INDIAN STUDIES

LONDON . HUMPHREY MILFORD

OXFORD UNIVERbl'l'Y PRESS

0

INDIAN STUDIES

Jn l,onor of

Charles Rockwell Lanman

HARVARD UNIVERSITY PRESS

Cambrrbgt, Jttassa:cbusttts

1929

COPYRI , 19119 ,,. BY THE PRESIDENT FELLOWS {ilf HARVARD COLLEGE _,.,

PRINTED AT THE HARVARD UNIVERSITY PRESS

CAMBRIDGE, MASS , U S A

THESE STUDIES, THE WORK OF COLLEAGUES

AND FRIENDS IN MANY LANDS, ARE OFFERED

IN GRATITUDE AND AFFECTION TO

~barlt~ l\odt\tltll 1.anman

CONTENTS

ARRANGED ALPHABETICALLY BY AUTHORS

DINES ANDER',EN

Miscellanea Pahca

SHRIPAD Km•mNA IlELVALKAR

Srngsnc Eldborat10n m Sakuntala, Act III

MAURICE BLOOMFIELD

D1mmutive Pronoun~ m the Native &rarumars .J

PAGE

81

187

7

WALTER EUGENE CLARK 217 Hmdu-.\rah1c N umerah, .J

CAROLINE A FOLEY RHYS DAVIDS ' ?

The Well-TO EY

/FRANKLIN EDGERTON

Notes on Jama MahJ.rJ.~tri J

CARLO FORMH Ill

On the Real l\leanmg uf the Dialogue beh~een l aJndvalkya and Ma1treyI

KARL FRIEDRICH G.M,DNER

Das V 1pii.ndm 1m R1gveda

Sm GEOIWE AuRAHAM GRIERSON

The Birth of Lorik

EDWARD WAsttnvnN HoPKJ~S

Buddlust1c Myst1usm

. J i/A VALENTINE "ILLIAMS JACKSON

Three ln<lo-lramdn Notes

HERMANN JACOBI

Mimiimsii und Va1ses1ka

JULIUS JOLLY

Ueber die Spatere Entw1cklu11g des lnd1schen Staatsrechts

ARTHUR BERRIEI>ALE KEITH

Dan<,lm and Bhiimaha

103

75

69

243

113

255

145

237

167

viii Contents

TAIKEN KIMURA 89 The Date of Vasubandhu seen from the Abh1dharma-Ko§a

STEN KoNow 58 Remarks on a Kharosthr lnscr1ption from the Kurram Valley v

SYLVAIN LEVI 85 L'lnscr1ption de Mahliniiman a Bodh-Gaya

PAUL MASSON-OURSEL 141 L' Autonom1e Spmtuelle selon la Pensee lnd1enne

ANTOINE MEILLET 8 La Flex1on Panthah en Ved1que, et les Nommat1fs en -ea du Latin

GENMYO ONO 93 The Date of Vasubandhu seen from the History of Buddh1st1c Philosophy

Louis DE LA VALLEE PoussIN 135 Extase et Speculation

EDWARD JAME~ RAPSON 49 The Date of the Amohmi Votive Tablet of Mathurii

ARTHUR WILLIAM RYDER

How to Live Happily on N othmg a Year

JYAN TAKAKUSP 79 The Date of Vasubandhu, the Great Buddh1:,t Plulosopher

FREDERICK '\V1LLIAM THOMA~ 198 A Ramiiyana Story m Tibetan from Chme~e Turkei,tan

HAKPJU U1 95 Ma1treya as an H1stor1cal Personage

JAMES HAUGHTON '\VooDs 187 Integrat10n of Consciousness m Buddhism

Cont,ents

GROUPED BY SUBJECTS

LANGUAGE AND GRAMMAR

La Flexion Panthah en Vedique et les Nominatifs en -es du Latin ANTOINE MEILLET.

Diminutive Pronouns in Jama Sanskrit MAURICE BLOOMFIELD

Notes on Jama Maharar;,tri FRANKLIN EDGERTON

Miscellanea Pahca DINES ANDEUSEN

INSCRIPTION<;

L'Inscr1ption de Mahanaman a Bodh-Gaya SYLVAIN Livi

The Date of the Amohmi Votive Tablet of Mathurii EDWAUD JAMES RAPSON

Remark!> on a Kharo.,thi Inscription from the Kurram Valley STE?. KONOW

YEDA

IX

3

7 --27

31

35

49

53

Das Viplinam im Rigveda 69 KARL FRIEDRICH GELDNER

On the Real Meanmg of the Dialogue between Yajnavalkya and Mmtreyi 75

CARLO FonMICHI

BUDDHISM

The Date of Vasubandhu, the Great Buddhist Philosopher Jy AN TAKAKUSU

The Date of Va.,ubandhu seen from the Abh1dharma-Kosa 'l'AIKEN KnrnUA

The Date of Vasubanrlhu seen from the History of Buddhistic Phi-

79

89

losophy 93 GENM\O ONO

Maitreya as an H1stor1cal Personage 95 liAK.u1u U1

The Well -TO EY lOS CAROLINE A FOLEY RHYS DAVIDS

x Contents

Buddh1sttc Mysticism 118 EDWARD w ASHBURN HOPKINS

Extase et Speculation 185 Louis DE LA VALLEE PoussrN

Integration of Consc10usness m Buddh1,;m JAMES HAUGHTON Woons

COMPARATIVE RELIGION

L' Autonom1e Spmtuelle selon la Pensee Indienne p A UL MAS80N-0UR'IEL

CLASSICAL PHILOi>OPHY

Mimiimsii und Vaises1ka HERMANN JACOBI

CLASSICAL POETRY

Dandm and Bhamaha ARTHUR IlERRIEDALE KEITH

I, ,,, Srnganc Elahorat10n rn .:,akuntala, Act III

SHRIPAD Km-.HNA BEuALK-\R

A Ramayana Story m Tibetan from Chmese Turkestan FREDERICK \YrLLIAM Tno\HS

How to Live Happily on N othmg a Year ARTHUR WILLIAM RYDER

l\IATHEl\lA TICS

Hmdu-Arab1c Numerals w ALTER EUGENE CLA!tK

LAW

Ueber die Spatere Entwicklung des lndt'>chen Staatsrechts JULIUS JOLLY

VERNACULAR

The Birth of Lonk Sm GEORGE ABRAHAM GRrERbON

lNDO-lRANIAN

Three lndo-Iraman Notes A VALENTINE \V1LLIAMS JACKSON

187

141

145

167

187

193

213

~17

237

INDIAN STUDIES

LA FLEXION DE PANTHAII EN VEDIQUE, ET LES NOMINATIFS EN -ES DU LATlN

BY ANTOINE MEILLET

IL suffit de se reportc-r a la deuxi<'me edition du Handbuch der latei­nzschen Laut-und Formenlehre de M. Sommer, pp 371 et smv , pour

c-onstater que la qut'1,bon de l'or1gme des nominabfs lahns en -es, du type de sedes, n'c>st pas resolue

A rapprocher le type latm uolpes du ht liipe, comme l'a fait M. W Schulze, on ne gagne r1en le type htuamen eio>t un ancien type en -ic-. ht zemc repond a v sl :emlJa Au !'.'>Urplus le baltique a -e et non -es. Et, s1 l'on rnpproche fide.'I de ~cdcs, comme l'a fa1t Brugmann, on ne comprend pas ce qm peut Hre exphqu(, par la sedes, sedis et .fides, fidei ont des flexions entierenwnt d1stmcte!,,

Le vocalu,me rnontrt' que scdes n'a m'n a faire av(•c gr too5, ni moles avec molestus Du reste aucune forme de nommabf <le theme en -s­tels que io>eraient shies et moles n 'ex1stt• dans aucune langue indo­europrenne, It' latm lui-rneme ne connait que le type honos Quant a pubes, nen n'auton!,,e £i, y chercher un th&me en -es- pilbcr a un ancwn r, ti·mom pubertiis

L'hypothese que l\I Sommer met m pn•n11e1e hgne, a !'.'iavotr un amcwn nommat1f en -£~(,)s de thc\me en -t- !'.'iera1t plaus1blC' 1\ila1s M. Somnwr ne s1gnak•. hors du latm, nen de comparable Or, pour l'exphcahon d\nw forme, ce qm nnporte avant tout, c'est de trouver une forme qm Im c-orre-,pon<le

II y a en mdo-uanien un nom dont la flexion rappellt' le type scdes. C'est celm dont, en vedique. le nommatif singuher est pti.nthah, l'ac­cusatif smguher panthiim. Ie nommabf plunel panthiih - ]'instrumental plurwl pathfbhih, le locabf pluriel pathirn - le gembf-ahlahf singuher pathah, l'mstrumental singuher patha, l'accusahf plunel pathti.h, le ge­mtif plunel pathdm Il n'y a pas a tenir compte des formes en -n­de l' Atharvave<la le nonunabf plunel panthdnah qm se trouve sept fois, est une format10n analogique grace a laquelle eta1t ev1tee la con­fusion avec le singuher panthah, a l'accusatu Slnguher, ou aucune con­fusion n'ava1t heu, !'innovation n'a pas prevalu panthanam se lit une seule fo1s contre 19 pantham (snr tous ces fa1ts, v. la precieuse etude de M. Lanman, Noun-inflection in the r eda, pp 441 et 470). Les monu­ments iraniens confinnent le caractere indo-iranien des formes vediques:

4 Antoine M eillet

l'Avesta a panta au nominat1f singulier, pantqm a l'accusatif et pa0qm au genitif pluriel, pa06 au gembf srngulier et a l'accusatif pluriel; la fonne en -i- est attestee mdirectement par l'accusatif singulier paOim du v1eux-perse, qui montre une normalisation de la flexion dans le parler deJa tres evolue qu'est le perse des inscriptions achemenides; d'autre part, on voit, par un nominatif pa.id, par un accusatif pa0qm des giithii, que la flex10n fortement anomale du mot s'est alteree de bonne heure en iramen

L'exemple de pdnthiih est umque en indo-1ranien Carved. rruinthiih, dont on n'a la flex1on que parbcllement, apparbent a une racine dis­syllab1que, I' -ii- a chance d'appartenir a la racine, et le cas est, par suite, d1ff{>rent de celm de panthiih

Hors de l'rndo-1ramen, une flex10n du nom du " chemin " pareille a celle de ved panthah n'est pas attestee. En slave et en baltique, l'his­to1re commence trop tard pour qu'une flcxion auss1 archa1que smt con­servee Toutefo1s, le contra&te de v sl pQtif " chemin " et de v pruss. pintis " chemin " montre que le JCU du vocahsme radical observe <lans vfd. panthiih, pathah, pathisu a ex1ste clans le domame slave et baltique. On entrev01t par Ii\ que la fle-x10n connue par l'mdo-iramen y a existe auss1. - Gr. 1ra.rns "chemin "et 1rovros "mer" sont des themes dfr1ves par lesquels on aurait evitc le v1eux nom anomal du " chemin " La parentc avec skr pdnthah, etc , e&t du reste mcertame -L'ann. hun "chemin" n'enseigne rien -Quant a lat pons, dont le sens ne con­corde pas exactement avec celui de !-ikr p<inthiih et de v sl. pQtt, et dont par smte le rapprochement est conteste (L Havet a toujours soup<_;-onne, a tort ou a raison, qu'un rapprochement avec le groupe de pendo pendeo, ou le -d- est seconda1re, rendra1t compte du mot), la fiex1on est la meme que celle de mons et de mens, et 11 n'y a, en tout cas, rien a en tirer 1ci - Le mot n'est clairement atteste que dans la parbe orientale de l'indo-europeen en indo-iranien, armemen, slave et bal­t1que.

Deux traits caracterisent le type latrn a nominatif en -es-. Le premier est que des fonnes en -i- s'y 1uxtaposent a des fonnes de la flexion con­sonan tique et au nommabf en-es-· le mot uates, uiitem a ungembf plur1el uiitum (a cote de uiitium), tout comme le vedique a pdnthah, panthiim, ma1& pathdm, pathisu. Et la formc en -i- est attestce par un nominabf uiitis chez Plaute et par la flexion reguliere en -i- de 1rl. faith. - En face de ,,edes, sedem, le genitif pluriel est sedum. - La concordance du type latin avec le type mdo-iranien de skr. panthiih est complete.

Le second trait est celm-ci · clans ved panthiih, pdnthiim en face pathah, pathhu, l'alternance vocalique est la meme que clans ddn,

La Flexion de Panthah en Vedique, et les Nominatifs en -es du Latin 5

ddntam, en face de datah, datsu. Or, la voyelle longue radicale de lat. sedes, mJij[es, en face de sedeo, molestus, s'explique seulement au nomi­natif singuher La longue a ete generahsee en latm ou les alternances vocaliques ont ete beaucoup simphfiees, mais on aperc;oit un vocalisme

,de nominatif singulier, reste d'une alternance comparable a celle qu'of-frent les formes vediques.

Si le rapprochement propose est exact, la flexwn de uiites, uatum, uiitibus et de :~edes, .'ledum, scdibus, pareille a celle de ved. pdnthiih, patham, remonte a l'indo-europeen. On peut &e dispenser de rechercher comments' est forme le type c' est un probleme de prehistoire de l'indo­europeen, les donuees sont trop peu nombreuses pour qu'il soit sage de l'aborder.

En indo-framen, ved. panthah est une survivance unique En latin au contraire, le type sede.~ est repr(·sente par un bon nombre d'exemples; v. Neue-\Vagener, Latcmische Fonnenlehre, I p. 400 et smv. Ces exemples sont pour la plupart nouveaux et tiennent a la d1fficulte que faisait le nommat1f des themes consonantiques. Le latm a pare a cette difficulte dans beaucoup de mots a l'aide de -i-, et II a iuuenis en face de gen plur iuuenum, cf skr y(wan-, ou apis en face de apum, etc. Ailleurs 11 s'est servi de -cs, et c'est ainsi que, a cote de plebs, il a plebes; a cote de nubs, nubes. Du theme ca11-etabh par abl sg cane, gen plur. canum, 11 y ales <leux fonnes de nonunatif, l'une usuelle cams, pare1lle a iuuenis, l'autre canes, rare, pare1lle a uiifes

Si lat pons est apparente a skr panthiih - cc qui demeure probable - on s'explique ams1 le nommat:u pons au hcu de la forn1e ancienne le type en -es- de sedes, uiites ne s'est mamtenu ou etendu que h\ ou il etait utile pour ev1kr de1, formc1, obscures et courte-,, tdles quc sera1ent *ses, *uiis, or un anc1en *pontis, qm a passe a pons, ne faisait pas plus de drfficulte que *mcnti.'I (mcns) ou *montis (mons).

Commc H s'est produit des mterferences entre le type *uiitcs, *uatom (gen. plur ), *uati-, 1dent1que au type ved panthah, pathiim, pathfsu, et l'ancien theme consonantique *sed- (norn d'action radical de *sed ) qui a rec;u un nonunatif sedes par smte de l'obscurite qu'offrira1t un nomi­natif de la forme *ses, il est impossible de faire un depart entre les origines des divers subs tan bf s latms a nommahf en -cs L' etymologie ou la formation de la plupart sont du reste inconnues 11 serait malaise d'expliquer pourqu01 l'on a niibium en face de nubs, niibcs (niibum etant exceptionnel), tand1s que l'on a sedum. lei comme a beaucoup d'autres egards, le latm offre l'aboubssement d'un developpement com­plexe, traverse par des influences multiples, dont le detail ne se laisse pas restituer.

6 Antoine M eillet

Abstraction faite des innovations analogiques plus ou moins recentes qm ont trouble la nettete du type, le latin conserve ici, plus largement que l'indo-iranien, une flexion qm caracterise bien la complication de la morphologie mdo-europeenne Atteste tardivement, le latm n'a pas l'archaisme general du vedique, les alternances de la flexion de ved panthah ne sauraient s'y retrouver. Mais, au lieu d'un exemple umque du type, ii en a toute une serie. Les exemples du type uates, uatum sont de ces survivances par lesquelles le latin est demeure si instructif pour l'h1stor1en des langues mdo-europeennes.

Note additionnelle - L'article c1-dessus a ete envoye aux orga­nisateurs du recue1l longtemps avant la publication de l'etu<le de M H Pedt.>rscn, La cmq111eme declmaison latwe, Copt>nhagut.>, Hl2G Depms, 1\1 Wackernagel a, ~\ son tour, trait<'.· de skr panthah, K Z , vol LV, pp 104 et smv (1927) Vo1r au'lSI Stolz-Leumann, Lat Oramm (IH28), p 232 Sur plus1eurs pomt& e&-.entiel1>. Je sms heureux dt.> constater que mes vue'l concordent avec celles de ct.>s emm<>nts lmgmstes A certains egards, la doctrme exposee c1-dessus <'n diff ere, il e1>t 1mposs1hle d'entrer1c1 dans une d1'lcuss1on qm sera1t gauche Je note seulement que le vocahsme, surement anc1en, de ved panthdh, pantham, av pant&, panfqrn ne se comprendra1t pas l-11 l'a Hait un flement essent1el du mot; car l'element' radical sera1t alors au degre zero CommC' dans lat uatc.v, etc , l' -e.'J final du nommahf est un eleme-nt advenhce Des lors.l' 1 de ved pathibh1h pafhisu n'a pal-I a Hre ccms1<lere comme Hant t•n alternance avec l'a de panthah Le genre femmm de v pC'rS<' pa01m ne prouve pas que l'on ait 1c1 un derive en + tout nom de genre amrne peut etre ou masculm ou femmrn en rndo-europet.>n !>mvant la fa<;on dont 11 eta1t corn;u, or, la notion de la "route" pouva1t etrc con<;ue comme fcmmme, amsi que le montre avec evidence gr bobs, femmm, cf aussi lat uia, etc SL pQit dmt etre un ancien theme en *-i-, car on n'y signale aucune trace de flexion consonantique

CoLI..EGE DE FRANCE

ON DIMINUTIVE PRONOUNS IN JAINA SANSKRIT

BY MAURICE BLOOMFIELD

AREADER of AJitaprabhasiiri's <;antinatha Caritra 1 cannot fail to notice the large number of pronouns with suffixed or m:6.xed ka

which on closer inspection show some kind or other of d1mmutive func­tion This text of 4890 c;lokas con tarns sixty cases, and such cases appear, to be sure much more sporadically, m other narrative Jama texts, begin­ning with Hemacandra (Tn~astu;alakapuru"a Cantra, and Paric;ista­parvan) where they are quite rare For example, Hemacandra's <;an­tmatha Cantra (Tri'la~ti, Fifth Parvan), contaming 2143 1_;lokas, does not show a smgle diminutive with ka, as contrasted with the sixty cases of AJ1taprabha's treatment of the same theme

The quas1-biografies treated by Hemacandra m his great chromcle, the Tri~asb and its appendix, the Pam,i~taparvan, as well as other per­sonal chromcles (Car1tras and Prabandha::,) not touched upon in these two text::., are taken up by the later literati (Siin'>) with a degree of eagerness that would be furious, if 1t were not so pious Especially, each of the 24 Jama Saviors 1s agam and agam the theme of a Car1tra (Canta), Mahiikiivya, Kathanaka, or Purana, whose author - with all due and boundle::,::, respect for Hemacandra, "the all-knowing m (this) age of bra::.s" 2

- seem::, to be nnpellcd by the desire " to go him one better " I have listed a goodly number of sueh works on Parc;va­natha, at the begmmng of my d1ge1,t of Bhavadevasiir1's Parc;vanatha Car1tra, pubh5hed under the title of "The L1f e and Stories of the Jaina Sav1or Par<;vanatha" (Baltimore, The Johns Hopkins Press, 1919) 3 The "history" of the sixteenth Sav1or <;;antmatha is not only written by Hemacandra and A]Itaprahha, but I have also m my hands a <_;antrnatha Mahakfn•ya by Mumbhadra siir1, an elaborate work in nineteen sargas and 4349 i.tanzas 4 Guermot' s Essa1 de Bibliographic

1 Published by the Jamadharmaprasiir.ikasabhii at Bhavnagar m 1917 (Virasam­vat 2443, V1krama~amvat 1973)

2 KahkiilasarvaJf'lai;rihemacandra 3 Hertel, m the lntro<lucbon to his Tramlabon of Hemav1Jaya's Kathii.ratnakara,

p. xv (Le1pz1g, 1920), mentions m addition a Piir~va-<'antam by Hemav1Jaya, d also lus 'lnd1sche Erzahler.' vu, 173, and Charpentier, ZDMG, lx1x, 321 ff

4 Edited, as nr 20 of Yashov1Jaya Jama Granthamala, by Shravak Pandit Har­govmddas and Shravak Pandit Bechar<las, Benares, Virasamvat 2437 (Ao 1911)

8 Maurice Bloomfield

Jaina reveals other (Jantinatha Caritras one by Sakalakirti (p. 88); another by Devacandrastiri (p. 84); 1 and a (Jantinathapura:r;ia (p. 55). In the same way Hemacandra's brief account of the female Savior Mal­linatha, the nineteenth Tirthamkara, in 266 stanzas, 2 Is followed by Vinayacandrasur1's Mallmatha Car1tra m eight sargas containing no less than 3783 stanzas. 3

The chief d1ff erence between Hemacandra and his successors is that the latter have taken to "enwombmg" -garbhita as the Ilindus say 4 - more or less lengthy stories from the general stock of Jaina :fiction, m illustrating the moralities which the Arhats and others are made to preach in the course of their spiritual careers These rather extraneous stories often really make up the bulk of the book, the bio­graphy of the hero holds together these stories by a very slender thread Thus the long story of Nala and DavadantI. 0 illustrates samyaktva, or perfection, in all but th<.' first four <;lokas of the sixth Sarga m 561 ~lokas of Vmayacandra\, Mallmatha Car1tra My digest of Parc;va­natha shows up this habit very clearly

But these texts differ also m ~tyle and hterary pretentiousness It IS a far cry from the Mahakavya style of Pradyumnaei'irya's (Jahbhadra Carita written in the most advanced alamkara dict10n 6 to the much later Paficadandachattraprahandha, 7 which 1s written m popular Sanskrit, much dashed with Prakr1hsms The texts differ m this regard, as well as m the extent to which they employ popular words (dec;i) or words and grammat1c forms which they have drawn from Sanskrit ko<;as and vyakaranas b And they d1ff er also m the Iingmstic habits of the md1v1-dual writers, of which AJitaprahha's predilection for d1mmutive pro­nouns is a good 1l1ustration.

1 Cf the utatwn of t"'o verses from the Ciintmiitha Carita of Devasiin (p 75), apparently the same text It "'ould appear from p 339 that Deva~rm translated his work from the Priiknt

2 Tr1~a~ti, Parvan v1, ~01 b ff 3 The Malhniitha Car1tra of Shrt>C Ymaya Candra Suri, edited hy Shravak Pandit

Hargovmdda~ and Shrnvak Pamht Bechardas Benare~, Veer-Er.i. 2438 (A n 1912) 4 Sec the colofom, at the end of several of the sargas of the Mallmrttha of the pre

cedmg note 6 So the Jamas call Damayanti, the ;.tory follows clm,ely that of the Kathiiko~a,

pp 195 ff. of Tawne;i, 'i, Transl<1t10n See Wmtermtz, Gesdmhtc der lnchschen L1tcra­tur n2, 325

6 See the author JAOS xlm, 262 ff 7 Edited and translated by A Weber, Abhandlungcn der Komghchen Akadem1e

der W1ssenschaften, Berlm 1878 8 Sec my article, "Some aspects of Jama Sanskrit," m "Festsclmft Jacob Wacker­

nagel," pp 2~6 fl'

On Diminutive Pronouns in J aina Sanskrit 9

The following texts and their sigilla are at the base of this exposi­tion of the diminutive pronouns in Jaina Sanskrit which will doubtless be modified but httle by further mstances derived from the same sf ere

HEM. MALL. Hemacandra's Mallmatha Caritra. Page 201b ff of Parvan v1 of Tn~astu;alakapurusa Caritra.

MAH Hcmacandra's Mahavira Caritra Parvan x of Trisa~ti<;alii­kapurusa Car1tra

PAR Hemacandra's Sthaviravali Carita, or Pari~i~taparvan. Edited by Hermann Jacobi Calcutta, 1891

VIN MALL The Mal1matha Car1tra by Vmayacandrasilr1 Benares, Virasamvat 2438 (AD 1912) See above

<;ANT The <;antmatha Caritra by AJitaprabhasuri Bhavnagar, 1917. See above.

PAR<_;: The Parc;vanatha Car1tra by Bhavadevasiiri Benares, Virasamvat 2438 (AD 1912)

SAMAR The Samarad1tya Samksepa by Pradyumnacarya. Edited by Hermann Jacobi, Ahmcdahad 1906

<;AL The (ahhhadra Car1tra hy Dharmakumara Benares, Vira­samvat 2436 (AD 1910) Elaborated by the author, JAOS xlm, 257-316

The followmg pronommal d1mmut1ves are gathered from these texts, they are to be com,1dered m connection with forms found in Hmdu Grammar'>, and al:-,o m Bhadrabahu Caritra, below pp 22-24, and see also the old Ve<l1c pronommal d1mmutives which have been treated by .Edgerton, JAOS xxx1, 93 ff Their relation to the Jain forms 1s du,cussed below

First per'>on. mayaka and iivakabhyJm Second person tvakam and tvayakii Demonstrative sa and ta sakah, sakii, takam, takayii, and takau. Demonstrative stem 1ma 1makam, 1maka1h, and 1m1kam (fem). Demonstrative stem adas a~akau Relative stem yaka Pronommal a<lJecbve anyakat.

GENERAL MATTERS

The primary use of a dunmubve 1s, of course, smallness, hut I have noted only one example of such use, to wit, <;ant 6 264 c;ighram eva kumare~a prachannam Jagrhe saka (sc suvarnak1iikmimalii), "As qmckly as possible the prince surrepht10usly grabbed that small (chain of golden bells) " Later on, sts 267,271, the cham of bells 1s expressly spoken of as kinkinimahka, a diminutive of which saka kink1~imala is a parafrase.

10 Maurice Bloomfield

Comporting with the persistent moralizing drift of Jaina texts, the use of the diminutive pronoun is almost entirely mental or spiritual, being also largely connected with matters of Jaina religion. And the entire mass divides itself not unevenly into the two grand categories of pejorative and meliorative, that is, the d1mmutive pronoun expresses something bad or good Two consecutive passages impose upon the relative pronoun yaka part of the burden of contrasting people who are able to resist the lure of the senses with people who succumb to their senses the pronoun 1s used effectively at pretty nearly the opposite poles of spiritual quahty

~ant. 3 406 pravrajyam prabpadyapi syur yake vi~ayai~inal]. ... ghore te patanti bhavarnave,

"Even they who have wandered forth as ascetics, yet, weak creatures, seek the pleasures of the senses, fall into the gruesome ocean of (re­peated) existences "

~ant 3 407 syur yake mrapek~as tu vi~ayesv arthita api ... te 'tra bhavanti sukhabhiiJinah,

"They who, even whC'n tempted (sohcited), remam loftily mdifferent to the pleasures of the senses, partake here of bliss "

One passage has two dimmutive pronoun1-i, one after the other, in different nuance1-i, one of which clearly expre~ses blame (of one's self), the other praise of some sort

~ant 3 25n mitral} provaca he subhru natmartham mayakii tvakam, anita kim tu m1trasyamaradattasya hetave,

"Mitra (Mitriinanda) said (to the prmcess) 'Not for myself, 0 beautiful-browed lady, have I perfidw11sly earned you off, noble lady, but m behalf of my friend Amaradatta,"' For the story see Kathako<;a, Tawney's Translation, p 154

PEJORATIVE UsEs OF THE DrMINUTIVI~ PnoNOUN

These divide themselves, without hard and fast Imes, into five kmds

1 :Execration of sm and improper conduct. 2 Execrat10n of inherent or congemtal depravity or evil 3. Ineptitude, or unworthmes~ 4 Self-deprecmhon, or modesty 5 Misery, unhappmess, or ill-luck

I Execration of Sin and Improper Conduct

In this class are placed ~uch cases as represent direct infraction of Jaina teaching The sin that is reported furnishes a commentary on

On Diminutive Pronouns in J aina Sanskrit 11

the diminutive, or, vice versa, the diminutive adds a touch of scorn or reprobation to the narration of the sinful deed

<_;ant 6 321: vi<;vaste vyakule dine . . praharanti yake papa dhruvarh te yanti durgatim,

·• The wicked smners who strike a confuhng poor man in trouble cer­tainly come to grief." Papah, after yake, shows what the pronoun really means

Samar 4 330 tad api pratipannam ea krtam ea mayalca taka tada baddham maya kanna parmame sudarunam,

"I wickedly assented to that, and did thus, and estabhshed for my sinful self in consequence a grievously hard karma " The speaker has been persuaded by his mother to eat of the flesh of a "dough-cock" which she ha'> killed .Mayaka and &udaruna illustrate each other. For this extreme example of ahmsii see my "Life and Stories of the Jama Sav10r Par<;vanatha," pp 195 ff , Hertel, "Gebch1chte von Pala und Gopala," p 84

Samar 2 278 utth1to 'ham miirch1tam tal.-arn viksya k1mkrtya­miidho 'stharn, VIfii<l uparatal} sa tu,

"Ar1smg, I bcheld that 11ile person m a famt, and was at a loss as to what to do Thm he died from the pmson " The per&on who dies has tried to poison the spcaker hy &prinklmg one of two '-'Wt'etmcats with poison Ily nm,take he ha'> h1mbelf eaten the poisoned one

(.lnt I 14(1 ii.ha rnaiigalo mantrmam prat1 yady ava<;yam idam kftryam nwyak<i karma mrghrnam

·· Maiigala &,11d to the 1mmbter 'If th1& hemous deed 1s unavoidably to be done Ly v1lla111ous 11u· '" (then I make th(' followmg stipula-tion) Note mrghrnarn after rnayak:i

Samar 4 QOI tac chrutva 11w1Jaka pr&tah prahhuh, "Havmg heard this the Lord (a l\f urn) Wd'> a&ked by u11ckrd me " The per'>on abkmg 1s a cruC'l man. devoted to hms:i, who h,1.., crw<l out,·• Slay thP&ew1cke<l men," when ccrtam ernnmalb arc bemg led to execut10n The Sage tells him that, m a former lnrlh, he had calummah'd an mnocent ascetic

(rrnt G 251 athava k1m v1kalpenadr<;) arupo 'nayfi baha, gatva­ham ap1 vikse tad yat karoty a.,akiiu khalu,

'' Why need I be m doubt? I sh,tll also m mvis1ble form go along with her, and find out what that nle woman 1s domg" Prmce Guna­dharma &uspects the prmct•&s Kanakavati of Riil . ..,asi-practlc<.'S

Samar 4 141 hrtii. trailokyasftrakhya ratnavaly agasamuna vadhyas tena stena irasakiiu,

"The necklace named Tra1lokyabara has hem robbed by that villam .. therefore that wretch must be executed hke any thief."

12 Maurice Bloomfield

Samar. 4. 513 baddham trtiyaprthivyaril ea narakayur na casakiiu, dharmacintamamm mohaparayattii prapadyate,

"Life m hell is ordained for her in the third earth; and that vile woman, overwhelmed by delusion, does not obtain the thought-jewel of rehg1on" The woman in question is the Queen Nayaniivali who lives in adultery with a leprous slave, and trickily chokes to death her hus­band who ism a famt, pretendmg to fall upon him in grief over his con­dition See my "Life and Stories of the Jama Savior Pari;vanatha," p 196, Hertel, "Geschichte von Pala und Gopala," pp 84 ff

<;ant 3 41 yah ka<;cid avaninatham jatayamam I vadisyati, bha­visyah sako 'vac:;yam akale 'pi yamahthil),

"Whosoever shall say that the King 1s superannuated, that rascally person unfailingly, even tho his allotted tune has not yet come, shall become Yarna's (Death's) guest'' From a proclamation forlnddmg reference to grey hair on the kmg's head

<;ant (3 1056 punah prsta vadhuti ta1h k1m bhadre tvayalcii saha, rosasya karanam k1mem na jfttam day1ta .. ya te,

"They agam asknl the wife 'Did not, 0 good woman, some cause for anger am,e between your beloved (hmband) and yonr ref rartory self?'" QueE>t1on put to a wife who ha:,, heen left by her hm,band

2 Eucration of Inherent or ('ongenital Deprarity or ENZ

In a i,maller group of eases the evil or w1ckedne:,,<, that u, excoriated by 1mphcabon of thP peJorahve pronoun hes rn the very nature (wa­bhava) of the subject there i:,, no mfract10n of .my rehgwu:,, h'net The subject acts badly because he 1s had and cannot be otherwise

<;ant 6 411 da.,tal) <;resth1suto naga1<; caturbl11r imakii1h saha, "The merchant's E>On wa:,, bitten by these four enl :,,erpents at the

same bme" <;ant 3 463 dadhave khadgam adiiya prsthe kopapariisalcau, '' That wwked (Siren) with a sword ran after (the two adventurers)

m high dudgeon " A siren runnmg after two escapmg merchants, to kill them

<;ant 5 282 athanena (sc sarpena) E>arnadista tadartham pakl?il).i salcii, caficva krtva murnocamam nitva kvapi mahahrade,

"That evil bird was bidden to that task, she took the frog into her bill, and left him m some spot of a great lake " The bird cooperates with her mate, a serpent, and carries a frog, the serpent's friend, to a great lake, whence he is to bring food (other frogs) to the serpent, which is perishmg from hunger in his own pool that has dried up during a

1 Emend Jlitayiimam to yiitayiimam Priikr1t1c pronunciation

On Diminutive Pronouns in J aina Sanskrit 13

drought. The bird-mate of the serpent (a sarika) is naturally as base as her husband

Par. 2 231. mata provaca gal).ikam apatyc vair11).i tava, yakiibhyiim udarasthabhyam mrtyudvare 's1 dhar1ta,

"Her 'mother' said to the hetaera. 'These two children of yours are your enemies, because the vile creatures, even when they were m your womb, brought you to the door of death '"

<;ant 6. 585 apanmye sakii tena c;irortir mantravadma, "That racl.:ing headache was removed by the spdl-rnonger."

3 Ineptitude or Unworthiness

In a fairly large number of examples thf." ~uffix express<:'s the quali­ties of the capt10n The boundary hne between this cla:,s and imphca­t10ns of nnp10usnes~ or mtrm~1c <lefic1eney 1s, of course', not always absolute

(.'rmt 4 386 margayamiba tiin pafica kaniln Jye~thavadhu:rh tatah, palyantariH <;amaniyarpayama:,a .rnlca p1 t.ln,

"(The merchant) a-,ked the eldest "'1fe for those :hve grams (of rice), and that wept woman took them from a sack and handed them to hnn" From the "parable of the talents" m \\h1ch the olde:,t daughter­m-law of a certam merchant fail., to mcrease the gram intru:,ted to her keepmg Cf my "Life and Stones of the Jama Sav10r Parc;vanJtha," p 120

1\Iah I 236 tairac;cam mJ.nusam dffyam m;uthunam mayaka pura yat krtam , yubrJJ.m1 tat,

"The se~-hfe that I. poor crcainrP, have led as annual, human being or god that do I repudiate " The glo:,:,ator knows no better than to say mayakii, mayJ

Pare; 1 478 kmuara mnya!,·a putramrn~e~e 'pi yat tvay1 durJana­hivacastivravisav1hvalacet<1sft viruddham id<1111 ilrebhe . ,

.. 0 prmce, the host1htv which h,t~ been praeh:,c<l by fooli.~h me upon you, evt'n tho you were like a son to UH.', because my mmd was led astray by the strong pmbon of the :,erpent :,peeeh of a rascal ( that I will atone for) •· V1hvalacetasft bhows JU~t what mayaka means

Pare; l 419 acmtayae ea yad aho paksmo 'pidr<;i mah};i., upahar­tum tada martyabhave k1m krtam anyakat,

"(The merchant) then reflected 'Behold, 1f a mere bird (parrot) has such a d1spos1hon to do good, then what other poor deed (is fit for one) who is in the estate of man?'" The merchant has obtained a fruit of immortality from a beneficent parrot, and 1s dec1dmg not to use 1t for himself, but to have 1t planted for the good of all mankind. See the sequel. The glossator merely says anyakat, anyat.

14 Maurice Bloomfield

~ant. 3 210 tatac; cajiiapito rajiia rak1?akas tvayakdc;u re, so 'tra baddhva duriicaro val)ig aniyatam 1ti,

"Then the king ordered the baihff: 'Sirrah, you wretch, promptly bind and bring the misbehavmg merchant here!'" Tvayaka and re explam each other

Hem Mall, stanza 223 kathac; ea dul].katha eva bhavadgul)aka­tha:rh vina, yakiibhis tittirir iva vagbhir vipadam ac;nute,

"All tales are poor tales if they leave out the tale of your virtues; -tales by whose wretched words one goes to destruction, like a (chatter­ing) partridge." From a i.totra of the Sav10r Malhnatha.

Samar 2 185 tad viksya mayakiidhyayi . tad 1to nipatamy aham . . pat1to amatra gatrabhaiigampiditah,

"See mg this, I, poor fool, thought therefore I shall take a fall from here I fell, and am (now) here at home, afflicted with broken hmbs " An elefant (mayaka. the unlucky victnn) is tncke<l by his enemy mto takmg the bhrgupata (smc1de by throwmg himself from a precipice), m order that he may nd hunsclf from his ammal exu,tence and become a V1dyadhara

4 Selj-deprerwtwn, or Modesty

Qmte common are the mstancC''l m which the speaker states some act which i'l really worthy, but behttl<'s or disavows hi:. m<.'r1t He tht>n resort1> to the dimmubve mayaka, m the senst' of "by my humble self" (or German, "be1 memer wemgke1t"), and there 1s, m, a rule. nothmg else m the• 'l<.'ntence to m<l1cate th1'l rather subtle touch It would ap­pear sometnnes as if the real pomt were, to call att<.'nt1on to the sp<'ak<.'r's merit by an understatement wh1eh would naturally ehc1t precisely the oppm,1te effect m the mm<l of the hearer Thus

(ant 4 794 1ttham (,'dnhJme<;varasya cante a5tarnaJ.i prokto 'yam mayalai bhavah.

"Thus the eighth C'XI:.tence (of the luturt' Arhat) ha1> been set forth in the chromcle of th<.' Jma Lord (.'anti (('antmatha) b,7J m7J wnrorfhlj self" Smularly (i Hi31 1ttham c;ant11me<;vara~ya mayakii prokto bhavo <lvada<;ah

Par<; 2 874 svam1padadvayirenukal).ena mayalr<ip1 yat mrvyii-dham,

'' What has been planned by my 1111worthy self who am a grain of dust on the feet of Your Ma1esty " In the speech of a numster who has managt'd well his kmgdom during its kmg's enforced absence The compound precedmg mayaka rivets its meanmg

On Diminutive Pronouns in J aina Sanskrit 15

(;ant. 2 Sl4 · so 'vadat · mayaka devi kim na prapta:rh mahitale, dr~e tvaddar~ane puJ_lyarah1tiinam sudurlabhe, "What object, 0 goddess, that can be gotten on earth have I, unworthy person, not ob­tained, in that I have beheld you who are very hard to reach by them that have no merit?" Mayaka and puJ_lyarahitanam iUustrate each other m this speech of a person who has appealed to a divinity.

(;ant 3 156 · uvi'ica sutrakrc cakr(' pri'isiido mayaka kila, "The arch1t('ct said 'To be sun', this palace has been erected by my

unworthy self ' " The palace is splendid the architect deprecates his merit

(;ant 5 313 eko 'vocan mayfuvedam caru ratnam uparJitam, dvitiyah smiiha ma.7Jakii tvam lobham kuruse vrtha,

"One (merchant) said 'By me alone has this precious Jewel been obtained ' The second r,aid 'B1/ my good self you show greed m vain '" Two merchants quarrehng over the pm,s('sr,1on of a Jewel while on a river, both tumble m and drown

(ant 1 112 kuto 'py aniya mayahi mukto bhavati balakal;t, sa mantrm bhavatii grahyah,

'' Th(' boy has beC'n taken by my qood srlf from somt' place or other and left, you, 0 mIDJi>ter, mui>t get hold of him " Sp<'ceh of a <l1vm1ty who finds a subi,,htute boy for the mm1r,ter's son who 1s leprous, but has been ordered to marry a prmcer,s

('ant 6 1455 akkadva <:r<'-.thmah putrah kac:c1d atr,"iyayiiu pure, arpitam 11rn1Jaka tar,yopanadyugmc1m manoharam.

'' l\lotht'r, to-day a eertam merC'hant\., r,on has come here to the city to-day a fine pair of i>hoes \\as given hun b1J lll!J good sf'lj" The speaker, a shm•makn, tr1ek1ly mtends to clwat thP nwrchant's son out of all his property m pa~· for th<' r,hoer, See for tlm, r,tory, Vm l\lall 8 735 ff , Ilertd, Indu,che Erzahler, vol vn, pp 1'!-t ff.

5 M w·r.7;, r11ltapp111c,·.,, or Jll-l11rl·

In tlrn, group the subjective element 1s altogether wantmg the person to whom the dmunubve pronoun refers 1s not rc>i>pons1ble for the t•v1l md1cated hy 1t Thi., cl.is'> of dm11nut1ves. then, unphe& pity, rather than cr1hc1-,111, nevertheless th<'re are m it, here and there, fine shades of blame for neglect. nnprovidence, lack of foresight, ck which are at the bottom of the sorry plight of the sub3ect. Thus, from the pomt of view of the writer the peJorative touch Ii> not altogether want­mg mdeed, here as elsewhere, any strict compartmental groupmg of shades of dmnnutives rests upon convemence rather than prec1s10n m the circumstances which give rise to its express10n

16 Maurice Bloomfield

We may introduce this class with an example in which the ka of the pronoun is remforced by another ka in the noun to which it belongs, to wit

Vm Mall. 8 214 sakhyad devemakam kasthavahakam drutam anaya,

"She (the queen) said 'Your lllaJesty, bring this wretched wood­carrier here quickly ' " The queen has recogmzed her husband in a former birth, who, m his turn, is passmg his present existence as a low­caste man

The following hsts this rather large class (ant 6 914 tatac; ea kath1ta tena khadgavismrtiJa kathii, tathapi

dand1to rii.Jfianarthadande krte sakah, "Then he (Samrddhadatta) told what had happened through his for­

gettmg his sword, but even so the 11njortu11ate man was pumshe<l by the kmg, the pumshment bemg undeserved " Samrddhadatta hm, forgotten his sword outside his home, thieves find it, conumt murder with it, and suspicion fash•n~ itst>lf UllJU"tly upon Samrddha<latta

(ant 6 91!) grarnapar5ady upavu,to yi""tvad ibit sako 'nyada, "As that unlucky (Samrddhadatta) was s1ttmg once m the village

hall " Tht• same Sarnrddha<latta a,-, m the preet·dmg item (Jnt J 377 tatac; ea <levariiJena bhiihlrnJJ Lharnto 'nuph. ganta­

vyam frayahinyatra muktva me Vi',ayam punah, "Krng Devarap said to hi~ brother 'Go, wretched IJOUth, to another

country, moreover out of the reach of my ~enses''" DevarJJa i., ex1lrng his youngei brother VabariiJa for no othn reason than that the people love hnn best

(ant 5. ~380 devi pro, aca he v at'la yady evam fraljakii saha, aganu~-yamy aham api dhruvarn,

"The queen ~aid (to YahariiJa) 'If so, then I will go there with you, poor clnld "' Sequel of the prt'C'<'dmg VatsataJa's mother (as well as DevaraJa's) propm,eb to 1,hare his exile

(ant 5 465 iiva!.iibh1j<im adhanyabhyiim ya<l a<lyJpi sames1 na, "If even now you will not come with Ub two wretrhed women "

Avakabhyam and adhanyiibhyam m reciprocal comment (ant 6 {i31 iti prokte 'pi sa yavat sulasam tyaJati sma na,

svayam evakkaya tavad ity abham sako 'nyada, "When (the hetaera), thus addressed, did not abandon Sulasa, then

the bawd (mother) qmte by herself spoke to that wretched (impover­ished) man (namely, Sulasa) one day"

Cant 6. 715 svayam vipannasairibhya}:i [sic; gloss, mahi~ya}:i] puccharh tasyarpayat salcah,

On Diminutwe Pronouns in J aina Sankrit 17

"That unhappy (Sulasa) himself handed the tail of a dead she­buffalo to that (tricky conjuror who was plotting against him) "

(;ant. 3 552: so 'vocan mayaka muktam ... mogham abhud idam, "He (Damitiir1) said. 'This (discus) has been hur]ed by unlucky me

in vain.'" In a duel between Damitari and Tr1pr~tha, m which the former is ultimately killed by the latter Mogham and mayaka illus­trate each other

<;ant 6 328 bhaycnii1tasya mayakii viva.ho neps1ta};i khalu, "Out of fear I, wretched girl, do not at all desire to marry him"

Bhayena and rnayakii illustratt' each other <;ant 6 1574 param r;ar;aka no gantum patt1bh1J.i suniyantrita}:t,

evam asthiit tatra kalam k1yantam a\aldill tathii, "But he could not t''>capt', because hl' was carefully kept confined by

the sold1en,, so that he abode therc mi~erabllJ for some tnue " The sub­Ject 1s a Pulindra (Pulmda), a forest-dwdler who ha'l been taken to his palace by a kmg out of gratitude for an important service The Pu­lindra longs for hu, natrve fore~t. and latt'r make'> his esrap0

(.'ant 6 1410 ekiik<iah k1tarnh kaf;c1d atho ratnakariitmaJam, 1ty uvaca ~ahasrei_rn dravyasya mayal;li, kila, svanetram tvatp1tu}:t pfm;ve muktam grahanake \t1 bhoh,

"A certam one-{'y<>d g.unbln said to the ..,on of Hatnftkara 'I my­self, nwerable man, deprn,1ted my (other) eyt:> as a pledge (for the loan) of a thousand with your father, 0 good Sir 1 ' '' ( 'f Ym Mall 7 H(i ff ; Hertel, Ind1schC' Erzahler, Yol vu, pp l 'i?!l ff

(ant 6 421 yady evc1mndhayJ, knyayasakau s::untisthate tato Jivaty anyathii mnyate dhruvam,

"If that unfortunate m.m will engage m ~uch (pt-mtenhal) practice, then he will live, othenH,e he will surcly die" Part of a convers10n­story

<;ant 3 61'2 anyt•dyur da ... yuna1kenJ,tipracandcna mJyma, agrh­yenalahitena musyate ~ma puri .wl,a,

"One day a certam very cruel robber, who hemg skilled m magic could not be caught becau~e ht• was not 1'.t'en, plundt•rcd that 11.'retched city"

<;ant 6 344 udbadhya taru<;akhiy;un atmanam martum udyatam, imikam aham ad1aham tvadv1yog<.•na sundara,

"I beheld tins wretched girl who had hanged herself on a tree branch, determined to die on account of sepa1 ation from you, 0 noble man!"

<;ant 3 39 lh cintilvi~anl).asyarn patun dr~tva sanarmavak, uva­caivam puna raJfii tadbhavav1dura saka,

18 Maurice Bloomfield

"Thus, perceiving that her lord's face was clouded with sad reflec­tion, the queen pleasantly addressed him, not understanding his state of mind, she, poor lady." King, son-less, has discovered a white hair in his head, and 1s aflhcted by vairagya, queen, misunderstanding his state, deJectedly g1ves him adv1ee that is not to the point.

~ant 6 291 evam uktvasakau vamakaranyastac;irodhara, cintaya vigatotsaha babhuvadhomukhi ksanat,

"Thus saymg, that unhappy girl, her neck reE>tmg upon her left hand, weak from sorrow, for a moment cast her eyes to the ground"

~ant 3 682 v1c1kitsiim imam krtva mrtva cayul)-ksaye salca, sam-3ata yatra taditah sthanam samkirtayamy aham,

"Havmg entertamed this doubt (about Jaina religion) that unhappy woman died at the expiration of her life's term What station she-was born m, that will I now relak " The woman suffers much m her next existence Saka may here reft--r to impiousness, rather than to evil fate.

Vm Ma11. 7 370 yaldibh,yam eva karnabhyam c;rutam svagm;ia­varnanam, re piiradJrika iti tabhvam eva hi <;uc;ruve,

"With the very same wretrhed ears with which he had heard pane­gyrics of his virtue-,, w1th those !>elf-same ears wa!> now ht'ard the cry 'O you vile pursuer of other men'!> wive!-.1'"

Vm Mall I 101 atraivana<;anam krtvft paralokah sunirmalal)-, upiirJaniyo bhavena no kiiryam rnayahiparam,

"Right her<:' I must fa.,t to death and r<'ach by nl<'chtation the wholly undefilc·d othe-r world, my poor self cannot do otherwi!>e " The speaker has been told by a Vidyadhara sage that he haE> JllSt five days to live, and he 1s lookmg for some one to reeite for him the "Five-fold 01)('1-,anc('" (namo 'rhadbhyah etc )

Vm Mall 7 1052 sii tm,thau nyagmukhii biila raJanyJm iva pad­mini, uvaca c,rt--1>thmim mfilam iivayor duh1tfisakau lalaniyi"i palaniya,

"The (prmceE>!> Vasurnati) stood with hn eye!> ca'lt down, lile a lotus by mght Tht> (mrrchant Dhanavaha) said to his wife Mula 'That poor cli1ld rnm,t be petted and brought up by U!> aE> a daughter '" The merchant haE> taken the prmcess from thc> E>l.we market to adopt her as his child, E.ee the digest of this remarkable story, Bloomfield, JAOS xlm, ~6.5, note 18

('al 5. 13 rapsarpah prasarpantam yakam Jt-gilyate ... bhog1lo­kam,

"The anaconda (double entente, royal serpent, meanmg Kmg ~remka) who devours the miserable serpent-folk (double entente, his people who are devoted to the senses)." See the author, JAOS. xhii, <J,77.

On Diminutive Pronouns in J aina Sanskrit 19

Par~. 2. 409 · tatra yuyam tu no dr~tal;i prsta~ ea munayo na tail;i, uttaram dattam ity artya prayukto mayakavadhi}:i,

"I did not see you there, and when I asked the Sages they gave no answer. Then, in distress, I, miserable, employed avadhi msight." Arty.a and mayaka illustrate each other

MELIORATIVE UsEs oF THE DIMINUTIVE PRONOUN

Mehorabve uses of diminutive pronouns are less common than the peJorative, but not less certain, or eff ecbve For the most part the meliorative function 1s found at the opposite pole of the pe3orative (pious impious), but 1t produces also shadmgs of its own The entire class may be treated under three heads

1 Approval of piety or good conduct 9l Express10n of excellence of social station or character 3 Conc1hation or caJolery (captatw benerolentiae)

I Appro1.'al of Pief.1/ or Oood Conduct

The two antithetic examples which reflect m two successive <;lokas of (;ant. 3 406, 407 yake m the sense of '' what impious people," and yake in the sense of•· what pious people," are stated m full, above p 10 Other ca'i<"S m which ka-dmunut1ves nnply approval of piety appear m the followmg

(ant. 6 754 siidhayanb yake sarvln yogan mrvana<,adhakan, ... tan bhadra sadhun nama~kuru,

'' Tho~e pious men who acquire all the ascebc practicl."s that secure emancipat10n (mrvana) them, my good slf. do you revere'" Here yake and sadhiin illustrate each other

<:;ant 3 324 vH;uddharn palay1tva tan mrtvagad amaralayam, tata<; cyuteyarn sarn3utii <;re~thim, te duh1ta :mka,

"Having preserved her virtue, then died, she went to the abode of the immortals Thence the pwus woman fell and became your daughter 0 merchant 1"

<;ant 4 791 dharmarn tadanbke <;rutva pratibuddha}:t sako 'pi hi, rajye <;atabalarn putrnm mve<;ya vratam adade,

"Havmg heard the Law expounded by hnn, that pww,1 kmg was converted and placed his son \atabala upon the throne, and also took the vow"

(;ant. 6 1495 c1ram bhoga<;riyam bhuktva Jiitaputra];i sako 'nyada, <;u<;rava sadguro}:t par<;ve dharmam pravraptas tatah,

"Having en3oyed for a long time the bhss of the senses, having begotten a son, that pious man listened one day to the teaching of

20 Maurice Bloom.field

the Law by a noble Teacher, and then went forth into the life of an ascetic ''

~ant 6. 394 · tarhi tan aham, kridayiffami naga:ris tu madiyan kric)aya tvakam,

"Then I shall make these serpents perform, and do you, holy Sir, make mine perform." Nagadatta is addressing a Muni (tvakam) who has come to convert him The serpents are the four ka~ayas (sins): anger, pride, guile, and greed

~ant 6 1621 tasyam c;1laya:rh kalena bahvyal;t samyatakotayal_i, siddhac; cakrayudha:rihr1bhya:rh yakii purvam pav1tr1ta,

"Upon that holy (yaka) mountam which had been prev10usly sanc­tified by the fe!:'t of (the Sage) Cakrayudha, as time passed, many crores of ascetic:,, obtamed bh:,,s "

2 Expression of Excellence of Social Station or C'harader

We may mtroduce this clas~ fitly with a description of the s1ddhi, or persomfied "bh:,,s," the :,,tate of the soul after 1t has been freed from the round of existences

~ant 6. 1553 pmde ea yoJanany asta,u madhyabhage sakd punal_i, Jiiyate maks1kapatratanvi cante,

"And the noble (siddhi, emancipated soul) is, as regards her frame, eight yoJanas m the e>..tent of her middle, and at her end thm as the wing of a fly" Prev10u:,,ly, 6 1550, the quest10n 1s asked, kidrgriipa siddhir bhavaty asau, "Of what form i:,, that siddhi?"

Thr remammg cases are of consl(lerably varied character they deal with noble station, noble character, and heroism

~ant. 4. 7. kathitas te mahibhartuh :,,uputra1anrnakathanat tenapy ahladita salai,

"(The queen) told her (fourteen dreams) to the kmg, and that noble lady was gladdened by him hy telling her that she should bear a glor10us son "

<;ant 6. 8 tasmnic; ea samaye devi sukhasupta catur<lac;a maha-svapnan dadarc;a . salc<i,

"And on that occasion the noble queen, sleepmg blissfully, saw the fourteen great dreams ( which herald the birth of a Tirtharnkara) "

~ant. 3 623 sva:,,amipe samayantam nanama sa mahipatil_i, kutal_i sthanad agato 'sity alalapa salw 'pi tarn,

"That kmg revered the ascetic who had come to his presence, and courteously asked him whence he had come "

Vin Mall 2 96. atha ri1Jye 'sak<iu nyasto bhiipo 'bhiit padmac;e­kharal_i, punyam sahaciiri]).i videc;e 'pi mahatmanam,

On Diminutive Pronouns in J aina Sanskrit 21

"Then that noble Padma~ekhara was placed upon the throne, and became king The virtues of noble men accompany them even to a strange land " Padma<;ekhara, sleepmg under a tree, is selected by the minister Subuddhi as succt•ssor to a kmg who has died without heir.

<;ant 6 I IS8 tato ra1asamad1stacebbh1J;i snapita saka, "Then that noble lady was bathed by tire-women appointed by the

king" Vm Mall 5 79 antahpuravadhiilokii1r mud1tfuh snap1tasakau, "That lovely (princ<.>ss) was bathed by the delighted women-folk of

the zenana " Ht>r father, the kmg, afterwardb asks the guardian of the zenana, whether he has ever seen so lovely a girl

<;ant 6 1284 siipatyapy amunii saldi . nita puriiraksasya man­dire, and, 6 1285 sii nmye raJamandire,

" That noble woman with her child was takm to the palace of the city guardian," and, later, "She was taken to the king's palace "

('ant 3 530 iimetI takaya prokte tiiv abhiit:im svariipmau, drstva 1agada sa ciiham yu<;ma<liiJfiiikari khalu,

",vhen the lo2•ely (prmcess Kanakac;ri) consented, the two (princes Anantavirya and ApariiJ1ta) i,howed them<;elves m their true form The (princess) seemg them told them that she was entirely at their dis­posal " The prmces have been d1sgmsed ai, nautch-guls; Kanakac;ri is in love with Anantavirya

('ant 3 5-H (m alternate readmg in foot-note) tiim c;astrani ta-sya<;u pratH:;astrii1r mahabhujau v1phalicakratus takau,

"These weapons of him (Dan11ldr1) the'le two great-armed heroic (princes) quickly made futile by means of counter-weapons "

<;ant 3 480 . . vahta. vyantari ~a.tha yak'>o 'p1 vahtah salcah, k~am1tah krtakrtyena <;resth1putrena bhaktitah,

"The VyantarI (Siren) returned, and so did the lmzdly Yak'>a who had been conciliated by proper perforrnanct•s by the merchant's son (Jinapahta) m devotion " The Yaksa here, as elsewhere in fiction, figures as a benefactor

<;ant 4 697 utt1sthotti~tha he tata yaty adyapy asakau nara);i., athasya prchato 'darc;i tena chaya <;arira3a,

"Rise, rise, 0 Father' That good man 1s gomg to-day also Then, when (the father) asked, he showed him his own shadow" The father has previously been led by that same son to suspect that a strange man is visiting his wife, see Hertel, lnd1schc Erzahler. vn, 130 ff.

Vm. Mall 5 14 tasyal,i striratnamukhyaya ayurgranthau mahi­pate, vidhiyate sako 'py uccair ni);i.sima);i puspamudgara);i. =

Maurice Bloomfield

Hem. Mall , stanza 62 · tasya}:t striratnamukhyaya ayurgranthau vidhiyate, sa pu~pamudgara}:t ko 'pi yadrk svarge 'py asarhbhavi.

The glossografer to Hem. glosses pu~pamudgara}:i by pu~pagucham, and ayurgranthau by ayul;isiicikayam granth1riiparekhayam.

The passage from Vin. Mall. may be rendered, "In the life's knot of this foremost Jewel among women, 0 king, that wonderful, grand bou­quet of flowers is clearly formed." Hem. adds that such a bouquet does not exist even in paradise The mimster Subuddhi says this to King Pratibuddhi partly m praise of the maiden Malli, the future 1.'irtharh­kara Mallinatha saka and nil_isima illustrate each other

3. Conciliation or CaJolery (captatw benevolentwe)

The cases are few, but their special shading qmte obvious Cant 3 182 punal_i sa bha:i:utakkaya sarvatha sevaniyo 'yam he

putr1 tvayaka naral]., "Again the bawd (akka = kuttani or kuttmi) told her (the hetaera

Vasantatilaka) 'By every means that man (Mitrananda) must be cul­tivated by you, my good girl '"

Cant 3. 189 mitranando 'vadat tarhi tadagre kathaya tvakam, bhadr<" gunotkara}:t . yasya <;ruta5 tvaya,

"Do you, my good _qirl, relate to her (the princess Ratnamafijari) the many virtues which you have heard about hun (Amaradatta) " Mitrananda is trymg to brmg together his friend Amaradatta and the prmcess RatnamafiJari, the obJect of Amaradatta's desire

<:'ant 6. 1405 idam asya hi mulyam yad vm,tubhis fvayakoditail_i, yanam te piirayi~yamo gachato nagaram m3am,

"'fhis mdeed is the price (we shall pay for your ship's load), namely, we shall fill your ship with goods designated by your good self " See Hertel, lndische Erzahler, vii, 121.

RARE INSTANCES IN THE CLASSICAL LANGUAGE

The Western Lexicons quote or cite diminutive pronouns practi­cally only from the Veda, on the one hand, or the Hmdu Grammarians on the other hand I presume it is Leumann that has contributed a number of such words to the Lexx from the late Jinistic Bhadrabahu Canta, which is not at hand ahakam, 1 109, which does not appear in my collection, but is known to the Grammarians, below; mayaka, 2 49, tvakam, 1 64; tvayaka, 4 9, and case-forms from stem taka, 4 22, 151 The form Prakrit tayam, = Sanskrit takat, occurs in the Maha­ra~tri tale of Aga<;ladatta, Jacobi, Ausgewahlte Erzahlungen, p 85, lme 37 (stanza 320), in the sense of "this damnable (act)." J. J. Meyer,

On Diminutive Pronouns in J aina Sanskrit 23

Hindu Tales, p. ~86, note I, quotes another case of tayam from the Kaliicarya Kathanakam. Both texts are Jinistic. Judging from my own experience as well as from the Lexicons, the Classical language shows them so rarely, as to render them well-nigh negligible for the entire period that bes between the Veda and the writings of Hema­candra. Bo., in the smaller Pet. Lex, Nachtrage (vii, p. 314), quote asakau from <;i~upalavadha 7. 53, Sahitya Darpai:ia 49 ~

"-J DIMINUTIVE PRONOUNS IN THE HINDU GRAMMARS ~

,:;:i L' The Hmdu grammarians treat familiarly a qmte remarkable s­

sortment of dunmutive pronouns, some of which coincide in form a use with those hsted above from the Jaina texts They range over aJl the personal and demonstrative stems and some of the pronominal adJectives Pamm 's general treatment of the dimmutive suffix ka is at V, 3, 71 ff first pnson smgular alzakam and the stem-forms makat­and mat/ea- m rnakat-pitrka or matka-pitrlca, see PataiiJah, Maha­bluisya I, 1, G (p 97 of K1elhorn's edition) on Pan I, 1, 39. The m­strumental sg. mayaka, very frequent m the Jama texts, and the solitary mstrumcntal dual ifral.<ibhy<im (p 16) are not ref erred to, but Bo Nachtrage 5, quotes ahakam from Bhadrahiihu Carita I, 109.

Second person singular fral.·am and the stern form:. tvakat- and t'l'afka-, m ti•akat-pitrka, m,atka-pitrka, JI ahabhas,ya, loc cit The stem saka figurei,, along with esal.-a, at Pan VI, 1, 132 schol, where the statement is made that their nommahvci, a.re sako and esako m ex­ternal eufomc combmat10111,, m which the non-dnmnuhve forms ap­pear as sa and esa Thu, 1s borne out unfa1lmgly, as far as salco is con­cernPd, by the Jama texts Note that tht' femnunp-, sakii and yaka are the only exceptwrn, to the rule that the fcmmme secondary (tad­dhita) suffix /,ii does not change the a- i,tem of a noun combmed with it to 1 (ikii), Pan VII, 3, 45, Vop 4-, f> This 1s borne out negatively by the pair imalw, but utnl,ii m the Jama texts above

:From the stem:, e~a, eta, the Peteri,burg Le),_Jcon, a.t the end of its arhcll' etad, and Bohtlmgk's Lexicon under etal.a and esaka, cite the unquotable femmmes esakii and esika, etakii and etil.a with reference to Pamm VII, 3, 47, and }us connnent.1,tors No dmnnutive forms from esa, eta have thm, far come to light m Jama Sanskrit

Panm1 VII, I, 11, 2, 112 and his commentators and successors permit the stem wzal.:a (also anwka) to form all cases except the nomi­native ayalwm The forms mentioned are, unal.·ena, unakayoh, and imakaih He does not refer to the fem imilca, which is, however, im­plied in siitra VII, 3, 45. Forms from masc imaka and fem imilca are

24 Maurice Bloomfield

fairly frequent in the texts above. The instrumental plural imakai~ is, however, to be appraised as the direct dimmuhve of the Prakritic form imaih, which is peculiarly frequent in Jaina Sanskrit Par. 8, 520; Parr;. 1, 885; 6, 767; 7, 398; Samar. 4, 508, 619; 6, 385; 8, 520; Qant. 6, 411

The nominative asakau, very frequent in the Jaina texts, appears in the varttzlca to Pan VII, 2, 107, but amukazh, which appears in the company of imalcaih m Pan VII, 1, 11, schol , has so far not turned up m my texts.

The var10us derivatives m the Jama texts from stem yaka arc sup­ported not only by Vedic mstances, but also by Pan VII, 3, 45 (with salca), Vop 4, 6

Of stem anyaka (Vedic and Jama, above) the grammarians take no note, but the masculme plural ,'farvalce (with mr;11alce), Pan V, 3, 71 (schol ) , and the fem sarvika in thf.' comments to Pan VI, 3, 35, VII, 3, 44, Vop 4, 6, are supported by the d1mmuhve sarvakam, Av. I, 3, 6.

Max Muller, Sanslmt Grammar for Beginners (London, 1870), p 131, section 274, referrmg to Pan V, 3, 71, and Szddhanta Kaumudi, vol. I, p 706, groups together. as denotmg contempt or dubious rela­tion the forms tvayak<i, yuvaka_yoh, asmakabhth, ayakam, and asal.au

It is possibl{' that some of these grammatical forms, as well as some of the Jama forms, are retrograde Sanskrit forms, based upon Prakrit models. Prakrit forms of aham, such as ahaam, lzage, hagge (hake, ahake), hsted by P1.-,chel, Grammatzk der Pralcrzt-Sprachen, sect10n 415, presuppose ahalcam, but this form is too solitary to be the basis of the entire movement Thne goes with 1t no indication of dnnmutive funct10n. Qmte beguiling 1s the form imtd = *imikd m Hemacandra (Siddhahemacandram) 3, 73, but this lea-form from stem ima 1s also too isolated to permit correlation with the free imaka formations of the Jaina texts and Panim's full set of case-forms.

CONCLUSION

Professor Franklm Edgerton, in his excellent Johns Hopkins doctor's dissertation, "The K-Suffixes of lndo-lraman," JAOS. XXXI, 93 ff , has treated exhaustively the diminutive uses of this suffix in the Veda (pp 125 ff ) including, very observantly, the diminu­tive pronouns He has, however, for various reasons, not extended his researches to Classical Sanskrit From the negative evidence of the Lexicons and western grammars the occurrences of dimmutive pro­nouns in the Classical speech must be, at best, exceedingly rare. Be­tween the far-off Veda and the Jaina texts lies the native grammar,

On Diminutive Pronouns in J aina Sanskrit 25

which treats, as we have seen, the diminutive pronoun as a familiar fact without any indication that it 1s practically restricted to the Veda (chandasi). What, then, is the basis of the Jainistic usage?

There seems but one answer hkely The Jaina writers have re­stored the d1mmutive pronoun to literature In my paper, "Some, Aspects of Jaina Sanskrit," Festschr1Jt Jacob Wackernagel, pp 226 ff., I have shown that Jama writers make copious use of the Vyakarana and Ko<;a literature 1 I repeat a remark made there, namely, that words which occur in Jama texts and, besides, only m Lexicons and Grammars, are no more " quotable " than the grammatical forms of the Bhattikavya As an additional illustration, I may mention the common Jama use of the pronommal adJectives m kina, mdmakina, tiivakina, iismiikina andyiiusmiilci11a. They are treatedm Pan IV,3, 1-3, but there are only a very few forms of that kmd quotable, 2 begmning with the somewhat different miiltina, Rv VIII, 27, 8 On the other hand, a small group of Jama texts yield the following

mihnakina Vm Mall.7,560, (ant 1,287, 2,29~), 4,298, 6,1465. tavalcina Vm Mall 7, 124, (ant 2, 312, 6, 1154, 1404 iismakina Vm Mall 7, 677 yiiusmiikina Par 1, 153, Vm Mall 7, 34, 917, ('al 6, 25 The Jamas use these words farmharly, not because they know them

from literature, their recrudescence 1s due to the Jamas' large control of grammatical and lt'x1cal science In the futur(' 1,tudy of the native vyakarana and ko~·a literature, the Jaina texts will take a large, almost controlling part, because they have revived much that 1s lost m Classical literature

1 Indeed, Jama scholar1>h1p to tlus dav kC'epi, 1b grip on thei,e subJects as stoutly as chd the gredt Hemacandra Of rCC'ent 11111,1,1\ cs from India, through the kmdness of their Pontiff, Jamacharya Slm Y1iava Indra Suri, 1 mav ment1011 a beautiful edtt10n (with word-mdex) of Hemacandra's Abhidhiinacmtiimam, elaborated by those two wonderfully productive, 1,cholarly Pandit~. Shravak Pd11d1t Hargovmda~ and Shravak Pandit Becharda1,, published as no 41 of the Yashov1Jaya Jama Granthamala, Bhavnagar, Veer Era 2441 (An 1915), and the Dhann.1dip1kJ by Nyaya V1sharad Nyaya T1rtha Upadhyaya Mangal V1JaydJI, Bhavnagar, Virasam,at 2451 (AD 1925)

2 See the Western LexKons

JOHNS HOPKINS UNIVERSITY

NOTES ON JAINA MAHARA~'J'Rl BY FRANKLIN EDGERTON

IN the course of a number of years of class-work on Jacobi's Ausge­wahlte Erzahlungen in M aharastri, I have collected the followmg

notes on matters mainly of etymology or mterpretabon, which now seem sufficiently numerous to Justify pubbcahon for the benefit of other users of that admirable and useful manual, and of students of Prakrit in general. Except m a few cases where the contrary is specifically mdi­cated, they all concern matters which seem to me msuffic1ently or in­correctly explamed both m Jacobi, and m J J Meyer's Enghsh transla­tion entitled lhndu Tales I refer to Jacobi by the m1bal J.

1. " V edisms" in .M <iharastri

khambha "pillar" = Ved1c skamblza (J stamblta) ✓ J11nna ''old" = Vcdic (RV) plrna (Skt Jirna) The u-forms of this

root arc d1stmctly Yed1c tlzama "station" = Vedic s~n (m Skt the word means

" ") ,. ,J -- ' powN ,.y..-'. ..=--"' i•a.nma "dwellmg" = Ved1c i1asman. Cf also fa1Jarn below, under 2

2 8nnani1c and Etymological Notes

ahi,ydsci = adhz1·dsa'lfah, "cause (divme power) image), invest" (sec Edgerton, J AOS 33 158 ff) " t " er ragen

to dwell in (an J adhyiisayati,

ablweiina "having experienced," from a+blwJ "enJoy" and so "experience "

ucchahya "&1ch erhebcn" = Skt uc-chal (root §al) J utthal0•

uvvaddha-pindw = Skt udbaddha-pind1ka "with stout, swollen calves", the identical Sanskrit compound occur&, Tantrakhyay1ka p 64, hne 7. J udvrddha- "krafbg", Meyer has the correct etymol­ogy but does not note the actual occurrence of the word m Sanskrit, which determines the meamng precisely and mdubitably.

kammar,ia 56 31 = karman, "medical treatment", so in Caraka. See below under muli

karat)a 37 5 "Ger1cht"; in this sense I thmk the word 1s a back­formation from l.aranilca (Pkt lciirarJ,iya) "Judge," which is m actual fact a derivative from Skt. karana, "cause," and not from kararJ,a.

28 Franklin Edgerton

khiimei 25 15, causative of kiam, means "say good-by," literally "excuse oneself"; cf Russian prostit' s' a, literally "excuse oneself," the regular Russian expression for "say good-by."

gosa "mornmg" is derived from go and a derivative of the root si (RV. saya "unloosmg" etc.), and equals Skt. go-visarga, literally "time of lettmg the cows loose "

COJJa = Skt tiscarya, I would explam the phonetic development thus <iscarya >acce13a > (ac)co11a by dissimilation of e to o before Jj. (J gives the correct etymology, of course, but does not call attention to the mterestmg case of dissimilation)

tayam = takat, 85 37 Meyer notes the etymology but fails to observe the clearly deprecatory or imprecatory use of the ~uffix -lea, which may almost be called a Vcdism The use of the "dnmnubve" -ka, particularly m peJorative senses, with pronommal stems, 1s espe­cially characteristic of some phases of Vedic diction See Edgerton, The K-Suffixes of lndo-Iranum, p 45 f

tavovahana 61 17 "excellent ascebc1/>m ''; J translates "Askese" and does not analyze the word The second member is clearly upa­dhiina = insesa, "excellence "

niyanta "sehend, beobachtend '' (not explamed rn J ) 1s snnply a pres ppl to the root ni, with meanmg mfluenced by nefra, nayana, "eye", because of these fam1har derivatives, the root ni 1s felt as mean­ing "to see "

niru11ei with saranam, 63 20, = "meditate on your (religious) refuge," "call on your sectarian deity," that is "prepare to die." See now Edgerton, ''The Hour of Death, its importance for maru's future fate, m Hil1du and western rehg1ons," Annals of the Bhandarkar In­stitute, 8 219-249, esp. p 2~4

payai 26 5 = Skt prakrti, "mm1ster" or "power" of a king (a techmcal Artha&iistra term), J derives from padiiti.

paraddha 7 35 = pralabdha, "seized" ( with regret, pacchiiyiiva-); J "gequalt" (DK pul:ita); Meyer "seized," no doubt understandmg the word as I do, but without statement of etymology

pula1ya, H. pulaei = drs, J/> an interesting case of vowel-metathesis; 1t is evidently Skt pralolrnyati (Pischel §§ 104 and 130) with metathesis of the vowels of the first two syllables.

bhan<J,ana "Streit" 13 28 = Piih bhandana and Buddh1stic Skt. bhandana.

manahara-kala 78 25 (metr gr for mano 0) = "night" ("pleasant

time") mari 35 ~4 = "pestilence" (Skt miira and lex. mari). J. "Morder."

Notes on J aina M <ihiirastri 29

muli 56.Sl = Skt millin, "root-doctor." The sentence alavanam pi hu chey<ina kamma7J,am kim ea millihim means "the conversation of the clever, I tell you, is true physicking (see kammana above), and what is the use of root-doctors?" J renders muli by "reich "

laJJhai· on this passive to labh see now Tedesco, JAOS 43.368. lambei = Skt lambayati "hang up, post," 18 21, 24 J "verbrei­

ten ", Meyer "supply," which 1s bad, but Meyer in a footnote suggests as an alternative the interpretation which I give

vaccai "gehen, wandern" seems to me clearly Skt. vrtyate, passive to root vrt

vaccha-tthala = Skt 0 sthala, "place" J 0 tala saccaviya 13 32, 15 19, 27 32, "~een," is not explained etymologi­

cally by Jacobi, who separates 1t from the other saccaviya, the p p p. of satyapayaii, "bewahrhe1ten." But 1t is obviously the same word as this, for the development of meanmg cf German wahrnehrnen

sulasd, n pr , i~ not explamed by J ; of course it is Skt Sudrs<i. sura = Skt £ura seems to me usC'd at 56 2 as an abstract, "hero­

ism", Skt siira 1s so used m Ind Spr 5756 haHai 16 22, 75 23 is given by J as meamng "hindern, abwehren,"

quotmg Hem 4 134 msedhati But Hem's definition 1s mexact, or rather, free, the word 1s based on the onomatopoetic hakka, a call used in stoppmg an elPphant 16 22 hakkw 1.-urndrena lwri Ongmally "to call 'whoa'' " it comes to mean" to call, yell at" m general, so in 75 23 a thief 1s the object of the shoutmg (Meyer here rendC'rs "call," though at 16 22 he follows Jacobi, "warded him off") See the next

haklcarai has the same meanmg as the precedmg, and a hke deriva­tion; it contains of course the element -kdra (as in humkdra etc) Per­hap& contraction or haplology has taken place m 1t (from hakkakara, cf. andh<irzyam, 22.3, = andhay<irz/jam, Skt andhal,dntam). but more hkely the ongmal exclamation wa~ sunply ha/.:1 In Sanskrit halc-ldira is quoted lexically In 71 33 this word 1s usC'd of an elephant (hakkario gaindo), but 1t also, hke hakkai, 1s used by extension of human beings (63.19 and 33, 72 20) The word has, of course, nothmg to with iikiirayati (J )

3 Omissions in Jacobi's Glossary

Most of these are perfectly simple and obvious words, their omission being clearly due to mere accident In some cases Meyer has noted the omission Some of them, however, appear to be due to errors of inter­pretation

ii7J,ii = Skt ii]fiii, "command," 63 6

30 Franklin Edgerton

kalaha = Skt kalabha, "elephant" (Meyer), 81 34. /calla, kallam, "morgrig, morgen," 50.20, 60 29 (Meyer). J. only

"gestrig, gestern." carana, "good conduct, morahty," 3 15 (tatsama). calana, "foot," 23 25 (tadbhava) cetthai, 59 22, 27, probably = ci((hai, Skt tis(hati (rather than root

cest-); so clearly at 59 27 where it is resumed by the participle (hio in the next line

tanzya 86 18, ace to Meyer = taniyas, compv to tanu. pan ta = Skt. priinta, "last, extreme," 33 29 pavara = prarara, by a 1,lip labellc>d "ts" (tatsama) in J. pii_gaya = Skt priikrta, '' commonplace," 2 28 manunna = Skt manop1a, "charmmg," 4 16 (l\'Ieyer) iia.JJlli = Skt 1Hidyate, "is made to sound," 40 21 vibhasci 65.8 ace to Meyer ts , but perhaps rather mbhiisci'P Uncer­

tam ,'liimanna 13 25 "gcwohnhch" is placed by a shp under so.manna =

Skt .frcimanya suhi = Skt. sulml, "fnend," 84 7 (where J wrongly emends to sahi)

and 8,236 (where he mterprets it as = sukhrn) Meyer has the correct interpretation m both places

4 A few Emendatzons

At 59 14 read pNhap~ sii,laln (with v 1) "on -.pits," for mulah1 At 39 16 read uvm•tiiyam (with v I , = Skt upapiiditam) for

uvaiyam At 37 27 read probably Janaya-kl,hao = Skt 1analw-haya, "par­

ricide," mstead of Jana0, supposed to mean "people-E>layer, murderer"

(but the use of the wordJa?J,a ,n such a connexion seems cur10w,) Kara­kan<.Ju at this point was about to attack h1& own father, though he did not know it It is a version of the Hildebrand motif The speaker wishes to avert the horror of (unw1ttmg) parricide

YALE UNIVERSITY

MISCELLANEA PALICA

Bv DINES ANDERSEN

1. KANTA

The followmg verse is found Ja VI, 370 contammg a riddle.

llanti hattlwhi piidclu mukhaft ea panrnmbhati sa ve, riiJa, piyo hoti Iran tcna-m-ablupa.ssa.~1 I!

