STUDIES IN TH E TANTRAS

PART I

BY

PRABODH CHANDRA BAGCHI, M A Dr es L ettres (Par s

LEC T U R ER CALCUTTA UNIVERSITY

PUBLISHED BY THE

UNIVERSITY OF CALCUTTA 1939

,

STUDIES IN THE TANTRAS

STUDIES IN THE TANTRAS

PART I

BY

PRABODH CHANDRA BAGCHI, M.A., Dr. es Lettres (Paris)LECTURER, CALCUTTA UNIVERSITY

PUBLISHED BY THE

UNIVERSITY OF CALCUTTA 1939

PItIWTPpD It n!>lA

PtllNTED AND r U W ISH ED V iffiTORDit ALATj llAtiRlET

AT TH E CALCUTTA U K lV liJlfilT i' M tESB , SIT A T E Ut JUBE, CALCUTTA

Ie, Uo. 88GBOcfcob&r, 1939.E,

CONTENTS

Preface K ...

P age

viiOn some Tantrik texts studied in Ancient Kambuja 1Further notes on Tantrik texts studied in Ancient Kambuja .. 16The Sandhbh and Sandhvacana 27On the Sdhanaml ... 34On Foreign element in the Tantra . . 45Some Technical terms of the Tantras ... 61Some aspects of Buddhist mysticism in the Carypaflas ... 74Notes on the word Parvrtti ... 87

A ppe n d ix

Detailed Notioes on ManuscriptsI. Niivasa-tyttva-Samhiia' 1 t ... 93

II. Sammoha Ta-ntra ... ... 96III. Brahmay amala ... 102I Y. Pingalmala ... 105Y. Jayadrathaymdla ... 109

PREFACE

The present book is a collection of articles which were published m the Indian Historical Quarterly and the Calcutta Onenlal Journal between 1930 and 1934. They contain the results of my examination of manuscripts of the Nepai Darbar Library collection in 1929 1 have thoughtit fit to publish them togethor m the form of a book, as my future studies m the Tantras will follow the lines suggested therein. As many of the manuscripts referred to m the articles have little chance of being edited and published for a long time to come, I have added in an appendix detailed notices on them

CALCUTTA, I S e p te m b e r , 1939 J P. C. BACCHI.

ON SOME TANTRIK TEXTS STUDIED IN ANCIENT

KAMBUJA

I

The inscription of Sdoh hah Thom1 (discovered in the province of Sisophon, Cambodia) mentions the introduction of the mystic cult of Devavja along with some Tantnk texts in Kambuja during the reign of king Jayavarman II who came to the throne in the fiaka year 724 (= 802 A.D.). The inscription is a long one and contains the chronicle of the religious foundations of Kambuja during a period of about 250 years. The inscription is not dated, but the last date mentioned therein is saka 974 ( = 1052 A.D.). The king, Jayavarman II, came from Java to rule over Kambuja, and the new cult was introduced shortly after his ascent to the throne. The story of this introduction is told in some details in the inscription.2 The high priest of Jayavarman was a Brahmanieal sage named fmvakaivalya. This Brahmin was enjoying a piece of land in the village oE Bhadrayogi in Indrapura given to his family

1 BEPEO X V , pp 70-71.2 Ibid :

(51) Hiranyadma-duija-puhgavo gryadhr ivvjayonify learunrdra gatah i

(52) ananya-labdh&rjt lihalu siddhim dart prakaymsa manbhitam prati u

(53) sa bhdharendrnumato grajanm sa-sdhanrii siddhim adiksad asmai i

(54) hotre hiiaiknta-manah-prasatrh samvibhrato dhma-mvrnihanya n

(55) stran Sirascheda-Vinikhkhyaih Bammahanmpi NayoUarkhyam

(56) tat Tumvuror vakira-catuslcam asya siddhyeva vipras sumadarsayat sah u

(57) dvijas samudclhrtya sa stra srath rahasya-haualyadhiy sayatnah i

(58) siddhirwahanh Itila dcvarj-hhiihyrp. vidadhro bhuvanarddhi-vjddhyai || A. XXVI-XXIII

2 STUDIES IN THE TANTRAS

long ago by the tings of Bhavapura (founded by Bhavavarman who was ruling about the middle of the 6th century) He was the guardian priest of a vnlinffa installed in a temple in that village. ivakaivalya, chosen as the priest by the new king, subsequently accompanied him to different cities founded by the latter. Now, a Brahmin named Hiranyadma came from Janapada (supposed to be some place in India) to the court of the new king Jayavarman and began to exercise a great spiritual influence on him. The king then authorised him to teach the new lore to fdiva- kaivalya, and to initiate the latter to the new cult. Then Hiranyadma gave Sivakaivalya Siraso/ieda, Finsihha, Sam.moha and Nayotlara- the four fiSstra which were the four faces of Tumvuru. livakaivalya was also initiated to the cult of B em m ja (Kamralen Jagat ta rja in Cambodian). Thus was the new cult introduced in Kambuja, The king Jayavarman, much attracted by it accepted it as the religion of the kingdom, and ordered that the yatis of the mcUriamka1 of ivakaivalya only would he the legitimate guardians of this cult and would have the power to perpetuate it in future.

The Devarja was a s/ralinga and it was one of the most celebrated deities of Kambuja. But nothing is known as

TNTRIK TEXTS STUDIED IN ANCIENT KAMBUJA 3

The classification, according to the lorantas, is not seen in the oldest Tntrik texts, for example the Ttnalas, as far as I know. On the other hand, such classification seems to be arbitrary and of late origin. The names of 64 Tantras attributed to each of these hrnts seem to be fictitious to some extent. So they do not thiow much light on the texts introduced into Kambuja in the beginning of the 9th century A.D. It is therefore necessary to go back to older and more authentic sources.

According to the oldest traditions known to me 1 the Tntrik literature is classified according to the tirotas (=current, tradition), plthas and mia//as. The Srotas or currents are threefold: dalcsina (light), vma (left) and madhyama (middle). There are the three energies (sultti- trayaih) of iva. Besides these three currents which issue from Siva, we have reference to other currents : like Bkairava-srotas from which distinctive Tantras have issued forth. The classifieatiou into plthas is four fold : vidyjnt.Iia, mantrajntha, mudrptha and mandalaptha. The third classification, o iz , that into ant nay as, is more common than the first two. The number of mnyas vanes. But generally they are accepted to be five in number, issuing from the five mouths of Siva.1 2 3 Siva is lepresented as having four faces turning towards the four cardinal points and one on the top. The eus fern ( pnrva mHi a) spoke the Vedas, the western (pa'seima), southern (tlahsina), noithern (uttara) and the upper (TtrddJtva) mouths spoke the different kinds of Tantras. There is no trace of any classification according to the krnts. The faces of itva represent his five aspects. They aie known as Fwarleva, Tatpurusa, Jyhora, Sadijojta and Imna facing the noi ih, east, south, west and top and representing the aspects of Isa, Isna, Isvara, Brahma and Sadsiva respectively. The oiiginal Saiva canon, the ffamas, are classified according to the faces which proclaimed them (see Hindu Iconography, II, Pc. II, pp. 366 ff,). We should note in this connection that the Sadyojta mouth which represents the Brahma aspect is the western face, and naturally proclaims the Pascmmnya.

One of the oldest Tntrik texts preserved in the Nepal Barbar Library is the AUsvsataUva-Saiphit written m the Gupta script of the

1 Tina discussion is mainly based on the 30tli chapter of the Brahmaymala calledSrolamrnaya. The MS. of tbid ymala which I have examined is that preseived m the Nepal Darbar L ibi ary. I t wat. copied in Uie Nepal Sam 172 1052 A.D.

3 Bhskara Bya. in his commentary on tbe Vmalevara Tantra (see nandram Ed., p. 24) quotes from Bbagavn Paiasurma i Ta-ncmnyn parainrtha-srariipn prarunya iti

S t u dits s in t h e tn tb s

8th century A.D.1 It is a collection of five slras which form a completo whole but each may he also counted separately and has its own chapters. These five atas are (1) Lauhika-tUiarnia, (2) MlasTdra, (8) Ubtamsfolia,(4) NagasUtra and (5) Guhyastra. The last of these five is more extensive than the others together and the first hnvltthi dharnia is really ignored hy the text itself in counting on folio 27b of the text :

prathamarh mlasulrantu cLvilyam disamjfhiam I tftyain prathamain nma caiurtham pvastrahaih li

Thus the four texts are called (1) Mlastra, (2) ihsitra^ ttav m itra, (8) Pratlmna Nagastra, (4) Prva Guhgasiitra. The UbtamstrcL contains the names of 18 old iva-isras :

wjayarh prathamam [hy c]sm nivsam tadanantarai I svyambhuvam ata aaiva vthularp, ladananlaratH n vrabhadram Ui hhytarri rauravath mltuis laih \ virasam oandrahsam ca jnnafii ca muhhavimhahim || prodgliarn lala caiva siddha-sanlnam eva ca i sarvodgiLaii aa vijneyarh lraiiain pramc&varam II (fol. 24u.)

The same list with some slight variations is given by the Brahma- gUinala of which a MS. written in 1052 A.D. is preserved in the Darhar Library.2 In the 39th chapter (fol. S69) we find mention of the following texts :

vi jay am caiva nisvsam svyamblmvam alali param i vthulam [vrabhadram ca rauravam mhutatath'j n vrea ca tath cnyarh- tata rdhvam nibodhata I candrajnnah ca prodgtam lalitam tath || aiddhisaltvnakam caiva sarvodgitam alali param i /ciraam ca vidhdevi pramevara eva ca u

The second list is corrupt to some extent, though the MS. of the Braimaymala which we have examined is generally a very correct one. Apparently the tradition about these texts was more living in the time

! The Exalted Ba] Sara Hemarja Sarml who has handled these MBS. for a long limo is also of the Baine opinion. Mahmahopiidhyy& H. P. S&stri has noticed it in his catalogne, Variar Library Oat., Vol I , p. 187. In hia introduction to the Catalogue, p. lxxvu, he also says that this S. is written in transitional Gupta character which may be a century older than the Prcimehara Tantra copied in 8f>9 A.D,"

5 See H. P. atr, Nepal Variar Library Catalogue, I I , p, 60.

4

TANTBIK TEXTS STUDIED IN ANCIENT KA BOTA 5

when the Ntsvsatatlva-Sarhhita was composed, but it was not so when the Bmlmiaymala was copied. In the inteival the TSntrik literature had become a fairly extensive one. The Tantras which were therefore considered bo be authoritative in the 8th century and even prior to it were: (1) ftjaya, (2) Fisvasa, (3) Svyambhuvamata, (4) Viithnla, (5) Trabharlra, (6) Raurava, (7) Mkuta, (8) Viram, Vi resa (?), (9) Candrahsa Caiidra (?), (10) Jm, (11) M nicha v mil a ica, Pimba (?), (12) Pro Agita t (13) Lolita, (14) Siddhi, (16) Santana, Sattvna (?), (16) Sarvodgta, (17) Kira^a, (18) Pmmcsvara.1

