). The Sanskrit original of the above-cited Tibetan verse is the following one:
dyauh kşamāvāyur adityah sāgarāḥ sarito disah/ antahkaranatattvasya bhāga bahir avasthitaḥ //
- Vakyapadiya III, 7, 41 (p. 200) 3 (cf. antahkaranadharmo vā bahir evam prakāšate /- ibid., III, 6, 23. p. 170). In the Arya-Lankávatāra-vrtti by Jñanaśribhadra there are the following ver
*
ses.
(fr. 2) de bas na Bha dra ha ris byed pa bstan pa las smras pa / .
grub dan ma grub ji sñed pa/ bsgrub pahi phyir ni bstan par bya / rten pahi rim (23 b) gyi ran bsin phyir /
de ni byed par brjod pa ho // şes pa ste / byed pa dpyad zin to // Tr.--"Therefore it was told in a treatise written by Bhadrahari
In so far as a thing attained or a thing not yet attained should be
denoted in order to be established, On account of its essence being the order to be resorted to,
it is declared to be an action." (=yāvat siddham asiddham vā sādhyatvenābhidhiyate /
asritakramarūpatvāt kriyeti pratiyate //)* (cf. kriyāyaḥ parinispattir yad vyāpāräd anantaram/ vivaksyate yada tatra karanatvam tada smstam // Väkyap. III, 7,
b, 89. p. 237.)
(fr. 3)5 tshans par smra ba po bha ta ha ri la sogs pa yan smra ba!
thog ma med pahi tshans pa rtag / yan dag sgra ni yig hbru gan / gan las don dños hbyun ba dan / hgro bahi rab tu byed pa yan //
3 Vakyapadiya, ed. by Pandit Ramachandra Sastri Koţibhāskara (Benares Sanskrit Series), Benares 1887 f., vol. 2, p. 200. According to the citation of the verse by Dr. Chakravarty antahkaranadharmasya instead of antahkaranatattvasya (Prabhatchandra Chakravarti: The Linguistic Speculations of the Hindus, University of Calcutta, 1933, p. 197). The reading of the Benares text corresponds with the Tibetan version.
* Vakyapadiya IIT, 8 (kriyāsamuddeśa), l. p. 298 (Benares Sanskrit Series). However, the verse cited by Dr. Chakravarti readz vyapadisyate instead of pratiyale. (P. Chakravarti: op. cit. p. 237)
5 All citations henceforth are taken from the Peking edition,

Page #3

124
HAJIME NAKAMURA
(fr. 4)gan yons rtogs pa thams cad la/
(127 a) snan bar yan ni mi gnas te/ rtog pa lun dan rjes dpag pas/ rnam pa du mar kun brtags paḥo //
(fr. 5) hdas kyan ḥdu ḥdzi byed pa dan/ dnos dan dños med rim mi rim/ bden dan brdzun par sna tshogs bdag/ dben pa las ni rab tu snan //
(fr. 6) hbyuň po rnams kyi nan rgyu ba/ hgran dan rim duhan snan ba ste/ thar pa rab tu grol ba de/
grol hdod mams kyis bstan par byed //
(fr. 7) de yi tshor ba gcig po yan/
rnam pa man por rab tu hbyed/ hjig pa hbyun bahi me mdag dan/ chu yi phun por chu bshin du //
(fr. 8) de phyir da rod rigs gnas pa/
gsal bahi gron rnams hbyur ba ni/ sprin rnams char dan Idan pa rnams/ rlun las ḥbyun ba bshin duḥo //
(fr. 9) gsum gyi gzugs su de snan ba/
mur smra tha dad rnams la ni/ mchog gi yons su hgyur ba ste/ Ita ba tha dad kyis ni bcins //
(fr. 10) shi shin rigs bdag tshans pa ste/ de las byun bahi ma rig pa/ des ni srin po bshin du med/ gan smra mi nus rnam dag go//
(fr. 11)ji Itar nam mkhaḥ rnam dag par / rab rib ḥkhrul pahi skyes bu yis/ ri mo rnam pa sna tshogs kyis/ kun tu gan bar sems pa pa // de ltar tshans pa bdud rtsi ḥdi/ hgyur ba med pa ma rig pas/ rñog pa lta bur byas pas na/ tha dad gzugs su snań ba ste //
(fr. 12) lha sa rlun dan nam mkhaḥ dan/ ri dan chu dan phyogs rnams kyan/ nan du byed pahi yan dag gi/ cha rnams phyi rol bshin du gnas //
(fr. 13) tshul ni gcig ñid gan yin las/
tha dad mthu ni rnam gnas pas/ mthu las tha dad ma yin yan/ tha dad bshin du ḥbyun baho //
(fr. 14) tshans sgrahi hdi ni sprul pa ste/ sgra yi mthu las gnas pas na / sgra tsam las ni byun ba yan/