Fausboll's edition has two rrm,prmts m this passage r<iJii for r<iJa, and abhipassati for abhipaswsi (cf pp 37G, 15, 377, 4, 378, 9-28). The English tram,lation of Cowell and Rom,c renders the la5t two sentences everywhere with the words "he 1s dear, and grows dearer than a hus­band", and at the same time adopts the readmg kantena, instr of kanta, m "a husband " It H, not ea1,y to see how the' translator under­stood abhipas.sati, but the trarn.lation 1s probably tak('n from the words of the Commentary p 37G, 25 itz so tas.\a evarupe kale piyataro hoti, tath<i pituno But it Is to he observed that pifuno may be genitive, parallel with the precedmg tas.s<i (gemt1ve femmme), hence "dearer to her, as also to his father" This translat10n 1s made nece5&ary by tathii, which renders the two members parallel tasMi, and pituno - kanta (masculme, femmme, neuter) = "lovely, charmmg," 1s never found m Pah as a ma'>culme noun ( = a hu1,band), and the emend,1tion kantena for lean ten a must be reJected, all the more a& ka m tena ahhipassas1 gives exactly the right rneanmg Passatz = v1dere, but ablnpassati = m­tclhgere, mente p<·rc1pere, exphcare, mterpretan, etc , therefore we must tran&late what do you understand by that? (qmd 1llud mterpre­tari putas?) (Answer the child m his mother's lap)

2 l'J°ANTE

In connect10n with the above 1t 1s mterestmg to look at kante, Ja V, 486, 13 ( = santike, Comm) which the English translator of Ja V (Francis) fairly correctly render& "at man-eater's hand [I] go forth to die" ( = ,qaccham' aharn porwiidassa kante, v I natte) PTS's Diction­ary, which has also adopted the emendation kantena, Ja VI, 370, pro­poses here a very astomshing conJecture porisiidassak' ante ( "). But the correct readmg 1s evident v l fiatte 15 simply to be read nante, Just as the parallel passage Ja V, 26, a shows; nante 1s Sanskrit nyante, ind. = near to = santike. Abh 706 has by mistake nattarn instead of fian­tam (also Childers).

82 Dines Andersen

S. IssA (isA)-PHANDANA

Jataka No. 475 (IV, ~07) contains the story of a wild animal (lcalasiha) and a Phandana-tree (spandana) In the Gatha's this animal was called isa, m. (cf Abh 612 issa, "a bear, a sort of lion," sa r<;ya, nya) There is doubtless an allusion to this fable in Dhp-a I, 50, 18: issa-phandananarn viya, kiikolukanam viya, which is rendered by Bur­lingame, Bu Leg I, 174, as follows "hke the Snake and the Mongoos, who trembled and quaked with enmity, hke Crows and the Owls" (with a reference only to Paficatantra V). It appears clearly from the paral­lelism between the two expressions that the translation should be. "hke [the enmity of] the Bear and the Phandana-tree [or] hke [that of] the Crows and the Owls " The translation of istm-phandana (PTS's Dictionary s v. phandana) is therefore inadequate, the translation of Rouse (Ja V, 131), on the other hand, 1s correct.

The followmg stanza 1s found Ja I, 464

4 VYAHARATI

Yatlu'i vcica va bhufi1assu yatha blwttafi w v11ahara,

ubha7Jam te na mmeti vaca bhuttmi ea, Ko~1ycl

The Enghsh translation renders this as follows "You may act or eat, which shall it be? for you can't both, my Ko~iya," the German translation has "So wie du sprichst, so i.-,s Jetzt auch, und wie dem Mahl ist, handle auch, denn be1des mcht zusammen passt, die Rede, Ko.-,iya, und das Mahl " - Both translations miss the verb vyaharati, which, on the basis of Comm's periphrastic explanation _qehe kattabba­kammam karohi, 1s rendered in the German translation with "handeln" (arbe1tcn) But this meanmg of 1•yiiharati cannot be found anywhere in Pah literature It always means "to say, tell, speak, pronounce," a confusion with voharati (vy-avaharati), m the origmal meanmg of this verb, is hardly possible On the other hand, voharati, through vohdra ( which is both = vyiihara and vya1'ahiira, and means among other thmgs "usage" (as to language) but not "pronunciation" has also ac­quired the meaning of vyiiharati, but we never find the reverse develop­ment Vyaharati therefore means "to say," - but not "arbeiten," and that fit11 the thought of the stanza perfectly the Brahman woman's talk about her bad stomach is in contradiction to her over-indulgence in sweets (ya ea te ayam vaea 'vata mam VtJJhanti' ti yan ea te idam patJ,ita­bhojanam bhuttam, idam ubhayam pi tuyham na sameti), therefore the

Miscellanea Pdlica 33

words viicd and vydhara are in the same relation as bhuttam and bhun­jassu.

Either eat, as you speak, or speak, as you eat! Both do not go to­gether: your complaints and your way of eating, my girl.

5. APAKATANNU (A[P]PAKATNNU)

The Pali word a-katafi:iiu (ungrateful) 1s lackmg in PTS's Diction­ary, but - to this Helmer Smith has called my attention - through a curious misunderstanding it is inserted as a-palcatafiiiu (p 50) = "un­grateful" (Vin II, 199, 9) This form 1s Just as chimerical as "un-pro­grateful" ['] It is of course to be unden,tood as a-palmta-fi:nu (sa aprakrta-Jiia) = "who does not know the matter m question" Also the translat10n in SBE XX, 25G ·• ignorant of what he had in hand" is somewhat obscure. Trenckner quotes m add1bon Vin IV, 112, 7 bha­gavatii sikkhiipadam apanfiattam, te t•ii bhikldtu appakatafiiiuno, Dhp-a I, 143, 2 Va1Jiputtalcehi naMlcehi appakatanfiuln (novices havmg httle gratitude" [1], Burlmgame'1, translat10n), T'in I, 312 22 gocare ap­palcatafifiuno (opp. pakatafi11uno, concermng newly-arrived monks), Ps. II, 430, 13 vinaye appakatannu.

UNnERSITY OF CoPENHAGEN

L'INSCRIPTION DE MAHAN.AMAN A BODH-GAYA

ESSAI D'EX£G:BSE APPLIQU£E A I.,'£PIGRAPHIE

BOUDDHIQUE

PAR SYLVAIN LtVI

J 'AI deJa eu l'occasion, il y a vmgt-cmq ans, d'appeler !'attention sur une inscnption de Bodh-Gaya qm avait ete publiee par J. F.

Fleet en 1886 clans !'Indian Antiquary, et de nouveau ed1tee par lui clans le beau recue1l des Gupta Inscr1pt10ns en 1888 (No 71). Cette inscription commemore, en vers sanscr1b d'une facture savante, la fondat1on d'une chapelle du Bouddha, dans le v01smage du Bodhi­manda, par un mome nomme 1\Iahanaman, ongma1re de Lankadvipa; Mahanaman y expose '>a genealogie spmtuC'llc en rC'll1ontant de proche en proche, par Upasena, l\Iahanaman, lipasena, Ra.hula, Jusqu'a Bhava La date, expnmee en symholes numenqueb, e.-.t l'an 269 d'unc ere qm n'est pas autrement prec1-.ce L'i·d1teur de l'm.-.cnption, Fled, apres avo1r mterprete cette date en prt"nant pour pomt de depart l' ere Gupta ~HS/319 (269 = 588/589 AD) a finalement hes1te entre cette ere et l' ere Kalacun de Ced1 249 A D ( 269 = 539 / 540 A n ) J e crus pouv01r contester ces deux mtcrpretat10n5, et JC proposa1 de ramener la date <lonn ee a I' ere ~aka 78 / n) ( 269 = S48 /349 A n )

J'avai.-. en effet dccouvert, dam une compilation chm01se du VIie 51ecle, un fragment de la Relation du voyageur chmo1.-. Wang Hiuan­tb'e qm v1.-.1ta l'lnde plm,1eur.-. f01::, au miheu du VII' <,1ecle, cc frag­ment raeonte comment, au temp-" du r01 Samudragupta, deux momes de Ceylan envoye'> par le ro1 Sri l\ieghavarna avaicnt accompli le pelermage de l\lahabodhi, l'un de.-, momes s'appelait l\lo-ho-nan, tradmt en chmo1::i "Grand nom", l'autre Ou-po, tradmt en chm01s par "ense1gnement, prophetic " Le prenuer porte done le meme nom que le mome de l'mscnpt10n <latee !:W9, le second, Upa, pcut-etre sou.-, une forme normalement redmte, un Upabena 11 e5t tentant dt' supposer que le Mahanaman de l'mscr1ption, d15c1plc d'un Upasrna, et disciple a la seconde gen6rat10n d'un autre Upa..,ena, e::it Hkntique au l\laha­naman qm vmt en compagme d'un Upa-, <le Ceylan a l\lahabodlu sous le regne de Samudragupta Vmcent Smith, dans un article de l'lnd1an Antiquary (The fo5cr1pt10ns of l\iahanaman at Bodh-Gaya, April 190~, pp. 19~-197), s'eleva contre l'hypothese que Je proposa1s, 11 lui

86 Sylvain Levi

opposait des raisons chronologiques et des raisons epigraphiques. Samudragupta etait bien sur le trone en 269 saka = 848/9 An., mais Meghavarna eta1t mort depuis quinze ans, si on admet la chronologie singhalaise. Sur ce point, Vincent Smith s'est plus tard rectifie lui­meme· dans une note de son Early History of India, 3• edition, p. 288, il assigne au regne de MeghavarJ,la les dates 352-379 Apres avoir peche par exces, l'hypothese pechera1t en ce cas par insuffisance. Je n'msisterai pas, et Je m'arretera1 plutot aux raisons paleographiques, qm sont graves L'aspect des caracteres employes dans }'inscription de Mahanaman suggere une date plus basse que le JVe siecle

Jene rcprendrai pas ici une discussion vouee d'avance a la sterilite, faute d'un fait nouveau. Mais JC voudrais, en m 'aidant des progres accomphs clans la connaissance du bouddhisme depms quarante ans, revemr sur certa1rn, details de la traductwn donnee par Fleet, signaler des questions qui ont du Im echapper, et proposer a la critique des solut10ns nouvelles

L'inscription debute, comme il convient, par une stance en l'hon-neur du Bouddha

vyiipto yeniiprarneyah salwla§as-1rurii sari 1vatah saftiadhatuh ksunniih piisandayodhas sugahpatharudhas tarkasastriiblmpd,t<ih .~ampurnno dharmakosah prakrtmpuhrtah siidhito lokabhiityai sii.'ltuh §akymkamndlwr JJayah ciratararn fad yasassiirafanttram

Fleet tradmt · "V 1ctorious for a ve1 y long time 1s that doctrine, replete with fame, of the Teacher, the chief kinsman of the Sr1kyas, by which, lustrous as the full-moon, the inscrutable primary substance of exist­ence has been pervaded in all directions, by which the warriors, who are heretics, obstructive of the path of beatitude, have been broken to pieC'e!>, bcmg assailed with the weapon of log1<', (and) by which the whole trea!>ure of religion, that had been stolen by the enemy which is origmal nature, has been recovered for the welfare of mankmd'"

La traduction est irreprochable, a ne considerer que les mots, mais le sens suggere (dhvam) est l'ame de la poes1e indienne, et nous sommes ici en presence d'une composition poetique, ou le versificateur mame avec une elegante a1sance Jes comphcatwns du metre sragdharii Dans une stance bien fa1te, la suggestion se precise par des touches succes­sives, et le dermer root dmt a la fois porter l'ensemble et en fournir la clef Le poete a manifestement chmsi a dessein pour cette place le mot tantra, qui est parhcuherement riche en evocat10ns. Le D1ctionnaire de Petersbourg n'enumere pas moins de 18 sigmfications (sans y faire entrer en ligne le feminin tantri). Au sens propre, c'est }'instrument a

L'lnscription de Mahaniiman a Bodh-Gaya 37

tisser: le metier ou l'ensouple, ou la chaine; il en derive des sens secondaires dont les principaux sont, selon !'interpretation des lexi­cographes sanscrits svariis tracint<i "les preoccupations de la royaute," kutumbakrtya "les affaires de famille," pradhiina ou mukhya "prmc1pal, essentiel," paracchanda "dependant," itikartavyata "regle a suivre," siddhanta "doctrine, ecole," siistra "tra1te technique," siistrabheda, §rutisakhiintara "tra1te special, branche particuhere de la tradition sacree"; dans ce sens, le root a pr1s une vakur tres precise, il s' applique a une categorie extraordmairement ahondante d'ouvrages ou la mys­tique et la mag1e se combment et qui a fleuri clans la bouddh1sme non moins que clans le brahmam:,,me Fleet a, parmi toutes ces s1gmfi­cations, legitimement choisi celle de "doctrme", mai:,, 11 faut observer pourtant que tantra amsi entendu 1mphque par un hen intirne le "traite," le "hvre" OU la doctrme e<,t enoncee Justement le root dharmakosa, a la tr01s1eme hgnt', :,,emhle preparer par anbc1pation un double sens de ce genre. Fleet l'a tradmt par "treasure of religion," et c' etait son dro1t, l' exprc:.1>10n a ce sen1, par exemple clans l\Ianu I, 99:

brahmano Jiiyamii.no hi prthu·yam adhi Jiiyate is,,arah sarmbhiitanam dharmakosasya guptaye

l\fms clans le domame du bouddh1smc, le mot ne peut manquer d'evoqucr le titre d'un ouvragc celehre, adm1s comme une autorite canomquc par toutes les ecole1>, l'Abhidharmakosa de Vasubandhu. Que les deux termes pmsst'nt alterner, JC n'en veux pour preuve que ce seul fmt J'm eu la chance de decouvrir a Katmandou, lors demon sejour en 1922, un manm,cr1t mcomplet des Kunka de l' Abh1dharma­kosa (du Jtr au IV 0 kosasthana), le feuillet m1tJal portmt l'md1cabon: Dharmakosa, et sur la f01 de ce titre, le Pandit S1ddhihaqa, detenteur de ce manu:.cnt qu'1l n'ava1t pmais lu, s'etmt 1magme que l'ouvrage etait un vocabulmre des termes techmques du bouddlusme, a la fa<;on du Dharmasamgraha l\Ims l'autt'ur de l'Abludharmakosa, clans la vrtti qu'1l a lm-meme ecr1te pour exphquer les kfmka, mterprete ce mot ains1. "L'ouvrage s'appelle Abludharmakosa, parce que l'Abh1-dharma y est, quant au sens, mtrodmt pour l'essenbel; ou b1en parce que l' Abhidharma en est le con tenant" idam tu siistram lwtham Abhi­dharmakosam ui tasy<irthato 'sminn iti yatlzapradhanam antarbhiltah atha vii etasyasrayabhiltah 1 Le commentateur Yasom1tra 2 glose en ces termes: "Le demonstratif (tasya) des1gne le traite technique qm porte le nom d'Abh1dharma, et qui va etre enonce immediatement, quant

1 Abh1dharmakosavyakhya, ed de la B1bhotheca Buddluca, I, 10. 2 Ibid. ad locum.

38 Sylvain Levi

au sens (arthatah), non point quanta la lettre; le sens y est incorpore, mais non pas la totalite, c'est cequ'il indique par }'expression "pour l'essentiel" (yathapradhanam), ii est incorpore (antarbhiita), intro­duit. C'est pourquo1 ce traite que Je compose tient heu de kosa au sens, est comme un kosa pour lui; clans ce cas, le mot est un compose du type tatpurusa fonde sur un rapport de genitif Abhidharmakosa veut dire alors le kosa de I' Abhidharma Le kosa (le fourreau) clans lequel entre l'epee est le kosa de celle-ci Ou bien encore l'Abhidharma, c'est a dire le Jnanaprasthana et autres textes analogues, est le con­tenant de ce traite que JC compose De cet Abh1dharma canomque, mon traite en a ete tire, quant au sens, comme il a ete d1t plus haut. Ence cas, Abhidharmakosa est un compose du type bahuvrih1 qm s'ex­phque amsi L'Abhidharma est le kosa de ce (hvre-) ci; ce d'ou on bre l'epee est le kosa de celle-ci, mon livre a pour kosa (fourreau) l'Abhidharma Il md1que par la qu'1l a rendu mtellig1ble le sens ge­nera] de cet (Abh1dharma) "Nous voila b1en loin du sens vague et flou dont Fleet ava1t du se contenter "tresor de la rehg10n ", nous avons affaire a un hvrc connu, celebre, cons1dere comme l'expo-.e classique de Ja doctrme bouddluque au point de vue plu]osopluque Trouverons­nous clans le contexte de nouvelles suggestions qm confirment celle-ci? L'auteur de l'Abhidharmakosa est Vasubandhu, cc nom est generale­ment rendu en chmois par 1!;t :t l "apparente (par alliance) au monde," quelquefois aussi par ~ .!r.... "apparente (par alliance) a (de<;) cheu(x) " En tibctam, il est rcndu par dby1g giien "apparente au(x) tresor(s) ", dbyig, qm ren<l ic1 le sam,crit vasu, a aussi le sens <le sara, hiranya, T'GJTU d'apres IC's d1ebonnairt'S t1beta1rn, sanscrits. L'ele­mt'nt bandhu parait It la quatr1eme hgne de l'mscr1ption, dam le mot Saltyaikabandhoh qm y C'st Pmploy6 pour designer le Bouddha Le morns qu'on pmsse dire de ct'tte appellat10n, c't'st qu'elle ebt etrange Le Bou<ldha re<;o1t fr6quennnent, meme clans lcs textcs canomqucs, l'ep1thete d'Adztyabandhu, en pah Adicca 0 "apparent6 au Soled," quelle que soit l'mtcrprctation qu'on &Oit tente <le donnt'r :\ Ct'ttC appellation chargee d'un long passc de mytholog1e 1\1:.w, Je ne me rappelle pas d'exemple de Sakyabandhu, t't Jene puis me defendre d'une irnpress10n <le surpri'>e a rencontrl"r cc mot Le Bouddha est Lien le Sakya par excellence, le sage dt's Sakya (S0muni), le hon des Sakya (S0s-imha), le taureau des Sakya (S0pumgava), tom; ces termes Im a!>signent une place d'honneur clans le clan. Mais a le designer comme "le parent, le parent umque des Sakya," 11 me sernble que les valeurs se renver­sent, comme si l'honneur lui venait d'etre apparente ace clan. Je suis done amene a croire que le poete a mtrodmt cette periphrase parce

L'Jnscription de Mahanaman a Bodh-Gaya 39

qu'il avait clans l'esprit un autre nom ou figurait l'element bandhu. Le terme loka, clans le compose lokabhutyai qui precede, evoque la tra­duction consacree en chinms du nom de Vasubandhu · "apparente au monde." Et c'est encore une autre des valeurs usuelles de vasu que suggere le mot sara accole a tantra, sara correspond au bbetam dbyig "objet de valeur" qui rend le mot vasu dans le nom de Vasubandhu. En outre il evoque tout naturellement, comme son v01sin tantra, une idee d'ordre htteraire. Sara est le correspondant indien de la "somme" med1evale que L1ttre defimt "Titre de certams hvres qm tra1tent en abrege de toutes les parties d'une science", et nous avons vu que l'auteur de l'Abh1dharmakosa pour Justifier ce titre allegue que son ouvrage est un abrege de }'ensemble de l'Abhidharma

Dans l'hypothese que je propose, les autres traits v1ennent se co­ordonner sans effort autour de l'eqmvoqut- centrale. Le Bouddha, "lummeux comme la lune en son plem, a penetre de toutes parts le monde des creatures qu'on ne peut mesurer" L'Abhidharmakosa qui donne une description physique et psychologique de tous les etres (sattva) clans tous les mondes (dhatu) a d:r01t au meme eloge Le Bouddha "a ecrase les soldats de l'heresie qm barraient le passage a la Bonne Destmation, en le1i a1,sa1llant avec les armes du raisonne­ment" L'eloge, ici, convient parfa1tement a l'oeuvre de Vasubandhu; le phdosophe y refute chemm fau,ant les theses condamnees par l'or­thodox1e, le dernu~r chap1tre se presente meme expressement comme unc "Refutation" dmgee contre les doctrines qm adherent a la per­sonnahte (pudgala) Le mot sam piirna "remph" qm accompagne l'expre&Mon dharmako§ah n'est pas morns &uggestif, c'est le terme qui marque reguherement la fin d'une compos1t10n, d'un manuscnt Quant a l'ep1thete prakrtmpuhrtah, Je DC cro1s pas que Fleet l'ait mterpretee exactement "stolen by the enemy which is ongmal nature", Je ne vo1s pas que le bouddh1sme ait denonce la Nature comme l'ennem1; c'est un trait propre au Samkhya, et que le bouddhi&rne ne lm a pas emprunte. Le Grand Yehicule a meme fim par enseigner que les etres sont de nature (pralcrtyii) en etat de NirvaJ)a Ll' sens le plus simple de prakrtiripu, c'est "cnnemi nature!'', les ennemis naturels du salut, sans doute Mara et ses aux1haire&, avaient detrmt le Tresor de la Loi, puisque la L01 est condamnee a d1sparaitre au bout d'un mtervalle fatal apres chacun des Bouddhas success1fs Le Bouddha a reussi a retrouver (siidhita) ce tresor pour le bonheur du monde. Mais il est probable que l'epithete prakrtiripuhrtah dmt aussi s'apphquer a l'Abhidharmakosa; on peut supposer que l'Abludharma, le systeme et aussi la litterature de ce nom, avait apres la grande epoque des ou-

40 Sylvain Uvi

vrages classiques: Jftanaprasthiina etc .... passe par une eclipse d'ou le genie de Vasubandhu l'avait degage. Ainsi d'un bout a l'autre, la stance se developperait comme il convient en deux couches paralleles de significations; en exaltant le Bouddha, elle exalterait aussi ce maitre salue couramment comme un Bodhisattva, Vasubandhu, et la question de la date prendrait une nouvelle importance en presence des opinions divergentes sur la date de Vasubandhu.

La seconde stance est un e]oge de Mahakasyapa; ici, nous sommes sur un terrain plus i,,O.r, et la traduction de Fleet aura a sub1r de fortes retouches. Le textc porte:

nairodhim subhabhiiMniim anusrtah sarnsiirasamklesajin M aitreyasya kare ,,imuktwas1tii yasyadbhutii ,,yalqtii mrviiniivasare ea yena caranau drstau muneh pa,,anau piiyiid vah sa munindrasiisanadharah stutyo 1l ahiilciisyapah

Fleet traduit. "May he, Mahakasyapa, who is worthy of praise, pro­tect you, - he who observed the precl'pts of (Buddha) the chief of saints; who practised that auspicious habit of abstract med1tat10n which 1s of the nature of a trance, who overcame the angmsh of suc­cessive states of existence, whose wonderful subjugation of the pas­sions in final emanc1pat10n (is to be) d1i,,played m tlw hands of Maitreya, and by whom the two pure feet of (Buddha) the saint were behold at the time of attainmg Nirvana."

Les deux dermeres hgnci,, de cette i,,tance sont parfa1tement claires, OU presque, l'ep1sode evoque par la tro1s1eme hgne est b1en connu. c'ei,,t Mahakasyapa qui a vu Jes p1e<ls du Maitre a l'occas1on du Nir­vana, le bucher qui devait consumer lPs restes sacres ne pouva1t pas s'allumer avant que ]e grand d1sc1ple "le fils aine du Boud<lha" comme l'appeJle ]e Vmaya des l\fahasamglnka, ne fut venu salucr respcctueusc­mcnt les pie<ls du Bouddha. 1 C'est un episode commun a toutes les traditions, Miilai,,arvastivadm, Sthavira (pah), Mahasamgh1ka. A la quatrieme hgne, l'ep1thete munindrasasanadharah que Fleet tradmt: "he who observed the precepts of (Buddha) the chief of saints" a cer­tamement une valeur mieux defime. L'adJcct1f dhara, combine avec Jes noms qm des1gnent les parties mtegrantes du Canon, ind1que une competence spec1ale dans teUe ou telle branche · on est Vinayadhara si on possede a fond le Vinaya, Matika (Matrka) dhara si on connait les rubriques de l'Abhidharma, Dhamma (Dharma) dhara, si on est verse dans l' ensemble. Mais Mahiikasyapa dispose d'une connaissance

1 Cf Przylusk1, le Par1mrv1lna et les Funera1lles du Buddha, clans Journ As1at 1920, I, 19-20

L'lnscription de Mahanaman a Bodh-Gaya 41

incomparable: il connatt tons les textes sacres, et c'est ace titre qu'il est designe a l'unanimite pour presider le concile qui fixe pour la pre­iere fois le Canon. L'epithete de Sasanadharah rappelle expressement cet honneur privilegie qui est echu a Mahakasyapa

La seconde ligne fait allusion a un mcident qui n'a pas echappe a Fleet, quoique le passage, 11 l'avoue, Jui para1sse obscur Fleet rappelle en note, ad locum, que "le Buddha, sur le pomt d'atteindre le Nirvana, a remis a Mahakasyapa sa robe Jaune (ka~aya), en le chargeant de la transmettre a Maitreya, quand celui-ci attemdra1t la condition de Bouddha," et 11 renvoie a la traduction de Hman-tsang par Beal 1

L'etude des legendes du Bouddha Maitreya a fa1t de notabls progres depuis la publication des Gupta Im,cript10ns Mr Matsumoto Bunza­buro a publie en 1911 une remarquable monograph1e sur la Terre Pure de Ma1treya (Miroku Jodo ron) qm a ete savamment analysee, discutee et complctee par Noel Pen dans le Bulletin de l'Ecole Fran­<_;aise d'Extreme-Onent, tome XI, 1911, pp 439-457 l\f Przyluski a rcuni divers textes sur le Nirvana <le l\Iahakasyapa dam, un articJe sur "le Nord-OuC'st de l'Inde dans le Vmaya <lC's Mula Sarvashvadm, Jour Asiat. 1914, II. 522 sqq. 1\I. Ernest Leumann a pubhe une Iongue etude sur un ouvrage red1ge en As1e Centrale, "Ma1treyasam1ti, das Zukunftideal der Buddlrn,ten," 1919, 2 fasc , dans l'mtroduchon il a resume lcs prmc1paux textes de la htterature maitreycnne du Canon chinms, p. 11-23 M P Dem1ev1lle, en rendant compte de ce travail dans le Bulletm de l':Ecolc Franc;a1~e d'Extrcme-Onent, tome XX, pp. 158-170, a Jui-meme apporte de nouveaux documenb Un autre ouvrage or1gmaire de l'A~1e Centra]e, le 1\Iaitreyasannti-nataka, connu par des fragments decouverb a Tourfan, a susc1te plus1eurs travaux dus a 1\11\[ Sieg et S1eglmg et ,\ 1\1 F. W K 1\luller, et pu­blic-, dans les Compte~-rendm, de l'Academ1e de Berlm, 1908 et smv

S1 nous acceptons la traduct10n de Fleet pour Ja scconde hgne de la stance, il est JUste de nous demander comment "Ja merve1lleuse sub­jugation des passions (de Mahaka~yapa) clans l'emanc1pation finale (doit etre) etalee dans ll's mams de 1\fa1treya" Le geste de Maha­kasyapa remettant au nouveau Bouddha la robe de ~on predecesseur Sakyamum, ne suffira1t pas a "mamfester cette merve1lleuse subju­gation." D'a1lleurs, un detail de plulologie nous avertit ic1 cl'une erreur probable. Le mot vyakrta a, commc toutes Jes formations tirees du verbe vyakar0

, une valeur speciale et bien definie dans la langue du bouddh1sme, ii s'agit touJours en ces cas de propheties dues au Bouddha, et qm revelent l'avenir d'un ind1vidu, d'une Jocalite etc.

1 Buddh Ree West World, II, 142sqq

42 Sylvain Levi

La carriere de Maitreya, y compris sa visite a Mahakasyapa, est tou­jours enoncee, puisqu'elle appartient a l'avenir, sous la forme d'une prediction communiquee par Sakyamuni a ses disciples. II n'est done point douteux que, clans le passage en question, vyiikrta doit recevoir son sens spec1fique "predit ", on aurait alors pour la hgne entiere: "Dans la mam de Ma1treya, la subJugation des passions (de Maha­kasyapa) clans }'emancipation finale a ete pred1te (par le Bouddha)." La traduction obtenue est absurde et s'elimine d'elle-meme.

Arriverons-nous a un sens me1lleur par la voie de I' expression vimuktivasitii. La vasitii est la maitr1se absolue exercee par le saint, Arhat ou Bodhisattva selon les V ehicules, sur des categories determi­nees de domames La Mahavyutpatt1 (XXVII) donne une liste des 10 va§itii des Bodhisattva, la vimuktivasitii n'y figure pas, ma1s on y releve un mot tres analogue a vimulcti, l'adhimukti qui constitue la sixieme des dix va§itii Sommes-nous en presence d'une alternance de termes? Un passage curieux de l'Abhidharmakosa, mamtenant ouvert a la cur10site des chercheurs par le labeur intrep1de de M. de Lavallee­Poussm, semble le suggerer Au hvre VII, 52 Vasubandhu mstitue prec1sement une d1scuss1on sur le cas de la surv1e de Mahakasyapa; il v1ent d'examiner les conditions <l'existence des etres susc1tes par mag1e (nirmita), ils <lurent grace a l'adhisthiina de celm qm les a crees. M. de Lavallee adopte le mot "protection" pour tradmre adhisthiina; je cite sa traduction, p 120. "Ce n'est pas seulement pour la duree de sa propre vie que le createur est capable de 'proteger' une chose de telle manicre qu' elle dure, &a 'protection' peut faire auss1 que la chose dure Jusqu'apres sa mort C'est ams1 que par sa protection (adhisthiina ou adhimolcsa, resolut10n) Kasyapa le Grand a fait que ses os dureront Jusqu'a l'avenement du Bhagavat Maitreya Non pas a l'endr01t de ce qm n'est pas dur C'est seulement la chose dure qui est susceptible d'etre 'protegee' pour une longue duree C'est pour­quoi Kasyapa le Grand ne 'protege' pas sa chair D'autres maitres d1sent: Non. - Le corps protege par la force de la 'resolution' ( adhi­molcsa) n'est pas capable de <lurer au-dela de la mort. S1 Jes os de Kasyapa durent, c'est par la 'protection' (adhisthana) des d1eux."

L'alternance adhisthiina-adhimoksa, ind1quee par M de Lavallee sans autre explication, est fondee en fait sur la Vyakhya de Ya&om1tra encore inedite pour cette partie du texte Mon manuscrit lit (p 272b): iiryamahiilciisyapiidhisthiineneti. iiryamahiilciisyapiidhirrwksenety arthah. Amsi c'est la vertu de l'adhimok~a de Mahakasyapa qui fa1t durer ses os, et ses os seulement sans la chair. M. de Lavallee, si famiher avec !'ensemble de la htterature bouddhique, n'a pas manque de c1ter en

L'lnscription de Mahiiniiman a Bodh-Gaya 43

note sur ce passage une ligne du Divyavadana, 61: M aitreyah. Kiisyapa,<Jya bhiksor amkopitam asthisamghiitam daksinena piinina grhitva ... "Ma1treya, prenant dans sa mam droite }'assemblage des ossements du blnk-,u Mahaka~yapa qm n'ava1ent pas bouge ... " Ce detail sau,1ssant coincide de mamcre frappante avec Jes termes de notre stance "En ce qm concerne la mam de Mmtreya, la ma1trise absolue de survie (de MahakaAyapa) a ctc predite (par Sakyamuni)." Le passage fa1t partie, danb le D1vya, du l\fa1treyavadana, en rcahte, ce rcc1t prophetique de la carriere de Mmtreya a ete decoupe dans le Vmaya des MulaSarvai>tivadm, 11 y est msere dans l'O'ladh1vastu, Chap VI de la version chino1sP 1 Mmtreya, arrive a la Bodhi com­plete, entoure d'une foule de d1sc1ples, se rendra au mont Gurupada, ou la masse des ossementi, de (Maha) ka~yapa le bh1k-,u reste sans bouger, la montagne s'ouvnra pour J\taitrcya, alorb l\Ia1treya prendra dans ba mam dro1te l'as<,emblage des Obbemcntb du bh1ksu Kasyapa qui n'ont pas houge, 11 les po5era sur sa mam gauche, et c'est ain6i qu'1l ensc1gncra la Loi "

La comc1dence de l'1m,cripbon rwee le Vmaya des l\f ulaSarvasti­vadm est frappante, die J'c.,t davantage encorP quand on constate que danb la htt(>rature ma1tr(>ycnne, pourtant ahondante, le detail ne reparait nullc part a1lleurs Dans un certam nornbre de rec1ts, Maha­ka~yapa tire de son recue1llement par l'arrivee de l\fa1treya, Im remet le vetemcnt (:..amgluit'i) du Bouddha que le l\Iaitre Im avmt donne pour le pa'lser a son 1succes5eur, 2 l'cp1so<le manque danb la version abregee mibe aussi b0US le nom de KumaraJiVa, 3 il manque auss1 U la redaction vers1fi(>e tradmtc par Y1-tbmg ( et dont J 'a1 retrouve l' or1gmal sanscrit: Ma1trcya-vyakarana), 4 dam le Koan l\11-le p'ou sa hm cheng king, trachut par Tchou Fa-hou 5 Maitreya, parvenu en prc">ence des restes de Mahakri~yapa, prend lm-meme la samghati que celm-c1 avmt re<;ue du Rouddha, le Vmaya des l\IfllaSarviishvadm enreg1stre auss1 cette forme de la legende dans la 1seet10n du Ksudrakavastu, chap 40, Tok. XVII, ~. 94a col 3, M Przyluski a tradmt ce morceau du Vmaya dans le Journal Asrntiqul', Hll4, II, 527, c'cst ams1 cette forme qm a ete incorporee clans l'Asokavadana, v Przylusk1, La Legende de l'Em­pereur Asoka, p 334.

1 Tok XVII, 4, 21b, col 5, texte tihetam <lans le Dulva, vol II ~ M1-le ta tch'cng fo kmg, trad KumiiraJi\,a NJ 209, Tok IV, 546b, col 3 3 M1-lc h1a cheng kmg, NJ 205, Tok XXV, 9, 31b, col G 4 M1-le h1a d1eng tch'cng fo kmg, NJ 207, Tok IV, 5, 50-51 • NJ 208, Tok IV, 5, 48b, col 12 = Tseng-y1 A-ban (Ekottariigama) chap 44,

NJ 543, Tok XII, 3, 34b, col 16.

44 Sylvain Lki

Les deux formes de Ja legende se ramenent peut-etre a la confusion, accidentelle ou volontaire, de deux mots tres voisins: dans un cas, ii s'agit de la samghiiti, la robe (du Bouddha); dans l'autre, il s'agit d'un samghiita, ou plutot d'un samghiita, un assemblage (d'ossements) ou une charpente (d'ossements); le texte du Divya donne samghiita, mais samghiita ne serait pas moins correct, et le pali clans }'expression atthi­samghiita a normalement la cerebrale. J'ai deja signale, et aussi clans le D1vyavadana, et dans un morceau qu'1l a egalement emprunte au Vmaya des MiilaSarvast1vadin, une confusion analogue 1 dans l'enonce du 82• s1ksapada, !'expression anirgatiiyiim ra1anyiim correspondant a anikkhantariiJake du pacitbya 83 clans le Vmaya pali; ra1ani "la nuit" a rem place riijaka "la personne royale." 2

J'ai adm1s 3usqu' ici la substitution du root adhimukti au mot vimukti dans le texte de l'inscripbon. La lecture donnee par Fleet est cependant exacte Mais la lettrc v ne se d1stmgue de la lettre dh, dans l'ecriture des Gupta, que par un petit trait horizontal trace au sommet de la panse commune aux deux caracteres Une confus10n, du fail du scribe ou du graveur, est touJours possible. L'a initial du root adhimukti dmt s'ehder apres la finale du mot kare. Mais meme s1 on ecarte cette correction, l'alternance vimukti-adhimukti peut se justifier par de nom­breux textes Vasubandhu Iui-meme se scrt de l'un des termes pour definir l'autre 3 La vimukti "dehvrance" est de <leux sortes, composee ou simple L'adhimoksa est la vimukti "composee," et la Vyakhya de Yasom1tra glose· adhimoksah samskrtii vimulctir iti dhiitvarthaika­tvat "parce que la racine (muc) a le meme sens (clans les deux mots)." La V1bhasa, dans un passage cite par Lavallee Poussin, porte: 4 "Le Bhadanta dit que les vimolcsa sont ains1 nomm6s parce qu'ils sont ob­tenus par la force de l'adhimolcsa" A1lleur1, cncore, 6 Laval1ee Poussin cite Samghabhadra, pour l'exphcation du mot adhimukti· "D'apres d'autres mai:tres, adhi sigmfie" superiorite, souverainete, "mukti sig­m:fie tmnolcsa " 11 serait mseux de multiplier ces temmgnages.

Qu'est-ce done que l'adhimukti ou adhimoksa? M de Lavallee Poussm a reum et discute plusieurs defimtion1, t1rees des traductions chinmses; mais nous avons mamtenant l'avantage de pouvoir utihser dans son or1gmal une definit10n donnee par Vasubandhu lui-meme, et

1 Makand1kav, p 543 sq ' Kuchean Fragments, p 361, clans R Hoernle, Manuscript Remams of Buddhist

Literature found m Eastern Turkestan. 3 Abh1dharmakosa VI, 76ab 4 lbui , VIII, 32d, p 207 6 Ad II, 24, p 155

L'lnacription de Mahdndman a Bodh-Gaya 45

dans l'ouvrage classique qui passe pour couronner sa carriere. La vrtti sur la karika I O de la Trimsika <lit: adhimokso niseite vaatuni . . tathaivdvadhiiranam. niseitagraharJ,am aniseitapratisedhiirtham. yuktita dptopadesato vd yad vastu asamdigdham tan niseitam yenaiviikiirerJ,a tan ni§eitam ity aduhlchiidyiikiirerJ,a tenawiikiirerJ,a tasya vastunas eetasy abhinivesanam evam etan ndnyathety avadhiiranam adhimoksah. sa e<isamhiiryatiidiinakarmakah. adhimuktipradhiino hi svasiddhantiit para­praviidibhir apahartum na sakyate "Adhimoksa, c'est apropos d'un ob1et decide, l'affirmation qu'il est exactement amsi Il <lit 'decide' pour eliminer tout ce qui n'est pas decide Decide veut dire un oh1et qui est hors de doute en vertu d'un raisonnement ou en vertu d'une autorite competente. De quelque nature que cet obJet ait ete decide, par ex­emple exempt de douleur etc ... , la certitude ancree clans l' esprit que cet obJet est prec1sement de cette nature-Ja, c'est l'adhimoksa. II a pour effet de donner l'mebranlabilite Celm chez qui l' adhimukti domine ne peut pas etre detache de sa doctrine par le~ champions d'autres doctrines" Yasom1tra, dans sa Vyakhya sur le Kosa II, 24, p. 154, rapporte d'autres definitions qui ne s'ecartent pas sensiblement de celle-c1. Adhimoksa est done la "conviction" Un exemple frappant de ce qu'est l'adhimoksa est fourni par le Samyuttanikaya I, 116. 1

iikankhamiino ea pana bhante Bhagavii Himavantam pabbatarii3am suvannam ty eva adlmnuecey11a, suvannaii ea pan' assii ti. 'Bhagavat, pour quelque fin, eta1t convamcu que I'Himavat est en or, la montagne sera1t en or." J<:t de meme Kathavatthu II, 608. iiyasmii Pilinda­vaeeho rarino .M iigadhassa Seniyassa Bimbisiirassa piisiidam suvanrJ,an teva adhimucci suvanno ea pana iisi "L'ayasma P1lmdavaccha eut la conviction que le palais du roi de Magadha, Seniya B1mb1sara, etait en or, et le palms fut en or " La conv1ct1on d'un saint ne saurait etre fausse, et c'est ains1 que l\Iahakasyapa s'etant convamcu qu'il devait attendre l'arr1vee du futur Bouddha avant d'entrer dans le N1rvii.J,1a, scs ossements se sont mamtenus intacts (at"tkopita) La premiere ep1thete de Mahakasyapa va mamtenant a son tour se pre­ciser: nairodhim subhabhiiMniim anusrtah Fleet traduisait · "who practised that auspicious habit of abstract meditation which is of the nature of a trance." Mais chacun da ces termes a sa valeur technique. Nairodhim e&t un adjectif derive de nirodha; le nirodha, c'est le "bar­rage" de l'intellect et des dharma de l'ordre de !'intellect (eittaeaittii­niim nirodha~). L'exercice spmtuel (bhiivanii) de l'ordre du nirodha,

1 SamyuktAgama, Tok. VIII, 4, ~8, col 1, cl. auss1 Mahiiyanasutriilamkara. XVIII, 88

46 Sylvain Levi

c'est la nirodhasam<ipatti, etat analogue au Nirva:r:ia (nirv<inasadrsa), 1

mais qui s'en differencie parce qu'il est susceptible de rechute, la conscience verbale (samjn<i) et la conscience de sensation (vedanii) ont alors d1sparu II conshtue le dernier terme de la chaine des vimoksa qui sont au nombre de huit; 2 et l'exerc1ce des vimoksa a pour frmt, entre autres, "le pouvoir par lequel un samt transforme ou fait <lurer (adln ~th<ina) ", 3 Yasom1tra, dans son commenta1re sur ce passage, mtrodmt le terme adhimoksa cette f01s encore pour gloser le mot adhi­sthiina du texte tath<idhimoksiid ad/zisthiinam stlnrasya vastunah iyantam kalam amfHthatiim ifi.; ams1 la nairodhi bha,,anii de Mahakasyapa precede en fait, comme dans la stance, sa t•1muldwas1tii (adhim 0

)

Qu'1l s'ag1sse, clans le cas parhcuher de Mahakas,vapa, de la nirodha­samiipatti, c'est ce qm est prouve par le texte du l\ia1treyavyakarana tradmt par KumiiraJiva 4 A l'arrivee de Maitreya dans sa caverne,

l\fahakasyapa sort fJt" S: lF "de la mcd1tat10n de totale extmc­

hon "; c'est la exactement la traduchon com,acree du terme mrodlwsa­m<ipaffi 5 Le terme subhabhar,a11ii, employe metri causa, est une pen­phrase exacte de samiipatft, car bluivana est exphque par Vasubandhu comme sam<ihilam ku.folam 6 "le blC'n al' etat de recue1llement", chacun de ces deux termes est un eqmvalent ai>sez exact de chacun des deux termes subhabhai•ana Il n'est pas Jusqu'au mat mwsrtah d'aµparence assez anodme pourtant, qm ne semble expmner urn' notion d'ordre techmque Un adJechf derive du verbe anusar 0 se combmc avec sraddha et dharma pour des1gnt>r les deux prermers degrc<; de la samtete §raddhanusiinn et dharmiinusiirin 7 le stage smvant est le £raddhii­dhim ukta, pms v1ent le drst1priipta, et ensmte le kayasaksin lequel a realise la mrodhasamiipatti On voit clairement que la 5mte des epi­thetes s'enchaine clans une relatmn de causahtc conforme aux exigences de la theor1e ascolasbque

Le reste de l'mscription est essenbellement narrabf, je ne rcben­dra1 que le premier mot, m1s en tete de la stance qm i>mt l'mvocabon a Mahakii~yapa Les disciples de ce samt y re~OlVent l'ep1thete de Samyuktagaminah Fleet traduit "endowed with a connected trad1-t10n of doctrme " Le sens attribue a samyulcta par Fleet est pour le moins douteux. Mais Samyuktagamin ne peut manquer, clans un texte d'mspirabon bouddh1que, d'evoquer le nom sacre du Samyukta-

1 Abludh Kosa II, 44d 2 Abh Kosa VIII, 33 3 Abh Kofa VIII, 34 7 Abh Kosa VI, 29 et 62-63

4 NJ 207, Tok IV, 5, 46b 6 Cf p ex Mahavyutpatt1 CIV, 67. 6 Ka&a, IV, 123 c. d

L'lnscription de Mahanaman a Bodh-Gaya 47

gama correspondant au Samyutta nikaya du pali, un des quatre Agama (en pali N1kaya) qui constituent les plus anciens recue1ls de siitra. Amsi les disciples de Mahaka.§yapa etablis a Ceylan se reclament du Samyuktagama, le trait n'est pas assez caracter1stique pour preciser leur ecole, les Mahasamgh1ka, les Sarvasbvadm. les MiilaSarvastiva­din soot d'accord pour placer en tete des Agama le Samyukta, que les Sthavira classent seulement en tr01sieme hgne 11 y a la toutef01s un ind1ce de plus pour nous empccher de ranger sommairement Ma­hanaman parm1 les Theraviidm, Ceylan, nous le ~avons par plus d'un temo1gnagc, eta1t lorn d'etre le domame de cette ecole exclm1vement clans les premiers siecles dC' I' ere Au reste, la queoa;t10n de I' Amradvipa ou Mahanaman est mstalle en residence (adhmism) est lorn d'etre resolue Fleet ava1t admis sur la foi d'une commumcation de Cunnmg­ham que "le nom d'Ile de la l\Tangue est un autre nom de Ceylan, qui aurait ctc designce amsi a cause de sa forme semblable au frmt du mangmer" J'1gnore s'il s'ag1t l:'t d'une fanta1sie p<'rsonnelle de Cun­nmgham, Je n'ai 1amais rencontre cette designat10n d'Amra<lvipa clans aucum texte pah ou sanscnt l\lai~ Je ne pu15 m'empecher d'ob­servcr que 51 on admet ce sens pour ce pm,sage, l'ep1thete Lankadt•"ipa­prasutah "ongmaire de l'ile de Lanka" para1t tout a fa1t 01seuse :Mais 3e ne veux pas m'engager 1c1 clans C<'tt<' d1scuss1on d'ordre g6ograph1que Je me sms propose seulement <le montrer, par un exemple cho1s1, a quel pomt l'ep1graphw bouddh1que est mseparable de l'etude de!i textes, quelle lurrnere C'lle peut en recevoir et auss1 leur apporter Je suis heurC'ux de dcd1er cette tentative a mon tres cher am1 Ch R. Lanman lt qui l'mdiamsme et les md1amstes do1vent tant d'obhgabons, le maitr<> de Henry Clarke Warren, l'fditeur de cettc magmfique Har­vard Oriental Serws qu1 a debute par la Jatakamala de Kern et qui v1ent encore de noui, donncr en trms beaux volumes la traducbon du commentaire du Dhammapada par E W Burlmgamc.

COLLEGE DE FRANCE

THE DATE OF THE AMOHINI VOTIVE TABLET OF MATHUR.A

Bv EDWARD JAMES RAPSON

IN MY chapter on the "Scythian and Parthian Invaders " in the Cambridge History of India (Vol I, pp. 563-585), I accepted without

hesitation "year 42" as the correct reading of the date m the Amohini votive tablet, which was inscribed while the <;;aka ruler Co<Jasa was governmg Mathura as Great Satrap, and I noted (page 576, note 1) that Buhler also had originally been mclined to read the decimal figure as 40,1 but had subsequently changed his opmion and preferred to re• gard this symbol as representing 70.2 From a careful exammation of the photo-lithograph which illustrates Buhler's edition of the inscrip­tion in Epigraphia lndica (Vol II), I was convinced, as I am still con­vmced, that Btlhler's first reading was correct, and that the sign m question should be read as 40.

Professor Sten Konow m his article on the Takht-i-Bahi inscrip­tion 3 says: "Professor Rapson has evidently overlooked the important article by Professor Luders, Ep Ind, IX, pp 243 ff, where it has been conclusively proved that the numerical symbol used in the in­scription stands for 70 and not for 40."

It was evidently undesirable m a work hke the Cambridge History of India to discuss mmute pomts of scholarship, and I contented myself, therefore, with an expression of my own opinion and that of the editor of the im,cript10n concernmg tins much debated point. But the views of a d1stmguishe<l epigraphist hke Professor Luders deserve serious consideration, and I am glad of an opportunity of explaining why m this particular mstance I feel unable to agree with 1nm

I have now most carefully studied the article of Profo,sor Luders, with the result that, while admiring the accurate and scholarly manner in which he has collected and arranged all the relevant evidence, I cannot accept the conclusion which he draws from this evidence.

The Briihmi numerals for 40 and 70 which occur m the inscriptions of the <;;akas and the Kusiinas either very closely resemble the Brahmi aksaras '-" = pta and ;! = pil, or they are conventionalized (cursive) forms which are manifestly derived from them. The aksaras them-

1 Epigraphw lndwa, 11, 199 2 lbul' IV, 55 3 Acta Onentalia, m, 57, n 1

50 Edward James Rapson

selves agree in their upper portions ( = pa), but they are so very differ­ent in their lower portions, that 1t would be surprising to find precisely similar forms derived from each.

Buhler, to whose great work, lndische Palaeographie, we all owe so much, seems to have overlooked this necessary consequence of his statements on page 76. "40 = pta . . . der m1tunter curs1v in ein Kreuz .. verwandelt wird ", 1 and on page 77 "70 = pil . neben dem cursiven Kreuz " 2 The two cursive form& here cited are, as repre­sented m the Plate, practically identical Each 1s what Professor Luders very conveniently calls "a St Andrew's cross." It 1s difficult to bd1eve that ther<' can have been m use in the same script and at the same period forms for 40 and 70 which were so similar as to be almost ind1&tmguishable.

Profr&sor Luders, hm1tmg his observations to tho..,e Brahmi in­scriptions wluch come from the ne1ghborhood of Mathura says (page 244) "There are two symbols, the St Andrew's cross and the pta, one of which must represent 70 and the other 40 " But surely this do<'s not follow a5 a necessary consequencP It I& at least possible that the St Andrew's cross may be, a& I &uppo&e 1t to be, merely a cursive form of the pta

A& a record of the history of Brahmi numerals the corns of the Western K'>atrapas are most mstruchve They supply m with a re­markably continuou& sequen<'e of dat<."s rangmg from 1 [oo] to 31[x]. On them we have numerou'i examples of the numeral 40 durmg the decades 140-149 and 240-249, and of the numeral 70 durmg the de<'ades 170-179 and 270-279, 3 and there is never any poss1b1lity of confusion between the two symbols, as the dates of the rulers who struck the corns ar<' well known Jn my volume of the Urit1&h l\f useum Catalogue of Indian C'oms, Andhra Dynasty, I give the normal form for 40 as 1{, and the normal form for 70 as~ (Introduction, page ccvm). It 1s evi­dent that the 40 &till resembles 'h = pta, and it is evident also that, 1f the two constituent elements of this or any other form of pla the pa and the ha, were made more angular, a figure resemblmg a St. Andrew's cross would be the result In the 70, on the other hand, while the pa of the ~ = pii remains recognizable, the bend of the stem to the left and the horizontal stroke to the right, which together represent ii, have become a loop. And this loop on the coms is characteristic of

1 Taf ix, Col v, A 2 /Ind , Col v11

• Br1t1sh Museum Catalogue of Indian Coins Coins of the Andhra Dynasty, the Western Ksatrapaa, etc, Plates XII, XIII, XIV, XV, XVII

The Date of the A mohini Votive Tablet of M athurii , 51

70. It never appears in 40 or in any other numeral except 70; and on well struck coins it 1s never wanting in 70.

It is quite certain, then, that the St Andrew's cross is never used to represent 70 on the coins. What evidence is there to prove that it is sometimes so used in inscriptions?

After examining minutely the mstances quoted by Professor Luders in support of his view, so far as they can be tested by the photo-lithographs given in the Epigraphia I ndica, I am persuaded that there is no certam evidence.

Professor Ltiders relies, in the first place, on the "Mathura Stone Inscription, dated Samvat 74" which he edited in Ep. Ind., Vol. IX, pp. 241 ff. In note 5 of his remarks on the text he observes (page 242):

"Owing to a flaw in the stone, a small portion of the lower left cross-bar of the symbol has disappeared In the impression somebody has tried to re&tore the m1ssmg portion by addmg in pencil a hook turn­ing upward&, but there is nothing to warrant this restoration. There can be no doubt that the symbol had the shape of a plain St. Andrew's cross, Just as m the other in&criptions."

The photo-lithograph "from a damaged rubbing" unquestionably shows either that there 1s a break m the stone, or that the rubbing has failed, on the lower left of the decimal figure in the date, and that someone m attemptmg to fill the lacuna has traced a portion of the curve which, 1f contmued, would make a loop such as is characteristic of the numeral 70 on the corns But 1s it certam, as Professor Luders seem& to as&ume, that the restorer, whoever he may have been, had no authority for his restoration? L, it not more probable that he felt Justified by traces vmble on the stone but not reproduced m the rub­bing? Unfortunately it appears not to be possible at the present time to decide this pomt from an mspection of the actual mscript10n.

The date of the mscr1ption must undoubtedly, as Professor Luders rightly pomts out (page 244), be Sam.vat 74, smcc the name of the kmg begins with Vasu -, and any date mcludmg the decimal 40 would fall in the reign of Huvi~ka; but I cannot agree that he has "conclusively proved" that 70 1s here represented by the St Andrew's cross. To me 1t seems more reasonable to regard the sign as a mutilated specimen of the looped form which 1s seen very distmctly m the "l\Iathura Inscrip­t10n of Sam.vat 79," 1 and which 1s essentially identical with the un­doubted form for 70 found on the coins of the Western K~atrapas.

1 Ep Ind, vol 11, p 204, no 20 and Plate, Professor LUders (p 245) reads the decimal figure as 40, but I thmk the evidence of the coms is quite dec1s1ve on this pomt

52 Edward James Rapson

Professor Luders quotes seven inscriptions from Mathurii or its neighborhood, in all of which he proposes to regard the St. Andrew's cross as representmg 70, and not 40 as had been supposed previously by himself and other scholars. There is, so far as I can judge, no internal evidence in any of these which will enable us to determine this question beyond poss1bihty of doubt. But, as Professor Ltiders himself ob­serves (page 245), one of the inscriptions (his number 4) "mentions the mahiiriiJa ra3iitiriiJa devaputra Hiiv1ska, but not in connection with the date, the inscription simply recording a gift to the mhiira of that king." Evidently this affords no proof, but there 1s, I think, some probability that the donat10n recorded may have been made to a vihiira recently founded by the reigning sovereign To this extent, even in the absence of other considerations which I consider to be far more cogent, I should rather prefer the readmg 40 which would give a date in the reign of Huv1~ka.

On the whole, therefore, I consider that the case for 70, though pleaded with great learning and much mgenmty, has failed. while I thmk that the forms for 40 and 70 on the corns of the Web tern Ksatrapas afford a safe cr1ter10n by which we may d1btinguish these numerals in the inscr1pt10ns The St. Andrew's cross 1s easily and naturally ex­plamed as a cursive form of 40, that is, of '-I, = pta, or of 1t'l mod1fi­cat10n, 1{, which appearb on the coins I find it hard to beheve that it can possibly be a cursive form of 70, ·when 1t lacks all traces of the stem bendmg to the left with its horizontal stroke on the right which are the d1stmgmbhmg features of the ;i = pii, or of the loop at the base which has taken their place in its num1bmatic representative ~-

Accordingly my faith m the correctness of the reading "year 42" = B.C. 17-16 on the Amohmi inscription is still unshaken Much of the perplexity which scholars have found in their attempts to arrange the chronology of the (;akas and the Kusanas beemb to me to be due to the belief caubed by the reading "year 72," that the Great Satrap Qoc_lasa ruled thirty years later, that is to say, so late as the second decade of the first century A.D.

CAMBRIDGE UNIVERSITY

REMARKS ON A KHAROSTHI INSCRIPTION FROM THE KURRAM VALLEY

BY STEN KONOW

IN the Annual Report of the Archaeological Survey of India 1917-18, pp. 31 f., we read "The only epigraphical discovery made in

the Frontier Circle was an inscribed copper rehc casket, the property of one of the sons of the N awab of Landi. It i~ stated to have come from the Kurram valley and is shaped hke a mimature stupa with harmika and umbrellas, all complete The inscript10n consists of four Imes in the Kharosthi script of the second century A D , punctured on the four sides of the ba~e of the casket, and 1s dated on the 20th day of the month of A~vina, the year bemg doubtful. In the opimon of Mr. V. Nat1sa Aiyar, Archaeological Superintendent, Frontier Circle, the record refers to the enshrmement of relics of the Buddha Sakyamuni in a monument belongmg to the Sarvast1vadms." 1

During my ~tay m Peshawar in l\Iarch, 1925, I examined the photo­graphs of the casket m the museum, which were qmte sufficient for determmmg tlw readmg of the greater portion of the record Some few passages were, however, 1lleg1ble, and as the record seemed to be of some nnportance, I called on the owner, the Arbab Muhammed Ahas Khan, who kmdly allowed me to take the mscr1bed portion of the casket to Tax1la m order to get 1t reproduced. In Tax1la I showed the inscription to my friend Sir John Marshall, and we both went through it together and chequed my prov1s10nal readmg

There are, as has alrC'ady beC'n stated, four mscribed sides, which I designate as A, B, C, D, respectively A and D each contams four, B and C each three Imes of l\.'liaro~thi letters, executed by means of punctured dots, as m several other inscriptions of the Indo-Skythian period. The arrangement of the epigraph is such that l 1 of A 1s fol­lowed by I I of B, and so on From D 1 we have to turn to A", and thence we proceed on till D 111

, after which comes D•v(and finally A•v). The begmnmg of A 1 1s defac<'<l, but Sir John agrees with me that

there cannot be any doubt about the actual readmg There is a hole in the casket at the beginnmg of C 1

, where three or four aksaras have been lost. In other respects the inscription is in an excellent state of preservation.

1 Cf Annual Report of the Archaeological Survey, Fronl!er Circle, 1917-18, p 2, and N G MaJumdar's List of KharosthI Inscr1pl!ons, No 26, Journal of the Asiatic Society of Bengal, N S, xx, 10

54 Sten Konow

The characters are Kharosthl of the Kani~ka period, and only the shape of the letters and their transliteration necessitate some few re­marks. I have explained elsewhere why I render the sign which is usually considered to be the common Kharosthi sign of the aspirated voiceless palatal as ks and not as eh. In the first place it only occurs where the corresponding Sanskrit word presents ks, and in the second place another Kharosthi letter is used where we have to do with an old or derived eh I do not suppose that any epigraphist would think it likely that we should have two letters for one and the same sound, and smce an examination of all the passages where the two signs occur in lndo-Skythian inscriptions and in the Kharo~thi manuscript of the Dhammapada 1 shows that they are always correctly distmgmshed, we have no right to transliterate both m the same way. I look on such words in which the letter corresponding to Sanskrit ks occurs as Sanskr1tisms, that is, as loan-words. For 1t seems to me that the regular development m the North-Western Prakr1t was from ks to kh 2

A peculiarity of the Landi record 1s a slanting vertical above the letter km samkara B 11

, kamdhasa D 111• In both cases k is derived from

sk, cf Skr .<Jamskiira and skandha respectively ·we have very unsatis­factory material for Judging about the treatment of the compounds contammg a k preceded by a s1b1lant m the old North-,Vestcrn Prakr1t. The Dhammapada manm,cr1pt has sakhaca, Skr * satskrtya, 0 3, kan­hana, Skr skandhiiniim, B 13, puskara, 0 21, nikhamadha, Skr. niskriimata, A 11 4, nskhama, Skr naiskramya, B 25 In the inscrip­tion& we have, if we abstract from foreign names such as Kaneskasa in the Mamkyala, Kaniskasa m the Zeda, [Ka]neskasa m the Kaniska casket, Kaneskasya in the Sue V1har and Hoveskasya m the Wardak vase epigraphs, perkarini m the Pathyar and Jukarani in the Kaldarra records Kadhavaro on the Mathura Lion Capital 1s uncertam, be­cause we do not know whether this word represents a Skr skandhiiviira or, as suggested by Professor Ludcrs, contams the noun kantha, town. We get the impression that a sibilant tends to aspirate as following k, but that this aspiration 1s so httle pronounced that it is frequently left unmarked I would therefore transliterate the vertical as h within

1 I quote this important text after the ed1t1on of M Senart, Journal As1abque, x1, xu, 1898, 192 ff , markmg as O the fragment published by S Oldenburg, St Peters­burg, 1897

2 Cf from the Dhammapada manuscript bhikhu, Skr bhikau, B9, &c , pradi­mukhe, Skr pri.iiimokae, B I 7, cakhuma, Sk1r cak,u,,mat, A"' 4 In face of suchforms I thmk that we must consider bhiksave, B 53, bhikaavi A11 5, ksaya A• S, cakauma, A 1i1 8, kairavayo, Skr ksirapaka, B 54, as foreign loans and transliterate the letter m quest10n as ka and not as eh In ordmary speech kh was probably substituted.

Remarks on a Kharosthi Inscription from the Kurram Vall"y 55

parentheses and see in this sign the mark of a slight aspiration. In favour of this transliteration I may add that a similar vertical appar­ently marks an aspiration also m other cases

The sign which is usually transliterated as f in Gudufara in the so-called Takht-i-Bahi inscription and elsewhere consists of an ordinary v with a slanting vertical attached at the upper end to the right, and Professor Lu<lers wiu, probably right m suggesting 1 that it he trans­literated as t•h. The consonantal compound which is usually trans­literated as st has a vertical r1smg from the cro..,..,-bar m the H1dda inscription and m some mstances in the Central Asian Kharo~thi record..,, where Mes..,rs Boyer, Rapson and Senart tran..,hterate slh

I therefore prov1s10nally transliterate samk(h)ara, k(h)amdhasa, respectively, but I w1llmgly admit that thi-, rendermg i.., far from being certam

A similar remark holds good with reference to the curved line found above s in tasa A 111 A snmlar curve is not mfrequently m,ed in the Dhammapada manu..,cr1pt, where old dhy is, to take an instance, some­times represented by Jh, sometmws by J and sometimes by J with a curvt:'d hne above, thus pa<l:wi;hu, Skr prat·widhya, B 29, ;ai, Skr. dhy<iya, B 4, 34, Jana, with the curve above J, Skr. dhyana, B 16. It seems probable that the curve marks a slightly pronounced aspiration, and I therefore transliterate the word m A 111 tas(h)a, suppo..,mg 1t to have been spoken with a E>hghtly m,pirated s.

The readmg of the mscript10n does not present great difficulties. It runs a:. follows.

A• [Sam 20 masa]sa Avadunakasa <l1201fa ki,unamm1 B• Svt:'davarma Yasaputra tanuayamnn raiia mi-C• m1 acaryana Sarva:.t1vadana pari-D• grahamnu thubc1.mn11 bhagravatasa Sakyamumsa A11 ~arira prachthaved1 yatha uta bhagravata B 11 av1Japracagra samk(h)ara :.amk(h)arapracagra v1fiana C•• v1fianapracagra namaruva namaruvapracagra sadaya­D11 [dana] sadayadanapracagra pha<sa. p[h],tsapracagra A 111 vedana vedanaprc1.cagra ta-.(h)a ta<s(h)apracagra uvadana Bill uvadanapracagra bhava bha,·apracagra Jach Jad1pracagra C 111 Jaramarana~ograparidevadukhadormanastauvagrasa D 111 evam m,a kevala..,a dukkak(h)amdhasa .',atnuda:. bhavab D•v sarvasatvana puya1' aya ea patica:.amupada A•v hkhida 1\Iah1phatiena sarvasatvana puyas.

1 JRAS, 1909, pp 655 ff

56 Sten Konow

It will be seen that no cerebral 'fJ, is used in the record, and further that the penultimate in the locative termination ami is always fol­lowed by an anusvara. Thus ksunammi A1, tanuayammi B 1, parigra­hammi C 1

, thubammi D 1• For the latter reason I do not think it likely

that the final three aksaras of B I should be read as one word and ex­plained as the locative of rana, which might represent a Skr. ara7J,ya. I would take rana as the equivalent of Skr. rajna and see in mi the beginnmg of the Rajan's name It 1s, of course, impo~sible to say which letters have stood in the beginning of C 1

, where there is now a hole m the casket. We might think of some such thing as rena katam, so that the whole would he ran'a Mirena katammi, made by the Rajan Mira, and compare the name of the Mira Boyal).a mentioned in the Gudufara mscription. But it is of httle use to make guesses.

With regard to the readmg, I would mention that there are distinct traces of sam and 20 in the begmnmg of A1

, and the readmg of the month 1s absolutely certam The photographs in the Peshawar Museum seemed to pomt to the reading asadasa instead of Sam 20 masasa, but an examination of the origmal has C'onvmced me of the correctness of the reading adopted above I absolutely fail to under­stand how Mr Natisan could read the name of the month Asvina

The reading Svedavarma in the openmg of Bi seems to be certain, but d is probably miswritten for t

The form sarvastivadana, of the Sarvasbvadins, C•, should be com­pared with forms such as sarvastwatana an<l ,mr,,astivatasa 1 on the Mathura Lion Capital.

It will be noticed that the context 1s interrupted m the midst of A 11 and only taken up again in D•v The intervemng space is occupied by a quotat10n from the Buddhist scriptures, viz , the famous formula of contmgent origination. The wr1tmg is consistent throughout, only p has by mistake been written instead of ph in the second phasa Dil. This shows that the orthography is not absolutely reliable, so that Sve<J,avarma Bi may be miswritten for ,qvetavarma and Jf ahiphatiena A1v for Mah1,patiena.

The interpretation of the record presents few difficulties. The era used 1s the so-called Kani~ka reckoning, for which Dr. van W1jk and myself have tried to estabhsh that its imtial point was A.D 134.2 The month Avadunaka is, of course, the Macedonian Auovva'io5, correspond­ing roughly to December, which has not formerly been met with in

1 The Sarvii.sbvii.dm Budh1ta 1s said to be khaiula for the purpose of teachmg the foremost Mahasiiqighikas prama, read prama fiavllave as two words

2 Acta Or1entaha, 111, 54 and ff

&marks on a Kharosthi Inscription from the Kurram Vallty 57

Kharo~thI records. Other Macedonian months occurring in Indian inscriptions are Ila.11Eµos in the Patika plate, ~aLu,Kos in the Sue Vihar inscription and 'ApTEµLuios in the Wardak Vase and the Lahore box lid epigraphs.

The only word which is not immediately intelligible is tanuayammi B 1• There can be little doubt that 1t is the same word which occurs as tanuvae in the Taxila silver scroll inscription of Sarp 136. Sir John Marshall 1 explained tanuvae as the name of a locality, and M. Boyer 2

saw in it a form corresponding to Skr. tanut'yaya, while I 3 interpreted it as the genitive of a female name Tanut 1a The new inscription shows that none of these explanations is likely to be correct. We do not know from which locality the Landi ca1>ket originally came, but it cannot have been the Bodhisattvagrha excavated by Sir John. An original tanuvyaya would become tanut•vaya, and the v could not have been dropped And, finally, tamwas mu1>t be the locative singular, and not the genitive of a female base, as proved by tanua,yammt It seems necessary to explam ianumya, ianuaya a1, adJect1ves quahfymg the locality where rehcs were deposited, and I therefore thmk that the word is derived from tanu and means "contammg a tanu, a relic." I therefore tran1,late the record as follows:

Anno, 20, the 20. dav of the month A va<lunaka, at this instant Svegavarman, the son of

0

Yasa, deposits a relic of the Adorable Sakya­mum in the rehc stupa ( erected by Kmg 1\1. . ) , the property of the Sarvastivada teachers - a1, it has been said by the Adorable one· contmgent on ignorance (are) the forces, contmgent on the forces perception, contmgent on perception name and form, contingent on name and form the six sense1>, contmgent on the six senses contact, contingent on contact sensation, contmgent on sensation thirst, con­tingent on thirst grasping, contingent on grasping existence, contmgent on existence birth, contingent on birth age, death, sorrow, lamentation, misery, downcastness and despair Such is the origm of this entire mass of misery - m honour of all being~ And this contmgent origi­nation has been written by Mahiphatika in honour of all beings

The record is of interest in more than one respect It falls withm the reign of Kani~ka, but we are not informed about the identity of the local ruler who held sway in the locality where the Stiipa was situated, nor about the position of this locality. The historical and geographical information which can be gleaned from the record is accordmgly small.

1 JRAS, 1914, 973 ff. a Ep. Ind., xiv, !l88

2 JA' XI, 1915, !l81 ff

58 Sten Konow

We learn a httle more about Buddhist literature in North-Western India about the middle of the second century A D. We knew before now that a version of the Dhammapada in a vernacular belonging to that part of the country existed ma comparatively early time. Now the Landi mscripbon brmgs what is evidently a quotation from an­other canomcal work written m practically the same language Have we a right to infer that extensive canonical texts in that form of speech were in exii,tence? And with which Buddhist sect 1,hould they be connected? From the mention of the Sarvastivadmi, m this as in other Kharo~thi records, such as th<' l\Iathura Lion Capital and the Shah-Ji ki Dheri casket, we might be mclmed to thmk of them. But then we know that their canon was later written in Sanskrit It may however be pos&1ble that thq had, at the same or at an earlier period, also canomcal books m the North-Western Prakrit. The Landi m­scr1ption raises tlus question afre&h

Its chief interest, however, rei,ts with the fact that 1t is compara­tively long and thus allows m, to Judge about some features in the dialect.

We have every reason for assuming the existence of a fairly um­form language spoken m North-We&tern India m the lndo-Skythian period It 1s represented by a number of Kharo-.thi mscr1ption& and also by the Kharosthi manuscript of the Dhammapada Nobody has as yet attempted to give an exhaustive :-ketch of thi& form of 1,peech, and 1t would not be pos1,1ble to do &o w1thm the limited &pace of a con­tribution to a complimentary volume But I 1,hould hke to off er &ome remark& about &uch details which are further elucidated by the new inscription

I said that the language 11, fairly umform. That does not, however, exclude the existence of mmor d1alect1c variat10ns, and a& a matter of fact such appear to exii,t both m inflex10nal forms and m phonetical feature&.

To the first cla&s belongs the nommative 1,mgular of masculme a-bases. In the Dhammapada manu&cr1pt 1t regularly endi, mu, which o 1s shown by the metre to be long If we turn to Kharo~thi m&crip­t1on1, we find the same o-form m m1,cr1ptions hailmg from the country to the east of the lndus, and ali,o m those found about Mount Banj (Dewa1, Kala Sang, Mount BanJ, Shakardarra, Yakub1) In the West,

,on the other hand, m Ara, Zeda, Machai, PanJtar, PaJa, J>eshawar, Charsadda, Jamalgarh1, Loriyan Tangai, B1maran, Hidda, we find e to be the termmat10n. In the W ardak vase inscription we find yo amdajo, which might seem to point to o-forms, but the words are probably

Remarks on a Kharo~thi Inscriptionfrom the Kurram Valley 59

neuter and not masculine, and a priori we should expect Khawat, where the vase was found, to belong to the e-group. The Landi inscription is also of the western variety, cf. samadae, that is, samadayah, D 111

•

The new record contaIDs qmte a series of locabves ending ID ammi · ksunammi A•, tannayammi B•, parigrahammi C•, thubammi D•. Similar forms are found ID inscriptions hailIDg from Hidda, Wardak, Ara, Ohind, Panjtar, and also from Taxila, where the silver scroll of Sam 136 has gahami In the Patika plate, on the other hand, we apparently have imasi in I. 5. This latter form is qmte common in the Dham­mapada manuscript, where we find forms such as a.wi 1 /,oki parasa yi, in this world and the other one The form a~vi ID collocation with parasa, i e , as shown by the metres parasS1,, showb that these forms are derived from such endIDg ID a.Ymi, Just as is the cast' with the ammi forms The manuscript shows that sm qmte regularly became sv and further 8.'f, thus svad1 ,2 Skr smrti A IV 2, Cro 42' tasa, Skr tasmat, yasa, Skr yasrnat, 0 16, sadana, Skr smrtanam, Cr0 43, etc The change of sm to mh ID samhasadi B 13 seems to belong to a different stratum

Now the locative forms in mi, or rather, as shown by the writing mT'I ID the War<lak Vase IDscription, mhi, prebuppose this very change, and one would be indIDed to thIDk that we are here agam faced with dialect varieties w1thm the North-,Yestern vernacular There are, however, some considerations which seem to make buch an assumption unlikely. The locative ID asi is apparently used in the TaXIla plate of Patika of Sam 78, while the termIDabon ami is found in the silver scroll of Sarp 136 coming from the same locality :Moreover forms such as ksunammi are of frequent occurrence in the dated Kharost}ii records from N1ya ,,e cannot, accordingly, find any geographical delimitations of the bpheres of each form On the other hand 1t is difficult to thmk that one and the same termmat10n asmi can sometimes become amhi and sometimes asS'I withm the same period and in the same dialect. Moreover the change of asmi through ast't to assi seems to be in best agreement with the phonetical sybtem of the North­Western tongue. We must therefore, I think, assume that the amhi forms are a later development of new asmi forms, reintroduced in the barbaric Sanskrit of another sect, at a time when S?n no more became av, ss, but mh.

1 M Senart reads asmi, but the sign which means sv m svaga, 1 e , svarga, 0 5, cannot be transliterated otherwise than sv

2 For the reason stated m the precedmg footnote I cannot accept M Senart's readmg miati

60 Sten Konow

There is also another feature where we are apparently justified in speaking of a phonetic development within the dialect. There has been, as is well known, some doubt about the distribution of the dental and the cerebral n in Kharo~thi records. In the Asoka inscriptions the two sounds are clearly distinguibhed, and also in later inscriptions the late Professor Buhler never confounded them in his transliteration. Several scholars, however, have been accustomed to transliterate both the n-letters as n, and M. Senart in his edition of the Dhammapada manuscript and :Messrs. Boyer, Rapson and Senart in their Niya in­scriptions use the dental n throughout. In my paper on the Kharol?thi manuscript m the Festschr1ft Wmd1sch, pp. 85 ff., I showed that n and n are, m the Dhammapada text, distmguibhed accordmg to a defimte rule, which 1s also traceable in a series of modern dialects every initial n and every doubled n remams as a dental, but every smgle un­compound n becomes n between vowels.

This seems to be an old characteristic of the dialect, and it can be traced m some Kharo~thi mscr1ptions. We find the two n-sounds cor­rectly, or approximately correctly d1stmgmshed in the Patika plate, where the except10ns to the rule, the words Panema and nwhadanapati, can be explamed ab loan-words m the Marnkyala inscription of Sa111 18 1 and the Shakardarra epigraph of Sa111 40 Rut m mobt cases the rule 1s not observed The Mathura L10n Capital, the H1dda Inbcriptlon and the record on the Kaniska casket use the dental n everywhere, both as m1bal and between vowels, Just as the Landi casket. The case seems to be similar m the Ohind, Yakub1, Palatu Dheri Jars, and Jamalgarhi stone inbcripborn,, where no m1tial n occurs

On the other hand the Taxila silver bcroll, the Dnnarva and the W ardak Vase records use the cerebral n both as an mitial and as medial, and the same letter 1s also m exclusive use in several epigraphs where there is no inbtance of an mitial n those from Tirath, Pa3a, Takht-1-Bah1, Kaldarra, Pan3tar, Dewai, Zeda, Macha1, Kala Sang, Fatch Jang, Skarah Dheri, and others In the Ara mscr1ption the dental n occurs only in the proper name Kanisl.:a.

Fmally n and n seem to be used promiscuously in the same word dana. In the records from Charsadda, Lor1yan Tangai and Janha, and also m the Kharosthi records from Niya and neighboring oases the distinction between the two sounds seems to have been largely dis­carded

1 The form dadanayago can owe its n to the uncompound nayago, apanage can, m a similar way, be explamed as alpaniige, and etranana may be one word, the gemllve plural of an adJeclive etrana, formed with the suffix iina from etra and havmg the same meanmg as Skr ataatya

Remarks on a Kharo~V,,i lnscriptwn from the Kurram Valley 61

From this state of things we can draw only one inference: the dif­ference between the two sounds was felt no more by those who drafted the inscriptions. If we now bear in mind that the oldest record where they are confounded, the Mathura Lion Capital, 1s a Saka epigraph, and that the language of the Sakas or their Iranian cousins in Chinese Turkistan did not distingmsh n and n as m the manuscript, we must, I think, infer that the phonetics of the North-Western dialect became modified in the mouth of the Indo-Skythians.