The second text of the list, the Fisvsa, seems to be the same as the Fisvsatattva-Saihfoi. We have already noticed that the independent chapter-division of the 4 sutras constituting1 the text points to the fact that they were studied separately. The word sa'ftifot also may indicate that it was simply a compilation of different texts. Amongst the four texts, the ad aud the pralhama are the Utlara and Faya-sUiias, Their very position in. the traditional computation : mla, ad, pratfiama and piirva point out to their intimate mutual relation. It seems quite probable that they together constituted our Nayoitara introduced in Kambuja in the beginning of the 9th century (802 A.D.) The contents of these two siiti as amply show that they were indispensable for the guidance of the priests. We can understand fiom them how the Fayottara could be useful to ivakaivalya of Kambuja for conducting the newly introduced cult of Devarja. The Uttar a siitra has five sections : (i) sivlayasthpana, mirlt, homo,, (ti) and (ni) abhska and diks, (iv) and (v) jfinayoga. The NayasTitra has 4 sections : (i) ysa praharana (yaga-?), (it) pralrltvicra, (Ui) rpavtcra, (iv) p a iammfta-sadbhvamcia (cf. also Sstrl, Barbar Library Catalogue, I, pp. 138 f. ; his notice is however incomplete). Their date of composition cannot be ascertained at present. But it was certainly composed long before the date of compilation of the four texts together, which also was done much earlier than the date of copying the present text. Thus roughly it may be said that the FaycttarasTUras were *

* The same list occurs also in the Kmkgama (p. 1), published from Madras by Alagappa Mudaliar. Gf. Also G-opinth Rao, Hindu Iconography, I I , P a t I , pp. 367.68. Some of these texts exist in very old MSS. The Kirana and Parametvara lontra are preserved in Nepal. The Kiraqa was found by H . P. Sstr in a private collection at Bhatgaon. The MS. is very old and was copied in 921 A.D. See Darbar Library Catalogue, I I , pp. xxiv and 99. The Paramesvaramatatantra is preserved m the Daibar Library, That MS was copied between the 11th and 12th century A.D. (Setr, ibid, pp. xxi and 16). Prof. Bendali mentions an older MS. of that tantra copied in 859 A.D. preserved m the University Library, Cambridge.

G STUDIES IN TUE TANTHAS

composed in the 6th and 7th centuries A.D., and compiled with the othertwo texts in the 7th and 8th centuries A.D.

The Brahnaymala (loc. oil.) says that the iVivtmm and the other Tantras mentioned above came out of the middle murent and were communicated by the upper month of Siva (madhyasrota-sambhut rdhvavakirt mnirgat). In another place (fol. 200a) the Brahma/mala distinctly says that the three texts known as Samnioha, Nayottara and Sir ascheda issued from the left current [mmasrotas) :

Sammahafl oa tath prolctam tath calva NayoUaram \ [Sirachedam] 1 tath proli ta m vcimasrotd vinirgatam n

The same texts are also mentioned m a supplement to the Brahma- yam ala namely the Jayadralhaymala (Sslr, I, Oh. 40, sec infra) :

savyasiolasi siddhni iraoliidra* bhaytmakaih i nayottararh mahraudiath mahsammohonam lath i Irilcam ctat mahdevi vmasrolasi mrgalam I

The fact that the JSayotkira is here attributed to the vaiMsrotas (left euneut) whereas elsewhere it is, as a part of the Bikousalaulra attributed to the madhyania-srotas (middle current) should not be considered as a serious obstacle in accepting the identification proposed. We have actually mention of texts coming out of the combined current of vma- mad/iyamay ( Fmamadhyamay calva coditene tathaiva hiBrahmayamala, fol. 200a). There are reasons to behove that the classification according to s oias was not very well defined.

The manuscripts already discussed, as we have seen, mention two other texts, viz., the Sammoha and the Sirascheda of the ) Vim u, (5) Yama, (6)

1 Though this portion u indistinct in the MS. the reading is supported by the text of the supplement, which repeats the same tradition.

a I t is evidently a mistake of the copyist for iraeheda.

TANTRIK TEXTS STUDIED IN ANCIENT KAMBTJJA 7

Yyu, (7) K u ver a, (8) Indra.' I t is true that these Tmalas are not mentioned in the Nihvsa-tantra, but in the BrahmaySmala of which we get a MS. copied in 1052 A.D. So they all had come into existence long before this last date. But it is possible to determine the date of their composition more precisely.

The Brahmymiala has its supplements and two of them are preserved iutheDaibar Library(1) Pmgalmata, ( l) J ay adrathay amala. The latter is a very extensive woik containing about 24,000 slokas divided into 4 sathas of 6,000 Slokas each. The MS. of the Fingala mata was copied in the Nepal Samv. 294=1174 A.C. There can be no doubt about the fact that it is a supplement2 to the Brahmaymala and is connected with the Jayadratha-. In the very first ohaptei of the Pimgalmata (fol. 26) it is said ay/a lavtrasya Pmgalmatasaihjn. PialtaM&lalpaih Jayadral/nid/i ikram Brait m at/a mal any a ann yah, Fing al b haUriky hminittaih. The Pvhgal amala therefoie presupposes the existence of the ,Taj/ndral!iaija makim and professes to be inspiied by the Brahma-. The Jayadratha- was therefore written long before 1174 A.D. Though the two MSS. (satins 1 and II) noticed by H P. fsastri are of the 16th and 17 th cen tm ies 2

I Tiarymaymala, eh. DO, Sroiammaga, fol. 109a llmlraymalam anyat1 ca Lrith tat Kantlayti,inalavi \ liiahmaymalalam anca visytiynmalam eva ca I Yamaijmalalat'n cnyam Vyuymalam eva cal Kniet atjmaian c,atva Indraymalatn eva aa I Ilhairavitsllam tat VidySpitlid vmtrgalatti I Y rim al am tath casta u mrgatm na safiayali I

The names of the eight Bhairavas also mentioned in this connection, ere : Sacehanda Bhairava, /frodila, U n m a tta U g ia , Kapl", ,Thankra, 8elihaia, Vijaya.

1 Ot the satins of this text, pieservel m the Darbar Library, stri haB noticed only two, T and IT Tho M 3, of the satin T is dated in Nepal Sam. 813=1723 A.D. (and uot Nepal Sam 847 as stated by Slain, Barbar Library Cat., II , p. J) and the MS. of the natila IT a dated N S 702 = 1(112 AD Tho writing of satka I I I appears to be of the aims period but the sati.a IV le preserved in an older MS. The colophon of this Ms. (fol. 3390-340) mua thus .

adhigate-salalaistrasyo yoginivmdavanditacaranayuqalnsyavivtdhamdyvidyoLtthtatlaranasyamahaliirnnikasyamnlSrjtlhtrjatrmaiiayacandradevapptasya KulcSr = yasndiiarampdevunSmaiheyasya tisyena yanditarjamadevena hklntam iti,

Jayaeandra here mentioned seems to be the same as king Jayacandra of K&n&u] who (ell before the Muhammadans towards the end of the 12th century. The MS. was therefore

8 STUDIES IN THE TANTBAS

The Jlyadrathayanuda is distinctly called Biraacheda. We have already discussed the texts which mention irachada as being handed down by the left current {vamasrolan) and communicated by the vmavahtra (the mouth turning towards the left). All the colophons of the Jay adrafdiay amala run thus :

i t i Bhairavasrotasi vidypltjie ira'schecle Sri Jayadrathaymala- mahStantie caturvimatisasre Srhlasamharsanym, eta. {of. Sstn, toc. cit.).

The frame work of the present tantra is as follows; Jayadratha the husband of JDuryodhana's sister and the king of Sindhudesa renounced the world and settled at Vadarikrama in the Himalayas for the purpose of practising austerities. He propitiated the goddess Prvat! who introduced him to Siva. The interlocution between these three is the substance of the Tantra. The first question asked was the nature of Mukti (salvation) which was explained according to the Snhhya system but 8iva said that the telling on rosary the formula of Klasafbicarmn was the easiest and the shortest way to salvation (strI, Darbar Library Catalogue, II, p. 2.). As regards the names mentioned in the colophon strl remarked in 1905 {Catalogue, I, p. xii) that these are a string of names, the import of which, if it existed at all, is lost w but in 1915 (Catalogue, II, p. 114) m connection with the TaUvasiuWutvalanira he explains them thus, It is called Bhairavasrotas because Bhairava is the speaker and his speech began after he had snatched away the topmost head of Brah m3 and put it above his four heads, It is called Pidyplha (sic. vidypthd) because it treats of the goddess Sund ari. But this explanation is not quite correct.

(i) Bhairavasrotas, as we have already seen, means the Bhairava current or tradition. There are 8 Bhairavae from whom emanate the 8 yamalas. So other Tantras of the Bhairavasrotas either must have been supplements to those 8 ymalas or inspired by them. The Jayadratha- ymala emanates from the same Bhairava {viz., the Unmalta Bhairava) who narrated the Brahnaymala, Bhairava is conceived as an aspect of 8 iva.

{ii) We have already discussed the significance of the 4 plthas. Vidyapltha is that method of sdhan which relies on the vidy or mantras.

copied either towards the end of that oenluiy or the beginning of the 13th century A D, The script supporta it. The Bjguiu Hemarja Sarm would attribute to the script used in this MS. a Kanaujiya ehaiacter ((Kanyaltubltya-srotas).

tantrik texts studied in ancient kambuja 9

In the ease of the Jayadra thay amala it is the Klaamlcarsan mantra which is of importance.

(ih) The significance of Siraseheda is still unknown to me. I have not been able to trace the explanation offered by Sstr anywhere in the texts but some Pan'.hts of Nepal who are acquainted with the Tantras confirm his explanation. Some of the Purnas indeed preserve the story of Sivas cutting the head of Brahma, but in a little diffeient way. In the Kmapnrina it is stated that Brahma was once boasting himself as the greatest God in the universe, Siva appeared on the scene and claimed that position for himself. Brahma was, however, obstinate. Thereupon Siva got angry and ordered his Bhairava to cut off that head of Brahma which was reviling him. Siva thereby committed a sin of which he got rid by going oil pilgrimage to Benares. The story of this rivalry between Brahma, Siva, and also Visnu is told also in the hngapurna, Emapmna and Sivapurna. In those texts, however, there is no question of cutting the head of Brahma, Siva establishes his superiority over the other two as the greatest architect of the universe and proves that Brahma and Visnu are only his different aspects (see Gopinath Rao: Hindu Iconography, Vol. II, Part I, p, 105 if. and p. 296 ff.). The same story, evidently taken from the Purnas, is told by Alberuni (Sachau, II, p. 147); (t Brahman was in shape four-headed. Now there happened some quarrel between him and Satikara, i.e., Mahdevaand the succeeding fighthad this result that one of the beads of Brahman was torn off.......Thusthe head of Brahman was dishonoured by the hand of Mahdeva, who tookit always with him wherever he went and whatever he d id ......After hehad entered Benares the head dropped from his hand and disappeared/ This is briefly the story about Sivas cutting the head of Brahma, but I bave not yet found any reference to Sivas putting it on his own hand and communicating a class of Tantras through it. But it seems probable that the legend has some bearing on the Sadyojaia (lit. just-born) face of Siva, which represents his Brahma aspect and through which he narrated some Tantras (see Gopinath Rao, loc. a t., Part II, pp. 3G6-77).

