BUDDHIST DEITIES AND MANTRAS IN THE HINDU TANTRAS

GUDRUN BUHNEMANN

BUDDHIST DEITIES AND MANTRAS IN THE HINDU TANTRAS:

I THE TANTRASARASAM. GRAHA AND THE IS ANASIVAGURUDEVAPADDHATI

1. INTRODUCTION

At various stages in its development Buddhism incorporated Brahmanical and Hindu deities, but in its Tantric form Buddhism has also influenced the Hindu pantheon. The Tantric period is characterized by mutual influences between the two religions.

A. Sanderson has provided evidence for the influence of the Tantric Saiva canon on the Buddhist Yoganuttaratantras or Yoginıtantras.

Sanderson 1988, pp. 146–147 and 1994, pp. 94ff. demonstrates that passages from the yet unpublished Saiva Tantras, such as the Brahmayamala (Picumata), the Tantrasadbhava, the Yoginısam. cara of the Jayadrathayamala and the Siddhayogesvarımata, were incorpor-ated with little or no modification into Buddhist Tantras of Sam. vara, such as the Laghusam. vara (Heruka bhidhana), the Abhidhanottara, the Sam. pu.todbhava, the Sam. varodaya, the Vajradaka and the D. akarn. ava.

Sanderson shows that it is unnecessary to explain existing similarities between Tantric Saivism and Buddhism by postulating a common source (often referred to as `the Indian religious substratum') from which the two traditions are assumed to have derived. Addressing the influence of Brahmanical iconography on Buddhist Tantric iconography, Banerjea 1956, pp. 558–561 highlights similarities between the forms of Siva and the Bodhisattvas Sim. hanada, Nılakan.t.ha and others. As is well known, Buddhist Tantric texts such as Abhayakaragupta's eleventh-century Nis.pannayogavali (NY) include Brahmanical deities, such as Ganesa, Karttikeya, the directional guardians and heavenly bodies, in the periphery of the deity man. d.alas they describe. The reverse, namely the influence of Tantric Buddhism on the later Hindu Tantric pantheon, is studied by B. Bhattacharyya.1 However, Bhattacharyya 1930, p. 1277 and 1932, p. 109 goes too far when he draws the general conclusion that the Buddhists were the first to write

Indo-Iranian Journal 42: 303–334, 1999. ^c 1999 Kluwer Academic Publishers. Printed in the Netherlands.304 GUDRUN BU¨HNEMANN Tantric texts and that the Hindu Tantras are borrowed from the Buddhist Tantras. Bhattacharyya addresses not only the iconography but also the deity mantras on the basis of such texts as the Sadhanamala (SM).

He concludes that Chinnamasta and the eight manifestations of Tara known as Tara, Ugra, Mahogra, Vajra, Kalı, (the Tantric) Sarasvatı, Kamesvarı and Bhadrakalı were adopted by the Hindu pantheon from Buddhist Tantric sources (1930, pp. 1278–1279 and 1932, pp. 148– 149, 156–157). I summarize our present state of knowledge on the adoption of Chinnamasta into the Hindu pantheon in Bu¨hnemann 1999, section 1.6.3.3. In Buhnemann 1996 I demonstrate on the basis of textual evidence how a form of Tara, called Mahacınakrama-Tara, was adopted from an eleventh-century Buddhist Tantric sadhana by Sasvatavajra into the Hindu Phetkarin. ıtantra. The Phetkarin. ıtantra's description of the goddess as Ugratara – along with her surrounding

deities and elements of typically Buddhist Tantric worship procedures and mantras – became the authoritative description of the goddess and was incorporated into many Hindu Tantric texts, such as Kr.s.n.ananda's Tantrasara. Bhattacharyya 1933 traces the adaption of Bhutadamara into the Hindu pantheon by examining the two extant Bhutad. amara-Tantras, one belonging to the Buddhists and the other to the Hindus. Bhattacharyya 1930, pp. 1295– 1296, 1932, pp. 161–162 and Pal 1981, pp. 102–104 examine the adoption of Man~jughos.a by the Hindu pantheon. It is usually not easy to determine when and how a deity was adopted from one pantheon into another. The importance of the cases of Bhutadamara and Ugratara lies in the clear understanding they offer of the process of adaptation of a deity from the Buddhist Tantric texts into Hindu Tantras.