In such circumstances we might ask ourselves whether it would not be just as well to follow those eminent scholars who do not make any distinction in their transliteration of the two Kharosthi letters I do not think, however, that such a procedure is justtli.ed It would come to a suppression of evidence, and the history of the Dhammapada manuscript show show careful we must be m venturing on such un­dertakmgs.

With regard to the treatment of intervocalic stops some inference can be drawn from the forms occurrmg in the Landi rnscr1ption, which partly corroborate, partly modify the results derived from the Dham­mapada manuscript

In the latter text an mtervocahc k sometime~ remams, is sometimes dropped and bOmetimes replaced by y Thus nwyaka, Skr. mocaka, B 31, savaka, Skr sriivaka, A 1

" 4 ff., u1uo, Skr rjuka, A 1" l, muya­

madia, Skr 1nrgamiitrkii A• 6, sagaraU<J.asa, Skr. samkiirakiite, Cro 3, ksiravayo, Skr. ksirapiika, B 54.

It 1s difficult to think that all these writmgs correctly render the actual sound, and though the 1,urrounding sounds may have something to do with the different representations, and though Sanskrit and other Prakrits may have exercised a certain influence, we should expect to find a fairly consistent treatment and be inclmed to see in the different resulting letters various attempts at rendering a sound which was not covered by any of the usual aksaras of the alphabet.

It may be of mterest to compare the treatment of mtervocalic g

I in the manuscript, because we know the general Prakrit tendency to soften intervocalic surds I abstract from a form such as baho1agaru, Skr. bahujiigara, A 111 15, because the g of this word has also been pre-served in other Prakrits. If we do so, 1t will be seen that the rule is that g in such posit10n is written k Thus nadakara, 1 Skr na</.agiira, A 11 4; parakada, Skr piiragata, B 4, raka, Skr. raga, B 35; urako, Skr. uraga, B 41 ff.; vikaya, Skr. vigiihya, B 42 In the word muyamadia,

1 M Senart reads nalagara, but I have no doubt about the correctness of my own reading

62 Sten Konow

Skr. mr_gamatrk<i, quoted above from A 1 6 we have the same change to y as in the case of mtervocalic k. In other words, intervocalic k and g are both treated in the same way.

If we now further compare forms such as idria, Skr. indri_ya, B 17; svihao,1 Skr. sprhayan, B 20, dhoreka, Skr. dhaweya, Cv0 Sl, udakavaya, Skr. udayavyaya, B 13, C•0 18, it will be seen that ancient intervocalic y seems to be confounded with intervocalic le 2 We must, I think, mfer that the sound in question was a spirant with a slightly pronounced guttural timbre

It is evidently the same sound which is intended by the writings kr m bhakravato, Skr bhagavatah; nakraraasa, Skr. na,qarakasya; samanumotakra, Skr samanumodaka, on the Mathura Capital, and gr in bhagravada, bhagravatra, Skr bhagavatah, on the Bimaran Vase; gadigrena, Skr ghatikena; kadalay1gra, Skr krtalayika, nati,qramitra­sambhatigra na, Skr jnatikamitrrsambhakt1k<inam, etc, on the Wardak Vase.

The Landi casket furnishes several instances of the treatment of intervocalic k and g 3 The state of thmgs i:-. evidently exactly the same as in the manuscript, except that the writing differs, gr having taken the place of k The new inscription also show~ that mtervocahc y was pronounced in a similar way, Just as was the ca~e m the manm,cr1pt. Thus we find pracagra, Skr pratyaya, B 11 ff ; umgrasa, Skr upay<isa C 111

, both in quotations from the Canon. It is evident that we are faced with a common feature m the enunciation of the North-Western language, at least of the form which had become used in Buddhist scriptures

With regard to stops we may still note the treatment of intervocalic p. It becomes v m namaruva, Skr niimariipa C 11

, Just a~ in other Pra­krits. In thubammi, Skr. stiipe, D•, however, 1t is replaced by b. We might dismiss thi~ form as a ~1mple miswritmg, 1f it were not for the fact that we also find it elsewhere, v1z m the Hidda, Wardak and a Taxila copper-plate inscription, while the Mathura Lion Capital, the Mamkyala stone and the S1lula Vase have regular forms with v instead. It is therefore evident that there was a dialect from thuba in addition to thuva.

The form ruva shows that the different treatment cannot be due to

1 M Senart's readmg amihao cannot be upheld 2 Cf also babaka, Skr biUba1a, cvo 81, where the Pah pabba1a 1s due to m1sunder­

standmg of an old babba3a 3 Cf afJadunakasa A1, Mahiphatuma A•v, wgra, Skr soka, C 111, bhagrafJala A11;

bhagravatasa D 1

Remarks on a Kharosthi Inscription from the Kurram Valley 63

a difference in the surrounding sounds, because these are identical in rilpa and ,<ttilpa. On the other hand b cannot well be explained as representing v, because the two sounds are carefully distmguished in Kharo~thi

I cannot see more than one explanation of the form thuba: it repre­sents thumba and 1s derived either from thumva or from thumpa, with the well-known substitution of a nasalization for a long vowel The only question is whether the form ha& been derived from an older thilva or a thilpa. Both explanations seem a prwri to be possible

We know that mv occasionally becomes mb m the North-Western dialect. Thus we find ,<Jambatsare, sambatsarae, for samvat1,are in the Takht-i-Bahi, Ara and Hidda epigraphs, but sammtsare m Sue V1har, samvatsarays in the Patika and Pa1a 1m,criptions Similarly va, Skr. iva, regularly bPcomes ba after an anusvara m the Dhammapada man­uscript. Thus siha ba, Skr. simham wa A1 6, nadakara ba, Skr. na<J,a­gdram ·wa, A 11 4 1 On the other hand mv becomes t', that is, probably a doubled nasalized v in forms such as savcwo, Skr. samvasa Cv0 37; savasi, Skr samvaset, A 11 2.

If we examine the possib1hty of deriving thuba, that is, thumba from thilpa and not from thu,,a, we apparently have full justification for doing so in a promment feature of the language of the Dhammapada manuscript, m which a nasal produces certam modifications in a fol­lowing stop. The general rule 1s that voiceless stops become voiced, while vmced ones more or less coalesce with the nasal. The details are, as follows.

nk appears as g; thus alagido, Skr alamkrtah, B 39, sagapa, Skr. samkmpa, A 11 5, etc In the latter passage the metre shows that both the first and the second syllables are long Sagapa accordmgly stands for samgappa, 2 in other words nk becomes ng. In the same way mkh becomes gh, that is, mgh, thus saghai, Skr. samkhyaya, B 27, saghara, Skr. samskdrah, A 111 lff.; ta gu, Skr. tam khalu, B 21,

1 It 1s evident that such features are of great rmportance for the mterpretabon of the text Thus 1t 1s 1mposs1ble to explam avalasa va A lit 15 as representmg abafuivam iva, though the Pii.ti text has abalassam va The form avala§a must be the accusative plural, and the passage shows that the common source was wntten Ill a language where the accusative smgular could not easily be d1stmgmshed from the same form of the plural m masculme a-bases, that 1s, the latter form ended m a nasalized a, as seems to have been the case already m V ed1c dialects Cf W ackernagel, Altmd1sche Gram­mabk I, (§ !?79, ba)

2 Here as m numerous other mstances the metre shows that Sir George Grierson was not right m doubtmg the existence of doubled consonants m the North-Western Praknt. See his paper m the JRAS, 1918, pp 141 ff.

64 Sten Konow

ng appears as an ordinary g, that is, as mg, in sagamu, Skr. Ba1!1,­grama~, C•0 6, where the r after g accounts for the preservation of the g. Also in kadigaru, Pali kalingaro, Cvo 14, the ordinary g is used, but here ng seems to be derived from nk; cf. kadankara, Pa:r;iini V. i. 69. In all other instances the letter g has been modified. In saga, Skr. sanga, B 3, 37, and in astagachadi, Skr. asta,,,,,gacchanti, 0 14, it is provided with a hook at the bottom, and m athagw, Skr a1tangika~, A111 4, there is a hook above the g. The result of old ng cannot, ac­cordingly, be an ordinary mg. Similar hooks are, as we have already seen, used to mark an aspiration or a spirantic pronunciation, and we are therefore probably justified in inferrmg that a voiced guttural be­came spiranbc after a nasal, the more so because ngh is treated in the same way. Thus we find sa_qha, that is, samgha, with a curve above g in A •v 6, and saghi, with a hook to the right of the bottom in Cro 13;

./ nc becomes J, that is, iiJ, and nJ ii, that is, nn, thus paja, Skr. panca, B 37, lei Ji, Skr kup,cid, 0 15, kuiiaru, Skr. kun1arah, A 1 5, 11 4;

nd becomes n, that is, nn; thus dann, Skr danda, B 39, 0 18, 28; panidu, A 111 16, where the metre proves the first syllable to be long;

nt becomes d, that is, nd, and nd n, that Is, nn; thus anadara, with long syllable before d, Skr. anantara, 0 13; miyadi, Skr mriyante A 111 12, nwinadi, Pah nibbindati, A 111 l, china, Skr chinda, B 37. Similarly ndh appears as nh, thus bamha, Skr. bandha, 0 27; kanhana, Skr. skandhiinam, B 13,

mp becomes b; that Is, mb, mb m, that is, mm, and mbh bh, that is, mbh, or mh, thus sabasu, Skr sampasyan, Cv0 26, anuabisa, Skr. anukampinah, C•0 16, avaramu, Skr. apalambah, A•v 2, udumaresu, Skr. udumbaresu, B 40, sabhamu, Skr sambhava, A• 2f.,017, 1 gam­hira, Skr. gambhira, B 6. We would a priori be inclined to consider the latter orthograph the more correct one.

Now we naturally expect to find similar rules prevailing in the Kharo~thi mscriptions of the Indo-Skythian period, the more so because M. Jules Bloch has been able to point to a parallel development in modern North-Western vernaculars 2 An old thupa might accordmgly become thumpa and further thumba. In order to Judge between the two possibilities it will, however, be necessary to examine the state of things m the North-Western mscriptions.

The materials for such an investigation are rather meagre, but, so far as I can see, decisive.

1 The latter passage shows that M Senart's explanation of the word as representing Skr aambhrama cannot be accepted

1 JA., x, xix, 1912, 881 ff.

Remarks on a Kharo1thi Inscriptwn from the Kurram Valley 65

We have already seen that Skr. samskara appears as saghara, that is, samghiira, in the manuscript, a form which shows that the modifica­tion of the guttural after a nasal is of more recent date than the de­velopment of the earlier sk to kh. The same word occurs, as we have seen, as samkara, with a vertical above k, in the Landi inscription. There does not, accordingly, here appear to be any trace of a softening after the nasal.

The only other instance of a guttural preceded by a nasal in the in­scriptions is the word samgha and its derivatives, where everything points to the conclusion that ngh was pronounced as m Sanskrit.1

If we turn to palatals, we find pamcame m the Patika plate; pam­[cada]se in the Paja inscription, Rajula on the Mathura Lion Capital.

Of cerebrals we have dadanayago in the Ma~1kyala inscription; a[m]daJo and avasadigana, Skr iipiisiin</:ikiiniim, on the Wardak Vase.

With reference to dentals we may compare puyayamto in the Patika plate; a[m]tara on the Wardak Vase, Balanamdi in the Sue V1har plate, Dharmanadisa m Jauh!, k(h)amdhana Landi D 111 , and kadh­avaro on the Mathura Lion Capital, whether this form is derived from Skr. skandhiiviira or contams the word kantha, a town. The form Mumjanamda side by side with Jfujavada on the B1maran Vase is doubtful, the latter form bemg probably the correct one.

Of labials we have nirvanasambharas in the Hidda and sambhati­grana in the Wardak Vase mscription

It will be seen that nothing seems to pomt to a state of things similar to that of tht> manuscript. We must therefore abandon the explanation of thuba as representmg thU1nba from thumpa, thiipa, and derive the form through thum1!a from thiiva.

The precedmg exammat10n will have shown that the North­Western dialect 1s not absolutely umform, and that the Dhammapada manuscript stands alone m its treatment of stops after nasals. The existence of corresponding features m modern North- 'Western ver­naculars prevents us from simply ascribing the state of thmgs m the manuscript to the mfluence of the old lraman tongue of Chinese Turkistan, where we find identical rules preva1lmg But it is possible that this mfluence has been mstrumental in regulatmg a tendency inherent in the dialect. In connection with the nominative in o and perhaps also the locative m assi such features will some day enable us to localize the Dhammapada dialect with greater prec1s1on within the territory once occupied by the North-Western Prakrit.

1 Cf. 8amgharama m the Patika plate and on the Karu~ka casket, aagha,a, 8agha­nama and maha8aghwna on the Mathurii. Capital, etc

66 Sten Konow

Such minor dialectic d1ff erences which may be traced in our ma­terials are, however, of minor importance, and in one important feature all North-Western varieties agree and make the language stand out as a clearly defined md1v1dual Prakrit. in the treatment of old sibilants. The three s-sounds are d1stmgmshed throughout, and every attempt at interpreting a Kharosthi record which does not pay due attention to this fact 1s bound to be a failure 1

In the manuscript we have some mistakes, owing to the fact that it is translated from a different dialect and probably written in a locality where the home tongue was an lraman form of speech. Thus we find sa for sa in pacha sa, Skr pascat sa Cr0 39 (but correctly pacha SU A 11 3), where the Iranian home tongue has sa, and saga for sanga B 3, 37, Fragm Cxxxvinv 0

, where I cannot give any satisfactory ex­planat10n. Similarly we alway& find suyi m&tead of §uyi for Skr §uci, where the origmal cannot have distmgui&hed between the different s-sounds, so that the translator was led astray and connected suci with the prefix su. That the origmal did not d1stmguu,h the dental from the palatal s is evident from the etymological play in samairya §ramano di vucadi, 0 16, which 1s meanmgle&s in a dialect which dis­tinguishes s and §.

The Landi mscription again shows that the three s-sounds are carefully distinguished. 2 It also corroborates certain inferences about the changes undergone by certain compound& containmg a sibilant, which can be drawn from the manuscript and other mscriptions.

There are no instances in the record of the compound fr, but we know from other sources that 1t became s or rather ss, thus samerw, Skr. §ramanah, B 39, and similar forms m several other passages and in mscriptions. Writings such as Sravanasa in the PaJa, Kaldarra and PanJtar records are Sanskritisms. The corresponding change of r§ to §, ss, on the other hand, which we know from forms such as phusamu Skr. sprsamah, B 25; phasai, Skr. sprsati, A 111 10, etc., 1s illustrated by the form phasa, Skr spar §a, D 11

•

Of compounds contaming an old s we have perhaps the well-known change of sy to § in i§a A', if thi& common form contains the pronominal base i and the element sya. S1J, becomes sin the manuscript. 3 I have

1 E g , the attempt made by some emment scholars at explammg the akaaraa which I read ya yetiga m I 5 of the W ardak mscr1ption as aaaetiga and further as aam­aveda3a Such a development 1s entirely agamst the dialect

2 Cf Sakyamuniaa aarira D 1 f , aadayadana C 11 f , aarvartivadana C 1, etc. 3 Cf taaa, Skr tr~tid, B 45.

&marks on a Kharo~thi Inscription from the Kurram Valley 67

already mentioned the corresponding tas(h)a in the Landi inscription, A111, which seems to show that the resulting ~ was slightly aspirated.

Of s-compounds we have skin samk(h)ara B11 ; k(h)amdhasa D111;

st in sarvartivadana C 1; thubammi D', pradithavedi A11

; sy in masasa avadunakasa Ai, etc The form dormanasta m C111 cannot, accordingly, be derived from daurmanasya but must contain the suffix id.

On the whole the Landi inscription 1s of considerable importance for our knowledge of the North-Western Prakr1t and also of the history of the Buddhist Canon, and a prehmmary account may, I hope, interest my d1stmguished American colleague and friend, to whom this volume is dedicated.

UmvERSITY or Osw

DAS VIPANAM IM RIGVEDA

BY KARL F GELDNER

ES IST ein allgemeiner ?Iau be_ der lnder, class gewisse Vogel - es werden der Kruiic (eme Re1herart) und besonders der Ha:qisa

(Schwan und indische Gans) genannt -die Geschickhchke1t besitzen, aus einer Mischung von Wasser und Milch die Milch allem als das Bessere herauszuschlurfen und das Wassn ubrig zu lassen. Diese Fabel, denn etwas anderes ist es nicht, obwohl 1rgendeine reale Beo­bachtung zu Grunde liegen mag, d1entc Dichtern und Philosophen oftmals als Gle1chnis In ernem ebenso feinsrnmgen wie gelehrten Aufsatz im Journal of the American Oriental Society xix, 2, 151-158 sucht der Jubilar, Profes~or Lanman, d1ese Fabel auf em naturhches Faktum zurtickzuftlhren, und er verfolgt 1hre Spur durch die ganze Literatur ruckwarts bis m die Ze1t der ve<l1schen Schr1ften Nur vor der allerletztm Pforte, vor dem Rigwda macht er Halt und verwe1st auf die Autor1tat Bloomfields

In der alten L1teratur wird diese fabelhafte Kunst, Milch aus dem Wasser herauszutrmken. <lurch t•i-pa und t"lpcina ausgedruckt. vi-pa bedeutet zunachst nut Unterschied trinken, 1m Trinken einen Un­tersch1e<l herausfinden, so A1t Br. 3, 29, 5, oder die E~senz von etwas trinken, wie RV 3, 53, 10 ,,f p1badht1am kulrzkcih somyam madhu Dort geht 1m ersten Stollen der Vergleich m1t dt>n Ha msas voraus und de r­s e lb e Vergleich schwebte vermuthch auch im letzen Stol­len be1 r-i pibadhmm dem D1chter noch vor

In 7, 22, 4 wird das~elbe Verb figurhch vorn Presstem gebraucht, der Somapflanze und Somasaft sche1det, letzteren gle1chsam aus der Pflanze heraustrinkt lTnd darnach ist auch 4. 16, 3 zu verstehen. Die wicht1gste Stelle 1st aber 10, 131, 4-5

yut•am surdmam ast•ina namuca,, asure S(l('Cl I vipipanh subhas pati fndram k<irma.<tt_, avatam I I putram iva pitarifo asvinobhendrat•<ithuh krwyair damsanabhih I ytiJ, surdmam ,,y <ipibah s<icibhih sar<u1vati frci maghamnn abhisnak I I

Es steht fest, das diese be1den Strophen in enger Beziehung zur soge­nannten Sautrama:r;ii-Fe1er stehen, bei der statt Soma v1elmehr die Sura, d. h. der Branntwein eine Rolle sp1elt Beide Strophen haben nach Asv. Ss. S, 9, 3 in dem genannten Ritual ihre feste Stelle. Damit ist aber nicht gesagt, dass sie schon von vornherein fur d1eses gedichtet

70 Karl Frwdrich Geldner

seien, wie Oldenberg annimmt. Sie sind vielmehr aus dem RV. in das spatere Ceremoniell ubernommen worden, we1l sie auf eine alte Sage anspielen, an der sich Jenes Ceremoniell anrankt. In die~er Sage kommt eben der Sura eme besondcrc Bedeutung zu. Durch die Anleh­nung an das RVLied hat jenes Opfer von dem dort in Str 6 genannten indrah sutrdmii seinen Namen erhalten Im RV. aber steht das Stro­phenpaar m andercm Zusammenhang Das Lied hegmnt m1t der Bitte um Schutz gegen Feinde (1) und um Verteilung von dcren Besitz (2). Dazu muss man lndra zum Freund haben (3) - dicse Freundschaft des lndra wird A V 3, 3, 2 m Verbmdung m1t der Sautramani crwahnt - wie lndra selbst m <lem N amuci-Han<lcl an <ler }'reun<l1-,chaft der Asvm und der Sarasvati emen Halt fand (4-5)

Die Sage, welche die R1tualtc>xte zur Begrundung des Sautramani­Opfers m1tkilcn, ist <ler Kampf zwischen Gott Indra und dem Damon Namuci Die ausfuhrhche Dan,tellung d1eses Mythm, m Sat Br. 12, 7, lfg. verknupft die Namuc1sage ummttelbar m1t dem ebenso be­kannten Konfl1kt dn be1den Gotter lndra und Tva~tar, nach der Er­mordung d('s Tvastan,ohnes V1svariipa <lurch lndra Ob d1ese Verk­nupfung alt 1st, mag dahing('stellt ble1ben Tvastar brm,k1ert lndra, d1eser begeht m Erwidnung em Sakrilegmm und busst zur Strafe seme unbezwmghche Kraft em Gerade m d1e5e Ze1t semer Schwache fallen seme Handel m1t Namuc1 Der Damon rnerkt, dm, lndra s1ch noch mcht nholt hat (Sat Br 12, 7, 1, 10) und beschhe1-,st dm lndra Jetzt ganz zu Fall zu hrmgen, indem er <lurch Branntwem r,eme Krafte und semen Somatrunk wegmmmt (1b 12, 7, 1, 10, l':l, 7, ~3. I) lndra erleidet einen volhgen Zusammenhruch und unterhegt Im Rmgkampf gegen Namuci Er mur,r, 1,1ch mfolge des Betrugs zu Jt.'lll'lll Entwaff­nung1-,-und Sicherhe1tspakt verstehen, der auch m TBr 1, 7, 1, 6fg. genau erzahlt wird Die Gotter wenden r,ICh an die A~vm als Arzte und die Sarar,vati als das He1lm1ttel mit der B1tte, ihn zu heilen (Bat. Br 12, 7, 1, 11) und 1hm die verlorene Kraft und den Somatrunk wiederzubrmgen (1h 12, 7, 3, 1) Be1 dIC·sem Punkt r,chemen nun die oben z1tierten Strophen des RV cmzust•tzen

Das Sat. Br crzahlt 12, 7, ~3, 3fg den we1teren Verlauf der Ge­schichte, den Ersatz des 1m Vertrag verbotenen VaJra lndras <lurch Schaum. Indra schlagt darauf dem Namuc1 das Haupt ah. Dadurch wurde und blieb der von Namuci weggetrunkene Soma m1t Blut ver­m1&cht. Die Gotter (lndra, die Asvm, Sarasvati) ekelten s1ch davor und erfanden das andhasor vipiinam, das Sondertrinken der be1den Fluss1gkeiten, und machten auf d1ese Weise den Soma wieder rein und schmackhaft.

Das V ipanam im Rigveda 71

Das ganze Material der N amucisage hat in unerreichbarer Grund­Iichkeit Bloomfield im JAOS. 15, 146fg zusammengestellt und kritisch beleuchtet. Ich pflichte Bloomfield darin durchaus bei, class das Schlag­wort suray<i der Brahma1a1astelle (Sat Br. 12, 7, l, 10, vgl. 12, 7, 3, 1) nicht zu Ubersetzen ist: zusammen mit dem Branntwein, sondern: mit Hilfe des Branntweins. N amuci macht den Indra durch dieses 1hm ungewohnte Getrank total betrunken

Wenn dann weiter (a o o. 153) Bloomfield den Satz aufstellt, dass die Gesamthc1t der vedischen Bucher eme EI n he It bildet und stets die Brahma1a1as und Sutras sow1e das gesammte rituelle Spruchmaterial fur die Erklarung der vedischen Hymnen heranzuziehen ist, so wird Jeder Verstand1ge d1esen Satz 1m allgemeinen btlhgen Bloomfields Arbeiten sind m dieser Hmsicht geradezu vorbildlich. Aber auf mytho­logischem Geb1et darf dieser Satz doch nur cum grano sahs gelten. Die mytluschen Erzahlungen der Brahmanas smd Ep1goncnarbe1t, die zwar me1st noch <len Kern des Mythos richtig erfasst, im Emzelnen aber vieles m1s1,deutet oder umdeutet. Der R1gveda steht sagenge­schichthch betrachtet auf alterer und hoherer Stufe. Die Gotterge­sch1chten lebten in dem Gedachtnis der Dichter noch gestalt-und gehalt­voller als be1 den Ep1gonen Nur schade, dass diese aus 1hrer inbmeren Kenntms so sparhche und abgerl.5s<'ne M1tte1lungen machen. Die Namuc1sage des RV ist em we1teres Be1spiel h1erfur.

Bloomfield verlegt nun Rv. 10, 131, 4 m den Schlussakt der Namu­cigeschichte, wobe1 er 1,ich ganz an den Gang der Erzahlung im Brah­ma1a1a anschhesst. Deren einzelne Phasen sind; 1 Surarausch des Indra und Verlust seiner Starke; 2. Pakt mit Namuci Wiederherstellung Indras durch die Asvin und Sarasvati, und Umgehung des Vertrags und Totung des Namuci, 3 Zuruckgewinnung des weggenommenen Soma und <lessen Remigung durch die Sarasvati Bloomfield kann sich dabei besonders auf l\Iahidharas kurze Analyse der Sage in <lessen Kommentar zu VS 10, 33 berufen.

Die hauptsachliche crux interpretum ble1bt das nur in den beiden RV. Strophen belegbare Wort suramam. Bloomfield nimmt Roths spatere Deutung 1m ktlrzeren Petersburger WB. an. Nach Roth ist surama Kompositum aus surd + ama = Surakrankheit, die iiblen Folgen des zu starken Branntweingenusses. Und zwar bezieht BI. diesen surama in der emen Str. au£ den ersten Akt der Namucisage, in der anderen aber auf deren Schlussakt, und ubersetzt (a. a. o. 148):

"Ihr beiden Asvin habt, indem ihr bei Asura Namuci euch selbst eine Sura-Uberladung antrankt, dem Indra bei seinen Taten geholfen, ihr Herren des Lichtes."

72 Karl Friedrich Ge'ldner

"Wie Eltern dem Kind so haben beide Asvin, o Indra, dir mit ihren Zauberklmsten geholfen. Als du, o Maghavan, mit aller Macht Sura bis zur Erkrankung getrunken hattest, da kurierete dich Sarasvatt."

N ach Bloomfield handelt es sich eigentlich um zwei verschiedene Vorgange. lndra trinkt sich unabsichtlich bei Namuci einen bbsen Rausch an in Sura, die ihm sem Trinkgenosse Namuci im Beginn der Gesch1chte beigebracht hat. Die Asvin tun dasselbe absichtlich, um den im toten N amuci verunreinigten Soma zu reinigen, und ihn so dem lndra wiederzugeben (a a o. 159, Mahidhara zu VS. 10, 33). Daher, so meint BI., steht vi-pii das eine Mai 1m Med. das andere Mal im Akt. Der Untersch1ed zw1schen Med und A.kt i~t aber ledighch <lurch die verschiedene Konjugation bed1ngt m-pd nach der seltenen Pras. Klasse 3 ist Med , vgl vipipiya, vipipiie Jaim. Br. 3, 228, dagegen ist vi-pii nach Kl 1 (piba) ausser RV. 3, 53, 10 stets Akt.

Hart 1st auch das Auseinanderreissen der beiden Vorgange in den zusammenhangenden RV. Str und die Beziehung der Worte namuciiv asure sacii, auf den to ten Namuci, wahrend der Lok. bei sacii in Ver­bmdung mit pd und ahnlichen Verben stets den Gast- und Opfergeber bezeichnet piba - tugrye saca 8, 32, 20, srusti,qau saca 8, 51, 1; ayau miidayase sdcii 8, 52, 1, ktpe miidayase saca 8, 4, 2, kanr1esu .m saca piba 8, 4, 3 Der Fehler aller Interpreten ist ebcn in vipip<ina, vyapi­bah zu suchen, das nichts anderes ist als das spatere m-pii Das hat schon Oldenberg (Nachr der Gottmger Ges. 1893, 343) geahnt, ohne klar zu sagen, wie er die Stelle verstanden haben will Als Schlagwort der Namucisage wird es ausdruckhch in Sat 12, 7, 3, 4 <lurch andhasor vipanam bezeugt, 1st aber in der spateren Fassung der Sage nur noch halbverstanden und an die falsche Stelle gekommen Das andhasor vipanam in Sat. ist dasselbe wie andhaso vipiinam Tan<;). Br. 14, 11, 26, wo wohl gleichfalls andhasor zu lesen 1st. Dazu bestatigend andhasi vipipite Jaim Br. 3, 228. Die beiden andhas smd nach der letzten Stelle das dawyam und das miinusam, also Jedenfalls Soma und Sura. Und so ursprunglich auch in der Namucisage.

Damit fallt aber Roths ge1~tvolle Deutung von surama. Suramam bezeichnet Jene eigentilmhche Mischung von Soma und Sura, die N amuci dem lndra vorsetzte, und die in Erinnerung an jene Sage im spateren Ritual nachgeahmt wurde, VS. 19, 1 (he sure, tvam somena samyojayami Mahidhara). Auch das hat Oldenbergs Scharfsinn (a. a. o. 348) herausgefunden, aber seine Deutung des Wortes ist verfehlt. Um zu emem brauchbaren Sinn zu gelangen bietet sich eine doppelte Er­klarung. Es kann aus surd+ ama zusammengesetzt also Bahuvrlhi sein: die heftige Wirkung des Branntweins habend, <lurch Branntwein

Da8 Vipanam im Rigveda 78

stark gemacht, vulgar 'geschnapst.' Dann ist somam zu erganzen. Man kann in diesem Falle sura-ama oder surd-ama zerlegen. Ersteres nach Wackernagel II, 1, § l13b,B, letztes nach § 115c. Oder surtima ist kollektiver Dvandva und eine Kurzung aus surasoma, das dann dem andhasi entsprechen wurde Diese Erklarung kommt der von J. Brune bei Oldenberg zu RV. 10, UH, 4 vorgeschlagenen nahe. Nun­mehr Ubersetze ich:

10, 131, 4: "lhr Asvin habt bei dem asurischen Namuci den stark­geschnapsten (Soma) durch 1'rinken geschieden, 1hr Herren der Schbnheit, und so dem lndra bei seinen Taten beigestanden.''

"Wie die Eltem dem Sohne so standet ihr Asvin (dir), o Indra, bei mit euren Erfindungen und Meisterkunsten; als du mit Geschick den starkgeschnapsten (Soma) <lurch Trinken sondertest, <la he1lte dich, Maghavan, die Sarasvati "

Die Strophen !.pielen auf eme Ep1!.ode im Namuci-lndra-Streite an, die !.ich 1m RV. sonst mcht w1ederfindet und aus eigener Kom­bination oder Phantas1e, wie man es nennen will, vervollstand1gt wer­den muss. Es fragt sich zunachst, wo d1ese Episode im Zu!.ammenhang der ganzen Namucierzahlung emzuordnen 1,ei Doch wohl vor der eigenthchen Katastrophe, ungefahr bei Sat 1~, 7, I, 11-12, da wo die Asvm und SarasvaU zum ersten 1\Iale angeruft>n wnden, aber noch vor dem volligen Zusammenbruch des lndra

Gottergesch1chten !.md nur das Spiegelbild menschhcher Verhalt­nisse lndra und Namuc1, der starhte Gott und der starkste Asura seiner Zeit, zwei grosse Rivalen und mnerhch spinnt'femd, versuchten zunachst sICh auf diplomabschem und gesellschaftlichem Tone ausser­hch gut zu stellen Jedenfalls haben mcht nur die Kommentare son­dern auch das Epos recht, wenn s1e von der Freundschaft beider sprechen (s Bloomfield a a o. 147). S1e luden 1,1ch zu Gast und Namuci versuchte dem Indra <lurch Alkohol ein Bern zu stellen, indem er in den Soma den Branntwem mi!.chte, m das gotthche Getrank das damom!.che. Die List gelang mcht sofort Die gottlichen ,,·under­manner, che Asvin, machtt>n dem lndra die Kunst vor, aus der Trank­mischung den Soma allein herau!.zutrmken (Str 4). Indra tut das Gleiche, tauscht also den Namuc1 und Sarasvati he1lt die ublen Folgen (Str. 5). Wie es schhesshch <loch noch zur Katastrophe kam, bleibt dann eint> ungeloste Frage. Es ware darum m engerem Anschluss an die Brahmai:iaerzahlung, auch moglich, class lndra gle1ch anfangs dem listigen Gifttrank des Namuci erlag, und <lass d1eser nicht nur einc bestimmte Somahbation, sondern den Somatrank uberhaupt m1t Sura vergiftet hatte, die Asvin aber nach der Rekre1erung lndras durch

74 Karl Friedrich Geldner

Sarasvati diesem vor dem entsche1denden Kampfe die Kunst des Vipana beigebracht haben. Denn fur diesen Entscheidungskampf brauchte lndra seine volle Starke und den Soma. Dann bleibt aber der Ausdruck ruimucau saca hart. Klarhe1t ist nicht mehr zu erreichen.

Zurn Andenken an d1eses bedeutsame Ereignis, an die Rekre1erung Indras und die Rem1gung des Soma wurde die Sautramanife1er geshftet als cm Reimgungsopfer, das besonders von einem entthronten Konig (Katy Sr 19, 1, 3) zu begehen 1st und von emem, der sich in Soma ubernommen (1b 2), also modern ausgedruckt, Mch eme Alkohol­vergiftung zugezogen hat.

UNIVERSITAT, MARBURG.

ON THE REAL MEANING OF THE DIALOGUE BETWEEN YAJNAVALKYA AND MAITREY1

BY CARLO FORMICHI

EVEN a superficial reader of the Upanishads knows the famous dia­logue between YaJfiavalkya and his wife Ma1treyi and 1s ready to

quote it as one of the loftiest passages m the whole rehg10us hterature of ancient India But the meaning of the sage's words 1s far from being clear "Behold, not indeed for the husband's sake the husband is dear, but for the sake of the self, the hui,band 1s dear Behold, not indeed for the wife's sake the wife 1s dear, but for the sake of the self, the wife 1s dear. Behold, not for the soni,' sake the sons are dear, but for the sake of the self the soni> are dear." And so on, m the text the thought is em­phai>Ized that <'Verythmg dear to U<, as property, brahmanical or ksha­triya d1gmty, heavenly worlds, Gods and creatures arc not dear m and for themselves but only for the sake of the self.I

Two such d1stmgmshed and authoritative mterpreters as Deussen and OJdenberg are at vanance m the rendermg of the meanmg of the text The former states '' dies Ji,t mcht etwa die Proklamation des Standpunktes emes extremen Ego1smus, sondern bedeutet (<la der Atman das erkennende SubJekt m uns 1st), <lass wir alles in der Welt nur rnsofern erkennen, bes1tzen, heben konnen, als w1r es als Vorstellung m unserm Bewusstsem tragen " 2

Quite opposite to this ('Xplanat10n 1s the one grven by Olden berg. "w1r lieben, sagt YaJfiavalkya, m Wahrheit allem unser Selbst alle andre Liebe fhes!ot ans d1eser Liebe und chent ihr wobe1 das m der Tat deuthch genug ausgesprochene Bckenntms zur Selb!othcbe <loch wohl aufgefasst werden darf als vert1eft <lurch den m1tkhngen<len Gedanken, class das e1gne Selbst das Allselbst 1st " 3

I thmk that great help may b(' drawn from the followmg passages "The atman 1s dearer than a son, dearer than wealth, dearer than any

other thmg, because what is mmost 1s the atman, and whosoever, to one saying anythmg else is dearer than the atman, declares "you will lose

1 Hrhaddranyalm-Upam.\had, II, 4, 5, IV, 5, 6 2 Sechz1g Upanishad',y des Veda, p 415 3 Die Lehre der UpanUJhaden und dtc Anfange des Buddh1.~mus, p 197

76 Carlo Formichi

this your dear thing," he lordly proclaims what will necessarily happen. Therefore one should worship as dear only the Atman, for who worships as dear only the atman to him no dear thing is any longer perishable." 1

"When brahmanas know the atman, then freeing themselves from the desire of offsprmg, from the desire of wealth, from the desire of heavenly worlds, they lead the hfe of wandering mendicants, for desire of offspring means desire of wealth and desire of wealth means desire of heavenly worlds " 2

"Who knows the atman becomes a .Muni and wanders about m search of Him alone as in search of his own world This 1s why the for­mer sages did not desire offsprmg and sa1d to themselves 'what is the use of offspring to us for whom the atman is the whole world ;i' Freeing themselves from the desire of off spring, from the desire of wealth, from the desire of heavenly worlds, they led the hf e of wandermg mendi­cants, for desire of offsprmg means desire of wealth and desire of wealth means desire of heavenly worlds " 3

YaJiiavalkya, who has known what the atman is, gives up all mun­dane affections and resolves to lead the hfe of a wandermg mendicant. What he says to his wife is a trui&m that 11-con1-tantly met with in the Upanishads there is only one reality, and this is the rndescribable, transcendent atman which knows no change, no sorrow, no death, no to-day, no to-morrow, no here no there, and he who for the sake of this one reality forsakes the world and 1b 1llus10n1-will transform all perish­able thmgs dear to his heart mto unpenshable one&

I agree with Oldenberg that in YaJf1avalkya's words there is no trace of the atman bemg das SubJect des Erl.ennens and a'i such der Trager dieser ganzen Welt, as Deus&en mamtams, but at the &ame tune Olden­berg's assertion that, accordmg to YaJfiavalkya, we love only our self and every other love flows out from this love, seems to me 'itrongly obJectionable At any rate, YaJfiavalkya's acknowledgment of 1,,elfish­ness would be a most strange one Can a man he called 1-elfish for re­nouncmg the world, hvmg on alms and a1-pmng to a supreme rcahty which, though not persomfied, appeal& to hi-, heart as strongly as any personal God to the hearts of his worshippers? In the Upamshads the word atman always means somethmg that is even above the Gods, and to attach to it the idea of selfishness 1s to forget what the Upamshads try constantly to demonstrate. the meffable transcendency of the Atman.

1 Brhadar Up I, 4, 8 a Ibulem, IV, 4, 22

2 Ibulem, Ill, 5, I.

Dial-ogue between Y8jnavalkya and Maitreyt 77

There is, accordingly, no relation at all between Yajiiavalkya's ex­pressions and those contamed in the Samyutta Nikaya, I, 75, where to the question, "Is anything dearer to you than your own self?" twice, "No" is given as answer.1

No religion has hitherto dared to acknowledge selfishness as its basis, and the Upanishads make no exception to the general rule.

1 Oldenberg, op c1t, p 851, n 124

UNIVERSITY OF ROME

THE DATE OF VASUBANDHU, THE GREAT BUDDHIST PHILOSOPHER

BY JY AN TAKAKl SU

I T 1s more than twenty years ago that I proposed a probable date for Vasubandhu, the author of th{' Abhidharma-lwsa (realism) as well

as of the J'i1niinarniitraid (1deah~m), m thP Journal of the Royal Asiatic Society (January, 1905) The date of Vasubandhu then proposed by me was A n 420-500. M Sylvam Levi at fir~t proposed a later date than mme, namely, the first half of the sixth century He has, how­evn, smce abandoned that date, m hi~ translation of the Siltra-alankiira, published m 1911, and has expressed hi5 view that Asanga's activity cowrs the whole of the first half of the fifth century As m those days htnary act1v1t1e~ m India seem to have been at once reflected m Chma, the date~ of the Chme5e translations of Indian works will serve as a reliable gmde m fixmg dates m Indian rehg10us or ph1losoph1cal history Actmg upon this conv1ction, I first published a translation of Paramartha' s L1fc of r· a.rnbandlm, and, as an appendix to 1t, Kue1 Chi'~ ver~10n of the controversy between the Ruddhi5t and the Sankhya philosophers m the T'ounq Pao (1904), and subsequently proposed m 1905 the date Jm,t nwnt10ne<l

Smee thm several scholars have taken up this quest10n Our lamented friend 1\1 N Pen.1 after a nunute mw~tigation, suggested AD 350 as the dak of Vasuhandhu's death, while Professor U. Wog1hara 2 propo-;ed AD 390-470 for Vasubandhu, and AD 37J-450 for Asanga In the meantmw tht' dates were vigorously discussed by several 5cholars m Ja pan Professor B Shno proposed a still more early date (AD 270-3.50) than that of M Pen, whereas Professor R. Hikata satisfied himself with the date gIYen by Professor Wogihara.

Only lately Professor 'J' Kmmra, takmg a different ba51~ of dis­cussion for his mvestigat10n, confirmed the date given by me. Still later, Profe~sor H lh, while d1scu-,smg the e:\.istence of a philosopher called Maitreya besides the would-be Buddha of that name, sheds a side-light on the dark points of several important questions, though

1 "A propos de la Date <le Vasuhandhu," Bulletm de l'Ecole Fran,;a1se d'extreme Orient (1911), p 339

2 Asanga's Bodhisattvablzum1, em dogmatucher Text der Nordbuddhisten. Le1p• z1g, 1908

80 Jyan Takakusu

his own date for Vasubandhu is A.D. 320-400, or one hundred years earlier than mine There are yet several other savants standing at opposite poles, the one being Professor S. Funabashi, who advocates the earlier date (fourth century, before Kumarajiva), and the other, the group includmg Professor E. Mayeda (after Kumarajiva, who was in China 383-414), Professor S Mochizuki (between 433-533), and G Ono (415-515), who favors a later date, somewhat later than even my own Now we can well draw our discussions to a conclusion, as we have almost exhausted our exammation of the materials at our command Besides, we may perhaps look forward to some fresh in­ternal evidence, now that new material has been discovered m Nepal and is being prepared for pubhcation by M Sylvam Levi I mean the Sanskrit text of Vasubandhu's r13iiaphtrimsaka, with Sthiramati's commentary on it As for ourselves we shall now sum up our studies and see if we can settle the quebt10n once for all

My arguments rest chiefly on evidence adduced, first, from some Chmese biographers, becondly, from travellers from or m India, and especially from the hfe of Vabubandu written by Paramiirtha, and thirdly from the dates of translation of the works of Vasubandhu and his contemporaries

1 KumiiraJiva (383-414 m Chma), who wrote or tranblated the biographies of A~vaghosa, NagiirJuna, and Aryadeva, did not write the hfe of Va&ubandhu, though one catalogue ment10ns by mistake the hfe of Vasubandhu

2. Fa hien (399-414 m India) doe& not seem to know of our philoso­pher, none of whose works is found among his tranblations

3. Paramiirtha (500-,569, 539 mvited from India, MG-569 m Chma) is the first who wrote a b10graphy of our philosopher He him­self is an 1deahst philosopher and ment10ns the death of h1& predeces­sor at the age of eighty, at Ayodhya, which must have occurred before 539 (the date of mntat10n to Chma), or, at the latest, before 546 (the date of his arrival m Chma) This fact alone made M Sylvam Levi's first proposal nnpossible Besides, Vasubandhu 's Mahayiimst1c works, which were written m the last part of his hf e, are subsequent to his conver&1on to the Mahayana and the death of Asanga, which occurred at the age of seventy-five If his elder brother was seventy-five, his age would be about seventy or more Durmg about ten years of his conversion, he seems to have written mneteen works as I pomted out in my Life. Some may hold the ten years as incredible, but as he said himself that he was too old to dispute with Sanghabhadra while he was as yet Hinayanist, the remaimng years of his life could not have

The Date of V asubandhu, the Great Buddhist Philosopher 81

been much more than a decade. Many of his works were translated into Chinese as early as 508, 509, 508-511, 508-538, 529,550,563, and by Bodhiruci (An 508-535 in China), Paramartha (A.D 546-569 in China), and others The compilation of these works was naturally earlier than the dates of translation

I thought it quite reasonable to take A D 500 as terminus ad quem and to fix his date as covermg three quarters of the fifth century. Con­sequently A.D 420-500 was proposed as a possible date for Vasubandhu This hypothesis of mme doe1-, not confhct with any date from the Life itself, namely, a controversy between the Buddhist and the Sarikhya philosophers (Buddhanntra versu.~ Vmdhyavasa (lsvarakr~na), a patronage of the Kmg V1kramad1tya and Balad1tya, a dispute With Vasurata, a grammarian of the Candra school, also with Sarighabhadra, Vaibhas1ka philosopher and the author of the Samayapradipika. Paramartha's date given for Katyayaniputra and A~vaghosa 1s "in the 500 years of Nirvana" (that is, 500-599). while that given for Var~aganya and Vmdhyavasa (f~varakrsna), the elder contempora­r1es of Vasubandhu, JS "m the 900 year!:> of N irval).a" ( that JS, 900-999) These two dates too are not made 1mpossJble by any statement m the Abludharma-kofo itself

In the Abhidharma-kosa 1t 1s i'>a1d " ,Yhen we know that the hfe of the Buddha's Law 1s about to end, reachmg almo-,t the throat, and that It 1s the time when vice is m power, seek your deliverance, be d1hgent " This would mean that the hfe of the Buddha ~asana, which lasts for 1,000 years, IS about to breathe its last Accordmg to Y a~o­mitra, 1 the or1gmal 1s Eram kanthagafapranarn rulitrti munz,qtisanam, maltinti,n balal,alarn ea na prmntidljam munwksubhth (" It bemg known that tht' hf e breath of the teachmg of the sage [is about to go out and] reached to the throat and 1t 1s the prevailmg tune of vices, those who seek deliverance should be diligent ") The r· IJtiklzya of Yasom1tra does not explam 1t But Hmen tsang's explanation puts the durat10n of the saddharma as definitely one thou!:>and years ,ve can thu!:> ai-.sume that we are nearmg the end of the period, that 1s to say, we have passed nme hundred year!:> and more smce the Buddha's Nirvana This 1s m perfect agreement with Paramartha's statement " m the mne hundred years," wluch means neither '' w1thm mne cen­turies" nor "m the mnth century" ,ve may better express it as "m the years nme hundreds," that Is, from 900 onward till 1t reaches 1,000.

1 L' Abhidharmakosa de J' asubandhu, tradud et annote, by Professor de la Vallee Poussm (1925), chap. vm, 40, p 224, note I

82 Jyan Takakusu

Now as to the calculation of the Nirva:r;ia Era. That a compara­tively trustworthy tradition as to the date of the Buddha's Nirva:r;ia was current m India among the Buddhists about Paramiirtha's time can be seen from another source, quite independent of Paramartha's Life The famous " Dotted Record " of Indian sages, which was brought to Chma by Sanghabhadra, mdicated 975 dots from the Nirva:r;ia to An 489, one dot having been added every year to the Vinaya pitaka after the Var~a ceremony was over

Sanghabhadra translated Buddhaghosa's Samantapasiidilra in An

489 in Canton, and kept his Varsa retreat there, thereby addmg the last dot to the Record Accordmg to this Record, the date of the Nir­vana would be B c 486 1 The 500 yearf.i after Nirvana as the date of Asvaghosa would mean some time falhng between A D 14 and A D 11~3, while the 900 years after Nirvana as the date of Vasubandhu would fall somewhere in An 414-513 This 1s an adcht10nal corroborat10n which cannot be rejected without some mcontrovertible faet

Now M Peri took up all the books of a dubious nature, which I purposely reJected as such, and proposed an earlwr date for Vasu­bandhu His arguments are d1v1ded mto eight heads (1) Bodh1ruci and Chm kang s1en Jun, (2) the list of patriarchs, (~~) the Mahayana­avatiira and Sth1ramat1, (4) Chronology, (5) the Sata sastra, (6) Bodh1cittotpiida siistra, (7) KumiiraJiva; and (8) D1sc1ples and com­mentators of Vasubandhu These titles alone seem to md1cate that he has exhausted nearly all evidence that can be adduced from the Chme,;e side Though his pamstakmg research incorporates many of the discuss10ns then gomg on m Japan, and is marked everywhere with a high scholarship m Japanese as well as Chmese Buddhist litera­ture, nevertheless there are no poinb, I am sorry to say, which really compel us to accept his date

}""or my part I have been trymg for some time to carry on anew my research mto the materials hitherto not sufficiently exammed by my­self The deeper I go into the matter, the more I feel convmced that, after all, my date of Vasubandhu 1s nearest the mark So mstead of laying a lengthy d1scuss1on before my readers, I shall confine myself to givmg a few important pomts that tend to undermme the founda­tion of all earlier dates proposed for Vasubandhu Among others there are three important facts on which the proposition of earher dates is based and m face of which my date seems to fall to the ground at once

1 For the details of the Dotted Record, see my "Pah Elements m Chmese Buddh­ism," J R A S, July, 1896

The Date of V asubandhu, the Great Buddhist Philosopher 83

First, the Chin kang sien lun, a work on the Vajracchedilcii, either translated or written by Bodhiruci (A D. 508-535), which M. Peri, Professors Shiio and Ui make very much of Secondly, the Bodhisattva­bhiimy-iidhiira, a portion of the Yogiiciira-bhumi, translated by Dhar­marak~a (A.D 414-421), along with the Bodhisattva-bhadrasila, also a portion of the Yogiiciira-bhumi, translated by Gm;iavarman (A D. 421). on both of which the arguments of Professors Wogihara. Shiio, and Hikata arc based To the same category belongs the Siitralaitkiira, said to be composed by Asanga, the origmal of which was published by Professor Sylvam Levi, who based upon it a new date for Asa:riga. Thirdly the Samyukta-abhidharma-hrdaya, by Dharmatriita, translated by Sanghavarman (AD 424). The name of Vasubandhu contained in the work was and is still a centre of dispute, especially among the Japanese savants

Let us now examme these points First as to the Chin kang sien lun The work is said to be a commentary on the Va;racchedikii, written by a man named Chm kang sien (perhaps Va1rar~i m Sanskrit), who is supposed to have been a pupil of Vasubandhu, and trarn,lated into Chmese by Bodh1ruci (A D 535) At the end of volume 10 of this work there is a pararnparii of transmission to the following effect· 1

" Maitreya composed a commentary on the V a1racchedikii and the Bodhisattva-bhumy-iidhiira, and handed them down to Asanga, who in transmitted them to Vasubandhu. The latter agaui wrote a sub­commentary on the r a3racchedikii, and handed it down to Chin kang sien (perhaps Va1rar~i) Then Chm kang s1en to Wu t<.m yi (Aksaya­mati), Wu tsm yi to Cheng tsi, Cheng tsi to Bodhiruci. thus transmit­ting it successively without interruption until to-day for about two hundred years " Such a paramparii would be very important if it came down from Bodh1ruci himself, or 1f we knew anythmg about the names to verify the truth But 1t seems to be all in the negative First of all, the work is stated to be the translation of an Indian original; but that it is not a translation can easily be seen by a perusal of its content If it is not an Indian work, it can only be a comp1labon in Chma, either by Bodhiruc1 or by some other hand In any case the work was in existence soon after Bodhiruci, because it was quoted by K1 tsang (A D. 548-623) and Hm yuan (died AD 589) Ki tsang's quotation m the commentary on the J'a3racchedikii, however, differs somewhat from what we have in the present text of the Chin kang .men lun. The work may have undergone an alteration by a later hand. In the style of composition and the method of annotation we notice

1 My edition of the Clunese Tnpuaka, ,ol X..XV, no 1512, p 874

84 Jyan Takakusu

many points that give the work the stamp of Chinese authorship. As to the authenticity of the work there was some dispute among Chmese savants of old, and the learned Ki, 1 a pupil of Hiuen tsang, finally passed a sentence saying: "As to the sastra Chin kang sien (Vajrar~i), 1t is said that he [Chin kang sien] was a man of Wu in the southern section [of China], and his work is not a true sacred teaching." This means that Chin kang sien was not an Indian but a native of Wu in the south, and 1s not giving us a true teaching. It would amount to saying that the work is a Chinese forgery As we know that learned forgeries were rather habitual early in the Six Dynasties and Sui Dynasty, and that quotations from forged works can be found even in some serious books, we can almost safely assume with the learned Ki that the work in question belongs to this category Apart from this, the paramparii itself is of a dubious character Between Asanga-Vasubandhu and Bodhiruc1 only three names are given, and yet a period of two centuries is said to have elapsed This is hardly credible, as Mr Ui rightly thinks. Besides, those three transmitters are mere names unidentified with any known personages in India or el~ewhere Not one of the catalogues of the Chinese Tripitakas, thirteen m number, has recorded the work either as a translation or as a Chmese compilation Havmg these facts before us, we can m no way utilize a work hke this as a basis of proposmg a date, even if I may be too harsh in regardmg it as a forgery

Next as to the Bodhisattva-bhilmy-adhiira and the Bodhisattva­bhadrasila, translated into Chinese AD 413-421 and 431 respectively. These two are, doubtless, portions of the large Y ogacara-bhumi. If Asanga is the actual author of all the three texts, a date anterior to that of the translation will be suitable for him, and my date for Vasu­bandhu ought to be given up at once From this pomt of view Pro­fessor Wogihara proposed Ao 390-470 for Vasubandhu and A.D.

375-450 for Asanga, though the reason why he puts the difference of twenty years of age between the two brothers by the same mother is not qmte clear to me Professor Sylvain Levi, too, on this account abandoned his former date and proposed a new date, holding that Asanga's activity covers the whole of the first half of the fifth century, as referred to above. Concerning the authorship of these four books just mentioned, Professor Ui rendered us a great service in his research

1 K1 1s the youngest but most learned pupil of Hmen tsang Formerly I called him Kuc1 Chi and M Peri, Koue1 k1, but we were mistaken, his real name bemg simply K1 His words here quoted are from his commentary (Tsan shu) on the Vajra­cchedika.

The Date of V asubandhu, the Great Buddhist Philosopher 85

on Maitreya, in which he concluded that Maitreya was the actual author of these works, not simply an inspirer of Asanga. I have made an abstract of Mr. Ui's research from his Study of Indian Philosophy, and give it here as a supplement to this article. According to him most of the works hitherto assigned to A~anga, inspired from above by the Bodhisattva Maitreya, are in reality to be attributed to Maitreya himself In that case Asanga would be only a transmitter, direct or indirect, of Maitreya's Mahayana doctrine. Chi yuan, 1 who went to India with F,a hien and Pao yun (A D. 399-414), mvited Buddhabhadra to China, and on bis return home translated several Sanskrit texts. As he had some doubt as to the Vinaya practice, he started once agam for India by sea some time after AD 427, and after his arrival in India he saw several Arhats and inquired about his doubt No one, however, could give a decisive answer. Thereupon an Arhat ascended, while engaged in meditat10n, to the Tusita heaven and mqmred of Maitreya hving there as to Chi yuen's doubt, and found that the latter was right in his un<lerstandmg He was satisfied with it, and on his way home he reached Ki pin (Ka~mira), where he died without a disease, aged seventy-eight. If Maitreya lived on earth at all, he ought to have been m activity about that time Some of the fiw Dharmas which are attributed to Ma1treya m Tibet are assigned to Saramab m China Whether Saramati can be identified with Maitreya or not remams to be seen. If, in any case, a scholar named Maitreya be found to be the author of those works hitherto attributed to Asanga, then the date of the latter ought to be shifted later, at least by one generat10n, if not more The ground for an earlier date for Vasubandhu should give way altogether

Thirdly, as to Dharmatrata's Samyukta-abhiilharma-hrdaya, Pro­fessor T. Kimura has shown in his Study of the Abhidharma Literature, I thmk, with a successful issue Accordmg to him, Vasubandhu sys­tematized Dharmatrata's work Just mentioned, and developmg it with the Sautrantika views, wrote the Abhulharma-l-o.fo as a textbook of the realistic school. Dharmatrata's work mentions the name "Vasu­bandhu," who represents a philosophical tenet of his time. This Vasubandhu is explained to be an older Vasubandhu by Fu kuang, a pupil of Hiuen tsang, but those who advocate an earlier date will hold 1t to be a mistake on the part of Fu kuang

Fortunately, however, this point can be traced m Yasomitra's Ab­hidharma ko§a vyiikhyii,2 which takes that old Vasubandhu to be a

1 "Chu san tsan clu ts1" (NanJ10, 1476) vol XV, s 7 2 Professor de la Vallee Poussm, Vasubandhu et Yascnn1tra, pp x1x and 159.

86 Jyan Takakusu

Sthavira and the Upadhyaya of Acarya Manoratha, who is, according to Paramartha's Life of Vasubandhu, an elder contemporary of Vasu­bandhu. Accordmg to the internal evidence adduced by Mr. Kimura from the Ahh1dharma literature, a succession of these principal Abhid­harma works 1s to be assumed, and Vasuhandhu's Abhidharma-kosa is to he placed at the concludmg period, and has to come subsequent to Dharmatrata's work, which was translated thrice - A D 418, 426, and 434 Had there been m existence Vasuhandhu's Abhidharma-kosa, which is ever so much clearer than Dharmatriita's work, why should they translate the latter work so often, one after another, one Chmese (Fa h1en) and four Indian (Buddhabhadra, I~vara, Gu:r:iavarman, and Sanghavarman), all well up m thmgs Indian, having been engaged in it As for further detail, the reader is referred to Professor Kimura's discuss10n, an abstract of which is also given here at the end

The above three points made clear, 1t will he quite sufficient for our purpose, because all other chronological data from the Chmese sources are often contradictory to each other, and the lists of transmis­sion of the ldeahsbc &chool, from Vasubandhu down to Bodhiruc1 or Dharmapala, are also m conflict with each other What remams for me to do 1s to translate all the important materials and to lay them before my readers, a task too heavy for me at present The present article 1s to mdicate simply that the date proposed by me some twenty years ago still holds good, now that so many !.cholars have been in the arena and so many earlier dates have been proposed It is grabfymg to me to see that Professor R Garbe, m the second edition of his Sankhya Phiwsophy (p 74), follows my date throughout, givmg a reason why he does so. The followmg hsts of Indian philosophers may help us to see that the interval between Vasubandhu and Bodhiruci or Dharmapala can in no way he so wide as two centuries

I

LISTS OF TEACHERS, SHOWING THEIR RELATIONS TO VASUBANDHU

TEN COMMENTATORS OF VASUBANDIIU (Vi3niinamatrata)

I. Bandhusri, contemporary and commentator of Vasubandhu, circa, AD 420--500. 2 Citrabhiinu, contemporary and commentator of Vasubandhu, circa AD 420--

500 S Gunamat1, his Lalc.,ananwiara, translated by Paramiirtha (came to Cluna AD

546) Teacher of Sth1ramati (4), OJ Late Years, Itsmg, AD 671-695 4 Stluramab, elder contemporary of Dharmapiila (6) AD 528--560, pupil of Guna­

mab (S), Of Late Years, Itsmg 5 Su<ldhacandra, contemporary of Sth1ramat1 (4), commentator of Vasubandhu's

Paramiirthasaptati

The Date of V asubandhu, the Great Buddhist Philosopher 87

6, Dhannapiila, died AD 560, aged 32, teacher of Silabhadra whom Hmen tsang met at his advanced age of 106, AD 633, OJ Late Years, ltsmg.

7 Nanda, teacher of Jayasena known to Hmen tsang AD 629-645 m India, com­mentator on the Y ogacarabhitmi

8 V1sesam1tra, pupil of Dharmapala (6), commentator of Ma1treya's Yogacarabhumi; otherwise called Jmam1tra by NanJ10

9 Jmaputra, pupil of Dharmapiila (6), commentator of Ma1treya's Bodhiaattva­bhumi

10 Jiiiinacandra, pupil of Dharmapiila (6), lnmg m T1ladha v1hJra rn Itsmg's time, AD (,71-69j

NB Of the aho\e names. No~ I and 'l are ~aul to lmve live<! 111 the 900 wars p N, at the same tune a!. Ya.,uhandhu, v. lule No'> 3 to 10 hvcd dft(•r ll00 ~ears P N

The long mten al I', utterly unaccount.ible Either one or the other 1s a mistake As to Dbarmapala, ltsmg makes 111111 a contemporan of Bhartrhar1 v.ho died forty years beiorc 1nm, that 1s, circa ~ n 630-bjQ Tlu., !-.eems to he an error

II

(a) PoI.\n, OF DH,PLTE

I Bandhu~ri

3 Gunamab, dern mg lu~ Hlea from B andhu;ri. a!,serts two dn 1s1011s of menta function

4 Stlura11Mt1. pupil of Gunam<1.t1 (3). one dn 1~1on of ment.il ftml t1on

7 Nand.i, bto cln1s10rn, of mental funct1011

(b) PoI:\:'T.., OF 1)1-,Pl TE

2 C1trabhJ11u (a) Dmu,1ga. dcrn mg 111, 1rlrn from C1tr<1hhJn11 (\?), ,1s<;crt,, three dn 1<;wns of

mc11t,1] tunct1ou,. <1uthorit, ol lu1u111lv1i (logic)

(i DharmapJl.t, four dn l'>IOII!, ot ment,tl fum-twu, pupil of DmnZtga (,L{Cordrng to T1het.tn trad1t1on)

(b) S.wJ....ir.1<;1,1m1. pupil ,rnd <;11ccp<;<;or of Dm1i.1!(.1 (a) m luluvuh;a (logic) (c) I'r.i'a,t.ip,lda, Ya1~t',1J...a pl11!0.,oplwr defrnclrng humelt ,1gamst DmnJga's

hetumh;ii (log1e) . .i,,ert<; umc y111w11 of \tm:1 (d) Par,1111,irtha. cla•rl \ D jh!) aged il quok;, nrne yu1,a, ot the l'a1M'~11a,

write;. the L1fr of Yasuharnlhu. tmn!ol.tte, tl\o \\Orh.<; of Dmn.ig.i. (a) .ind ol!e work of Gunam,iti ('.l)

(c) l'oIXT'- OF l)J-,Pl,TE

(e) Can<lrapiila, adnut<; onl) pfln ,l-\'d!oanZI m the ,1la~ a-1·1Jfi,ln.1

7 Nanda, asserts nava-vJ.san,"i (J) Jayasena, follower of Nanda (7), ll'>!,erb nava-, ii!.anii

6 DharmapJla, a<lnuts pflrva-, 11s.i.na a~ v.ell a!o na, a-vii,anJ

88 Jyan Takakuau

III

LIST OF OPPONENTS OF V ASUBANDHU

(1) Sanghabhadra, opponent m Abh1dharma, Of the Muldle Age, ltsmg, two works agamst the Abhulharma koaa.

(2) Bhiivaviveka, opponent m Prajfliipiiranntii, Of the MuMle Age, ltsmg. (8) Vasurata, opponent m Vyiikarana, one work agamst the grammar of the Abhi­

dharma-koaa (4) Isvarakr~na, opponent m Sankhya, agamst his Sankhya-saptaii Vasubandhu

wrote Paramiirtha-saptati

IV

CHINESE THREE ScHOOL'> OF IDEALISM (Vi3fiaptimatratd)

I T1 lun tsung (Da5abhum1 school) Founded by Borllunic1 from North India (A D 508 to China), probably rcpresentmg the North school of the 1deahstic philosophy

II Shueh lun tsung (Mahayilna-samparigraha '!chool) Founded by Paramartha from UJJaymi, West India (A n 539 mv1te<l to Cluna), probably representmg the famous \'alablu school of the 1dealu,t1c philosophy

III Hu fa tsung (Dharmapiila school) Founded by Hmen tsang (A D 629--645, m India), representmg the then flour1shmg Nulanda school, m'ltituted by Dhar­mapiila, with the appearance of tlus school the tl\-o old sc-hools became well­rugh extmct

V

THREE PREDEcEssoRS OF VARUBANDnu's RE-tLISTIC. PHILOSOPHY

I. Dharmottara (or Dharma&r1) His work on realism, Abhidharma-hrdaya (Nan110, 1288, translated AD S48-S44 or 384)

II UpaMnta Hts work, Abhidharma-hrdaya (enlarged) (NanJto, 1204, translated AD 568)

III DharmatrJ.ta His work, Samyukta-abhul.harma-hrdaya (NanJtO, 1287, trans­lated [l] AD 397-418, [2] 426-481, [3] 484)

IV Vasubandhu His work Abhidharma-ko8a (NanJ10, 1269, translated An. 563-567, 1267, AD 654, }270, AD 651)

N B For further details, see Professor Klffiura's supplement

VI

TARAN1THA's PARAMPARA OF TEACHERS

I. V asubandhu II Dmniiga, pupil of Vasubandhu

III Dharmapii.la, pupil of Dninll.ga

IMPERIAL UNIVERSITY, TOKYO,

THE DATE OF VASUBANDHU SEEN FROM THE ABHIDHARMA-KOSA

TnE Foun TEXTS

By TAIKEN KIMURA

THE earliest epitome of the Vaibhas1ka philosophy is the Abhi­dharma-hrdaya, written by Dharmottara or Dharmasri (NanJio,

1288). It treats of important questions of the abhidharma philosophy, yet it 1s, as a system of philosophy, nothmg but a crude writmg m form and content. Then comes Upa~anta's Abhidharma-hrdaya (Nanjio, 1294). It 1s a new work, enlarged and developed from the above text Next we have Dharmatrata, who again enlarged on and developed Upabanta's text His work 1s called the Samyulcta-abhidharma-hrdaya (NanJio, 1289) This work was agam revised and developed by Vasu­bandhu, then a great realistic phi.losopher, and is what we know as the Abhidharma-kosa (NanJIO, 1269, 1270) For a detailed discu~sion, see Kimura's Study of the Abhidharma Literature, part 5 Now let us examine the dates of the C'hmese translations of the above four texts which have appeared one after another.

CHINESE TRANSL',.TIONS OF ABTIIDHARMA HRD.HA AND OF ABIIIDHAIDIA KOSA

I

Dharmottara's Abhidharma-lmlm;a (NanJJO, 1288) Translators Sanghadcva and Tao an Dates var10usly given AD 34:J-344 (see Nan110, 1688, vol Y), AD 384 (see

NanJIO, 1476, , ol II), or .._ n 391 (~ee NanJIO, rn88)

II

Upasanta's Al1h1Jharma-hrdaya (XanJIO, 121)-J.) Translators Narendra~ J.Sas ( ... Il 537-368), and Dharma praJiliirur1 (an assistant) Date AD 568

III

Dharmatrlita's Samyukta-abhulharma-hrdaya (NanJIO, 1287) There are three translat10ns

(a) The earliest translat10n (now lost) Translators Fa h1cn, AD 899---414 (m ln<l1a), and Buddhabhadra, AD 898-421

(m Chma). Date An 897--418

(b) The second translation (now lm,t).

90 Taiken Kimura

Translators Uvara (A.D. 426-481), 10 chapters, and Gunavarman (A.D. 481), last 3 chapters. (c) The third translation (Nan310, 1287)

Translator Sanghavarman (A D. 433-442) Date AD 484

IV Vasubandhu's Abhuiharma-koaa

(a) The first translation (N.inJJO, 1269) Translator Paramiirtha, author of the L1f e of V asubandhu Date A D 563-567

(b) The second translation (NanJJO, 1267) Translator Hmen tsang Date AD 654

(c) The Abhulharma-1.oaa-lcarzkli (NanJio, 1270) Translator Hmen tsang Date AD 651

Among these four treatises, which successively develop the fonner, one, that by Dharmatrata (Ill), with its three translat10ns, is most important for ascertainmg the date of Vasubandhu, for it existed nn­mediately before the work of Vasubandhu, and was translated by five compeknt priests who were versed m thmg1, Indian of that period, that 1s to say, A D 400-434 Fa h1en, who, as we all know, travcled m India about fifteen years (AD 3!)9-415), does not ment10n even the name of Vasubandhu, and had he known of such an autht•nbc text as the Abhidharma-kosa, why should he have tran1,latcd Dharmatrata's imperfect work instead;, The rernammg four of the translators, who were all from India, arnvmg m Chma between 390 and 43~J. did not even speak of Va1,ubandhu or Asanga and, them1>elves translatmg the imperfect text of Dharmatrata, chd not brmg to hght the system­atized work of Vasubandhu There would be no reason whatever to translate three tmies one and the same older te'.\.t, had there been before them a new revised work of our great philosopher.

Two V ASUBANnnus

We know from the Life of Vasubandlw that Vasubandhu had two brothers, V1rifici-vatsa Vasubandhu and Asanga Vasubandhu, and we have to acknowledge the existence of yet another teacher named Vasubandhu anter10r to our Vasubandhu, the author of the A bhidharma kosa.

In the preface to the Samyulcta-abltidharma-hrdaya of Dharma­trata, it 1s said "Several teachers have written Abhidharma-hrdayas, which are not similar in scope, either extensive or concise. The work

The Date of V asubandhu seen from the Abhi,dharma-Kosa 91

of Dharmottara (or Dharmasri) is extremely concise, while that of Upasanta is much larger, being of eight thousand §'lokas. Yet there was another teacher whose work consisted of twelve thousand slokas. These two are said to be extensive ones Vasubandhu annotated the dharma in six thousand slokas, and his work was wide in extent and deep in meaning, not being inclined to one of the Tripitakas Such was the Asmn.<:Jlcrta-iikiisa-sastra." Vasubandhu here referred to must be an earlier teacher of that name, and not the author of the Abhidharma­ko§a. He must be anterior to Dharmatrata, who quotes him, and to the dates of translation of Dharmatrata's work, that IS, AD 418, 426, and 436, as we have seen above 2\I Pen considers that "six thou­sand ~lokas " here may be a mu,take for " SIX hundred slokas,'' because the Abhidharma kosa of our Vao;uban<lhu contaim only so many, he not bemg aware of the fact that It m reality referred to another older Vasuban<lhu. as Pu kuang hmted l\I Pen's proposition thus falls to the ground as soon as we rl'cognize Vasubandhu's Abhidharma-lwsa to be subs<·quent to Dharmatrata's work For convemence sake we will call the author of the Asa11rnkrfa-iikiisa tht' old Vasubandhu, and the author of the Ahhulharma-lwsa the new Vasubandhu

In the A bhulharma-kosa It..,elf ( Chapter 9) we come-across a sl'ntcnce explaining And_ya m the followmg word!> •· It 1s said by another com­ml'ntator that the unreasonable mana8hira as mentioned in other sfitras 1s the cause of Avidya" .

The l\ords "another commentator•· here are rightly explamed by Pu kuang, a pupil of Hrnen bang, as the' older Va..,ubandhu Further, Ya~om1tra too, m his Ablndharma-l.-ofo-1·y6.f,hyii, explams It m the fol­lowmg word-, "Stl1a11ro T'asuba11dhur Aciirya Jla11orathopiidhyiiya eram aha " See Prof e<,sor de la Vallee Pom!>m, l' asubandhu et Ya.fom1fra, pag<' 159 Thuo; the old Yasubandhu, so-C'allC'd by Pu kuang, was, accordmg to Ya~om1tra, the upiidhyiilja of the Acarya Jlanoratha, who was a contemporary of our Vasubandhu and was quoted, though without g1vmg h1!> name. m our Vasuhandhu's Adhi­dharnw-1,o.fo 1to;elf So we C'an assume with perfect safc>ty that Vasu­bandhu, the author of the A..rnnurkrfa-<il,dfo, quoted by Dharmatrata, Vasubandhu, a stha1·1ra and the upiidhyi(lJa of 1\Ianoratha, mentioned by Yafomitra, and Yasubandhu, hmted at m the words "another commentator" by the new Vasubandhu m his .tbludlwrma-Xosa and explamed by Pu kuang as the old Vasubandhu, are one and the same personage, and one and all refer to the old Vasubandhu, qmte dif­ferent from our Vasubandhu So we should be very careful not to assign a date to Vasubandhu without d1scr1mmabon

Taiken Kimura

CONCLUSION

Judging from the dates of translation of the Chinese Abhidharma texts, we can say that the Abhidharma-kosa was not yet in existence in A.D 430, and consequently during this period the new Vasubandhu had not yet entered upon his career as the author of either the Realistic school or the Idealistic school. Those who assign him a date in the third or fourth century are, I think, confusing the old with the new. I, for my part, consider the date A.D. 420-500 to fit our great philoso­pher best.

IMPERIAL UNIVERSITY, TOKYO

THE DATE OF VASUBANDHU SEEN FROM THE HISTORY OF BUDDHIST PHILOSOPHY

BY GENMYO ONO

THE theory of the three persons (kaya) of Buddha and the eight faculties of mmd were developed and brought into perfection by

V asubandhu These theories may be taken as peculiar to him In old days, as m the present day, a new thought current in India used at once to be reflected to Chma The translatmg of an Indian original was at the same time the transplanting of the idea set forth in it. Vasubandhu's two theories Just referred to cannot be traced in the works, either siltras or stistras, translated prior to the commencement of the fifth century In the Maha-parinirz,ana-siltra (Nanj10, 113), translated by Dharmarak~a AD 414-421, and the Srimali-sinhaniida­siltra (Nanjio, 59), translated by Gm;iabhadra An. 436, we have for the first tune the du,course about the nature of Buddha (Buddhasva­bhava) and the embryo of Tathagata (Tathagata-garbha) There are, however, as yet no theories of the three Buddha-kayas or the eight vi3fi<inas In the S uvarna-prabh<isa-siitra (N anJio, 127), also translated by Dharmaraksa, An 412-426, and the Tathiigatagarbha-siitra (Nan110, 384), tran~lated by Buddhabhadra AD 4!20, no theories of Vasubandhu are found as yet It 1s true there exists in the former a section, Tri­kaya inbh<iga, which treats of Buddha-lcayas This section, however, is a translation mterpolated by Pararn.artha, An 548-569 We have for the first time an enumeration of the eight 1'ZJii<inas, m the Lank<i­vat<ira-siitra translated by Gunabhadra AD 443, though the functions of 1•13nana.~ are not clearly defined In the works translated by Bo­dh1ruci and RatnamatJ A D 508-535, we see the idea gradually ripened and are confronted often \\1th the three kayas and the eight rz;iianas, though the theones are not yet s;ptemahzc<l Such treatises are the Sandln-nirmocana-siitra (NanJIO, ~4(i), the Pu tsang pu chien siitra (NanJIO, 524), the Dasabhum1 sastra (NanJIO, 1194), the ra3racchedika stistra (NanJJO, 1168), and the Saddharma-pundarika-s<isfra (NanJio, 1233), the latter three bemg Vasubandhu's own works

Fmally, thm,e treatises of Asanga and Vasubandhu more or less professing to set forth the theories were translated by Paramartha, Buddhasanta, GautamapraJfiaruci, and other~. An 531-563 These are the Mahiiyiina-samparigraha (Nanj10, 1184), of Asaiiga, the Vi-

94 Genmyo Ono

jnapti-mdtratd (Nanjio, 1238), the Karma suldhi (Nanjio, 1222), the Buddha svabhtiva (Nanjio, rn20), the Mahiiyana-samparigraha-siistra (Nanjio, 1171), the Madhyiinta-vibhiiga (Nanj10, 1248), all of Vasu­bandhu; the Dasa-bhilmi-siistra (now lost) of Ma1treya, the Trikaya sect10n in the Suvarnaprabhiisa, and the Wu slwng i sittra (Nanj10, 259), the latter five bemg the translations by Paramartha, A.D. 548-569. With these works we are first mtroduced to the systematized theories of Asanga and Vasubandhu, especially with regard to the three kayas and the eight t'1Jfiiinas

To sum up, the theory of the eight t''lJfiiinas is established out of the idea of the Buddha-gotra and the Tathagata-qarbha and with the development of the eight t'iJfianas the theories of the tri-kaya and the catur3fianas are completed If we examine the whole of the translated texts, we find no trace at all durmg the fourth century At the begin­ning of the fifth century, we notice some genns of them appearmg, but from the middle of the fifth century down to the begmnmg of the sixth century, we feel the ideas fully developed and ripened Since the systematization of the theor1es belongs to Asanga, and t>specially to Vasubandhu, their activity may be considered to cover the middle of the fifth century and after

KOYA.SAN UNIVERSITY

MAITREYA AS AN HISTORICAL PERSONAGE

Bv HAKUJU lTI

THE Bodhisattva Maitreya was probably an historical person at the begmnmg, and afterwards was believed to be the coming Buddha.

This belief seems to be pretty old, but the Maitreya here alluded to is the instructor of Asanga,especially in the doctrme of Yogacara,namely, ViJfianamatratii. We shall r!.'strict our discuss10n to this personage. For convenience' sake we sha]] use the name Maitreya without any title for the instructor of Asanga, and therefore a histoncal person; while we shall add the title Bodhisattva for the would-be Buddha, who may be considered a fabulous person I presume that scholars will generally be1ievc with me m the historical existence of l\Iaitreya, yet nobody from our side until to-day has set forth his opinion expressly on this point, and consequently the authorship of many a work as­signed to l\f aitreya, or to Asanga, is not qmtf' settled This is a ser10us question in the history of the development of Buddhism m India, which, if settled once for all. vall clear up a dark passage m the origin of ideah&m and its development and will enable us to fix the age of activity of Asanga and Vasubandhu

Generally speakmg, those who speak of the relat10m.h1p of ~anga with Maitreya are Asanga hnmelf and Yasubandhu Those who handed do"\\n such a tradition are three fin,t, Bodhiruc1, who came to Chma m .\ D .50{; and stated that Maitreya compo,1,c-d the• commentary on the rapacchethl,6 and the Bodlmaffl'a-bhi"om;-adhara, and taught them to A'ianga, secondly, Paramiirtha (499-509), who reached Chma m Ao 546 and compiled the L1Je of J'a.rnbandltu, and thirdly, Hmen tsang ((i00--664), who started from Chma m AD 6~9 and on his return home m A D 645 compiled his travel&, Sz yu /. i In Tibet there are trachtions much later than those handed down by these bavellers Accordmg to the Life of l'asubandlw, by Paramartha, Asanga was a native of Purusapura (Pe&h"\\ar), and JOmed the order m the Sarva­stiviida school Learning of the doctrmc of V 01d of the Hinayana, and bemg unable to comprehend it. he deternnne<l to kill hnnself Pmdola of the east V 1dc·ha came and mstructed him m it Not bemg satisfied with it, he went up to the Tus1ta heaven, where he "as mitiated mto the doctrme of V 01d of the Mahayana by l\lmtreya Afterwards he now and agam went up to the heaven to learn the l\Iahiiyana siitras, and on descendmg he preached them, but people would not believe

96 Hakuju Ui

him. He asked Maitreya himself to come down and preach the Law. Thereupon the latter made a descent on earth and lectured on the Sapta-dasa-bhumi (Yogiiciira-bhumi) for four months. During the lecture no one but Asanga could approach him, though all could hear him from afar. Asanga in the daytime repeated and explained the lecture of the night before. Then people began to believe the Mahayana. Asanga could through the power of meditation understand even that profound Avatamsaka-sutra. According to Hmen tsang, he joined the order m the Mahisasaka school, but was afterward converted to the Mahayana While in a sanghiiriima near Ayodhya, he often went up to the Tu~ita heaven at night and learned from Maitreya the Yogiiciira­bhumi, the Siitriilankiira, and the Madhyiinta-vibhiiga, and in the day­time he repeated and expounded them.