It is now quite clear that the Jayadraihaymala is a Siraseheda text. Not only the colophons of the text but the text itself clearly speak of it. Thus towards the end of the Hatha III, (fob 215a) we find the following text :

Evam elan viaykhylum tantram etad anuliamam iVasisthcna purdhitam naranryanair api ||

2

10 STUDIES IN THE TANTRAS

Brahman malwmpe tu irachcdai mahbalam i Bahumantraguxilrnapy tavdya yraliaiilqtam \\

Here the text is quite clear about the identity of the irascheda and the JayadralhayStmla. Now the J ay adraihay afflala, as we have seen, is mentioned in the Ptyalmaia (of which we have a MS. dated 1174- A.D.), and apparently m the existing version of the Bra/muymala (Chap. l9j fol. 169a) as catushasatjcabhedas tu (This MS, of Brahma is dated 1052 A.D.). A part of the Jdyadratha itself is preserved in a MS. of the 12th-18th century A.D. If we admit that it is this Sirac&cda which was taken to Kambuja in the beginning of the 9th century (802 A.D.) then the date of its composition would go back to the 8th century A.D., and necessarily the earlier versions of the Urahn ay amala and other ymalaa would go back to a still older period.

An apparently later tradition mentions another work as Biracheda. It is the Karavrayot/a,, also called Pauilanka. A MS. of the lautra written in Saka year lGl (1008 A.D.) is preserved in the Darbar Library, The colophons of some of its chapters are to be noticed in this connection :

fol. 21bHi rmahsrotasi sirachode Icaravlrayogc paratati Irehuma (?) clvdaashasrihciym klikramaniri.uiyasulram;

fol. 25aiti irbhairavasroiaai irachede Itaravirayoge pamtanlre pai amvthanirnayah ;

fol. 30ah srbhairavasrotasi iraohedo mahtiaruvirayoge partantrc leali hulahramali samptalj,.

On fol. la the following words are put into the mouth of the goddess :

Sarvasrcto'dbhavaVn jnndm prasd[d] brutan maya iymalstahapurvantu laftvrmy elravidhnila (?) n iracheda ca bahudh mah-sanilina-saflcalcaih \ pamtanimm ca ciilcra (phetkra ?) sgarmbhmahsanam. , .n oaturvimsali vai laltsh sarpena tu dhrin[ ?] n 1

1 Karaora weans a cremation ground ; of. Hemacandras Abludhnacintmani, IV. 65" iwainoA harawam spt." But the word is also used as the name of particular imaiana at least in two places In Nepal a Smaiamt was known as ha.ravira, (of. S. Lvi, Le Npal, I I , p . 282). In the Tibetan texts we find Karamra as the name of a VtMia in the city of Mlapura (Mafigalapura?) in Udijiyno (Pag Sam Jon Zang, p. 137, Index, iii ; of. also Taranath, tr . Schiefner, p. 324). I t seems probable that in Uddiyttna also there was a particular cremation ground with that name which has been wrongly called a mhra. Some Tntnc practices were to be performed only in the imainas. Trace s of snch practices are also found in Bnddhist sources. Cf. Lanhavatnrn (Nanjio), p. 808

SimyBgSre inaiane va Vfhamle gnhSsti v [ palle'Vhyataltie ca yogi vsam ptakalpayet ||

TANTRIK TEXTS STUDIED IN ANCIENT KAMDUIA 11In this list the Kamvirayaga apparently mentions the texts of the

school to which it belongs. It mentions the original Sirascheda : the ma/iasantJaimmtlcaknm. The text here referred to seems to be no other than the Jayadrathay amala containing the extensive sathas. Thus it is evident that the original Sirascheda text was the same as the Jayrulralha- ymala. The Karavragoga apparently drew .its inspiration from that text.

We have already seen that the Jayadrabh ay amala. came out either of the vmasrotas or the vmamadhyamasrola

12 TDIES IN THE TANTRAS

of Phetkri Bitairava fiom whom issued the? Phetkra hot Ira- Consequently we ave justified in supposing that there was a tanka named Vinsikha tantra which was connected with the name of Pina'sikha Bhairava. If this identification is accepted then the VinUsikha tanka was intimately connected with the Sirabcheda, i.e., the Jayadraihayamala as the Kambuja inscription also would make ns believe.

The texts thus being identified, it remains to be seen which is the god mentioned as Tumluru and why are the four texts called the four faces of Tumburu.' Dr. Cbatterji says that Tumbuiu is the name of a Gandharva and thinks that he had something to do with the Gandharva lantra. But the context has no bearing on any tantra connected with the name of Tumburu. The inscription would have us believe that all the four texts were connected with that god. Tumbaru or Tumburu is recorded in all the lexicons as the name of a Gandharva but no detailed information is available on him. It is the name of one of the Yaksa worshippers of the Jiiia (see Hemacandra, AbhidhUna-cintdnian'1 > P 4.1, where the commou- tator explains the word as tunihnii ardati vtghnu lumlmruh). The Buddhist texts mention him as the lung of the Gandharvas, Thus in the MahSsamaya Suttanta (Dialogues of Buddha) Part ?, p. 288) amongst the Gandharva chieftains are mentioned Paficasikha and Sunyavaccas, the daughter of Timbaru. Then again in the Sahhapauha suttanta (ibid, pp. 802, 803) Buddha being enchanted by the music of Paficasikha questions him, whereupon the latter tells him the story of his love for Bhadd Suriyavaccas, the daughter of Timbaru, the king of the Gandbabbas. In the stoiy Paficasikha figures as a great musician who had a lyre of yellow Beluva wood. The Gandhabba Timbaru is also mentioned in the PsdihasuManta. In the Chinese translations of these Sutras the name of the Gandharva is transcribed as Tan-feou-lu *tm-bieu-ru= *tamburn and as Teou-fvoti-lou= * Teu- h\m-xa=*tu{m)hum \cf. Trijpitaka, New Tokio Ed., Yol. I, pp. 80, 033]. These forms show that in the corresponding Chinese versions the names presuppose the forms Tamburu, and Tumburu, and not Timburn as preserved in the Pli texts. The Mahbhrata refers to Tumburu on several occasions: in diparva (65.51): snpiy ccdibhus ca vikhytau ca haha h huh I Tumburns ceti catvrah smrtli Gandharwsatlamh || and again .di (159 5!) : Gandharvaih sahitali srmn prgyatas ca Tumburuh, In the first verse Tumburu is evidently used as a general designation of the four Gandharvas : Biipnya, Atibhu, IW m , Hh, whereas in the

TANTEIK TEXTS STUDIED IN ANCIENT KAMJSUJA 13

second verse it is used, in all appearance, as the name of one particular Gandharva, who was a musician, Whatever it may be, the number four seems to have been connected with the name of Tumburu, though it is difficult to determine at present whether it was originally the geneno name of the four Gatidharvas or the name of a particular Gandharva with four faces.1 But there is no doubt that Tumburu was pur excellence a musician. He is mentioned as an authority on the musical scieneo.

The SaihgUlolca, while mentioning the oldest authorities on music, cites the name of' Tumburu. These authorities are: (1) Brahma, ( i ) Siva, (3) Nandikesvara, (4) Siv, (5) Rarnbh, (6) Tumburu,etc. (. ... Sivamnclilcevara sivSramMslM/ lumhmih.........cf, Castri,Catalogue, "Vol. II, p. 7, and also Introduction, xxxv). These aie the names of gods who revealed music to the mortals. Nandikesvara is another name of Siva ; Tumburu is a Gandharva. A stringed musical instrument, Tamburo, is connected with his name. Though

1 Tlieie seems lo have been u time when Siva, was four-faced. The Mahblirnla preserves its traces (Anti

1 4 STUDIES IN THE TANTRAS

fchore is no definite text to fall back on, still it scorns probable that J Tamburo was no other than Siva himsolf. Both of them are repre

sented as having four faces, and both of them are authorities on music. If this identification is aceepted (hen a new light can be thrown on the text of the inscription of Sdok Kak Thom. The four tantras : Sira-

J iscJieda, Vmtk/ta, Sammoha and Nagottara are said to be 1'umburor vahtracatiisham, not because they constituted the four faces of that god but "because they issued forth from, or were communicated by, his four mouths (valUm). Besides we should note that valUra really means mouth and not face. The four Tntrik texts therefore seem to have represented four different wugas, connected with the four faces of Ttimburu, who was, in all appearance, an emanation of Siva himself like the Bhaimvas.

The introduction of the four texts throws some light on another problem, viz., that of the relation of Kambuja with northern India, Dr. Ohatterp has already tried to trace some of the elements of Kambuja culture to North Indian origin (Indian CuUural hijluencc in Cambodia, p. 253 f.). Now there are reasons to believe that the four Tantras brought to Kambuja by Hiranyadma were of North Indian origin. The Pingalamala, which we have seen to be a supplement to the Braltniagamala and to be connected with the Siraschcda-Jagadralha- g amai a, is very clear on this point. On fo lio 5 b we find mention of the countries where the Siva-sdkan waB in vogue. It is no doubt_ Othe country of the ryasthe rgvarla :

Vindhyottaragatenaiva Magadhccparena tu \Himdrer daksine bhge patcll prvalas iath u Aryavaria iti hhytas ladbhavcryasdhalcau i Agrajanmakulodbhtafy aarvasdhratyo yataJi n Viesana ca iath brmi agrairahtivcahah \Saktyaiitai janitam janma janmagrety abhidliiyate il Ka-pi vxtavinirniuld[] anyadeodbhuvvap ^Emarpanoa Icmirau khhgau hanhuiiodbhavau II Kmiiio&

TANTRIK TEXTS STUDIED IN ANCIENT KAMBUJA 15

Koala, Kverl-rstra (?). This shows that the oldest Brahman ical tantras which included the 18 texts mentioned in the vsataUva- samhit, the 8 ymala and their supplements, all originated in Northern India, The four texts, the 1Sfayottam, SiracJteda, Vtnsik/ia and Sammo/ia, taken to Kambuja in the beginning of the 9th century A.D, would therefore be of North-Indian origin.

Thus we see that the four Tantrik texts mentioned in the Inscription of 802 A.D. during the reign of Jayavarman II are partly preserved in old MSS. in the Nepal Darbar Library. The Nayotiara was probably the same as the Nay a- and Uttam-stras which form a part of the NisusataUvammhii, now preserved in a MS. of Gupta writing of the 8th century A.D. It was composed much earlier than the date of the MS. and may be safely placed in the 6th-7lh century A.D. The SmAcheila was iu all probability the same as the original JayatlraUta- ymala of which an extensive text copied in the 12th-13th century A.D. exists in the Darbar Library. The Viusik/ia seems to have been a supplement to the Jayadrat/iaymala and the S ammo/ta, the original, on which the later Tantras of that name were based. The four Tantras were of North-Indian origin.