Page #4

BHARTRHARI'S VERSES
de dag ñid du sha bar hgyur // . (fr. 15) rab tu byed pa gcig hbyed la/
man por rab tu phye ba pa /
de ni (127 b) lun ston rtogs pa las / • tshans pa mchog ni chud par hgyur // shes rgya cher ro // Tr... . Bhatahari and others, who are brahmavādins, also say:
(6.3) The beginningless, eternal Brahman, whose essence is Word and Syllable (akşara, undestructable). .. , from That, things have evolved, and the function of the world also has appeared."
(=anādinidhanam brahma śabdatattvam yad aksaram/
vivartate'rthabhāvena prakriyā jagato yatah//– Vakyap. I, I) (fr. 4) Whatever has appeared by dint of imaginary assumption (parikalpa) does not abide.
Imaginary assumptions appear in various forms by dint of ratiocination (tarka, vitarka), scriptural dogma (agama) and reasoning (anumāna).'
(cf. Vākyp. II, v, 143, p. 140; III, 4, 1-2, p. 139; III, 14, v. 199. p. 550) I* (fr. 5) 'Separation and connection (samsarga), being (bhāva) and nonbeing (abhāva), order (temporal sequence) and non-order (simultaneity), truth and falsehood, various things are manifested from Ātman, the Truth.
(cf. avirodhi virodhi vā sann asan vāpi tattvatah/ u kramavän akramo vāpi nābhāva upapadyate /--Vakyap. III, 3.66. T p. 130) (cf. II, 317. p. 214.) No 4* Hocamsen: 3:
24 (fr. 6) '(It is the internal cause of beings (bhūta), which put forth effort and appear in time (temporal sequence). Deliverance is to be saved and relieved. This is taught by those who wish for deliverancc.'
(fr. 7) The sensation (vedanā) of it is divided in a uniform and multiform way.
It is like water in contrast with the multitude of waters and like flames which flare and vanish. (cf. Vākyap. II, 252; 480) X
(fr. 8) Therefore a man who now belongs to a superior clan becomes (a member) of a distinctive family (kula) as when many clouds bearing water are born of winds.'
(fr. 9) It appears in three forms.
To (those thinkers who adhere to various extremes, it (=one of these) becomes the supreme one. They are bound by various heretical opinions (drs!i).'
(fr. 10) 'Brahman is Atman, which is to be known by tranquility, and from
S

Page #5

126
HAJIME NAKAMURA
which ignorance (asidyd) appears.
By means of it (it) does not exist like raksasas (demons), which are of such a nature as can not be explained.'
(fr. 11) The Sanskrit verse corresponding to this verse and parallel verses are found in several works. We shall discuss it later.
- (fr. 12) Gods (deva), earth, the wind, the ether, mountains, water and directions are what the internal essential portion has manifested as if they were things external.' (cf. fr. 1.)
(fr. 13) Whereas there exist various potentialities (sakti) different from the entity of one form, yet that is not different from potentialities. They appear as if they were different.'
(ekam eva yad ämnätam bhinnam śaktivyapāśrayat/
apṛthaktve 'pi śaktibhyah pṛthaktveneva vartate // Vakyap. I, 2.)
(cf. Vakyap. II. v. 22, p. 78; v. 25, p. 79; v. 476, p. 279; III, 1, v. 20. p. 23; III, 14, v. 567. p. 714.) 570 in AL
(fr. 14) This is the transformation (emanation) of sabdabrahman. Abiding by the potentialities of Words, they become tranquil in themselves, although they have arisen from Words alone.'
(sabdasya parināmo 'yam ity amnayavido viduḥ /
chandobhya eva prathamam etad viśvam vyavartata? // I, 121)
Vakyap.
(cf. Vakyap. II, 31. p. 82)
(fr. 15) As the one entity is divided into many by means of various creative processes, the highest brahman will be attained by resorting to (the science of) grammar."
(=Yad ekam prakriyābhedair bahudha pravibhajyate/ tad vyākaraṇam agamya param brahmadhigamyate//
-Vakyap. I, 22.)
So the problem has been discussed in detail."
(fr. 16) (42. 138a) de bṣin du nam mkhahi yan nam mkhahi kho na ste/ bro ba la sogs pa du ma daň sbyor bas yul tha dad par brtags pa ste / man pohi gnas na nam mkhaḥi kho naho shes bha dra ha ris smras pa ho//
""Likewise there exists ether alone for ether. The object (place) is regarded as differing with adjuncts and with desires of various kinds.
Although there exist many things in ether, there is ether alone in reality'
The negative character of the absolute was stressed by Yajnavalkya, the Upanishadic thinker; and later in the Käthapa-Up. (II, 6, 9; 12).
7 It seems that the Tibetan translator took vyavartata in the original for nyavartata.