Taranatha's tales, mythological as they are, give similar traditions as to Maitreya's instruction, Asanga's lecture and his compilation of siistras Thei-.e three traditions agree in substance, making Asanga re­ceive instruction from Ma1treya But who 1s l\1a1treya, a teacher or a Bodhisattva? Naturally we had to presume that the tradition would contam an historical fact, and Maitreya was the actual teacher of Asanga. We shall now examine what we are told emanated from Maitreya

I

THE YOGA.CA.RA BIIUMI (NanJIO, 1170)

The Y ogiiciira-bhiimi, said to have been preached by Ma1treya, was translated by Hmen tsang, AD 648 This text was partially trans­lated m an earher period, as a separate siitra, as follows

(a) Bodhisattva-bhiimy-iidhiira siitra (N anJIO, 1086); translated by Dharmarak~a, A D. 414-418 (vol xxxv, chap 1, vols. xhx, I, chap 5, of the Y o_qiiciira-bhiimi).

(b) Bodhisattva-bhadra-sila siltra (N anjio, 1085), translated by Gunavarman, An. 431 (the same chapters as above of the Yogaciira bhumi)

(c) Bodhisattva-sila-karmaviica (N anj10, 1197), translated by Hiuen tsang A,D 649 (vols. xl, xli, xlu, chap 10 of the Yogiiciira-bhilmi).

(d) Bodhisattva-priitimolcsa (NanJIO, 1096); translated (1) by Dharmarak~a, AD 414-421, (2) by Hiuen tsang, An. 649 (the same chapters as above of the Yogiiciira-bhumi). ,

(e) Chie ting tsang (Nirnaya Sangraha), (NanJJO, 1235); translated by Paramartha, A.D. 550-560 (vols li-lvii, chaps. 1-7 of the Yogiiciira­bhiimi).

Maitreya as an Historical Personage 97

(j) Wan fa chang li (R<ija-dharma-ny<iya), (Nanjio, 1200); trans­lated by Hiuen tsang, A,D. 649 ( vols. lviii-lxi, chaps. 1-4 of the Y og<iciira­bhumi).

All these six are portions of the large Y ogacara-bhumi, and most of them are said to have been expounded by Ma1treya for Asanga.

Only one of these is said to have been composed by Maitreya. Hiuen tsang himself alludes to the Yogacara-bhumi in his S1 yu ki (vol. x), and says that it was composed by Ma1treya The internal evi­dence, too, indicates that the whole ·of the Y ogacara-bhiimi existed first, and extracts were drawn from 1t afterwards, thereby makmg them separate siltras In Asanga's Hsien yang shang chao lun (Nanpo, 1177), he says expressly that Ma1treya 1s the propounder of the Y ogacara­bhiim1, and adds "Formerly I, Asanga, heard it from him, and now settmg together the important pomts of the Bhumi will here illustrate the holy teaching " This shows again that this sii<Jtra 1s a com­pl'ndmm of the Y og<ic<ira-bhiimi propounded by Ma1treya The com­pendmm is Asanga's \\'Ork, while the or1gmal from which 1t was abridged could not have been his own Therefore the Y ogiiciira-blzilmi must be a work of Ma1treya, himself Both Vasubandhu's commentary on the Jfahiiyiina-sampar1graka and Jmaputra's commentary on the Yogaciira­bhiimi pra1M:· m thelf openmg verses Ma1treya as the expounder of the Y og<iriira-bhiimi Thus no one has t.'Ver regarded Asanga as the author of the whole or a part of that work The Y oqifriira-bhumi, one hundred volume1, m all, 1~ d1v1ded, the prmc1pal portion of 1t mto seventeen bhumis, and the rest mto three chapkrs concermng the Tripi(aka

Therefore the text 1s oftt'n called Sapta-dasa-bl11lnu-siltra, or siistra. Bhilv1veka called tht• VIJfiilna-matra philosophers by the name of Saptada~abhium bastrm, while Paramartha names the work itself Sapta-da.fo-blulmi-siltra The Bodlmatfra-bhiimy-iidlziira (K anJIO, 1086), a portion of th<.' l'o_qiiciira-bl11lm1, as we have seen, 1s th<.' most important of all the seventeen bhumis This text 1s quoted twice by Asanga in h11, commentary on the T'a;racchedzkii (KanJIO, 1167, 1208), which was agam commented on by Vasubandhu This clearly md1cates that the Bodhisatfra-bhiimi was known to Asanga and Yasubandhu and was m ex1t-.tence before them Thus we can safelv conclude that neither the Bodhisatfm-bhiimi nor the l"ogiic<i.ra-bhiimz 1s the work of Asanga They can only be the works of J\fa1treya. who transnutted them directly or indirectly to Asanga

98 Hakuju Ui

II

THE YoGA-VIBHAGA-SASTRA (now lost)

This text does not exist either in Sanskrit or in Tibetan or in Chi­nese, but from the quotations found in the Mahayiina-samparigraha (NanJlo, 1183, chap 4), the Abhidharma-sangiit' (Nanj10, 1197, chap. 6), and the 1'1Jfiapti-miitrata-.nddhi (NanJ10, 1197, chap. 9), we well know that it existed in those days The purport of the passages quoted is actually found in the Madhyanta-vibhiiga (NanJio, 1244, chaps. 1, 4, 5), which also belongs to l\faitreya Ab a similar passage exists also in the Y oga-11ibhaga section ( chap 6) of the Sandhi-nirmocana-siltra (Nanj10, 246, 247), we may assume that it was composed by Maitreya on the same basis as siltra

III

THE MAIIAYANA-SUTRALANKARA (Nanjio, 1190)

For this work we possess fortunately both Sanskrit and Chinese texts Accordmg to the catalogue Kai yuen lu (N anJIO, 1485, A D.

730), and the Chmese editions of Sung, Yuen, Mmg, and Kaoh, it was composed by Asanga Rylvam Levi, the editor of its Sanskrit text, probably followmg these traditions, attributed 1t to Asanga The mis­take was ongmated by the Kai yucn lu catalogue, and it comes as fol­lows Prabhakara mitra, who was a pupil of Silabhadra m Nalanda, and was well versed m the Yogacara, came to Chma, leanng his pupils, PraJiiiidhara varman and oth<·rs, behind, and was engaged m transla­tion of Sanskrit texts till AD 633 A translation of this text was fin­ished A D (}30-632, and presented to the Emperor Tai tsung m 633, in which year the translator died, aged 69 Hrnen tsang was m India, and Just m that year saw Silabha<lra, at the advanced ag<' of IOG, m Nalanda L1 pai ye wrote a preface to the translat10n by the Imperial order, and stated m it that it was compiled by the Bodhisattva Asanga. I do not thmk that he meant to say that Asanga was the actual author of the text He had probably md1cated by the word "compiled" that Asanga heard it from Ma1treya and afterward arranged what he had heard, as was generally believed Hrnen tsang, too, tells us exactly so. If so, the statement of the Kai yuen lu that Asanga " composed " it is an error on the part of its author Hm chao, a pupil of Hmen tsang, in the commentary on the 1"1Jfidna-mafrata, says expressly that the Kiirikii was composed by Maitreya and the commentary by Vasu­bandhu.

Maitreya as an Historical Personage 99

Hui chao (circa A.D. 690) is much anterior to the author of the Kai yuen lu (A.D. 730), and we can take it to have been derived from his teacher, Hiuen tsang. Sthiramati, in his Mahiiyana-avatiira (Nanjio, UM3), says toward the end " Thou shouldst know that the M ahiiyiina is the origmal teachmg of Buddha as it is expounded m the Siitralanlcara of Maitreya " Thus, accordmg to Sth1ramati, the text 1s by Maitreya, not by Asanga A s1m1lar tradit10n is handed down in the Dharma lak~ana school m Chma, though whether the authorship of the com­mentary belongs to Asanga or to Vasubandhu 1s still a question

IV

THE MADHYANTA-VIBHAGA (N anJio, 1245)

The Kiiril.ii of the J"ibhaga was composed by l\fa1treya, while the commentary u, by V a<;ubandhu This ex1~ted therefore, hke the other extracted texts, before A~anga and Vasubandhu Ko explanation is wanted a'> to its bemg first propounded by l\Iaitreya to Asanga, who in turn handed 1t down to Yasubandhu

V

THE VAJRACCHEDIK1-P1RAMITA-&ASTRA (KanJJO, 1167, 1168)

l-brng ~ays m the appendix to his translat10n (l't31) "Accordmg to a trad1t10n m India, Asanga received from :\faitreya the Kankii of 80 verses and Vasubandhu commented upon it " It i'l thus a work of Mmtreya Bodh1ruci (:\ D 308) attributes the l{cirika as well as the commentary to Va..,uhandhu, "bile I-bmg (,\ n 711) assigm, only the conunentary to Vasubandhu I-tsmg puts m the teAt A5anga as the compo-,er of the Kank{i but this contradicts hi~ own statt'ment above quott·d So we ought not take the "ord ' composer ' m its strict meanmg

VI

THE ABIIISAMA y A-ALA11i"'K..lR-\.

Though it i~ not known m Chma, there exists both m Sanskrit and in Tibetan the Ablnsa1naya-ala11k<ira The Sanskrit text 1s found in Professor Takakusu's collect10n, besides Professor ·wog1hara's copy. The authorship evidently belongs to l\fa1treya (Kritir iyam Arya­Maitreya-niithasya), but few notice the colophon It 1s attached to the commencement of the 25,000 Prajna-p<iramitii, and was pomted

100 Hakuju Ui

out by Hara Prasad Sastri, who discussed the question of the historical character of Maitreya. Having obtained a hint from his article, I searched through all Chinese texts emanating from Maitreya. In the Tibetan traditions the Abhisamaya-alankiira is sometimes attributed to Asanga and at other times to Maitreya But the Tibetan catalogue assigns it definitely to Maitreya and makes it have no connection what­ever with Asanga

Besides the above six texts, there are two Tibetan texts attributed to Ma1treya, namely, the Dharma-dharmata-vibhanga and the Maha­yiina-uttara-tantra Vasubandhu wrote a commentary on the former, while Asanga wrote that on the latter Of these two a translation of the latter by Ratnamati, A D 508, is found m the Chinese Tripitaka (Nanjio, 1236), m which the Kiirikii 1s said to be by Ma1treya and the commentary by Asanga Fa tsang (died 699) m his commentary on the Wu cha pieh lun (Avisesa-sastra, Nan110, 1258), says

"Accordmg to an Indian tradition handed down by Devaprajfia, the translator of the sii.<lira the author of the ]f ahiiyana-uttara-tantra, is Saramati (Kien hm), who also wrote the Wu cha pieh lun Yen tse, a Korean pupil of Hiuen tsang, m hi~ commentary on the Sandhi­nirmocana (NanJIO, 247), speaks of Saramab as the author of the Uttara-tantra."

On this point this commentary 1s quite different from the Tibetan tradition, in which the above two texts are classed among the five dharmas of Maitreya

I have now taken up almost all the texts that are connected with Maitreya, and shown that seven of them are by the hand of Maitreya himself.

If he has so many works to his credit there can be no doubt what­ever as to his historical existence Moreover, Asanga never quotes any other siistras than those of Maitreya, quotes by name the Bo­dhisattva-bhumi ( a portion of the Y og<iciira-bkilmi), the Y oga-vibhiiga, the Mahayana-sutra-alankara, and the Madhyanta-vibhaga, in his work, Mahiiyiina-sampari,qraha (Nanjio, 1183, 1184), and uses them as the foundation of his own discussion. He would not use his own work as a basis of his own theory.

Accordingly the Y ogiicara-bhumi and other works above referred to were not composed for the first time by Asanga, but had existed as separate works before Asanga and Vasubandhu. Maitreya, a worthy human philosopher, wrote several works on his own idealism, and taught Asanga probably personally. Asanga further systematized his own philosophy on the basis of his teacher's instruction. This being

Maitreya aa an Hutorical Peraonage 101

the case, it would not be very difficult, especially in India, to derive his doctrine from the Tu~ita heaven, where the Bodhisattva Maitreya is believed to live, thus identifying Ma1treya with the coming Buddha. The idea of the heaven-descended philosophy would add great weight to the Mahayanistic faith, for the faJth m Maitreya in the Tu~ita heaven was already established both in India and in Chma. Tao an (A.D. 314-385) is said to have had such belief, and hoped for a birth m the Tu~ita heaven

Though Asanga's authorship of these works has been shifted to Maitreya, it would by no means lessen the greatness of Asanga, who remains still the author of eleven works of importance. His figure will come all the more to the front, for his position as an idealistic philosopher 1s thereby clearly defined, from his predecessor Maitreya and his successor Vasubandhu. The lineage of the Vijfiana-matra school would be thus without any fabulous element. first, Maitreya, second, Asanga, third, Vasubandhu - with these three generat10ns the systematization of the idealistic school came into completion.

As to the approximate date that can be proposed for Maitreya, the dates of translat10n give m, a clue The Bodhisatfra-bhumi was translated by Dharmaraksa. An 414-418. We can reasonably assume that the Y ogiic<ira-bhiim1 was m existence before A n 400 Sth1ramab 's Mah<iy<ina-avatara, which quotes Ma1treya's Siitr<ilankiira, was trans­lated, according to one account, by Tao ta1, A n. 397-439, but accord­ing to another, by Tan Yao, A D. 462 Tao tai went to the west of the Onion range and obtamed the text of the M ahiii,ibh<isa, which he 1s said to have translated with Buddhavarrnan, A n. 437-439 But a preface by Tao tmg puts the time of translation as A n 425-427. The Mahayana avatdra, then, must have been translated after 1t, and this falls probably in An 427-437, but the text must have been obtamed before his return home, about An 425 ,Ye can assume that the Y ogiicara bhumi and the Siitra-alankiira had existed before A n 350. From these facts we can fix the termwus ad quem for the date of Mai­treya at AD 350

illPERIAL TOHOKU lTNIVER.'>lTY.

102 Hakuju Ui

NOTE

Professor Ui's date is too early. The date of translation is A.D. 414 and after, and the compilation of the origmal is not necessarily to be put back so far as 64 years Even 1f it be reasonable to put 1t back so far, 1t is only the terminus ad quem for the two works, not for Maitreya. Hereafter Professor U1 goes on to discuss the dates of Asanga and Vasubandhu His dates are as follows Ma1treya, 270-350, Asanga, 310-390, Vasubandhu, 320-400; about 100 years earlier than mme. He acknowledges 150 years between Vasubandhu and Dharmapala, instead of the 200 years of Peri If there be an error of calculat10n, 1t must he m the length of this mterval See the hsts of teachers above

J. TAKAKUSU

THE WELL - TO ET Dr C. A F RHYS DAVIDS

My teacher Croom Robertson was one who often walked hand in hand with pain. And he would say, when lecturing in ethics on

happiness - say it too with a wry quarter-s1mle as one of 'them who know' - 'Some say, why look ahead to pleai;;ure or a neutral object? It is sufficient to be wanting riddance of pam We can resolve to do without positive pleasure, but we cannot live with pain. Much of our action is to av01d pam, and no ·'calculus" 1s necessary here.' 1

This 1s to word 'the well' negatively. For pleasure, happmess 1s but the feelmg accompanymg the state of bemg well When we are well, the body 1s 1n a pleasureful state, the mmd, the will, is m a happy state, aff am, our little world, are m a prosperous state But there is the 'man,' who 1s neither tho'ie mstruments nor that little world, the 'man' of whom tho'ie are adJuncts, vehicles, and these the occas10n, the arena, the medmm of self-expresswn, self-direction It is the very man, the very 'he' of whom we can mo!'it truly say, he 1s well, he 1s better, he is unwell, he seeks to be well, he v:zlls the well

Now Buddhism, m its old Pah !'!Criptures, worded To tV very largely in that negative way Its s;pintual hygiene 1s mamly taught in nega­tive terms That 1ts code of moral fundamentals wa!'i negative 1s not a d1stmctive feature Even to-day we do not word a Christian code of them m accordance with the teachmg of Jesus: 'Ward thy fellow­man a!. thyself Ward his property as thme own "ard a woman's husband as brotht>r, a man's wife as sister and mother, ward the young as brethrm Let speech be truthful, km<l. courteous, m,eful Keep sober' But the Buddhist U!'ie of the negative term for the thmg!, mak­ing for man's 'Mll' goes far beyond the 'silas' Here are some, all of which are leadmg terms m the holy hfe

The goal, the summum bonum is l\7zbb<ina, rmsarana, (bham-)ru­rodha, auwta, khaya, anuppiida, nmuttz, ak·uppa, ak utobha.lfa, The way l!, myy<inil,a, nekkhamma, palui11a, mbb1da, alobha, adosa, amoha, avyiipiidn, wwbl11JJlui, a111111tfta, suiiiiata, appanilnta The saints are mbbuta, <rnasam, khin6.sara, without chanda, panpunnasankappa, kata­kararfi,ya, brahmManyavussit<want, pa11nadha1a, ohitabhiira, m,ttanha, and more might be added All might be com,1dered as summed up in

1 Mind, January, 1893

104 Caroline A. F. Rhys Davids

the phrase samma dukkhass' anlakiriyaya - 'for the perfect making an end of pain.' Once or twice the founder is recorded as summing up his teaching, and once it is in these words· 'Both in the past and now do I, even I, declare just this pain and the destroying of pain.' 1 And that which he was said to have beaten out under the Bo-tree was the making-to-cease (nirodha) the coming-to-be (samudaya)

Earnest disciples were taught that endmg of ill could only be end­ing of becoming (bhaM). The word 'life' was neither appreciated nor depreciated, 1t was not used doctrinally The thing to make cease was the being reborn and the being redead. These were the milestones of the woes of all the worlds And the man who has spmtually 're­jected, cut-down at the root, the body' of his desires and craving, 'made it hke the stump of a palm tree, made 1t something that has ceased 'to become, so that 1t cannot grow up again in the future' - he alone is happy. So are the Arahants happy- not because before them lay well-warded the way of the worlds toward the goal- in that it was the way of and to the d1vme Well-Willer Safety ahead was merely the outlook of the convert, the entrant, the First Path wayfarer His was the slogan. Khinanirayo 'mh1: perished for me is purgatory' 2 not that of the samt Khinii Jiih perished 1s birth! The samt had done with the Four Ways, he had 'cro5.sed over' His was the happmess of Lucretms's coast-spectator He was safe, but his view was seaward at what he had eome through. He was not lookmg landward at what now lay before him. That 'before' was as 'the track of bird m air un­traceable ' 3

This that doth ne'er grow old, that d1eth not, This never agemg never-dymg Path -No sorrow cometh there, no enemies, Nor 1s there any crowd, none famt or fail. No fear cometh, nor aught that doth torment 4

To th1~. the Path Ambrosial have gone - Full many

He was in K1rval)a; of the 'beyond' he did but say, it was utter Nirvai;ia parinibbiina For

Nowhert> 1s measure for one gone to obhv10n That whereby we speak of him - that exists no longer Wholly cut off are all forms of our knowmg, Cut off the channels of speech, every one 6

1 Manh 1, 140 2 Sarpyutta, u, 70, etc 3 Dhp ver 92, Theraglitha, ver 92 4 Therigatha, ver 512. 6 Sutta-N1pata, ver 1076

The Well - TO ET 105

His happiness was on the one hand so retrospective and on the other so barred from any forward view into the future, that it might have served Croom Robertson as a fit instance of a well-being in terms of riddance of pain It may have been with such 'as with one who after long toil and much peril reaches home, and is content with that for the day, whatever hfe may yet give or ask for on the morrow. They had won up out of the maelstrom of samsara . . to something ineffable, that now is, but is not to be described m terms of space or aftertime; and resting they sang We will leave it at that.' 1

It will be said This is the 'well' held up m the teaching for the few, for those whose faces were set toward the highest, for those who, after ages of slow maturing. were near maturity. Of these I have said quoting Emerson 'of immortality the well soul is incurious He is so well that he is sure 1t will be well.' 2 But, the obJector goes on, the teach­ing for the many, for the believing layman 1s more positive, less austere, more human, more suited to those who have not turned their back on the world that they know, on life as they know it

This 1s true 3 No creed on earth may be said to sit so lightly and pleasantly over man's conduct and ritual observance as the Buddhist layman's sasana It may seem the creed of world-orphans It 1s so Yet is 1t less so than 1t seems He too seeks the r nseen "' ardmg, for is not the quas1-de1fied Teacher one ma chrome process of ever-wardmg Buddhas? ls not the Teachmg, m some way not understood but ac­cepted m faith, a world-gift to man for his salvation? Are not the holy almsmen, albeit very unequal as to holmess, an ever-present m­fluence wardmg off 111, producmg merit? The layman could afford to word 'the well' positively Worlds lay before him, but by a worthy life here, he could earn guarantees that they would be 'bright,' not 'dark' He could afford to <;peak of thmgs pleasant as pleasant, and not as m truth pamful 4 He could speak calmly of death, for it was not the end-all, but JUst the common lot

All this we know And if the worthy Buddhist layman is not al­ways consistently cheerful over the last-named matter, neither are we when we, hke hnn, find nothmg better to comfort ourselves withal Small blame to the inconsistency The blameworthy thmg for him and us 1s to find nothing better

1 PsalmJI of the Sisters, ver 51 l, 512. 2 PsalmJI of the JJrcthrcn, xlvm 3 De la Vallee Poussm empha~1zc~ tlus d1stmclion, so , 1tal to a Just appra1se­

ment of (early) Buddlm,m as a "hole I\' 1mina, l!)!:ZJ 4 Sukham sukhato, Sutt.a N1piita, ver 7.59, transl m Buddhu;t Psych, p 86

106 Caroline A. F. Rhys Davids

But there is one point about this layman's gospel that we forget to bring out It worded 'the man' more worthily than did the monk, the bhikshu There is no denial of 'the man,' no anatta, in such discourses to laymen or laywomen as that to Sigala, to the thirty kumaras (who were advised to seek 'the man'),1 to Visakha, to C1tta and other 'gamams ', to Anathapirn;hka There was no robbing the word man (puggala) of havmg any reality, of bemg a merely conventional label. There was no whittlmg down the word to 'a nammg of something that exists (body and mind) by something that does not exist ' 2 The lay­man was reminded that things are transient and that ills abound, but the third monkish slogan anatta - a word not of mondial import, but derived (and misused) from a local, a temporary protest in early Buddhism, was not brought into his gospel, to worry and undermme his conv1ct10n that there was n·ally and truly 'a man' who worked karma of thought, word and deed, and who reaped the harvest thereof here and hereafter

And with this more direct, less sophi:,ticated, worthier wordmg of 'the man,' the Buddhist lay-&asana, be it noted, worded more ration­ally 'the well' of the man than did the monk This lay m .,eekmg his good (htta) and that of others, his advantage (attlw, amsamw) and that of others (parannaii ea) both here and hert>after Before hnn lay the two 'gomgs,' the wc-11-bourne and the c-v1l-bourne (sugat1, duggati). Neither was eternal, though either probably lasted long Certamly the former, the sagqa lolw, d1<l And how far Nibbana - word as vague as our 'heaven' - was m any way du,tmct thcrefrom he did not a&k Kot till M1lmda made out a case of a <l1&tracted mankmd m suspense about 1t 3 do we come across any worrymg over the subJect

Kor had that other word for the monk's ideal, !"1mutfi, liberation, any charm for the layman Negative term though 1t be, 1t has come to appeal strongly to u&, who are the heirs and the record-readmg wit­nesses of ages of struggle:, for liberty, national, rehg1ou&, social Scarce any word thrills m, more &trongly than JU&t this word of riddance.

Libcrte, liberte chcrie' 4

It could tlmll the monk, the nun no less The po&1tlve thmg they had got ml of wa:, so fearfully pos1t1ve - awful even .ts fire in the turban,

1 Ymaya 1, 23 2 Vi1Jamanena aVIJJamanassa pandiplfd Ablndhammattha-sangaho1, vm, 14 3 l\Iln S~S That there v.as any worry m the snnply put query of SaIT1yutta, 1v,

251, 261, l'> not dearly nnphed 4 The 'l\larse1!1,use '

The Well-TO ET 107

fire in the house, fire in the jungle - that the very riddance itself stood for salvation, for peace, for 'the well.' The more usual Indian 'riddance-word' of salvation as purity (suddhi) does not reverberate in the early Dhamma as does the paean of liberty.

But it did not appeal to the Indian layman As a religious slogan it does not appear in the 'three' Vedas M oltsa is a later development, due either to the influence of Buddhism, or to that which caused Buddhism to dt>velop 1 That other rehgiorn, make little of it 1s deeply sigmficant, maybe, of their more po~itive outlook 2 It is true that Aryans m general can thrill to a negative shibboleth, witness amrta, ambrosia, immortal, but the idea of bemg spiritually set free 1s too unworldly to come to the front m any world save that of the recluse

Well then, we have put forward these two features in the gospel of the Buddhu,t recluse man 1s not worthily worded, man's well is not worthily worded The one and the other are n('gat1vely worded. The one is declared to be a fiction, the other 'is,' 3 but i1, entirely meff able.

And let the apologist of the dual gospel m Bud<lh1.'>m remember this It 11, the gobpel of the recluse which i!, and will b<' looked upon by people of other l.mds and other creeds as the original. the venerable, the gf'nume "Buddhi!,tn" Not because the layman's gospel is not very worthy, or some later devdopments no less so But the Pah canon holds the field yet as the archetype m Buddhist literature And in 1t the hfe, the welfare, the world, of the monk outwC'igh~ and dwarfs altogether the h fe, welfare and world of the 'manyfolk ' \V 1th monks as recorders, a.'> comp1lers. as editors, a~ 'libraries,' nothing else could well have been expected

There 1s yet another defect m wordmg, mbmately bound up with those other two defects But 1t I!, a feature m both gospds l\Ian wills his 'well ' But m Pah there is no fit appra1bmg of 'will' ~uch as our European Aryan tongues enable u.., to make There is the .'>1gmficant approach to 1t m classmg all man's self-expression - deed, word, thought - as activity, as work (kamma) But when the factors of that 'bclf' are analyzed, dynamic terms, approachmg 'will' m nwaning, fall mto the background, and no discernment 1s shown of this that to teach religion as a Path to a Goal to be trodden by ('ach man, we must, to malce it intelligible, show man as choosmg, as w1llmg to walk therein

1 We only find the notion, as developed, m the Svet and Mu1tr1 Up. 2 Cf Ency Rcltg10n and Ethics on its absence The artide Moksa was mserted

by special suggcst10n made to the editor a Min 270 atthi mbbanam •.•

108 Caroline A. F. Rhys Davids

But into this defect I have gone elsewhere. 1 Let us keep here to those first two defects

(1) In the tenet called anatia man is not worthily worded. The doctrine was m the first mstance a protest, not without reason, against what had become a distorted emphasis in the brahmanic teaching. The 'man,' namely, worded more usually as self or soul, was in fixity, immutability, d1vimty, identical with the world-soul. Hence he had, not to grow, to werden, to become, he had to come-to-realize Hence he was not so much a growing plant, as a Jewel or star to be cleared of all that hid or dimmed.

But in time this doctrme of protest degenerated into the harmful dogma, never worthily reasoned out, 2 that the 'man' is, not only not immutably divine, but non-existent; that there 1s no one who thinks, speaks, acts, but that there 1s only thinking, <;peak.mg, domg It 1s not just to say, as do some apologists, that this demal of the man was part of the early protest I venture to hold that the early protest was, after a wavermg start, the pos1t10n taken by the founder 3 witlwut the denial. But there can be no doubt about the much more n<'gative attitude that grew up among the after-men for one who reads, m a historic perspec­tive, what they came to say

We must read both what they said and between the Imes of what they said Pah literature is for us still a very new study There is too much takmg up or reJecting en bloc The evolution of the brahmm as an ammate book, the evolution of the mantras he handed on as such -we cannot get at the base and back of these But the evolution of the Pali canon, the evolution in its animate libraries - this 1s of more recent growth l\famly we can only surmise, yet we can get nearer to the conditions under which the phenomena of Order and of P1 takas came to be and to grow And we can be more d1scernmg accordmgly

We can discern, m the stereotyped, madequate, 1ll-fittmg anatta riposte,4 somethmg like an extmct coal, an archaic corpse It belongs to the early protest, but it 1s applied to the later demal of the 'man.' It is no more alive, it cannot meet the query of the puzzled hstener, 6

as Gotama would have met 1t had he indeed been the teacher on that occas10n We can discern, how needful it became, to the mamtenance

1 Bulletm of London School of Oriental Studies, 1926 Buddhist Studies, ed by B C Law, 1926

2 There 1s an almost Humean approach to such an attempt m Sa111yutta, m, 230 3 Cf Vm, 1, p 23 with p 13, MaJjh, 1, 232, Samy , m, 66 4 Cf among many repetitions, e g , Sarpy , m, 94 b Cf ibul, p 88 MaJJh, 111, 19f.

The Well-TO ET 109

of this unworthy wording of the man, to buttress it about and around with every stone of support that could be brought together. 1 We can discern in Buddhaghosa's Commentaries how this succeeded, how argument had died and dogma stood firm. We see the man, who is central in the early gospel as the way-farer, 'you and I' 'faring on, runnmg on' from world to world, the man who to body and mmd is as the forest to the faggots, borne thence to the fire, the man who grows or sICkens m his karma and who stands after death before Y ama (his fellow-man, as were all devas) to be confronted with that karma and judged by it, 2 this man, the very Thou, neither body nor mmd - we see him, m those commentaries, whittled away mto a momentary com­plex of five khandhas We see the current word for hu, personality· attabhiiva, • self-state,' treated as a mere concession to the conventions of the multitude, the 'many-folk ' 3

(2) And m the tenet called nibbana, nibbiinadhatu, or asa khata dhatu, we see · the well,' the surnrnurn bonurn unworthily worded How­ever the word be mterprete<l, it remams a negation and mdicates a tendency away from the positive, a tendency to which so many other Buddhist terms bear witness We may, we should appreciate the re­serve m early Buddhi~m which refused to commit itself to any descrip­tion of that Goal 'Who, at this tnne of day, 1s content with the worth of such descnpt10ns m other scnpturcs;i But 1t does not follow that the meffable 1s fitly worded only by a negation We may rest too easily complacent m the contentment ~hown by the Indian over his neti, neti! It hmts at that racial weakness which found no word for 'will,' whence we derived our 'well,' which never developed its Aryan root war m the wav we developed the twm root wal

In other word~. what man wilh, strongly, he tends to name posi­tively. He does not cry 'No-land'' when he ~1ghts the longed-for sea. 1 hat contentment with the notion of riddance rather than with the notion of what we were to put m place of the thmgb got rid of - that contentment said 'not-diseased,' aroga, arogya, where our stronger positive words say irylHa., .rnlus, health, .rnnte And 1t found a com­plement to that negation about man's body m a negation about the very man, soul or self, when utterly well In n1bb<ina 1s implied ex­tmct10n of the heat of spmtual fevers Even the slightly more positive sitibhuva, the coolness of the saint, does but md1cate riddance of fever. There is no fit word for the state that is won. 'Growth' we meet with

l Esp Ill Katha Vatthu, I, cf Bud Psychology, mu, eh XIII. ~ Devadiita Sutta (M, m, A, 1). 8 Asl. 808, cf Vis. M 810.

110 Caroline A. F. Rhys Davida

here and there; the notable causative of werden, bhavand, we also meet with But when the long work in all that process is consummated, when the man becomes that which he has willed to be, - 'well,' - for this word, daughter of will, we find no name. We do not even find a forced use (as m the Greek ro tv, and my 'the well') of su, to wit au­bhava. Sotthibha11a does occur, but very rarely, and not with emphasis. Arahatta is weakened to 'worthiness to receive off ermgs' !

Men cannot eviscerate religion of so much as did Theravada Bud­dhism and yet preserve the kernel. That 'will' was badly worded was a national defect which Buddhism did what it could to remedy But unworthy wordmg of 'the man' and of his goal was a mistake of its own makmg In the far East more positive conceptions somewhat remedied matters But in Theravada Buddhism the double negation has survived And certain features 1t presents to-day are deepened m s1gmficance when we consider them as possible consequences of that survival For is. not this true, that if 'the man' be unworthily worded, the will by which he seeks the ultimate 'well' will certamly be un­worthily worded, and that 'well' will be worded no better, will be practically blotted out;,

Consider' Theravada Buddhism negated man, the willer through body and will, and negated any concept of 'well' as willed by a Will er - source, worker, end- of the man To-day we w1tm•s& how 1t has gone on losing substance as a rel1gion, how 1t has become practically a system of ethics on the one hand and, on the other, a system wardmg a dead world- dead !>Ocial cleavage, dead language, dead literature

For the way of the worlds, the larger hf e of both the s.een and the unseen has faded out of its perspective Its ancient gospel threw open the gates to the long v1shl!>.

Apiirutii tesam amatassa d11iirii!

Around and about its votaries, devas, no longer deities, but fellow­men of other worlds, stood wardmg, watchmg, adv1smg, pra1smg, re­provmg Clairvoyance, clairaud1ence - to see and hear as could those devas - were not banned as dev1hsh, hut were welcomed by it as means of access to fellowmen here and over there Man was shown as wayfarer m a Way, a way not only of present 'best' livmg, but a four­staged way leadmg to the goal

All of these opportumties and vistas - all that made his cult a very living religion. that is, the havmg heed to the unseen - the Thera­vadin has virtually laid on the shelf among the venerable things of his

The Well-TO ET 111

past. His attention, as layman, is concentrated on this one only of his many lives. The Just-so-much as was given him of the way of the worlds in which we are wayfarers he has lost, and he has learnt nothing since wherewith to word it more worth1ly, more truly.

Or, if he be monk, his forward view is otherwise blurred. For as monk, he has ever worded not only 'man' and his 'goal' unworthily, but 'life' as well He never had any hope of the hf e of the worlds, for everywhere that meant rebirth, redeath of the body And body, he held, was, with mind, essentially 'the man ' Hence birth and death stood out over-large, over-fearful He believed m growth only when the life had been cut off from the general hfe of the worlds But surely the growth of 'the man,' 1s not the way of the growth of any of his bodies. It 1s the slow advance toward that immortal adolescence, m the consciousness of which growth our worthiest septuagenarians will say. 'I would not exchange my seventy-five for your twenty-five''

l\J ay the faithful and kmd scholar-friend, to whom we herewith will well of youth eternal, gladly echo Lord Haldane's recent birthday saymg 1

Life, man, will, well· herem was weakness, herem lay a falling be­hind In all four words we have worthier beacon-lights than early Buddhism had How are we letting them shme?

What are we domg with our word 'hfe '? We feel after the life of the racP, but we measure the whole hfe of 'the man' by this one little earth-span and its body "ith that hodv we grow old, and the will, compliant servant, age:. with it when 'we' are not even 'grown up.'

What are we domg with our word 'man'? In the school and acad­emy we have thrust him out, rPplacmg him with his mstruments, body and mmd, mea~urmg his growth by these. In the churches, in the world we speak of him as 'havmg' a soul, or not, as if he were some­thing else And when we bury body, we call 1t hun 1

What are we domg with the word 'will'? We have put it, m the school, on the shelf Or we have screwed it down to mean conscious resolve We refuse to admit that man-as-actmg many way whatever is usmg w11l, we do not discern that man ·s only way to 'the well' lies in will For will is commg to be, not only commg to do

What have we done with the word 'well'? We have not, even, with our wiser European neighbors, put it on its own feet as noun ,,·e buttress it up with affixes well-being, wel-fare Then, taking from the meanmg where we add to the letter, we he it down to the body, we tie it down to earth-life, we tie it down to race-betterment. But this is

112 Caroline A. F. Rhys Davida

not so unintelligible a cosmos that we must see in only a stage of race­betterment the uttermost, perfected well of 'the man ' To do this is to reason, believe, hope with our forward view limited to earth. The man, 1t 1s true, can only grow towards the immortal youth of his Well by work for the betterment of men But in all men is 'the man.' And in the perfected well of each man and the way thereto bes the welfare of men.

CHIPSTEAD, SURREY.

BUDDHISTIC MYSTICISM

BY E WASHBURN HOPKINS

IN considering the question of Buddh1stic mysticism it will be well to clear the ground by a preliminary exammation of two factors

which appear to be of mystical character But these factors, of which both form essential parts of saintly trammg and one at least 1s indis­pensable to perfection, imply m their turn a belief m Karma, that is, in the meluctable recurrence of trarn,m1gration as conditioned by a severe ethical standard, a standard apphed not only to physical acts but to thought& and ideas, though, to the Buddhist, thought itself 1s physical and, so to speak, a thmg

But there is nothing mystical m the doctrme of Karma itself. It is a combmation of popular belief m metempsychosis, adapted to a sy..,tem that reeogmzes no p..,yche (for which rea.,on the term metempsychosis should be reserved for Brahmam..,m, while the Buddh1..,t notion is better expre&sed by transnugrat10n or relmth) and the equally popular pre-Buddlust1e belief m a theory of future rewards and pumo;hments, adapted to a system that recogrnzes no d1vme Judge of morality com­petent to a.!>.',Jgn rt'war<ls and pum.!>hmenh, For general u..,e m the Bud<lh1stic Brotherhood Karma became an automatic etl11eal force of nature and apparently 1t was in this form an unque&tion<.'d dogma The 1mphcat1on of or<lmary ethics I.', mdee<l greatly refined in the case of th(• samt or Worthy (Arahat) who ''pa..,..,es beyond good and evil", but it 1s by no means M.·t a1,1de, for even the mo1,t exalted sdmt 1s still under the mfluence of an etl11cal Karma, though 1t mw,t be remembered that to the Bud<llust, as to the Hmdu gener,tlly, wi~<lom and knowledge are moral qualities But mysteriou~ a!-> appear~ to u~ the whole Karma process and mcred1ble as it seems that ..,uch a moral power of nature should have been accepted without <l11,cus:.,1on and even made an im­perative article of faith (for faith m Karma and its workmgs is as ex­plicitly demanded of the Bud<lhi1,t as 15 faith m Buddha), there is yet no more mystic1&m m the behef than there is m the Brahmamc assump­tion of God and soul But smce the machmery of Karma affects the d1scuss10n of my..,tic1..,m, there remains the nnportant pomt as to whether Karma was thus accepted as an article of faith from the very begmnmg, that 1s in Buddha's own formulat10n of his creed.

It will probably be conceded that the early discussions and dia­logues called Suttas are committed to the behef in Karma and that the

114 Edward Washburn Hopkins

later literature is steeped in the same theory, ostentatiously promul­gated in the Jatakas (stories of rebirth as conditioned by previous morals), the Cariya Pitaka, Peta-vatthu, and so forth; but it has been pointed out that in the most ancient exposition of original Buddhism, that is, m the statement of the Four Noble Truths, there is no explicit mention of Karma and that in the certainly antique theory of the scheme of causation (the Chain or Wheel) there 1s the same silence. From this bas been drawn the rash conclusion that Karma was a theory added on to Buddhism as expounded by Buddha and was not an essen­tial of the founder's own !>ystem.

From that conclusion a careful examination of the facts must com­pel one to dissent The earliC'st texts we possess contain, in the same form as was accepted later, the doctrine of the Four Noble Truths and scheme of causat10n and, if the texts are accepted as they now stand, the scheme of causation is as prmutive as the Four Noble Truths, al­though ,t would still be possible to argu<' that it was borrowed and adapted from Sankhya philosophy by Buddha bims<·lf. The theory of ('ausat10n explained m this sch<'m<' is, however, absolutely incompre­hen!>ible. it is even meanmgles!>, without the implication of etlucal transnugrat10n. All attempts made to fit 1t to a single hfe or to !>Ce m it a co1,mic analysis have failed utterly The only po!>sible <·xplanation of the schedule, wluch begins with ignorance and ends with birth and death, 1s that it mvolvcs the p:t!>!>mg over of certam element!> of one existence (cau!>e) mto another (effect). or, mother words, the doctrme of rebirth ·with that doctrm<' 1t become& at lea!>t, faulty as it is, a would-be logical ser1e1, leading from one hf e to anothn by a thread of persistent unetlncal ignorance

But no less clearly do the Four Noble Truths imply belief m Karma, even if the theory be not mentioned. Buth and life are suffermg, suffermg sprmgs from cravmg, extmct10n of cravmg cause!> extmct10n of the suffermg entailed by birth and life, this extmchon is attamed by following the Eightfold Path, of right views and so forth If this were a solution of the problem how to hve one morally excellent life, "birth" nught well have been omitted from the list of woes which the disciple (already born) is here schooled to evade But, qmte apart from this, the whole trend and meanmg of the Four Noble Truths are ex­plicitly to prevent the repetition in a future hfe of the su.ffermg entailed by cravmg m the present. The existence of an otherwise unendmg Samsara or round of births is the reason given for the scheme when 1t is promulgated and this reason cannot logically or historically be dis­joined from it. The trance, m which, incidentally, one may see one's

Buddhiatic Mysticism 115

"Previous births," implies the words of Buddha when, envisaging his own demise, he declares that they may be uttered by everyone who holds the Mirror of Truth: "Hell 1s destroyed for me and rebirth ... I am no longer liable to be reborn in a state of suffering " And equally cogent is the statement made by Buddha on proclaiming the Four Noble Truths and in closest connection with them. "It is through not understanding and grasping the Four Noble Truths that we have had to run so long, to wander <,o long in the weary path of tram,migration, both you and I. . . By not seemg the Four Noble Truths as they really are, long 1s the path traversed through many a b1rth When these are grasped, the cause of rebirth 1s removed, the root of suffermg is uprooted and there 1s no more rebirth" (MPS DN n 91) Again, in the Dhammacakkappavattana, the origin of 6uff ermg 1s given as craving for pleasure and hfe, which causes "renewal of existence" and the ariyo atthangiko maggo or Eightfold Path (called Aryan or Noble) is expn•s1,ly mvcnted m order to destroy the cravmg which causes re­newed existence. It 11, impossible to cut tht'se tt'xb, to pieces, take out somethmg sub1ect1vely regarded as "older" and re1ect as '' later added" whate\'er does not 1,mt the theory that Karma H, an after-thought of Buddh1i>m. Without Karma there would have been no Buddh11,bc "plan of 1,alvat10n "

In that plan, morality 1s md11,pensable, but only a .. a prelnninary to the age-long effort to C'ra<licate vital nw,ery altogether, not thC' misery of one life· but of countlei>s hws m repeated rebirths To Buddha sal­vation was the e'leape from new e:\.1stence as ammal, man, or godlmg, for the fatt> of i;,p1rib and gods 1s to pa1,s away and <lie like human be­mg~. Their ~upreme happmcsi> on dymg 1s to be reborn as men, that they may become <l1~c1ples of Buddha and i>O be saved from all future ex1-.tence At the core of all tl111, lies the conviction that the suffermg of hfe entailed by cravmg (evil cravmg, usually, but 1,ometimes moral cravmg, a& for chastlty), 1 11,, to the i>age, not the mere unease or dis­comfort typified by "not gettmg what one des1res '' and the d11>tress of poverty or d1~ea1,e, but the ab~ence of any 1,ettled condition, of any ab1dmg resting-place "Impermanent are all tlw bemg1, that come mto bemg, for masmuch as they arise they must pass away " There 1s no per~1stent entity underlymg phenomena, nothmg has 1,ubstance. To cease from nnpermanent bemg, from the mccsbant becommg~ typi­fied by rebirths, tlrn, 1s the salvation preached from the begmnmg by Buddha.

I Jt1v, 54.

116 Edward Washburn HopkiM

In the course of the training necessary to the attainment of this salvation, the would-be samt or Worthy acquires (this also apparently is the belief of Buddha himself) certain mysterious "accomplishments" and these are the factors referred to above as not really mystical. They are the powers known as lddhis and those involved in the trances or raptures called Jhiinas. The first might well be called magical powers or even divine, since they are expressly said to be the same powers that are exercised by fairies, gods, and so forth, and they are supernatural in that they are above the natural powers of ordmary man, though natural to gods and attainable by superior men. They are not powers restricted among men to the Buddhists, whose saintlmess or Worthi­ness endow& them with these faculties, but they are practised, though not &o capably, by other religious wizards, who also m part command this "Angelic wi1,dom " 1 Buddhist adepts can fly through the air, go through earth, on water, turn themselves mto other shapes, enter an­other's body, and so forth Yet these powers are attamable in their highest form only by Buddhistic trainmg m will, effort, thought, and inve1,tigat1on, through a process of "thmking, developmg, practising and accumulating" the power involved, that of mmd over matter. There must be, to gam this end, earnest and cogmt meditation and a constant "struggle agamst evil " In other words, we have here a re­finement on the common superstit10n ari1,mg from exaggerated belief in the mind's control of the body, the super&tition that a man of great spmtual (psychic) power must neces1,arily po&sess power over matter, usually exploited by a show of phy1,1cal feats Tlrn, bd1ef was current among Brahmans and Bud<llrn,ts, a1, 1t was a commonplace of European belief that a wizard could ride the air and a great samt could do super­natural deeds Later Buddhism disapproved of the pubhc exercise of the&e powers, even if properly gamed ""hen us<'<l, they tend to make the user vam, they arc valueless except for occas10nal opportumties to impre~s the multitude with rehg1ous marvels (when a Buddha might profitably show off his accomphshment1,), or for the strengthenmg of one's own mental growth Like the Eightfold Path they are a proper subject of med1tat10n.2 In the clo1,mg sct>ne of Buddha's life it 1s inti­mated that he might have utilized the ability to prolong his own earthly existence, if he had been requested to do 1,0 in good time Probably the popular belief was that &uch accompfo,hments were not so much attamed by conscious effort as they were the habitual concom1tants of superior wisdom or saintlmess, just as the ideal ~piritual kmg possesses

1 Gandhabba-vIJJii, e g , Thera G , IX, Then G , vs 232, and Kcvaddha S. 2 Thera G , vs 595

Buddhiatic Mysticism 117

"accomplishments,, essential to his nature, though these are but the lddhis of attractiveness, longevity, good health, and universal popu­larity. In this and other cases, noted by Rhys Davids, the Iddhis are not supernatural; but ordinarily they are powers above the range of common men and shared with gods. As such, m later works, they are joined with the supernatural powers of the "d1vme ear," of "knowing the thoughts of others," of the remembrance of previous births, and of the "d1vme eye," as one group under the designat10n "superior kn~wl~dges," Abhiiiiias, a group attamable by any deeply relig10us and mentally competent Buddhist (who, at the same time, extirpates all smful taints) Of one such we read that he became anx10us as to his religious qualifications and "strove and wrestled" until he acquired the Abh1fifiiis Another Brother, by virtue of the same powers that he had acquired, picked up two men as they fell from a tree and set them down unharmed. By his lddh1, Buddha made a boy actually present disappear and then, "by withdrawmg Iddhi," made him reappear 1

Such powers were regarded as the natural outcome of the religious hfe. Thus l\fatanga's 1,on 2 took orders under Buddha and "seemg the powers wielded by the Brethren he aspired to the same and by prac­tising exercises won the Abh1fifiiis." It is only in very unusual cases that the exercises can be dispensed with Sumana. "a child of seven but of ripe insight" got the powers, so that he could fly through the air cros1,-legged, as Buddha flies, or at will crosses water "by Iddhi " 3

Moggalliina, one of Buddha's chief disciples, was at first averse from mental toil, mclined to take the recommendation of Mlt'nt meditation as excuse for sloth, but bemg sharply reproved by Buddha with the caustic remark that "lazme<;s 1s not synonymous with thoughtfulness," he devoted himself to earnest thought and exerc1se1,, so that in the end he was pronounced "foremo1,t m I<ldh1" by Buddha himself. He could "create hvmg shapes," that Is, transform himself mto other 1,hapes,4

and he had "visions of the future and of all the present world " 5 hke a god's, as Anuru<ldha, who by Iddh1 could remember fourteen prece­dent births, says of his own power, "My v11,ion, hke a god's. is clari­fied." He then identifies the1,e powers, exemplified by the divine eye, with those of the trance m which one 1,ees beyond mortal ken.

I know the destm1es of other hves, "'hence bemgs come and whither they will go, Life here below or other-where of hfe­Steadfast and rapt m fivefold trances sunk 8

1 Thera G, clxu and ccxxv11 2 Ibui, clxx1v 3 lbul, ccxix and vs 1104 4 lbul, vs 901 (the "body wrought of mmd") 6 lbuJ,, vs 1183 6 Ibul,, vss 916, seq

118 Edward Washburn Hopkins

The Suttas give the "six supreme knowledges" as follows: 1. Be­ing one he becomes many, becomes invisible, goes without obstacle through a wall, through solid ground, on water, in the sky, touches sun and moon, reaches to the heaven of Brahma. [These are the usual eight or nine Iddhis ] 2. With the divine ear purified and surpassing that of men he hears sounds heavenly and human, far and near. [So by his "ear divine" Buddha hears a conversation miles away ] 3. He un­derstands with his mind the minds of other persons and beings and knows the passionate, the bound, the freed, as such. 4. He recalls in all details the various temporary states m which he lived in preceding births. 5 With the divme <'ye he d1scerm, beings farmg according to their deedf. 6. He lives in the attainment and reahzat10n of sane free­dom of lwart and mmd. These six Abhififias should be realized by the Worthy, as 1s said, for example, m the Dasuttara Suttanta.

No check is put upon the exercise of such powers m so far as they subserve the cause of traming and are helpful to oneself or to others. The eye d1vme must itself be purified to giveth<' VIbIOn of other worlds and of bemgs not born of parents (Payas1 S ) , as Buddha wlwn dying saw spirits mnumerable crowdmg so close before lum that many occu­pied the "space of a hair-tip " For edification Buddha lll the Patika Sutta performs wonderb, but he knows the deadly peril of .'>Uch exhi­bitions in th<' case of ordinary men and so 1s made to say "I see danger in the practice of these accomphshments, I loathe and abhor and am ashamed of them" (Kevagdha S ) , the true " accompli.'>hment [he add.'>] 1s that of self-mastery " In tlw same vem, the Vmaya lay,<, 1t down as a law of the order that a Brother "bhould not d11,play before the laity the wonders of Iddh1 1,urpassmg the power of ordmary men; to do so 1& to be guilty of an offens(• " Perhaps, as the late master of Buddhism, Rhys David.'>, suggc1>ted in connect10n with this pa5bage, the feelmg against the use of wonder-power/> wa-, only gradually brought to a pomt where the practice was forbidden He nught have instanced the parallel m Brahmamsm, where it u, only the later Yogi who 1s exhorted not to make a display of his btmilar powers "To posse1,i, the power 1s noble; to show off i.<, ignoble "

The "d1vme eye," instead of being acquired by effort, may in ex­traordinary cases be an mnate power ref>ultmg from Karma Thus m the first Buddha, V1passi, whose name suggests vis10n, is said to have been mamfcf.ted the d1vme eye at birth, for it was "born of the result of his Karma, by which he could see as far as a league by day and by mght" (Mahapadana S ) . This is not, as compared with other exam­ples, a very remarkable power, but 1t 1s noteworthy in that it 1s obv1-

Buddhistic Mysticism 119

ously merely a physical strong sight that is here lauded, whereas usu­ally the divine eye is mental, connoting clairvoyance, and is thus dis­tinguished from the "third eye," which may be rendered "insight." As early as the Upanishads there appears the notion of the" eye turned inward," of which perfected msight 1s the full capacity and salvation is the result of using it. This corresponds to the Buddh1&bc "eye of wis­dom": "Knowledge is born as the eye of wisdom and by the attain­ment of this eye one 1& freed from misery " 1 With the divine eye one "discerns the pageant of bemgs farmg according to their deeds," but with the eye of \'\-tsdom one acqmrcs the final absolute knowledge of those Worthies who

Discern "'hat 1s from whJ.t has been, See ho"' to pass beyon<l the seen Of loathed existence, who <lesITe, A~ smnethmg better, sometlung lugher, The end of all ex1~lence. "'here Suh~tance nor birth nor bemg are 2

This insight 1s agam to be d1stmgmsh('d from that "suffusion of consc1ousne5s" whereby one comprehends th(' thoughts of others As explaim·d m the Da5uttara S and elsewhere, this 1s the power recog­mzed (or <lemed to-day) a5 telepathy, the ab1hty to read the mmd5 of othcr5 at a distance If one choose5 to call clairvoyance and telepathy mysticism, or to apply the 5a111e term to relig10m faith (in Buddha, and so forth), or to bee It m the tranC'e-VI5IOil of "infmite bpace," then there 1s notlung more to be 5a1<l 13ut 1£ one confine5 the u&c of myst1c1sm to the meanmg of orn.·ne55 with reality and the power (or <le51re) to effect 1t, then It lb obvious that the5e Abb1fifi~ib are not powerb belonging to my&ticu,m hut r.ttlwr to a snnple faith on the one hand and to a gro­pmg expcrmwntat10n with scientific fact-. on the other All these powers are developed through a M:'Vere cour5e of nwntal trammg They are not myi,,tical g1fb hut •· accomph5hmcnts" pamfully gamed, on a par with the trance-knowledge wl11ch begms with a plea..,urable mtelhgcnce and passe5 by degrees mto rapt mddference J<'mal felicity m the trance 1s attame<l not by umon or commumon but by •· the knowledge that after tlns present worl<l there 1s no beyond." 3 In 1b prnmhve form the trance adds nothmg to the adept's prcv1ou& knowledge though it leads to higher realms of consc10usness It 1s only in the later psychology that these higher states begm to be grouped around a more persistent ego than is adnuttcd in the earlier texts. The trance 1s never mvolun-

l ltIV, 61. i ltiv , 49. 3 DN. 11 68.

1~0 Edward W aahburn H opkina

tary obsession or seizure or cataleptic swoon leading to union with higher reality or Power, during which knowledge is revealed. The adept "enters spheres of space and reason regarded as infinite" and goes on to "the sphere of nothingness" and beyond that to the sphere "where there is neither consciousness nor unconsciousness," and so finally to a state of suspended perception and feeling, every stage being mastered in order and then in reverse order, so that the saintly adept can "lose himself in any one of them and emerge from any one of them whenever he chooses and for as long as he chooses," as 1s stated in the summing up of the Mahanidana S. The final trance is one of absolute equanimity devoid of the sense of sukha and dukkha ( well-being, ill­being). A seven-day exercise in this trance-absorption is recommended as bringing the assurance of \Vorthiness, or at least the assurance that one will not be born agam on earth. 1

1 The ancient (Ved1c) asceticism, which tormented the body to msure

1spiritual power, is thus replaced by mental discipline Meditation, designed from the begmnmg of the course of training to attain the result of mastering phenomena, becomes finally a tranquil absorption. In the first stage, the subject experiences a state of" pleasurable mterest," with his mmd still applied and active. In the second, this merges into a state devoid of pleasure but not without happmess, though now with­out application of mmd In the third 1,,tage, a state of seremty super­venes, with "happiness and fixedness of thought" (sulcha and citte­kaggat<i, without pit-i, pleasurable interest) In the fourth, there is left a neutral state of sublime indifference Each succes:-.1ve 1,,tate is (m general) produced by eliminatmg the mfer10r elements of the preceding. The pleasurable mterest of piti 1s lost m the succeedmg happmess (well­being) of sukha and, as the second stage eliminates mental application, leavmg only three elements, so the third 1s charackrizcd only by happi­ness and attention, and the fourth, devoid of pleasure and happiness, remains as a state of fixed but indifferent attention (the first trance 1s sometimes subdivided, making five in all).

The effect of th11,, course of trances Is first to extingmsh all craving, 'then to consolidate knowledge and produce unlnmted happiness (one has part in others' bhss), and finally to bestow absolute tranqmlhty. Before this final result, are mamfested the powers, such as the dibbasota and dibbacakkhu (the d1vme ear or hearing and divme eye or sight). An ecstatic contemplation devoid of reasoning merges into a sort of intuition (the power of seeing things as they really are) before passing into the deep mdiff erent neutral state. There are no physical hmita-

1 Mahiisat1patthiina S , D. N. 11. 814.

Buddhistic Mysticism 121

tions that the adept cannot ignore at will. He not only sees things as they are mentally, but he sees things in earth and heaven and hears all sounds, understands a1l thoughts, knows the past and present and even (in the case of Buddha) knows the future. He can perform all the "ac­complishments" and destroys all evil taints in himself. It is a marvel­lous and purifymg process, but at no point in this discipline or in the exercise of the "accomplishments" does the adept 'leek either to realize any union with a higher power or to "merge subject and object into one." Not only is no attempt to do this perceptible in the voluminous disquisitions on the trances but it is impossible that the adept should make the attempt, because Buddhism does not admit that there is any subject' The adept waits on no impulse from any exter10r power to enter into the meditation called Jhana He himself chooses his time for meditation, enters upon it by means of a mechamcal fixation of mmd (staring at a circle while sitting in a certam pose and the like), determines m advance when to begin, how to follow out each trance­state, and how to mvert the process, C'ommg back from the fourth (or fifth) trance to the first, and so to come out of his trance-state alto­gether. Each trance m its several stages 1s a self-induced state of hypnosis, the end of which is settled bt>fore it begins and the content of which before it arrives is already well known The adept orders him­self, before entermg the trance, to quit it at a certain time, so many hours or day1- ahead, and 11-aware of what will happen m each stage before he enters upon it.

The perfect peace of the final stage is entirely negative It is the absence of all that makes life worth hvmg, the pa1-srng away of all that binds to hfe, a foretaste of "salvation from existence." of N ibbana. The only my~ticism 1s that of hypnotism, wluch, finally 1s not incumbent on the Worthy, it 1s a practice which may be dispensed with, though recommended That the primitive Buddhist seeks to free himself from earthly or worldly bonds in order to "umte with the One," or to unite with reality by 1,mkmg back into the stream of hfe 1s an explanation later than primitive Buddhism It combines the soul-idea with the idealism of a time when Buddha had become a metaphysical abstrac­t10n, an Absolute Even as late as the third century B c there 1s no such theory, only the idea that Buddha was a supernatural being and that man, despite all that Buddha had preached to the contrary, had a sort of soul or persistent individuality. In the Buddlusm of the Great Vehicle, a development of the heresies of the Great Congregation of circa 800 B.C., is found the real mysticism of metaphysical theology, Gautama the man being one with a Holy Spirit, who is a form of the

122 Edward Washburn Hopkins

Absolute. But in the records of the primitive Congregation there is no mysticism and nothmg to warrant any discussion of the early Buddhist as a mystic seeking to realize himself in a new and wider world. The only mysticism 1s found in the trance, which is pre-Buddhii.tic, is not a necessary exercise, and 1s not really mystic (it connotes no "umon with reahty" on the part of the sub1ect)

Yet it ii, commonly accepted that, as Professor Hocking puts it, "Buddha referred all desire to the single cravmg which he described as the cravmg for ind1v1duahty or separateness of being", that, agam, in Buddhism the love of power of the Vedant1st "has taken the form of an ai.pirat10n for metaphysical status with all the power over one's own destmy (and over other men'i. minds) therem nnphed," to which Pro­fci.sor Bennett objects that the a1,pirat10n 1oeem1, rather "to he real -without qualification " 1 But it is a Bud<llustic scholar who has em­phasized most strongly tlrn, mterpretat10n of the Buddlusbc ideal as the overcoming of all "separatene1,s" m mysbeal umon with the One. Rhys David1o m lw, "Amcr)('an Lecturei-" 'Says "It 11, a belief com­mon to all schooli. of the Buddlust'S that the ongm of 1,orrow 11, pre­cisely identical with the origin of md1vidu,thty Sorrow 1s m faC't the result of the effort wluch an md1vidual ha1, to make to keep separate from the rest of ex11,tenee . . The urnty of force1, which con1,tltutes essential Bemg must sooner or later be du,1,olved ,vherever an md1v1dual has bC'come separate from the re"t of existence, then im­mediately du,ease, decay, and death begm to act upon it \Yherever there 1s md1viduahty there must be l11mtat1on, wherev<>r there 1s hmi­tabon there must be ignorance Men overlook the fact that they are really no more separate than a hubbh.· m the foam of an oc<'an wave 1s separate from tlw sea . It is not st>paratcness you 1,hould hope and long for, says the Budd}111,t, It is umon - the sense of onene,;,s with all that now 1s, that has ever been, that c,tn ever be Leap forward with­out fear You shall find yourself m the ambrosial waters of Nirvana and sport with the Arahats who have conqul'rcd birth and death" 2

Now it is qmte true that the Buddlust urge1, one to renounce the idea of Me and Mme and destroy the false notion of an ind1v1duahty bai.cd upon a permanent 1,ubstance m the Me But the Buddhist speaks only of impermanence. "There is no substance, no permanence m any being." To know th1i. is to destroy for one's own good what seems to be permanent, the well-mgh meradicable root of md1viduahty.

1 W E Hockmg, Iluman Nature and Its Remaking, pp 75 and 334, C A Dennett, A Philosophical Study of Mysticism, p 52

2 Rhys Davids, Buddhism (American Lectures), pp 124-129

Buddhistic Mysticism 123

The apparently logical conclusion that individuality implies a sense of separateness and that separateness must again imply desirable union with the rest of existence, oneness with the All, as the goal of effort, is never drawn, and the outcome can be only that there was no such goal, was no such conclusion in the mmd of Buddha or in the view of the primitive Congregation.

It is not a matter of logic, applied to what is really the Vedanhc (not Buddhistic) view of the soul as a bubble thmkmg itself different from the sea, a5 the spark really one with the fire, which leads to the idea of the soul 5eparated from the RC'al, from God It is a question to be determined on the evidence of the earliest Buddhishc texts, of which a number of examples have already been c1ted to 5how that Karma underlies the1r thought But these 5ame texts show also that in the Buddhism of the primitive Congrl'gabon the desirable goal was not union but mastery over Bemg, to the l'n<l that the ·worthy (Arahat) should cease to exio;t There 15 no rl.'al sub1ect, there 15 no perc1p1ent apart from percept10n There 15 only a ser1e5 of transitory states of consciousnesb Before death one can by trammg acquire mastery of intellect and fePlmg through development of purpose, effort, and so forth At or even before the death of the perfected Worthy all aggre­gates pao;s away (Karma no longer -work5), the elements of md1v1dual­ity cease to be The dt>struct10n of md1v1duahty 15 the clt·sired end be­cause that de1-,truct10n automatwally debtroys the pam of impermanent existence There 1-, no hmt that md1v1du.1hty 1mphes "separateness" of any 5ort The hmitatiorn, of inchv1duality are of course those of a bemg concht1oned, that 1s lmukd, but only b;v the process of Karma, not hunted becam,e separated from any umversal One The only One recogmzC'd was the one process of birth and death, the stream of bemg, and from that one the early Buddh15t sought to free hm1self There was too the idea of "separateiieb~." but 1t I!> not a bad separateness from a One but a desired and toiled for separateness from the Karma­stream, detachment from the world m every sense (compll'te i·n·eka). One of the means of becommg separate or detached was the practice of trances and the mastery over matter gamed thereby. In the h1ghebt trance 1f anywhere there should be a premomt10n of the sense of mystic "umon, onenebs with all," of winch Rhys Davids bpeah, but, on the contrary, the adept in thi!> final experience, when he 1s rn a state of complek sclf-posse5s10n and equammity, without pam and without happmesb, "Mts suffused \\Ith a senbe of pur1ficat10n, of translucence of heart, so that there 15 no spot m his whole frame not buffused there­with," and m this state he simply enjoys the •· accomplishments," that

Edward W aahburn H opkina

is, he becomes invisible, duplicates his form, travels cross-legged through the sky, hears all the sounds of heaven and earth, with his divine ear, sees the hearts of others, remembers his previous births, sees with his divine eye other beings and their estate according to their Karma, knows the taints (lusts, and so forth) as they really are, and is thus "set free" and has knowledge of his freedom, his emancipation, and realizes that rebirth has been destroyed and that "after this present hfe there will be no beyond" This (Samaiifia-Phala Sutta) 1s one of rnnumerable passages m which is summed up the fruit of a well-spent life. Such a hfe leads to "the destruction of all rebirth."

'fhere 1s not a word suggesting any mystic union with a One or an All On the other hand, at the end of this d1scour:;e, which is solemnly repeated m the Kassapa-Sihanada, there occurs the followmg state­ment "This is the frmt of a Buddhist recluse and there 1s no fruit of his hfe higher and sweeter than this," or, as worded in the second dis­course: "There is no other state of blissful attainment m conduct and heart and mmd which 1s higher and sweeter than this," namely the at­tainment of subJugation of evil and of the five modes of intmtion characteristic of the perfected samt A description of the eight stages of deliverance (above) does no more than explam the states of con­sc1ousnebs as apprehension of :;pace as mfimte, of the non-reality of things, and so forth, till one reaches the summit of consc10usness with the conviction that to tlunk at all 1s an mfer10r btatus, and then sen­sations and ideas cease altogether It 1s not soul that is functioning here but states of consc10usness and, smce all com,ciousness 1s imper­manent, when all fleeting :a.tates of conbCiou:a.ness cease there can be nothmg left to umte with anything.

In contrast with the lmuted created state of ind1v1dual bemg the texts speak of the state of freedom as uncreakd, aJiitam and later psy­chology speculates m regard to consciousnebs m that state, Just as some texts speak of the bhss of N1bbana in contrast with the misery of bondage. N 1bbana was "a bhsbful cessation of misery"; hence it be­came a bh:a.:a.ful state after death It was partly tlus mgenuous playmg with terms that helped to undermme the pnm1t1ve faith, Just as the natural hankermg after contmued life helped to reebtabhsh the perma­nent ego under a veiled form, despite th<' gravity and clarity of the early texts, which refute th1:a, mterpretat1on and emphasize the fact that there 1s no ego and that bliss is merely" freedom from all existence." The many Brabmans Joinmg the first Congregation wou]d perhaps un­consc10usly shape back to normal the abnormal not10n of a hfe abso­lutely endmg, as they succeeded in ideahzmg the subJective notion of

Buddhiatic M ystici.am

time and space countenanced by Buddha, who declares that conscious­ness has no existence independent of the body, which is impermanent, and that when the intellectual faculties cease, both the individual and the elements cease to be: "The world and the waning and waxing thereof is within this mortal body endowed with mind" 1

Again, it is not unusual to cite the Ten lndeterminates as proof that Buddha, after all, may have had a dim belief m the ten views which he refused to discuss. But Buddha was surrounded by sophists whose themes were these very problems, considered by him unfit for discussion, such as the duration of the world and of man's soul. In view of his own fundamental tbes1& that man has no soul, that nothing has any substance, anatta, it 1s certamly ot10se to ask whether his scorn­ful refusal to discuss soul with the sophists may be construed as a tacit admission that man may have a soul and a hfe after death If forced to answer, he would have said that, when anyone save a Worthy dies, the elements makmg for future existence on account of Karma will result ma defimte future hfe, 2 but at the death of a 'Worthy (who may have already enjoyed N1bbana) nothing survives. He has "escaped the yoke" and I& "devoid of hope", he passes away "without desire for ex1&tence or for non-existence" and his N1bbana is "deliverance from all ties " 3 The destruct10n of misery can be attained only by him who comprehends the All, but of what nature 1s this comprehens10n? "He that comprehend1, the All, sabba, and whose thought about it is of renunciation and abandonment, can attam destruction of nusery." 4

The Buddlust concerns himself with the All only m the sense that he abandons all lush," for the purpose of ms1ght and thorough knowledge " 5

The world of the All, a& we know 1t, Is itself but a temporary phe­nomenon, though It is divided mto <lrfferent parts as world~ of radiance, and so forth, but all, mcludmg the highest heav<?ns, are transitory. The perfected &age passe~ out and beyond them all The Agafifia Sutta gives an early Buddlustic view regardmg the world of men and its connection with other worlds and with human fate. "'Yhen after a very long period tlus world passes away," there comes later on a rebirth from thc-world of radiance, of bemgs horn mto the new world as human bemg1,. They elect a man to keep them m order and so mvent kmgsh1p. They settle down to var10us pursmts and so dev1~e castes [not here as

1 Subha S , Ang N., 11, 48 2 Lolncca S "Wrong views lead to pnrgatory or ammal rebirth" 3 It1v, 53, 55, 102, yogdt1go, nirdso, sabbaganthapamocanam 4 lbul, 7 6 lbul, 86

126 Edward Washburn Hopkins

the result of Karma], namely, priests, soldiers, tradesmen, slaves. When a member of any of these castes lives evilly and has false views (one defect is as permcious as the other), he will be reborn in a tem­porary hell, but one who has lived a good hfe and held right views will be reborn in heaven (a bright and happy world); and one who has lived a hfe of mixed good and evil "will be reborn to suffer both joy and sorrow" (as beast or man), and if, as man in the next rebirth, he has what are called the" two and thirty lucky marks," these. according to the Lakkhana Sutta, are to be explained as caused by good deeds m the past (1t is not, as usually said, the lucky marks that determine the future) By h11, mherent greatness Buddha himself, through previous Karma, wm, first reborn after death m a bright and happy world and then, reborn on earth, acquired the thirty-two lueky marks as the frlnt of deeds m his former life, but now m this hfo, having freed him­self from all ties, he i,ays of himself "The cravmg for future life of any sort 1s rooted out, that which leads to renewed hecommg is destroyed and there 1s no more birth for me." And what tlus means ii, md1cated by what was said when Buddha <lied "He pa.<;sed away m that utter passmg away wluch leaves nothing whatever to remam" A common formula 1s that used abow, "after this prC'sC'nt hfe there will be no be­yond " All parts and powers of a man dissolve and to brmg them mto this state' "1s bhss, that 1s N1bhana " In a word, to cite another preg­nant dictum "Cessation of Karma 11, N1bbana" Had aught remamed m the future, any union or '' <;elf-rcahzabon," 1t 1s mcrechble that it should have escaped ment10n m these explanatory utterancC'i,

The "mii,ery" of hfe 1s often exp lamed m tl){' early texts as no more than Mmple unease The modern mind 1s apt to i,trei,s the philo­sophical side of Buddhism or to make the mi5take of mterpretmg early Buddh15m m term., of an ethical code It wm, mdeed cth1eaJ, because there can be no true knowledge m an unethical naturc Thii, umty of ethies and wisdom is common to all Indian thought (though occas10n­ally demed by extravagant 1,oph1sts) and 15 very beautifully formulated in the Sonadanda Sutta. ''"'here there IS uprightness thne 1s w1dsom, and where there 11, wisdom there 1s upr1ghtne1,s, to the upright belongs w11,dom, to the wise belongs uprightness, WI5dom and goodness are declared to be the bei,t thmg m the world " And 1t was mdeed ph1lo­soplucal, with its persistent elaboration of the thesis "all is Imperma­nent " But 1t 1s noticeable how, m teaching men to escape from the cond1hon of nnpermanence, the emphasis is laid not 5o much on the misery of impermanence as on the physical and mental suffering m­volved in hvmg. The Four Noble Truths urge escape not from im-

Buddhistic Mysticism 127

permanence but from a permanent condition of bodily pain and sorrow extended through "transmigration's weary round " Karma itself is permanent till one escapes what 1t enforces, that is, a permanent con­dition of impermanency'

The three fundamental doctrines of Buddhism, formulated in the words aniccam, dukkham, anattam (impermanence, misery, unsubstan­tiality), are of course emphasized differently m different passages and it IS true that one of these may remam unmentioned while another is propounded and that m general the three together seem to be implied in all statements of doctrine Yet it Is rather surprising to find the Four Truths explained m detail on their first promulgation in a form wluch sugger;ts that the whole weight of Buddha's "discovery" lies in the anmh1lation of phy!->1cal and mental misery Birth, old age, grief, suffermg, despair, not to get what one desires, all this is pamful, all hfe il', pamful, a round of r;uch painful hvei> 1s only linked misery long drawn out It 1s nowhere hmted that one may rise to a higher plane of living devoid of misery The only remedy offered is to stop hvmg and so stop suffermg Incidentally, it may be addl'd, 1t 1s nowhere sug­gested that hfe has anythmg to counterbalance its misery The Bud­dhist !'ll'l'S no lovdmess m the human body, he is bidden to study it as a d1sgm,hng mass of mtestmes He never asks whether old age may not be happ~', whether death 1s worth crymg about, whether happiness Is not as common as unhappmesi> He !-.eei> only 1lli> in hfe and 1gnorl's its JO)'i>, as he sees only human uglmesi, and dei>pisc!-. 1b phyi,ICal beauty 1

Such a one-!-.1ded view of hfo wa,; evidently not natural This is shown by the peri,Ii>lence with which 1t I!-. urged Thl' <l1'ic1plc had to be drilled mto takmg hfe at Buddha's estimate of 1t The refuge would be i>U1c1dc, but, "'1th thl' unphc1t belief m Karma, tlus would be only to prolong the agony Hence one mui>t learn how to get the better of Karma, how to commit a smc1de that would have pl'rmanent re!-.ults; how to Ci>Cape unpermancncy of bemg. "luch 1s perpetually reshaped to fresh m1i>ery Thui> nnpermanencc of md1v1duahty became the greate:-.t ill of all, because hardei>t to get rid of, but such an ill made its strongest appeal only to the <;age For the mass of humble Buddhist recluses the weight of argmm•nt rrmamed rather m the ills of dmly hfe, till the mo!>t lowly could solemnly recount that, findmg it too weari­some to plough and reap, he gladly bec-ame converted to a hfe of mo­nastic ea~e '~ We may 5Ui>pcct that it was oftl'n these Ills that appealed

1 er c g, Theri G , 19, 380, 466 2 Thera G , ,·s 43

128 Edward Washburn Hopkins

most to the lower orders. Yet it is still remarkable how the sage also treated life's ills as if they were of equal importance. Life is not worth living (a) because it is painful and (b) because it is impermanent. Even in the scheme of causation the note is the same. For instance, at D. N. ii, 32, where are omitted the two elements of ignorance and Karma and the scheme is set for a single life: "From cessation of birth comes cessation of decay, dying, grief, lamentation, ill, sorrow, and despair; such is the cessation of this entire body of ill" (that is, life). Other passages treat impermanence as the great ill; to put an end to imperma\lence is the goal· "Cessat10n from rebirth 1s the escape from what has come into being and is condit10ned (because 1t) has arisen from a cause," 1 a passage that illustrate!. also the fact that "condi­tioned" 1s not "hmited by separat10n" from the All, but is equivalent to "bound in the causal nexus of Karma" (implying rebirth) On the other hand, the once-returner (a sage who has almost "laid low his burden" but has to be reborn once more) 1s not said to be near the goal of permanence but near the goal of "makmg an end of sorrow" (hfe's miseries) One might 1magme that the practical pessimism of this attitude sprang from the uncertainty as regards the kmd of rebirth, for one 1s liable to any one of the "five ways of destmy," that 1s, one may be pumshed m purgatory, or be born as an ammal, or function as a spectre, or be reborn as a man or as a god The decadent Buddhism of later centuries copied a leaf from Brahmam!,m and avoided any such uncertainty by teaching that a pious gift to a Buddhist or Brahman (the Brahman says a gift only to a Brahman) result!. m ment enough to escape any except the most desirable lot hereafter, a passage so philosophically naive that it deserves to be cited complete "A moral person may decide by aspiration upon his form of rebirth when he offers a pious gift Such a person may give a gift to a Buddhist recluse or to a Brahman m the shape of food, drmk, ranncnt, and so forth He hopes to receive somethmg for the gift He sees a nobleman Ii ving luxur10usly and he thmks, 'Oh. when I die might I be reborn as a wealthy nobleman ' He holds this thought fixed m his mm<l and ex­pands 1t and this thought of lns, s<."t free m a lower range and not ex­panded to any higher range, conduces to the (desired) rebirth within that lower range. This applies only to a moral person, for the mental asp1ration of a moral person succeeds because of its pure ~ingle-minded­ness " 2 In the same way a moral person may become a god m the next life, 1f m this hfe he give a pious gift with this aspiration. But

1 Dasuttara S. 2 Sangit1 S

Buddhistic Mysticism 129

such puny rewards for earthly generosity are despicable to the sage and were probably a late invention even for the humble and ignorant who were likely to be tempted by them. The reward of course is only for the laity, an inducement to them to be generous to the Buddhist mendicants.