Tumbaru appears to have been an emanation of Siva himself, who is represented as having communicated the four texts through his four mouths.1

1 Eor furtliei details on Tumbaru see infra pp, 22 ff.

f u r t h e r , n o t e s o n t a n t b i k t e x t s

STUDIED IN ANCIENT KAMBUJA

The Affama! anil their influenae.-The inscriptions of Kambuja abound in references to the Samte canon. Various inscriptions refer to Sivaastra,1 SitivTigamafl Sarvgamu,8 and Saiva-vg/carana.1 * 3 4 5 6' Agama means the oldest Salvile canon which conformed to the Vedas and had not entir ely separated from the Vedic religion like the later aiva sects. SUsira was a term synonymous with Agama. Agamas are generally believed to be 28 in number bub we have already discussed the tesi, of the NTsvsatattva SamhitS, itself an Agama, which mentions only 18. We have also pointed out that Illeso 18 Agamic texts must have existed long before the 8th century. The references to Agamas in the inscriptions of Kambuja, the oldest of which go back to the beginning of tho 9th century, confirm the same view. One of the four texts mentioned in the inscription of Kambuja, via., the NngoUara, at least belong to the Agama proper wliilo the three other belong to tho canon which grew later on under its inspiration. In the inscription of Angkor vat > we find another reference to an Agamic text- it is tho Pmmesvara (tasmin Kuril niahadgagarh, i/atholctam pramesvare) This is the 1 Tira mas u ara la n tra also called Pramesvaramalatantra which is one of the 18 Agamas mentioned in the Nisvsasarhht list. It is the 25th of the 28 Agamas mentioned in later literature.0 We have already seen that there is a MS. of the Pramevaiatanlra copied in 859 A.D. The work was certainly much older, as it is mentioned in the IShmisaeamhita list of which we have a manuscript of about the middle of the 8th century.

In my last article I have tried to show that the original aivite canon which contained the 18 Agamas was of North-Indian origin as, according to them, the best Sivcryas were the Brahmins of ryvarta. But the people of the surrouuding countries, Kmarpa, Ksmira, Kalinga, Konkana, Kci, Kosala, Kvcr-Ratfa were not eligible to that position for their physical deformities. By physical deformities we have

1 Inscription ol Phnom Sandal, of about the end of the flth century A.D. Beijgaigne,Insoriptum de Oampa et du Cambodge, TL, p. 157.

3 Inscription of Anghor vat, tbul, p. 392. Ibid, p. 389 1 Ibid, p, 392.5 Beigaigne, Inscription, etc., p, 390; also p. 384 with the note of Barth.6 Gopmath Bao, Hindu Iconography, I I , P a rt I, pp. 367-08,

FURTHER NOTES 17

to understand that their statures did not follow the prescribed standard and were either too tall or too short (aUtlirgfia ati/irasvaka). Such a conclusion is also substantiated by other evidences. The Tantrasra which is a famous compendium of Bengal Tantrism says on the authority of Kriysrasamuccaga, mala and Vaisampyana-samhil that the persons with physical deformities of various description, and persons who are diseased, immoral, etc., cannot be gurus (ibid, p. !3)M a ninyagmmha

Kriysra-samiccayeSviir caiva galatkuslhl nctrarog ca vmanah i kunaltl yvadanta ca slnjilaS cdlukhgakali n hinngah kapatlmgl bahvi bnhujalpakah i clair dosair villino yah sa guruli isyasammata\ il

Ymalcabhisaptam apuh'a ca kaanjam ln ta vani lathi kriyhinam sathan cpi vttmanaiii gurunindakam 11 jalamklatlcran ca varjayen matlmn sacl i sad matsam-samyuktam gurum tantreria varjaycl n

Vaiampyana-samhitymaputro mrtap utra ca kuslhl ca vmanae talh- n

The same compendium again says on the authority of Jbla (quoted by VidydharcSrya) that the quality of the gurus differ according to the countries in which they are born. According to it the best gurus are found in the countries of Madhyadesa, Kuruksetra, Nata and Kohkana (or Nata- Konkana?), Antarvedi, Pratisbhna, and Avanti, The Madhyadesa is ryvarta. The gurus of the second quality are found in Gauda, lva, Sura (?), Magadha, Kerala, Kosala and Dasrna. The worst gurus are those who belong to the countries of Karnata, Narmad- Rstra,1 Kaccha, Klmda, Kalamba and Kamboja2 (ibid, pp. 10-11) ; talh Vidgdhar- cryahptam Jblavacanam

Madhyadcaa-kiiruksetra-nalakonkariasambhavah tAntaruedi-pratisthnvanty ca gurttamli n

1 I t is evidently the same name as quotai! in the list of the Pifigalmata. Through mistake I connected it with ICaveri and took it to mean Kaven-rditra, I t seems to be a different country and probably is meant for Surstra

2 Konlana, which is amongst the forbidden countries in the Agama list, here is placed in the fust rank. Nata-Konkunn may howevei be a mistake for another country. The countries of Kailmda and Kalamba are not known. Kahnda (certainly not Klindi)

3

18 STUDIES IN THE TANTRAS

Madhyadea, ryvartaGaudh lvh surscaiva mgadkJj. Icorls talh i Koalca dasrnca guraoah sapta madhynmh n Kartyta-mmiad-mstra-liacchiiLrodbhavs laih i Klind ca halambs eu hamboj cdhum nudh il

This list was certainly drawn up at a time when the authoiity of the orthodox gamas was a little undermined by a rise of the heterodox schools. But it still shows the old tendency according to which the cryas of North Indian origin were given the first place.

This throws some unexpected light on the recruitment of Sivcryas in different countries including ancient Kambuja. We have seen that Hiranyadma earns with the new Sastras from a janapuda, which was most probably a jampach in India. The family of ivakaivalya, who was initiated to these stras, was long established in Kambuja. The history of this family, recorded in the inscription of Sdok kale Thom is of great interest. The members of this family enjoyed the priesthood of the king through succession since the timo of Bhflvavarman (middle of the 6th century A.D.). They were ioryas aud were guardians of Unga established in different places. The succession of the priests was determined according to the nuUrvamba i.e,, maternal lin eag e (taimbrvamse yatayas striyo va j t vidy vig-viltnuiia-yiihlabiuu,h \tad-yjaks syuh............BEFEO, 1915, p. 62) which implied that thesuccession was to go to the children of the sisters (bhgineya) or to those of the daughter of the sisters, or the elder brother. There are several cases of such succession recorded in the inscriptions (ibid, p. 54). It is difficult to explain the necessity of such an arrangement. Barth in 1901 thought that such an arrangement was necessary because the royal priests used to take the vow of celebacy and therefore they bad to choose their successor from the line of their sisters. But M. Finot (ibid, p. 56) says that it is difficult to accept this explanation as we hear of priests (though of very late times11th century A.D.) who were married. Tt is however clear that the intention was to avoid difficulty in finding a successor because when the branch lines are counted the family 1ms an unlimited scope. But what was the necessity of sticking to a particular

seems to be a mistake for Kulinda. Kamboja does not Beem to be the ancient country of the Kamboja-Q andhara group. I t may be the country of the people called Kam-po-tsa mentioned in the Tibetan sources and located m Assam, These people seem to have been the predecessors of the modern Koch.

FURTHER HOTES 19

family for the selection of priests ? The only explanation that occurs to my mind is that according to the gamas the Sivcryas had to be chosen preferably from the Brahmanical families of North-Indian origin. Such families were not numerous in Kambuja. The family of Sivakaivalya was probably a rare one and priests had to be chosen from that family and its branch lines, as the members of them alone were fit to be tUvcUryas, In the inscriptions of Kambuja we have several other references to the families of North-Indian origin, of which the members attained the position of royal chaplain. Thus we hear of the royal chaplain Bhat^a Divkara who came from the banks of the Klindi (Yamuna) and was thus an expert in the Yedic sacrifices (Bergaigne, Inscription, I, p, 81 fE.). In an inscription of Angkor vat we are told that the royal priest Sarvajnamuni, who was a special adept in the aivite rites, came from the Aryadesa (Bergaigne, Inscriptions, etc. , lxv, 9, p. 388 : ryyadeesamipannasi Sivrdhanatatparah i yo yogengatah Kainvudehe ,), In the same inscription we hear that a descendant of Sarvajamuni filled the country called Madhyadesa (here a part of the ancient Kambuja) with Brahmins versed in the Veda and Fedanga (lxv, 22 . eakara desam namne inaiti mailhyade'sam jankulam i vedavedngavidmpram,.,). There seems to be a reference here to the immigration of Brahmins from India. In the inscription of Prah vat we find mention of a Brahmin, named Agastya, related to the royal family, who originally came from the ryudesa (Bergaigne, Inscriptions, etc., xhv, 5, p. 181 ailta dvijo- gastya iti pratto, yo vedavedngavid ryyadese...). Such practices were known m India too. The great Cola king Bjendra Cola who built the KSjarjevara temple at Tanjore is stated to have " appointed Sarvaiva Paiiijda-ivcrya as the priest of the temple and have ordered that thenceforth his iSisyas and their iSisyas alone, belonging to the Aryadea, the Madhyadesa and the Gaurladusa, shall be eligible for the office of chief priest (South Indian Inscriptions, II, 1, p. 105, wrongly referred to as II, 2, p. 153, in Hindu Iconography, II, 1, pp. 5-6). We also know that the Malia kings of Bhatgaon (Nepal) had Brahmins from Bengal as their priests. These Brahmin families used to come to Bengal from time to time to contract marriages in order to maintain the purity of their family tradition. This was however the custom most probably in the pure Stvasdhan, i.e., gamnta Saivism. With the heterodox aiva sects like the Psupatas and others the praetiee was different. Thus in Nepal the priests of Pasupaiintha were recruited only from amongst the South Indian Brahmins (S. Lvi, Le Npal, I, pp. 864-65).

STUDIES IN THE TANTRAS0

The influence of the Agamas can also be traced m the fciaivite cult practised in Kambuja and Campa. There are ample evidences in the ancient inscriptions to prove that the constructions of the Sivahtigas were made according to the piescnption of Ike canon. According to the gamas the litigas can be of two lands, the cala, i.e., movable and the acala, i.e., immovable. The cala litigas are again of different types : mrnniaya, earthen; lohaja, metallic; rainaja, of precious stones ; dnija, wooden ; sailaja, of stone ; and ksanika, those made for temporary worship. The lohaja, i.e., metallic litiga* are made of 8 metals : gold, silver, copper, bell-metal, iron, lead, brass and tiu and the ratmja ones are made of pearls, coral, vaulmya, topaz, emerald and bluestone.1

The acala or sf/icivara litigas are of 10 kinds, Svyambhuva, Parva, Daivata, Gnapatya, sura, Sura, rsa, Rksasa, Mnusa and Bna. The Makutfigama calls them Slhira litigas and divide them into four classes : Daivika, rsaka, GSnapa and Mnusa.

In ancient Campa fSaivism was the predominant religion and iva was worshipped mostly in the form of a litiga. A litiga established by king Bhadravarman towards the close of the 4th or the beginning of the 5th century A,D. became a sort of national deity for the people of Campa. This litiga is differently called in the inscriptions Bhadresvara, ambliu- Bhadresvara, and most probably also as Srlna-Bhadresvara (see R. C. Majumdar, Campa, pp. 177 fE.). Tbo inscriptions do not generally speak of the materials used for the construction of the litigas. Many of them, specially the mikhaUhgas, were certainly carved from stone. But we have some references to other types of litigas too. An inscription of Po-nagar, dated 965 A.D. (Majumdar, II, n47), speaks of the gold and stone images of the goddess, i.e., Bhagavatl (haiml and sailamagi pralim; insev. n 45 : Italadhautadeha), erected by king Indravaiman. This shows that both gold and stone were used in the construction of the images of deities in Campa. Another inscription is more explicit on the point. The Tang Tikuh Inscription of Indravarman I (dated 7l aka799 A.D.)contains two stanzas which have not been correctly interpreted till 1 2

1 See G. Reo, Hindu Iconography, I I (i), pp, 75 ff The gamas from which he derives the information are : Suprabliedgama, Krangama, Knnhgama, Malcutgama and Knangama ; ej. ibid, I I (a), App B, p. 3 ff.