Page #6

BHARTRHARI'S VERSES
127 ...So spoke Bhadrahari.”
(=saryogidharmabhedena deśe ca parikalpite /
teșu deseșu sämänyam ākāśasyāpi vidyate /---Vakyap. III, 15. p.20) (42. 138b-139a) mtshan ñid kyi tshig ces bya ba ni smra ba rnams kyi rjod pahi don du ḥbyed pas na rtsod pa yan man ste / yi ge ma gtogs tshig gshan ni // yod pa ma yin brjod pa ni/yi ge tshig ni ma gtogs par (139a) cun zad kyan ni yod pa min shes dpyod pa can gyi rjes su mthun par bya ste / Bhadra haris bśad pa / (fr. 17) smras pahi sgra kyan ḥdus pa ni/
rigs ni hjug par hgyur ba ste / gcig bu yan lag med pahi sgra/
blo yis rigs su bsdu ba ho ll (fr. 18) thog mahi tshig ni so sor kun /
bltos dan bcas te de Itar yan / brjod pahi phyir ni tha dad blo/
tshul smra rnams kyi rnam pa man // shes te /
“Bhadrahari says:
(fr. 17) "Spoken Words, having been gathered, are classified into classes (jati). A single, undifferentiated word is put in a class by the mind (of hearers).' (=svä jätih prathamam sabdaih sarvair evabhidhiyate /
tato 'rthajätirūpesu tadadhyāropakalpanā // Väkyap. III, 6.) (fr. 18) The first syllables (akşara) have each of them something to be seen (visible aspect) and express (meaning) in such a way, that the capability (of hearers to interpret them) is multiple. There are many kinds of those expressing ways.'” Sekasminn api drśye 'rthe darśanam bhidyate prthak /
kälāntarena vaiko 'pi tam paśyaty anyatha punah // ekasyäpi ca sabdasya nimittair avyavasthitaih / ekena bahubhis cartho bahudhā parikalpyate // Vākyap. II, 138; 139, pp. 138-139.)
(cf. Vākyap. 11. v. 33. p. 82) Bhavya, the Madhyamika philosopher, commenting upon Nāgārjuna's Madhyamaka-śāstra IX, 3 cites the following verse: 8
8 This verse is not found in the corresponding passage in the Chinese translation by Prabhamitra of Bhāvaviveka's Prajnapradipa (Az
1 .
H. FFF WA, Taisho Tripitaka, vol. XXX, p. 82c). According to Prof. Nozawa it is likely that this verse was omitted by tnc translator.

Page #7

128
HAJIME NAKAMURA
(a
(fr. 19) lon ba rkan pahi tshod dpags kyis/
ñam nahi lam du rgyug pa Itar / rjes su dpag pa gtsor ḥdsin pa/ mam par Itun ba mi dkaho //
*********
shes bya ba Ita bur hgyur ro she na / de ni rigs pa ma yin te/
Avalokitavrata comments upon this as follows: don de ñid bsal bar bya bahi phyir slob dpon bha ndri ha ris rjes su dpag pa sun phyun baḥi khuns bstan pahi phyir/(a. a) shes bya ba smras te/dper na lon ba rkan pahi tshod dpag gis lam bde ba spans te ñam ñaḥi lam du rgyug pa sin la sogs pa dag gis brdos pa la rdeg ḥchaḥ śin rnam par ltun ba mi dkaḥ ba de bshin du/yid ches paḥi lun spans pa rjes su dpag pahi tshod dpag gis rtog ge skam poḥi lam du rgyug pa dag kyan tshe ḥdi la yan ḥkhrul pahi gcod par hgyur la/tshe phyi ma la yan hbras bu mi ḥdod par rnam par' Itun bar hgyúr ro shes zer ro/
The Sanskrit original of the cited verse spems to have been the following:
.
"hastasparśadibädhena visame (=girimärge, Comm.) 'py abhidhāvată/ anumānapradhanena vinipato na durlabhah //" (Vakyap. I, 42.)
"Just as a (blind) man who is running quickly on a dangerous pass (in the mountains) may fall down if he fails to cling by hand, so it is not impossible for a man who chiefly relies upon his own reasoning (lit. inference) to fall down (into a hell)."
The Tibetan version seems to have a free translation for the first quarter of the verse. Bhartṛhari asserted that the knowledge derived from scriptures alone is of absolute value, whereas that derived from other sources is of relative value. He denounced the validity of reasoning. (cf. Vakyap. I, 30; 31; 41, etc.)
Among the above-mentioned verses, fr. (1) and fr. (12) are nearly identical to each other. It is likely that the same original verse was cited in a slightly different way, as was mentioned, by the two scholars Asvabhāva and Jñānaśrībhadra, who lived in different periods. It is note-worthy that we can find the apparent Sanskrit originals of some of the fragments in the Vakyapadiya of Bhartṛhari. As for some other fragments also we find in this work verses of similar purport, as has been pointed out by the writer.résent
Moreover, we should add that a verse which might be regarded as the original of fr. (11) is cited in the Tattvasamgrahapanjika of Kamalasila. In the Sabdabrahmapariksa of the Tattva amgraha by Säntiraksita, the doctrine which admits Sabda-brahman as the world-principle is attacked. Kamalašila, the commentator, explains that the author attacks Bhartṛhari's doctrine in the verses beginning with v. 128, and that he attacks "the doctrine of the opponent"
Ad v. 144, p. 72 (ed. by Embar Krishnamacharya, Baroda 1926, GOS, 30).