In this paper I am concerned with the influence of Buddhist Tantrism on Hindu Tantras as evident from the adoption of Buddhist deities, mantras and elements of typically Buddhist Tantric worship proced-ures. Many Tantric texts have not been edited at all or at least not critically. Often we have no information about the period in which they were written nor who their authors or compilers were. Since much work remains to be done before one can attempt to draw conclusions of a more general nature based on primary texts, it seems best to begin with a study of select Tantric texts. Part One of this paper examines Buddhist influences in two closely related texts, the Tantrasarasam. graha and the Mantrapada of the Isanasivagurudevapaddhati. Part Two will address Buddhist deities and mantras in two later compilations, the Srıvidyarn. avatantra attributed to Vidyaran. ya Yati and Kr.s.n.ananda A gamavagısa's Tantrasara. I will not discuss the origins of these deities nor address questions as to whether they were originally tribal or folk

deities who were assimilated into the Buddhist pantheon. I use the term `Hindu Tantras' instead of Saiva, Vais.n. ava or Sakta Tantras/Agamas to indicate the non-sectarian character of most of the texts I examine.

The Tantrasarasam. graha (TSS) is a compilation of mantrasastra by Narayan.a, a Kerala Brahmin who resided in Sivapura on the banks of the river Nila. He was the son of Narayan.a and his wife Uma. The work, which is called a Tantra in the colophons of the chapters of the text (e.g., 32.67d, 70c), is divided into thirty-two chapters. It is popularly known as the Vi.sanarayan. ıya, since its initial chapters (2–10) deal mainly with mantras to counter the effects of poison (vi.sa).2 Goudriaan 1977, p. 160 states that the TSS is not the Narayan. ıya quoted by Raghavabhat.t.a in his commentary Padarthadarsa on the Saradatilaka (ST); however, this turns out to be incorrect. The work referred to and cited by Raghava is indeed the TSS.3 Since Raghava completed his commentary in 1493, the TSS can safely be assigned to the fifteenth century (eliminating the possibility of the sixteenth century)4 or earlier. The TSS was printed with an anonymous commentary (vyakhya) which cites the Mantrapada of the Isanasivagurudevapaddhati.

This Mantrapada (MP) forms pada 2, pa.talas 15–52 of Isanasivagurudevamisra's Isanasivagurudevapaddhati (ISP), which is also known as the Tantrapaddhati. The ISP is a Saiva manual of temple worship in four padas and is assigned to the last part of the eleventh (Dvivedi)5 or to the twelfth century (Unni 1987, p. 9). One should, however, distinguish between the ISP and the MP inserted into it. At least part of the MP seems to be of a later date than the other parts of the ISP. Since MP 47.20cd is cited (without specifying the text's name) in Raghava's commentary, p. 865, 22 on the ST, the MP must have been extant in its present form before 1493. Unni 1987, pp. 38–65 summarizes the contents of the MP, and Dvivedi 1995, pp. 189–190 supplies a list of the texts quoted in the ISP.

Most chapters in the MP correspond to chapters in the TSS. The relationship between the MP of the ISP and Narayan.a's TSS is discussed in Goudriaan 1977, pp. 158–160 and by Goudriaan in Goudriaan/Gupta 1981, p. 128. Goudriaan considers it possible that either chapters 15–38 of the MP are recast and shortened in the TSS, and chapters 39–52 of the MP are based on the TSS, or that the MP and the TSS are based on one source. In addition, Unni 1987, pp. 19–22 has demonstrated that chapter 41.2 of the MP clearly refers to the TSS (cited as the Narayan. ıya) and that chapter 49 of the MP draws on chapter 30 of the This confirms that the TSS is among the sources of the later part of the MP. V.V. Dvivedi 1992, p. 35 draws attention to the fact that the ISP describes the worship of the Buddhist deities Vasudhara and Yamantaka.

The two sections are reprinted without changes in Dvivedi 1992, pp. 36– 41 (= MP 26.1–64) and 1992, pp. 42–44 (= MP 47.1–39) from the only available edition of the MP. Dvivedi inserts the titles Vasudhara-sadhana and K.rs.n. ayamari sadhana, which do not appear in the text of the MP.

Dvivedi 1995, p. 184 believes that the mantra of Yamantaka in the MP is taken from the Kr.s. n. ayamaritantra (6.13). He does not discuss possible sources for the description of Vasudhara and her worship.

Taking Dvivedi's discovery and his brief discussion of it as a starting point, I have identified additional material of Buddhist origin in the MP of the ISP and in the TSS, which Dvivedi does not consult. In this paper I discuss the adoption of the two-armed earth goddess Vasudhara; of the god of wealth, Jambhala, who is widely known as the Buddhist counterpart of Kubera; and of Yamantaka and his mantras. Several other Buddhist mantras in the TSS and the MP are included in an appendix.

2. THE EARTH GODDESS VASUDHARA LAKSMI

The TSS and the MP extensively describe the mantras of the earth goddess Vasudhara and their ritual applications. The goddess's name Vasudhara means “a flow of wealth,” and is suggestive of her being a form of Laksmı. This is indeed supported by her classification in the two texts.