2 cs. ibid, I, Inscr. I I , TO ; XV, B ; 26, XYII, B. 26 i XVIII. D. 27, B. 24 ; and H , Inner. L X I, 0 .11 .

E WITHER NOTES 21

now. The stanzas in question are (see Majumdar, LI, n 23, viii and ix)

tnfsfcj irrfctft i

cra i^ i^ tr xi;=i [ V i l i ]

c t# r ^ ^ftsixHlw*! if% siri s d a p m iM il [ f X ]

Dr. Majumdar translates the stanzas thus : after Bergaignc and Barth : Indravarman also installed an earthen hfiga of the God, which therefore came to be known as Indrabhadresvara. He also established in the year of the Sakas asi yam dn (721), two treasures for the god, the one composed of movable and immovable property, and the other moveble and with a mouth (priests ?). 1

The last part of the translation is evidently unintelligible. There is no question of " property " iu the text and t( a movable treasure with a mouth (?) does not convey any moaning. Ko'sa here, as in many other cases in those inscriptions, should be taken in the sense of hnga-ho'sa, Ko'sa was apparently an outer covering of the lingo, and was used probably for decorative purposes. The inscriptions of Campa very often record the gifts of iosa made by the kings to the ling as. Those hosas were often golden and decorated with costly gema. The hosas had sometimes faces and kosas with six faces are twice spoken of. We find mention of TTrddUva- Icosa which was most piobably a detachable one (see Majumdar, Campa, I, p. 182). If in the present case we take iosa in the sense of linga-hoki, the text becomes clear. It should then be translated: Indravarman also installed an earthen {pattuiva lingo) of bun (the god) which therefore came to be known as Indrabhadrevara, He also established, in the Saha year sasi-yama-adri (721), two hosas, one cara, i. e., movable, and the other siliira, i.e., immovable. The movable (cara) kosa bad a face (or faces)/ The Unga was an earthen one (prt/nva) which corresponds to the mrnmaya-linga mentioned by the gamas and it had two hosas, of which ono was movable and the other, probably a simple cylindrical one, was a fixed one. The cara ko'sa had a face (or faces) and thus when fixed to the hngas used to convert it into a mukhalingo. The two words cara

* Bergaigne, Inscriptions, etc., I I , p 33 et 37VII1-IX Sri-Iildovarman a ug ausai un tinga tenestie de ce dieu, qui a t appel dsormais dun autre nom Indra- bhadrevara. Il a aussi constitu poui lui deux tisois ; l'un compos de biens meubles et dou d'eloqueftca., (les prtres du templeBarfch).

22 STUDIES IN THE TANTRAS

and slltira naturally remind us of the two types of Unga, mila, movable and acala, immovable, also called sthira or sthoara in tbc gamas.

In ancient Kambuja the Ivhgas used to be made of metal as well as precious stones. We have references to hngavi hama'sobham, suuarna- magalinga, svarnalinga, ltaladJmtla-linga, sjthatihahnga and mamhiiga. The materials used for the construction of these Ivngas therefore were chosen m accordance with the prescription of the gamas. They all were of the type known as calaknga and fell under its subdivisions : ailaja, lohaja and ratnaja.

The four faces of Tumburu,I have already tried to establish that the four Tantnk texts tsirascheda, vtnsiha, sammohana and wyottara mentioned in the inscription of Sdok Kak Thom, weie authentic aivasstras, being studied in Indiani the 7th and 8th centuries A.D. if not earlier. These texts constituted the valciracafoislcam"1 of the god Tumburu and were introduced in Kambuja for establishing the mystic rites known as devarja( sidd/nh....... devambMhhg), Tumburu evidently had some sort ofconnection with the L evarla cult. Devarja was a phallic representation {linganija) of Sivaand we have already seen that Tumburu was an emanation of Siva himself. The inscription of Sdok Kak Thom tolls us that the f irs t temple of Devarja was built by Jayavarman II (802 A.D.) in his new capital Mahendraparvata (Phnom Kulen), and tho royal chaplain ivakaivalya was appointed priest. The deity was subsequently taken to Hariharlaya where the capital was shifted. Afterwards when tho king Paramasivaloka (t. e,, Yasovarman : 889-910 A.D.) built his capital at

1 Wc have now a definite text beforo us which supports tho identification of Tumbum with Siva. In the Yogavsiqtha-RmSyaxia (Ntnna-pralarata, I , X V III, 28 20), compiled before the 9th century A D the following verses :

tty astaoarya-ijukl.s t wtaro raudra-eespitli 1 ladoinmitci vyomni sanali Icenpi hetuw |j utsavaih 'paramani calmili paramrthapialaleam | wmasrol oyats, et s Turniiumih Rudiam Sritali || pUjayilv jagatppjau devait Tumbim-Blmravau | vicitrrthli kath oalrur~madirmadato^h ||

These things aie spoken of the eight mtrlcs who were one day out for amusement, They arc heie characterised as the followers of 'the left current (vUmasrotogai) and lelated to lludra who is Tumburu, i.e., the Tumburu aspect of Eudta {Tumhuntfn, Rudram). The eight-mat'l's ara here made to worship the two gods Tumburu and Bhnirava. This passage cleaily mentions Tumbiuu as an aspect of Eudra, I t Bhould also be noted that Siva is often referred to in the msouptioiis of Knmtmja as Caturanana, Caturmukha, etc. Cf. Borg,ugne, Inscriptions, etc , H, n LXIV (p. 377) ; xT XLIV (p. 183) i n LV (p. 210).

FURTHER NOTES 23

Yosodharapura (Angkor Thom) he brought, the deityto tho new capital and placed him in the temple of Vmm hnll (lit. the central moun fc, which was built in the centre of the city for receiving the deity.

This central edifice erected by Yasovarman was for a long time believed to be the Bayon which is situated just in the centre of Angkoi. But M. Pinot in his recent studies (Etudes Asiatiques, Vol. I, p. 245 ff.) has tried to show that the inscription of Sdok Kak Thom has told a lie- A detailed examination of the sculpture of Bayon has led M. Pinot to believe that Bayon could not have been originally a Saiva temple. He thinks that the newly built capital of Yasovarman was not placed undei the protection of the Unga Devarja, the national deity of Kambuja, but under that of the Bodhisattva Avalokitevara. Necessarily he was led to conclude that Angkor Thom and Bayon were not built by king Yaso varman, as the inscription would bave us believe, because he was a $aiva, but by his predecessor Jayavarmau II who was a Mahynist. Yasovarman according to him played the part of a vandal and changed Bayon into a sanctuary of the tinga. The principal reason for starting this theory was that the sculpture of Bayon is almost entirely Buddhist. But it might be argued that the temple was begun as a Buddhist one and finished as a aiva one. But to this objection M. Pinot answers that even in seveial niches of the towers the central figures were originally those of Buddha. They were later on deliberately destroyed and replaced by tinga. Another serious difficulty remained to be explained away. Each tower of Bayon is decorated with four colossal faces turning towards the four cardinal points. In 1911 M. Pinot interpreted them as the architectural translation of a catnmuhhaliga. He, however, gives up that explanation in the light of later researches and now thinks that they represent the faces of the Avalokitevara. He is aware of the fact that no such architectural representation of Avalokitevara is at present available but he still supposes that the architect wanted to represent Avalokitevara as looking in the four directions and thus protecting the city on all sides. M. Pinot would therefore conclude that the inscription of Sdok Kak Thom has distorted the facts. Bayon was not originally a Sai va temple and Yaovarman, who was a staunch Saiva, could not be its founder. It was founded in the time of Jayavarman II (802-869 A.D.) who was a Buddhist king.

But M. Philippe Stern in his study on the evolution of the Khmer Art (Le Bayon dAngkor et l'Evolution de l Art Elmer, 1927) has questioned the hitherto admitted chronology of the monuments of Angkor on grounds of style. According to him, Bayon did not exist in the time of

24 STUDIES IN THE TANTRAS

Yasovarman. Therefore, the central mount (V m m kantl) of Yaovarman has to be searched for elsewhere, He thinks that it should be identified with the PMmmkas which, in all appearance, occupied the central position in the old city. The city developed m course of subsequent centuries and its centre was naturally removed. According to the chronology proposed by him, Bayon could not have been built before the time of Udayditya- varman II (1049-1052 A.D.) or that of his predecessor Sryavarman I (1002-1049 A.D.). The outer walls of the city would belong to this period.

But M. Coeds in a recent study (BEFEO, XXVTIT, pp. 81 ff.) has tried to prove that Bayon was built still later during the reign of Jaya- varman V II (1182-1201 A.D.). According to him the outer walls of Angkor Thom and some other buildings, which are of the style of Bayon, were constructed in the same period. Thus both M. Stern and Coeds agree in placing Bayon and the outer walls of the city in the same epoch though they do not assign the same date to their construction. Both of them disbelieve the testimony of the inscription of Sdot Kak Thom which clearly attributes the foundation of Yasodharapura (i.e., Angkor Thom) and "Vnarn Kantl, 'the central mount, to Yasovarman.

This inscription, we have already seen, was composed in aka 974 (=1052 A.D.). The dale of its composition therefore falls in the period to which M. Stem would attribute the construction of Bayon. It seems strange that a contemporary inscription would mean by Fnam Kantl any other edifice except the Bayon. What is possible is that the traditional history of religious foundations, which it records, is confused. Its attribution of the foundation of Bayon to Yasovarman may therefore be easily questioned but Bayon was certainly considered as a sufficiently old edifies in the middle of the 11th century for affording scope for confusion about its real founder to the author of the inscription.