The Dasuttara Sutta gives nine "perceptions to be cultivated " One of them is of the ugliness of hfe and one 1s of impermanence and "of suffermg in impermanence " The relative value of these percep­tions may be judged better perhaps if one turn from the philosophical expositions, where aniccam and dukkham are apt to stand side by side, to the &1mpler cogitations of the hermits, as voICed in their pious verses. Here there 1s, as strikes the attention at once, a formal acceptance of the doctrme of impermanence It is so formal in fact that it occurs repeatedly m the same phrase "thereat aro&e in me the deeper view, attention to the fact and to the cause," 1 followed by the statement that the mu,ery of hf e then became manifest, and there are verses keepmg the close connection between the ills of hfe m terms of pain and of impermanence 2

When one by wisdom doth discern and see Th' imperma11e11ce of everyth111g w life, Then one at all life's suffering feels disgust, Lo 1 herem hcs the '\\a) to purity

"hen one by \\ 1s<lom doth discern and i,ee

That e1Jen1thing 171 lzf e IB bound to ill, Then one (and 1.0 forth, sabbe sam/Juird anzccd du/J.h<i anattd)

The hernut &truggles to convmce himself that "mmd and body should be held to he 111 ", then he adds "and understand imperma­nence to be 111" and cite& the words of the Great Phys1c1an, who likens the heart (mmd) to an ape leaping for frmt uncertamly from tree to tree, with tlus conclu&10n

l\fany and s\\eet, cntrancmg, are the lusts \\ hercm the ignorant maJtmty Entangled he Thcv do but \\ ish for 11! \Yho seek to In e agam, Led by their heart to perish m the Pit 3

1 See Thcra G, vss 20D, SOI, 318, 409, 464 2 Ibul , vss 676-677 3 Niraye, ibul, vss llll, .~eq The poetical translat10ns are taken wholly or

slightly changed from l\Irs Rhys Davids' PsalmB of the Brethren (TheragiithJ.)

ISO Edward Washburn Hopkins

The same poet in a subsequent stanza gives the essence of his thought thus:

Mountams and seas and rivers, earth itself, The quarters four, the mtervemng pomts, The nadir, yea, and e'en the heavens above­All are impermanent and all forlorn Where ranst thou then, my heart, find ease and rest 1

Further on in his poem the recluse answers his query by saying that when his mmd (or heart) 1s fully trained 1t will become "devoid of all craving for any form of future existence" and "pass beyond the stream of bemg," which means that he will rest happy m the knowledge that his hfe is absolutely ended for all time. Another poet 2 thus v01ces his dec15ion (bhaven' amhi anatthiko)

There 1s no life that lasteth evermore, Nor permanence m tlung!. from causes sprung The fortors of our hfe to hemg come And then d1s~olvP In that tlwv pa!.s aVI a), Tlus 1s their 111 I seek no more to be 1

He who has overcome cravmg, the poisoner of hfe, the giver of pain, sits free with mind mtent, "rapt m ecstasy of thought, and no higher bliss 1s given to man than this," says Bhuta, 1 who5e whole "psalm" is a repebt10n of tlns phrase, the meaning of wlueh 1s a-, obv10us in its implication as m its express utterance. Man's h1ghe5t fehc1ty 1s to have the certainty that he will hve no more after death Here, of all placc5, would have been where one would expect some suggestion of a faith m a life beyond, had the poet who compm,ed the vnse5 been un­orthodox enough to harbor hope of this

Yet already such a hope was begmnmg to spring up, though for the most part veiled m negat10n such as that of the cheerful rakr who squan­dered all his wealth on a harlot and bemg de5btute rept>nted and 1omed the Buddlusts He smg5 how he has given up all de51re and 1s "now faring on to N1hbana, where at our Journey's end we grieve no more," 4

gacchiima dani nibbanam yattha gantrii na socati One sees, N1bbrma has become a place, as 1t were, to go to, a negation (of sorrow) con­ceived spatially. The one who renounC'es the world "de5troys its grief and pam" 0 and it was not a far cry to the thought of the "beyond" as another world where all is bhss A httlc freedom of translation per­fech th15 and N1bbana is tran5formed mto a comfortable heaven.

1 Ibid, vs 1133 4 Ibid , vs 138

2 Ibid, VSS Hll-Hl~ 6 Ibid, \-S 195

3 Ibul' VS 519

Buddhistic Mysticism

Oh, great, Oh wondrous is N1bbana's bhss, Revealed by Him, the Utterly Awake 1

There comes no gnef, no pass10n, haven sure, Where 111 and a1lmg perish evermore 1

131

Litterally, "very happy mdeed, as indicated by the perfect Buddha (awakened), Is N1bbana dcvmd of grief and pasi,,ion, peace, where un­happiness disappears" But even to thr Buddhist, whose notion m the early texts was never that of "entermg" N1bbana as a place, there must have been ID <,uch ven,es the reflection of a dim feelmg that N1bbana was not altogether the mere extmct10n of paID and of ex11,,tence but had a more positive content From havmg been the negation of a mortal hfe of sorrow It became gradually "the immortal not-sorrow­full," nay more, It became "the bhsi,,ful" because It 1s "deliverance from sorrow" and "<lehver,mce 1s bh1,s " Ai,, one who gams the bliss of N1bbana never returns to sorrow it 11,, natural to call N1bbana "im­mortal'' The veri,,es of the tlmd century n c (or perhaps a bit earlier, only not so old ai,, the firi,,t d11,,couri,,ei,,) show dearly the trend, and the Great Velude doe:,, but follow this trend when it even ventures to ID­terpret N1bhana as that flow of bemg which, as Samsara, 1t was ID­vented to a,·md I A verse ascr1b£>d to Buddha but abo allotted to a lesi,,er authonty J gives the origmal coneept10n

"]1050 "1tl11n this goodl} du,c 1phne Shall lon11;• \\1th d1hgt·nte to h.110\\ the truth He lmth ·~ eternJ.l round ~hall put a\\ .iy

And end .111 µ.1m and sufferrng for <1J e 3

1 Ibid , , 5 227 2 Ilnd , , s 2,;7 3 The text of , cr~e~ 227 all(! 2.n 15 as follm, ~

su~uh.h.irp ,·atn mhbannrp ~.uumiisamlmrldhade51tarp asok.i.rp , 1ra1am khcmam vattha du],.J,.h.tm mruJJhati (v~ 227) ,o 1masnun dh.inuna,rnn\(' npp.tm.ilto ,1hesMb pnhiiya Jii tisam;,:iram dukkhass' antam h..inss.ib (, s 2 j7)

The mnal dcsC'r1pt10n of N1bb:m,i 1~ 111 terms of pre;,enl m~1ghl and (_ool happi­ness or pe.t(_e .ind 1>ecunt:,, yogal.hemam anuttaram, sa11t1, v. hen ,t 15 not simply the extmd10n of cr.i,mg and of hfe The po;,Jll\t' a~pect refers, as ma\ be seen from the Guthus, to the state of mmd atlamed before death b) one confident of post mortem emnnupalton from further hfe, one who h.i~ "overum1e e'\1~tence" The sense of µeac.e may have bct·n heightened b~, the le:,s plulo~oplucal but practical belief that "existence," if not o, ereome. enta1\cd torment m hell, hut this thought could not have aff1..x·ted the view of the "all but" perfected samts, though such pas­sages as Theri G 4Jl and 501 show that 1t l\RS not without s1gmficance m their general estimate of hfe's misery

182 Edward Washburn Hopkins

The appeal here is simply to the feeling of relief from suffering; it offers no hope of future hfe or of any sort of existence, nor does it lay any stress on the motif of impermanence A certain Gotama, not the Master, in another poem unites the two theses in their natural suc­cession He says that he became a Buddhist mendicant because he had often suffered of old (in previous lives) both in purgatory and as an animal, not to speak of existences in which he suffered as a spectre ( a kmd of spirit that is always famished), and after this, "glad enough to be reborn as a man" he still suffered in human form; and even as a god (after his human hfe) he was not satisfied, for though in successive turns he "hved m the realms between consciousness and un<'onscious­ness," yet to him "all this becoming" (impermanent being) at last appeared "void of real value," so that, on being born as a man again, he turned to his ~ole refuge santi, pea{'e (escape from rebirth), which is h1s N ibbana. 1 The misery in hfe 1s recognized first, then the misery of livrng.

To the laity 1t was enough that the l\fastcr had ~aid "transient are all existmg things " They repeated this formula dutifully while still eagerly seekmg "transient thmgs," such as children and wealth One of the good mendicants sneerei at them for doing !-.O, or, as the commen­tator says, he speaks compassionately of them, in verses which may be freely rendered thus (kiirnii amccii iti)

The word alone hut not the thought the pious laity msp1res, Forever saymg with their hps "tran~1ent are all the world's de1,1res," The whtle they seek the very th1ng5 they (log1eally) should despise, W1ve5, c,h1ldren, wealth Poor pass10n's slavrs are thev, unable to be \\1sel (vs 187)

The doctrme of impermanence did not trouble the laity. They were taught to look for nothing better than a life hereafter in a "bright and happy world," whence they could return to earth agam until their hearts were weary of change and they should seek the supreme JOY of non-existence To them apparently the sorrow of life did not over­balance the JOY of hvmg The teachmg of the l\Ia5ter in their case was confined to msbllmg moral rules and ~omewhat slowly inspiring them with a behef that life was sad rather than joyful. Until the imperma­nence-doctrine was grasped, the sorrow-doctrme was comparatively unavailing. But neither to the laity nor to the mendicant disciple was Buddha's teachmg that of a mystic. In regard to all speculations con­cerning a future existence he said simply. In the case of the laity (the mentally untrained), 1f good, they go to heaven; but in the case of the

1 lbul, , vss. 258, aeq

Bud.dhistic Myst,icism 133

wise who know the truth, there is no use in inquiring whether they exist hereafter, for "such questions implicitly assume the heresy of a sub­stantial ego, of a real self (apart from states of transient consciousness); to believe that one has a self, a soul, permanent, lasting, eternal, which will continue after death, is to walk in the jungle of delusion, and bound in this delusion a man 1s not free, he is not wise, he is not emancipated from sorrow, he will still go on in the weary round of birth and death." 1

The highest praise given by Buddha to his contemporary Gavam­pati is that he was a "surpasser of becommg, '' that 1s, he had surpassed Karma and passed beyond the state of future existence, praise much higher than that accorded to him by the multitude, who acclaimed GavampatI because he had by his "accomplishment" prevented a river from overwhelming a Buddhist settlement (he stopped the river and "made it rear up like a mountain-peak " 2 To conquer death and life and so to "abolish ill forever" was the goal of the early Buddhist, as Sarabhanga 3 1-ays "The Buddhas who taught the law, and were themselves the law mcorporate, showed the abolishment of 111, whereby one at death may become free from rebirth, wholly emancipated " The cessation of becommg is cessation of bemg, as S1vaka 4 says:· "Birth 11-but woe agam, ever again ... stayed in me now 1s all further rise of consciousness, blown to nothmgness will 1t be even here" (be­fore I die), words also attributed to Buddha (t•ipariyiidikatam cittam idh 'eM ndhamissat1)

With this agrees the frequent image of the passions and hfe (for the future) of a perfected samt bemg "extingmshed hke a fire." An ingemous scholar, argumg for some spark of future hfe m the ideal of prim1t1ve Buddhism, has entered a plea for 1t on the ground that N1bbana, the "blowing out," of the flame does not mean complete extmgm1-lunent, because the Brahman priest held that fire was an im­mortal god and when produced by the fire-sticks 1t was merely called mto action agam, the fire had been lurking m the fire-stich. So the N1bbana of the Bu<l<llust may imply a lmgermg spark immortal still. But, 111 the first place, the Buddhist N1bbana, as Rhys Davids has shown, refers primarily to the cxtmct1on of the very elements (lm,ts, and so forth) which the adept 1s bent on <lestroymg utterly and, sec­ondly, the flame of the Brahman's immortal fire is not lurkmg in a lamp but in the fire-sticks and N1bbana refers only to a lamp. No Brah­man priest ever thought that the flame of a lamp was not really out when it was "blown out" One of the Buddhist Sisters says (vs. 116), in order to visualize her final goal· " I press the wick right down

1 Sabbiisava Sutta 2 Thera G , vs SS 3 Ibul , vs 493 4 Jbul, , vss. 183, aeq

184 Edward Washburn Hopkins

[out goes the flame]; behold the lamp's Nibbiina!" She has attained freedom from craving, which lights the lamp of life. She certainly does not hope for re-illummation. Mrs Rhys Davids says very truly of the Sisters "Their verses do not seem to betray anythmg that can be corn,trued as a consciousness that hidden glories . are awaitmg them. There 1s nothmg pointmg to an unrevealed my5tery" Negative evi­dence of this sort extendmg over all the earlier literature 1s a fornudable argument agamst the assumption of primitive belief many future state whatever m the case of the Arahat He i5 parinibbuto, "completely extingmshed," explicitly a5 to cravmg, nnphcit1y as to future conscious­ness (the frmt of craving) and any state of being.

The temporary states of consciomness (replacmg "soul") which represented md1viduahty had to be d1bper<sed m order to the attam­ment of the Buddhist's goal. At, or before, thl' final phyb1cal death the labt element leading to any future life lb extmgtm,hed The fleetmg states, hitherto formmg the unagmed ego, haw ceased to be, the "hut of belf" has been broken up Thoughb and :icts still cx1bt m the umverse but only as content of the monwntary hveb of othnb, they are not "mme ", no I exists to thmk or act Rcmovc the caw,e of "becommg" and the very existence of the ego CC'aseb In the flood of other t:>Xlstence can survive at mo5t only the dtsJecta membra of the pseudo-ego The m<l1v1dual ex1sh, no more, no lugher lJemg exists with wluch a 1,urv1vmg con5c10usm•<,~ could umte and tlwre can be no consc10usne1,1, (even were there such a Bemg) when once the maker of consc10usnl'ss is anmhilated ·what wa~ once regarded ab the subject has been "blown to nothmgnebb" The <;urv1vmg pu,r;qalo of the later psycholog15ts 11, an abortion born of the de<sire to fit prnmtive Buddlusm mto a 1.ystem that practically a1,1,erb, what Buddha categorically demed N1bbana 15 freedom and relem,c and only as such <loeb 1t have a po1>1tive content, rde.i1>l' from grid and fear for one whobe sorrow has been hfe and who1,e fear has been that hfe nught contmue:

mokMwmlu vwamtine idam a;aram 1dam amaram idam a;artimaranapadam asolam asapattam ammbtidham aVwlztam abhayam nzrupattipam "'Tu, fn•cdum (from desire), rclca~e from age, from death, from sorrow, From rivals, crowds, and stumbling, fear, and hurt 1

i Theri G, vss. 506, 512.

YALE UNIVERSITY

EXTASE ET SPECULATION

(Dhyiina et PraJiiii)

PAR LOUIS DE LA VALLtE POUSSIN

T'ANGUTTARA (iii, 355) contient un texte isole clans la vieille D litterature bouddhique ma1s tres sigmficatif " Les moines qui pratiquent le recueillement ou extase (les jhayins) blament les moines qui s'attachent a la doctrme (Jes dham,ndyogas), et rec1proquement. Ils dmvent au contraue s'esbmer les uns les autres rares en effet les hommes qui passent leur temps en touchant avec leur corps (c'est-a­dire, en connaissant d'une connaissance immediate, en "realisant," en "experimentant," reali:ing, experiencing) l'element 1mmortel (amatii dhatu, c'est-a-dire, le N1rvana), rares ausst ceux qui vo1ent \a profonde reahte (atthapada) en la penetrant par la PraJfia (c'est-a-d1re, par la conna1ssance specufabve qm <liscerne le caractere vra1 des choses)

2. 11 resulte de ce texte qu'il y eut confht entre deux tendances, entre deux ecolPs, entre deux categories de rnomes.

D'une part les hommes du recue1llement (samadhi) ou de l'extase (dhyana, 1hana) - Ils henncnt en mediocre esbme la connaissance a proprement parler mtellectuelle, le savo1r d1bcursif qm elabore une metaphyl,ique, etud1e la nature des choses, s'attache aux verites bou<l<lhiques <le la <louleur ou <le la " production en raison des causes " Toute leur confiance c:,t <lan:, les v1s10ns et sensations mystiques qu'on obt1ent <lans les ctats plus ou morns hypnotiques nommes "extases" Pour obtemr ces etat5, une <l1ete severe et des exerc1ces d'ascehsme (contemplat10n du cadavre) et <l'hypnobsme (fixation du regard, etc) sont m<l1spenbablel> L'ascete qm les pratique obtient, en memf.' temps que les pouvoirs mag1qucs, l'O:'il d1vm, la connmssance de scs anc1ennes existences, etc l\fa1s, b1en plus precieux, eleve au <lessus de humame nature 11 entre en contact, au cours de~ recue11lernents les plus pro­fonds (" recue1llement de la ces:,ation de la pensee et de la sen<;ation ") avec cette chose mde:fimssable qu 'on nomme Nirvana

Ce vaut la de grands avantages Les "ho1nmcs du dliyana" se recrutent necessa1rement parnu Jes

momes le m1eux places pour entrer en recue1llcment "forestiers," qui t d "h d . t' " t d " ' t t " son auss1 es ommes u cnne 1ere e es pcm en s ces momes

v1vent a cote de la rcgle commune, pratiquemcnt d1~penses des cere­monies conventuelles, de l'etude, du preche

186 Louis de la Vallee Poussin

Le danger, pour eux, est de chercher l'extase pour l'extase, et con­fondre l'extase avec l'hypnose L'hypnose n'exige pas la purification de la pensee, la suppression du des1r: des recettes mecaniques y suffisent - fixation du regard, suspension de la respiration, serrer les dents et coller la lan:gue au palais Beaucoup d'ascetes croient obtenir par ces procedes les pouvoirs magiques et la possession du Nirvana· le plus grand nombre se fait illusion, car les pouvoirs magiques comme la possess10n du Nirvana suppose la samtete que l'hypnose ne peut dormer.

S Ainsi pensent sans doute les "hommes du dharma," moines fideles a la discipline conventuelle, ennemis des exagerations ascetiques, qm recitent l'Ecnture, qui prechent, qm reflechissent sur la nature des choses. Ces momes se souv1ennent que le Bouddha a defim la saintete comme "suppression du desir," et explique qu'on supprime la soif en etudiant la nature douloureuse et impermanente des objets qui pro­voquent la smf

Les hommes du Dharma ou "philosophes," condamnent done les hommes du dhyana, ou mystiques.

4. Le Bouddha met tout le monde d'accord - Sans la Prajna, connaissance des verites bouddhiques, la saintete et le Nirvana sont impossibles D'autre part, les exercices d'hypnose sont tres bons. quarrd ils sont pratiques par un homme detache des choses sens1bles, 11s prodmsent l'extase (dhyiina) Dans l'extase l'ascete peut prendre con­tact avec le Nirvana

Le canon nous apprend qu'il y a deux especes de saint celui qui est "dehvre par la Pra1:fia," qui, a la mort, entrera dans le Nirvana; celm qui est "doublement dehvre," c'est-a-dire, qm, etant deJa "dehvre par la PraJ:fia," a cult1ve Ies recue1llements et pr1s, des cette vie, posses­sion du Nirvana.

UNIVERSITE DE GAND

INTEGRATION OF CONSCIOUSNESS IN BUDDHISM

BY JAMES HAUGHTON WOODS

DHARMAPALA'S comment (P. G Mundine Pitaka Press, Ran­goon, 1909, p. 214) on Visuddhi .Magga, Book vii. 2038, unpub­

lished in America or Europe, throws light on the problem and method of the integration of consciousness as conceived by a Buddhist

What is sought is an insight into the umty which is presupposed in change. A distorted vision of hfe results if 1t be assumed that existence consists of independent, momentary, and permanent entities separated by discrete units of time. Our hopes, our feelmgs, our perceptions, pre­d1spos1tions and activities are obstructive and painful 1f we clutch at them, as a man might snatch at grasses, reeds, creepers and shrubs which overhang a !;W1ft-flowing torrent in which he is swept along (Sau­yutta, xxu, 93-4, vol iv, 139-140) Such inconsequential efforts prevent us from graspmg the interconnect10n of all exi!;tence, and thus umfying our mmds A bemg who can comprehend the world Without segregat­mg it mto groups which are mcompat1ble and who can trace the tran­sitions w1thm and without such groups 1s hke the lotus born m the water, stand mg m the wakr, yet unspotted by the water This typical being 1s the Tathagata who is unspotted by the world and not swept away by thr stream. In ordmary bemgs therr 1s the constant effort to at­tam stability, to overcome mner contrad1ct10n, and so to become more orgamz< d and more coherent The Tathagata 1s not disturbed by the succe!;sIOn of thmgs He 1s not caught m parts of the mecham!,m For he discovers the nature of the mechanism and ulenbfies hunsdf w1th1t The order of nature stands (Sariyutta, 11, 25), the mterrelations are defimte, each concretion consists of mstances of relations The Typical Bemg is fully enhghtened and under!;tands this and sets 1t forth and makes it plam An ordmary bemg who can discern the TathJgata discerns the orderliness and mterpenetrabon of thmgs m time Keepmg the law thus means not breakmg the umty of things by errors of 1udgment. This retention of eqmp01se extends to mternal balance of emotions, passions, and thoughts and to actions and words Without the control of conduct there can be no eqmp01sed mental event, and without poise no ms1ght. The passage in Dharmapala's comment shows how the in­terplay of thought, emotions, and actions 1s reqmred in the higher types of personality.

138 James Haughton Woods

Ettha ea vijjasampadaya Satthu pailfiamahattau pakasitau hoti, caranasampadaya karu:r;taya mahattau. Tesu pafifiaya Bhagavato dhammara11apatti, karu:r;taya dhammasauvibhago; paililaya sausara­dukkhambh1da, karunaya sausaradukkhasahanau, pafifiaya paraduk­khav11ananau, karu:r;taya paradukkhapatikararambho; pafifiaya parin­ibbanabh1mukhabhavo, karu:r;tiiya tadadhigamo, pafifiaya sayao tara­nau, karu:r;tiiya paresau taranau; pafifiaya Buddhabha vas1ddh1, kar­unaya Buddhak1ccasiddh1, karu:r;taya bod h1sattabhum1yau sausara­bh1mukhabhii vo, pafifiaya tattha anabh1rati, tatha karuniiya paresau abh1usapanau, pafifiiiya sayau pareh1 abhayanau, karu:r;taya parau rakkhanto attanau rakkhati, pafifiaya attanau rakkhanto parau rakkhati, tatha karunaya aparantapo, pafifiiiya anattantapo, tena attah1tiiya pat1pannad1su catusu puggalesu [A 11 95] catutthapug­galabhiivo s1ddho hob, tatha karuniiya lokaniithatii, pafii'iaya attanii­thatii, karu9iiya e'assa ninnatabhavo, pafifiaya unnamahhavo, tathii karuniiya sabbasattesu 1amtanuggaho, pafififtytmugatattii na ea na sabbattha v1rattae1ttata, pafifiaya sabbadhammesu vuattae1tto, kar­unanugatattii na ea na sabbasattanugahataya pavatto Y athii hi karuna Bhagavato smehasokav1rah1tii, evau pafifiil ahafikiiranrnma­karavmnnutta tJ afifiamafifiav1sodh1ta pararnav1r.,uddhi1 gunav1~esa VIJJaearana-sampadah1 paki1~1tii b datthabao

Here agam the range of the Teacher'5 ,, t5dom 1, d15plaved m the Jwrf<'cbon of ]115

mtelhgeme, and the range of Im, rnmJJ.1;,;,10n b~ the perfec t10n of )11;, rnndurt llv tins \\lsdom the Exalted One .i,U.i,m;, the J..mg<lom ol nghteou51W55, and I,, tln~ comJM5SJOn he d1;,tnhute;, r1ghteousne5s llv \\ 15clom he dl'5m'5 to he free from the d1,torhon5 of the round-uf-rdnrth5, hv compa;,;,10n he endure;, the d1;,turt10n;, of the round-of-re­births, Lj \\1;,<lom he under;,tand5 the pam, of other,. hv comp.i,5,10n he 1'\.c·rt5 h11115elf to find ITmed1Ps lor tlw pam;, of other;,, by v. 1;,dom he ;,el;, Ju, i.tce tuv.,ml, final de­hverarwe, b~ compa;,510n he .i,due,e;, frnal dehver.i,me, In ,,1,dum he ,ine, lmm,l'lf, by compas;,1on he SJ.\ e;, other,. hv \\ 1;,clum he atta1115 tu Budclh.i,hoocl, b, eompa5,1on he hrmg, the aet1on5 ol a lluddha to pcrfect10n - By lomp.i,,,1011 \\ lnle m the st.ite of a Future Buddha he wa1, facmg the round-of-rebirth',, h, v. 1sdom he take!> no delight m that pa!,t hfe, l1kev.1!,e hJ compas51on hC' rPlrams lrom producmg any fear mother,, hy \\ 1sdom he lumsdf feels no fear of other!,, h,. lompa~~1011 \\ lule protec trng others he protelts h1m~elt, by \\ t5dom v. lulc protcctmg lum,ell he protPet5 other5, like\\ 1se by eompasstou he gne5 no tormPnt to other'!, hJ Wl'idom he gnc~ no torment to hun'lelf Thus he attam'l to the 1,tate of the fourth type m smh J. h~t a~ that of the four tvpes of man [A, 11 9.3] Smularl, by compu5;,1011 he heeome5 Lord of the World, l,y \\1sdom Lord of Self. by eompa5510n he cea'ies to aba'le l11m1>Plf, bv \\ 1sdom he ceases to feel pride L1ke\\1se hv compass10n he does helpful act!> toward., all hemg1, Becam,e Pompa;,s1011 11, comhrned \\ 1th w11>dom, he 1s not attaC'hed any\\ here Bv v. mlom !us mmd Ill unattaehed to any thmg llelause wisdom 1s eomhmed with compa,s10n, 1t operates m helpfulnes<; to all hvmg hemgs For Just as the eompass10n of the Exalted One 1s without all personal attachment or repuls10n, so his wisdom 1s qmte free from

Integration of Consciousness in Buddhism 189

vamty and seU-re£erence Thus we should understand that it has been shown that the peculiar excellences of the Exalted One are purmed m the highest degree by the per­fection both of his mtellect and of his conduct.

This integration reaches such fullness and completion in the Typical Being that when one discerns the order of thmgs and their interrelation one discerns Him, and when one discerns Him one is aware of the co­herence of all existence

HARVARD UNIVERSITY

L'AUTONOMIE SPIRITUELLE SELON LA PENS:EE INDIENNE

PAR PAUL MASSON--OURSEL

T'EUROPE scolastique a d1scute sur la nature des "purs esprits"; .l.....J elle a tout au moins adm1s !'existence d'un "esprit pur" - Dieu. En outre les theories de la liberte, depms Biotin JUsqu'a Secretan et Renouvier, ont eu a cteur de montrer ce qu'il faut enlever ou aJouter a l'homme empirique pour qu'1l se reahse en sa veritable essence et, le cas echeant, pour qu'il fasse retour au prmcipe divm

Les <leux problemes ne s'1mposerent pas moms ii l'Inde, comme en Occident la religion fraya la v01e a la ph1losoph1e Mais la rehgion fut plus une ascese qu'une theolog1e, a l'mverse de ce qm se prodms1t chez nous, le chr1sbamsme s' ctant moule tres tot clans ladogmatique grecque.

S1 le Bouddh1sme n'eut 1mplante darn, la pensee md1enne l'obse~s1on de l'esclavage qm rcsulte de la transm1gration,Ia doctrme des Upani~ads et1t regne sans conteste elle pretend rendre l'ame md1v1duelle homo­gene :'t l':lmc absolue par chmmation des pomts de vue relatifs S'il est perm1s de s'expnmer en termes spmoz1stes, d1sons que nous sommes Dieu non en tant que modes, ma1s en tant que substance; et le fa1t est que substantiaht{, se d1t iitmatii '' Celm qui sa1t " s'ass1m1k, a aat, cit, ananda l'etre, la pensee, la beatitude

Mais l'1<lee de transmigration, extra-brahmamque par ses or1gines, introdui~1t une toute autre concept10n l\lfr1tes et dernentes acquis au cours des vie~ prcce<lentes comme pendant cette vie nvent l'ind1v1du, selon les Bouddh1sres exempt <le substantiahte, a une existence sans fin. Cependant le savoir, 1c1 encore. ouvre la voie du salut Il faut comprendre que la transmigrat10n resulte <lu desir chez qui ne desire rien le karman cesse de s'accroitre et, par la simple usure du temps, il s'epmse Les Jamas precisent que par l'ascese on en "force." on en hate la" maturation" D"ou un absolu non plus statique et dogmabque comme I'atman des bralunanei:., ma1s dynamique et negatif, eschatolo­g1que et non theologique le nirvdrJa. N1 Hrc, m non-etre, encore moms action; mais absence d'egoisme et par smte extmct1on <le toute relativ1te. Sat, cit, ananda n'offrent, pour decrire cet absolu, aucun sens.

Ces deux solutions extremes, mais pas s1 opposees qu' elles ne se concil1ent clans la notion de brahmaniri•iir,w (Gita ll, 72, V, 24), mar-

Paul M asson-Oursel

quent les deux poles de la speculation indienne. La gamme des ontolo­gies s'egrene, depuis l'ideahsme des vijniinavadins Jusqu'au materia­lisme des carviikas, car ii y a mille fa«;ons, inegalement realistes, d' affecter par de la relativite - upiidhi, karman, r;arira, etc. - I' existence spiri­tuelle. Pour celle-ci realiser son integnte, ce peut etre, comme clans les duahsmes Jama ou vai«;e~ika, s'isoler de ce qui n' est pas elle, realiser le kaivalyam des Samkhyas; et ce peut etre echapper a !'illusion, soit parce qu'on s'y soustrait, s01t parce qu'on la domine en la compre­nant, soit parce qu'on aper«;o1t qu'on la cree soi-meme par ce jeu divin, "la danse de <;iva," rcphque hindouiste du nirmiinakiiya bouddhique.

L'ascese des yogi'ns mene egalement a l'une ou a l'autre des solutions extremes. Les simples ne vo1ent clans l'affranchissement qu'une puri­fication morale aboutissant, par mamtes restrictions et rigueurs, a l'apathie Dhyiina, samiidhi sont des fa<;ons de vider la pensee, afin d'atteindre ii la paix clans la vacmte II s'agit moins de sauver ]'esprit que de le volatiliser Mais les introspectifs diale'cbciens, les yo_giiciiras s'achemment, a travers des difficultes toujours plus ardues, vers la conquete de "terres" tOUJOUrs plus precieuses ils defricht'nt un sen­tier sur lequt'l les porte une" methode," un veh1cule Ceux-la n'isolent ni ne suppriment l'esprit, ils le realisent Ils parviennent au-dela du stade ou cet esprit s'oppose a quelque chose d'autre; ils depassent le bien comme le mal, le vrai comme le faux, mais sans arreter le progres dans l'etabhssement d'une transcendance Ils n'esbment pas que la negation des oppositions marque le terme de !'effort liberateur cet effort se poursmt en siidhana, en mcorporation de ]'esprit affranchi clans un monde par Jui transfigure, OU, SI ]'on pref ere, dont ii est l'auteur et qui done est d1gne de Im. Le transcendant peut redevenir immanent - dehin, r;aririn, etc - car aucune boue ne saurait souiller le lotus

Ains1 l'Inde envisage de differentes mameres la question que nous lui posons· qu'est, ou que fait, ou que devient l'esprit erige a la posses­sion d'e sa veritable nature (S'1,1abhiim). La reponse est zero, ou etre plus qu'etre, OU vacmte - ce qm ne veut pas dire neant, - OU fecon­d1te infime. Divergences d'a11leurs auxquelles le mystique se montre moms sensible que le log1cien, car le plus humble yogin, qu'il soit brahmane ou miidhyamilm, se £latte de posseder tout ensemble la science de l'illus10n et des pouvoirs surnaturels par la vertu de ses macerations

L'lnde ne fut Jamais assez bouddhique pour renoncer a poursuivre l'absolu; c'est meme clans le Bouddhisme qu'elle en a cherche les plus audacieuses approximations Elle ne fut non plus jamais assez boud­dh1que pour expulser radicalement de sa notion d'absolu 1'1dee du karman. Quoiqu'il soit surabondamment certain que c'est l'acte qui

L'Auwnomie Spi:rituelle aelon la Penaee Indienne 148

asservit, ii doit y avoir un acte qui n'asservit pas, mais au contraire traduit l'autonomie spirituelle Au dire des Jainas c'est celui qui s'ac­compagne de la le<;yii blanche, sans aucune cooperation des facteurs materiels de vie Selon les Bhagavatas c'est le svadharma de chacun d'apres sa caste, si on l'accomplit sans ego1sme, par amour pour Bhaga­vat. Dans les doctrines d'epoques diverses apparues sur les confins iraniens, c'est la luminosite propre de l'&me, que de l'opacite peut voiler, mais que r1en ne saurait eteindre, et qui, exaltee, dev1ent splendeur im­mesurable, amitiibha, en une transposition imagee des theories euro­peennes sur !'argument ontolog1que, disons que cet eclat se mamfeste lui-meme en mamfestant les tenebres. Prestigieuse metaphore solaire, aussi ind1enne que gnostique, et qm symbolise a merveille l'efficace de la connaissance l'atman ou le V~Jii<ina, comme le vou,; d'Aristote, est en principe toutes choses, quo1que des conditions de fait restreignent sa portee les Vai<;e~ikas expr1ment cette ver1te en compensant !'omni­science de l'atman par l'atom1c1te du manas, organe necessaire de toute perception

Sous ce biais comcident l'etre et la pensee en leur reahte supreme, comme !'existence sensible (dharma) est l'obJet nature] de !'esprit ernpir1que (manas) Tant vaut la connaissance, tant vaut l'etre, l'une et l'autre sont des aspects de l'acbon. L'acte absolu est celui qui ex1ste en droit, l'acbv1te relative celle qui ex1ste en fa1t Asservis en fait, nous sommes libres en droit, si nous ne meconnaissons pas notre essence Ce qm oppose, pour parler comme la ph1losoph1e europeenne, !'existence a !'essence, c'est le p01ds du karman accumule. Ce karman, a la difference du sem1tique pecM or1ginel, s'use pendant que le temps s'ecoule, qumqu' a mesurc 11 se reconst1tue, sauf chez le dehvre-v1vant (jivanmukta). D'autre part, a la difference de l'v>..17 des Grecs, 11 ne re­presente nullement la contre-partie logiquement necessaire de l'energie propre it l'mtelhg1ble alors qu 'Aristote ne pouva1t adrnettre de formes sans rnabere, l'Inde con<;mt des dharmas nnmateriels, <;akti ne recouvre pas MvaµL,; Pour nous soulager du fardeau karm1que vertu et intel­ligence suffisent, tand1s qu'ailleurs seul un Dieu peut effacer la tare prirnord1ale et faire de l'homrne un elu. Pour amener a l' existence les phenomenes c'est assez qu'entrent en Jeu les conditions de la pensee empmque, point n'est besoin d'une volonte divine qm se surajouterait a I' entendement parfait Ces raisons d1spensent la speculation indienne d'elaborer une theologie, ou, lorsque les sectes en e}..igent une, inspirent une theologie autre que celle d'Occident Toutes ces differences result­ent de ce que l'espr1t tel que l'Inde le con<;01t, au lieu de combiner des "idees," des "images" relativement exterieures a lui, comme nous le

144 Paul M asson•Oursel

prejugeons traditionnellement depuis Socrate, Platon - et Democrite, consiste en une fonction operatoire, creatrice de ses objets. Artha ne designe-t-il pas le terme provisoire d'un processus, un but, une fin, au lieu de connoter une substance, une chose etrangere a l'intelligence? L'lnde se 1oue de la contradiction. De meme qu'a ses yeux le dharma ne tend qu'a se rendre inuble, elle n'a denonce la servitude du karman que pour faire sais1r quelle "verite "appartient a l'acte par excellence, celui qui clans l'autonomie comprend et cree. Quelque information sur l'md1amte fait croireque cette civihsation n'a enqu'unetheorienegative de la delivrance, une mibation plus profonde montre que la-bas aussi, que la•bas surtout l'absolu fut hberte

UNnERSITE DE PARIS

MIMAl\lSA UND V AISESIKA

BY HERMANN JACOBI

DAS Mimarpsa Sutra unterscheidet sich in einem wichtigen Punkte sehr wesenthch von den ubrigen philosophischen Sutras wahrend

namlich diese darauf angclegt sind, die betreff enden philosophischen Syst('me vollstandig darzustellen, finden sich nur wenige philosophi­sche Grundsatze im Mimarp5a Sutra,' und zwar im ersten Adhyaya, vornehmhch in de!,sen erstem Pada, der darum den N amen Tarkapada fuhrt, die ubrigen elf Adhyayas aber haben eme ganz andere Aufgabe: sie lehren die Grundsatze, nach denen aus den oft unbestimmten, ja zuweilen w1dersprechenden Vorschriften der Brahmana.s und Samhitas die richt1ge Darbringung des Opfers festgestellt werden soil :Es ist das keine Philosophic im eigenthchen Smne Die 1\Iimamsa-Plulosophie, wie sie als ems der 5echs philosopluschen Systeme bekannt ist, hat nicht den Verfasser de5 Sutra zum l:rheber, sondern ist <lurch die Tatigke1t scmer Kommentatoren ms Leben gerufen worden Die Grundlage bilden die philosophischen Erorterungen des al ten V rtti­kara, welche Sabarasviimm im Bhasya zu l\I S 1 I, 5 z1tiert oder viel­leicht nur 1m Am,zuge mitte1lt Der Verfasser der al ten V ritt1 1st nicht wie Keith memt, 2 Upavarsa, da er d1esen als eine Autor1tat anfuhrt, wegen 5emer Polem1k gegen den Sunyavada muss er spatn als ~00 n. Chr gesclmeben haben 3 Das vollstand1g ausgebaute System der Mimamsa-Plulosoph1e gehort emer v1el spateren Ze1t an, es hegt in zwe1 }_,a5sungen vor, diem manchen Emzelhe1ten von emander abwei­chen, als Gurumata des Prahhakaraguru, der etwa um 600-650 n Chr. anzusetzen l5t,4 und ah, Bhattamata des Kumiirilabhatta, der wahr­scheinhch m der ersten Halfte des 8 Jahrh n Chr lebte.

Der Gegenstand des 1' I S , dem etwa 14 Funfzehntel des ganzen Werkes gewidmet !.m<l, I!.t, me oben angedeutet, eine techni!.che D15z1plm der Opferpr1ester D1ese D1sz1plm mus5 sehr alt sem, so alt wic das vedanga Kalpa; denn ohne diem 1hr entw1ckelten Grundsatze der Auslegung der Brahmana und Samhita zur richbgen Am,fuhrung

1 Als Sigel fur die plnlosopluschen Si.itra bed1cnr 1ch m1ch lner w1e m J A O S. xxx1, p 1 ff, der Anfangsbuchstaben 1hrcr Namen MimiiI]lsii Si.itra, Brahma Si.itra, Va1~e~1ka Darsana, Nyiiya Dar~ana, Yoga Si.itra

2 The Karma MimiiI]lsii, p 7 f 8 J A O s . XXXJ, p 24 4 Keith, l c , p 9, n. !l

146 Hermann Jacobi,

des Opfers hatten die Kalpasiltra nicht abgefasst werden konnen. Thibaut (Arthasamgraha, p. vi) definiert diese Disziplin: "as a body of rules enabling us to construct on the ground of the Veda a Kal­pasiitra or prayoga " Diese Arbeit war aber bereits geleistet vor der Zeit unseres MS; denn dasselbe erortert i 3, 11-14 (im Kalpasiitra­svata}:tpriimiinyadh1kara:i:ia), die Frage, oh das prayogasastra autori­tativ sei, und entscheidet dagegen. Daraus folgt, dass das M S weniger einem prakhschen Interesse client, als einem theoretischen. Dies zeigt sich auch darin, class in 1hm nicht wie in anderen vedischen Dis­ziplmen die Lehrsatze einfach vorgetragen, sondern nach einem festste­henden Schema (piirvapaksa, uttarapaksa, siddhiinta) diskuhert und bewiesen werden. Diese Methode der Diskussion im M S. pragt der Mimiirpsa den Stempel einer spekulahven Wissenschaft auf, die wohl Anspruch darauf machen konnte als eme Art von Philosophie be­trachtet zu werden So ist es verstandlich, dass ihre Anhanger den Antrieb empfanden, die wenigen philosophischen Grundsatze 1m M S. zu emem vollstandigen phdosoph1schen System auszubauen

Wenn wir nun untersuchen wollen, welche Stelle die Mimarpsa in der Entwicklung der md1schen W1ssenschaften emnimmt, musscn wir m erster Lime uns<:>re Aufmerksamkeit auf die philosophischen Lehren richten, die 1m 1. Adhyaya des M S erortert werden und die theore­t1sche Grundlage des Ganzen b1lden Es handdt sich zunachst um folgende zwei aufs Engste m1teinander verbundene Grundlehren: (1) Zw1schen dem Worte und dem, was es bedeutet, besteht eine uran­fanghche (d h nicht ze1tlich erst eingetretene) Verbindung i 1, 5 autpattikah sabdasya 'rthena sambandhah (2) Das Wort (und uber­haupt der Ton) ist ewig, d h das Wort ist immer und uberall latent vorhanden und wird, wenn es ausgesprochen wird, mcht allernst zum Dasein (<lurch ulpatti), sondern nur zur Erschemung (durch abhi­vyakti) gebracht, i 1, 6-23.

lch erwahne h1er noch ein weiteres sprachhches Problem, namhch ob das Wort das lndiv1duum (dravya) oder die Species (iikrti) bedeute, welche Frage i 3, 30-33 erortert und im letzteren Sinne entschieden wird.

Aus den be1den ersten Grundsatzen, welche die Ewigkeit des Wortes und seiner Verbmdung mit dem, was es bedeutet, lehren, folgt, class der Veda von Ewigkeit zu Ew1gkeit besteht, gewissermassen als das ewige Weltgesetz. Das begrtindet die unbedingte Gultigkeit seiner Vorschriften, auf denen der dharma beruht (codaniilaksano dharmah i 1, 2). Die Autoritat des Veda w1rd durch seine Ewigkeit gewahrleistet und nicht etwa dadurch, class die hochste Gottheit, der

Mimamaa und V aisefika 147

Hvara, den die Mimlilp.sakas leugnen, sein Urheber sei. So dient also die Lehre von der Ewigkeit des Wortes als theoretische Grundlage fur die theologische Disziplin, welche den eigentlichen lnhalt des M.S. bildet. Aber letzterer macht kemeswegs jene Lehre notwendig. Denn auch bei der entgegengesetzten Theorie, namhch class der Isvara den Veda verkundet babe, was die Yogins annahmen, wurden die in Adhy. ii-xii des M.S. aufgestellten Regeln und Grundsatze fur die Auslegung der Brahmana zur r1cht1gen Darbrmgung des Opfers unverminderte Gultigkeit haben. Die technische D1sziplm hatte off enbar schon lange bestanden; denn s1e war unentbehrhch fur die Opferpraxis. Erst nachtraghch empf and man das Bedurfms, sie auf eine theoretische Grundlage zu stellen. Und wir erfahren am, MS 1 1, 5, classes Bada­raya:i:ia war, der sICh zu der Ans1cht von der ewigen Verbmdung des Wortes m1t seiner Bedeutung bekannte Bedenkt man nun, class die Interpretation der Brahma:i:ias zum Behufe der Opferpraxis den Ausu­bern d1eser Kunst keme drmgende Veranlassung gab, das metaphy­sische Wesen des Wortes zu ergrunden, so hegt die Vermutung nahe, <lass s1e die Anregung zu derle1 Spekulationen von anderer Se1te em­pfingen Nun finden wir che5elben oben angefuhrten Grundsatze, welche im M S aufgestellt werdm, auch be1 den Grammatikem Katyayana begmnt sem Vartt1ka m1t den Worten siddhe §abdarthasambandhe, diese g1bt PatanJali mit Auflosung des Kompos1tum also w1eder: siddhe §abde 'rthe sambandhe ce 'i1, 1 und erklart siddha, m1t nztya Der Sinn 1st also. "Dm, Wort, seine Bedeutung un<l die Verbm<lung beider sind ew1g" Im Sarngraha, !'iagt PatafiJah, werde die Frage d1skutiert, oh das ,vort nd1Ja oder karya ~e1, und zugunsten von miya entschieden 2

Ka1yata bemerkt hierzu, <la~~ <ler Samgraha em hcsonderes Werk (granthat'isesa) se1, und Nage;a g1bt an, dru.s Vya<;h es m 100000 Sloken verfa!'ist habe, Bhartrhari 1m Yakyapadiya nennt da~ Mahabha~ya Samgrahapraflh•aiicuka 3 Am Sehlus .. e semer Du,kussion im :Maha­bhasya 4 enbche1det sJCh PatafiJcth dafur, <lass nitya nicht nur das unveranderhch Ewige bedeute, sondern auch das Bcharrcnde, <lessen ,ve~enhe1t nicht zugrunde gche iad api mtyam, yasmims tatit'am na vihanyate Kirn punas iatfrarnP tadbhams tattvam ah rtar apt tatitJarn na vihanyate '• Auch die Species (akrii) al~ solche geht mcht zugrunde" Dies fuhrt uns zur dr1tten der oben angefuhrten Lehren 1m 1\1 S , nam­hch, da~s das Wort die Species bedeute. Katyayana erortert zu i 2,

1 Mahabha~va, ed K1elhorn, vol 1, p 6, I 2 2 Ibid , p 6, I 21 3 S K Belvalkar, Systems of Sanskrit Grammar, p 81, n 9 4 L C' p 7, I 21 f

148 Hermann Jacobi

64 v. 33-59 die Frage, oh das Wort das Individuum (dravya = "!Jakti) oder die Species (akrti = jati) bedeute. Die erstere Ansicht vertrat Vya9-i (v. 45), letztere, der Katyayana zustimmt, Vajapyayana (v. 85).

Es steht also fest, dass die Lehren bezilglich des Wesens und der Bedeutung des Wortes sowie der Verbindung beider bei den Gram­matikern mit den in 1\£ S. vorgetragenen durchaus tibereinstimmen. Dafur, <lass sie bei den Grammatikern entstanden seien, lasst sich anfuhren, dass sogar schon vor Yaska die Frage, oh das Wort ewig oder nicht ewig sei, diskutiert worden zu sein scheint. Er erwahnt namhch und verwirft die Ans1cht des Audumbarayal).a, nach der das Wort indriyanitya sei, mnerhalb des Sinnesvermogens (des Menschen) Bestand habe, d h. nur in dem horbaren Laute. 1 In der grammatischen Literatur zwischen Pal).ini und Katyayana, die nach des Letzteren Andeutungen ziemlich bedeutend gewesen zu sein scheint, wurden die betreffenden Probleme weiter erortert und Patafijah zieht dann das oben angegebene Fazit. Die Vermutung liegt nahe, dass die Fragen uber das Wesen des Wortes zuerst von den Grammatikern aufgeworfen worden sind. Aber dass die Mima:rpsakas die betreff enden Lehren von jenen entlehnt haben, kann nur eme genauere Untersuchung wahr­schemhch machen. Zunachst lasst sich zeigen, class zwischen beiden genannten Gelehrtenkreisen eme engere Beziehung, bestehend in der wechselseitigen Rucksichtnahme der einen auf die Begriff e der anderen, obgwaltet babe Emerseits namhch werden im M S. i sprachhche Fragen mit besonderem Interesse erortert; ausser den anfangs her­vorgehobenen drei Grundsatzen wird m i 3, 24-29 uber die apabhram­sas und Kasusverwechslungen, i, 3, 8-10 uber Mlecchaworter, i 1, 24 ff. uber die Entstehung des Satzsinnes gehandelt Darum werden die MimiilllSakas, die in ihrer speziellen Disziplin es nur mit viikya zu tun haben, dennoch als padaviikyapramiinaJna bezeichnet. Anderseits berlicksichtigt Katyayana Vorstellungen der Mima:rpsakas m semer Erorterung Uber die Bedeutung des Wortes, so bezieht er sich zu i 2, 64 in v. 44 u. 47 auf die codana (vedische Vorschriften) und in v 89 auf das Dharmasastra.

Betrachten wir nun die Diskussion uber die oben genannten Grund­satze bei den Grammatikern eingehender. Katyayana erkennt die Ewigkeit des Wortes an (nitye sabde), ohne sich naher darauf einzulas­sen. Durch Patafijali erfahren wir, dass dieses Problem im Sarpgraha, als <lessen Verfasser Vyac# gilt, erortert und entschieden worden sei. Da aber dieses Werk uns mcht erhalten ist, so bleibt naturlich unent­schieden, ob die Diskussion in M.S. i 1, 6-23 auf ihm beruhte oder

1 L1eb1ch m ZDMG, Neue Folge Bd 2, p 211

Mimamaa und V aise1ika 149

originell ist. -Die Frage nach der Ewigkeit der Bedeutung (artha) wird von Katyayana nicht unter diesem Titel behandelt; ihre Beant­wortung ergibt sich aber aus seiner Untersuchung Uber die Bedeutung des Wortes i 2, 64 v. 35-59. Wie schon oben angegeben, stehen sich zwei Ansichten gegenuber, nach der des Vya<;h bedeutet das Wort das lndividuum, nach der des Vajapyayana die Species. In der alten Terminologie, deren sich Katyayana durchaus bedient, sowie auch das M.S. an der betreffenden Stelle, wud Indiv1duum mit dravya, Species mit akrti bezeichnet. Bei den Phdosophen und ilberhaupt den spateren Schriftstellern sind daftir die Ausdrtlcke vyakti und jati (samanya) Uhlich geworden. Es ist nun beachtenswert, dass sich Patafi.Jah neben der alten auch der neuen Termmologie bedient. Das dti.rfte darauf hindeuten, dass er junger als der Verfasser des M.S. ist.

Die Untersuchung uber die Bedeutung des Wortes hat nun bei Katyayana fo]genden VerJauf. i?Zuerst werden v. 35-44 fur VaJapya­yana' s Ansicht, dass das Wort die Species bedeute, mehrere Grunde geltend gemacht. Dann (v. 44) tritt der Verteid1ger von Vya<;h's Ansicht, class das Wort das lndividuum bedeute, auf mit zwei Grunden fur dieselbe (v 46, 47) und funf Emwanden gegen die des Gegners, (v. 48-52), von denen der erste eme grossere Tragweite hat und sich gegen die selbstandige Existenz der Species richtet. Dagegen zeigt dann der Anhanger ViiJapyayana's, <lass die vom Gegner fur seine Ansicht angefuhrten Tatsachen sich ebensowohl erklaren lassen, wenn das Wort die Species bedeutet (v 53-55), und widerlegt dann die gegen dieselbe erhobenen Emwande einzeln in derselben Reihenfolge (v. 56-59)

Viel kurzer wird derselbe Gegenstand in M S i 3, 30-35 behandelt. Im Piirvapaksa werden drei Grunde dafur angefuhrt, class das W ort nicht die Species bedeute, sondern das lndiv1duum, und d1ese werden im Uttarapaksa widerlegt. Die Diskuss1on, fur die nur die siitras, nicht das v1ele Jahrhunderte jungere Bha~ya in Betracht kommen dtirfen, bewegt sich ganz im ldeenkreis der Mimamsakas; die von Katyayana vorgebrachten grammat1schen Grunde bleiben unberuck­sichtigt. Dern Verfasser des M.S ist die Existenz oder Subsistenz der Species eine ausgemachte Sache, er schemt die Kontroverse vorgefun­den und semen Standpunkt gemass seiner Lehrc von der Ewigkeit des Wortes gewahlt zu haben. Dagegen macht die Darstellung der Kon­troverse bei den Grammatikern den Eindruck, <lass sie auf ihre Urheber zurUckgehe.

Die von Katyayana v. 48 erwahnten Grunde gegen und fur die Annahme, class die Species gesondert neben oder Uber den lndividuen

150 Hermann Jacobi

bestehe und in ihnen zur Erscheinung gelange, haben fur unsere Un­tersuchung ein besonderes lnteresse, wie sich in weiteren Verlaufe zeigen wird Der Gegner macht v. 48 (nai 'kam anekadhikaranastham yugapat) geltend, class, wenn es nur eine Species gabe, sie <loch nicht in allen ihr zugehorigen Individuen gle1chzeitig erscheinen konne, was PatafiJali dannt erlautert, class der eme Devadatta nicht gleichzeitig in Srughna und m Mathura se1. Die Widerlegung dieses Emwurfes erfolgt m v 56, das ich hier m1t Patafijah's Erklarung in Text und Paraphrase wiedergebe. nai 'kam anekadhikaranastham yugapad ity iidityavad visayah ("M1t der Behauptung, class) sich em Ding mcht an mehreren Orten gleichze1tig befindet, damit verhalt es sich wie mit der Sonne."

PatanJali na khalv apy ekam anekadhikaranastham yugapad upa­labhyata ity adiiyavad visayo bhavisyati; tad yatha: eka iidityo 'nekadhi­karanastho yugapad upalabhyate. - "msama upanyasah: nai 'ko drasta 'nekandhikaraastham adityam yugapad upalabhate." - evam tarhi:

it£ 'ndravad visayah tad yathii: eka Indro 'nekasmin l.ratu§ata ahilto yugapat sarvatra bhavati. evam dkrtir yugapat sarmtra bharnsyatz. "Dass nicht em Dmg an mehreren Orten gleichzeitig gesehen werde, damit wird es sich verhalten wie mit der Sonne Namhch die t'ine Sonne wird an vcrsch1edenen Stellen gle1chze1tig geE>ehen. • D1ese Erklarung passt mcht auf urn,eren Fall, mcht em und derselbe Beo• bachter sieht die Sonne gle1chzeitig an mehreren Stellen ' Dann muss man(statt ity iidityar,ad visayah) E.agen. iti 'ndramd visayah. damit ver• halt es sich wie m1t Indra N amhch em und derE.elhe Indra, der bei mehreren Hunderten von Opfern gleichze1t1g angerufen wird, ist an allm gleichze1hg zugegen. So wird auch die Species gle1chzeitig uberall sein."

Daraus ergibt s1ch als die Sachlage folgendes Man hatte den Genusbegriff mcht als einc Abstrakt1on aus den 1hm untergeordneten lndividuen erkannt, sondern dachte sich die Genera als transzendente Wesenhe1ten von ew1ger Dauer (generaha ante rem bei den Scholasb­kern) Jedes Genus (oder Jede Spez1es, was die m iikrti hegende Vor­stellungswe1se besser andeutet), tritt m1t allen lnd1v1duen m Ver­bmdung und hat nicht m ihnen semen Sitz ( avinaso 'nasntatvat v. 57). Die Spez1es, z B Kuh, ist nur eme, der mdividucllen Kuhe g1bt es zu alien Ze1ten eme unbeschrankte Anzahl, wie ist es aber denkbar, class die eine Species gleichze1tig m1t allen m Verbindung stehe, alE>o gleich­zeitig an vielen Orten se1? Man verwies dei.halb auf die Sonne, die auch nur eine ist, aber uberall zu sem scheint, Mit d1eser Erklarung gab sICh Katyayana zufrieden. Pataftjah aber fand, class die Analogie

Mimti:rriaa und V aiie~ka 151

mit der Sonne zutreffend sein wUrde, wenn derselbe Beobachter die Sonne gleichzeitig an mehreren Stellen sahe. Das ist aber nicht der Fall, deshalb kann mit der Sonne das fragliche Verhaltnis nicht erklart werden, und darum erklart er es mit lndra's gleichzeitiger Gegenwart an vielen Opferstellen. Der Punkt, auf den es hier ankommt, ist der, dass Katyayana einer Erklarung zustimmte, die Patafijali als unzu­treffend beiseite schiebt

Den eben behandelten Gedankengangen ganz ahnliche finden sich auch in der Mimarpsa, zwar nicht in der Untersuchung uber die Bedeu­tung des Worte1,, sondern m der uber seine Ew1gke1t Gegen dieselbe wird 1m Purvapaksa MS. i 1, 9 (sattvantare ea yaugapadyat) geltend gemacht, class dasselbe Wort gleichzeit1g an versch1edenen Orten gehort werde, es ware aber unmoglich, dass, wa'> nur an einem Orte ware, an verschiedenen Orten ersch1ene Das ewige Wort ist geradeso w1e die Spez1es nur ems und soil <loch allerorten sein Darum ist die Antwort auf den Einwurf d1eselbe h1er wie dort; iidityavad yaugapad­yam, i 1, 15. "Mit der Gleichzeitigkeit verhalt es sICh wie mit der Sonne " Hier haben w1r also die unklare, im emzelnen nicht durchge­dachte Analogie, die auch dem Katyayana genugt hatte. Darum halte ich, namenthch m Verbmdung mit den fruher angefuhrten ln­dizien, die Folgerung fur berechtigt, da1,s der Verfasser des M.S. und Katyayana derselben Periode angehoren, und <lass die Abfassung des l\I.S mcht spater als PataiiJah anzusetzen sei, v1elmehr etwa zwischen 200 und 300 v Chr

Zurn Schlus<,e dlf'ses Abschmttes muss noch eine Textschwierigkeit im Mahabha5ya erortert werden. Dieselbe Stelle des l\Iahabha~ya zu 1 2, 64, v 56, die eben behandelt wurde, l'>teht namhch schon vorher p 243 Doch lautet das vor dem bhasya stehende varttika bier: asti cai'kam anekadhikaranastham yugapat, und die Worte iti 'ndravad visayah, welche an der zweiten Stelle die von PataiiJali vorge1,chlagene Verbes1,erung enthalten, en,chemen h1er nach Kielhorn's Annahme als zweiter Teil des varttika, der also vom ersten <lurch em Stuck des bhasya getrennt ist. Das ganze varttika wurde somit lauten: asti cai 'kam anekadhikaranastham yugapad iti 'ndravad visayah. In dieser Form bietet das varttika weder syntaktische noch sachliche Schwier­igkeiten. Man versteht dann aber mcht, weshalb Katyayana es nicht auch an zwe1ter Stelle in d1eser Form gebracht babe, 1,ondern in einer, die rnit jener verglichen eine entschiedene Unrichtigkeit enthalt. lch nehrne darum an, class an erster Stelle das vartt1ka (40) mit bha~ya ein spaterer Zusatz ist, vielleicht auch v. 41, soda1,s sich dann v. 42 in naturlichem Zusammenhange an v. 39 anschliesst. An zweiter Stelle

Hermann Jacobi

(v. 56) aber kann das viirttika nicht entbehrt werden, da es die Erwide­rung auf den Einwurf in v. 48 enthalt.

Der Lehre von der Ewigkeit des Wortes musste der Umstand ein besonderes Gewicht verschaffen, dass sich die beiden angesehensten Gelehrtenklassen des alten Indiens zu ihr bekannten. Die Grammatiker galten namlich von je als die ersten Gelehrten (prathame hi vidvdmso vaiyakaranah, vyakaranamillatvdt sarvav'ldyanam. Dhvanyaloka, p. 47), und die Mimarpsakas, denen man die ehrende Bezeichnung padavii­kyapram<ina1na beilegte, durften ihnen als Gelehrte im Range nicht viel nachgestanden haben; denn die streng wissenschaftliche Methode, wie man wohl ihre Erorterung der Grtinde pro und contra zur Feststel­lung Jedes Lehrsatzes nennen muss, handhabten sie mit gleicher Meisterschaft. Aber trotzdem erhob sich gegen die Lehre von der Ew1gke1t des Wortes eine heftige, und abgesehen von den Grammati­kern und Mimamsakas uberall siegre1che Opposition. Dieselbe ging aus, soviel w1r sehen konnen, von den N aturph1losphen und Buddhisten. Erstere batten die phys1kah~che N atur des Schalles erkannt, womit die Lehre von der Ewigkeit des Wortes unvereinbar ist, und letztere stimm­ten ihnen bei, we1I nach ihrer Grunduberzeugung es kein ewiges, un­veranderhch Seiendes gibt (ausser den asamskrtadharmas: Raum und zwei Arten von nirodha 1). Die so aufgeworfene Streitfrage scheint lange die wissenschafthchen Kreise Indiens aufs lebhafteste interessiert und erregt zu haben, wie sehr, ersieht man nach der zutreffenden Be­merkung von E. Abegg 2 schon daraus, class in der ind1schen Log1k eines der gewohnlichsten Schulbeispiele fur den Schluss der Satz von der N1chtewigkeit des Tones, und fur den Fehlschluss der von seiner Ewigkeit bildet. Die Erorterung dieses Gegenstandes in dem M.S hat noch, ich mochte sagen, einen akademischen Charakter, und verrnuthch wares ahnhch auch im Sarpgraha des Vya9i gewesen, Aber den grossen Fortschritt in der Erkenntnis brachten er~t die Untersuchungen der Naturphilosophen. Wir lernen sie zuerst kennen aus dem Vaise~ika Darsana des Kanada Bevor 1ch aber die einschlagigen Stellen des V.D. bespreche, muss ich eine Bemerkung uber unser Verstandnis dieses schwier1gen Textes vorausschicken.

Die Erklarung des V.D. beruht Illcht auf einer alten, ununter­brochenen Ueberlieferung Die Bibliotheca lndica Ausgabe des V.D. von Jayanaraya:r;ia Tarka Paficanana (1861) enthalt den Kommentar des Sankaramisra (etwa 17. Jhd )3 und den des Herausgebers. Ein

1 Stcherbatsky, The Central Conception of Buddhism, p 106. 2 Festschrift fur Wackernagel, p 225 3 Vgl Keith, Indian Logic and Atom1sm, p 35 f.

Mimdmad und VaiJesika 153

noch jtingerer Kommentar ist der des Candrakanta Tarkalankara, Calcutta 1887. Alle diese drei Kommentare widersprechen einander in der Erklarung vieler siitras, und Sankaramisra weicht von der alten, nicht erhaltenen V rtti in manchen Punkten ab, wie er in seinem Kom­mentar an den betreff enden Stellen angibt In schwierigen Fallen, die leider nur allzu haufig im V.D. sind, verh.,gen s1ch also die Erklarer aufs Raten und folgen ihr Eingebung, offenbar we1l kein autoritativer Kom­mentar nach Art von Vatsyayana's Bhasya zum Nyaya Darsana be­stand. Und ebenso scheint es au<'h schon zu Uddyotakara's Zeit gewesen zu sem Denn wenn Ihm ein (fingierter) Gegner vorwirft, er setze sich in Widerspruch zu emem bestimmten siitra des VD, so ist seine stereotype Antwort: 1 "nem, weil Du den Smn des siitra nicht verstchst" (na, ,'futrarthapar13n<in<it) U<ldyotakara wurde wohl nicht eine falsche Erklarung dem Gegner in den Mund gelegt haben, wenn ein bha5ya oder sonstwie bebtelter Kommentar vorhanden gewesen ware, woraus die r1chtige zu entnehmen war. Dagegen war Jenes wohl moghch, wenn nur <ler Text der siitras ohne Kommentar vorlag Fur dieselbe Annahme spricht die Tatsache, class Prasastapada sem Werk Bha~ya nannte, obschon es kem Bha~ya zum VD 1st, sondern eine selbstandige systemat1sche Darstellung des V aises1ka Das wtirde er wohl nicht haben tun konnen, wenn bere1ts ein Bhasya oder eme V rtti zum V.D bestanden hattc 'W1r werden daher zu dem Schlusse ge­drangt, class die Erklarung des VD. mtmdlich m der Schule der Vai­se~ikas uberliefert wurde, ohne in einem Kommentar schr1ftlich niedergelegt zu werden Dazu schemt es uberhaupt mcht gekommen zu sein, vermuthch well das Auftreten der N aiyayikas das lnteresse der N aturph1losophen in neue Bahnen lenkte Be1 der MimiiIPSii scheint die Entw1cklung anfangs ahnhch wie be1m Va1se~ika gewesen zu sein Der alteste Erklarer des M S , der namhaft gemacht wird, ist Upavar~a, <lass derselbe tiberhaupt der erste gewesen se1, lasst sich ebnsowenig beweisen wie das Gegenteil Aber 1ch bin durchaus geneigt zu der Annahme, class die Erklarung des Sutra sehr lange mundlich in der Schule der MimiiIPsakas uberhefert wurde, ehe em schriftlicher Kommentar entstand. Denn der Wortlaut der siitras, namenthcb im ersten Adhyaya, ist zuwe1len unklar oder gar unverstandhch, und wird dann auch selten von Sabarasvamin, dem altesten uns erhaltenen Kommentar, erklart, wohingegen er immer m1t voller Besbmmtheit den Gedanken angibt, der m dem betreff enden siitra angedeutet sein soll. Dieser stand ihm offenbar fest und war ihm oder schon seinem altesten Vorganger, <lessen Werk er benutzte, <lurch die Tradition der

1 Nyaya Vii.rtbka (Bibi Ind), p 108, 16 iii, 18. SiO, 11. 846, 15.

154 Hermann Jacobi

Schule gegeben, weshalb der Kommentator es nicht fur notig hielt, den Gedanken aus dem siitra selbst, als seiner autentischen Festsetzung, dadurch zu entwickeln, class er die Worte des Sutrakiira in dem frag­lichen Smne erklarte. Bei anderen alteren Sutrawerken, solchen Uber Ritual, Dharma u. s. w, bei welchen der Stoff unverandert derselbe bheb, wie er vom Sutrakara dargestellt wurde, machte es nichts aus, wenn die Erklarung der mundlichen Ueberheferung uberlassen wurde, und smd zu ihnen sicher erst lange nach ihrer Abfassung Kommentare geschrieben worden. Aber philosophische Lehren tragen den Keim der Entwicklung in sich, und wenn sie nicht rechtze1tig schrifthch fixiert werden, dann ergeben sich die Schw1er1gkeiten, die w1r beim Studium des V D und, m germgerem Masse, be1 dem des M.S empfinden. Jedoch 1st es begreifhch, <lass man auch bei diesen beiden altesten philosophischen Sutras nicht von der Schulpraxis, namhch der aus­schliesshch mundlichen Ueberheferung, abwich, die bei Jenen anderen alteren Sutrawerken im Schwange war, bis die Periode der Kommen­tatoren-Literatur emgetreten war

Doch kehren wir zu unserem Thema nach die~er Abschweifung zuruck, die eben begrunden sollte, we6halb ich bei der Deutung der sutras 1m V D unbedenkhch von den Kommentatoren abweiche, wenn der Zusammenhang der sutras unter emander oder die Behand­lung des Gegenstandes m anderen Quellen dies empfiehlt

Die Frage nach der Ew1gkeit des W ortes w1rd 1m V ai;e~nka behan­delt. Der Gang der Untersuchung ist folgender V D 11 2, 23-25 w1rd festge6tellt, das6 der Ton weder eme Sub6tanz (dravya), noch eme Kraft (karman), sondern eme Ezgenschaft (.quna) 6eI Dann werden als Grunde dafur, dm,s er mcht ew1g sei, angefuhrt sato linga­bhiiviit (26) "we1l kem Anze1chen dafur vorhanden 1~t, <lass er besteht (auch wenn er nicht wahrgenommen wird)" (vgl MS i I, 7 a.~th<i.niit) - nityavaidharmyiit (27) "wegen semer Verschiedenheit von :Ew1gem (msofern er ent6teht und vergeht)." anitya§ ea 'yam kiiranafah (28) "der Ton 1st vielmehr nicht ew1g, we1l er auf Ursachen beruht " - na ea 'siddham vikiiriit (~9) Und class der Ton mcht-ew1g, d h verimder­hch sei, 1st eme Tatsache, wed er sICh wandelt, d. h er kann bald lauter bald Je1ser sem - abhwyaktau dosiit (30). N1mmt man an, dass der Ton mcht hervorgebracht, sondern als etwas immer Daseiendes nur zur Erscheinung gebracht werde, so )assen sich die vorher ange­fuhrten Tatsachen nicht erklaren - Nachdem so die Nichtewigkeit des Tones erwiesen ist, wird der Satz ausgesprochen: samyogad vibhagac ea sabdiic ea sabdanispattih (31). "Der Ton entsteht <lurch Verbin­dung (z. B. von Trommel nnd Schlagel), <lurch Trennung (z. B. beim

Mimdmsd und VaiJesika 155

Spleissenvon Bambus), und aus einem Tone (bei der Fortpflanzung des Schalles) "

Hiermit ist die physikalische Seite des Problems erledigt, es folgt die Anwendung des Ergebnisses auf das Wort lingiic ea 'nityah sabdah (82). "Das Wort ist nicht ewig, wegen semes charakteristischen Merk­mals (namlich, dass es aus Lauten, d. h Tonen, besteht) " Es folgen nun <lurch tu m 33 als solchC' angedeutet die Emwande des Gegners. dvayos tu pravrttyor abhii·viit (SS) "Es 1st aber ew1g, we1l sonst Ueber­gabe und Annahme (des Veda) se1tt>ns des Lehrer<; bezw Schulers nicht statthaben konnte," d h man kann nur etwas wirkhch Vor­handenes einem anderen ubergeben oder von 1hm annehmen - pra-thamiisabdat (34) ''Wegen der Benennung, "d1C' n..,te," d. h drC'1mal <ipr1cht er die er1,te sam1dheni-:Formel, worau1, i,ICh erg1bt, class, was er dre1mal spr1cht, em und dasselbe, mcht 1mmer w1eder Neues ist (Vgl 1\1 S 1, 1, ~O samlchyiibhifriit) - .<?amprat1pattibhiivac ea (35) "Und Well man ei- c,vort, Spruch, etc) al ... das1,elbe w1edererkennt, (wcnn da...,<ielbe 1,patn odt'r von emC'm anderen w1ederholt w1rd) "

Nun erfolgt die Ablehnung und W1derlegung des ehm Ge..,agten samdigdhiih sati bahuiN' (36) "D1ese Grunde smd meht emdeut1g (vgl m, 1, 17), (we1l <lw1,elben Erschemungen aueh <la emtreten), woes sich um mehrNe, w1rkhch VC'rsch1e<len{' Handlungen odn Vorgange han<lelt " Z B h{'I Tanz<'n man lehrt, lernt und w1ederholt emen Tanz, memand halt a her <larum den betrdfenden Tanz fur em ew1gC's, tram,zendentt'!> SC'1endes, <la1, dureh che Auffuhrung nur zur Erschemung gebracht (mamfe1,t1ert) werde' (vgl N D n, 2, 29 ua, an.lJafye 'py abh1;a.ws1;6 'parariit) - :mmkhyii.bhii1'ah Mi,m<inyatah (37) ("In 1,ol­chen Fallt'n Prklart s1eh) <lw Angahe emer Zahl au1, <lC'm Gememsamen "; d h das alien Auffuhrungl'D emes Tanzes ete Geme1m,ame, das emhe1thehe Selwma, gilt als die Emhe1t be1 dcr Zahl der emzelnen AuffuhrungPn, d h be1 1hrer Zahlung, und ehC'n!>O gilt <ler 1dent1sche Gedanke eme1, Spruche.., fur <lie Emhe1t der Zahl der "'1ederholungen desselben Spruehes, und .',O we1ter

Es kann kaum bC'zwe1fdt werden, <lass die Polem1k m dC'n angefuhr­ten !-ifltras de1, Va1ses1ka s1eh gegen <lie l\Iimiimsii r1chtet, und dass Kanada d1C' Dar-,tellung dessC'lben Gegemt,m<le1, Ill 1\1 S 1 1, 6-23 gekannt hat Aber eme direkte Bez1ehung der.'!dben auf das M.S. las..,t s1eh hochstens m seinem 37 sfltra erkl'nnen, des..,en Wortlaut samkhyiibluii,ah siimiinyatah <leuthch auf l\f S 1 1. 20 samkhyabhiiv<it hmwe1st Wenn Kanada mcht alle 1m 1\1 S aufgefuhrten Grunde einzeln widerlegt, so hat das wohl darm seme Ursache, <lass das Prob­lem zu seiner Ze1t oft verhandelt worden war und er deshalb den piir-

156 Hermann Jacoln

vapak~a nach dem damaligen Stand der Diskussion einrichten konnte. Ausdrticklich verdient aber hervorgehoben zu werden, dass die dem Vaise~ika eigene Argumentation, die physikalische Behandlung der Frage, dem M.S. und den Grammatikern vollig fremd ist. Dieselbe ist off enbar die Errungenschaft einer spateren Periode, die durch Ka:Q.ada eingeleitet wird.

Das bier Gesagte gilt auch fur die Behandlung des Problems von der Verbindung des Wortes mit dem, was es bedeutet, die sich V. D. vii ~. 14-~0 findet. Die Mimarpsa lehrt, dass zwischen dem Wort und dem, was es bedeutet, eine naturliche, ewige Verbmdung besteht (vgl. oben p. 146 autpattikah sabdasyd 'rthena sambandhah M.S. i, 1, 5) Die Widerlegung 1m Vaise~1ka ergibt s1ch aus physikalischen Betrachtun­gen; denn das Wort ist, wie wir eben sahen, physikalisch ein Komplex von Tonen Der Ton (§abda) ist aber eme E1genschaft, die der Luft (akasa) mhariert, er kann also ausserdem nicht noch eine zweite In­harenz (samavaya), namlich in der Bedeutung des Wortes haben. Aber auch samyoga, der andere physische sambandha, ist ausgeschlos­sen; denn da sam_yoga ein guna 1st, und guna keinen guna haben kann (agunavan 1 1, 16), so kann sabda (als guna) mcht samyoga (emen guna) haben Das 1st <ler Smn von V. D. vu ~. 14: gunatvat. Zwei guna konnen mcht m Verbindung (samyoga) stehen, das mtlsi.te aber der Fall sein be1 i.olchen Wortern, deren Bedeutung ein guna (z B. Farbe) 1st (guno 'pi vibhavyate 15) Da em samyoga durch Bewegung (karma) enb,teht und das Wort als guna ohne karma (nislcriya) ist, so mtlsste die Bewegung ausgehen von der Sache, die seine Bedeutung bildet, dann konnte cs keme Worter geben, deren Bedeutung Dmge smd, die mcht Sitz emer Bewegung smd w1e z B aka&a (mslcriyatvat 16). Und ebem,o konnte es keine 'Worter geben, die etwas N1chb,eiendes bedcuten, weil t>me Verbmdung m1t etwas N1chtse1endem undenkbar 1st. (asati nastiti ea prayog<it 17) Aus die!->en Grunden ergibt sich, <lass Wort und Bedeutung mcht in (physischer) Verbindung stehen: sabdarthav asambandhau 18, son<lern e.!> bcruht auf Konvent10n, <lass man em Wort m best1mmter Bedeutung versteht samayikah sabdad arthapratya_yah 20 Die Kommentare fas-;en die siitras 14-18 als pur­vapak~a, der Jeden sabdarthasambandha lt>ugne, und sutra 20 als sid­dhanta, der als den betrdfenden sambandha die Konvention (samaya) lehre. Die Sache schemt sich auf folgende Weise zu erklaren Mit sambandha schemt man ursprtlnglich das Verhaltm:, zweier tatsachhch zusammengehoriger Dmge (svabhavika odcr praptilaksana) bezeichnet zu haben, also Verbmdung (samyoga) und lnharenz (samavaya) und das dadurch Bew1rkte Daruber geht der Gebrauch von s<imbandha

Mimfi1fl,8ti. und V aiieri,ka 157

im V. D. nicht heraus. Im N. D. iii 2, 43 wird noch der ahaydhi­tasambandha erwahnt. Dann aber wurde sambandha auch Bezeich­nung von begrifBichen Verhaltnissen uberhaupt und die spatere Philosophie, namentlich die Scholastik, operiert mit zahllosen sam­bandhas, wovon man sich leicht <lurch den Artikel sambandha im Nya­yakosa uberzeugen kann. Bei sabda und artha nimmt man so einen vacyaviicakasambandha oder pratyiiyyapratyiiyakasambandha an. Die wirkliche Entw1cklung wird durch Heranziehung einer parallelen Erorterung im N. D. ii 2, 53-54 klar, worauf ich daher hier emgehe.