Page #8

BHARTRHARI'S VERSES
129 (paramata) in the verses beginning with v. 144. As an instance of the latter, he writes as follows:
yathoktam...
yatha viśuddham ākāśam timiropapluto janah / samkirnam iva mātrābhiś citräbhir abhimanyate // tathedam amftam brahma nirvikäram avidyayā /
kaluşatvam iväpannam bhedarūpam vivartafa // ji skad du/
ji Itar nam mkhah dag pa la/ rab rib kyis bslad skye bo yin/ rab rib sna tshogs dag dan ni/ Idan pa bshin du mon par rlom // de Itar bdud rtsi hgyur med paḥi/ tshans pa hdi ni ma rig pas/ rñog pa bshin du hgyur ba ste /
tha dad no bor rnamn par hgyur // shes bśad pa Ita buḥo / 10 (In the Sanskrit original of the foregoing Tibetan version, the final word seems to have been vioartate instead of vivartatah.) Ganganatha Jha's translation: ...
"This has been thus declared :: ..
'Even though Ākāśa (Space) is pure, yet obsessed by darkness, people come to regard it as limited and made up of diversified parts; in the same manner, though Brahman is immortal and unmodifiable, yet It appears to be sullied by Nescience and hence diversely modified'.11
The foregoing verses, moreover, nearly completely correspond to two verses of the Brhadaranyakopani şad-bhāsyal2 by Sureśvara, a disciple of Sankara. (In the latter, amalan occurs instead of amytam, prakāšate instead of vivartatah upalakşayet instead of abhimanyale ; otherwise the passages completely the same.) These two verses are also cited by Prabhācandra, the Jain scholar, in a passage of the Prameyakamala-mārtanda where he criticized 13 the doctrine of the sabdabrahmavādins. (In this text, amalam occurs instead of amstam, bhedarupam prapaśyati instead of bhedarūpam vivartalan). Once again, the same verse is cited by Abhayadeva, the Jain scholar, in his work Tattvabodhavidhāyini (p. 388), as a doctrine of the sabdabrahmavādins. (amalam instead of amplam, vivar
10 The Der-ge edition, Bstan-hgyur, vol. 293, 187b.
1 The Tattvasangraha of Santaraksita with the Commentary of Kamalasila, tr. into English by Ganganatha Jha, vol. I, Baroda 1937, p. 126. * 12 Ad III, 5, 43; 41. p. 1246, Anss. 13 p. 12b, NSP.

Page #9

130
HAJIME NAKAMURA
tate instead of vivartalaḥ.) 14 Most strikingly, in the commentary (urtti) by BhattaNārāyaṇakantha upon the Mrgendrāgama, a very important, fundamental scripture of the Saivas, those verses are ascribed to Bharthari, who is characterized as a vivartavādin, as follows: 15
tathā cāha tatrabhavan Bhartshari”...
yathā visuddham ākāšam timiropapluto janaḥ/ sarnkirnam iva mātrābhiś citräbhir abhimanyate // tathedam amptam brahma nirvikäram avidyaya /
kalusatvam ivāpannam bhedarūpe pravartatt // evam cābhinnam evedam param brahma paramātmalakṣaṇam, manasām hi samsāradharmaiḥ sukhaduḥkhādibhir yogah/paramātmā tu sūrya ivāmbhahpratibimbabhedair upādhibhir abhinno'pi bhinna iva pratibhāti /
The Sanskrit text by Nārāyaṇakantha seems to be the least affected form of the original verse.16 These verses seem to be of the same purport as the Mandükya-kärikā III, 8.
yatha bhavati balānām gaganam malinam malaiḥ /
tatha bhavaty abuddhānām ātmāpi malino malaiḥ // “As the sky appears to be soiled with dirt to the ignorat, so appears Ātman, too, with impurities, to those who are not enlightened".17 The same view was shared by Sankara also.18
What is most striking from the viewpoint of the history of ideas is that verses very similar to the foregoing are given by Dignāga, the, Buddhist philosopher, in vv. 31.-32 of his Trikalapariksă as follows : 19
ji Itar nam mkhah rnam dag la! rab rib kyis- ni blad paḥi mi/ skra sad kyis ni kun gan ltar / sna tshogs su ni mnon par rtogs // 31 de Itar ḥdir yan rnam śes ni/
1. The text is not now available to the author, who has read it in E. Frauwallner's Dignāga und anderes" (Festschrift TVintemnitz, S. 237).
15 Härānacandra Sastri : Sabdabrahmavādah. (G. Jhá Commemoration Volume, p. 93.)
13 As for understanding the meaning of the verse, Anandajñāna's tikā upon the Brhad. Up. Värttika is highly valuable.
17 The Agamalästra of Gandapada, edited, translated and annotated by Vidhushekhara Bhattacharyya, University of Calcutta, 1943, p. 53.
18 S. ad Bhd. Up. p. 152, 1. 24 ... na hi bālais talamalinatādibhir vyomni vikalpyarnāne talaunalinatadivisistam eva paramarthato vyoma bhavati (S. ad BS. I, 2, 8. vol. I, p. 18+, II. 5-7). vyoinniva talamaladi parikalpitam (ś. ad BS. I, 3, 19. vol. I, p. 279, 1. 7.). ākāse balās talamalinatādy adhyasyanti (s. ad BS. I, 1, 1, vol. I, p. 1l. Anss). Upadeśasahasri, II, 18, 22.
1) The author has cited the verse from the above-inentioned Frauwallner work.