According to M. Coeds and M. Stern, the construction of Bayon and the outer walls of the city would fall in the same period. The towers of Bayon and those of the five city-gates are all decorated with four colossal faces. What do these four colossal faces represent ? Are they the representations of the faces of AvalokiteSvara, as M. Pinot thinks ? Even admitting that Jayavarman II has not directly, but through his tradition, influenced the construction of Bayon, it is difficult to believe with M. Finot that he was a Buddhist king. M. Pinot takes him to be a Buddhist firstly, because he came from Java (or Srivijaya) which was a great centre of Mahyna Buddhism in this period andsecondly, because he

FURTHER NOTER 25

founded the city of Amarendrapura, formerly identified with the ruins of Bantay Chmar which is completely a Buddhist city to judge from the sculptures. But the identification of Amarendrapura with Bantay Chmar has been reasonably questioned by M. Stern (loc. cit ). We should also bear in mind-that the posthumous name of. Jayavarman II is Pcvamevara (the Supreme LordSiva). The cities which he built Mahendraparvala, Hankailaya and Amarendrapura are all connected with the names of Siva. The last name seems to be only a different form of Devarja. The priest whom he chose as his chaplain, ivakaivalya, was a Saiva and came from a Saiva family. It was again he who authorised Hiranyadma to introduce the texts of Saivgama along with the Saiva cult of Devarja into Kambuja. He really made it the religion of the state, erected its temples and granted lands to the priestly family for its maintenance. Besides it would he wrong to say that the sculptures of Bayon have no trace of Saivism. An important bas-relief of the first gallery of Bayon (seeCornaille, Guides ana' Enines dJnglcor, p. 13 >, n. 36) represents three temples in one row, of which the towers bear tridents (/risiila) and the deity in the centre is a ivalinga. In the face of these facts, it is difficult to admit that Jayavarman II was a Buddhist king and that be introduced Mahyna from Srlvijaya into Kambuja. There is no reasou to suppose that the four Tantrik texts brought by Hiranyadma had anything to do with Mahyna. Jayavarman II was a Saiva. If any one of the edifices (for example Bantay Chmar, Bayon, etc.), containing some Buddhist sculptures can eveu be pioved to have beeu constructed m the time' of Jayavarman II, the only possible explanation is either that he was a tolerant king and allowed Mahyna to flourish m the country, or that he had employed artists who had come from the neighbouring territory of Srlvijaya and had Mahymsb training. I t will be wrong to suppose that Mahyna Buddhism of the 8th-9th century ' D. was very much antagonistic to TSntrik Saivism. Though the sculpture of the temples partly seem to be Mahynist, the indwelling deity was no doubt Siva.

It seems difficult to admit that Bayon was not originally a aiva temple. The state religion of Kambuja was always the cult of Devarja. A temple like Bayon, which is situated just in the centre of the city,1 could

2 The question of the date i.nd position of the Bayou has become more complicated after the recent trial excavations made by MM. Cda and Goloubew at Angkor (Annual Bibliography of Indian Archeology, 1932, p. 40). I do not pretend to hold to the suggestion which has occuned to me m iegard to the dale and position of the temple in this article. I t IB, I hope, clear from the article that my main contention beam on the significance of the four colossal faces over the towers and gateways

26 STUDIES IN THE TANTRAS

not therefore have Keen meant for any other deity except DevarSja. If in some of the niches of the to were of Bayon the figures of Buddha havo boen deliberately destroyed and substituted by Unga we must attribute that work of vandalism to a period when the king was a very orthodox one and did not even tolerate the sculptural representation of Buddha in the temple of evarja, as his predecessors used to do, It is therefore necessary to go back to the older theory of M. Pinot that the four faces of the towers of Bayon (as well as those of the towers of the city gates) are the sculptural r> piesentation of the four faces of Siva. Bevaraja was in all probability a muldalinga and it was quite natural that the towers of its temple and those of the city-gates constructed in the same period would bear the mulihalinga symbol. This explanation seems to have a strong support in the inscription of Sdok Kale Thom which says that the four sad ras which prescribed the cult of Bevaraja constituted the /ou r faces of ike Tumbimi. It may nob be therefore improbable that the four colossal faces oil the towers are architectural translation of the four faces of Tumburu, Tummiror valet ra- catuham, mentioned in the inscription, because, it is through thoso four faces that the god originally oommnnieated the four fundamental texts which prescribed the religious rites of the king and his people. They arc the symbols of the different miyas of the Haivito canon.

THE SANDHBHASA AND S AN D li A V AO AN A

III

Prof. Vidlmsekhar SssLiI in an ai'fcicla published m the Indian Historical Quarterly (U 28, pp 287ff.) Ins tried to determine the exact meaning of the expression Santlhbhs. Ho lias oolleeted a large number of facts which justifies us in rejecting the old interpretation suggests 1 by Mah- tnahopdhyya H. P. stras the twilight language {londhn bh'). There can be no doubt that the proper reading of the expression is 8 and hlt hand, and not SancUty-bli^cl, though it occurs in a large number of badly copied Nepalese manuscripts. The large number of texts quoted by Prof. Vidhusekhir Sstr has enabled him to interpret if as ahpryiha vacano, or ncyr/ia vacano, i.e., intentional speech. This interpretation is in agreement with the Chinese translation of the word as " secret, hidden, and thus that of which the sense is to be made cleir (=znegrlfia). bhipryiha means that it is intended to imply or suggest something different from what is expressed by (.he words. (V. S. Sstr, loc. ei', pp. 293-294.)

I leave it undecided for the present wheLhor the expression SantHili- vacana was used m the ancient texts like the Sadd/uirma-pundtiriha to emphasise on any deeper meaning of the particular texts in connection of which it is used. Eut there is no doubt that it Was used to mean a symbolical language to signify something different from what is expressed by the words in the later Buddhist texts belonging to the Vajrayaa and the tlahaflym. It is evident even from the CVi.ryjcafyaviniseaya published by E. P. SSstrl. The expressions like nagara bhire domb tohori Icuili (p. 19), 11 h blio vota riiiulhel (p. I t ) , Kabkolapriyabola- melahaiaynanda splinrat-kiindarh | Sadyah sodhita slillitkarh kli]ars caknnah || (comm. p. 33), etc., cannot convey any meaning, if interpreted literally. We have, therefore, to assume that there is some hidden meauing in them.

The Ilevajrataidra, a canonical text of the Vajrayaa, not yet much studied, contains a chapter on the Sanclhbhs. It gives the clues to the interpretation of many symbolical words (Sandhmcana) used in the Vajraym and Sahajagna literature. It is difficult to determine the age of the Revajratantra at present but it must have been compiled before the

2 8 STUUlliK LN TU IH TANTliAS

Loginning of the eleventh ceti Liny A D., whon it was (.nuislafod into Chinese by Fa-hu (Tokio edition, XXVII, '!), There is also a Tibotau translation of the text. The great importance of the text in the Vajnum literature may prove that it probably belonged to the very early days of the VajragSm (7th-8th centuries A, D.).

The 18th Chapter'1 of the Tlevapa deals with the Sandhbha FIeva']m-sarvatantranidiia~smdkb/is nTirna patalah). In this chapter Sand/iab/ifisa is characterised as the mahsamaya ( = Chinese fa san-mei-ye, great Samara) of the yogis, and also as "the great language (waMbksam) and is " full of the meaning of doctrines (Sainaya-sanhela vistar am). The Chinese translation of the text is a faithful one. The word BandMb/ts is rendered as fang pian skuo. Fmg-pion means upyikam (cf. MaMvyufpatti, Sakaki, 6339). Rosenberg (Vocabulary, p. i2) interprets fang pieu s/iuo as aupacrilcah. Fang-pien is used here as an equivalent of Bamaya, i.c., special doctrine of the school represented by tlio Tl/ivajraknitra. The doctrine is summarised in one stanza which occurs in tho beginning of the tanti a for explaining the word Hevajra

Hekrena mahkarun vajraiii praji ca bhauyato |Prajnopytmakam tantram tanme liigaditarn srnu ||

This is the fuudamenlal doctrine of tho Vajraym as lias been explained in such works as the PmyTiopyavinisaaya-siddln of Anangavajra, rooently published by Dr. B. Bhatfcacharyya in the Graekwads Oriental Series.

The passage which deals with the Bandhbhm has been established from a collation of the three different manuscripts, and the Chinese translation :

[ D, fol. 9-9*j P, 36^-27^ ]

wroire

KftvtW

1 This conesponds to the 3rd Chapter ot the 2nd Kalpa ot the Sanskrit text. The Sanskrit MSS. ot the text are divided into two Kaipas, of 10 chapters eaoh. In Chinese translation, hswaver, the numhenng of the chapters is continuous! thus the 3rd Chapter of Kalpa 11=Chapter 13; I have consulted three MSS. of the Hevajratantra,ie MS. in the Nepul Darbar Library (>), tha t m the possession of Prof. 9 . Tucci who kindly lent it to me (T) and the last one m my possession (P).

SANDHBIIS AND SANDIXVACANA 29

n?s an s i ari nasi nfn ciEjT I nfitif:1 ats:2 sts: s apami 1 fifai h surfit: s'ai s# aff ^n i^st7 an i aw# 3 n# sttfosi flcf I!SRST fsfiwn SSti SHrSS I a i sfasf #4 rsts nretcfcss n ajsf sgwn stai gs stqftsrr wat; i ntsateRrsti =5 4 ^ stfri ni mtnfnntfii sis ftfisfti i^ s 10 i sa siifnfit11 ariti tra s tiisr i12 n i n l safss ai4 sw fs tfs i normst ssr a; farei sa sfifistt: # # s a i# atrat a# s a i# erst i assi 1 ksi# as fitat sm s# 1 * nsi n *a? st4ii# s ast nsc ffnfi?t: i arar a i nsa sa* gnfirai fss ^s# ii annnnsrasi ssrst stfir afa i s^nte arsa arar n*twts ntnt, n fsfafsitsa fs# a sfi isrnisst i nnsfsftsn ant siratt sis nas: il aasss si^sa fsfstsft i fnsfsni sfa fffsfs astati s nrsfs n nt ns fasi sas afe a ntssfs sa: i astiti suffis stfirer ag#3ai: il ^sa nstssifsamiatnisi ata ssa: a # s :15 i

1 T, afe, P, #st 8 n#: 1 T, agn, P, ara- 1 P, aast0-s P, facjfi:. " D, mita, T, aff, P, aslsr-7 D, snnir, T, nani, P, ansar:. 8 T, fsf, P, ^, D, snfiai.0 I), irpst, T, fsfen, P, fciRS. 1 0 T, l'at- 1 1 T, s ta i ;1 3 srifareti- 1 3 D, assi, T, sssrir, P, asfa- 1 4 D, arata# P, m m u15 Instead oE the reading SPSt (sanha) we have safiidhy o? sanclhy in all the texts.

I have, however, accepted the corrected reading of the word.

8 0 8TUD1HS IN TETU TANTAK

The words used in the Suulkb/iCisCi aro according to the preceding text the following 1 ;

mad ana (Ch. mo-ui,ei/) = madya ; Ch. huo die fruit-provision ? vaia1 2 (Ch. -

SANDHBSA AND SANDHAVACANA 31

rajakl (Gh. Ian-jo-ki/i) Kamakuh ; Ch, ]de-mo-pon, karma dass.[Mudr (Ch. mu-to-lo) = ? miao leiten." ; 1 Ch. "will achieve i.e.

susitldlndh,1*]These are not, however, all the words of the Sanclkvaeam class, as we

have many others mentioned in different places either iu the commentaries on the Deltas or in the lleouj, alani ra. I will only mention a few of them here__

ali, haltthese two words literally mean vowels and consonants. See Baoot, La grammaire de Thomm Sambliota,p l , ali leali camini snrya putj'fwpyaIlewjni (Chap, I ); lohapmuena . . .lokbhsena ca Ourycari/avinimiya (com., p. 15).

Inland, rasane, avadhtvthese are the names of the three principal of the thirty-two arteries lalaii prajfisabhvena rasanopyasuinsiliit | (ivaiUili mudlyadese lu gr&hyagrhahavarjit h IJevajm (Ch. I).