In N. D 1i 2, 53 piiranapradiihapa,ananupalabdhe§ ea samban­dhdbhiivah wird gelehrt, dass kein sambandha zwischen §abda und artha bestehe, weil sonst beim Aussprechen der Worter "Speise," "Feuer" oder "Schwert" der Mund gefullt, verbrannt oder gespalten werden mtisse. Hier bedeutet zweifellos sambandha die materielle Verbindung des Wortes mit dem Gegenstand, den es bedeutet. Darauf sagt der Gegner: sabdarthavyavasthiinad apratisedhah. 54. "Dies 1st kein Bewe1s gegen (das Bestehen emes sambandha), weil zu Jedem Wort eine bestimmte Bedeutung gehort." Statt sambandha gebraucht der Gegner vyavasthana, off enbar we1l sambandha eme Bedeutung bekommen hatte, die hier unzulasi,1g ist. Die Widerlegung des Einwurfs erfolgt im fol­genden &iitra. na, samayikatviic chabdarthasampratyayasya 55 "Nein, weil das Verstandnis der Bedeutung dei. Wortes s1ch au& der Konven­tionahtat erg1bt " Der Wortlaut dieses siitra stellt die Bezugnahme auf V. D. vu 2, 20: siimayikah sabdiidarthapratyayah ausser Zweifel (die Ersetzung von pratyaya des V. D. <lurch tmmpratyaya ist <lurch N. D. ii, 1, 52 veranlasst). Vatsyayana sagt 1m bha~ya zu d1esem sutra: na sambandhakiiritam §abdadrthavyavasthanam kim tarhz samayak<iritam. "Die feststehende Zugehorigke1t des Wortes zu seiner Bedeutung wird mcht durch sambandha bew1rkt, sondern <lurch Konvenhon (samaya)." Hieraus erkennt man, w1e das entsprechende Va1sesika sutra (20) gemeint ist, und <lass die Kommentatoren m1t 1hrer Dcutung desselben als purvapaksa auf den lrrwege sind Vatsyayana fahrt nach der angeftihrten Stelle fort. yat, tad avociirna aby{!dam iii ~asthiv1s1stasya vakyasya 'rthav1se~o 'nuJnatah sab<larthayoh ~ambandha iti samayam tad avocame 'ti. "Wenn wir sagten (1m bha~ya zu u 1, 52): Die be­stimmte Bedeutung des durch den Gemtlv charakterlSlertenAusdruckes asya idam ( d h. der Gebrauch des Gen1tivs I in diesem Ausdruck) gibt einen Zusammenhang (sambandha) zwischen dem Wort und seiner

1 Der Gemt1v dnlckt allgemem 1rgendemen Zusammenhang (sambandha) zw1schen zwe1 Begriffen aus In v1elen Fallen w1rd er dadurch erk I art, dass zw1schen 1hn und das reg1erende Wort sambandhin gesetzt w1rd

158 Hermann Jacobi

Bedeutung zu, so meinten wir damit die Konvention (samaya)." Diese Argumentation ist nicht ohne Eindruck auf die Mimaqisakas gebheben, wenn sie auch naturlich an dem von Badarayai:ia gelehrten autpattika sabdasya 'rthena sambandha festhalten. Der Vrttikara zu M.S 1 1, 5, p. 12 antwortet auf den Emwand, class der §abdasya 'rthena sambandha (s1ehe I 12 mit d('uthcher Beziehung auf N D 1i, I, 53), nicht em (phys1scher) Zuzammenhang sem konne, w1e karyakarana­bha,,a, mmittarwirmttikabhava, asrayasray1bhiiva, tmmyoga u s. w., fol­gendes yo hy atra ryapaddyah sambandhas, tam ekam na t•yapadisati bhavdn: pratyiiyyasya pratyayakasya ea yah sam3niisam3nilaksana iti "Nur denJemgen Zusammenhang, der h1er anzugeb..,n ware, g1bst Du nicht an denJ<'mgen zwH,chen dem, was zur Erkenntnis gebracht werden soil und dem, was ~w bewirkt, welcht>r beE,teht in dem Namen und dem Benanntt-n •· Da~ 1st aber dt>s Vrtt1kiira We1terb1ldung der ursprunghchen Lehre, m dem Sutra 1,teht davon nichts

Hier moge mir eine Bemerkung daruber gestattet sein, wie 1ch mir die Entstehung der paradoxen Lehre be1 dt>n Grarnrnat1kern und MimamE,akas dt•nke Zu kemer Ze1t konnte e& zwe1fdhaft sem, class das "rort m1t semer Bedt>utung verhunden H,t Daman E,ich aber nur phys1sche Verbmdung vor1,tt,lkn konntC', 1>0 nahm man eme 1,olche auch zw1schen dem \Yort und dern, was e1> hedeutet. an In unklarer Form, darum abt•r mcht mmdt·r w1rksarn, ze1gt 1,1ch d1est' Vortsellungsart un prinut1ven Glauben bt'zughch Zauberformeln u ~ w Dit' altesten Sprachphilosophen gaben dwi-n noch unklaren Yorstellung t'men klaren, besbmmten Aui-druck und erklarten, class das \Vort und die dadurch benannte Sache, fur die <lann erE,t spater die Species gesetzt wurde, m w1rkhcher Verbmdung stt>he, dasb dwselht> uherbmnhch sei, erregte kemen An&toss, 1,ch1en viehnehr selb-,tverstandhch lw1 dem gehe1mrnsvollen We1,en de1> Wortes und der Sprache uberhaupt D1ese Denkart, dw noch den Erorterung<'n bei Katyayana und nu M S zugrunde hegt, wurde tat~aehhch uherwunden <lurch <lw erbtarkende NaturphiloE,oph1t', die wir zuer&t m1 V D und spater nur m Emzel­he1ten weitergebildt•t im N D kennen lernl'n Die 1,patt>ren Anhanger der Lehre von der Ewigke1t dt's Wortt'& und dt>bsen Verbmdung mit semt'r Bedeutung fonden s1ch m1t den Ft>1>tstellungen der Naturph1-losoplne durch emen Komprom11,b ab, msofern dem ew1gen Worte eme gle1ch eWige Fahigke1t (saldi), seme Bedeutung auszusagen, mnewohnt; dam1t war der viicyav<icakasambandha des Nyaya, von dem das MS noch mchts we1ss, tatbachhch anerkannt

Unsere bisher1gm Erorterungt>n gingen von der so gut w1e sICheren Annahme aus, class Kanada die Mimarpsa gekannt habe Zur Stutze

Mimiimsii und V aise~ika 159

dieser Annahme mdge der Hinweis darauf dienen, class er ihr seine Vorstellung tlber das karma, das in der Mimazpsa apiirva heisst, verdankt. Hieruber durfte eme kurze Ausfuhrung am Platze sein. Die Lehre vom karma 1st als relig10se Theor1e sehr alt und lasst sich bis in das Brhadara:i;iyaka zurllckverfolgen, aber erst durch die Mi­marµsa ist sie zu einer "wissenschaftlich" erwiesenen Tatsache erhoben. Das apurva wird durch die Opferhandlung hervorgebracht und besteht als das notwend1g vorauszusetzende Bmdeghed zwischen d1eser und ihrem spater, eventuell erst 1m Jense1ts, emtretenden Lohne. D1ese Theoric 1st m der Mimamsa 1m, emzelne am,geb1ldet, woruber Ganga­nath Jha in der Emle1tung zu ,wmer Uebersetzung des Slokavartt1ka ber1chtet Alles ist Iog1srh deduziert. naturhch unter der Vorausset­zung, dass der Veda absolute Autoritat bes1tzt An d1e1,er zweifelt auch das Va1~es1ka mcht So konnte Kanada das adrsta ( = karma) als ('twas Reale.,, uber Jed('n Zweifel Erhabcnes, betrachten und zur Erklarung selb1,t phys1kah1,cher Vorgange verwenden, wo die rein physikahsche Erklarung mcht au1,re1chte, WH' Dasgupta 1 ausgefuhrt hat \Venn d1e1,er Gelehrte aber glaubt, das'> das Vaises1ka eme alte Mimamsa-Schule se1, so wird das folgende die Unhaltharke1t dieser paradoxen Ans1cht dartun Auf ::,emen Vl'rsuch,2 m V. D 11, 2, 36-37 die Annahnw dn Ew1gke1t des Tones hmem zu mterprebcren, brauche ich mcht we1ter emzugehen und darf meme oben gegebene Erklarung d1eser be1den siitras fur s1ch ioielbn sprechen las-.('n

Die Ablehnung der Lehre von <ler Ew1gke1t de::. Wortes hatte fur Kanada '5 Philo1,oph1e weittragende Folgen Denn dam1t fie! auch <ler Satz von der Ew1gke1t des Veda, auf den die J\Iimiimsakm, d1C' ab­solute Autor1tat dC's Vt'da begrundekn, 3 und 5tatt de::,sen musste angenomrnen werden, dasioi er von emem Vrheber stamme (pauru-5ryati:a des Veda) Der N aC'hweH, findet .-,1ch Y D v1 I, I buddhipiirva vdk1jakr'1r 1•ede "Die Abfa&sung von Satzen 1m Yeda setzt cmen ver­nunftigen lTrheher voraus " 1\lan heachte, <lass t•iikya mcht bloss Satz m grarnmat1schem Smne (khandm•akya) 1st, sondern auch em Komplex von Satzen, dw ernen Gedanken zum Au::,<lruck brmgen (miihm•akya), <lahcr die Mimamsakas definieren apauruseyai·akyam vedah 4

Em Nikya setzt emen Verstand (buddhi) voraus, der den im vakya ausgedruckten Gedanken gedacht hat, buddhi 1i-.t aber em guna des dtman, folglich stammt der Veda von emem hoheren Wesen, und zwar einem allw1.-,.,enden, wed sonst der von 1hm geoffenbarte Veda keine

1 A History of Indian Ph1losophY, Vol 1, p 283 2 L c, p 284, n 1 3 MS 1 1, 27-32 4 Nyayako~a. p 735

160 Hermann Jacom

absolute Autoritiit haben konnte. -Die Annahme von hbheren Wesen wird folgendermassen begrtlndet. In V. D. ii 1 wird gezeigt, dass der Wind (vayu) kein sichtbares lndicium (linga) habe, sondern nur aus einem begrifB.ichen (samanyato dr~tat) erschlossen sei, weshalb wir nichts lndividuelles von ihm wissen (avisesa, 16.), also auch keinen Namen ftir ihn hatten, wenn wir ihn nicht aus dem Veda erfti.hren (tasmad iigamikam. 17). Dann fahrt das Siitra fort: samjniikarma tv asmadvisistiiniim lingam 18, pratyaksapravrttatvdt samjniikarma1J,a}J 19. "Namengebung ist aber ein Bewe1s fur hohere Wesen als wir, weil sie stattfindet auf Grund der W ahrnehmung ( des zu benennenden Dinges)." Derselbe Gedanke kehrt V. D. vi 2, 2 wieder: briihmane sam­jniikarma 8'lddhilingam. "N amengebung im Brahma:r;ia ist ein Beweis fur die Vollkommenheit (des Benenners)." Der Sinn scheint mir fol­gender zu sein Der Urheber des Veda muss Verstand haben, weil er m Satzen redet. In den Brahma:r;ias finden wir Satze wie 3yotistomena svargakiimo ya3eta; wer sie ausspricht, muss nicht bloss Verstand, sondern einen vollkommenen Verstand haben, weil er die uber den Verstand der Sterblichen hinausgehenden Opferhandlungen er kannt und benannthat.

Das 1st alles, was in dem V D. uber den Urheber des Veda d1rekt gesagt wird. Ka:r;iada aussert sICh mcht daruber, oh es eine absolute Gotthe1t, lsvara, g1bt, dem die Off enbarung des Veda zuzuschreiben ware (Lehre des Yoga), oder oh der Veda in Jeder Weltperiode aufs neue von einer zur Allwissenheit gelangten Seele verkundet wird (Lehre des Sankhya). lch glaube aber, dass Kai:iada letzterer Ans1cht zugetan war. Die Entsche1dung d1eser Frage hangt ab, w1e mir scheint, von der r1chtigen Erklarung der sutra V. D. 1i1 2, 19-21, die liber die V1el­heit der Seelen handeln. Dieselben lauten.

sukhaduhkha3nananispattyavisesad aikatmyam 19. vyavasthato niina. ~O. sastrasamarthyiic ea 21. N ach dem altesten Kommentar San­karamisra 1st der Smn des ersten sutra folgender es gibt nur einen einzigen Atman, denn das linga des atman sind sukha, duhkha und jniina, und d1ese entstehen gle1chmassig in allen Seden, fur deren Verschiedenhe1t kem linga da ist. W1e also der iilciisa, <lessen linga der Schall in versch1edenen Teilen desselben entsteht, doch nur einer ist, so sind auch die einzelnen Seelen nur Teile des einen atman Aehn­lich Candrakanta, mit anderer Begrundung, aber gle1chem Resultat J ayanaraya:Q.a N ach Sankaramisra und J ayanaraya:Q.a 1st dies der purvapaksa, und das folgende sutra enthalt den s1ddhanta, wahrend nach Candrakanta siitra 19. lehrt, was der atman (brahma) an sich ist, wogegen 20 die V1elheit der empirischen Seelen im vyavahiira-Zustand

Mima1]1,8d und Vaiaefika 161

erkliiren soll. Sutra 19 ist klar: "Die atmans sind von einander ver­schieden, weil jeder seine Besonderheit hat." Wenn das vorhergehende sutra den von den Kommentatoren hineingelegten Sinn haben sollte, so mtisste dem ndnd in 20 entsprechend in 19 statt aikatmyam stehen eka eva; oder wenn aikatmyam = atmaikyam sein soil, dann mtisste es in 20 nanatvam statt nana heissen. Aber aikatmya bedeutet hier auch nicht, classes nur einen einzigen a/man gebe; denn das beweist doch nicht der fur das azk<itmya angeftirhrte Grund, namlich <lass hinsicht­lich der Entstehung von Lust, Leid und Wissen kein Unterschied obwalte (zw1schen den einzelnen Seelen). Sondern es bedeutet hier wie auch sonst "Wesenseinheit," wom1t ausgedrUckt werden soll, class es zwar viele Seelen gibt, aber mcht versch1edene Arten von Seelen. Sutra 19 ubersetze ich daher: "alle Seelen smd gleichartig, weil un­unterschiedhch in allen Lust, Leid und Wissen entstehen konnen." Es g1bt aber viele Seelen. Dafur wird m 20 die vyavasthci, und in 21 scistras<imarthya als Grund angefuhrt, Letzteres legen die Kommenta­toren als eme Berufung auf die heiligen Schrift aus, als wenn das stitra lautete: srutes ea. Aber das kann scistras<imarthy<it nicht bedeuten; es ist dasselbe w1e scistrarthavatvat B. S. n 3, 33 und bedeutet: weil sonst das scistra (ved1sche Vorschr1ften zu opfern, u s w.) zwecklos (anarthaka) waren, (scistram caivam anarthakam syat. MS. vu, 2, 6) d. h. m unserem Falle. wenn nicht Jede Seele ihre besondere Existenz hatte, so wurde der Lohn 1hrer Opferdhanlungen nicht ihr zugute kommen, und dann wurde eme ved1sche Vorschr1ft wie agnistomena svargak<irno ya1eta kcmen Zweck haben In ahnhchem Smne wird B. S 1v, 2, 17 vidycisiimarthyat gebraucht weil andernfalh, "anarthi­kawa mdy<i syiit." - Wenn noch em Zweifel ubr1g bliebe, ob nach Ka~ada's Ans1cht jedem <itman ind1v1duelles, ew1ges Dasein zukomme, so wird er <lurch das we1ter unten anzufuhrende siitra V 2, 18 gehoben; aus demselben erg1bt sich, dm,s die Emzelseele m der Befremng (moksa) fortfahrt zu bestehen.

Kanada lehrt also ausdrucklich, <lass alle iitmans in 1hrem Wesen (qua atrnan) vollstand1g cinandcr gle1ch smd Wenn er auch hohere Wesen als wir smd anerkennt (V D. u, 1, 18 i.1ehe p 160), so han­delt es sich dabe1 nur um verkorperte Seelen, w1e aus der D1s­kussion uber die verschiedenen Korper 1v, 2, 5-10 hervorgcht: tatra sariram dvundham yoniJam ayonzJam ea 5, worauf er dann m1t funf Grunden beweist, dass es Wesen g1bt, die kemem Mutterle1be ent­sprossen smd santy ayoniJ<ih. Die hoheren VV csen haben aber (siehe oben, ii 1, 18-19) vollkonmmene Erkenntms, da 1hnen alles direkt wahrnehmbar (pratyaksa) ist. Unter ihnen konnte also auch der

16i Hermann Jacobi

Offenbarer des Veda sein, wenn nicht gar mit dem vieldeutigen sutra iv 2, II, vedalingiic ea dies ausdrUcklich gemeint ist. Jedenfalls kam Kanada mit der Annahme der Gle1chheit aller t'itmans zur Erklarung der von 1hm anerkannten Tatsachen aus und steht soweit auf demselben Boden w1e die Sankhyas, weJche ja auch die Gleichhe1t aller purusae lehrten Aehnlich scheint die Sache auch noch 1m ursprtinglichen Nyaya gelegen zu haben Wie sJCh dann in ihm die Lehre vom lsvara entw1ckelt hat und d1eser er'lt bpat als Urheber des Veda anerkannt wurde, babe 1ch an anderem Orte 1 darg<'legt. Hier erwahne ich den Nyaya nur, we1l er ze1gt, <lass ahnhche VorstcJlungen uber den oder die Verfassn de~ Veda w1e im Vaises1ka auch noch lange nach der Abfassung des V D m Geltung standen

Am bedeutsambkn aber ist der prmzipwlle Gegem.atz, in den sich das Vaises1ka hmsichthch des dharma- Begriffes zur Mimarnba gesetzt hat DI(> Mimamba erklart dab, wa~ dn Veda vorschre1bt, als dharma: codanalaksano dharmah 1, I, 2, was sICh naeh dem Bha~ya etwa so para­phras1eren lasst dharma 1st em Gut, deb~en charaktenbtisches Merk­mal der m ved1schm Vorsehr1ften heg<'nde Antr1C'b zu Opfnn etc ist D1ese Defirntwn. die uber das H'esen des dharma nichts am,bagt. mochte den Theologen genugen, rncht so den Ph1losoph<'n Ihr setzt Kanada eine andere entgegm yato 'bhyudayamhsreyasas?ddh1h, sa dha-rmah (1, I, 2) "dharma 1st da-., wodurch unser ze1tliches 'Wohl und ew1g<'s Heil zubtande kommt ", und <'r fahrt fort tadm('anad arnnayasya priimiin_yam (1, l, 3) "Weil C'r 1hn lC'hrt, hat der Veda Autontat." Das ist eme Umkehrung des Satzcs der l\firnam'la mcht we1I der Veda absolutt' Autor1tat hat, smd semP Gebote dharma, '>ondern we1l d<'r Veda dharma lPhrt, hat er Autontat' Nach Kanada's Defimt10n ist dharma die Ur&ache von abhyudaya und mh.freyaso, dN dharma im Smne der Mimii.msii hat e<; nur m1t abhy11daya zu tun, m1-,ofern all(' Opfer nur zur Erlangung ze1thcher Guter, &e1 <'S m diesem odn ('mem spateren Dasem, V<'rr1chtet werden Von <l<'m ew1gen Heile, mhsreyasa, ist in der Mimamsa uberhaupt rncht die RC'de, aber im Vm~es1ka 1st es geradt' die Hauptsache, der Zweek semer Lehr<' Das siitra 3 Jautet. dharmm,isemprasfitad dra11yaqunalwrmaNamariya1•?~emsam<U'iiy<i11iim pa­darth<inam 8adharm,ym•aidharmyiibhyam fatt1•0Jiianiiri m hsreyasmn "Durch eine bebondere Art von dharma wird dw r1chtige :Erkenntnis der sechs Kategonen nach 1hrer Aehnhchke1t und Versch1edenhe1t erlangt, und aus 1hr geht das mhsreyasa (die :Erlosung) hervor" In­sofern afao, als die Vai~es1ka-Philosopliie zum mMreyasa fuhrt, fallt sie unter den Begriff d<:'s dharma Dass es Kat;iada mit dieser Charak-

1 Jacobi, die Entw1cklung der Gottes1dee be1 den lndern, Bonn, 199lS, pp 47 ff.

Mimdmad und V aiae~ika 163

teristik ernst gemeint ist, ersieht man daraus, dass er im Adhyaya 6 einige dharmaa, die zum abhyuda.1/a fuhren, eingehend erdrtert, ein Gegenstand, dessen Behandlung sonst in einem philosophischen System gar nicht am Platze ware. Kal).iida steht durchaus auf dem Boden der off enbarten Religion und kann den Anspruch auf voll­standige Rechtglaubigkeit erheben; nur ~o konnte er die Berechtigung des jniinamiirga gegen die Mimiirpsa als Vertreterin des karmamiirga zur Geltung bringen, we1l auch sem Heilsweg sich unter den Begriff des dharma bringen lasst Aber er verstand den jfi<inam<irga <loch nicht so, <lass die Erkenntnis dessen, was er m seinem System ah, Wahr­heit erw1esen zu haben glaubte, unmittelbar zum Ziele fuhre, v1elmehr bedarf es dazu noch des yoga Daruber handeln v 2, 16-18 tadana­rambha <ifmasthe manasi duhkhabhiivah, sa yogah (16) "Wenn dies (namlich der Kontakt, samnikarsa, von <itman, indriya, artha und manas) nicht &tattfindet und da& manas nur m1t dem <itman verbunden i&t, dann 1st der Leib le1dlos das 1st yoga." Das folgende sutra lehrt, <lass der Auszug der Seele aus dem Leibe, ihre Einkehr m einen neuen Leib, die Verbmdungen (d. h Assimilation) von Spe1se und Trank und die von anderen Produkten (die priinas) durch <la.!, adrsta bewirkt werden - tadabhiive samyogabh<ivo 'pr<idurbha11a§ ea moksah (18). ,Venn da~ adrsta aufgehort hat, dann tntt d1l' Erlosung ein, welche in der vollstand1gen Abwef.enheit Jener Verbmdungen und deren N1chtwiederhervortreten besteht " - Ueber die Bedeutung des Yoga fur die Erlangung der .Erlosung (apm,arga) handelt ausfurhlicher N D 1v 2, 38-47, worauf bier zur Erlauterung der Andeutung Kal).a­da's hmgew1esen sem moge Dass dabe1 wirkhch an Yoga-ubungen nach Art der un Yogasutra gelehrten gedacht I.!,t, bewei.!,en deutlich N. D. 1v 2, 42, aranya_guhiipulinadisu yogabhyas6pade.foh, und 46: tadartham yarnamyamabhyam, 1/ogac ea 'dhyatmavidhyupayaih "Zurn Zwecke der Erlosung soll die Seele gereimgt werden <lurch yama und niyama (cf Y S n 30, 31), und Ausfuhrung der Vorschnften des Yoga, die au£ da.!, Selbst Bezug haben " (N ach dem Bhasya smd mit letzterem emige Uebungen gememt, die nur zum Te1l m1t den Y S. ii 29 aufgefuhrten yogangas identisch sind )

Ich fasse das Ergebm~ der vorstehenden Untersuchungen kurz zusammen. Das System der Mimarpsa, wie e.!> im M S. vorhegt, entstand in derselben Per1ode, m welcher Spekulahonen uber das Wesen des Wortes die Grammatiker unter den Vorgangem Katya­yana's aufs lebhafteste beschaftigten Denn dieselben Probleme finden sich auch im ersten Adhyaya des M.S Doch auch in formaler Bezie­hung hat die Mimamsa ein naheres Verhaltms zur Grammatik als

164 Hermann Jacobi

irgend ein Zweig der iilteren Literatur, insofern nur in jenen beiden Disziplinen die strenge Beweisftihrung mit Grunden pro und contra zur Festsetzung der Lehrsatze als Methode der Darstellung zur Aus­hildung und steten Anwe-ndung gelangt ist. Auch die Vorstellungswelt ist iID allgemeinen noch die der vorausgehenden vedischen Periode. Eine neue Weltanschauung, die VOID Lokayata ausgegangen war, tritt uns im Vaise!;iika entgegen, sie besteht in der naturlichen und realis­tischen Erklarung der physischen Erscheinungen und der abstrakten Begriff e, die in der Sprache zum Ausdruck gelangen. Dadurch tr1tt das Va1ses1ka in einen Gegensatz zu dem altertumlicheren Mimazpsii­System, <lessen Lehren von der Ewigkeit des Wortes und des Veda VOID Standpunkt der Naturph1losophie aus betrachtet und abgetan werden. Aber m religioser Beziehung sind beide Systeme einig, msofern auch das Vaise~ika auf dem Boden der offenbarten Religion steht. Jedoch macht sich seine fre1ere R1cbtung, natilrlich innerhalb der Grenzen der Rechtglaub1gke1t, <lurch eme veranderte Fassung des dharma­Begriffes geltend, wodurch auch s<>ine Erlosungslehre (jfianamarga) Anspruch auf Orthodoxie gewinnt

Es ist klar, class zwischen der Abfassung des M.S und der des V. D. ein langerer ZeitrauID liegen IDuss, m dem sich die angedeutete Entwicklung vollz1eben konnte. Wenn meme oben vorgetragene Ans1cht, class das M.S zwischen 300 und 200 v Chr. enti;.tanden se1, richtig ist, dann dlirfte das V. D. in das erste Jahrhundert vor oder nach Anfang unserer Ze1trechnung anzusetzen sein

Nachtrag. Seitdem vorstehende Abhandlung wiedergeschrieben wurde, hat unsere Kenntmss der altesten komIDentarcn-Literatur zur Mimamsa eine bedeutende Bereicherung erfahren durch Professor Dr Krii;,hnaswami A1yangar (Madras) in seinem Aufsatz "A School of South Indian Buddhism m Kanchi," der der Fourth Oriental Confer­ence (Allahabad, November, 1926) vorgelegen hat. Es wird namlich in dem anonymen Prapanchahrdaya (Tr1vandrum Sanskrit Series, No. 45, p 89) folgend~s ber1chtct Ueber das ganze Gebiet der Mimamsa, uIDf assend den tantrakiim;la (M.S ) , den devatakaJ,1<;la und den brahIDaka:i;i<;la (B S.) verfasste Bodhayana einen grossen Kommentar namens Krtakoh Aus dieseID hat Upavarsa in deID seinigen einen Auszug gegeben. Upavar~a's KoIDIDentar 1st dann von Devasvamin weiter gekurzt und auf die beiden ersten ka:i;i<;las beschrankt worden. Auch Bhavadasa schrieb ein "Jaiminiya-bhiiffa," Ihm folgte Saba­rasvamin, dessen Bha~ya nur den ersten ka1,19a umf asst Hieraus ist nicht ersichtlich, oh der Vrttikara mit Devasvamin oder Bhavadasa identifiziert oder von ihnen unterschieden we.rden soll.

Mima11J8d und V auefika 165

Mit den Angaben des Prapanchahrdaya stehen aher weitere von Professor K. Aiyangar beigebrachte Zeugnisse uber Kp:akoti nicht in vollem Einklang. In dem alttamulischen Gedicht Ma:i;timekhalai ist Krtakot1 Name eines Autors, ebenso im Trika~gase~a, brahmavarga v. 19, 23., wahrend in der Va1jayanti (Oppert's edition, p. 95, I. 808) Upavar~a als Krtakotikavi bezeichnet wird.

RHEINISCHE FRIEDRICH-WILHELMS-UNIVERSITAT,

BoNN

DAN])IN AND BHAMAHA

Bv ARTHUR BERRIEDALE KEITH

ALL dates given in Indian literary history," wrote W. D. Whitney ~ in 1879, "are pms set up to be bowled down again," and it seems not unfitting, in honor of one who took upon himself as a labor of love the complet10n of Whitney's version of the Atharvaveda, to seek to demonstrate once more the truth of this aphorism

With his wonted acumen H. Jacobi 1 has lately sought to establish within narrow hmits the dates of two mteresting writers, Dal).dm and Bhamaha His results have been accepted by the learned historian of Sanskrit Literature, M Wintermtz, 2 as well as by Dr. S K. De, 3 to whom we owe a valuable exposition of Sanskrit Poetics. The dates suggested, therefore, may now be regarded as well on the way to definite acceptanc•e, and 1t becomes necessary, accordmgly, to submit to a careful mveshgation the eVIdence adduced, in order to ascertain whether we have now reasonably assured results or merely plausible combinations

We certamly owe it to Jacobi that we have some definite evidence for the date of Bhamaha, beyond the admitted fact that a <'ommentary on his trl'atlse on poetics was written by Udbhata, who was a councillor of Jayapida of Kac;mir (779-813 AD) In his fifth chapter Bhamaha takes occasion to discus1' the nature of inference, a subject eagerly in­vestigated by the contemporary Buddhist logicians, and he mentions as a defimt1on of mference a doctrme thus expressed:

tnriipiil liiigato Jniinarn anurniinam ea lrecana

In Dharmakirt1's N,yriyabindu we have the defimtion · tatra tririipiil liitgad yad anumeye Jniinam tad anum<inarn. Nor 1s this all, in verses ~8 and 29 of the same chapter we find a reproduction m sense of three Sutras (138-140) of Nyiiyalnndu, in, and in them two verbal coinci­dences in definition. This 1s proof of a very strong kmd that Bhamaha knew DharmakirtI's work, and Dharmakirt1 certamly had not attained fame in the time of Hmen Thsang, while 1-tsmg, whose stay in India dates from 673 to 695 A.D, notes him as one of recent fame Jacobi further suggests, very ingemously, that m iv, 7 Bhamaha's ironical

1 SBA, 1922, pp 210 ff 2 Gesch d md Litt, m, 641 3 Sanskrit Poetics, 1, 63 ff

168 Arthur Berriedale Keith

words: gurubhi~ kim vi,vddena ? are a reference to Prabhikara Gum, the famous Mrmatisaka, who was doubtless a predecessor of Kumirila. 1

In this case, however, the evidence is inadequate, and clearly must be disregarded in a serious consideration of the date.

On the other hand Jacobi dismisses, without adequate ground, the important observation of Pathak 2 that Bhamaha's reference in vi, 86 to a N yasakara is an allusion to the commentary of Jinendrabuddhi on the Ka~ka V rtti, which may be assigned to c. 700 A.D. In doing so he rehes on the impression of Kielhorn 3 that the commentary in ques­tion used Haradatta's PadamafiJari, while tradition ascribes Hara­datta's death to 878 .A.D. There seems no real doubt that the recollec­tion of Kielhorn was at fault, and it may be noted that Winternitz does not follow him in refusing to see in the Qu;upalavadha, ii, 1 U a reference to the Nyasa. Without claiming certainty in the case of Magha's reference, 1t may fairly be said that the burden of disproving allusion to Jmendrabuddhi rests on Jacobi, and that his reference to Kielhorn is quite inadequate for the purpose Skepbcism in this case seems the more UDJUstified seeing that the dates accord so well; the use of Jinen­drabuddhi and that of Dharmakirti concur in suggesting 700 A D. as the earliest period for the author

When we come to the case of Dai:i<.hn, we find ourselves deprived of any precise guidance. Jacobi readily accepts the suggestion of Pathak 4

that the threefold division of karman into nirvartya, vikiirya, and prapya set out in the Kavyiidar<;a (iii, 240) is derived from the V akya­padiya (iii, 45-51) of Bhartrhari. But here we have a very different state of affairs from that m the case of the relation of Bhamaha and Dharmakirti, the probability that Da:i;idm is simply adopting a current doctrme, not derived from Bhartrhari's work, 1s very great, and to render this impossible it would be essential to prove that the doctrine was originated by Bhartrhari. Neither Pathak nor Jacobi establishes anything of the sort, and it is perfectly clear that, unless and until better reasons are adduced, this effort to fix an upper limit for Dan9m must be pronounced without value Equally without importance, as Jacobi in these cases recognizes, are guesses at the identity of the king Rajavarman or Ratavarman of Kavyiidar<;a, ii, 279, and efforts to solve the riddle m ni, 314 on the theory that it alludes to the rule of the Pal­lavas in Ka:i'ici. More serious is the proposal of Mahe«_;acandra Nya­yaratna 5 to find use of Bai:ia's Kiidambari in the Kiivyiidar<;a. Acom-

i Keith, Karma-Mimtinati, pp 9 ff 2 IA, xh, 235 Cf JBRAS, xxw, 19 ff. 3 JRAS , 1908, p 499 ' IA , xh, 237. 5 Cited by Peterson, Da<;akumtiracar'lla, 11, S, note.

Da,µ/,in and Bha:maha 169

parison, however, of the two passages yields no such result; the Kddambari (p. 102) has

kevalam ea nisargata eviibhiinubhedyam aratniilokocchedyam apra­dipaprabh<ipaneyam atigahanam tamo yauvanaprabhavam.

The Kavyddar~a (ii, 197) has

aratniilokasamhiiryam aviiryam suryara~mibhih drstirodhakaram yilniim yauvanaprabhavam tamah.

The assumption that the Kddambari was the source of the verse in Da1:uiin seems without possible ground, and none of those who have approved it have suggested on what point their claim of borrowing is based. If there is the relation of borrowing, every consideration sug­gests that Bai;ia is the person indebted, and that he has endeavored to elaborate and improve on his model. But it is really gomg too far to stress such a resemblance In the world of Kavis long before 600 A.D.

we may assume that many tried their hands on so obv10us and tempting a theme as that enshrined in the verse and in the Kddambari. Even were the verse the model of the Kddambari, it would not throw decisive light on the date of Dai:idm, as it may not be more than a quoted verse, which Ba:r:ia used independently.

Still less attractive is the effort to show that Kavyiidarr;a, iii, 302:

ratnabhittisu samkriintaih pralibimbar;aiair vrtah jniiw Laiike,;varah krcchrdd A iiJaneyena tattvatah,

is derived from ()u;upiilamdha, 1i, 4.

ralnastambhesu samkriintapratlmiis le cakiir;ire ekiikino 'pi paritah pauru~eyavrtii wa.

Jacobi himself candidly admits that a similar idea is found already in the Kddambari (p. 131), and nothmg 1s more unwise than to trust as evidential these variations of well-worn themes

This completes Jacobi's proofs for an upper limit of date; he accepts the fact that Vamana knew Dandm and used his work, and Vamana he assigns to the reign of Jayiipic;la. It follows, therefore, that for Jacobi the upper and lower limits of date for Dandm and Bhamaha are much the same From the arguments, however, adduced above against Jacobi's views, it follows that, while Bhamaha cannot well have written before 700 Ao. the date of Da1;u;lm 1s wholly undeter­mined, and that there are no external criteria which induce us to bring him into close relationship in time with Bhiimaha.

170 Arthur Berriedale Keith

Jacobi, however, claims that comparison of the doctrines laid down in the two works extant shows that Dangin was a critic of Bhamaha, not Bhamaha of Dal).gin, and he has in his favor the view of commen­tators on the Kiivyiidar<;a, such as Tarunavacaspati, who here and there asserts that Darn;lin is criticising Bhamaha 1 It is, however, clear that, apart from the fact that the commentators are of late date, and are often clearly wrong m their explanations of Da1;u;lin, no stress can be laid on such assertions as evidence of date. What the commentators were interested in was not the chronological sequence of doctrines; when Dangm criticized something which occurred or appeared to occur in Bhamaha, he was set down as criticizing that view without regard to the que&t10n whether Da:r;u;lm had Bhamaha before him or merely some forerunner who enunciated a view adopted by Bhamaha. And, fortunately, we are not left m doubt as to the exi&tence of a predecessor, whom Bhamaha cites no les& than twice, a fact sigmficant of his im­portance TlH& authority, Medhavin, has left no work for us, as usual m India, the more complete treatise has &uperseded the older, and it is lucky that we even know of his existence. Mcthodolog1cally 1t is obvi­ous that m any places where Dal)dm seems to criticize Bhamalia we are bound to ask whether 1t 1s not rather a cai,,e of dealmg with the views of Medhavm or perhaps rather some other predecessor of Bhamaha.

To assume that 1t must be Bhamaha, inmply because Bhamaha expresses similar views to those cr1tJc1zed, 1s logically quite madm1si.ible, and, if actual use of Bhamaha 1s to be established, it mui,,t be on the strength of precii,e parallels of very special quahty Such an example, for instance, exists m the case of Bhamaha and the author of the Bhatti-Ka,•ya Bhamaha, writmg of riddles, declares (u, 20) ·

kav,yiiny api yadim<ini vy<ikhy<igamyani ~·astravat utsavah sudhzy<im eva hanta durmedhaso hatiih.

Bhatti (xxii, 34) says of his own poem

,,yakhyiigamyam idam kiivyam utsa11ah sudhiyiim alam hat<i durmedhasa(} casmin ,ndvatpriyatayii mayii

There can be no doubt as to borrowmg here, and the borrower 1s shown decisively by the fact that the claim 1s as adequately motived by the character of Bhagi's work as it is artificial in the case of Bhamaha's apphcabon of the idea to riddles. The mstance is extremely mforma­tive; it proves absolutely that Bhamaha was fond of usmg predecessors anonymously, and renders it as natural to expect to find 1m1tation of

1 Cf Hari Chand, Kaluliiaa et l'art poMique de l'Inde, p 76

Dar.uJ,in and Bhamaha 171

Da1_19in and criticisms of his doctrines as to trace allusions to Bhamaha in Da:r:u;lin. Yet it may be noted, as indicating the complexity of argu­ments of this kind, that Sovani 1 regards the passage in Bhatti as an allusion to Bhamaha, and K P Trivedi 2 interprets the passage in Bhamaha not as a eulogy of riddles, but as a condt>mnation of lack of simplicity m poetry m general. It 1s clear, therefore, that no conclu­sions of value can be based on superficial investigation

The passage which seems dec1s1ve to Jacobi 1s that in which Da1_19m enumerates the ten Dosas of poetry The fo,t agrees verbally with that given by Bhamaha, but the latter follows it up with (iv, 3) ·

pratiJfiiihetudrstdntahinam dustam ea nesyate

On the other hand Darn;lm continues (m, 126)

pratt3fiiihetudrstiintahiimr doso na vety asau viriirah karka<;ah priiyas teniilidhena kim phalamP

Dan«;lm thus du,misbes al', unattractive and frmtless any discussion as to whether there should be recognized an eleventh Dosa m the shape of a logical defect He doeb not precisely refme to recogmze it, he accepts ten as cntam and leaves the eleventh problematical To Jacobi 1t seems clear that it was Bhamaha, who, bemg deeply inter­ested in matters logical, invented the eleventh Dosa The priority of Bhamaha would thus be secure, but there seems no reason to accept the assertion whieh is not supported by any evidence 'What is clear is that some authority proposed this new Do~a, and that it was the subJect, ru, every mnovat10n 1s, of considerable discussion. Da:r;u;lin thought the matter not worth pursumg, while Bhamaha was mter­ested m it, pr10r1ty on elther side i,eems excluded as a legitimate deduction from the evidence Indeed, if we imitate Jacobi's own pro­cedure 8 in the case of the lists of Alamkiiras, we would ascribe to Medhavm or some other predecessor of Bhamaha this mnovat1on, to which Darn;lm and Bhamaha should be deemed to have reacted in dif­ferent ways, as m the Cabe of the Alamkiira lists In that instance Jacob, holds that Bhamaha followed generally his source, makmg chiefly an effort to simplify, while Dai:idm developed new bub-d1v1sions in order to display his critical power

The second argument adduced by Jacobi for Bhamaha's priority rests on the remark made by Dandm (ii, ~~O) at the close of his illus­trations of the figure Atu;ayokti, hyperbole·

1 Bhandarkar Commemoration Volume, p 393 2 Ibid , p 411 3 SBA , 1922, p 218.

Arthur Berriedale Keith

alamkdrdntardiidm apy ekam dhu~ pardya1J,am vdgiqamahitdm uktim imam ati<;aydhvaydm.

This he holds to be a sarcastic reference to the doctrine of Bhamaha (ii, 85) which seems 1 to make out that hyperbole lies at the basis of every figure of speech ( vakrokti).

sai~d sarvaiva vakroktir anayiirtho vibhavyate. yatno 'syiim kavinii kiiryah ko 'lamkiiro 'nayii viniiP

Da:r;ic;lin, in Jacobi's opinion, reJected this doctrine of Bhamaha on the score that he himself held the view that the quality, metaphorical ex­pression, lay at the root of poetry (ii, 100).

tad etat kiivya.,arvasvam samadhir niima yo gu'}Jah kamsiirthah samagro 'pi lam enam anugacchati

This doctrine of Da:r:ic;lm's, m his view, marks an advance from the /point of view of supporters of the Dhvani theory of poetry from that of Bhamaha, in so far as metaphor has for poetry the function of con­veymg to us what 1s not and what cannot be expressed in words

The argument is ingemous, but scarcely capable of standing close examination. The suggestion that Dandm is attacking the doctrine that hyperbole plays a part in all figures by substituting for 1t the doctrine that metaphor lies at the root of poetry, has no foundation in Da:r:icJm's own words There 1s no trace of any connection between the two ideas in his mmd, still less of his feelmg them to be repugnant. He merely mentions that some hold that hyperbole 1s involved m figures without either endorsmg or denymg the doctrine, and in a totally dif­ferent context he exalts the importance of metaphor Of the Dhvani doctrine he shows no knowledge. Jacobi 2 claims that he 1s referrmg to the Dhvamkara in the first verse of his treatise where he alludes to earlier views m the words bhiiktam ahus tam anye. But Trivedi 3 is equally convmced that it is to Bhamaha that the reference is made, relying on Anandavardhana's observation 4 regardmg Bhamaha's doctrine of hyperbole in its relation to figures generally The only safe conclusion 1s that the passages in either author have no definite relation to the other; we must free ourselves from the delusion that what is preserved 1s all that existed, and remember that Dai:ic;lin and Bhamaha had before them a wide literature which for us is lost probably forever.

1 See Anandavardhana, Dhvanydloka, p 208 2 SBA., 1922, p 'l25 3 Bhandarkar Cammemoratum Volume, p 419. 4 Dhvanyaloka, pp 207, 208, 211

Da'IJ,</,in and Bhdmaha 17S

The danger of seeking to read Da1;u;J.in in the light of Bhimaha is strik­ingly illustrated by Dr. De's 1 interpretation of Kavyadar,a, ii, 220, which he interprets as an effort to reconcile his views with those of Bhimaha, a view disproved sufficiently by the fact that Da1,u;lin merely gives the view as one expressed by some, and thus does not make it his own.

A third argument at one time adduced 2 - perhaps now abandoned

- by Jacobi rests on the remark of Da1a1<;lin (ii, 363):

bhinnam dvulhd svabhiivoktir vakroktu; ceti viiftmayam.

which is, he holds, based on the dictum of Bhamaha (i, 30):

yuktam vakrasvabhavoktyii sarvam evaztad isyate

But there seems no ground for this claim. The term Svabhavokti in Dai:t<;lin applies to a figure, indicating minute description by character­istics, as his definition (11, 8) ·

n<inavastham padiirthanam rilpam saksad vwrnvati svabhavoktu; ea J<itu; cety adya salamkrtir yatha

and still more his illustrations show To suppose that he uses the term in ii, 363 in a d1ff erent sense of ordinary speech generally is clearly illegitimate. In Bhamaha, on the other hand, the usage is clearly other. We are definitely told (1, 36).

vakr<ibhidheya<;abdoktir ist<i vacam alamkrtih

and m n, 93 the recogmt10n of Svabhavokti as a figure 1s given as by other writers.

svabh<ivoktir alaml,ara iti kecit pracaksate arthasya tad avasthanam svabhavo 'bhihito yatha.

The two positions differ essentially, and Dan<;lin's appears to be the older. Bhamaha goes beyond the standpoint of Da1,19in, he insists that mere description, however truly 1t expresses the essentials, 1s not enough to make an Alarnk<ira. There 1s reqms1te an element of Vakrata, or from another pomt of view of Ati,;ayokti, thus Bhamaha preludes the position later laid down more completely, and now con­sciously, by the V akroktiJivitak<ira, which finds m J' akrokti the essence of poetry. From his standpoint Bhamaha 1s qmte m order in denying to Dai;t<;lin's figure Svabh<ivokti, the name of a figure.

1 Sanskrit Poetics, 11, 62, n. 21 2 ZDMG., lx1v, 755.

174 Arthur Berriedale Keith

But Da:r;H).in represents a much more natural view, and one which may with practical certainty be regarded as earlier than that of Bhamaha The point 1s worth closer consideration because it has been obscured by Jacobi's treatment, and misapprehension of It has ren­dered less effective his discussion of Vakrokti. 1 He holds still 2 that Dal).gm and Bhamaha use Svabhavokti in two quite different senses. In the one bense it is merely ordmary description, in the other it is a form of Alamkiira The nature of this form of Alamkara Jacobi asserts to be die Darstellun,q einer platonischen [dee, supporting this view by the term Jati which is also applied to it. But this really Is not m accord with the much simpler view of Dal).<}m, which fortunately is made clear by his examples as well as by his definition He illustrates in sequence mstances of J ati, K riyii, Guna, and Dravya, the first by a descr1pt1on of the characteristic features of parrots, the second by the movements of an enamoured dove, the third by the qualities of the contact with the beloved, and the fourth by a description of ~1va with his characteristic marks, and he remarks that this sort of description prevails in ~astras. What 1s meant, therefore, is nothing ph1losophical, but an exact description of essential characteristics Why is this called an Alamkiira by Dal).dm ? Because he has (11, 1) a wide definition of Alamkiira which makeb all things which lend beauty to a poem Alam­kiiras ·

kiivyar;obhiikaran dharman alamkiiriin pracaksate.

Dal).gin in fact is takmg the natural view which suggests itself on analysis of any Kavya It must contain descriptions which do not substantially deviate from those m ~astras, as well as what may be termed generically Valcrokti. Dandin, therefore, 1-,olves the problem of poetry for h1mbelf by acknowledgmg the plam fact, the stock m trade (viifimayam) of the Km•i consists of Svabhavolcti and Vakrokti, and the two fall under the generic head of Alamkiira In the former of the two divisions there is clearly no place for the use of double entendre, (;lesa; in the latter it is bpecially appropriate as the first half of Kavyiidar<;a. ii, 363 tells us

(}lefah sarviisu pusniili priiyo vakroktisu r;riyam.

Bhamaha refines on this, and in him we find Svabhiivokti in a new light; he does not deny that Kavyas must contain Svabhavokti, but he has adopted a theory of Alamkiira based on Vakratii or Ati(}ayokti, which, v

1 Cf ZDMG , bnv, 751, 754, 758, BernheJD1er, 588, n. I 2 SAA , 1922, p 225.

Da'f)</,in and Bhamaha 175

as we have seen, Dai:i9in merely records as held by some, and, accepting this theory, he cannot adopt Svabhavokti as an Alamkara. This is an obvious and simple explanation of the facts and one which prevents us from holdmg that Dal).dm in the same book of his work, and writing on the same topic, uses Svabhiivokti m two perfectly different senses without warnmg to the reader. Moreover we escape the necessity of reading mto DarnJin's Svabhiivokti, as Dr De 1 must do, a refined in­terpretation which is not suggested by Dal).9m, and which is plainly quite incompatible with his examples.

It is due to this misunderstandmg that Jacobi 2 claims, as a fourth clear case of borrowing by Dandin, Kiivy<idar,;a, ii, 235

hetu,; ea silksmale<;au ea 'l'iiciim uitamabhii.sanam

as compared with Bhamaha, u, 86.

hetu<; ea silksmo le<;o 'tha niilamkarafaya matah.

Here, however, there 1s obviously no evidence of borrowmg by Dal).din, and Jacobi later seems to regard the treatment of these figures by the two authors rather as mstance& of drlf erent reactions to an older list with which both dealt m their special ways As a matter of fact, how­ever, great importance attaches to the words which follow in Bhamaha and give his reason for reJectmg the figure& m question

samudiiyiibhulhanasya vakroktyanabhulhiinatah

It is clear that Bhamaha rejects the figures as instances of Alamkiira on the same ground that he reJects Smbhiit'okti, it is that they all lack the element of J' akroktt Smbhiit'Okfi expre&ses J ati and so forth, these Samudiiya, without an element of Vakrokti This deliberate use of Vakrokti 1s the determming pomt of a more mature and elaborate view than the simpler attitude of Dai:idm The process of development is clear some authority developed the theory that Ali<;ayokti was in­volved m Alamkaras, this was merely current, not generally accepted, in Dan9m's day, by Bhiimaha's time or by Bhiimaha himself the doc­trine was carried to the pomt of msisting that Ati<;ayokti was essentially involved in the concept10n of Alamkiira, and those forms of expression which did not contain this feature were refused rank as Alamkiiras

Jacobi, it may safely be assumed, would no longer lay stress on similarity of the definit10ns of Bhavikatva which in Dai:i9in (ii, 364) runs.

1 Valcrokti;ivua, p xiv, n 17 2 ZDMG., lx1v, 754, n 1, but see SBA., 1922, p 218

176 Arthur Berriedale Keith

tad bhavikam iti praku'IJ prabandhavifayam gu1J4m bhavah kaver abhiprayah kavyesv asya "Yavaathiti'JJ.

and in Bhamaha, iii, 52, presents almost the same first line followed by

pratyaksii iva drr;yante yatriirtha bhiltabhavina'IJ.

To this there are many other parallels 1 which could be added, but it is obvious that nothing can be made out of such similarities, which may merely be due to common derivation from current theories and contain no conclusive sign of relative priority.

Jacobi, however, still holds, 2 though with less confidence than formerly, to the behef that the discussion of the relations of Kathii and Akhyiiyikii in Kiivyadar,;a, i, 28-30 is aimed at that in Bhamaha, i, 25-9. Da:r;u;lin, with much good sense, dismisses the attempt to dis­criminate in essence the two forms of literature, while Bhamaha ac­cepts as valid criteria formal distmctions of no aesthetic value. Jacobi holds that Bhamaha cannot have known Dai:idm's work, since other­wise he must have taken note of his destructive criticism, and that Dai:ic;lin is, therefore, the later, unless indeed some predecessor of Bhamaha had similarly treated the subJect This he deems unlikely, especially as Amara has qmte a different d1stmcbon of the two classes of literature This is clearly unconvmcing. As Jacobi himself admits, Dai:ic;lin's criticism was not sufficient to keep Anandavardhana, Abhina­vagupta, Rudrata, Hemacandra, and others from upholding the dis­tinction of the two classes, and there is not the slightest ground for supposmg that Dai:ic;lm knew Bhamaha's text. The latter lays down for the Katha·

anyaih svacaritam tasyiim nayakena tu nocyate svaguniiviskrtim kuryiid abhiJiitah katham 1anahP

Da:r;u;hn has

niiyakenaiva viicyiinyii niiyakenetarena vii svagurJiiviskriyii doso niitra bhutiirthar;ansinah.

It is impossible from these passages to assert that Danc;lin, not Bha­maha, is the borrower, taken by themselves the statements may each be the source of the other by way of opposition, or they may be based on current controversy without definite relation A decision on the pomt really depends, apart from other considerations bearing on the relative priority of the authors, on the question whether in the im-

1 Cf Har1 Chand, op. cu., p 72. s SBA , 1922, p i15.

DaTJ.,c/,in and Bhamaha 177

mediately preceding passage we have a critique by Dai;u;lin of BhAmaha or by Bhamaha of Dai;u;lin, and reasons will be adduced below to show that the latter is the true explanation of the relation of these passages.

Jacobi does not adduce, and probably would not accept, certain of the arguments adduced by other scholars to establish the priority of Bhamaha. Thus Dr. De 1 holds that when Dai:idin (ii, 244) cites as a valid illustration of the figure Hetu

gato 'stam arko bhatindur yanti vasaya paksinah.

he is attacking Bhamaha (ii, 87) who reJects the figure and describes the verse as bad poetry, to which some give the name -Vartta. The conclusion, however, is wholly 1llegitimate. Bhamaha's mention of V artta shows that he is not attacking Dandm or at least has others in mind, nor is there any reason to show that Dal).~in is dealing with Bhamaha A common use of older material is here strongly suggested. It would indeed be different if we beheved that Dandin's examples were necessarily or even normally his own composit10n, but m the absence of any evidence on this head no reliance can be placed in the conJecture

Nothing again can be established from the fact that Bhamaha (i, 41) gives as an instance of the fault known as Aviicaka the words himapahamitradharair vyaptam vyoma, while Dandm (ni, rno) cites the whole verse a1, an example of the form of riddle called Parihanka. To see in this, with Dr De,2 evidence of criticism of Bhamaha by Dai:ii;lin, is clearly impossible; the curtailed citation nught be taken as proof of the contrary relation, but mdependent treatment of common sources is again a legitimate hypothesis

No other passage seems to have been adduced of any value for proof, and there seems no poss1b1hty of resistmg the conclusion that we have not a single passage in which we can say with any validity that Bhiimaha 1s probably cr1t1cised by Da1_1~m The question then arises whether any passage may be taken as provmg that the work of Da1,1gm was used by Bhamaha. As we have seen, those passages adduced to prove Bhamaha's priority have also been used for the purpose of estab­lishing exactly the opposite conclusion, and probably with about equal or even superior justification. Of other passages many have no pro­bative value, and need not be considered, but there remain points which must be assigned some value.

1 Sansknt Poetics, 1, 65. 2 Sanskrit Poetics, 1, 65

178 Arthur Berriedale Keith

Firstly, it must be observed that, while Bhamaha expressly tells us that he has composed many of the illustrations of the rules laid down in his Kavyiilamkiira, Dai;u;lin in no case refers to or criticises one of" these illustrations. The importance of the pomt is best realized by reflecting that both authors attach importance to their examples, and that, as we have seen, two of these examples are the object of divergent views Like all arguments ex silentw this consideration has no conclu­sive weight, but 1t may be set again&t the probability of use of Bhamaha by Dal).<;lin, and it &uggests, having regard to the importance of Bha­maha's work, that Dandin wrote either before it was written or before it became well known

Secondly, stress has always been laid by supporters of Dandin's priority on the fact that, while he adopts as vita] the d1stmction of the V aidarbha and Gauda styles, the attitude of Bhamaha 1s severely critical (1, 31, 3~).

V aidarbham anyad asfit1 manyant,e sudhiyo 'pare tad eM ea kila Jyayas sadarfharn api niipare Gaudiyam idam etat tu V aularbham ih kim prthak gatiinugatikanyayiin naniilchyeyarn amedhasam.

It is idle to deny the a prwri probability that this is a criticism of Dandin m the usual insultmg manner of Bhamaha Jacobi seek& to remove this impression by pomtmg out that long before Dal).<;lm the poetry of the Gaudas enJoyed no high repute, seemg that Bana char­acterises it as aksaradambara (Harsacarita, ver 7). This argument 1s only technicany vahd 1f one believes that Da])dm has been proved later than Bana, and as we have seen, there is no real evidence or probability of this. Nor is there any value m Jacobi's further observa­tion that Dandin, though setting out the two main styles, recogmzes that there are mtermed1ate shades and types, for even so Bhamaha's criticism would be justified, apart from the fact that Indian critics, like those of other lands, naturally feel themselves entitled to seize upon the salient characteristics of any doctrine which they condemn. ignoring saving causes which might modify their censure. Nor again is it any argument to say that Bhamaha recognizes the division as tra­ditional, for the pomt of the term gatiinugatikanyayiit may well be that he is reproachmg Dal).dm with following blindly a wrong tradition It must in fact be admitted that the view which sees in Bhamaha's verses a deliberate attack on Dan<;lm remains by far the more plausible. It does not attain certainty, but it is a much stronger argument than any adduced on the opposite side.

Dar,uJ.in and Bhamaha

Thirdly, it is argued that Bhamaha's verses (ii, 87, SB):

yad uktam triprakdratvam tasyah kau;cin mahatmabhi~ nindapra<;ansacikhyasiibhediid alriibhidhiyate sdmiinyagunanirde<;iit trayam apy uditam nanu.

179

are directed against the Kiivyiidar<;a, u, 30-32 where the three forms of Upamd are one after the other defined. Dr. De 1 contends that the attack cannot be addressed agamst Da:r;ic;lm, because he does not set up only three kinds of U pamii. This, however, is clearly an untenable view, resting on a misapprehension of the term tnprakdratvam which has a simple sense as directed to a connected group of three, and has nothing to do with the total number of divisions of the U pamii in Dan<;lin. It is also to be noted that Bhamaha (u, 32) reJects miilopa­miidih while Dam;lm (u, 42) accepts the M alii and other forms.

Fourthly, it 1s pomted out that the view of Cunas adopted by Bhamaha 1s a far more advanced one than that of Danc;lm, and is easily explamed as due to the obvious defects of the system of that author The historical process, as traced by Jacobi 2 himself, without apparent recognition of its opposition to his own view, 1s that Bharata gives the Dosas of poetic compos1t10n, and then treats the Cunas as their opposites. Da:r;ic;lm felt this to be inadequate, a5. makmg the essential quahbes of poetry mere negations of def eels, although his­torically 1t was natural to hegm with defects and then advance to the discovery of quahbe~ He, therefore, bases his d1s<'rimmation of the two Miirgas, l'mdarbha and Gauda, on the possess10n by the former of tho~e character1st1cs which make good poetry and winch are lackmg in the latt(•r. But the difference of the dehmtion of the several Gunas in Bharata and Danc;lm sufficiently showed how difficult 1t was to arrive at clear re~ults, and Bhiimaha found a solution by reducmg the number of Guna8 to three, which, though this IS not stated m Bhamaha, rest on fundamental distinctions of the manner m which the mmd of the reader was affected. The three Cunas are jf iidhurya, beauty, OJas, force, and Prasii.da, clearness, and Bhat ta Nayaka 3 bases the distmc­t10n on the melting, the uphftmg, and the expansion of the Cilia. We need not claim for Bhamaha a prec15,e apprec1at1on of the emotional states to which lus Guna.~ were to correspond, but the reduction to three must clearly have been based on some principle of tlus kmd, and in auy event the advance on Dam;hn Is enormous. It Is 5,1gmficant that the Dhvanikara and the majority of authorities from Mammata to

1 Sanskrit Poetics, 1, 68, n 2 a Cf DhvanyiUoka, pp. 68, 70

2 SBA , 1922, pp 223 f

180 Arthur Berriedol,e Keith

Vi~vanatha accepted the new division, and it is a strong argument against Bhamaha's priority that Dai;u;lin evidently ignores it entirely.

Fifthly, it is noteworthy that Bhamaha recognizes besides the or­dinary three topics of the <:astra, GurJ,a, Do~a, and Alamkiira, a fourth, that of the training of the poet. This is fully recognized and interest­ingly explained in later writers such as Vamana, Rudra ta, Raja~ekhara, and Hemacandra, but it is passed over in silence by Da:i:idin. It is, of course, a very natural addition to the ordinary stock in trade of writers on poetics, but the evidence is strongly in favor of its being an innova­tion after Da,;u;lin's period. Jacobi meets this contention by the sug­gestion that Da:i:ic;lin handled the topic, or meant to handle it, in the Kaliipariccheda referred to in iii, 171, perhaps as a counterpart to the treatment by Bhamaha in his fifth chapter of the doctrine of the means of knowledge. This clearly is a guess without weight, and that it should have been advanced without any authority of any kind may be re­garded as satisfactory proof of the difficulty of supposing that DaJ;Ic;lin was later than Bhamaha.

Sixthly, it is significant that Da:i:ic;lin (ii, 366) expressly alludes to the Lakfanas and admits them to the rank of Alamkiiras. These are, of course, the Kavyalaksanas of Bharata which are familiar from their description by V1~vanatha, but which, save for this mention by Da:i:ic;lin and a belated revival by Jayadeva, disappear from textbooks of poetics. This is a clear sign of archaism, and is significant as indicating the process of emancipation of the <:astra from connection with the drama, a process carried further by Bhamaha than by Da:i:ic;lin

Seventhly, against the fact that there 1s the late evidence of com­mentators apparently m favor of Bhamaha's priority must be set the fact that in Namisadhu's commentary on Rudrata's Kiivyalamkiira we find the phrase (p. 2): nanu Dandimedhavirudrabhamahiidtkrtiini santy eviilamkiira<;iistriini? The order in such a passage is naturally that of historical order, and this view is, of course, m some measure supported by the fact that we know that Medhavirudra or Medhavin 1

was actually a predecessor of Bhamaha It is at any rate of greater probative value than the view of the late schohasts. Nor is it in the slightest degree inconsistent with the not infrequent passages in which Bhamaha appears in phrases 2 such as '' the old writers, Bhamaha, and so forth," for Bhamaha was unquestionably for these later writers the head of a school, and it is significant that the phrase sometimes adds to him Udbhata, his exponent and follower in the tradition.

1 Cf P V Kane, JRAS , 1908, pp. 545 f ' Har1 Chand, op cit , p 70

Da1J,</,in and Bhamaha 181

Eighthly, there must be noticed a passage alluded to above which presents considerable difficulty of interpretation. In the Ka:trgiiilart;a (i, U, 22) Dai;t<;lin has:

gunatah priig upanyasya niiyakam tena vulvifiim nircikaranam ity esa mcirgah prakrtisundarah. vant;avirya,;rutiidini varnayitvii ripor api tajjayiin niiyakotkarfavarnanam ea dhinoti nah.

Bhamaha has (i, 22, 23):

nciyakam priig upanyasya van,;avirya,;rutcidibhih na tasyaiva vadham brilyiid anyotkarfavulhitsayii yadi kiivya,;arirasya na sa vyiipitayesyate na ciibhyudayabhiik tasya mudhiidau grahanam stave.

Dr. De 1 seeks to avoid any contact between these two passages by holding that the meaning of the latter is merely '' disapproval of a dis­astrous ending, perhaps in conformity with a similar conventional prohibition in the drama." Now Bhamaha is far from partial to dra­maturgy and it is somewhat surprismg if 1t was from this source that he derived his rule. But what makes this view impossible is the ex­traordinary similarity of language in the two stanzas; it cannot be an accident that Bhamaha has niiyakam priig upanyasya, van<;avirya<;ru­tiidibhih, and anyotkarsavidhitsayii m a context where they directly recall Dan9-in's priig upanyasya niiyakam, varu;aviryan,;rutiidini, and niiyakotkarsat>arnanam. The possibilities of accidental likeness are far exceeded here The argument of Bhamaha i~ quite simple, you must not make a hero of a man by extolling his race, heroism, and learning, and then destroy him to exalt another person as hero, the person who is to come out victorious at the end should be extolled also in the be­ginning. We have a direct attack on Dan9-m's doctrine and it becomes much more probable that the followmg passage in Bhamaha, which in itself is not conclusive, may be really directed agamst Dandin. ·

The conclusion, therefore, seems to be that there 1s sufficient evi­dence to turn the scales strongly in favor of the view that Bhamaha actually knew and attacked Dai:i<;hn. The arguments in favor of this view are much stronger than those adduced against 1t and at the least render the assumption of the priority of Bhamaha extremely hazardous. Possibly the mistaken idea that Bhamaha belonged to c. 600 A.D.,

which was once suggested by Jacob1,2 may have encouraged the view that he was anterior to Da:i;i9-in, and this has been adhered to even when that inaccurate opinion was corrected by its author.

1 Sanskrit Poetics, 1, 68, n !l 2 Bham.,att.akaha,p 54•,n I.

18i Arthur Berriedale Keith

It remains to consider, disregarding the date of Bhamaha, the upper limit to be assigned to Da:J)c,lin. We may assume that the Kavyadar<;a and the Da<;akumiiracarita are by the same hand; the evidence 1 ad­duced against this theory is clearly of no substantial value. We are assured by Rajac;ekhara 2 that Da:J)c,lin wrote three works. The guesses at the third are not happy, Pischel's M rcchakatika, Jacobi's Chando­viciti, may be dismissed, nor is Hari Chand's Anamayastotra worthy of serious consideration. The Kaliipariccheda 1s prima facu more plausi­ble, since the form in which it is referred to by Dal)din (iii, 171) is rather suggestive of a treatise of his own But the very form of the title seems to indicate merely a chapter additional to the three which make up the Kiivyiidar<;a as we have it. If ever written, it may have served as the model of Bhamaha's work on this topic, which ii. alluded tom the Kiimadhenu commentary on Vamana (p 29) The riddle of the third work remains, accordingly, still unsolved The Da,;akumiira­carita necessarily offers us httle material for dating its author, yet something may be deduced from its contents. Its picture of India sug­gests that 1t was not composed under the Gupta empire or that of Harsavardhana, and that 1t may, therefore, fall m the period between these two events 3 The style is certamly not mconsistent with the view that Dandm was a predecesi.or both of Subandhu and of Bana It may be granted that, as the Kiivyiidarr;a 1c, sufficient to show, Darn;lm was well able to display skill in the Kavya manner, and that Ucchviisa VII with its avmdance of any labial letters 1s a dIStinct tour de force. But that does not rn the lt-ast alter the fact that Dal)dm writes a far more natural i.tyle than either of the writers named, and that 1t is much more probable that he preceded than that he followed Bii:J)a.

Efforts, as we have seen, have been made to prove use of Bai;ia, but obviously without any cogency. To these may be added the suggestion of Hari Chand 4 that Tarunavacaspati 1s right when he holds that Da:J)dm's statement (i, 25), that there are exceptions to the rule that the Akhyiiyikii should be narrated by the hero, is an allusion to the Harsacarita of Bana m which the story of Har~a is recounted by Bai;ia himself. As a matter of fact we have not the slightest reason to adopt this suggest10n, which naturally occurred to Tarunavacaspati, because he, like ourselves, had not before him the older .Akhyayikiis and Kathiis on which the rules of Dal)<;hn's predecessors were based. Whether we

1 G J Agashe, IA , xhv, 67 f 2 Har1 Chand, op cit , pp 78 ff 3 Collms, The Geographical Data of the Raghuvamsa and DaAakumaracarita, p. 46. 4 Op cu, p 81

Da1J,</,in and Bhamaha 18S

take Bhamaha's account or that criticized by Da1_19in, we have every assurance that they were framed before Ba1_1a wrote or at any rate be­fore his works became of importance as models. In Rudrata we find that the Harsacarita and the Kadambari have succeeded in winning their way into recognition so that the descriptions of the two types conform generally to their nature. 1 Da1_1din and Bhamaha on the contrary preceded this result, were 1t otherwise, we must have found a very different presentation of the case. We can easily understand by observation of the accounts in the authors of treatises on the drama how defimbons of species of literature were drawn up; they rested on imperfect mductions m which minor points were treated as fundamen­tal; Da:r;i9m reacted against this, while Bhamaha, who is often more wedded to tradition, was content to accept the tradition as handed down.

We may, therefore, place Da:r;ic;lin with some confidence before Ba1_1a and Subandhu That he was later than Kahdasa is indicated by certain hints Thus in Kiivyadar,;a, i, 45:

pra.<tadavat prasiddhiirtham indor indivaradyuti laksma laksmim tanotiti pratit'tSubhagam vacah

there has been seen a reference to the ()akuntalii, i, 20.

malinam api himiinr;or laksma laksmim tanoti.

Again in Kiit 1yiidarr;a, n, 129:

sundari sii bhm,aty evam vwekah kena Jiiyat,e'i! prabhiimiitram hi taralam drr;yat,e na tadii,;rayah.

there is a hint of ()alwntalii, 1, 25

miinusfsu katham vii syiid asya riipa.<tya sambhavahP na prabhiitaralam jyotir udeti i 1asudhiitaliit.

Or agam for Ka,,yiidarr;a, n, 286.

yasyiih kusumar;ayyiipi komaliingyii ru3iikari siidhu;ete katham tarwi hutii,;anavatim citiim'i!

a model may be found m the pathetic address in the Raghuvan,;a, viii, 56.

navapallavasamstare 'pi te mrdu diiyeta yad aiigam arpitam

tad idam visahisyate katham vada viirrwru citiidhirohanam'i!

1 De, BSOS , 111, 515.

184 Arthur Berriedale Keith

These instances might be increased, but, without strictly proving de­pendence, they do give substance to the belief that Dal}.<;lin either him­self used, or cited poets who used, Kalidasa as a model. If Dal}.<;lin chose, he could doubtless easily have written these verses, and it may well be, though we cannot prove it, that a considerable proportion of his illustrations is of his own composition.

If Dai:i<;lin is later than Kahdasa, it is only natural that he should know Bhasa, and there is no real doubt that it 1s from him,1 and not from the M rcchakatikii, that he takes the famous verse

limpativa tamo 'iigiiwi varsativiinjanam nabhah

The conclusion thus suggested, which places Dai:i<;lin some time before 600 A D., would, of course, be entirely overthrown, were we to accept the mgenious theory of Ramakrsi;ia Kav12 which finds a third work for Dai;i<;lin in the Avantisundari, a Kathii, and from that frag­mentary text, supplemented by the metrical Avantisundarikathiisiira, deduces that Dan<;lin was the great-grandson of a certain Damodara, who was the protege of a PaIIava king Smhav1snu, and who was identi­cal with the famous poet Bharavi It must be remembered that the date of Bharav1 1s not defimtely ascertamed, and 1t 1s suggested that Sinhav1~:r:iu reigned about 500 A.D., so that it may be held that the date which would thus be attamed for Dandm as his great-grandi>on would be plausible enough But it 1s clearly 1mposs1ble to accept this evidence seriously, for two reasons, either of which must be conclui>1ve, Firstly, there is the fact that the fragments of the Avantisundari do not in any way identify Damodara with Bharavi, and, secondly, it 1s not in the least clear that even the Kathiisiira, which has no independent author­ity, does anythmg of the sort It merely says (i, 22) of Damodara:

sa medhiivi kavir vidviin bhiiravih prabhavo giriim anurudhyiilcaron maitrim narendre J isnuvardhane.

Nothing but ingenuity will enable us to see in bharavih in this stanza a proper name or B1ruda, and, even if it were <,o to be taken, there is nothing whatever to indicate that the author of the K iriitiirjuniya 1s meant. It may be added that, so far as one can Judge from the de­plorably mutilated Avantisundari, there is no reason whatever to accept identification of the author Da:r:idin, son of VIradatta and Gauri, grand­son of Manoratha, with the author of the Kiivyiidar<;a and Da<;akumiira­carita.3

1 Carudatta, 1, 19, Balacanta, 1, 15 2 Ed of Avantisundari (DaksmabharaU Series, no S, 19H). 3 See S K De, Ind Hist Quarterly, 1, SO ff

Da'f}-</,in and Bhamaha 185

The relation of the Bhat(i-Kiivya to Da:Q.<;lin unfortunately remains dubious, as does the actual date of Bhagi's work, though the mention of King (_;ridharasena of ValabhI assigns it roughly to a period between 550 and 650 AD Moreover, it must be remembered that we have not Bhatti's authority for the names of the figures which he illustrates; these rest on MS. tradition or the commentary J ayamafigalti The result of Jacobi's own researches 1 is to show that Bhatti used a different source, or rather sources, from Bhamaha and Da1:u;lm, whose versions may conceivably go back to a common origm Nothmg is adduced by Jacobi 2 to establish the priority of Bhatti to Dal).dm, and there appears in fact no reason to assume any relation of dependence on either side. In the case of Bhamaha, as we have seen, there is no real doubt as to Bhagi's priority, but 1t is not suggested by Jacobi that in his treat­ment of Alamk<iras he served as the model for that writer

1 SBA , 1922, pp 218 ff 2 ZDMG, lx1v, 139, but see SBA, 1922, p 217, and cf Kane, IA, xh, 208

UNIVERSITY OF EDINBURGH.

S~NGARIC ELABORATION IN SAKUNTALA ACT III

Bv SHRIPAD KRISHNA BELVALKAR

IN an attempt to rise above the recensions of the Sakuntala and to formulate something hke an approximation to its Ur-text 1 the crux

of the critique, no doubt, is the problem of the Srngaric elaboration that we find in the Kashmir-Bengali families of MSS as compared with the Deccan and South-Indian families of MSS In its shortest form, as given by Cappeller, the scene occupies S stanzas and 23 lines of prose, from the exit of the two friends at page 34, hne 14, to their re­entrance with Gautami at page 35, line 24. Patankar's so-called "purer" Devanagari text differs from Cappeller's only by the reten­tion of the stanza " Gandharvena v1vahena," etc., which Cappeller finds "uberflussi& ~ (Einleitung, xn) On the other hand, P1schel's edition of 1877 eHr'arges the scene to about 13 M~!lZ~§ and 2~ prose line~!-w~tle the -~exf" of _the Kashmir MS , apart from shght variations in ~eadings, differs frQ.m Pischel's ecf1tion only by the addition of .l!_!._e -proselinesatferPischers stanza 79, while 1t omits Pischel's stanza so and tlie-two pro5e lines ter;re It, as also-Hie one-lme address behind the cu-rfam totiie C~k~;~ika---1emale after stanza 89 The longer text is accordmgly about four times as large as the shorter, and there is so much scope, and even prima jMie justification, therem for forming a subJective evaluation that I had long despaired of ever bemg able to reach a satisfactory solution of the problem. A chance remark made by that acute French scholar, Professor Sylvam Levi, in his epoch-makrng work, Le Theatre Indien, note 1 to page 182 (Appendice, page 37), st't me, however, on the track, and I believe that 1t is now possible to arrive at a text of the scene neither too short nor too long, a text which retains only the dramatically essential elements, and at which no " Puritan of the type of Momer Williams " needs any longer cavil. I do not propose, for obvious reasons, to give here the actual tentatively constituted text, but only discuss the broad lines along which it can be formulated

The strongest argument of the champions of the Kashmir-Bengali text has been of course the fact that the stanza" Carui;i.a sphuritena,"

1 The prmc1ples underlymg such an attempt, and certam results to be obtamed by thell' apphcabon, I have discussed ma paper recently contributed to Asia Maior, vol. 11, fasc 1, pp 79-104

188 Shripad Krishna Belvalkar

etc. (Pischel, 88), which occurs only in the longer version, has been quoted by Visvanatha, the author of the Sahityadarpar.w,, 1 who belongs to the 6rst half of the fourteenth century. Saradaranjan Ray 2 also points out that in the twelfth century Vardhamiina quotes the latter half of stanza 83 of the elaborated version as of Kalidasa. The re­joinder on the other side is that this simply proves that the " tamper­ing " with the text was already in Bengal an accomplished fact by the twelfth or thirteenth century, and that it is, therefore, no wonder that Bengal writers like Visvanatha show their acquaintance with it. But if now it can be shown that even so early a writer as Sriharlila in the Rainavali (first half of the seventh century) has 1m1tated the longer Srngaric version, 1t will have to be in that case concluded that some part of the " elaboration " probably comes from Kalidiisa himself. Now, Sagar1ka m the plantain-bower, in the second act of the Ratnavali, is a close enough parallel to Sakuntala m the cane-bower in the third act of the Sakuntala. Both are the suffering v1ct1ms of the God of Love, and are mtroduced m the scene with the standard paraphernalia of a bed of lotus-leaves together with a wreath and wr1stlets of lotus­stalks, and so forth The hero m both the plays describes the bed and the other obJects m the bower from which the herome has just de­parted The heroine m Sriharsa's play returns to the bower and over­hears the words of the hero a.-; he places upon his bosom a lotus-stalk trinket, which she had unw1ttmgly dropped down In Kalidasa's play this happens only in the longer 1•erswn of the scene This should es­tablish a prima Jacie presumption m favor of the longer version, pro­mded, of course, that 1t can be proved that Sriharsa was a close student and an 1m1tator of Kahdasa Now, that Kahdasa's works were dili­gently and appreciatively studied at the court of Srihan,a can be safely concluded from the encommm which his court-poet Biil).a bestows upon Kahdasa (l-larsacarita, lntroduct1on, stanza 17)

Nirgai,iisu na va lca.,ya Kiil:ul,iisas1Ja aiilctU1u I Pruir madhurm,iindriisu mafi3ari8vwa Jiiyate 11

And as to Srip.~a'~imitating Kahdasa, we have been able to gather, even in a hurried reading of Srihar~a's Priyadarsika 3 for the purpose, the followmg clear-enough instances, besides the imitation of the " Bee episode " pomted out by Sylvaiili.,_v1

1 N1r Sag , ed of 1922, p 846 I 2 SiiJcunta/,a, 6th ed1t1on, Calcutta 1922, p 284 11 3 Parallels are also to be found m the Niigananda, but 1t 1s not mtended to be

exhaustive on the pomt

Srngaric Elabo'l'atwn in Sakuntala, Act I I I 189

Pnyadarhk4, i, 8. Priyadariikd, 11, s "Sansli sanse raJJantJ, duve ettha um­

mattA " (Priyad., m, 4, 16).

" Diaham ratt1m v1 tuJJha anurii.o " (Priyad., p. 89, end, Vam Vtlas ed, 1906).

"Kusa dava edam uttantam mved1a saJJhaveanam via dukkham kar1ssam" (Priyad., p 87, middle)

" Kamahniba<ldhanuriio v1 mahuarao malad1m pekkh1a ahmavarasassii.da­lampado," etc (Priyad , p 39, end)

" Saalapudhavipar1tti'inasamatthena V ac­charii.ena par1ttii.anti," etc (Pnyad, II, 8, 6)

Mala,nkagnim?Jra, 1, 11; Vikramorvaiiya, 1v, 7;

"Savve sagandhe visasadi; duve v1 ettha iirannaA" (Sakuntala, v, 21, 25),

Also, "Duve ettha ummattaa" (Vi­kram, 11, 8, 8)

Sakuntalii's love-letter, " TuJJha na il.ne ruaam mama una kamo diva. p1 rattun

p1," etc

" Samv1bhattam h1 dukkham saJJhavea­nam hod, " (Sak , m, 9, 4).