Page #10

BHARTRHARI'S VERSES
131
mam par mi hgyur ma rig pas/ chu rnog bshin du shugs pa ni /
tha dad gzugs su rnam par hjug // 32 (Somc noteworthy differences: skra fad kyis (kesondukaih) instead of mätrabhih; Adir yar Tham ju ni (atrapi vijñānam cva) instead of idam amplam (amalam) brahma.)
Those verses cited in Tibetan versions should be considered as having probably been ascribed to Bharthari. Reasons: (1) Tibetan versions mention the name of the author of the verses as Bhadrahari, Bhadrihari, Bhatahari, Bharitehari or Bhandrihari. These names give ample testimony to the supposition that the original Sanskrit name of the author of those verses was a little difficult to transcribe with Tibetan characters, and so the name must have been Bhartshari which could easily corrupted into any one of them. b.
(2) We can find in the Vakyapadiya by Bharthari the Sanskrit original texts of at least 10 of these verses. As for some of the other verses we find similar verses in the same work.
(3) Bhatta-Nārāyanakantha ascribed (fr. 11) to Bharthari. Although some of the above-mentioned verses can not be traced in the Vakyapadiya, the BhartȚhari ataka or in collections of lyrical songs ascribed to him, it is possible that he might have composed 2 other works which are not extant now or that some of the verses might have been ascribed to him in later days.
There are on the other hand some difficulties in identifying Bhadrahari etc. with Bhartshari. However these difficulties will be easily solved by the following considerations :
(1) It is a well-known fact that Bharthari adopted the standpoint of emanation-theory (pariņāmavāda), whereas the standpoint made clear in (fr. 11) is, as Nārāyana-kantha asserts, manifestation-theory (vivartavāda). These two standpoints were strictly distinguished from each other by later Vedāntins.22 It seems that the author of (fr. 11) would be different from Bhartshari, the author of the Väkyapadiya. However, in the days when these verses were composed these two philosophical standpoints were not strictly distinguished from each other. Bhartshari himself used the two terms (pariņāma and vivarta) as nearly the same meaning, 23 just as the Brahma-sutras used many similes which
30 The Epigrams Attributed to Bharthari, ed. by D. D. Kosambi (Singhi Jain Series, vol. No. 23). Bombay 1918. This work contains not only the famous three Satakas, but also the Vifatpita, Vijñānajataka and many other poems ascribed to him.
Il Professor Louis Rcnou told the author that, according to the information by the late S. Dasgupta the editions of the Vakyapadiya published hithertofore do not contain all the verses of the work, and so it might be possible that those verses not identified by the author are found in unpublished MSS.
Paul Deussen: Allgemeine Geschichte der Philosophic I, I, 4 Auri., Leipzig 1920, S. 63 f.
The author discussed the problem in a Japanese work of his (IT109. D IA'S DEDI p. 2397). Cr. Paul Hacker: Vivarta, Akademic der Wissenschaften und der Literatur. Abhandlungen der Geistes- und Sozialwissenschaftlichen Klasse, Jahrgang 1953 Nr.5.