Some of these technical wordsbodhicilla, samarasa, harm, giti, evarh- Mm, etc., have been explained by M Shnbididluh with the help of the commentary in his Dos Chants Mystiques (Paris, 19&8, pp. 9-10).

In the chapter {palala) which follows the preceding one ( T indili thanma patalah, Chap. 1 of Part II of our \ISS. of the lleoajra = Chap 14' of the Chinese translation, Tokio, X XV II, 4, p. 74a) the Lord explains some doctrines m the Sandhbhm. The passage is faithfully Iransciibod in the Chinese translation (ibid, p. 75a, 11-13). We will first give the original text as collected from the three different manuscripts already mentioned and then the Chinese transcription, as given iu the translation .

fhHX 13^ 1 WNTITO ftffts f 3PTO 3rf5tfa TO KfaT ||cif to isT^r *t o i fftTO I

qPuTO: ^ N to h

TOTOf fWt RTO tlTO I

*tT!3 Wffi gitiuf% w WI4 II TOm lite 3TO ^ fllfalTO I

fqs TO TO^TO TOKlft UpflTO I

VPSTTO TpS TOS ^ L^lTO IIs

1 Muto tcheng in the Chinese text is wrongly taken as an equivalent of the word mudr. The original text means that the five mudmsdombi, nati, caudali, dvtj und rajah! conduce to perfect stddlit.

D, T, nrtsrrl ,^ -D, P, f%fro, D, vnsrt ; T, ; P, fvn? ;I), to*T steiT, T, TOtflll D,P, ^i , T, Olnnese , T, tsl^l , D,P,

32 .STUDIES IN THE TANTRAS

Chinese transcription (p. 74a, 17-20).

kal-lo-yi-li che-a mo-l mou-m-li k-ko-lo | km ki-pi-cha ho wei-jo-yi lne-lou-ni-ki a-yi-lo-l || ta-hi t&o-lo kie-jo-yi ng-che mo-ye-n pi-jo-a-yi | ho-lei k-leng-jo-lo po-ni-a-yi uei-nou-ion wei-eul-a-yi || tso-wu-san-mo kie-chtu-li si-lo+h kie-pou-lon lo-yi-a-yt | mo-lo-j1 yin-ta-na so-lei ta-hi po-lon h-yi-a-yi || pi + leng-kie-na kie-cha kic-lcng-ti shou-ta nou jo-ni-a-yi | m-lien-shou ying-ye tso-a-wei-a-yi

Lan-lii ji-sa-l pou-a-wei po-ni-a-yi | mo-lei-yi-ji "hong-no a-lou we.-oha-yi

nhiQ-ni-mo to-hing wei-jo-a-yi | |1

Most of the words in the above passage belong to the Bandhbhm, The other vvoids are in their Afabiama forms [ha, to stay, from stM ; bjjat" to beat (the drum) fiom v&tlya,', Vuijjai " to eat from 1ihiula ; pijiat "to drink" fiom piva ; fa tim i, "to bow into" f rom pra-nama ; bajjiai " to abandon " from uarja ; latat " to take " ; k/iaiai, Itaraule, mimmi, cambiai, pawat, baiai, etc., also aie to be explained as ApabJuamba forms which we find in the Dohiis. The Saiulhvaeaita in the passage are :

halrhola =padma "lotus."kibida from hr pita, = damarnka, i.e., "drum.bala = mama, meat.maane from viadana = madjja, u wine."

fqfam , T, f i lm , D, ^ , D, Rifqrast, D, ; T.T. y D, T T5 ,P, , D, omits , T, =(TWR;=fr, p , f f , T, f q f , p , 5rrsn, D, TremiS W , D, Ultore , T, T lfo w tf f , P, T TfT T fe, D, T O T ; TOT; P, 5^T ,Ch. Birre ; D. tT , T, res , P, igz , D,T, , D, T , T, T t Ptos ;P, T TTtam , D. , P, fTKl? , D, m , Cil, TTRtT , P, TTTTS ; P, T'grfTTi ; D, TTTlk'q^T ; T, grercrequcrs , P, TTXfq , D, Tqf ; T, TTTIT, D, Tis?, P, sis ?, D. T, stfs, P, nf?, T, P, ttH s 1

1 net, nou, w , rung, m stand 01 ancient ndei, ndou, nda, etc , enl for ya, there is often confusion between mo and wet, for the Chinese characters aio similai. The vowels are lengthened according to the indication in Chinese. The combination of two consonants has been lepresented by the plus sign.

SANDHABHASA AND SANDHAVACANA 33

hlinjara= bJiavy aduddnra= ab/iavya(tasama from catusmma = rjnthasilha from silhaha = svayamb/t haryipra= snjctamalai indhana vy apatiaprenlchana = yatilihtta=ffaniramma = asthylharanamalayaja milam "union, this word, as we have seen is

omitted in the Chinese translation. The meaningis the same as that of hnnduru which means dvindriyasamyoga.

di ndima ~aspar sa.

I do not, however, pretend to say that by the substitution of these equivalents we uan arrive at a clear meaning; the meaning will still romain obscure. That obscurity can be removed only when the text will be interpreted in the light of the Vajrayna doctrines. Unfortunately our knowledge of the Vajrayna is still very meagre.

ON TUE SDII ANAML

The present volume which forms the Tome XLI of the Gaekwatls Oriental Series, contains the remaining Sdhanas, 142 in number (Nos. 171-312). Dr. Bhatfcaoharyya is to be congratulated on the successful completion of his edition of the Sdkanaml. Most of these Sdhanas are lying scattered in Tibetan translation in the jBstan //gyur and the worth of the edition would have been certainly enhanced if they had been compared for settling the text. But still our indebtedness to the editor is not minimised in the least, for he is doing the work of a pioneer in this field. The greatest importance of these texts is ieonographical and ritualistic. Dr. Bhattacharyya in his Indian Buddhist Iconography has done justice to the ieonographical aspect of the work, but the ritualistic aspect still remains to be studied.

In a learned introduction the editor has discussed some of the most important problems of Buddhist Tantras bearing on the present text. This introduction deals with : (1 ) Magic in ancient India in which he finds the origin of the Tantras j (2) Tantras in which he gives a general description of the Hindu and Buddhist Tantras ; (3) Origin and development of Vajrayna; (4) Chronology of the Vajrayna ; (5) Leading tenels of Vajrayna ; (fi) Aims and objects of the Tantras ; (7) Authors of the Sdhanas ; (8) Vajrayna deities j (9) Iconography. Our remarks will be eonfined to the most important of these sections, viz., 3, 4 and 7, which deal with the history of the Vajrayna.

It is not true to say (p. vi) that Buddha gave instruction on Mudm, Mandala and Tantra. The oldest images of Buddha, of course, represent him with Midr but thoy go back only to the 1st and 2nd centuries A.D. There is nothing to support the statement that Buddha incorporated Tantric practices (p. xvii) into his system of religion. There is no work on dhians translated into Chinese early at the beginning of the Christian era. The oldest translation of something like a dhran, which I know of, belongs to the beginning of the 3rd century A.D. (Nanjio, Catalogue 478 is lost. Another dhran is said to have been translated between 223 and 253 A.D. by an Indo-Scythian monk, Tche Kien (Nanjio, No. 355 Anantamuhha-sdhaliadhran ?). But there is an interval of 100 years between these two translations and the translation of dhrans, which took place towards the end of the 4th century A.D. Both the translations are registered for the first time in the catalogues of the 6th century A.D.,

ON THIS SADIIANAM AI jA 25

and tliorefoie doubts can be easily raised about their authenticity ; it can be, however, safely asserted that the genuine translations of d/iaratyis go back to the end of the 4th and the begiuning of the 5th century A.D. It may be pointed out in this connection that the chapter on dhran in the Lanknalrasnlm is not found in its first Chinese translation made by Gunabhadra in 4,43 A.D. but occurs only in its later Chinese translations.

Some interpretations of the Vajrayna doctrines given in 3 cannot be navely accepted at the present state of our knowledge about Vajrayna. On p. xxii Dr. Bhattacharyya says,Vajraynists went beyond due limits in thoir spite against the strict rules of morality, and they violated all of them and plunged headlong into the worst immorality and sin. This conclusion is based on a literal interpretation of some of the verses of the Prapi02i/mmicagasddJn of Anangavajra, edited by Dr. Bhattacharyya himself. The verses in question are the following :

utrmffraT tail ^!vt i

ferir sran m u

Kiwi T O W R i m u

strerofiiflvraT ? ''rawlwt i i^sflerf ut[wpit ? ] ^ % iut fount

36 s t u d ie s in t h e tantras

Dr, Bhattachavyya himself (p. xliii), contains many a clue to the interpretation of the VajraySna doctrines. Let us try to see how the terms 61551, U5T, etc,, on which the interpretation of the versos depend, are explained in this Tantra, In the very opening chapter of the text called Vaj'rakula- patala we find a description of the 32 niids (arteries) of which three are the principal, visi., lalan, rasano, and aw d/adi, corresponding to the id, fingala and snsumi of the Hindu Hatshayogins. The nature of these three nds is further defined in the same chapter as :

53^ g ru h

The nature of lai ana is, therefore, Prajh and this Prajn becomes an active force when lalan is purified. I think it is in this light that we have to interpret the first two of the verses quoted where Prajnjpramil is said to be existing in the forms of lalan (lalanrEpam stga...). What is then BrJmanclikulotpann minim ? Dr. Bhattacharyya knows (p. lx) that there are five hulas (families) representing the fivo clhgn Buddhas Altsobhya, Vairocana, AmitSbha, Ratuasambhava and Amoghasiddhi. For further explanation we have to refer to the 5th chapter, Taitvapatala, of the Ilevajratantra where the live mudrs, Dombl, Natl, ltajakl, BrShmani and Candll are said to be belonging to the five kulas.

55T ira iwirsrtffi g*rt i5 # is t flnifiuftat II55* to' usti toito 5 1 i r a i f r T O f w v r c r f t t o t s t 11 T O '

UN Till! SADHANAMALA 37

It follows from tins that the live muclrs originating from the hulas beginning with Brhmana ( Brhmandi/culotlbhtctm) are respectively : Brhnani=Tathgat, Can4=Ka.tiia, J)om b=fajm , Rajahi= Karma, and N a ti= P a d m . What is MudrS? Mudr is explained in the same chapter of the Hevajra as these are,therefore, different gestures of the fingers by which the Yogin pretendsto evoke the Vajra, i,e,, Prajna' (Fajram tiiudryale anetia,......). Til udr istherefore not a woman.

The same chapter of the Hevajra explains the different words etc., thus :

AST amfcl fl'wrafflat IPinlfrr cfstr urn fww i.......j jw

38 STUDIES IN THE TANTRAS

forms in the Sanskrit texts whereas the second falls back on Odi and Odivia, i.e., Odra and Aiiruvianyu ( = Orissa). The latter is generally transcribed by the Chinese writers as Wu-lch'a, i.e., Uda (the phonetic value oi teha=n\a) and tho former sometimes as Yue-tt-gen {i.e., U-ti-gana). Wu-tchang is the older form and is based on *U-dia>ig (the phonetic value of tchangdang) winch was wrongly translated as garden5' and hence restored as Udyna. Hiuan tsang transcribes the name as llOi-tehang-iia, i.e., U-diang-na. I do not know if M. Levi has identified this country with Kasgarh (and Dr. Bhattacharyya does not give any reference) but on the contrary M. Levi has located it in the Swat Valley ( / . As. he catalogue gographique des Taksa, pp. 105-112) as all the available sources of information wonld indicate. M. Levi has adduced good grounds for this identification and it will suffice to reproduce some of them here.