The well-known song m Act v of the Sakuntal,a "Arunavamahuloluo tu­mam," etc,

"Tumam diiva asahiimi Jiie pudhavr­nii.tho samive vattad1 " Sak , (P1schel ed , p 59, end)

But why multiply mstances? Nothing can carry conviction if the cases already cited have faded to evoke 1t

A careful study of the contexts in which the above parallels oc~ur should indicate to us the way m which Srihar~a modifies the ideas aµd l

images of _K!!:.h~J!,sa to smt his own purpose I choose Just one case by way of illustration Sanskrit poets are m the habit of mentioning a number of creepers that blossom in the sprmg, among others the Madha'l'i, which puts forth flowers m the month of Madhava, - Va1sa­k.ha, or April-May, and the Nm 1amalika, whose flowering season comes about a month later, m what 1s known as the Grisma-rtu Now, it will be remembered that the Sakuntala opens at the very begmmng of the Grisma, and hence the creeper N at1amalika 1s descnbed m the opening act as navakusuma Jot'Vand, or " showmg forth her youth in the form of new blossom.'{gi'he other_cree~__, J/adhai 1i, 1s not at all mentioned m__tliz Devanagari recens10n of the play In the Kashmir recension ( which m my opm1on preserves the most genume version of the garden­scene) the creeper is mentioned, but only to motivate the heroine's .. movement away from the tree behmd which the hero was standmg concealed, a movement which Pnyamvada stops with the words, Cittha idhayyeva difoa baularukkhasa'mi11e Tae samivatthuliie ladasanadho via baularuklchao padibhadi The _»_~11g_ah y~rs_i_~I! 9{ the scene, as I have elsewhere 1 shown, considerably disturbs the sequence

- ' -1 Aina MaJor, vol 11, fa11e 1, p. 101, also Sir Asutosh MookerJi Silver Jubuee, vol.

iii, pt. i. p 856.

190 Shripad Krishna Belvalkar

of the speeches and is responsible for large interpolations, amongst others the referenc_e to the ~&lhavilata a~l its _1?l_~s-s011:1in_g out of sea.sol\.fAsamae kkhu esa iimulii.do maululii miihavUada). In the G.n~ma­rtu the Madhavi can be properly described only as past its flower­ing season (adikkantakusumasamaii), - as the Kashmir MS. in fact describes it, - although the creeper may still put forth a few late buds now and then. If now we turn to Srihar~a's Ratniivali, we see the way in which the two-creeper motif has been adopted by that poet and turned to a slightly varying use The second act of the Ratniivali opens in the vernal season with the Miidhavi (for which the Queen in the play affects a partiality) in full blossom A N avamalikii creeper in the same garden is the favor1te of the King and it has not yet com­menced its normal flowering season The King, however, secures an akalakusumasamjanarJ,adohala, or a prescription for inducing the creeper to yield flowers earlier than its normal penod, so as to prevent the Queen from securing an easy victory over him m that respect. Here the trend of events follows the regular botanical sequence, which the extra passage in the Bengali recension altogether subverts, there­by betraymg the hand of the unskilful interpolator, who may, possibly, have taken his cue from Sriharsa's Ratniivali itself.

Reverting now to the Sri1g~ric p~sage in the third act of the Siikun­tala, the most cogent object10n against its shorter version is the fact tha~pl~ ~~t and loose _wi~ the time-indi~_ations of the scene The act opens a little before 2 P M , alter the conclus!9~ the mid­day lihation~t-amt--ihelieai~yis still unabated 2 whe11 th_e ~ng objects to ~akuntali'sgoT:rig- out of l~hower~---ff this indicates, let us say, abour 4 P.M, Kalidasa here seemito-take nearly seven pages of Cappeller's edition to cover this per10d of two hours When Gau­tarni later enters at the conclusion of the Srngaric scene, it is already evening, and the poet, following the shorter vers10n, requires scarcely half a page of the same edition to cover this subsequent period from 4 P.M. to sunsetlllt is not, of course, meant that a scene should take as much time i~he acting of it as would be required m actual hf e for the happening of the action represented m the scene. This would be absurd. But within the same scene a careful poet is expected to ob­serve the same time-ratio. The shorter version of the present scene fails to do so. The longer version, on the other hand, does observe the proportion and, what 1s more, mterposes in the latter part of the scene a sentence indicative of the flight of time: Diniivasiinacchiiyeva puro­miilam vanaspateh (Pischel, 81). Contmuing the scene from this point,

1 Cappeller, p t8, I. ll. 2 Aparinin,atw dttJaaah, 1b1d , p. S4, I. ff.

Srngaric Elaboration in Sakuntala, Act III 191

it will b t noted that Pischel's stanza 88 has to be retained on the au­thority of Vardhamana, as also stanza 84, which is adopted in the Ratndva,'i. If, farther on, we have to retain stanza 88, which is quoted by Visvunatha, that necessanly involves the retention of the majority of at least the prose speeches leadmg up to it, mcluding the episode of the rety ng of the lotus wristlet The essential features of the Srngaric scene hr.ve accordingly some kind of a warrant for their existence. On critical grounds I beheve it possible to reduce the version of the scene as given by Pischel by the dropping of some five or six stanzas and twice as IDR'llie:nes or prose; but agamst the genumeness of the scene when thus sl:oietned I know of no valid arguments that can be urged except the pm I) l ubjective ones. I am not here ignoring the cogent remarks of Pnnel 11e1 S Ray, 1 which would prove fatal to the longer version if it were,, o .. ',?ossible to meet them. He says "The passa~~ribes at great length how the mrnalavalaya was picked up by Du~yanta and put back oh the wrist of Sakuntala. Tliis, howcve_r,. cofilradicts_ the poet; hr later on we find the mrniilavalaya still lying in _th~ 9!0Ve. Offi\pare H astad bhrastam --uram bi.~abharanam ity iisaJjamaneksarJ,O Nirgantum sahasii na vetasagrhiid iso 'smi siinyiid api (infra), which is undoub •edly autfic, bemg common to all the recensions "

No"', as agai this argument it can be urged m the firstplace that, in spite of he remarks of the commentator Raghavabhaga anent the lme, Stananyastosiram prasithilamrrJ,iitaikamlayam, namely, Mrniilasya ekam mukhyam valayam yatra . elmin ity anena valayiin­tariisahatvam dhvanyate, we have really no right to conclude that Sakuntala wore only one wristlet on each hand· there must have been more than one so worn;- only, as the word mrniilaika points out, they were all made of lotus-stalks alum·. But even supposmg that we agree to concede his point to Raghavabhatt.', there was, one may urge, the other hand in any case, from which the equally loose wristlet could drop away any ti:me - even subsequent to the retymg of the wristlet by Du~yanta on one of the han~ Fmally, - and this is the most probable hypothesis, - just as the lotus-leaves placed on the bosom of the heroine to allay the heat had to be frequently replaced by fresher ones, so must it have been even with the wristlets Compare Pischel's stanza 73 (iisuvimardita - v. I, vimrnita - mrrJ,iilavalayiini), and particularly the stage-direction m the Ratniivali at the beginning of the bower-scene - Nalinipatraih sayaniyam mrniilair valayani (note the plural) ea racayitvii parisistani nalinipatriirJ,i Siigarikiiyd hrdaye nik~ipati, as also the stanza, Sthitam urasi visiilam padmini-

1 Op cit, p 284

19i Shripad Kriahna Belvalkar

patram etat, and so forth, 1 in which the king describes the con:dition of one of the cast-off lotus-leaves. There was surely no dearth of lotus­stalks, so that the friends could prepare only two wristlets and no more. Ray's objection need not be held, therefore, to outbalance thie weight of the other internal and external evidence thus far considered.

Finally, a word as to the way in which deliberate interpolations which are of the nature of weak paraphrases of the words of t 1he origi­nal, or minor variations of a poetic theme successfully introduced by the author of the original, come to be made and adopted as _genuint> parts of the text. That this is mostly the work of the cleveNr ll{'t of students and of self-opimonated pandits who combine in thtenselves the functions of a second-rate critic and a third-rate poeqymay be almost taken for granted; and in connection with this I recnember how, in the upper clases of the high schools, we were often re<f1ired by our Sanskrit teacher to versify simpler passages from the Kadambari or the Dasakumiiracarita, or put some well-known stanzas from Kali­dasa or other classical wnters into metres different from the Jriginal, while retaining intact (and even occasionally improving) the central idea of the verse and as many as pos111ble of its words Wi,.h some effort, I believe I can still reproduce some of my Juverule ePorts, of which I thought very highly at the time and which I often wrote down in my own copy of the origmal Such exercises in compositrnn have been all along current m India, and 1t is conceivable that from the margins of some of the MSS these JCJUne product10ns got ador,ted into the text of the origmal in the course of successive transcr1pt10ns from MSS, with perhaps the addition of the prefatory api ea, or ' .• hii hi, to mark their adventitious character, although it would 1 e too rigorous an applicatwn of the prmciple underlymg thi~ fact to reject as non­genuine every single stanza C1f a text that has these prefatory par­ticles, as Cappeller s«>ems to have done m his "kurzere Textform" of the play In dealmg with all such mterpolations, especially when they are of a respectable age as bemg vouchsafed for by more than one MS., the conscientious editor has often to fall back upon" higher criticism," which is often sub1ective m its nature, and the editor can inspire con­fidence m the conclus10ns that he may thereby reach only in propor­tion to the success that he may have already attamed in applying to the same text the more obJecbve canons of textual cr1tic1sm.

1 Ratnavali, u, 12

DEcCAN CoLi:..EGE

A RAMAYA~A STORY IN TIBETAN FROM CHINESE TURKESTAN

BY FREDERICK WILLIAM THOMAS

THE te:x:t which forms the subject of this modest study is by no means one that would have been expected to emerge from the now

famous hidden library of Tun-huang (Ch1en-to-fung). The Indian lit­erary works hitherto recovered from Chinese Turkestan, whether San­skrit or Prakrit originals or versions m other languages, are almost exclusively of a Buddhist character. But here we have manuscripts exhibiting the story of the chief Brahmanical epic, the iidi-kiivya, with no infusion of Buddhism. From the extreme east of the region, the very border of China proper, comes a Tibetan version of the story of Rama.

The first document (A) is one of the numerous rolls of thin yellow­colored paper which were inscribed with Chmese translations of Bud­dhist siltras It is quite similar to those illustrated in Sir Aurel Stein's Serindia, plates CLXVI-CLXVIII, and along with them is deposited in the India OffiC'e Library It 1s of considerable rxtent, measuring 15 feet I inch by 10½ mches ('26 cm), and the Chmese siltra for which it was onginally pieced together fills with its regular columns (c. 18 5 cm ) of finely written characters the whole of one side The Tibetan writmg, 439 Imes (~5 cm m width), in a rather cursive, but for the most part elegant, hand, occupies the greater part of the reverse

This arrangement of the two languages, which is abundantly exem­plified m other cases, affords ground for chronological deductions We have many Tibetan documents, letters, memoranda, short treatises, and the hke, similarly associated with Chinese texts. In all cases the Chmese was the prws It 1s clear that durmg one period disused or appropriated MSS of Chmese siltras were freely used by the Tibetans as stationery. The period of Tibetan domination m the eastern part of Chmese Turkestan extends from about 700 to about 850 A D , when the region passed under the control of the Turki-speakmg U1gurs We may safely conclude that most of the Tibetan writmgs from those territories belong to period 700-900 A n , and that the Chmese books which were turned to such use were of a somewhat earlier date.

The second document (B = Ch 80 IX 3), hkew1se fragmentary at the beginning, is of rough. yellowish paper, measurmg 5 feet ll inches in length by ll; inches ( c. 29 cm.) in width. The one side is inscribed

194 Frederick William Thomaa

with 99 lines, of the full width, in a large hand, which varies between a square formal type of characters and a more cursive style: it contains not a few additions and corrections in a smaller, cursive writing, which also appears on the back of the document. There we find in that hand. but, as regards the latter half, with larger and more formal characters, which may be due to the scribe of the obverse, 39 lines widthwise (c. 25 5 cm), plus one lengthwise hne presentmg another part (C) of the story and here also we find one correction similar to those on the ob­verse. Upon the evidence which we have of the variation of style there is nothing to show that a single scribe is not responsible for the whole.

The corrections on the obverse side correspond, so far as the com­mon ground is concerned, with the readings on the reverse. Elsewhere they to a certain extent correspond with those m document A.

The reverse contains also, in a much blacker ink and partly in a different hand, some notes, drafts of letters and the like

The three 1 documents are mutually mdependent. C corresponds rather closely with A, but m such a manner that the two must be dif­ferent translations of one original. B covers for the most part different ground from A, a precedmg part of the story But in the common part it clearly goes back to the same or1gmal What was this original?

The story, as told, is in form and substance wholly Indian, and the interspersed verses are unmistakably Indian in style and sentiment. But we should seek in vam for an Indian version of the RamayaJJ,a to which the text closely corresponds. It follows the general lines of the narrative in the Jf aha-Bharata (T'ana-Parvan, chapters 274-290); but the incidents and the nomenc1ature differ widely, and indeed surpris­ingly. A few examples may be given

The demon Malyavant is the son, not of Sukesa, but of the "Yaksa" Kore (Kuvera), who is distmgmshed from Vai~ravana Dasaratha has only two wives (not named), and only two sons, whose names appear as Ramana and La~ana ( the latter always so spelled, the MSS showing no cerebrals). Sita is a daughter of Riivana (always named DasagrJva); and, when cast away and found by Indian husbandmen, she is in no way connected with king Janaka of V1deha, who, m fact, is not men­tioned The account of the search for Sita differs in many details from what we find m the M aha-Bharata and the Ramayana. The birth of Lava and Kusa is prior to Sita's bamshment and is quite differently narrated. The scandal which causes the banishment is reduced to a single saying overheard by Ramana, who has an mterview in quite popular style with a washerman's wife, and receives instruction con-

1 Concernmg a fourth document see the addendum (p 212)

A Ramaya1J,a Story in Tibetan from Chinese Turkeatan 195

ceming the nature of women. The recovery of Sita and her children takes place upon the earnest representations of Hanumant.

Among the variations of nomenclature, we may mention that Kumbhakan;i.a is replaced, in his first occurrence, by Amalakan;i.a, or Utpalakan;i.a(?), Kaikasi by Mekesina or Megasina or Mesina, Vibhi­lia~a by Birisana or Birinasa. Uma is Umade or Upade, Hanumant's companions are Pagsu and Sintu (not Arigada and Tara, as in the Ra­mayaT),a), the two eagles are Pada and Sampada (not Jaµiyu and Sam­pati, as in the M aha-Bharata and Ramayana); Marica becomes Maruce; the monkeys who make the bridge are Maku and Damsi; and so forth.

Certain incidents have a distmctly popular tone. Besides that al­ready mentioned, we have Rava~a carrymg off Sita along with the plot of ground or estate (sa gzi); 1 the monkeys enter the cave holding by each other's tails, Sugriva in his fight with Balm has a mirror tied to his tail; Hanumant, when captured by Ravana's forces, begs to be killed as his father was killed.

We have therefore a highly peculiar Ramiiyar.ia story. Whence and how did it come to the Chmese frontier of Turkestan? No Rama story known from India exhibits a majority of the above features, but there are certainly some attachments The name of Ravana's father is given (B) as Ratana, which reminds us of the Ratnasravas (for V1sravas) of the Jama Riimayana (sarga I, ll. 132 sqq ). Ravana's interview with Vi~1a1.u has some resemblance to that narrated m the Uttara-kiinqa of Valmiki's poem (c. 24) The story of the washerman's wife recurs in Tulasidasa's Ramiiyana (Lava-Kusa kiinqa d. Bhagavata-Purana, ix. 11. 8 sqq ) But the largest amount of similarity is found in the first of the two Rama narratives (IX c 51 and XIII c. 107) contained m the Katha-sarit-8agara

There are two indications which perhaps point m different direc­tions. The first is the form of the names of Dasarite (Dasaratha), Pra­haste, the Yaksa Kore (Kuvera), the nominative in e may be due to the language of Khotan There is also another name which has a rather Central Asian appearance This is Manlyapanta, or Malhyampata, or Malyapada, for Malyavant It 1s impossible to suppose that the Tibetan concocted these forms or derived them direct from a Sanskrit origmal. A Chinese source being excluded on the ground of the general correctness of the other names, and the 'Khotani' and 'Tokhari' for similar reasons, an original in one of the indigenous monosyllabic languages of Chinese Turkestan is not out of the question.

1 This 1s, perhaps, due to the representations showing her on a sort of platform.

196 Frederick William Thomas

A second possibility is perhaps indicated by the reference to the man Litsabyid Dri-ma-dag-pa, whose wife's utterance led to Slta's banish­ment It looks as if we here were dealing with a Licchavi rajaka (washerman). In that case a Nepalese source is suggested The sugges­tion has no antecedent 1mprobab1hty During the seventh and eighth centuries Tibet was in close relations with Nepal, and the soldiers or Nepalese auxiliaries may easily have carried the story with the armies which overran the eastern part of Chmese Turkestan. In that case we are dealing with a popular form of the Rama story having a currency m Nepal. This would furnish the direct connection with India that is demanded by the undiluted Indian character of the narrative.

In any case we have m these documents testimony to the early cur­rency of popular Rama narratives following the general Imes and scale of the Maha-Bh<irata and departing freely from the classical version of Valmiki Some such narrative may have found a place in Gu1,1adhya's Brhat-kath<i, which m this pomt may be reflected by the Kath<i-sarit­s<igara; the extant portion of Budhasvamin's Sloka-samgraha does not seem to give it.

The Tibetan writmg in the documents, though it presents some pe­cuhar1bes, found also in other specimens from Turkestan, 1s in all es­sentials identical with that still m use The language also is what we are accustomed to m inscriptions, edicts, letters, the writmgs of Mi-la­ras-pa, and so forth. There are words and phrases not to be traced in dictionaries, and the irregularities in the initial consonants, the abund­ance of homonyms, and the insufficiency of syntax, which obscure the meanmg m all Tibetan writmgs not guaranteed by versions in other languages, are here also in full operation But the story is simply told, and the whole might, if it were worth while, be edited and translated, except that the verses would create a difficulty In the two abstracts given below, the third document C being useful only as supplymg a fragmentary part of A, I have inserted most of the verses (liwratim as regards all peculiarities of spellmg, metre and so forth) WJth renderings of a qmte tentative, and in places even con1ectural, character

DOCUMENTS A AND C

1-3 The Yaksi.J:ii Megasina [Kaikasi] finds favour with the Muni [Ratnasravas or Visravas], who with her begets three sons, Dasagrl'.va, Amalakarna (distmguished infra from Kum­bhakarna), and Birinasa [Vibhisa1,1a] Upon Dasagriva, the eldest, Brahma confers ten heads, and his strength is in pro­portion.

A RamayarJ,a Story in Tibet,an from Chinese Turkestan 197

4-22. (fragmentary). Man-lya-pan-ta (Malyavant) proposes to (C 1-9) the "sons of gods" [Devaputras] that they should combine

and go to Lankii-pura, the realm of their uncle [Vaisrava:r;ia]. Seeing the prosperity of Vaisravana, he weeps. Being asked the cause of his tears, he explams that he remembers his father, the Yaksa Kore [Kuvera], who had been expelled by Vaisrava:i;ia and sent down to hell. He implores the assistance of the 'sons of the gods,' who are wdlmg to help, but profess their inability

22-30. At the suggestion of Malyavant the "sons of gods," desiring (C 9- 15) to conquer the gods, worship Brahma with asceticism and

so forth during one hundred thousand years. Brahma, how­ever, knowing then evil obJect, accords nothmg They con­tmue their efforts, and then beg of Brahma boons, namely: (I) that everyone at whom they shoot an arrow shall die; (2) a life-charm (g-yun-drun = svastilca) to prevent their be­ing killed by others, and (3) sovereignty of the three worlds. Brahma replies that he Is himself the only sovereign of the three worlds and he cannot be reached by an arrow

30-33. The sons of the gods try to propitiate Mahadeva But he (C 15-19) also, knowmg their evil purpose, will not appear, even when

Dasagriva cuts off one of his own heads and makes it into a burnt offering

33-41 Mahadeva's wife Upade [C Umade = Uma], bemg very (C 19-23) compassionate, appeals to her lord; and upon his continuing

obdurate she herself comes before the 'sons of the gods,' and advises them to give up Mahadeva and propitiate her They declining upon the ground that she is a woman, she grows angry and prophesies their overthrow by a woman

41-47 Similarly Mahadeva's mmister Prahasta IS rebuffed, as being (C 23-30) a monkey, and prophesies overthrow by a monkey

47-54 After a long time, as Mahadeva still disdains to appear, the (C 30-36) goddess of speech [Sarasvati] takes the form of a lump on the

tip of the tongues of the sons of the gods and so modifies their requests, making them demand (I) sovereignty over the gods, (2) a life-amulet preventing death at the hands of any being who does not first cut off Dasagriva's horse-head, (3) death of any being against whom they hurl the first arrow

54-56. Thus the gods are defeated by the sons of the gods, and in

198 Frederick Willwm Thomaa

(C 86-89) Laili-pura gods and men are destroyed, and the demons who fill the island make Dasagriva their king.

56-60. The gods who rule the three worlds take council together and arrange that a human being capable of destroying the demons must be born as a daughter of Dasagriva.

60-65. A wife of Dasagriva gives birth to a daughter, who, as the sign-readers declare, will ruin her father and all the demons. So the child is enclosed in a copper vessel and committed to the waters She is found and adopted by Indian peasants, who name her Rol-riied-ma [Lilavati].

65-72 The king of Jambii-dvipa, by name Dasaratha, being without a son, prays to 500 l;lsis dwelhng on Gans-di-se (Kailasa). They send him a flower, stating that he should give it to his chief queen. he should then have a son Out of compassion the chief queen causes half of the flower to be given to the Junior queen. Two sons are born The son of the junior queen, three nights the semor, is named Ramana The chief queen's son is named Laksana

72-82. Dasaratha, returning wounded from battle on the side of the gods, decides to abdicate and discusses with the chief queen the enthronement of Ramana or Laksana He is in perplexity through consideration for the chief queen, and is seriously ill; which Ramana perceiving prays that his father may live on condition of his own retirement to a hermitage He departs, and Lak&ana is crowned, after which Dasaratha dies

82-89. Laksana VISits Ramana and offers to resign the overlordship [cakravartitva] of the four Dvipas, to which he 1s unequal. Ramana declmes, whereupon Laksana takes a pair of Ra­mana' s shoes [chags] and instals them, himself playmg the part of minister.

89- 108. The girl Rol-rfied-ma having grown up and being very beau­tiful, the peasants send everywhere to seek for her a suitable mate They find Ramana, whose appearance greatly im­presses them, they off er him the girl (11. 95-105) -

skra ni mthon tin gyasu !J,khyil/dmig ni 'ud dpal la// tshans pa l:!,i dbyans ltar kha dog rnam par dag / !J,phral ba rgyan mchog mdzes pa dbyisu §is / / dpal ldan dri myed pad mo mchog las skyes / lus ni yan lag yons su tshan / gser gyi gzugs la/nor bu phyis pa !J,drab,/ / phyogs kun b,od kyis rnam par gsal bar gdab, /

A Ramaya1J,a Story in Tibetan from Ckinese TuTkestan 199

lua la ea !J,dan .rgyal po spos kyi dri / !J,gro Man lha rdzaa yid !!,on brjod pa!J,i take/ g-yar nas 'ud dpal la l!,i nad.kyan rgyun du ltan// !J,di ni gan du mdzum zin dgod pa !J,i dus de na/ rol mo agra siian nag ni l!,byun bar !!,gyur / / bud myed rin chen b,3ig rten !J,dir byun ba/ / tha mal kun kyi dban du gyur ma lags// khyod ma gthogs pa myil!,i !],jig rten na/ / !],di h-i bdag po gzan na gchig ma mchis / mtahan brgya!J, !rt sku gzugs mdzes mna!J, ba!J,/ / yon than sna tshogs ldan ba b,i skub,/ / khyod kyi :;a snar bu mo !J,di !J,bul na/ sna tshogs lo ma yan lag rgyas rgyur chin/ sin lo men tog mdog sdug rgya.'l pa l!,i mdog / / rgyan dan bchas/sin yid du !Jon/ rin chen dar bas grubs pa b,di bzes sig /

Hair dark and curlmg to the right; eyes like blue lotuses, Like the sound of the Veda (brahma-ghosa), of colour (articulation)

ziarna) utterly pure, In herself, beautified by excellent ornaments, auspicious of form; Brilliant, born of the best of soilless lotuses; Body m all its members complete; Like a gem set m a shape of gold, She brightens with radiance all the regions; In her body, 0 king. the fragrance of burnmg sandal, In speaking. 0 lord of the world, while she speaks what is accept-

able, At her mouth the aroma of lotuses falls continually; Whenever in Jesting she smiles, at that time Her voice 1s music agreeable to hear, This jewel of a woman born m the world Should not be at the disposal of any ordinary person. Except yourself there is in the world of men No other lord for her. To you, whose gracious form has a hundred good points, Whose body possesses all merits, We offer at your feet this girl. All her limb-tendrils fully developed, Leaves and flowers of fair hues fully expanded, Made more acceptable as equipped with ornaments, -This girl made of Jewels in quantity be pleased to take.

!tOO Frederick William Thomas

- and he, giving up his asceticism, marries her under the name Sita, and becomes king.

108-118. The five hundred ascetics being near the realization of their aims, a minister of the Y aksa Kore, by name Maruce [Marica], red-haired, tall, large-eyed, with inverted toes (?), his head adorned with human skulls, all his limbs smeared with blood, creates hindrances. The defeated ascetics write and circulate a statement that whoever shall dispel the obstacle shall have a boon. Ramana accomplishes this. Maruce reappearmg, Ramana throws a ring at him and de­prives him of his eyes, whereupon he flees The ascetics give Ramana a blessing anyone at whom he aims a shaft shall perish and be reborn as a god.

118-133. Dasagriva' s sister Phurpala [Surpanakha - description] appears before Ramana (11. 123-125).

bdag ni l!,jig rten rnams kyis rtag tu bsnags/ bsod nams ldan bas/kun du rnam par snan/ !!jig rf,en lta bus noms par myi/ mkhab, la dben gyi sprin tshogs bkrigs pa la/ gfii dro snan sar g:ml bar gyur pab,i/ / gzugs bzans mchog !!,di/ /khyod lcyi l!,bansu l!,bul/ /

Ever lauded by the worlds, Through being possessed of merit, everywhere resplendent; As not content with the world alone, Upon the troops of clouds gathered away in the sky Shining at the appearance of the warm sun's light -This fair, excellent form I offer at your service

Bemg rejected by Ramana, she withdraws (II 127-129) -

dgyes pas lhun brtsegs brtan ba l!,i ri bo la/ chab _qan g-yun drun gfii dro b,od !!,char du// mkhal!, la dben ba l!,i sprin tshogs l!,khrigs pa !!,is/ gfii dro gsal yan ljchar ba l!,i dus skabs myedj /

When joyously on the mountain with its firm, towering frame The light of the warm sun is reflected in Manasa, full of water, For the cloud-troops gathered away in the sky, Though they shine with the warm sun, it is no time to appear

[rain?].

She returns to her brother m Lanka and suggests the abduc­tion of Sita.

A Ramaya1,1,a Story in Tibetan from Chinese Turkestan !!01

188-168. Dasagriva consults Maruce, who dissuades him from the attempt; when Dasagriva persists, he suggests the ruse of the deer. Ramana being persuaded by Sita to go after the deer, Maruce interposes a storm between him and his wife and brother. Hearing a call for help, Laksana at last, after re­proaches and insinuations from Sita on account of his reluc­tance to leave her, goes after Ramana; and Sita is left weep­ing. Dasagnva appears first in the guise of an elephant and then in that of a horse; finally, when Ramana and Laksana are on the point of arriving, he carries off Sita along with the plot of ground (sa gzi); he fears to lay hands on her person.

163-188 Ramana and Laksana, returning, are dismayed to find Sita and the plot of ground vanished Worn out by a long search. they fall asleep on the spot Next day they depart in grief, and come to a black stream ma valley. La~ana is about to drink, when Ramana warns him that it is outflow from some stricken creature Advancmg into the valley, they find that it flows from the eyes, mouth and nostrils of a huge sleeping ape, who, being questioned, explams that he is Sugriva, younger son of the kmg of apes His brother Balin is king, and, bemg the stronger, has maltreated him Questioned as to Sita, he explams that he himself, bemg 111, has not seen her In his service arc three apes, who have fled at the corn­mg of Ramana and Lak~ana They are up on the mountain and may be questioned Bemg reassured, the three huge monkeys descend and relate that early on the previous morn­mg they have seen a ten-headed man, on whose front head, a horse's head, was a woman placed upon a plot of ground, who, crying out that she is Sita, wife of king Ramana, and implormg any merciful person to rescue her, was carried off. Ramana proposes alliance with Sugriva, offering to make him king, if he will arrange to find Sita

188-207 On the morrow the two monkeys fight in Ramana's presence without result Ramana says that next day he will be a par­ticipant and not merely a spectator. Balin's wife gives her husband good advice Purther fighting, in which a mirror tied to Sugriva's tail plays a part. Balm is slain by Ramana's arrow and becomes a god.

208-211. Ramana arranges a meeting with Sugriva, and, when Sugriva's army fails to appear at the appomted spot, after three years

Frederick William Thomas

he sends him a verse message inscribed on an arrow, warning him of Balm's fate (ll. 209-210) -

dam la gnas byas mgrin bzans po/ llbal!, li lam du ma lJ,gro §ig I I gan du lf bal!, li bsad pa lJ,i lam// de ni legs pa ma yin no/

Abiding by his compact, let Sugriva Go not upon Balin's path. Where Balin met his death, That path is not a good one;

whereupon Sugriva, alarmed, comes with his army. ~11-236. Sugriva nominates three powerful monkeys, Pagsu, Sintu

and Hanumanta, to go in search of Sita, and Ramana fur­nishes them with a message and his signet-rmg. After a long and vain search, they are thirsty They happen to see two ducks alightmg in a hole m a rock, which they regard as a sign of water Followmg the ducks and one holdmg the other's tail, they enter the hole, where they discover a fine apartment Askmg whose it is, they are told that it belongs to a daughter of Sri Devi, by name Gtsug-rgyal-sgeg-mo [Maya, daughter of Meru?] They show respect to her and tell her where they are going. She bids them wink their eyes Taken in an unknown direction, they find themselves on the shore; and, as they gaze at one another, a great black mountam appears. They notice that it is shaking; and, coming near, they see a black bird with its feathers injured. Upon their enquiring the bird explains that his father was AgaJaya, king of eagles his own name is Pada, that of his younger brother being Sampada disputing about the sov­ereignty, they made a compact that whichever should fly away qmckest from the summit of Mount Meru should have the sovereignty Being the more speedy, he looked behind, and he saw his brother's feathers burned by the sun. Turnmg back, he went to the assistance of his brother and had his own feathers singed. Thus maimed he could not obtain the sov­ereignty, and so remains on that spot. To the monkey's inqmry concerning Sita, he replies that at the time when she was carried off, his father, being a relative of Ramana, shot into the sky two of his feathers, which fell to earth. Being struck by the feathers, the demon was tired and set down

A RamdyarJ,a Story in Tibetan from Chinese Turkeatan iOS

Sita. He then threw a red iron ball, which Pada's father took for food; having eaten it, his heart was burnt and he died. The demon then made off with Sita.

286-254. The monkeys take counsel as to proceeding to Lanka, which is in the middle of the sea. Sintu avers his own inability to make the passage, while Pasu (sic) thinks that he can go, but being ill, could not return Hanumanta, having waited to be asked, professes readiness and leaps to Lanka. He finds Sita guarded by an army in a doorless fort with nine circular walls He approaches very quietly and presents the latter and the signet-ring Sita i~ overJoyed and reads the letter (ll. 245~54)-

Men tog IJ,dab rgyaa lte bu bi slcu mdog la/ anun kyi than bas gduns par ma gyur tam / stag ris bun ba lte bu !J,i !J,khro I.din la// thugs kyi grwa nas/bren bren myi brtse b,am/ bdag m bslu ba lJ,i ri dags de bkum slar mchisna/ / sa gzi bchasu bzan mo myi bzugs, nas// mye nan yid la gduns pa lJ,i lus mdog ni /

lo IJ,dab lhags pas gduns pa !J,i !J,3on sin bzin / I mdzes sdug sel mdog IJ,dra ba !J,i than grogs de myed nas/ / nur pa b::in du phyogs !],tshams kun du mol/ / chab gan g-yun drun sems kyi thag bsrins pas/ /,o zla man po M dus kyan yens la !J,da!],s / da ni gans kyi sen ,qe rtsal phyun mthus gthugs pas// sdo ba lJ,i dgra gzan ran fiid !J,kham ste !],chi/ lhun po ri riser rin chen byin sar na/ skar tshogs mkha!J, la rgyu ba lJ,i !J,od dan bral,/ / de bas bdag la d_qons tdan brtse ba lJ,i nan iiid kyi/ dkyiL yans rgya mtsho lJ,i dkyil du g'ljur liar yan// pku chab lte bu lJ,i yid kyi dba!J, rgal dag/ / rgya bskyed dgons pa lJ,i lhun ni tshim par mdzod/ /

Upon that body like a flower with expanded petals Has there been no blighting by the drought of sickness? Has anger, alighting hke a bee, with its axe (?) Not cut away little by little from the angle of your mind? When I came back from slaying the illusory deer, And the fair one along with the plot of land was not there, My body's hue was blighted by grief at heart, Like a tree with its leaf-petals blighted by frost·

!!04 Frederick William Thomas

As in the absence of its loved, crystal-hued mate A duck wanders about in all directions, Postponing the thought of Manasa full of water, Many years and months have passed in restlessness. Now that I am equipped with strength beyond the might of an

ice-lion Other venturesome enemies, failing of themselves, perish. When the splendid jewel rises on the peak of Meru, The stars moving in the heavens lose their light. Therefore your affectionate nature in thought of me May expand widely its horizon, as though it were the horizon of the

sea. Crossing the waves (agitation) of mind, as of a mountain stream, Let the mass of your expanding thoughts be satisfied,

stating Ramana's grief and his intentions. 254-284. Sita impresses upon Hanumanta the necessity of prudence;

but he does not listen He goes into Rava:r;ia's park and in­verts all the trees and so forth This is reported to Rava:r;ia, who commands many of his servants to capture Hanumanta, who however kills them all Dasagr1va then sends his eldest son with a net made of sunlight Hanumanta will not ap­proach the net, but, when the demon prays to the Siddha gods, these order Hanumanta to enter the net When the demons propose to kill him, Hanumanta prays them to kill him as his father was killed, namely by tying to his tail cloths dipped in oil, and settmg them on fire. Hanumanta leaps on the forts and houses of the demons, sets them on fire, kills many of the demons themselves, and then, removing the cloths from his tail, makes off and seeks Sita. He says he is departing and begs for a message She eventually complies, giving him a Jewel to convey and Hanumanta brings the letter to Ramana, who is overJoyed and reads it (ll. 276-ft84)

yon than yid bzin man po 1J,i IJ,phra tshogs kyis / mdzes par brgyan pa!!,i sgron ma lte bu sku gdun la// snun gyi ser bus ma bskyod la me zes/ gus par rmas pa b,i Zan tshigs snan pa dan/ / brtse dgans b,phrin yig sems kyi dan rig tshig / mnen !J,jam lcug pa !J,i phyag rgya no.!J,tshal te/ dba brug (?) sdu_q pa IJ,i jbris bab,i g-yar lam na/ snan gyi pha ma gna!J, !J,da!J,s phrad chin smyis pa bzin / /

A Rdrrw:gatJ,a Story in Tibetan from Chinese Turkestan 205

bdagi dran ba. '!Ji si,obs ni dban myed pas I I bphana pa bzin du baams pa nas/lha nid la yan/ lhag par bren bren .dran ba 1J,i mthu ma mchia/ thugs rjebi gzuga kyia/yun du mi gtan zin/ mnal!, than rkyen kyi l!,khor tahoga man po la/ bdag !J,dra IJ,i myi dpen / dman ba ma mchis kyan / chab gan l!,khor rgyug re brtan ya rabs gzun / thugs la dgons par gnan ba gthan rag l!,tahalj

"In that frame, like a lamp finely adorned With ornaments of many attractive virtues, The chill of illness does not range, I trust" this Respectfully spoken, agreeable answer And affectionately meditated letter know to be my heart's mean-

mg Recognizmg the supple, finely-turned (?) seal, In the presence of the dear ... writing (?), It was as 1f I dreamed of meeting my old parents long passed away. My memory's strength, uncontrollably, As 1f shot forth from my mmd, 1s w1th my lord alone, And I have no strength to remember anything more For that with compassionate heart you have not in this long time

given me up, When there is a numerous court subject to your authority, But upon one like me, lord of men, who, although not lowly, Am of the middle rank of the retinue dependent upon you, You deign to bestow thought accept my thanks.

284-293 The army of monkeys and men having rejomed, all start for Lanka, but a great ocean bars the way. Ramana commands the monkeys Maku and Dams1 [N ala] to make a bridge. They set to work, tearing up mountains and trees As Damsi takes the mountains on his knees and builds the bridge, while Maku stands, they quarrel about their respec­tive strength, whereupon Ramana reproves them (a verse, 11. 291-293). The bridge being finished, the army crosses.

293-299. They having arrived at Lanka-pura, the time for giving battle having been fixed, Amalakarna, who is wise, gives good advice to his brother Dasagriva not heeding, Amala­kar:r;ia goes to join Ramana, whose suspicions he dispels by quoting a verse (II. 291-293). Ramana being placated, Ama­lakarna becomes his adviser.

i06 Frederick William Thomas

299-808. A demon named Rum-rna (Kumbhakan.ia) had by austeri­ties and through the intervention of SarasvatI in the manner stated above obtained the boon of perpetual sleep. Dasa­griva and the others succeed by pouring molten metal [?] into his ears, making thousands of elephants smite his body, beating great drums near him, in awakening him. In re­sponse to Da.sagriva's appeals he devours the army of men and monkeys. but Ramana he is not able to devour. Ha­numanta, whom he seeks to devour, is sometimes at his ear, sometimes in front of him, sometimes in his eye· so he gives up, and the two separate. At last Kumbhaka111a is exhausted and falls asleep again.

808--809. Amalakan;ia states that on mount Kailiisa there grows a herb Ubri-ta (Amrta), which could heal the whole army: Hanumanta is sent to fetch it. Not bemg able to find it, he pulls up Kailiisa, gathers the Ubri-ta, and replaces the moun­tain. With the [Jbri-ta the whole army of men and monkeys is healed.

809-820. A time for battle is again fixed. Da.sagriva with his first arrow kills his younger brother Birinasa, who is flymg Ra­mana reflecting that, if he himself went forward and died, Sita, even if saved, would have no compamon, Laksana is placed in the van. He being struck by many arrows, Ramana goes forward himself. Dasagriva, concealing himself by magic, kills with his arrows many men and monkeys. Ra­mana challenges him to show as much as a toe; and, when he does so, aims an arrow at where his horse's head should be and cuts it clean away it then floats about above the army of men and monkeys Ramana says 'do you not know how to die: must you go on fighting?'; and, rising above his own army, kills most of the demons over the great body.

820-828. The demon army having been totally destroyed, Ramana breaks with his arrows the nine walls of the fort in which Sit.a is imprisoned and liberates her He then with medicine re­stores Laksana to consciousness

828-326. Sugriva with his army of monkeys departs to his own king­dom. Ramana goes to Jambu-dvipa, where a great feast is held. Sita bears a son, to whom is given the name Lava

826-888. Hanumanta is appointed Minister to Sugriva. The two invite Ramana with his brother and Sita, and make a great feast. Hanumanta and Ramana are special friends, and keep

A RamayaTJ,a Story in Tibetan from Chinese Turkestan 207

up a correspondence. Sugnva having died, the monkeys offer the sovereignty to Hanumanta, who refuses (in verse, II. 831-388), but upon earnest supplication (in verse, ll. 834-837) consents

888-850. A tiff between Ramana and Hanumanta. The latter apolo-gizes (II. 343-848) -

khri tan nons ses gnon zin l!,gyod / / non bu las kyi phan mthos na / ri bo sin mod drin brjed bzin// dmal!, ba mthos na t.shon ma mchis / I lha dpal khyod lcyis nons se !!,am/ khens te l!,gyin ba ma lag.'I kyan / / dpyid ka !J,i chad na chus blugs na/ be ba br3ed bzin g-yens par gyurd/ mt.sho la brten pa l!,i nan dan nur/ / gud du b,phur du gnas ma mchis / / spre l!,u rigs dmab, ba bdag b,drab, la/ I mtho nas byams pa/ khyod las dkon /

"A myriad times I am known to have offended," that I regret with shame.

To elevate an off ending person is waste of labour Forgetful of kindness, he 1s like tree on a mountain. In elevating the low there is no profit Though by you, great lord, I am not regarded as offender Nor despised as puffed up, A fish flooded with water m the spring time, Forgettmg ... , becomes agitated, Goose and duck, which have their home m the lake, If they leap out on to the bank, are not m place. To one like me, a monkey of low race, Kindness from the high is, except from you, rare,

and they become friends as before 850-365. A vassal of Ramana, by name Benbala, revolts, and Ramana

leads an army against him, depositmg the queen mother and her son meanwhile with 500 ascetics on Mount Malaya [ri ma la ya error for Himalaya?]. He fails to return in time, and the queen, becommg anxious, wanders in search of him, depositing the child with the hermits. The child, how­ever, goes after his mother, and the hermits do not know what has become of him. Has Lava fallen into the water or

!!08 Frederick William Thomaa

been carried by friends? Considering Ramana's love for Lava and his consequent grief, they decide to create a sub­stitute in the Kusa grass. this they do, and by their power the created child receives consciousness and is in form indis­tinguishable from the real son. Sita upon her return finds with the hermits a boy like Lava and asks his name. They reply 'it 1s Kusa,' and Kusa becomes his name. Sita is con­tent to have the two similar sons.

865-410. Ramana returns, after reducing Benbala to subjection. On one occasion, while wandering about, he sees aLitsa-byed, Dri­ma-dag-pa (Licchavi Malaputa? RaJaka?), and his wife quar­relling. The former says 'This harlot is unlike other women. Where is a moon-image hke you, who, not content with her husband, sleeps with other men?' She replies, 'what do you know of other women? For instance, Sita devl, the excellent wife of kmg Ramana For 100,000 years she was with Dasa­gr1va, king of the demons But see, she was rescued all the same and is a dear wife. Do you know the nature of all women?' Ramana 1s troubled and determmes to find out from the woman how the nature of woman differs from that of men He arranges a secret meeting with the woman, and asks her the question. She explains (II 384-389) -

rims nad tsha bas !!,dugs pa!J,i myi// rgyun du gan,</ chab dran ba b:.in// bud med !!,dod chags rgyun du ,q-yo / skyes pa bzan po rtag du dran/ gzan gyis mthon sin thos pa dan / dpya!J,s par dogs pa ma mchis pa Ml/ sa phyogs dben par bstegs .<1lan chad/ su yan !J,khyigs pa ma mchiste / phyis /cyan !J,dod pa !,ii don sgrub bo / skyes pa rnams ni khrel myed pas// phrad dgu!,i g:an la smras na yan// dpyas par dogs /cyan myi !J,dzem par/ da dun rg71ag pa ma gzigs sam/

As one tortured by fever's heat Constantly remembers icy streams, A woman, ever agitated with desire, Constantly remembers a handsome man. So long as she is seen by others and heard,

A RamayaTJ,a Story in Tibetan from Chineae Turkeatan 209

No blame attaches to her: When she has come to a lonely place, Unrestrained by anything, She effects even with outsiders the object of her desire. Persons again, being unabashed, Though the people speak unfavourably of them, When blamed are not shame-faced, But are all the more proud - see you not?

The king believes her, and tells Sita that she may go where she pleases (II. 393-396). She departs with her sons to a her­mitage. Ramana remams in his palace.

4lo-439. Ramana mvites the monkeys Hanumanta is surpnzed to have no evidence of the Queen, and Ramana tells him the story. Hanumanta by showmg in what circumstances he found Sita demonstrates the absurdity of the suspicions Ramana is convmced and sends for Sit:a and her children They give a great feast in honour of Hanumanta, who returns to his own realm Ramana and Sita and the children hve happily m their palace

DOCUMENT B

LI. 1-20 Vaisravana, expelling the Yaksa Kore (Kuvera), 1s made by Brahma kmg of the Three Worlds and the Four Dvipas. [The gods] pray to the Devar~i and Sri Devi, who agree to beget a son without bodily contact and each gazing steadily m a 1mrror ultimately they have a son, who is called Vai­sravana He expells the demons from La:rika-pura, sendmg them down to hell, and fills the land with men and gods

20-52. After many generations of the demons the Y aksa Kore has a son Mal-hya-pa-ta (Malyavant), who is found in a quilt. The people of the country ask who are his father, mother and relatives, and he mqmres of a Brahman Ratana, who m­forms him that his father is the great and powerful Y aksa Kore, who has been deposed by Vaisrava"9-a Desirous of retribution, but powerless, he decides to practice austerities in the garden (8in-rta1.J,i-tshal) of the Vaijayanti palace of Brahma's son Svapasina [sic, for VaisravaQ.a?] He lauds the Devar~i, as ruler of the Worlds, son of Brahma and so forth (verse, II. 31-33), and the Devar~1 inquires the object of his austerities. After three days he replies that to the kind R~i,

210 Frederick William Thomas

who warms the world like a sun, he wishes to give his daugh ter Mekesina (11. 87-88) -

lJ,3ig rten dag la gfii ltar dro ba dkon / / bdag la byams par khyod las gzan myi bzugs / / drin bzo lan du bdag gyi bu mo lJ,di// Me ke si na lJ,bul na b:es su gsol / /

In the worlds a warmth like the sun is rare. For affection to me there is none other than you. In return for kmdness this daughter of mine, Mekesina, I offer take her, I pray.

The R!j1 is displeased at an idea so inconsistent with his vows, but at the same time loth to violate the maxim of accepting what is offered m kindness: so he says nothing. Malyavant presses him (ll. 42-44) -

kun la bde mchog lon spyod ni / / gzi mdans mdzes mal palJ,i gro.<JS dan bsten pa lags// dpal brjid b,phrul gyi bsten grogs lha mo b,di/ khyed la !}tshams :es kun bstod sku myi nas// mgo phan dmalj, zin/ /.-tkabs kyi zur myig chen/ yid tsam IJ,dzum zin //g:og .<1tegs tshul la mkhas// b,jo sgeg che zin/byi byad da,9 la brtson// lha mo !µli lJ,i nan na/dkon ba JJ,di Mes sig

Everywhere the enjoyment of happiness (?) Depends upon a loved companion as lady of the house (?), As companion to support your wonderful, brilliant greatness, this

lady Is a match for you - further praise is unnecessary (?). In rank infermr, quick to see what is opportune; With a smiling mind, wise in the ways of playfulness; Of great fascination, earnest m doing what is to be done; -In this lady are these rarities take her I pray;

and the R~i at last consents. To him and Mekesina are born three sons, Dasagriva, Utpa[la]kan:ia (?) and Birisana. To the eldest the grandfather Brahma gives ten heads and cor­responding strength and so forth.

52-68. Mal-hyan-pan-ta proposes to the 'sons of the gods' to go to Lanka, the country of their uncle, and they consent. Mal­hyan-pan-ta advises them to obtain a boon from the gods;

A RamiiyaTJ,a Story in Tibet,an from Chinese Turkestan 211

but the gods, discerning their evil purpose, do not grant it. However, a teacher whom they had first worshipped and who had been made by Mahadeva goddess of speech asks for their success; and under her influence the lord of the gods gives them the rule over the gods. From that time they quell the gods, and, defeating the gods and men of Lanka-pura, fill the place with demons. They make Dasagriva king· with a court of gods and Na.gas he enjoys himself, and the kmg [Dasagriva ? Vaisravana ?) goes where he pleases.

68-77. Dasagriva's Ministers suggest to him that to dominate others is nothing there is the great Vi~nu, lord of the world. To Dasagriva's inqmry as to where Vi~J).U is they reply that he is in the Ocean of Milk in the north Dasagriva goes in his chariot, and his coming is reported to Vi~l).U, who says "Let him come in." Dasagriva says he has come to fight and why then should he come in? Visr,m says 'We two, being superior to others, ought not to fight at once hke dogs Today come m and sleep tomorrow we will fight' Dasagriva enters; but as Vi!jlnu does not rise to greet him, he becomes enraged and demands to fight. Vi!jlnU says 'Nothmg else 1s necessary; take one of my earrings ' Failing to effect this, Da&agriva is de1ected, and, having bowed before V1s1:m's feet, goes back to Lankii-pura.

77-89 The gods, assembled m heaven, take counsel lndra mquires of HJig-rten-gyi-phyva (the fortune[-teller] of the world), who states that Dasagr1va, king of Lanka, has enslaved the gods and 1s doing mischief He asks who first empowered him and is told that it was Mahadeva The gods go to Ma­hadeva, who upon inqmry by Brhaspati says, 'I did not em­power him I do not know that Dasagriva is so. If he is harming the world, well, I am occupied with a vow appeal to Vi!jlnu' They go to Vis1;m, who says,' At present there is a king of Jambu-dvipa, by name Dasarite [Dasaratha] He having no son, I will appear as his son and quell the demons.' Vii:mu appears as Dasaratha's son Ramana, and Vi!jli;tu's son as the younger brother Lakfana. The gods also are born variously A woman capable of destroying the state of the demons is conceived by a wife of Dasagriva.

89-9~. The child of Dasagriva's wife is born, sent away, found by husbandmen and named Lilavati (as in A).

212 Frederick William Thomas

93--99. Dasaratha pays respect to 500 R~is living on Kailasa and begs a son. They send him a flower, bidding him give it to his queen. The chief queen gives half of it to the junior queen, and two sons are born, the son of the junior queen being the elder by three days.

ADDENDUM

A fourth document (D = Fr. 63 = Vol. 56, fol. 11) has come to light. Similarity in paper (though that of D seems thicker), script (recto ll. 51 of Tibetan, verso ll. 31 of Chinese), size (Tibetan c. 25 cm., Chinese c. 20 5 cm.) and distribution of lines, suggests that this is in reality a (pr10r) part of A The subJect matter corresponds to ll 1--47 of B, preceded by a laudatory description of the country ruled by the Yaksa Kore, which would be Ceylon The agreement with B is not literal, and iri the proper names, as rendered into Tibetan, there is some divergence· thus for Sin-rtafri-tshal 1t gives Sgyed (skyed) mo­tshal, for Svapasina Bi.surasena, and for mekesina megasina - F.W T.

OXFORD UNIVERSITY.

HOW TO LIVE HAPPILY ON NOTHING A YEAR Translated from the twelfth book of the M ahabhiirata

BY ARTHUR WILLIAM RYDER

I MAGINE that what is Does not exist at all;

Then will you not be grieved, However low you fall.

Your deeds of yesterday And those that went before

Are past and gone; for them You need not sorrow more.

What was, no longe-r is, What was not, will not be·

The past need bring regret To none from blindness free.

Where is your father now? Where may his father be?

You do not see their hfe, Your hfe they do not see.

And you, 0 King, and I. With every foe and friend,

Will surely cease to be, Since all things have an end.

The men of twenty years, Or thirty years, or more,

Will all be dead when once A hundred years are o'er.

And even should riches cling To you, do not repine,

But seek for comfort in The thought, " They are not mine."

214 Arthur William Ryder

If man leave not his wealth, Then wealth the man will leave.

Since this is surely so, Why should the prudent grieve?

And poor men live to-day Who calm a nation's fears

By wisdom and by strength, Your betters or your peers.

They do not grieve hke you; Then cease to grieve at length,

Surpass or equaJ them In wisdom and in strength.

Consider what the past And what the future teach,

Not grieving at events, Indifferent to each.

Desire the things you may, Not those you may not gain;

EnJoy the gifts of fate -Those lost deserve no pain.

And he is surely fool Who curses God a,nd weeps

For what he had, and lost -Ingrate for what he keeps.

And be not troubled if Men show unworthiness

Of wealth they have; for thus Your sorrows grow no less.

Endure though riches smile On all but you alone;

For men of sense enJoy The wealth that others own.

Yea, brave and righteous men In willing sacrifice

Abandon wealth and home, Knowing salvation's price.

How w lifJe Happily on Nothing a Year

Even kings a kingdom leave And count their loss a gain:

In pain's extremity They seek the end of pain.

From such men learn to find In penury, relief:

Grief often comes as joy, Joy wears the form of grief.

Nay, who would set his heart On gold that ends a~ dross,

On life that ends as death, On love that ends as loss?

The pole-tusked elephant Is like the sage, for he

Lives lonely m the woods, Gladly, and frugally.

UNIVERSITY OF CALIFORNIA.

i15

HINDU-ARABIC NUMERALS

BY WALTER EUGENE CLARK

THE numerals which we call Arabic are so called, not because they were invented by the Arabs, but because the Arabs transmitted

them to Europe So much is certam Arabic literary tradition, as generally interpreted, declares that the nine numerals with zero and place value were invented by the Indians, and that they were adopted by the Arabs during the last quarter of the eighth century A o. In conformity with this Arabic tradition these numerals were generally called Hindu all through the mediaeval and Renaissance periods in Europe They were commonly called Arabic only from the sixteenth century. The general opmion of mathematicians, Sanskr1tists, and Arabi<' scholars, based on this Arabic tradition and on the Indian evi­dence itself, has been that these numerals with zero and place value are to be traced ultmately to India

Durmg the past few years Kaye has written a senes of articles 1

m which he disputes this general opm10n with greater and greater vehemence and certamty He claims that he is the first one to apply a stnctly sc1entrfic method to this particular problem His method con­sists m denymg all vahd1ty to Indian hterary tradition and to Indian manuscripts previous to the date at which the manuscripts them­selves were written The only admissible evidence is that of mscrip­tions and coins On this basis he tries to prove that the numerical symbols with zero and place value were unknown m India untI_l the end of the mnth century A D , and that Indians and Arabs ahke must have taken them from some third source Having reached this funda­mental position he makes use of Indian literary trad1t10n only m so far as it does not contradict his scientific conclus10n :An passages which are defirutely m contradiction with thii, conclusion are later m­terpolations. The Arabic literary evidence is handled in the same way All passages which suggest an Indian origm or praise Indian ac­complishment in any branch of knowledge are legendary Such pas­sages as criticize the Indians and belittle their knowledge or methods

1 Journal of the Asiatic Society of Bengal (hereafter cited as JASB), 1907, p 474, 1908, p 111, 1908, p 293, 1911, p 801, 1912, p 349, Biblwfheca Mathematica, x, 289, Journal of the Royal, Asiatic Society (hereafter cited as JRAS), 1910, p 749, Indian Mathematics, Calcutta, 1915, Scientia, xx1v (1918), 53, and xxv (1919), 1; lsu,, 1919, p 826.

!?18 Walter Eugene Clark

are accepted as just and reasonable} On the other hand, it is very noteworthy that Kaye fails entirely to apply this same hypercritical method to Greek, Latin, and Chinese literary evidence. This is ac­cepted as valid without criticism and without the support of inscrip­tions.

The whole problem of the Hmdu-Arabic numerals was admirably summed up in 1911 in the little book of Smith and Karpinski, The Hindu-Arabic Numerals An article by Ruska 1 has shown pretty conclusively the superficial nature of Kaye's treatment of the Arabic evidence, and his conclusion (JASB, 1907, p. 498) that there is abso­lutely nothing Indian m al-tlwarazmi

After his elimination of the Indians as mventors of the numerical symbols with zero and place value, ~aye suggests a Greek or__1K!_~ for these and for most or all of Indian mathematics, but his reasons for this conclusion are expressed with great vagueness The most tangible passage is the followmg 2 "It was during this period also that Damas­cms, S1mphcms (mathematicians of some repute) and others of the schools of Athens, having heard that Plato's ideal fonn of government was actuaUy realised under Chosroes I of Persia, emigrated thither (circa A n. 532) They were naturally disappointed, but the effect of their visit may have been far greater than historical records show." There is no citation of authority for this statement, no critical analysis of its historical value, and no reference to mscript10nal evidence, as is demanded in the treatment of Indian literary evidence The only good authority seems to be Agathias (n, 30-31), a sixth-century Byzantme author, who wrote a history of his own times The passage states that these philosophers were so disgusted with the ideas and practices of the barbarians that they very soon returned to Greece. There Is no mention of numerals or mathematics, no suggestion that the brief visit of these Greek philosophers had any appreciable effect on Persian mathematics or Persian thought

Carra de Vaux,3 independently of Kaye, arrived at similar con­clusions, but expressed more definitely and emphatically and with additions which were welcomed with approval by Kaye in a later article. 4 The new theory of the origin of the digits 1s the following:

1 "Zur altesten arab1schen Algebra und Rechenkunst," m Sitzungabenchte <kr Heidelberger Akademie, 1917

2 Indian Mathematica, pp 15-16, 45, Scuntia, xx1v, 54 3 Scuntia, xx1, !'1178 4 Scientia, xx1v, 54. Kaye had anticipated the first two conclus1ons of Carra de

Vaux and vaguely suggested the tlurd Cf JASB, 1907, p 489, 1911, p 801

Hindu-Arabi,c Numerals 219

Firstly, the evidence for the Indian origin of the numerals with place value is entirely legendary and this legend is Persian. Carra de Vaux even goes so far as to say that the fables of Bidpai have nothing to do with India, but were developed in Persia by the Neo-Pythagoreans. For this amazing conclusion there is Just the bare statement, without the citation of any evidence Secondly, that the word hindi or hind is a mistake for handas'i, or is a parallel derivative from the same root, and means "measure, arithmetic, geometry," etc It refers, there­fore, not to Indian signs but to arithmetical signs m general. 1 (Thirdly, that the numerals with place value were invented by the Neo:..Platon­ists or the Neo-Pythagoreans, were taken by them to Persia, revealed to the Persians, and passed on by the Persians to the Indians and Arabs, and by the Arabs to Europe~ These conclusions are based entirely on vague generalizations. No effort 1s made to prove that the Neo-Platonists or Neo-Pythagoreans knew numerals with place value, and not the shghtest evidence is produced to show that these numerals were known m Persia before the eighth century The only Greek or Latin text known to me which throws any hght on the activi­ties of the Neo-Platomsts m Persia is the Solutiones eorum de quibus dubitavit Chosroes Persarum Rex of Pr1scianus Lydus, 2 one of the Neo­Platonists who went to Persia m the reign of Chosroes I This work deals at considerable length with questions of metaphysics and of natural science, chiefly on the basis of Am,totle and his school, but it does not contam the slightest suggestion of mathematics or of numerals with place value

So far as I know, the only early evidence that Neo-Platonists or Neo-Pythagoreans had any knowledge of numerals with place value

1 Granted that the word handa,si has often been m1stram,lated as 1f 1t were hindi, still such early mediaeval works as the f.9_!:!rte<:nth-:.C'.elltur.Y '1t'J'l>oq,opla KaT' 'Iv8ol,s of./ ~~. the Lwer Abaci of Leonam__m l'lBil (120~ An), the Algorilmi de numer_q_bi.­dmum of the twelfth century (a translat10n of the eight-century work of al-!Jwaraznn), and many of the other early European works on algorismWJ, seem to prove clearly that the Arabs themselves regarded India as the place of or1gm of the d1g1ts, and under­stood the word hindi to mean Indian The attempt to explam away the word hindi seems to me to be futile The further content10n that m Europe m the mediaeval period the word India was a term of very general meanmg and does not necessarily refer to India itself might be vahd for mdependent works m Greek and Latm m Europe, but here we are dealmg with works which are translat10ns from the Arabic, or based directly on Arabic works and trad1l!ons, and Arabic hmdi could not have had the same vague meanmg to the Arabs that the word India had m Europe See also Ruska, loc. cit, p 114

2 Preserved only m this Latm version, and published by Bywater m the Supple­mentum AT1Stotelicum

Walter Eugene Clark

is contained in a passage of Boethius (about A D. 500). fn the Geome­try of this author are given nine numerical symbols which are called apices, and the statement is made that they were used by the Pytha­goreans for calculation on the abacus, which they had invented and named mensa Pythagorea m honor of Pythagoras. A great contro­versy has been waged over the authenticity of this passage. It is still unsettled The description of these numerals does not occur in the Arithmetic of Boethius, where it would be in place, but in the Geome­try, m the midst of a discussion of angles, the subject is changed ab­ruptly to a discussion of d1ff erent classes of numbers, and then 1s given an account of the abacus and a representation of the nme numerical symbols used by the Neo-Pythagoreans m connection with it None of the early successors of Boethms, who used his work and quote him, make any mention of this important passage (None of the manu­scripts of the Geometry are older than the eleventh century and there 1s no other trace of numerals with place value in Europe m the earlier Greek and Latin literature, or in later literature until a Spamsh manu­script of A.D 976, m which they are de:fimtely called Indian 1) If the same method which is employed by Kaye when dealmg with Indian literary evidence 1s applied to this doubtful passage of Boethms, we must admit that it cannot be used as certam evidence for numerals with place value earlier than the cl.e.Y.~l!lh ce!!tl.U",Y, the actual date of the manuscripts themselves After the hypcrcntical method used m demohshmg the theory of the Indian origm of the digits, the vague and slipshod method employed rn bmldmg up a pos1hve theory of Greek origin 1s most unfair and biassed

The,e is a curious passage quoted by Nau 2 from the well-known Syrian writer Severus Sebokt The date is A,D 662. Sebokt speaks of " the subtle discoveries of the Hm<lus m astronomy, discoveries which are more mgemous than those of the Babylomans, and their clever method of calculation, their computation which surpasses words, I mean that which is made with nme signs If those who think that they have reached the acme of science Just because they are Greek had known these thmgs, they would perhaps have been con­vrnced, although late, that there are others who know somethmg." Kaye tosses this passage aside 3 as obviously worthless, with the words,

1 Hill, Archaeologia, lxn, 151, 170 Another Spamsh manuscript of AD 992 gives the same characters

2 Journal AsiatiqUe (hereafter cited as JA) (1910), 11, 225-!2!27, Revue cle l'Orwnt Chretien, 1910, p 250

• lndum Mathematics, p 81

Hindu-Arabic Numerals

"but his authority makes such erroneous statements about 'Indian' astronomy that we have no faith in what he says about other ' Indian ' matters." He does not point out the obviously erroneous statements. Surely the oldest known mention of Indian numerals deserves a more critical treatment than this. More certain reasons than Kaye's per­sonal prejudices and vague generalizations are necessary if the passage is to be cast aside as legendary or a later forgery. It may or may not be valid, but in the present state of our knowledge it certainly has as much value for the sub1ect under discussion as the passages of Agathias and Boethius 1

Not without possible value is a passage in the Chronicum Paschale: 2

'Ev TOLS XPOIIO~S T~S 1rvp-yo1rodas EK TOV ')'EIIOVS TOV 'Ap</>a.~ao av~p TLS 'Ivoos-0.JIEcjJO.llT/ uo<f>os aurpov6µos, 6116µaTL 'Avoov/36.pws, OS Kai UVJ/E')'pa.if.,aro 1rpwros

'Ivoo'ts aurpovoµi.av Whether or not this refers to Aryabhata, as has been suggested, is very uncertam 3 The chromcle ends with the year Ao 629, but 1s based largely on earher sources The date for which the statement is vahd 1s un'.certam, but whatever its date may be, the passage may have historical value as a partial corroboration of Sebokt and as provmg at a comparatively early date in Europe the knowledge that the Indians had cultivated astronomy

There is the same looseness m Kaye's treatment of Chinese mathe­matics which, according to him, had much mfluence on early Indian mathematics He expresses hmself as follows 4 " Mr Yo~h10 Mikami states that there is no evidence of Indian mfluence on Chmese mathe­matics On the other hand, he says, ' the d1scoverie~ made m Chma may have touched the eyes of Hin<loo scholars ' " This statement is. made on page 23 of M1kam1's book, The De1,elopment of Mathematics. ut China and Japan, but it 1s directJy preceded by the words "It is certam that the Indian learnmg exceedingly mfluence<l Chmese thought but at the same time " Further, in a chapter entitled On the Indian Influence (pp 56-61), Mikami repeatedly suggests the possi­bility of Indian inlfiuence on Chmese mathematics and astronomy. "Things Indian exercised supremacy in art and literature, in philoso­phy, m the mode of life and the thoughts of the mhab1tants, m every-

1 For an account of Sebokt and lus works, see Baumstark, Geschichte der syrischen Literatur, and Ruska, loc cit, p 46 Sebokt 1s not "a certam Sebokt," but one of the most famous and trustworthy of Syriac authors

2 Bonn ed1t10n of the Corpus acriptorum hUJtoriae Byzantmae, 1, 64 Cf Georg10s Kedrenos (1, 27), m the Bonn ed1t10n of the same series

8 Colebrooke, Essays, 11, 884-885, 425, Lassen, Indiache Alterthumskunde, u, 1148-1149, Abhandlungenfur du Kunde des Morgenlandes, 1v, 288

4 Indian M athematws, p 4 I

Walter Eugene Clark

thing. i It is even said, astronomy and calendrical arts had also felt their influence. How then could arithmetic remain unaffected? No doubt the Chinese studied the arithmetical works of the Hindoos." ... "Will it be too rash for us to assume that the problem of the circle-measurement had been transplanted from Indian soil to the fertile land of the Middle Empire? " . . " But all this, it must be confessed, remains mere conJecture, there is nothing positive that serves as an evidence of any actual influence upon the Chinese mathe­matics. In astronomy some of the Hindoo theories were studied, and there are still extant some part of them in translations and quotations But neither a single problem nor a single rule for the solution in the domam of mathematics now remains that is definitely known as of Indian or1gm The fact, however, that the Indian mathematical works had been studied m China can by no means be demed " Then he sums up by saying, " The possib1hty of the Chmese mathematics hav­ing been mfluence<l by the science of India may well be conJectured from the meagre account here given As for exact information, we have none."! ln the next chapter he shows that from about AD 618 on there were often Indian astronomers on the Astronomical Board in Chma, and that they composed calendars which were in general use. He also gives in Chmese the names of six Indian works on mathe­matics and astronomy which were translated from Sansknt into Chinese at an early date Mikam1's book 1s praiseworthy from the purely descriptive poutt of view, but m the matter of chronology he merely follows orthodox Chmese literary trad1t10n and Kaye blindly follows him 1 The Chou Pei is regarded as vahd for the twelfth century B c., the Arithmetic m Nine Sections for the second century B,C., with a revision m the third century A D,, the Arithmetical Classic of Sun Tsu for the first century A D., and the Sea Island Arithmetical Classic for the third century A n. It 1s very doubtful whether these texts in their present form are vahd for the dates assigned to them by tradi­tion. They may have been much reworked. For instance, Mikami gives a long and excellent analysis of the Arithmetic m Nme Sect10ns, but remarks that " Hayashi has given a summary of the same text which is utterly different from what I know of the book." Nothing definite 1s to be gained from the jumbled conclus10ns of Mikami; but Kaye's quotation of him is so unfair and misleading that one becomes skeptical of the rest of his " scientific " method.

The possibility, and even the certainty, of manuscript interpola­tion has always been recognized by philologists and has to be guarded

1 Indian Mathematics, pp 88-40.

Hindu-Arabic Numerala

against carefully; but there is no reason for assuming that Chinese and European manuscript traditions are any more exempt from inter­polation than are the Indian and Arabic manuscript traditions. By his method Kaye does undoubtedly eliminate some doubtful or worth­less material wh'.ich has been given undue weight by earlier scholars, but at the same time he undoubtedly eliminates much that is valuable and correct.

In the earliest Indian inscriptions we find numerical symbols with­out zero and without place value That 1s, there are separate signs for the numbers from one to nine, for ten and multiples of ten, for one hundred and multiples of a hundred, for one thousand and multiples of a thousan!d. St1ch numerical symbols begm m the third century B.c and are used exclusively in mscr1ptio;ns down to about Ao 600 (or, as Kaye insists, down to the end of the mnth century) After that time this method is used with decreasmg frequency, along with other methods which constantly gamin favor, down to the twelfth century or so At some time, and this exact date 1s the chief matter under dis­pute, a new system came mto use In this later &ystem there were only ten symbols, those for one to mne and zero These were used with place value so that they sufficed for the expression of all pm,s1ble numbers {fhe earliest supposed occurrence of any of the first mne symbols ism an inscription of A D 595, and between then and the end of the nmth cen­tury we have about twenty mscript10ns m which they are used: In­scriptions contaming the old symbols without place value are 'much more frequent From the errd of the mnth century the symbols with place value are med with mcreasmg frequency By the twelfth cen­tury the old &ymbo1s without place value have almost entirely disap­peared , The &ymbol for zero occurs first with certamty m the mnth or tenth 'century~, Of these twenty mscr1pbons some are regarded by Indian ep1graphist~ as later forgeries, some as doubtful, and some as genuine Most of them are land grants mscr1bed on copper plates. Grants of land or villages were often made by a kmg to some man or group of men for special service rendered, and s'uch land was exempt from taxation. Consequently there was great temptation to forgery of such grants, especially after periods of pohbcal disturbance; and there are many undoubted cases of such forgery, a large proportion made in the eleventh century A D. in southern India. These can some­times be detected with considerable certamty palaeographically by the unsuccessful imitation of the older forms of the letters used m the same

1 The occurrence m the eighth century referred to by Bayley (JRAS, 1881, p. 27) cannot be verified. No argument can be based on 1t

Walter Eugene Clark

part of the country, by the inferiority of execution, or by some inac­curacy in the genealogy or account of the older king as gained from a study of older and genuine inscriptions of the particular king named. Unfortunately our mscriptional material is fragmentary and has many gaps Since the discovery of undoubted forgeries epigraphists have been very skeptical and critical, and inclmed to mark as doubtful any grant about which there is the slightest suspicion. Kaye takes ad­vantage of Fleet's article in the Indian Antiquary for 1901 m which about sixty grants are critically discussed and hsted as spurious to de­clare that all of these early grants which contain numerical symbols with pl'ace value are later forgeries, or, at least, even if the text of the grant itself seems to be genuine, that the numerical symbols with place value have been added after the end of the mnth century. He makes one possible exception for an inscription of A.D. 813, but thinks that this too must be treated provisionally as a forgery 1 After the ninth century the material is so ample that the forgery of all the grants can not possibly be mamtamed The use of numerals with place value in inscriptions soon becomes the rule, not the exception The matter of these early grants 1s one tu be decided only by skilled ep1graph1sts as the material mcreases I shall not discuss 1t here, except to remark that it 1s by no means certam that all the mscnpt10ns in quest10n are forgeries At the most all that can be ~aid 1s that some of them are forgeries, that some are doubtful, and that the question cannot be decided defimteJy on the basis of the present evidence The problem is an open one

We have considerable fragments of a birch-bark manuscript called the Bakh~hah manuscript 2 discovered some thuty years ago m the extreme northwestern part of India These are part of a large work on mathematics g1vmg rules and problems and complete solutions m nine numerical symbols with zero and place value It bean, no date Hoernle tentatively dated the compos1t10n of the work between A D

330 and 400, and the wr1tmg of the manuscript between the seventh and tenth centuries Thibaut has expressed the op1mon that the manu­script was written between A D 700 and 900, while the work itself may be older. Kaye places the work and the manuscript m the twelfth century or later. 3 His arguments are not conclusive. The date of

1 JASB, 1907, p 481, Scuntia, xx1v, 55, JASB, 1910, p 756 2 Hoernle, Indian Antiquary (hereafter cited as IA), vol xvn, and V erhandlungen

des VII Internatwnalen Orientaluten-Congreaaea, Ar1sche Sectwn, pp 127-147

a JASB, 1912, p 249

Hindu-Arauic Numerals

the work and of the manuscript is unknown. It cannot be used as evi­dence upon which to build any scie'nti:fi,c conclusion.

Granted that the inscriptions and the Bakhshali manuscript do not at present furnish proof of the employment of numerical symbols with zero and place value in India earlier than the ninth century, there re­mains a considerable body of literary evidence which has been entirely ignored by Kaye, but which has, it seems to me, defimte historical value.

In the Viisavadattii of Subandhu 1 occurs the following passage: " The stars, because of the nothingness of this world of transmigration, are like cipher& scattered in the sky, as if on the ink-black rug of the Creator who reckons the sum total with a bit of the moon for chalk." The word translated " cipher " 1s §unyalJrindu, " the dot which repre­sents emptmess " The earliest form of the zero, as given m inscrip­tions of the ninth and tenth centuries and m the B'akhshali manu­script, 1s merely a dot Also, m the earliest occurrence of zero m con­nection with the Arabic alphabet (AD 87S) the zero is represented by a dot 2 The exact date of this work is uncertam, but it can be assigned with confidence to a date not far removed from AD 600 smce 1t is later than the Nyayaviirttika of Uddyotakara, which can be assigned with certamty to the sixth century, and earher than the Harsacarita of Bana which can be assigned with certamty to the early seventh century. Subandhu i'l named m the Gaudavaho of Vakpati, which dates from the beginning of the eighth century It 1s impossible for the V asa­vadatta as a whole to be as ]ah' as the tenth century 1 The passage in question is found m an edit10ns and m an manuscripts which have been reported Phi]ological1y there is not the slightest reason for consider­ing the passage to be an mterpolation of date later than the end of the nmth century.

Numerical symbols are not 1.1,<;ed in any of the old Indian works on mathematics and astronomy which have been preserved, except m the Bakhshali manuscript These works are all m verse, and in verse such numerical symbols could not be used. It was necessary to use the ordmary names for the numbers or numerical words or combmabons of consonants and vowels with numerical value Therefore the occur­rence of any of these methods does not suffice to prove with certainty that the author was ignorant of numerical symbols with place value

The commonest way of expressmg numbers in literary works from the sixth century on is that of usmg words with numerical meaning.

1 Ed Hall, p. 18!i! 2 Karabarek, m Wiener Zeitachriftfur die Kunde des Morgenlandea, x1, 13 1 Gray, Vtisavadattil, pp 8-1~.

Walter Eugene Clark

Words which denote pairs such as twins, eyes, ears, hands, and the like, mean two. Kha, §unya, iikiiaa, am.hara, viyat, etc., which denote "empty space, hole, sky, nothing," mean zero. The oceans are 4. The teeth are 82. The fingers or nails are 10. Such words denoting numbers from 0 to 49 are in common use. There is a host of synonyms m Sanskrit, so that an author writing m verse need never hesitate for a word which will fit into any place in any metre It is not possible to express easily in verse a large mass of numerical data by means of the ordinary words denotmg numbers.

The earliest Indian inscription which contains a numerical word is from the eighth century, or, if this and another inscription of A D. 818 are forgeries, from AD 945 But numerical words are used in Sanskrit mscriptions m Java m the eighth century and m Indo-China in the seventh century (beginnmg m A D. 604) 1 The system must have been m use in India earlier than m these distant colomes, unless, as Kaye suggests, 2 it "was mtroduced about the mnth century, possibly from the East " This opm10n seems to be based only on the pr10rity of the mscr1ptions of Java and Cambodia. It disregards completely the certam evidence of Indian literature There 1~ not a shred of positive evidence m favor of it

Although in our fragmentary mscriptlonal material m India there 1s no certam trace of this method until the eighth or mnth or tenth century, there 1s ample literary evidence for its earher use 3 Vara­hamihira employs this method m his Brhat Samhitii and m his Paficasid­dhiintilcii. The mstances are so many that 1t 1s not worth the trouble to enumerate them Both works can be dated with certamty in the sixth century AD It 1s used by Brahmagupta m his Brahmasphutasid­dhiinta, which can be datt>d with certamty at the begmning of the seventh century The date of Lalla's Sisyadhivrddhida, which also makes use of numerical words, 1s not certam Such evidence as there 1s pomts to the sixth or seventh century This same method is used in the Silryasiddhanta. The or1gmal text of this work was earher than A D 500, smce Varaham1h1ra gives an abstract of some sections of it; but our preserved text d1ff ers m some particulars from the one described by Varaham1h1ra and must therefore be a reworked text of uncertam date. It cannot be proved with certainty that the onginal text used numerical words V araham1h1ra in his abstract of the four

1 IA, xvm, 24, 48, Barth, "lnscr1pbons Sanscr1tes du Camhodge" in the Notwe, des manuacript, de la Biblwtheque Natwnale, xxvn, 81

2 Indian Mathematics, p Sl, JASB, 1907, pp 475 ff. 1 Buhler, lndiache Palaeographie, pp. 80-S!l.

Hindu-Arabic Numerals

other early Siddhantas also makes use of numerical words, but it can­not be concluded with absolute certainty that these Siddhantas used such words. It is barely possible that Varahamihira may be sum­marizing them in language of his own It seems unlikely, however, that a large mass of numerical data should have been expressed in verse with nothing but the ordinary names of the numbers. It seems to me likely that even the five old Siddhantas, none of which have been preserved m full form except the reworked Silryasiddhanta, made use of numerical words

In order to uphold his position, Kaye would have to maintain that the above-mentioned texts of Varaham1hira, Lalla, and Brahmagupta, m which numerical words are used commonly, are not originals at all, that we have only completely rewritten texts, versions dating from the tenth century or later, in which all of the numbers have been ex­pressed in an altogether different system from that of the origmal texts. It is impossible for any Sanskr1tist who has worked with any care through these early mathematical and astronomical texts to subscnbe to such a theory

Moreover the beginnmgs of this method can be traced back with certamty beyond A D 500, to still earlier texts for which no definite date can be given. The Vedanga on Metncs, which goes under the name of Pnigala gives examples m very simple form for the numbers 1, 4, 5, 6, 7, 8, 10, 11, 12 The Jyotisa Vedanga m four passages uses numerical words to denote the numbers 1, 4, 12, 27. The Srauta Siltras of Katyayana and Latyayana use the words qayatri and Jagati, which are the names of metres, m the direct sense of 24 and 48, the number of syllables which make up the metres 1 None of these texts can be later than the fifth century AD and the last two are almost certainly pre­Chnstian

This system of numerical words might of course be used in connec­tion with the old numerical symbols without place value But when m the Silryasiddhiinta (1, 29) the number 4,320,000 is expressed by khacatuskaraddrnaviih, that is to say, four zeros or blank spaces, the teeth (S2), and the oceans (4), surely a system with place value is im­plied. In giving numbers by this method the words are always given carefully m a de:fimte order m such a way that as read the numbers denoted by the words are to be apphed from right to left. Likewise, rn the Brhat Samhila of Varahamihira (vm, 20) the number 3,750 is

1 For all the texts Just mentioned, see Weber, lndurche Studum, vm, 166-167 and Vber den Vedakalendar Nament1 Jyoturham m the Abhandlungen of the Berhn Academy (1861), p. 6

Walter Eugene Clark

expressed by emptiness (0), the arrows (5), the mountains (7), and the Ramas (3). Surely this implies the recognition that the numerals which make up the number occupy four places. Under the old nota­tion this number would have been expressed by three symbols, that for 3,000, that for 700, and that for 50 It seems to me that the use of long compounds which describe large numbers of several places proves the existence of numerical symbols with place value and zero. The remarks of Woepcke 1 and Buhler 2 are still valid Whatever may be the date of the present Buryasiddhanta, and whatever system may have been used in the five old Siddhantas, Varahamihira carries us back to the beginning of the sixth century for the use of numerical words based on numerals with zero and place value.