Page #11

132
HAJIME NAKAMURA
could be interpreted in different ways.
(2) Hithertofore it has been generally supposed that Bhartphari died forty years before I-tshing sojourned in India, following the information given by I-tsing himself.2. Accordingly Bharthari was regarded by J. Takakusu to have died in 651-652 A. D. If we examine this passage more closely, however, it becomes clear that in I-tsing's information there is something amiss. I-tsing said that Bharthari "was a contemporary of Dharmapāla”, 25 and yet also that Dharmapāla composed a commentary upon Bharthari's verses "26 These two are not consistent with each other.) Let us discuss this problem. ' : Dr. Hakuju Ui?? took the passage as meaning that Bharthari died forty years before the beginning of I-tsing's stay in the Nālandā monastery. According to his calculation Bharthari died c. 630 A.D. In the above-cited passage, however, he is said to have been a contemporary of Dharmapāla who commented upon the Prakīrņaka of Bhartshari. The date of Dharmapāla is wellknown; he lived 530-561 A. D. Now, Bhartphari must have been somewhat Dharmapala's senior Suppose that the former was older than the latter by, say, ten years; then Bharthari should have lived c. 520–630. If we adopt Takakusu's view, he must be supposed to have lived c. 520-652. Needless to say, this supposition is absurd. We are brought to the conclusion that either (1) the information that "Bhartphari died forty years before I-tsing" is wrong or (2) the information that Dharma pāla commented upon Bharthari's work is wrong. We shall take up this problem in another light. Mr. H. R. Rangaswamy Iyengar28 made clear the fact that Bhartshari lived in an age not so remote from that of Vasubandhu. He asserts as follows:
"In the second Kānda of the Vakyapadīya, while describing how the science of grammar, which had been almost extinct, was restored and propagated by the great grammarians, Chandra and Vasurāta, Punyarāja, the commentator of the Vākyapadiya, mentions several times Vasurāta as the teacher of
Bharthari.29 In the Kärikä 490 of the Vakyapadiya 30 Bharthari himself seems . to refer to his teacher Vasurāta by 'Guruņā'as is evident from the words of
"It is forty years since his death (A. D. 651-652)." (J. Takakusu: A Record of the Buddhist Religion as Practised in India and the Malay Archipelago. Oxford 1896. p. 180) 25 J. Takakusu: op. cit., p. 179.
WinPINA 9, Taisho, vol. LIV, p. 229. 27 T55- vol. V, p. 130.
28 H. R. Rangaswamy Iyengar: "Bharthari and Dignaga" (Sri Atmananda Prakasa, published by " Jain Atmananda Sabha ", Bhavnagar, 1952, p. 27 f. The author obtained a copy by courtesy of Mr. Muni Jambuvijay).
1) na ten stadguros tatrabhavato Vasurătâd anyah kaścid imam bhäşyârnavam avagâhitum alam ity uktam bhavati (Punyarāja ad II, 486). kenacic ca brahınaraksasaniya Candracarya-Vasuritaguruprabhrtinim datta iti/te (taih?) khalu yathāvat vyákaranasya svarūpam tata upalabhya satatan ca siyanām vyākhyāya bahuśākhitvam nito vistararn prăpita ity anuśruyate (ad II, 489).
30 CC. II, 490b:... pranito gurunāsınākam ayam agamasamngrahah.

Page #12

BHARTṚHARI'S VERSES
133
Punyarāja prefaced to the verse." Again a Jain writer, Simhasürigani, who may be assigned to the beginning of the sixth century A. D., in his unpublished work, Nayacakratika, a commentary on the Nayacakra of Mallavadin the senior," which is not now extant, mentions, twice in his work, Vasurata the Upadhyaya of Bhartṛhari.33) This confirms the statement of Punyaraja and establishes that Vasurāta was a great grammarian of the day under whom Bhartṛhari studied and that Bhartṛhari often held views quite different from those of his master.
According to Paramartha, Vasurata was a Brahmin and brother-in-law of Baladitya, a pupil of Vasubandhu. He was well-versed in grammar. He defeated Vasubandhu, through the intervention of Chandra, another great grammarian. This means that Vasurata, Chandra, and Vasubandhu should be regarded as contemporaries and Bhartṛhari, the pupil of Vasurata, assigned to the fifth century A. D."
Another piece of evidence of much more importance was discovered by him. In the fifth chapter of Dignaga's Pramana-samaccaya, which is devoted to the exposition of the Apoha theory, the following two Kārikās are found:
thigs pa dan ni tshogs pa yi/ chu sogs rnams la rjod byed ni/
grans dan tshad dan dbyibs rnams la/ Itos pa med par hjug par byed // dbyibs dan kha dog yan lag rnams/ khyad par can la gan hjug pa / de yi yan lag la sgra ni/
rab tu hjug la dun asma yi //
[The above-mentioned Tibetan verses have been cited from Mr. Iyengar's article. According to the Sde-dge edition of the Tibetan Buddhist Canons kept by Tohoku University, Sendai, the sentences run as follows:
la lar dños su rjod par byed de dper na /
grans dan mtshan ñid dan ni dbyibs/ Itos pa med par rab tu hjug/ chu la sogs pahi thigs pa dan/ hdus pa la yan rjod par byed//
:
31 Cf. ibid. atha kadacit yogato vicārya tatra bhagavata Vasurataguruņa mamayam agamaḥ samjñāya vatsalyāt praṇita iti svaracitasya granthasya gurupūrvakam abhidhātum äha.
3: This Mallaradin should be distinguished from another Jain writer of the same name who is the author of Nyayabinduṭippani.
33 Cf. Nayacakratika, folio 272a: so 'bhijalpo 'bhidheyarthaparigrahi bahyaç chabdad anya iti Bhartṛharyadimatam / Vasuratasya Bhartṛharyupadhyayasya matam tu... Cf. ibid., folio 277a: evan tavat Bhartṛharyadidarśanam uktam / Vasurătaḥ Bhartṛharer upadhyāyaḥ
Cf. "A study of Paramartha's life of Vasubandhu and the Date of Vasubandhu" by J. Takakusu JRAS, 1905, pp. 33 ff.