(1) All the Chinese sources (Fa hien, Hiuan tsang, etc.) locate Uddiyna in the Swat Valley. M. Fouclier {Iconographie Bouddhique, pp. 121 and HS) has drawn our attention to the miniature of a Nepalese IviS. of the 8th-9th century A.D. which bears the inscription Yajrapni of: Mangakostha in O.JrJiyna. Mangakogtha is only a different name of iVTangalapnra {Mong-hie-li of Hiuan tsang), tho chief city of the Swat Valley. (3) In the more ancient Tantras, for example the Ilevajrntantra (7th patata), the order of the p it has is as follows :

igir Sfinir C1*N ^ I ifts rpnfkfV ^ n

So if OddiySna be at all near any place it was certainly not near Ktnarpa but near Jlandhara (4) In the Bomakasiddhnta Uddiyna is enumerated along with Sindhu-Surstra. (5) In the Thing annals (Chavaunes, Documents, p. 160) the boundaries of Uddiyna (Yue-ti-gen) are given as follows : India is on the South ; Chitral is on the North- West and it is situated to the North of the Indus (6) As regards the antiquity of the form OJiyna one may refer to an inscription of the year 77 of the ICushan era (Liiders, List, No. 62) which records the gift of a monk Jlvaka, a native of Odiyna. No authentic source of information of this period mentions Oda {i e., Odra) There can be, therefore, no doubt that Uddiyna is the same as the Swat Valley and that Odliyna, Udiyna Uddiyna O-rgyan, 0-(Ji-ya-na, U-rgyan, U-rli-ya-na and Wu-tchang and Yue-ti-yen are all different forms of the same name and are quite distinct from Orissa.

Dr. Bhattacharyya may still demand satisfaction on three more points : li) where was then Zahor, to the royal family of which intaraksita

ON THE SADHANAMALA 39

belonged ? As Indrabhfci, the king of Udyna married the sister of the former at Zahor, Zahor must be in the proximity of Uddiyna. (vi) Where is Lankpurl of which Jalendra was the king? As Jalendras son married the sister of Indrabhti, Lankpurl must be in the same zone as Udcltyna, (in) According to the Tibetan sources Lui-pa was an employee of the king of TJdrJiyna. But how is this that songs attributed to him are written in Bengali ?

Zahor is mentioned in the Tibetan sources in connection with the countries which Indrabhti visited after leaving Ultjiyna. He visited the cemeteries of Bidclha 1 country, a particular cemetery in Ks mir and another in Nepal and last of all the cemetery called Lanka in Zahor (Waddel, Lamaism, p. 382), Nobody has the right to separate any of these names from the context in which the mention of Ksmlr points out, beyond all doubt, that Zahor visited by Indrabhti is the borderland between Ksmlr and Nepal which is not very far from UcldiySna. The identification of Zahor with Mandi as proposed by Franche (Indian Tibet, Vol. II, pp, 65, 89-90) is therefore quite correct. If one refers to pp. 419 ff. of the Punjab Tribes and Gastes, Vol. I, he will be certainly surprised to see the persistence of the old beliefs in the country of ntaraksita. Lakpur! is a cemetery in Zahor. But we have a different description of Lankpurl in the history of the 84 Siddhas, where it is said that the country has two divisions, one is 8 ambiala of which Indrabhti was the king and the other Lankpurl of which Jalendra was the king (Taranatb, p. 325). The Tibetan accounts very often are confused and it is not uncommon that in them simple cemeteries have been converted into monasteries (see I.H .Q , Vol. V, p. 763 n.). Leaving aside the cemetery of Lankpurl in Zahor it may be admitted that a certain locality in UiJdiyna was also called Lankpurl. But the identification of Zahor with a place in Assam as proposed by Dr. Bhattachaiyya does not satisfy even his own position as he has been obliged this time to locate (Jikliytia in Assam ! There are, however, indications which justify us in thinking that some locality in North-western India was known as Lat'ik. Jayabhadra, a translator of Galerasamlai alantra into Tibetan, is said to have been a man of Lanka, also written Langa (Cordier, Catalogue, II, pp. 42-4:1), Ceylon was certainly not known as Lanka in this period and Oakra'sam- baratantra had probably nothing to do with Ceylon. On the contrary there 1

1 I am not ama about the identification of this place with Videha (?) piopoged by Waddel. I t may be very well pi-t'u (Binda) which Fa-lnen visited just after crossing Iho Indus.Cunningham, Geography, p. 178.

40 STUDIES IN THE TANTRAS

is ample evidence to show that the oui turo of ambaralaiitra was intimately connected with Sambhala country which is said to have been a part of Uddiyna. Lanka of Jayabhadra, therefore, seems to have been the same as tho Lanhpur of Jaleadra. Hinan tsang (Watters, IT, p. 257) speaks of a country of Lang-lue-lo in the lower valley of the Indus, where there weie 100 monasteries with 6,000 monks of both Hiuayna and Mahyna in his time. This name has been connected with that of the Langga tribe that still lives in the north of Baluchistan. This Laiiga tribe is distributed in different districts of the North-west and classed as a Jfc tribe in the districts of Dera Ghazi Khan where it was probably aboriginal or immigrant from eastward (Punjab Tribes and Castes, II, p, 30). Considering the immigrant nature of this people, it is not probably too much to think that they once occupied districts contiguous to the Swat Valley further to the north-west, and that their country was known as Lanka, The other form of the name of the native place of Jayabhadra, Lwhga, is to be taken notice of in this connection.

Now the last objection about the identification of Ut.ldiyna, remains to be answered. How could Lui-pa belong to Uddiyna and be a Bengali at the same time ? This question is rather complicated because it is connected with many others. The name of Lui-pa is in Tibetan Sa-lto-pa, i.e., Matsynti'da. Though Cordier ( Catalogue, II, p. 33) hesitates to take him to be the same as Matsyendranlha, lie cannot adduuu any plausible reason for doing so (see also, S. Lvi, Le Npal, I, p. 358, n. 4). I refrain from discussiug this problem for the present as Prof. Tacci has dealt with it in a paper to be shortly published in the Journal of the Asiatic Society of Bengal.1 It suffices for me to say that the mystic cult connected with the name of Matsyendrautha we ut far beyond the limits of Bengal and was in vogue in many distant parts of India. There is theiefore no wonder if his name is associated with both Uddiyna and Bengal. Moreover while studying the history of these sects we cannot overlook the importance of the great community which the Yogins from remote parts of India formed in ancient times and do still form. Nor can we overlook the fact that these extraordinary people coming from different parts of India still meet in the inaccessible shrines of the Himalaya to communicate their secret doctrines to each other and thus maintain the solidarity of the religion of which they are the representatives.

Dr. Bhattacharyya is to be specially congratulated for sections 4 and 7, Chronology of Vajrayna and the authors of the Sdhanas. He

Cf. J.A.S.B., 1930,1, pp. 132 IT.

ON THE SDHANAML. 41

has tried to collect available materials on the history of the teachers of Vajrayna and find ont an acceptable chronology, This chronology may have to be modified in the light of future researches but still a beginning had to be made somewhere. There is room for supplementation, but I refrain from it as that will exceed the scope of a review. I will however content myself in pointing out that Asanga, the author of Sdhana No. 159 cannot be identified with the great teacher of Yoga cara unless it is proved that the complicated ritualism described in this Sdhana already originated in the 4th century A.D.

One of the Sadhanas (No. 127), of which the importance has been reoognised by Dr. Bhattaeharyya himself (pp. cxxxv-oxliii), is of great historical value. It is the ISkyatsdhana, which is said to have been recovered by Arya Ngrjuna, (i.e., Siddha Ngrjuna, circa 7th century A.D.) from the country of Bhoba, i.e,, Tibet (rya-Ngr)iena-pdaih Bhotem uildhftam). In fact there are six sdhanas (Nos. 123-137) devoted to the goddess Ekajat. The description of Ekajat, as given in these sdhanas, closely agree with that of Mahclmltramatr iu Sadhanas 100 and 101. A comparison of these two goddesses show that they are essentially identical, the only diffeience being in the bija mantras, in the case of Mahacinahra- maitmi it is composed of three letters (tryaksarl vidy : orh Jinrn hum, of. Sdhana No. 101) whereas in case of Ekajat it is sometimes composed of 4 letters (om hrirh trim hierh, cf. Shanai 123, 125, 126, 127 and as hnin, trim hum phot in Sham 124) and sometimes of 5 (omhrim trim kuThphat, of. Sham 124).

Corresponding to these goddesses we find in the Hindu pantheon not only Tr, as supposed by Dr, Bhattaeharyya, but also Ugratr (of whom the worship was introduced by Vasistha and hence the same as (Mahciua- tra), Bleajal and Mahnlasrasvat. The dhyna of Tari) quoted by Dr. Bhattaeharyya on p. cxxxix is the same as that of Nllasarasvatl, as stated in the Phsthr Tantra (see Tantrara, pp. 514f.). Further it should be noticed that in Sdhana 101 it is said that Clnatr is to be worshiped in :

u w . il

In Sdhana 123 it is said that the Yogt should meditate on Ekajat in solitude, cremation ground, and street corners ( ) Thesame is true about Nlasarasvat. In Tantrasra (p. 506 quoting from

6

42 STUDIES IN THE TANTRAS

the Phetkri Tantra) practically the same verso as found in Sdhana, p. 101, is quoted with the interpolation of two lines :

Wfa# vpstr 3T vj-qutft SfT'r 1[ iTOfaft: *n s if *n gpisgftl ^TW3?fl sfai ^n*l *n IJ

W fteat il

The Tantrasra quotes from a number of authorities to explain the difference between the goddesses described before. Nllasarasvatl is

^ifarrcifl*iwtr (Nllatantra). She is a PaOcksan-vidy ; she is Ekajat while she is separated from Tara (i.e., the pranava) ; she is Nllasaiasvatl while in unison with Tr; and she is Ugratr while she is a nily of three letters. The presiding goddess of other vidj/s is Ekajat because Ekajat is her veiy nature.1

It is therefore evident that the Hindu Tantras considered Ekajat, Nllasarasvatl and Ugratr as different aspects of the same goddess. Buddhist Tantras do not speak of Nllasarasvatl but mention M. abaci na- hramatr instead. Both Ekajat and Mahclnakramatr are of foreign importation. The Hindu Tantras all have preserved this tradition. The hints of this is already given in the verse quoted above from the Nllatantra where Nllasarasvat is called Sarcabhfjma/jl, knovvor of all languages and Sarvmni/air tiattiaskrt, worshipped in all the traditions. The story recorded in the TrStantra (to which attention was first drawn by H. P, SsstrlNotice oj Sht. MSS., and fase. Vol. I, 8, p. xxxii and 152) describes how Vasisfcha went to Mahcna to get instructions from Buddha. He brought from there the cult known as Mahclnalci amacara which prescribes the worship of Mahclnatr, If there is any truth behind this