The words kha " sky " and silnya (" empty ") imply either the use of a symbol for zero or a blank space. It 1s possible, but not certain, that the use of a symbol for zero was later than the use of the other nine symbols. That is, it 1s uncertam whether the system implies place value in our exact sense of the word, or whether it merely im­plies an abacus, or at least a board divided into perpendicular columns, the columns having the place values of umts, tens, hundreds, and so on, while the column which represented an order of numbers that did not happen to be represented was left blank without any special symbol. Kaye demes that any form of abacus was used at an early date in India. I have not yet found definite evidence for the use of a board divided into columns which had place value, but the words kha and silnya, used m connection with the system of numerical words, imply, it seems to me, either such a board or a symbol for zero as early as the beginning of the sixth century.

There is considerable evidence in India for the use of a board and chalk, or of a board on which dust or sand was sprinkled. Such a board might have been divided mto columns and made to serve every purpose of the more highly developed abacus. The Arabic tradition, which is treated as legendary by Kaye, maintains that the Indians calculated on boards covered with dust or sand, or with white chalk on a black board, and refers to boards diVIded into columns The western Arabs called their numerical symbols gobar, "dust numbers," and derived them from India. In Brahmagupta's Brahmasphutasmdhanta, x, 62, 66, 67 the word dhiilikarma, " dust work," 1s used as a synonym of garpita, "calculation." This, it seems to me, is sufficient proof of the authenticity of the Arabic tradition. The passage quoted above

1 JA (1868) 1, 447. 2 As quoted by Bayley m JRAS, 1888, 28-!M

Hindu-Arabic Numerah

from the Vasavadatta proves the use of chalk and some dark surface on which calculations could be made by about A,D, 600. Varahamihira (Paiicasiddhantika, iv, 37) remarks that even an ignorant fellow can reckon with Imes made by chalk. There is also an important passage in the Divyavadana (p. 263) : 1 Bhilriko ga'l'),itre krtavi svetavar'l'),am grhitvd ga1J,ayitum arabdhah pasyati yatha Bhagavatd vyakrtam tat sarvam tathaiva: " Bhurika was skilled on the instrument for calculation. He took a piece of chalk and began to calculate. He saw that everything predicted by the Holy One was true " The passage deals with astro­logy, the prediction of the future of a child from the position of the planets at the time of his birth, and demands complicated mathemati­cal calculations. The suffix of the word ganitra denotes means or in­strument The word must denote some special contrivance for cal­culation, a specially prepared board, but whether this was divided mto columns is uncertam Elements in this text go back to the second or third century B c. at least, but in its present form 1t 1s later It must, however, be dated in the early centuries of the Christian era. Varaha­mihira's BrhaJ Jiitaka shows a great development of elaborate astro­logical calculations at the beginnmg of the sixth century 1 involving much arithmetical and algebraical work The words phalaka and ganana, meaning " wooden board " and " reckonmg " can be traced back to Jataka, 1, 451, Mahiivagga, i, 49 and other Buddhist texts, the earliest of which are probably to be ascribed to the period between the fifth and third centuries n.c. Although the use of a board and training in calculation can thus be traced back mto the pre-Christian period, there is no actual description of a board ruled into columns. But the fact that words meaning " emptiness " are used regularly in the later period to denote zero naturally suggests a board and a reckoning by columns with place value, before a symbol for zero was employed. After the discovery of place value, either such a board With columns or a sign for zero is necessary Whether the sign for zero appeared m India simultaneously with the discovery of place value 1s uncertain.

Granted, for the sake of argument, the use of a board ruled into columns,? it is not necessary to conclude that numerical symbols were employed. Calculations might have been made by placing a certain number of shells or counters in the columns But the regular name for the numerical symbols as a whole is anka, "mark." 3 This word is

1 Cf Fleet, JRAS, 1911, p 519. 2 Cf. Rodet, JA {1880), 11,468, Bayley, JRAS, 1883, p 29 1 Buhler, Induche Pa/,aeographie, p 78.

280 Walter Eugene Clark

used by Varahamihira at the beginning of the sixth century (Brkat Samhita, xviii, SS), and often in the Siiryasiddhanta, as one of the nu­merical words meaning nine. By the fifth or sixth century, therefore, calculations were made by means of nine numerical symbols (not with counters), probably without a sign for zero. The fact that anka was used as a numerical word for nme and not for ten seems to point to the conclusion that at first there were only nme symbols, and that the symbol for zero developed later 1 It is, of course, possible that the dot was already used to denote zero, but was not regarded as being an anka, since it represented nothing and was merely jotted down m order to prevent the fact being lost sight of that one or more of the orders of numbers was not represented at all

One of the oldest preserved works on mathematics and astronomy is that of Aryabhata The date 1s given defimtely m the text itself as AD 499. This work contams the Dasagitilca, which gives m ten very condensed stanzas all the numerical data of Aryabhata's system of astronomy, and the Aryiis(asata, which consists of 108 arya stanzas m three chapters, one on mathematics, and two on astronomy. In the Da§agitika, a peculiar method of notation is used 2 The consonants of the Sanskrit alphabet from k tom are given the numerical values of 1 to 25 The other consonants from y to h are given the values of 3 to 10 The mne vowels and diphthongs from a to au (no d1stmction 1s made between long and short vowels) are attached to thes0 consonants either to add zeros or to give them place value The stanza 1s as follows.

Vargaksarani varge '1,arge 'vargaksarany kat nmau yah khadvinavake svara nava varge 'varge navantyavarge va.

(" Beginmng with ka the varga letters are used in the varga places and the avar,ga letters are used m the avarga places, ria plus ma equals ya The nme vowels are used m the two nmes of places varga and avarga.") The last clause is left untranslated The words varga and avarga seem to refer to the Indian method of extractmg the square root 8 The number of which the square root 1s to be taken 1s divided off into groups of two digits each The varga or square places are the first, third, fifth, and so forth, countmg from the right. The avarga or non­square places are the second, fourth, sixth, and so forth. The words varga and avarga are used m this sense in the fourth stanza of the GarJ,i-

1 Cf Woepcke JA (1868), 1, 448 2 See Rodet, JA (1880), 11, 444, Fleet, JRAS, 1911, pp. 114, 121, Barth, Collected

Works, m, 182; Jacquet, after Whish, JA (1885}, 11, 118. 1 Cf Rodet, JA (1879}, 1, 409

Hindu-Arabic Numerals 231

tapcida. There is no reason for refusing to take them in the same sense here. The varga letters are those from k to m, which are always ar­ranged in five groups of five letters each The ai,arga letters are those from y to h, which are not arranged in groups The ref ore the vowel a used in varga and avarga places with varga and avarga letters refers the varga letters k to m to the first varga place, the umt place, multi­plies them by one The vowel a used with avarga letters y to h refers them to the first avarga place, the place of tens, mult1phes them by ten. In hke manner the vowel i refers the letters k to m to the second varga place, the place of hundreds, multiplies them by a hundred The vowel i used with avarga letters refers them to the second avarga place, the place of thousands, multiplies them by a thousand And so on with the other vowels up to the ninth varga and avarga places This makes it possible to express numbers up to one followed by eighteen zeros As a matter of fact the largest number expressed in this nota­tion by Aryabhata himself occupies only ten places. The last clause, which I have left untranslated, offers great difficulty It may give, as the commentator Paramesvara says, a way of expressmg numbers beyond the mneteenth place by means of an anusvara used with the vowels Fleet emends vii, to hau The words which I translate " in the two nmes of places " are translated by Rodet as " m the two nines of zeros " That is equivalent to saymg that each vowel adds two zeros to the numerical value of the consonant This, of course, will work from the vowel ion, but the vowel a does not add two zeros It adds no zero or one zero, dependmg on whethei 1t is used with varga or avarga letters. It seems to me, therefore, more likely that a board d1-v1ded mto columns is implied rather than a symbol for zero, as Rodet thmks.

This stanza occurs m the Da§agitilca, which, 1f the name 1s strictly accurate, ought to contam ten stanzas As a matter of fact, 1t contains thirteen. There is an mvocatJon to the Gods, this technical paribhasa stanza explainmg the terminology which is to be used m the Da§agitikti, ten stanzas givmg the numerical data on which Aryabhata's descrip­tive astronomy is based, and a colophon The first stanza contains the name Aryabhata The thirteenth stanza begms with the words "Having known these ten stanzas which describe the movements of the earth and planets m the celestial sphere " The paribhasii stanza is not counted. It 1s obv10usly from this thirteenth stanza that the name Dasagitika was derived I see nothmg susp1C1ous in the discrep­ancy between the name Da§agitikii and the number of stanzas found

Walter Eugene Clark

in most of the manuscripts. 1 Nothing is gained by eliminating the pa,ribhasii stanza as of later date and hence getting rid of the objection­able words varga, avarga, and kha. The explanation given above, or the slightly ddf erent ones of Rod et and Fleet ( which amount to exactly the same thing so far as the calculation is concerned), are the only ones which will make the numbers come out right, and could be deduced from the numbers involved in Aryabhata's astronomical elements even if this paribhii~ii stanza were not present. We can check Aryabhata's numbers by means of later works, especially that of Lalla, which ex­pressly state that they employ the numerical data of Aryabhata with slight modifications. These make use of the system of numerical words, so that there can be no mistake on our part in knowing approximately the numencal values of Aryabhata's combinations of consonants and vowels.

In the other sections of Aryabhata's work only a few numbers are given, and these are always expressed by the ordmary words which denote the numbers. Neither the peculiar notation descnbed above, nor the system of numerical words is used Kaye insists that Arya­bhata mvented his peculiar system of notation because no convenient system of numerical symbols was known to him. and because the system of numerical words had not yet been mvented. The matter is not as simple as that It is much more hkely that Aryabhata invented and used his peculiar system only for the practical purpose of giving in a very concise form a large mass of numerical data in verse All of this is crowded into ten brief stanzas The rest of the work 1s descrip­tive and contams very few and simple numbers. Numerical symbols could not be used m verse. The ordmary words for numbers, although usable for a few simple ones, are clumsy, and badly adapted to g1ving a compact mass of numencal data in verse. A system of consonants and vowels with numerical value allows much greater conciseness than the system of numerical words.2 For instance Aryabhata gives in one stanza the whole table of twenty-four smes which, as expressed in the Silrya.nddhanta by numerical words, occupies five stanzas.

No later authors follow Aryabhata's method, as would be expected if he for the first time had made it possible to express a large mass of numencal data in verse It seems to have been purely an individual

1 See Kem, Brhat, Samhitii, p 58 of preface Kaye (JASB, 1908, p 111) remarks that there are manuscripts which contam fifteen stanzas These doubtless correspond to the manuscripts described by Bhau DaJI (JRAS, 1865, p 897), who says that the two add1t1onal stanzas are not m the arya metre and are obviously a later addition.

2 JA (1880), 11, 440, 453, Barth, m, 18i

Hindu-Aral>ic Numerals 233

invention for a very particular purpose, and that purpose was not com­putation but description. Only Brahmagupta at the beginning of the seventh century actually quotes any of Aryabhata's combinations of consonants and vowels with numerical value. Brahmasphutaaiddhiinta i, l!t, xi, 5, and xi, 17 quotes Dasagitika, S, 1 and 4. In spite of its wonderful conciseness, this system could not become popular in litera­ture because 1t manufactured such barbarous and uneuphomc c-ombi­nabons of consonants. The use of it in the ten concise descriptive stanzas of the Dasagitikii is no proof at all that Aryabhata was ignorant of the system of numerical words or of numerical symbols with place value. It 1s quite possible that he used such symbols in his actual calculations It is pure assumption to conclude with certainty that his actual calculations must have been made by means of consonants and vowels with numerical value.

Kaye has great doubt as to the authenticity of the work as a whole, and in particular insists that the Ga'T),iiapiida at least is of much later date. 1 The problem of the two or three Aryabhatas is a difficult one. Suffice it to say that considerable portions of the work can be proved by quotations in Brahmagupta to be prior to the first part of the seventh century and to have been written by Aryabhata. As Kaye remarks there are no quotations from the Ga1J,ifapiida. This, it seems to me, is due to the fact that Brahmagupta quoted only such passages as he desired to criticize unfavorably. Either he had no criticism to make of the mathematical section of Aryabhata's work, or he did not take the trouble to criticize it because none of it was in contradiction with smrti. In practically every case where he combats Aryabhata, it is because the latter departs from smrti.

Later another system of numerical letters, called katapayiidi from a word in the stanza which describes 1t. came mto use, especially in southern India The letters k to fi and t ton have the values of 1 to 10, the letters p to m the values of 1 to 5, the letters y to l the values of 1 to 9. The vowels have no numerical values. The system employs place value This method allows the choice of much more euphonic combmations of lette1s, and skilled writers worked out words which had connected meanmgs, as m the case of the Semitic chronograms. The origin of the system is unknown. The earliest certain instance of it is in the colophon of a manuscript dated A D 1174. However, it is used in an astronomical work, the Mahasuldhiinta of a later Arya­bhata, composed between the seventh and eleventh centuries. It is also vouched for by the astronomical Jaimini Sutras (1, ~. 2) of un-

1 Btblwtheca Mathematica, x, 289, lndum Mathemati.cs, p 11

!!84 Walter Eugene Clark

known date. 1 The date of the origin of this system is too uncertain to allow of using it in connection with the problem under discussion. It is uncertain whether it is of independent ongm or is a modification of the system of Aryabhata.

If the secrecy of the Neo-Pythagoreans is appealed to as the reason why their knowledge of numerical symbols with place value did not leak out in Europe until a late date ( except for the doubtful passage of Boethius and for the supposed teaching of it in Persia), we may equally well insist on the same possibility in India. For instance, Varahamihira (Brhat Samhitii, xiv, 28) says "The teacher is to communicate these things only to a pupil of steadfast mind; and the pupil after having learned them is to make his astronomical contri­vances in such a way as to keep them secret from his own son even " Brahmagupta (Spas(adhikiira, p 45) after an elaborate explanation of the method of calculatmg the true places of the planets remarks. "This is not to be given, even under oath, to one who 1s not a son since 1t will destroy the good karma of the one who so gives 1t. and smce military expeditions, marriages, and horoscopes depend on the true positions of the planets." Many such quotations could be given In the early period in India much secrecy was drawn over astronomical learning The new numerical symbols may have been used by groups of mathe­maticians and astronomers for a long time before they came into gen­eral use and before they were employed in mscriptions W ntmg was known in India several centuries before it appears m inscriptions This fact alone is enough to make very dubious Kaye's method of de­termming the date of the mvenbon of the numerical symbols with place value solely from mscriptions Even m Europe after the time of Gerbert (circa AD 1000) numerals with place value did not come into common use, nor are they found on coms and mscr1ptions, for more than two centuries

In India from a very early period there was a preoccupation with large numbers and with arithmetical problems The enumeration of large numbers mountmg by powers of ten or a hundred was carried further in India than anywhere else in the ancient world To each place (power of ten or a hundred) a defimte name was given. The Y ajur Veda (probably as early as the eighth century B.C ) gives names for classes of numbers from one to a number which we should write as one with twelve zeros. The M ahabhiirata gives names for classes of

1 lndUJche Studum, vm, 160, JA (1885), 11, HB-125, JRAS, 19ll, p. 788, and 1912, p 459, Zeit f du Kunde dea Morgenlandea, 11, 425, Buhler, Pal,aeographie, §85b

Hindu-Arabic Numerals 285

numbers up to one with fifteen zeros. There are many early passages containing such enumerations. 1 The Buddhist Lab,ta JI istara, which even in its present form belongs to the early centuries of the Christian era, has a long chapter descr1bmg the contest at arms and scientific knowledge between the young Buddha and the other princes. The description of the contest in the knowledge of numbers and arithmeti­cal problems occupies seven pages. The names of the classes of num­bers up to ten to the ninth power are taken for granted as well known to everybody. Then names are given mounting by powers of a hun­dred for classes of numbers up to ten to the fifty-third power. A few still higher numbers are given 2

Early in the seventh century Brahmagupta remarks at the end of his chapter on mathematics. " These problems are given only for pastime. The wise man can mvent thousands of others, or he can, on the basis of the rules given above, solve the problems propounded by others As the Sun with its hght darkens the stars, so can the man who is skilled m these rules darken the fame of other mathematicians rn assemblies when he propounds algebraic problems or solves them " Such passages prove at an early date m India a great mterest m numbers hers and numerical problems. Such contests m mathematics among the learned carry us back in thought to similar contests m knowledge of the sacrifice and of philosophy which we find recorded m the Upani­shads in the sixth century B c

Such enumerations of numbers based on powers of ten lead easily to the discovery of place value, much more easily than the clumsy Roman method or the Greek system, with the myriad as unit, could do. All that was necessary was to write down in numerical symbols the values of the different classes of numbers as they were given

Kaye (JRAS, 1910, pp 7~9-760 and more positively Scuntia, xxv, 13) remarks. " Bhaskara speaks with disdam of his Hmdu predeces­sors, but cites certam anonymous " ancient teachers " as authonhes. If these ancient teachers had been Hmdus, he would most probably have mentioned them by name and indicatmg thereby certain teachers who were not Hindus " To a Sanskntist who is acquamted with Indian habits of quotation, these statements are so utterly absurd that it is is not worth the time to discuss them.

It seems to me that the Indian literary evidence proves conclusively the presence of a symbol for zero by A n 600 Before this could be

1 JA (1863), 1, 251, Zeuachnft der Deutschen Morgenlandischen Geaellschaft, xv, 182-140, lnd'!Bche Studwn, vm, 324-825

2 Ed. Lefmann, p 147 Cf JA (1863), 1, 248

iS6 Walter Eugene Clark

referred to in a work of general literature it must have had considerable history behind it. It also proves the knowledge of nine symbols with place value ( with either a blank column on the reckoning board for zero, or a symbol for zero) by the end of the fifth century A.D. at least. Beyond that the present evidence does not go. But this carries the Indian knowledge of symbols with place value back at least four hun­dred years earlier than the date assigned by Kaye.

As further proof of Indian knowledge of digits with place value at a date earlier than that maintained by Kaye reference should have been made to the Yogabha~ya iii, 18 (Anandasrama edition, p. 130, and Woods' translation, p. 216) yathaikd relchd satasthiine satam dasasthiine dasaikd caikasthane, "Thus the same stroke is termed one in the unit­place and ten in the ten's place and a hundred in the hundred's place" and to Sankara's commentary on the Vedanta-Siitras ii, 2, 17 (Ni~a­yasagara edition of 1917 with Bhamati, Kalpataru, and Parimala, p. 521, and Thibaut's translation, vol 1, p. 397) yathd caikii]YI, sati rekhii sthiiniinyatvena nivisamanaikadaiasatasahasraduabdapatyaya­bhedam anubhavati, "So, again, one and the same stroke is, according to the place it is connected with, spoken of and conceived as meaning either [one, or] ten, or hundred, or thousand, &c."

The Yogabhasya may be as old as the sixth century A.D. (Winter­nitz, III, 461) and Sankara is at least as early as 800 An.

HARv ARD UNIVERSITY

'OBER DIE SPATERE ENTWICKLUNG DES INDISCHEN STAATSRECHTS

BY JULIUS JOLLY

DIE epochemachende Entdeckung des Kautiliya oder Kautaliya Arthasastra hat zu weiteren Nachforschungen uber die Quellen

des ai. Staatsrechts veranlasst, so ist es dem unennudlichen Jayaswal in Patna, einem treffiichen Kenner der a1 Politik und Rechtswissen­schaft, gelungen drei Hss. des Rajanitiratnakara ausfindig zu machen und auf Grund desselben dieses bisher nur auszugsweise bekannte Werk im Auftrag der Bjhar and Orissa Research Society kritisch herauszuge­ben.1 Gedruckt war bis dahin von dem gelehrten Mmister Cai;i<J.esvara, der dieses Lehrbuch der Politik verfasst hat, nur das 1uristische Lehr­buch Vivadaratnakara, einer der sieben Te1le seiner grossen Enzyklo­padie des Dharma, in der Bibliotheca lndica. Die Enzyklopadie wird in dem Rajanitiratnakara (p. 9) unter dem Namen eines Dharmasastra­nibandha zitiert, muss also alter sem als der Ra1anitiratnakara, der im Auftrag eines Konigs Bhavesa von M,ithila geschrieben ist, welcher nach Jayaswal nach 1370 n. Chr auf den 'thron kam, als Ca1;u;lesvara schon ungefahr 85 Jahre alt war, was durch andere Beispiele von Lang­lebigkeit in <lessen Familie plausibel gemacht wird. Jedenfalls hatte Can<J.esvara, nach seiner eigenen Angabe in der Enzyklopadie, schon 1314 sein Gewicht in Gold an Brahmanen verschenkt, sein G6nner war damals der Konig Hansimhadeva aus der Kan;iataka-Dynastie, der um 1804 zur Herrschaft gelangte und um 1310 Cai;idesvara auf den Mmisterposten berief, den schon vorher <lessen Vater Viresvara bekleidet hatte. Es scheint also dass Cai;i<J.esvara erst als ein im Konigsdienst ergrauter l\finister das Bedurfnis empfand, seine politi­S'chen Erfahrungen in einem Lehrbuch fur Fursten niederzulegen.

Dieses Werk handelt m 16 Abschmtten (taranga) uber den Konig, die Minister, den l\iroh1ta, den Oberrichter, die Richterkollegien, die konigliche Burg, die Beratschlagung, den Schatz, das Heer, den Heer­fuhrer, den Gesandten, die Pfhchten des Komgs, die Strafgewalt, die Ubergabe der Herrschaft an den altesten Sohn des Konigs, die Uber­gabe der Herrschaft an den Altesten durch den Purohita, die Weihe des neuen Konigs. Originell und interessant ist die Unterscheidung

1 The Ra3anUi-Ratnakara, by Chandesvara, ed by K P. Jayaswal, M. A. Cal­cutta, 1924 28 8. 878., 8°.

288 Julius Jolly

von drei Arten des Konigtums: Oberkonig (samrcit oder cakravartin). tributpfilchtiger (,aakara) und tributfreier (akara) Konig, wobei mit dem sakara dem Herausgeber zufolge Ca1;).(;lesvaTas eigener P,atron, der Konig Bhavesa gemeint ist, der von dem muhammedanischen Kaiser von Delhi als seinem Lehe'nsherrn abhangig war. Auf die Kaiser von Delhi ware nach Jayaswal auch die Stelle uber die nur <lurch die Tapferkeit ihres Arms zur Herrschaft gelangten (kevalasaur­ycidyciptaraJyasya), der indischen Konigsweihe entbehrenden Fursten zu beziehen

Die Quellen unseres Werks kann man aus den darin vorkommenden Zitaten erschliessen, unter denen die 38 Zitate aus Manu an erster Stelle stehen und M. als die Hauptquelle erschemen lassen. Von anderen Smrtis wird Yajfiavalkya 19mal zitiert, Narada 16mal, der Herausgeber hat diese Narada-Zitate grdsstenteils in meiner Ausgabe der Naradasmriti nachgewiesen und bemerkt uber den Rest mit Recht, class diese Zitate 1hrem Inhalt zufolge nicht aus einem Gesetz­buch, sondern nur aus einem ebenfalls dem Narada zugeschriebenen politischen Werk stammen konnen, wie es von 2 solchcn Stellen auch ausdrucklich heisst Rajanitau N aradah, Nitau N aradah Da die gedruckte Naraoasmrb ihrer Einleitung zufolge nur uber e1gentliches Recht handelt, so kann sie <lurch ein Buch uber Rajaniti erganzt wor­den sein, das inhaltlich etwa dem 7. Buch bei Manu entsprach, auf das emes d1eser Zitate auch besonders hinweist raJanarn m•weserJ,a nijagada Manuh pura, vgl. M. 7, I. Ubrigens enthalt auch der unten zu erwahnende Kommentar zu Somadeva viele solche Narada .. z1tate uber Niti. Im Ratniik81ra erschemen ferner Katyiiyana~ Vasi~tha, Vi~nu, Vyasa, Harita u. a. Smrti-Verfasser, das Mahabharata 14 mal, Ramaya:r;ta 2 mal, wemge Pura:r;i.as, von juristischen Kommentatoren und ihren Werken Laksmidhara und sem Kalpataru, Kamadhenu, Kullukabhatta, Gopala, Mitak~a.ra, Srikara u. a Kulluka muss hier­nach fruher gesetzt werden aJs ich fruher annahm (13 oder 14 statt 15. Jh ). Viel geringer als die Entlehnungen aus dem Dharmasiistra sind diejemgen aus dem Arthasiistra, wofur ausser den schon genannten politischen Zitaten aus Narada fast nur diejenigen aus dem Kamanda­kiya Nitisara in Betracht kommen, der als Kamrundaka, Arthasastra, Niti und RaJanib z1bert wird, im ganzen 15mal, ferner die 3 Z1tate aus einer Sukramiti, die aber in der zuerst von Oppert herausgegebenen Nib dieses Namens nicht vorkommen, em neuer Beweis fur die Unecht­heit dieses spaten M.achwerks.

Aus dem Uberwiegen des Dhannasastra uber das Arthasastra in unserem Werk hat Jayaswal geschlossen, dass schon im Zeitalter

U eber die Spatere Entwicklung des I ndiachen Staatarechta !lS9

La]q1midharas, dessen Kalpataru CaI?,«;f.esvara stark benutzt hat, eine neue po]itische Literatur aufkam, die nicht mehr auf den alten Artha­sa.stras eines Usanas, Brhaspati und Kautilya, sondem auf den Lehren des Dhannasastra fusste und daher auch die alten Bezeichnungen des StaatS'rechts als Arthasastra und Da1p;laniti aufgab und den neuen Titel RaJaniti einfuhrte. Die Verfasser dieser Werke waren Juristen aus der Dha.rmasiis.tra-Schule und schr1eben Lehrbucher des Dharma. In seiner Zugehorigke1t zu d1eser jungeren politischen Richtung sieht Jayaswal die Hauptbedeutung des von ihm veroffentlichten RaJaniti­ratniikara.

Der Ubergang von arthasiistra zu raJaniti liegt allerdings schon im Kamandakiya Nibsiira vor, wo der lnhalt d1eses Werks als riiJavidya bezeichnet und schon dem Vorganger Kamandakas, dem Vi~J)ugupta ode.r Kautilya, das Verdienst be1gelegt wird, das Ambrosia des Niti­siistra aus dem Ozean des Arthasiistra herausgeholt zu haben (I, 6-8). Betr des riiJa in riiJaniti ist auch an das alte raJadharma "Konigs­pfhchten" und an riiJasas-tra (riiJavidyii) zu ermnern, da:s durch die Uberemsbmmung von Mhbh XTI, 58 m1t Abvagho~a I, 46 und Jata­kamiilii IX, 10 als alt erwiesen wird 1 In dem alten Kalpataru (12. Jh) lautet der T1tel dt>s politischen Teils rii(Jadharrmakanda, w1e aus der lnhaltsangabe bei Eggelmg I O III, 410 zu entnehmen ist Von Anfuhrungen enthalt ubrigens dieser riiJadharmakiinda nur solche aus den Smrtis und Puriinas, b1ldet den 11. Abschmtt emer 12 teiligen umfassenden Enzyklopad1e de's Dharma und entspricht also der Charakteris1erung,die Jayaswal von den spateren Systemen der Politik gibt Es soil nun noch an einigen anderen Werken d1eser Art gepruft werden, oh Jayaswals Annahme allgemem begrundet 1st

Betrachten wir zunachst den Nitimayukha oder RaJanibmayukha des Nilakantha, der zwar schon 1880 in Benares gedruckt, aber b1sher noch nicht naher untersucht ist, Diese ausfuhrhche Darstellung der Pohtik bildet das 5 Buch m dem um 1640 unter den Auspizien des Komgs Bhagavantadeva entstandenen Bhagavantahhiiskara, der eine grosse Hteili•ge Enzyklopad1e des Dharma 1st, w1e der Kalpataru. Aber der Nitimayukha schopft viel mehr aus der pohtischen Literatur als der RaJanitiratnakara und der Kalpataru. Grosser Wert wird hier allerdmgs auf die religiose Konigsweihe (abhiseka) gelegt, deren Be­schreibung die ganze e.rste Halfte des Buchs fullt Dann folgen eine Menge kurzer Kapitel, uber die 7 Elemente des Staats, die 18 Laster eineis Konigs, die taglichen Pflichten desselben, seine Diener, seine sechsfache Politik, seinen Harem, Prinzenerziehung, den Schatz, die

1 Vgl Wmtermtz, Geach d ind bit, III, 506 ff

!MO Julius Jolly

Burg und das Heer, Elefanten und :Nerde, Gesandte, Spione, Krieg­fiihrung, Spiel u. a. Hier werden nun sehr viele Stellen ans Kamanda.ka oder Nitisara angefiihrt, ich konnte uber 50 solche Zitate zahlen, so beruht die Darstellung der sechsfachen Politi.k wesentlich auf dem Nitisara. Selbst Ca~akya und das Kautiliya waren de,m Verfasser noch bekannt, wie das Zitat auf p 52 zeigt. sthalamrgayam aha CaTJ,alcyah supariksitarafcsitam tu simno laghuyiinas tu mrgiit,avim upeyiid iti, was ungefahr K. 1, 21, 43 entspricht, nur steht dort statt mrgat,avim das synonyme mrg<i:rwnyam. Auch Varahamihiras Yogayatra i.st reichlich benutzt, ein Zitat aus dem Mana,sollasa, einer alten Enzyklopadie fur Fursten, findet $ich auf p 58. Freilich werden auch die Smrt_is, Ma.nu an der Spitze, und die PuraJJ.as sowie Mhbh hiiufig genug zitiert, von Dichtem Magha und Kalidasa, aber den Grundstock des zweiten Hauptteilsbilden die Zitate aus dem Nitisara.

Ein ganz selbstandiges Werk uber die P:flichten eines Konigs ist die noch ungedruckte RaJabhii!;!al).i von Ramanathadik~ita, 1 die nach ihren Zitaten aus Kalpataru, Kamadhenu, Kulliika, Dayatattva und Divyatattva (von Raghunandana), Madhava, VivadacintamaQ.i, Viva­daratnakara u. a. Dharmanibandhas und Kommentaren zu schliessen kaum vor 1600 entstanden sein kann, also wohl ungefahr in die gleiche Epoche wie Mayiikha und RaJanitiprakasa gehort. Dieses Werk handelt im A:nschluss an die Smrtis uber die Gottlichkeit des Konigs, seine Straf gewalt, seine Rate, seine Schreiber und seinen Hauspriester, die Ehrung der Brahmanen, den kbniglichen Gesandten, die Burg, die Aufgaben und die Laster des Konigs, Karnpfregeln, Auszug in den Krieg, tagliche Pflichten des Konigs, V:erwaltung und Rechtsprechung, gefundene Schatze, Zeugenverfahren, Eide, Strafen fur Meineid, Mass und Gewicht, Schuldrecht, Hinterlegungen, Opferpriegt,er, Opferlbhne und andere Lohne, lnJurien, Sachbeschadigung, Diebstahl, Ehebruch, Spiel und Wetten, Grenzstreitigkeiten. Dann folgt ein kurzer Hinweis auf die im Arthaisastra (arthasiistre) behandelten Gegenstande: Land­mes91.Ulg, Kochkunst (supakarana), Prufung und Heilung der Elefanten und Pferde, auf die 18 Konigswissenschaften (riijavidyii), besonders Niti-, Dhanur- und Arthasastra, auf die 1m Mb.bh. genannten Autori­taten u. s w. Das Schlusskapite] handelt uber Konigsweihe (rajabhi­•?eka) und endigt m1t einer Reihe von Mantras aus dem Agnipura~a. Von son8t1gen PuraJJ.aS zitiert die Rajabhii~aQ.i das Skandapurii~a, Garu,;lapurana, u. a., die-Hauptquelle bilden aber die Smrtis, beson­ders das 7. B:uch des Manu und der Rajadhanna des Mb.bh. Das

1 Mllnchener Sanskr1t-Handschr1ft Nr 822, RaJ Mitra Nr 1207.

U eber die Spdtere Entwicklung des I ndischen Staatsrechts 241

Arthasastra ist in dieser Darstellung der Rajaniti nur durch die erwahn­ten kurzen Hinweise vertreten.

Hier verdient auch der angeblich von Konig Bhoja (11, Jh.) ver­fasste, 1917 in Calcutta gedruckte, ganz versi:fizierte Yuktikalpataru Erwahnung, der allerdings nur in seinem ersten .A:.bschnitt uber Politik handelt (iti samlcsepatah proktii riijanitih, p 17), weiterhin uber Bau­kunst, Hausgerate, Edelsteine, Schmuck, Waffen, Haustiere, Fahrzeuge, Schiffe, Schiffsbaukunst u. a. zum Arthasastra gehorige Gegenstande. Man kann demnach dieses Werk als ein Arthasastra 1 bezeichnen, ob­schon das Wort artha§iist11a nicht darin vorkommt. Als massgebend fur Nlti werden im ersten Abschnitt die Nitis von Brhaspati und Usanas bezeichnet, weiterhin allgemein die Nitisa.stras zitiert. Tat­sachlich finden sich hier ma,,nche A.nklange sowohl an das Dharma­sastra als an das Arthasastra. So ist p. 11, 72-74 = M. 7, 63, 64. 66, p. 15, 105 = M. 7, 20, p. 16, 113, 115 = M 7, 147. 149, p. 17, 118 = M. 7, 74. Anderseits 1st p. 8, 52 = Nitisara 16, 37, p 9, 63 = Nitisara 13, 26, p. 10, 71 = Nitisara 13, 33, p U, 81 f. = Nitisara 9, 1, p. 12, 84 = Nitisara 9, 28, p 13, 90 f. = Nit1sara 11, 23 f., p. 14, 93 f. = Nitisara 30, 29 f , p 5, 34 = K. A !?, 9, 23, p. 11, 75 (die drei Arten von Gesandten ) = K. A. 1, 6, 2-4. In den we1teren Abschnitten finden sich zahlreiche Zitate aus dem GarU<;la-Pura:i;ia und anderen Pura:i;ias, ausserdem aus Lauhapradipa und Lauhar:i;iava uber Metalle, Palakapya uber Elefantenkunde u. a Dieses Werk ist wichtig fur Kulturgeschichte, besonders fur Geschichte des Sphiffbaus, und zeugt fur das Fortleben des Arthasastra.

Allerdmgs konnte sich die unmoralische Staatskunst des bekannt­esten Arthasastra, des Kautiliya, nicht behaupten mit ihre'n Ratschliigen uber ra:ffinierten Steuerdruck, w1llkurliche Einziehung grosserer Ver­mogen, Tempelraub, staatliche Konzessionierimg der Trmkbuden, gerichtliche Tortur, Ablosung der Korperstralen <lurch Entrichtung von Geldstrafen, Erleichterung der Ehesche1dungen, schadhchen Zau­ber, Betrug und Hmterlist jeder Art. D1ese Lehren mussten schon bei der Umarbeitung des K. A in den popularen verstfizierten Grundriss Kamandakiya Nitisara emer einwandfreieren, wenn auch kemeswegs tadellosen Moral Platz machen. Auch in der Marchen- und Fabellitera­tur wie in den Kommentaren zu Manu wurde das K. A. als politisches Lehrbuch <lurch den Nitisara ersetzt, der besonders im H1topadesa ausgiebig z1tiert w1rd. Hatte doch bekanntlich schon B~:i;ia in der Kadambari (p. 109) das "Kautilyasastram" als ein ruchloses, von

1 Vgl Wmtermtz, Geach. der ind. Lilt., III, 53i.

Julius Jolly

grausamen Lehren stro'tzendes Werk (anrsa1[1,saprdyopadeaanirghroam) gebrandmarkt.

Eine ahnhche Entwicklung wie in der brahmanistischen zeigt sich auch in der Jaina-Literatur, wenn man Somadevas Nitivakyamrtam (10 Jh.) mit Hemacandras Laghu-Arhanniti (12. Jh) vergleicht. Das erstere Werk ist noch stark von dem K A. abhangig 1), wenn es auch kaum mit Ghoshal 2 als ein schwacher Abklatsch (a poor copy) des letz­teren Werks bezeichnet werden kann Die neue Ausgabe des Niti­vakyamrtam von N. R. Premi (Bombay, 1923) enthiilt emen alten Kommentar, der voll von interessanten Zitaten aus b1sher unbekannten Nitiwerken 1st, die Somadeva neben dem K. A. fur sein Lehrbuch benutzt hat, das Jayaswal 3 als em Gemisch von Eth1k und Pohtik, Wmternitz 4 als ein padagogisches Werk fur Konige bezeichnet, und das namentlich in dem letzten Abschmtt uber Vennischtes (379---405) die verschiedensten Klugheitsregeln enthalt. Die Laghu-Arhanniti 6

dagegen charakterisiert Ghoshal m1t Recht als ein Werk nach Art der brahmanistischen Smrtis, nur class darin die Ra1aniti auf den jam­ist1schen sagenhaften Komg R~abha zuruckgefuhrt w1rd. Der An­schluss an Manu u a Smrbs tritt besonders in der Lehre von den Prozesscn und den rehgiosen Bussen hervor 6

So hat Jayaswal die Tendenz der spateren ai. Politik im ganzen rich­tig charakteris1ert und hat sich die trad1tionelle Ethik als starker bewahrt als die blosse Nutzlichkeitsmoral des Arthasastra, gemass dem alten Grundsatz (Y 2, 21; Nar. I, I, 39), dass das Dharmasastra dem Arthasastra uberlegen ist und m Zweifelsfallen die R1chtschnur fur das emzuschlagende Verfahren abgeben soil.

1 Wmtermtz, a a O , III, 527 2 A Ilistory of Hindu Pohtical Theories, Calcutta, 1923, p 243 3 Jour B1h Or Society, XI, 66 (1924) 4 L C' 528 6 Ahmedabad, 1916 6 Wmtermtz, I < , 531

UNIVERSITY oF WuRZBURG.

THE BIRTH OF LORIK

BY Sm GEORGE A GRIERSON

IN the Indian Province of Bihar, and in the United Provinces of Agra and Audh, the Gowalas (Gopiilakas) or Ahirs (Abhiras) are well

known as an important pastoral tribe. Their caste-profession is cattle­keeping and selling milk and its products, and, though the milk they sell is not always free from suspicion,-witness many proverbs,-they are, as a body, looked upon with some consideration. There is a famous tribal legend concerning an Ahir named Lorik, which is most popular among them, and the folk-epic describmg his birth and adventures is sung at all their f estiv1ties. A Bihar proverb runs.

kltno Ahira hohi inyana Lonk chari na gawahi and

However learned an Ahir he, Nothmg but Lor1k smgetb he.

The cold weather of the years 1888 and 1889 is marked m my mem­ory by two incidents We were then stationed m the Gaya district of Bihar, and were made happy by a visit from Professor and Mrs Lan­man on their bridal tour through India Later on, m the same cold weather, while on my own official tour, I found myself m camp at the traditional scene of the birth and early adventures of this Lorik, and succeeded m gettmg copies of two recens10ns of the whole huge poem, taken down from the mouths of two reciters of repute It 1s pleasant to think that these two mcidents, occurring almost together, so long ago, find themselves agam associated with all their memones in these pages.

A not very complete, and not always accurate, abstract of the story of the cycle grouped together under the name of the Git Lorik is given by Beglar on pages 79 ff. of volume VIII of the Archaeological Survey of India This has been corrected and further condensed by Crooke on pages 55 ff of volume I of The Tribes and Castes of the North-Western Provinces and Oudh; so that my readers can mercifully be spared from its repetit10n On the present occasion I therefore content myself with giving the translation of the two versions of the first canto, describing the circumstances of the hero's birth. This will be found interesting by students of Indian religion. It 1s commonly asserted that among the Hindus of northern India, the worship of the V ed1c god Indra disap-

244 Sir George Abraham Grierson

peared many centuries ago, and was supplanted by that of Vi,y;,.u of Siva. Apparently this was entirely true only for those who lived in cities, for those who lived a learned or an ascetic life, or for those people who have been preserved to us in literature. Only with difficulty did the worship of Vi~nu reach the pastoral and agricultural classes, that is, the great bulk of the community, which was largely Munc;la or Dravid­ian in origin. I have dealt with this question at some length in a paper in the "Zeitschrif t fur lndologie und lranisbk" 1 and I now give one example that is not there mentioned, but that has direct connection with Ahirs.

According to the Bhagavata Purana, X, xxiv, the cowherds in Krsna's time worshipped Indra. When Knn;ia asks Nanda why he does this, Nanda replies (11 8-11), "Parjanya, the Rain-cloud, is the Lord Indra Himself The clouds are his dearly-loved forms. These pour down water for the dehght of all beings and for keeping them alive. Dear one, we and other m('n sacrifice to this Ruler of the Rain-clouds those things which spring into existence through his rain, and with what remains over after that, we men support ourselves." etc Here we see clearly that it was as a Rain-god that Indra was the chief object of worship among pastoral "and other" tribes. The rest of the story is well known. Knn;ia persuades them to abandon the worship of Indra, and, instead, to worship him (that is, V1snu) under the form of Mount Govardhana. There is a contest between lndra and Krsna, m which the latter is easily victorious. We have here a Brahmanical account of an attempt to spread the new, fashionable Vii;mu-worship among the lower orders, which, accordmg to the writer, was successful. If there was such a success, it was only temporary, for Indra, the Rain-god, is still a divinity of the Ahirs and other pastoral and agricultural people. In my paper in the Z I I. above ref erred to, I have shown in detail the state of affairs m modern times. The cowherd peasant of Bihar has, I need hardly say, his local godlings, and those to whom it is useful to appeal on special occasions. He also knows about Viigm and Siva, but they are misty characters far beyond his mental horizon. In practice his chief great deities are Indra and Durga, the latter acting as lndra's vice-regent upon earth The relationship of these two deities to each other, and to the other gods of the Hindus, is indefinite, varying according to tribe or locality. To the Ahirs who sing the song of Lorik, lndra, though sometimes addressed as a single person (for example, verse 267), generally appears as a group of seven brothers, inhabiting with their wives a heaven named Indrisana, and situated (verse 270) on

1 II, 188 ff.

The Birth of Lorik 245

Mount Kailasa. Durga is their sister, and at some unspecified time the eight had divided their landed property, - exactly as we now see every day in Bihar, - the world of mortals falling to her share, while her seven brothers retained IndrAsana. But here relations begin to be com­plicated. Durga is not only the sister of the seven Indras, but is also the sister of Mahadeva. There is no suggestion that she is his sakti, or his wife. On the contrary, she (126) addresses him as brother, and he (ISO) addresses her as sister, terms which no Hindu married couple could possibly use to each other. As in classical legend, Mahadeva is shown as performing arduous austerities, but, unlike the destroyer of Anai:tga, he displays no resentment when Durga interrupts him, or when she unmasks a trick that he has played upon her, by hiding Kr~:r:ia under his armpit

This brmgs us to Kr~:r:ia, whose name naturally turns up in an Ahir legend. He is not the Krsi;ia of the M ahabhiirata, the Yadava king of Dvaraka, but the youthful Krs:r:ia whose pranks among the cowherds of the VraJa-ma:r:i<)ala are a favorite subJect of Hindu story In our poem he is little more than a favorite servant of the lndras, for whom he acts as masseur. When Durga wishes to take him down with her to the world of mortals, they are most unwillrng to part with him, but finally, un1er pressure, give their consent.

The account given of the death of Kamsa differs materially from that With which students of the Bh<'igavata Purtina are familiar It is true that Kr~:r:ia on his birth is hidden away from him among the cow­herds, but thereafter the story skips straight away to the well-known episode of Kr~na' s theft of the garments of the herdmaidens 1 These girls complain about his conduct to Kamsa (337) Kr~:r:ia, when Kamsa remonstrates with him, retorts by threatening to cut off his head, and Kamsa agrees to be decapitated, provided Kr~na could perform two tasks -to twist a rope of ashes, and to weave a cloth of smoke Kr!jna performs both (the poem does not say how), decapitates Kamsa, and takes his kingdom.

Durga, having been allotted the world of mortals, descends from lndr!sana to take possession of her property. The mortals take her for a demon, and refuse to worship her, so that she has to spend the night sitting on a dunghill. According to the second recension, there were three persons in the world, whose wickedness she could not endure. These three, in subsequent cantos, appear as the chief villains of the

1 It may be noted that the herdma1dens are called by the general term Radhikaa We know that, m Sanskrit literature, the name of RadhA 1s later than the Rhagaf!ata Purdna

246 Sir George Abraham Grierson

story. Whatever the cause, she returns in indignation to the IndrA.sana and demands 1 that the Indras should give her Kr~:r;ia as her personal servant. After many subterfuges on their part, and many impish tricks on the part of Kr~J.la, she gets him, and arranges for his birth, in an avatara, - or, rather, in an avatara of an avatara, - in the person of Lorik, and so the first canto ends. The poem is called a Git, or song, because, in recitation, it is sung or, rather chanted; but, although uttered in short sentences, it cannot be said to be in verse. Like other folk epics of northern India, it is composed in brief clauses, the length and general swing of each of which are governed by the convenience of the singer's breath and by the rhythm of the chant. Whatever these clauses are, they are certainly not verse, whether we measure by accent, by length of syllables, or by counting syllabic mstants

The language of the poem, of course, varies from place to place. As recorded for me it was naturally couched in the dialect of the Gaya district, that is to say, in the Magahi dialect of the Bihari language, which, with a few shght irregularities, is the same as that described in volume III of my Seven Grammars of the Diaf ects and Sub-dialects of the Bihari Language, and on pages SO ff of volume V, part 2, of the Linguistic Survey of India Attention may be drawn to the common employment in the text of mterJections, usually little more than pleo­nastic. Note, however, that ho is used m addressing a male, and ge when addressing a female Ki re daiba, literally "What, 0 Fate!" is a common eqmvalent to our "Lo and behold'" or "What do you think happened next!" A typical Magahi interJect10n 1s re- O! In other parts of India the use of this word is more or less msulting, and, except in abuse, its employment is prohibited by good manners But, even in ancient times it was noted as a word of customary conversation in Magadhi Prakrit, 2 and m the modern speech of the Magadha country, or Magah, it 1s very common, and, especially in interrogative sentences, is used without the slightest tmge of disrespect There are many sto­ries based on this peculiarity of the Magahi dialect. For instance, we have the verse -

Magah des hai Kancan-puri des bhala hai, bhdkha buri

rahal-a Maggah, kahali 're,' tekarii-la ka marabe, re

1 She threatens to curse the seven lndras with impotence if they will not comply This is a favor1te curse of Durgii's, and 1s the Leitmotiv of a subsequent canto

2 Compare the Prakrta-kalpataru, II, 11, 28 (MagadhI sect10n), sambodhane .. 'ali' ea 'le li.' There 1s a som,ewhat s1m1lar use of rem modern Bengali

The Birth of Lorik

Magah land i11 a golden ecene, The country ' 11 fine, but the speech 111 mean.

I hved there once, and hence my "ri" Why do you drub me, ri, Sir, pray?

247

The unfortunate speaker of this doggrel had once lived in Magadha, and had acquired the habit of using re in every sentence. In some other part of India he does this and gets a drubbing for the insult; yet, so in­grained is the habit that, even while apologizing, he utters the objec­tionable word.

THE BrnTH oF LoR1K

The venerable Indras are seven brothers. The eighth is a sister, Devi Durga. The venerable Indras are the rulers of Indrasana. They divided (their kingdom) between the brothers and the sister. (5) The Indras took the lndrasana as their kingdom. They gave the world of mortals to Devi Durga. Devi Durga came to (her) kmgdom, the world of mortals She stayed the whole night m (her) kmgdom, the world of mortals. When Devi came to (her) kmgdom, the world of mortals, (10) no man utters the name of Devi m the world of mortals All night long she tarried in a lane on a dunghill No man repeats the name of Durga. TI?.is is what the (men) of the world of mortals say, - "She who has come is a Hm;idi, a Rak~asi (15) 0 children, she will devour you all "

She passed the whole night amongst the mortals At dawn she arose and went to lndrasana Lo and behold, she enveloped her whole body in fire Durga enveloped her whole body m flames (20) As a pan (of coals) was she blazing, and her matted locks were dishevelled Devi came before the court of Indrasana She was going along in furious rage Lo and behold. her body 1s quivering in wrath Devi Durga stood by the door-frame. (25) The Queens of the Indras raise their eyes and see her. Then out speak the Queens of lndra "Hear, 0 hear, Lord lndras Lords, what (useless) Cutcherry are ye holdmg Your sister is standing at the door (30) Lords, she hath already cursed you with a curse. Lords, ye seven brothers will now be childless Lords, this is now the curse which she hath uttered O Lords, the whole of lndrasana will be burnt to ashes Devi will not leave us even a place for sitting "

(35) The venerable Indras are seated There also 1s seated their Guru, Mohan Bariya Bhagirath (that is, the Ganges) Four brothers who were Prime M1msters, Headmen, were also seated there The venerable Indras then spoke out "Hear, 0 hear, Sister DevI. (40) To thee, Sister, divided we out thy share To thee fell as thy share the kingdom of mortals. To us fell as our share lndrasana. Lo, we re-

248 Sir George AINaham Grierson

mained in IndrAsana. Thee did we send to the kingdom of the mortals. (45) What kind of trouble bath come to thee among the mortals? Why, 0 Devi, hast thou returned to the court of IndrAsana?" To this did Devl make reply. "Seven brothers, did ye take IndrAsana for your kingdom. To me did ye give the world of mortals and sent me thither. (50) All night long did I tarry m a lane on a dunghill. No man in the city utters the name of Durga This is what the mortal people say. 'She who has come is a Hu1_1Qi, a Rak~asi. She will devour you all.' (55) These are the words the mortal people say.

Now will I not return to the kingdom of mortals. Until ye give me Acyutananda (that is, Kr~z:ia) as a servant, till then will I not return to my kingdom of the world of mortals." Tp that do the lndras make reply. (60) "Hear, 0 hear, Sister Devi. Sixteen hundred servants are there in Indrban. 0 Sister, select thou a servant, and carry him off by his topknot." The sixteen hundred servants did the venerable lndras make to stand (m a row). They made them to stand close to Durga. (65) Upon them does Durga cast her eyes Now out speaks Devi Durga. "Hear, 0 hear, Brother Indras. Wherefore show ye me these servants? Until ye give me Acyutananda as a servant, (70) Brothers, till then will I not return to the kmgdom of the world of mortals " Then out speak the venerable lndras. "Hear, 0 hear, Sister Devi Where wilt thou get Acyutananda? Sister, that servant 1s dead or is lost" (75) To that gives Devi Durgii answer. "Hear, 0 hear, Brother Indras This speech of thine believe I not. Brothers, give ye so much an answer as this. Wherever ye shall conceal my Acyutananda, (80) there will I have power to gain that servant So much consent. Indras, give ye to me, or else, lndras, will I search for the servant, and pull him forth " To that gave the Indras answer "Wherever thou mayst find Acyutananda, (85). Sister, from thence search for him, and pull him forth" Such was the command that the venerable Indras gave

Round and round does Devi search through Indrasana, but no­where findeth she Acyutananda. From thence went forth Devi Durga, (90) and behold she came to Nandan's forest In the Nandan forest are there sixteen hundred nymphs (pari) Acyuta was making the sixteen hundred nymphs to dance in the N andan forest Every nymph that does not dance in the Nandan forest, her head does Acyutananda cut off and make into a t/,igri (a kmd of musical instrument) (95) He cuts up her body, and makes of it a drum. He cuts up her thighs, and makes of them a fiddle. He cuts up her arms and makes of them a fiddle-stick. Thus was he making the sixteen hundred nymphs to dance in the Nan­dan forest, when, lo, there came there Devi Durga.

The Birth of Unik 249

(100) Acyutananda looks towards Devi Durga. He fled off to the Baill flower garden ( or to the garden of Bel trees). Devl Durgi asks from children for his traces. "Hear, 0 hear, master children. The man who was now superintending a dance, (105) Sirs, whither has that man fled?" To that now give the children answer. "Hear, 0 hear, Devi Durga. The man who was superintending the dance, his glance fell upon thee, 0 Durga, (llO) and, lo, he fled to the Baili flower garden." Durga arnved at the BailI flower garden. Round and round does Devi search through the flower garden, but, lo, nowhere does she find the darling Acyutananda.

From thence did Devi advance m further search (115) Behold, she went to Mahadeva's flower garden Round and round does Devi search through Mahadeva's flower garden, but, lo, nowhere does she find the darling Acyutananda From thence did Devi advance in fur­ther search. Behold, Durga approached Mahadeva. (120) The vener­able Mahadeva was seated m an arduous worship (pujii) Behold, ( under the rules of this worship) there was no perm1ss1on for him to arise. Now the venerable Mahadeva 1s burning a maund and a quarter of ghi and bdellium. Now the venerable Mahadeva 1s seated with his face to the East. Behold, he 1s seated in the prescribed position (125) T4en out speaks Durga to Mahadeva "Hear, 0 hear, now, 0 Brother, Mahadeva In an arduous worship art thou seated Lo, Brother, where hast thou put Acyutananda?" To that gives Mahadeva a stately reply. (130) "Hear, 0 hear, 0 Sister Devi. Hither came not Acyuta­nanda Behold, he would be m the court of Indrasana" Then out speaks Devi Durga "From hence, Brother, will I not turn to go (135) Until thou raise up both thme arms, till then will I not move a step back" So many words did Durga speak out, and behold the venerable Mahadeva laughed aloud The venerable Mahiideva raised both his arms, (140) and, behold, Acyutananda now came forth (from below them). When Acyutananda came forth, Devi Durga runs up and seizes him by the topknot Then out speaks Devi Durga "Ah, how much trouble hast thou given me, Acyutananda (145) Lo, Acyuta come to the kingdom of the world of mortals" Then out speaks Acyutananda. "Hear, 0 hear, 0 Devi Durga I will not go, until I have (first) gone to lndrasana. Lo, Durga, till then will I not move a step to the world of mortals"

(150) From thence went forth Devi Durga and Acyutananda. Be­hold, they both went to the court of Indrasana. He went, and Acyuta­nanda stood there He went, and stood in lndrlisana. Behold, he stood close to lndrasana. (155) The lndras raise their eyes and see him.

uo Sir George Abraham Grierson

In their hearts do the venerable Indras grieve. "What tyranny hast thou done, 0 Devi ! that, lo, 0 Dev!, thou hast sought out and found the darling Acyutananda. Lo, sixteen hundred servants were there for thee in lndrAsana. (160) Lo, from them thou mightest have selected a ser­vant, and have carried him off by his topknot. But thou hast selected and taken our peculiar servant. Lo, DevI, into our soul cometh lamen­tation. Behold, thou hast made widows of all (the women of) Indr!sana. Ah, thou art taking away for good our jewel. (165) We lndras are seven brothers. To the seven was he the servant doing service. A servant who used to press (that is, could press) a seer and a quarter of mustard­seed in the palms of his hands (to extract the oil). He used to massage the limbs of us seven brothers. Great was the service that Acyutananda used to perform. (170) Ah Sister, the whole of IndrAsana hast thou made desolate."

Then (again) out speak the lndras. "Hear, 0 hear, Acyutananda Forasmuch as our sister Devi bath seized thee by the topknot, Lo, go thou, a servant, unto the kingdom of the world of mortals." (175) Now out speaks Acyutananda. Weeping, weeping, does Acyutananda speak. With suppliant clasped hands does now Acyutananda make his petition "Hear, 0 hear, ye venerable Indras. Prepare ye the book with back of copper, (180) and thereon make ye a writing, 0 lndras. Whatever I shall tell you to write That, 0 lndras, write ye with pen and with ink-case." So many words speaks Acyutananda Then out speak the venerable lndras. (185) "Hear, 0 hear, Acyutananda. What kind of words are those that we shall write?" To that Acyutananda gives answer "Hear, 0 hear, ye venerable lndras This many are the petitions that I ask from you (190) Write ye these letters m the book (of fate) Until ye give me the horse Katar, Kr~na's steed, Until ye write for me two wives (seated) on my thigh Sixteen hundred nymphs are there m lndrasana, - Sixteen hundred with whom Kr~i:ia (once) sported on (the banks of) the Yamuna (195) Of these O lndras, two died virgins Two damsels who were washed away by the Yamuna and died. These two damsels write ye (as seated) on my thigh, and, vener­able lndras, write ye something more Until ye write a brother at my back, (200) till then will I not move a step towards the world of mortals. Until ye give me the sword Maigar, that belonged to Bhimasena the K~atriya, - unless ye write the sword Maigar for my arm,-till then will I not move a step towards the world of mortals. And, 0 venerable lndras, write ye something more. (205) What woman, and what man, may have performed austerities, What woman, and what man, may have slept on a mat of kusa grass, - they who have so slept for twelve

The Birth of Lorik i51

years, - in her womb write ye my birth. Lo, then, Indras, will I set my foot towards the world of mortals."

(210) The venerable Indras are seated. And GaurI's Ga~apati is also seated. The five PaJ}.9avas are also seated. There also is seated there their Guru, Mohan Bariya Bhaglrath. Lo and behold, four brothers who were Prime Ministers, Headmen, were also seated there. (215) They produce the book of copper (back). Lo, the four brothers, the Headmen and the Prime Ministers write First they write the horse Kata,r, Kf!g1a's steed. Then they write the sword Maigar. They write the sword Maigar, that belonged to Bhimasena the K~atriya (220) Then they write the two women Then they write a brother to be at his back. To that last do the lndras give an answer. "Hear, 0 hear, Ac­yutananda. They write not a brother to be at thy back. (225) Obsti­nate fool art thou, thou dost not heed our words. Lo, Acyutananda, we have written for thee a foster brother A ready made foster brother will we send for thee Behold, Acyutananda, he will be a foster brother unto thee." Then write the venerable Indras, (230) "West of Gaura lies Kalaundi village There dwell aged Kuar and aged Khulhan. Husband and wife, have they performed for twelve years asceticism in the world of mortals. 'Childless' and 'Barren' have the names been given them. While they have been doing asceticism, twelve years have passed. (235) Still the names of 'Childless' and 'Barren' have not departed from them. Dear One, now the thirteenth year hath begun. Now the man and wife are starting to drown themselves m the Ganges. In her womb have I written thy birth "

Then agam did the Indras write these words. (240) Then they write a kmgdom m the four quarters of the world Then do they write a crooked word. Lo and behold, they write a six months prison at Ko1la.

The writing and the reading were finished The accountant closes the book (245) At that very moment, at that very hour, Lo and be­hold, the wife and her husband are going to the Ganges to drown them­selves While the wife and the husband were on their way, Durga, as she slept, gave her a dream "O aged Khulhan, drown not thyself. (250) My servant Acyutananda of the Indrasana, - I have already had the reading and the writing done, and am come, - Behold, 0 Khulhan, he will take incarnation in thy womb " At that very moment, at that very hour, lo and behold, Acyutananda took up his abode in her womb. (255) When the aged Khulhan had carried him in her womb for nine months, behold, in the tenth month Acyutananda took incarnation. When he took birth in the world of mortals, the Brahmai:is Wt"fe sum­moned, and they calculated his name. Accordmg to the constellations,

Sir George Abraham Grieraon

his name was Acyutananda, (i60) and, lo and behold, his pet name was Lorik Khanjar.

Second Veraion 'Ram, Ram,' I cry when I rise, and when I sit. Had I known the

name of Rama, I would have erected a high sacnficial platform to him. I would have brought Ganges-mud from the Ganges, and have plastered with it the platform. At the time of :fleeing, would I call on Mother Bhageswari. (265) When night falleth, I would invoke Mother Durga.

(End of the Invocation)

Then up and spoke Mother Durga "The venerable Indra hath sent me to the world of mortals. When I came thither, I sat on one side. White ants settled on my thighs" (270) Then in the world of mortals up rose Mother Durga, and went to Kabilas. 1 She dressed herself in patchwork and in rags Spreadmg out her hair so that it would fill a basket, she goes along with a castor-oil-tree stalk for a staff. She went, and Durga stood at the door of Indrasana. (275) The seven lndra­brothers trembled when they saw Durga Then did the lndras address their sister. "In what thing art thou lacking that thou art come to Kailasa from the world of mortals?" ·

Then up and spake Mother Durga, "0 Indras understand. Brothers, I have fled hither from the world of mortals (280) I cannot endure three sins (that are practised there) Such warriors have ye created in the world of mortals, King Harewa, King Parewa, and the warrior Jodhi Paura King Harewa is so base and vile Every captive that he prisoneth in his prison, (285) if he be old, each one dieth there, if he be young, there becometh he an old man King Harewa's minister is Maniir Jil. He also 1s base and vile. He giveth each prisoner boiling water for his drink, (290) and for his eating, but a quarter of a seer each of cakes of broken rice and bran. Even from that the minister taketh the half for himself. Hungry and without food, man draggeth man and eateth the flesh of man. Such a man of violence is the warrior Jodhi of Bikatpur, powerful, but vile and base. (295) Sixteen hundred maidens bath he kept unmarried. When a wedding procession cometh to fetch the bride, he killeth the boy-bndegroom, keepeth the bride, and sleepeth with her for a single night. Once he bath destroyed her caste he aban­doneth her. In Gaura there was an aged Gowalin. (300) Her name is

1 The Hmdl form Kafnlaa for Kaila,a 1s very old We find 1t 1n the Padum4vah of Malik Mu}.l.ammad ( 16th century).

The Birth of Lorik 258

Khulhan. Each morn, as she riseth, doth she bathe in the Ganges and offer incense and oblation. That woman is known by the name of 'Barren.' 0 lndras, give ye me my hero."

Then Mother Durga stood up in the IndrAsana. Then all the Indras said, "carry thou off thy servant " (805) Then Mother Durga caught hold of Isvara Mahadeva. Then away did flee Isvara Mahiideva. Then did Durgii catch hold of :Oak and Diinava, but all the :Oliks and Dana­vas fled away. No one agrees to go to the world of mortals (~HO) Then went Durgii to the bank of the river Y amuna. There, in the form of a boy, was the revered Kanhaiya (Kr~~a) playing Then did Durga catch hold of him by the ann. "Come with me, Child, to the world of mor­tals." "O Mother Durga, I will not go for thee to the world of mortals. (815) Seven births have been mme in the world of mortals. Mother Durga, my first birth was a birth in the house of a fish Then did the sailormen net me, and support themselves by sellrng my flesh. Then, Mother Durga, was my birth in the form of a boar, and the huntsmen speared and killed me. (320) Then, Mother Durga, was my birth from the belly of Devaki; and, Mother Durga, my maternal uncle was ut­terly vile. He it was, Mother, who fettered my father and my mother, and bowed them in the prison-house My birth took place at midnight in the month of Bhado Then all the lndras sent me golden dolls (825) Then the Indras filled the lymg-m brazier with cummin and amseed. Then the news went to King Kamsa, and Kmg Kamsa came to search for me. Then JUst as used to rise the moon of the full-moon day, so, gradually, grew I up to be a youth (330) With my bamboo flute I had gone to the bank of the Yamuna. Then I, Kanhaiya, played the flute. and summoned the sixteen hundred Radh1kas Then the sixteen hun­dred Radhikas took up their curd-pitchers Then all the Radhikas went to the bank of the Yamuna (They cried) 'He hath eaten up my curds and broken my pitcher (335) He bath grasped my arm~, and hath pulled me about He hath torn my garments that I wear ' Then all the Rad­hikas went to Kamsa and made complamt 'Hear, Kamsa, 0 hear my words. Kanhaiya bath taken birth m Devaki's belly. (340) To me bath he done a dozen tyrannies ' Then did Kmg Kamsa go unto the bank of the Yamuna. Then sa1th Kanhaiya, exhorting Kamsa 'Hear, Uncle Kamsa, 0 hear my word. Thou hadst done tyranny to my mother and my father. (345) Now will I cut off thy head, and take thy kingdom to my hand' Then saith Kam.sa, exhortmg Kanhaiya. 'Hear, 0 Kan­haiya, hear thou my speech. Thou wouldest kill me, Kanhaiya, and take my kingdom to thy hand. Do thou fulfil for me these two things. (850) Twist thou for me a rope of ashes. Weave thou for me a cloth of

Sir George Abraham Grierson

smoke. Then cut thou off for thyself my head.' Then did I, Kanhaiya, twist for him a rope of ashes. And a cloth did I weave for him of smoke. (855) Then did Kanhaiya cut off King Kamsa's head, and take for him­sell his whole kingdom to his hand."

Then Kanhaiya took to himself the form of a bee. Then the bee settled close by a Bower. Then Mother Durga caught him by the arm and seized him. (360) "Mother Durga, if thou art confining me and taking me away, give the command as to what things I shall eat as food. What things, Mother Durga, wilt thou give me to drink? What weapons, Mother Durga, wilt thou give me to (bear upon) my shoulder? To whom, Mother Durga, shall I be married?" (365) Then Mother Durga gives reply. "For food wilt thou get, my son, eighty still-fulls (of spirit) each day, and eighty goats as a rehsh to it daily. A sword of eighty maunds 1 in weight will I give thee for a weapon. A shield of seventy-two maunds weight will there be for thee. (370) A dagger will I give thee, my son, of eighty-four maunds. The horse Kata,r will I give to thee to ride. Thy first marriage will be with Dauna Mijar. Thy second will be an elopement with Chandain. Then wilt thou rule over the land of Bikatpur Agauri (375) Then King Kannga will seize thee and imprison thee for six months Then for six months will he keep thee in a potter's kiln. For six months, my son, wilt thou remain in the River Gadanii. Though struck by iron, thou shalt not die In the fur­nace of fire thou shalt not die (380) In the river Gadana thou shalt not die I write not for thee any kmd of death. First, my son, shalt thou kill King Karma, and then King Harewii." Then Kanha1ya departed thence. According to the constellations the god lndra wrote his name as Banawar (385) And, by love, his name was the heroic Lor1k Now went forth Durga to the ancient Khulhan, and Mother Durga gave her a flower to sme11 Then the ancient Khulhan became pregnant, and on the ninth month Lorik came into existence (890) In the twelfth year Lorik became fully grown. Then his friend RaJal, the washerman, and Lorik used to wrestle by the pond, and there Saira, the chaste, and BiJadhar, when the two saw Lorik, used to smile.

1 A standard maund 1s eqmvalent to eighty pounds

THREE INDO-ffiANIAN NOTES

Bv A V WILLIAMS JACKSON

1. A Suggestion to Connect Old Persian AG•R•[ J with a Sanskrit Word.

THE fragmentary Old Persian word AG•R•[ J in the Bahistan Rock Inscription of King Darius, Bh 1 ~l (8), has long been

a puzzle. The passage in which it stands reads as follows eatiy [Dara]avahus :rsayaOiyah a"tar imah dahyavah martiyah

hyah AG•R•[ J aha1 avam hubartam abaram, hyah arikah aha' avam hujra.'ttam aparsam.

'Sa1th Darius the King Withm these countries the man who was-?-, him I supported with good support; he who was inimical, him I punished (ht questioned, inquisitioned) with good punishment (lit with a good questioning, inquisi­tion) '

The antithetical balance in the case of the ad1ectival participles (hubaratam and hujra.<ttam) 1s obvious, and since the evil attribute arikah is now generally conceded to sigmfy 'antagonistic, mimical,' or the hke (cf Av ayra-), its antonym must mean the reverse of that A few observations, however, must first be made with regard to the reading itself and to the explanations proposed to explam the meaning. The more important of these are presented in their chrono­logical order

Rawlinson's transcript of the cuneiform text, published in 1847 (JRA8 10 p xl, plate, cf p 199), gave agatii, with t partly broken and ii supplied, this was rendered, with a question mark, 'of the true faith(?) ' 1 Although Rawlmson'1-, t was a mistaken decipherment and should be r, as Kmg and Thompson's reexammat10n of the rock in 1904 proved, it is mteresting as showmg that his restoration [ii] seems to indicate he thought that only a single letter was required at the end of this word. Kmg and Thompson, The Sculptures and

1 Benfey, Die pera Keuinachriften, 1847 (issued shortly after Rawlmson's work appeared), p 9 (cf 71), translated by 'truglos,' but this and his etymology may be disregarded Oppert, JA 1851, p 35, suggested agaritli, cf Skt dgantr, 'amvant, etranger ' Spiegel, Altpera Keilinschriften (2 ed 1881), p 20~, wrongly emended as da'U8tii, ' Freund '

256 A. V. Williama Jackson

Inscription of Behistan (1907), p. 5, give the correction of Rawlinson•s t tor and read agar[ . ... ], with a footnote that the lacuna had space for two characters; they translate by 'whoever was a [friend],' as they similarly do in the case of the Susian (Elamitic) version (p. 97), and they render pi-it-ku-du in the corresponding Babylonian version (p. i62) by 'zealous.' As regards the question as to whether one or two letters be required to fill the lacuna in the Old Persian text, we must keep in mind below the apparent difference between Raw­linson and the view of King and Thompson. From experience one knows that it is sometimes not easy to determine such a space with precision. Unfortunately when I made four ascents of the rock in 1908, I did not have time to examine this particular word. Scholars have followed King and Thompson in assuming that two letters are missing, as will next be observed, but commented upon below.

Weissbach, when reviewing that work in ZDMG. (1907), 61.725, suggested to derive 'a-ga-ra . . .' from the root gar 'wach sein + ci,' and kept to this view later in his Keilinschriften der Achameniden, p U, § 8, where he reads agar[ta] and translates by 'der umsichtig war.' Bartholomae, WZKM. (1908), 22 72, proposed to associate this attribute with Skt gurta-, Lat gratus (comparing YAv. agre­maitis) and to read· the vocable perhaps as agar[ta], nom sg. <?f a noun of agency. Tolman, Anc. Pers. Lex and Texts, 1908, pp. 60-61, s.v. agartar, also takes it as a noun of agency, but prefers (like Weissbach) to derive it from a presumable OP. a+ *gar, 'to wake' (YAv. gar, Skt. gr, cf TPhl vigaraned) as 'a watcher, wakeful, zealous,' and translates (p 8) by 'watchful.'

The data presented above allow us to conclude that the legible part of AG•R.a[] 1s practically certain and that the general connota­tion of the word is fairly clear, whatever may be its etymology. In what follows I am going to assume, with Rawlinson, that merely one letter is missing at its end; but only a renewed study of the rock itself, with careful measurements of the space itself, can settle the question.

At the risk, therefore, of having the proposal rejected, I venture to off er a new suggestion. This would be to read agra"[9a" ], not agar[ta], and to associate it with Skt agrantha-, from the lndic root grath-, granth-, 'to be, knot, fasten, attach.' This Sanskrit word is at least found as an adverbial gerund in am (see PWb. 2 881; cf. Whitney, Skt. Gr. § 995) in Aitareya BrahmaJ;Ia, 5. 15 tad yatha punar agrantham punar nigrantham antam badhniyat, 'just as one would bind the end, tying (or attaching) it with a knot again, (and)

Three Indo-Iran-ian Notes 257

again tying (or attaching) it down.' 2 At any rate it gives us a hint, and my notion would be to take the presumable OP. agra"[8a,.] as an adjective, with passive sense, conveying in general the idea of being bound by the knot or tie of allegiance, like the somewhat rare Eng. 'alligate,' from Lat. alligo. If so, agra"[8a,.J would mean 'attached, devoted, loyal,' as opposed to arikah, 'inimical, hostile, refractory.' Whether this suggested etymology has anything in it, or not, remains for the specialists to decide.

!l. Turf an Pahlavi dasin, a Sanskrit Loan-word in the Manichaean Fragments.

The word TPhl. dasin, meaning 'gift,' occurs several times in the Middle Persian Manichaean texts brought to hght by A. von Le Coq through his memorable discovenes in the Turfan Oasis, Central Asia, early in this century, and first deciphered and published in part by F. W K Muller. Five instances of diisin are thus found in Muller's Handschrijten-Reste, Teil II, Berlin, 1904. These occur respectively in Frag. M. 47 recto, lines 4 and 5 (the latter written defectively as dasn); M. 74 verso, ll. 14, 16, 18; see Muller, op cit, p. 84 (bis), 76-77 (tris). In his glossary, C. Salemann, Manichaeische Studien, I.. p. 65, St. Peters burg, 1908, accepted the meaning of dasin as 'Geschenk,' but had no etymological explanation to off er for this signification. W Bang, 'Manichaeische Hymnen,' in Museon (1915), 88' !ll, gives a Turkish form dasin (but without precise reference) as likewise meaning 'Geschenk,' and adds in a footnote, 'ich darf das tuerkische Form des Wortes den lranisten empfelen.'

The Iranist naturally thinks of Av dasina-, Phi. dasin, as a related form, but in that case the meaning is confined to 'dexter, the right side.' The key lies in the hand of the Sanskritist, who can at once produce both the Skt adj. diiksina-, 'relating to the sacrificial gift or to a gift in general,' and the neut. noun diik~rJ,am, as denoting such 'a gift or collection of gifts' (comparmg also the fem. noun dak~rJ,a-, 'gift, donation, remuneration'). Thus the Turfan Pahlavi term diisin, 'gift,' in Central Asian Manichaeism is plainly a loan­word, borrowed from the Sanskrit, and its etymology is clear

2 Keith, Rigveda BrahmaTJ,as the Aitareya, etc (1920), p 287 translates, 'twm­mg 1t agam and agam mtertwmmg 1t '

258 A. V. Williams Jackson

S. Can the Word aiim in a Turkish M-anichaean Hymn be of Indic Origin?

In a Turkish Manichaean Hymn, the text of which was first made available by Le Coq, Tilrkiache Manichaica aua Ohotacho, III, Berlin, 1~!!, p 47, Nr. 11, recto, and later translated with comments by W. Bang, Museon (19~5), 38. 41-4i, there occurs the word aam. It is used in connection with the blessings brought to the world through Mani's having preached the gospel of salvation in his 'Evangelium' (Turk. awanglion, cl. TPhl. 'evangeliyon), here exalted as the 'Jewel of the Law,' by following the teachings of which the w&y of release is found. The particular sentence (ll. 3-4) in this passage to which at­tention is here called runs as follows in Bang's transliteration and translation oz-yu qutrul-yu yol yifiaqi-y ol nomta asitip uqar [am] sam, 'den Weg der Rettung und Erlosung in 3ener Lehre (Predigt?) hbrend, versteht man das Heilmittel ( :>) '

In his translation of the sentence we observe that Bang has trans­lated sam by 'Heilmittel ( ?) ' with a query, adding in his comments (p. 47H), 'Etymologisch ist sam ganz unklar.' Perhaps a suggestion might be made to help clear up the etymology.

Although my knowledge of Turkish in the Manichaean documents is extremely limited, I would hazard the con3ecture that sam may be a word of Indic origin, findmg its way into Central Asian Mani­chaeism through Buddhism. If so, especially as it is plainly a word full of religious significance, I would suggest comparing it with the familiar Sanskrit word §ama-, Pali sama-, m the sense of 'tranquility, repose, cessation, final emancipation ' In that event the sentence as applied to the teachings in Mani's 'Gospel,' would become pregnant with meaning, as follows· 'On hearing the way of rescue and release one comprehends (perfect) tranquillity,' i.e., the spiritual calm which brings final emancipation We are familiar with Buddhist elements in Mani's syncretic religion.

COLUMBIA UNIVERSITY