Page #13

HAJIME NAKAMURA
la lar yan lag ubah shig la hjug pa ma yin te / dper na /
dbyibs dan kha dog cha sas kyis / khyad par byas nas hjug pa yin/ sgra yis de yi cha śas la /
rab tu hjug pa drigs ma yin// A restoration into Sanskrit of these verses might casily correspond to the following verses in the Väkyapadiya:
bindau ca samudāye ca văcakah saliladişu / samkhyāpramānasamsthānanirapeksaḥ pravartate// (Vakyap. II, 160.) 35 samsthānavarnāvayavair visiste yah prayujyate / śabdo na tasyāvayave pravșttir upalabhyate // (Vākyap. II, 157.)
This means that either Dignāga took the Kārikās from Bharthari's work or both Dignāga and Bhartrhari took them from quite a different work. But there is no evidence to support the latter alternative. We learn from Jinendrabuddhi, author of Viśālāmalaţikā on the Pramāņa-samuccaya-urtti of Dignāga, that Dignāga, is here referring to the views of Bharthari.3s ) This evidently supports the former alternative that Dignāga is quoting from Bharthari.
The foregoing are the pieces of evidence set forth by Mr. H. R. Rangaswamy Iyengar. But whereas he has pointed out just one Tibetan citation of a verse by Bharthari, we have in addition pointed out many citations of his verses. Taking all these pieces of evidence into consideration, we have to conclude that Bharthari must have lived prior to, or at latest contemporary with, Dignāga, Asvabhāva (No-bo-nid med-pa), Bhavya (Bhāva viveka) and Dharmapala.
There has been much dispute about the date of Dignāga.37 Randle once said that “all that can be said with certainty is that he lived somewhere between 350 A. D. and 500 A. D."38 According to most scholars he lived about 500 A. D.39 Dr. H. Ui fixed his date as c. 400-480, on the ground that he
35 The first and the second halves are given in reverse order in the edition of the Benares Sanskrit Series (p. 185)
af Cf. Visälamalaţiki : Mdo. re, folio 331 b. 1. 6 ff.: kha cig tu gtso bor cha tshas rnams la hjug te / bha rite haris yis smras pa / cha sogs rnams la zes paḥi sogs pahi sgras. sa la yois su gzui no/(=kva cit tu mukhyā avayaveșu vsttih/yathoktan Bharthariņa salilădişv iti adisabdena Prthivyādinām parigrahah )
37 According to Taranátha, Dignåga was a pupil of Vasubandhu (Taranátha's Geschichte des Buddhismus in Indien, übersetzt von Anton Schiefner, St. Petersburg 1869, S. 131).
33 H. N. Randle: Fragments from Dinnaga. London 1926, p. 3. In another work he says: "Dinnãga's date shares the uncertainty attaching to that of his master Vasubandhu. IIc may fall anywhere between 400 and 500 A. D." (Randle: Indian Logic in the Early Schools. Oxford University press, 1930 p. 27).
39 Moritz Winternitz: Geschichte der ichenndis Litteratur, Bd. III. Leipzig 1920. S. 167. Louis Renou et Jean Filliozt: L'Inde Classique, tome II. Paris 1953, p. 380.

Page #14

136
HAJIME NAKAMURA
At the end of this article the author should not fail to mention the names of the scholars who kindly helped him in carrying on this study. Prof. Yamaguchi introduced him to Dr. Kensho Hasuba and Dr. Kyōgo Sasaki, who went over No-bo-nid med-pa's works in order to discover the suspected citations of Bhartshari's verses. Prof. Josho Nozawa kindly informed him of Bhavya's citation of a verse of Bharthari. Dr. Bunkyo Aoki and Prof. Shinya Kasugai did not spare efforts to give the author valuable suggestions. Here the author wants to express his sincere gratitude to these scholars, without whose kind help this article could not have been brought to its present state of completion. P. S. Referring to the Vijñānavāda, Bharthari says: 'kecid vyāvșttirūpām tu dravyatvena pracaksate/(Vākyap. 3, 1, 19. p. 22.)
This must refer to later Buddhist Idealists, not to such earlier ones as Asanga and Vasubandhu. If so, the date of Bharthari should not be fixed in a period much earlier than was set forth above.
The Sanskrit original of the above-mentioned fragment No. 5 is found in the commentary by Vrşabhadeva upon the Vākyapadiya. As the text is not now available to the writer, he wants to cite Dr. Hacker's explanation: “In einer Reihe von 12 Strophen, die der Kommentar zitiert, befindet sich die folgende (p. 7,1–2):
Vyalito bheda-samsargau bhāvábhävau kramå kramau Satyányte ca visvátmá pravivekāt prakasate,"
(Paul Hacker: Vivarta. Studien zur Geschichte der illusonistischen Kosmologie und Erkenntnistheorie der Inder. Akademie der Wissenschaften und der Literatur. Abhandlungen der Geistes- und sozialwissenschaftlichen Klasse, Jahrgang 1953. Nr. 5. S. 205.) Dr. Hacker takes this anonymous verse for a later development. Taking the above-cited Tibetan version into consideration, however, the present writer thinks that this verse also should be ascribed to Bharthari. He expresses his sincere gratitude to Dr. Hacker who kindly sent him copies of his valuable works.

Page #15
BHARTRHARI'S VERSES 135 must have lived a little earlier than Gunamati (1992).40 The latter must have lived prior to Paramartha's arrival in Kuang-tung in China in 546 A. D. and carlier than Sthiramati. Sthiramati must have lived in the sixth century 42 or according to Dr.Ui, c.470_550.43 Winternitz assigns him to the fifth century.44 Recently Professor Hikata has surmised that the date of Dignaga is about 440 520.45 It is generally admitted that Asvabhava lived earlier than Dharmapala and later than Dignaga. He lived, according to Dr. Ui, c. 450-530 A. D., 46 and, according to Prof. Hikata, c. 470-550 A. D.47 Bhavya was almost contemporary with Buddhapalita (c. 400-450?).48 He is regarded to have lived and worked at the beginning of the fifth century A. D.49 According to Dr. Ui, he lived c. 490-570.50 In any case he was an elder contemporary of Dharmapala. The date of Dharmapala (?) has precisely been fixed. He lived 530-561 A. D.51 So we can assuredly say that Bharthari must not have lived later than these Buddhist scholars. It has become clear that the statement of I-tsing that Bhartshari died some forty years before the date of his record is incorrect, It is, however, doubtless the case that Bhartphari lived after Vasubandhu, as is clear from the above-cited materials. Vasubandhu is generally regarded to have lived in the fourth century,52 or in the latter half of the fourth century:53 According to Dr. Ui, Vasubandhu lived c. 320~400.54 According to Prof. Hikata, he lived c. 400-480.55 Frauwallner asserted the existence of two Vasubandhus, ascribing the elder to 320/380 A. D. and the younger to 400480 A. D.56 Such has been the extent of the debate about his date. So, taking these facts into account, we are brought to the conclusion that Bhartshari, the author of the Vakyapadiya, must have lived c. 450-500 A. D. 40 * BIDW *', vol. V, pp. 142-145. 4 Cf. ibid., p. 136. . * Filliozat: op. cit., p. 380. - H. Ui: op. cit., p. 136. HM. Winternitz: History of Indian Literature, vol. II. University of Calcutta, 1933. pp. 362-363. 45 FS ALE: LEF# (CEDEPE MEL E . U os , h t , p.321). 45 H. Ui: op. cit., p. 147. 7 R. Hikata : ob. cit., p.321. 18 J. Filliozat: op. cit., p. 379. 49 M. Winternitz: History, etc. II, p. 362. 50 H. Ui: op. cit., pp. 148-149. 51 H. Ui: op. cit., pp. 128-132; R. Hikata : op. cit., p. 321. 52 M. Winternitz: History, etc. II, p. 355 f. 53 J. Filliozat: op. cit., II, p.380. 5. H. Ui: "op. cit,", vol. I, pp. 413-414. 83 R. Hikata : op. cit. pp. 315 f. 55 E. Frauwallner: Or the Date of the Buddhist Alaster of the Law Vasubandhu. Roma 1951. All different views about Vasubandhu's date have been mentioned in this work. . 11, p. 355 f. 51 H. Vi ut: op. cit., Ir