The sections TSS 22.19–41 and MP 26.1–64 draw heavily on Buddhist material. The seer (.rs. i) of Vasudhara's heart mantra om. vasudhara svaha (MP 26.5+) is specified as the Buddha (MP 26.5a) and the mantra's presiding deity is Vasudh ara Laks.mı/Srı. The practitioner is instructed to bow to the Buddhas and the Bodhisattvas in the beginning of the worship ritual (MP 26.8ab; TSS 22.20c). The five Buddhas are listed in MP 26.8cd–10 as Aksobhya, Vairocanaka, Ratnasam. bhava, Amitabha,6

Amoghasiddhi, and the eight Bodhisattvas as Padmapani (i.e., Avalokitesvara), Maitratma (Maitreya), Gaganadigaja (for Gaganagan~ja), Samantabhadra, Yaksadhipa (i.e., Vajrapani), Manjughosa and Viskambhaka. One of the eight Bodhisattvas is clearly missing from this list, which otherwise resembles the one found in texts such as SM, no. 18, pp. 49, 12 – 50, 2, which include the names Maitreya, Ksitigarbha, Vajrapani, Khagarbha, Manjughosa, Gaganaganja, Viskambhin and Samantabhadra.

The practitioner then visualizes the seed syllable tram. , which transforms into Mt. Sumeru. From the syllable vam. (TSS, yam. MP) a lotus is mentally produced, on which the practitioner visualizes himself (MP 26.11–12; TSS 22.21ab). He then recites the widely used Buddhist Tantric mantra (cf. SM, p. 218, 8–9), om. svabhavasuddhah. sarvadharmah. (sarvasamah. MP) / svabhavasuddho (sarvasuddho MP) 'ham – “Om. , all dharmas are intrinsically pure, I am intrins-ically pure.” This recitation is followed by the contemplation of the four Brahmic states (brahmavihara), loving kindness (maitrı), compassion (karuna), sympathetic joy (mudita, TSS, sam. to.sa MP) and equanimity (upeksa) (MP 26.13–14; TSS 22.21ab+).

The practitioner recites the mantra om. sarvatathagatanam. sarvasiddhayah. sam. padyantam (MP; om. sarvatathagatah. sam. sitah. sarvatathagatanam. sarvasiddhayah. sam. padyantam TSS) / sarvatathagatas cadhiti.s.thantam (MP; sarvatathagatas cati.s.thantam TSS). This mantra is recited by contemporary Japanese Shingon practitioners as: om. sarvatathagata (sic) sam. sitah. sarvasattvanam. sarvasiddhayahsam. padyantam. tathagatas ca adhiti.s.thantam (Miyata 1988, p. 16). Our texts classify mantras according to the categories hr.daya and upah.rdaya, which are well-known from Buddhist Tantric texts, and refer to the hand gestures samayamudra and vajramudra (MP 26.15–17ab; TSS 22.23).

The samayamudra is defined, for example, in SM, p. 4, 7 and the vajramudra in SM, p. 3, 5–7. Among the surrounding deities of Vasudhara are the yellow Amitabha Buddha, the white Lokesvara (MP, Yagesvara TSS) and the dark Vajrapani (MP, Vajrin TSS) (MP 26.24–26; TSS 22.28–30). The anonymous commentary on the TSS states that the Buddha should be visualized clad in a robe as described in the Buddhist scriptures (TSS, p. 314, 16–17).

MP 26.20–22 describes Vasudhara as a yellow colour, holding a pomegranate (dad. ima) in her left hand which is resting on her left knee. In her right hand she holds a lotus which contains a small vessel showering jewels. The goddess places one foot on a vessel which showers wealth. The TSS (22.25–27ab) confirms the above description and adds that the lotus in the goddess's hands is red. The iconographic passages in the two texts are:

hemanibham. pıvarakucakalasam. candramukhım alikularucikesım. </> kan~culikangım. kuvalayanayanam. carubhujam. tanutaravaramadhyam /</ 20a> hemakirı.tam. kanakaman. imayair abharan. aih. sucinivasanagandhair </> an~citamalyair adhigatavapus. am.

kantimatım. 7 pran. amata vasudharam // 208 tis..thantım. vinidhaya�n_ghrim. vasuvars. igha.t�ımukhe / akun~cita�n_ghrija�nusthava�mapa�n. isthada�d. ima�m9 // 21 ratnavars. igha.t�ıgarbham utpalam. ca�pare kare / syandama�na�rthadha�ra�d. hyana�lanirgatavallar�ım // 22 “Bow down to beautiful Vasudh�ar�a, who resembles gold (in colour), whose pitcher-like breasts are fleshy, who has a moon-like face, whose hair resembles a flight of (black) bees, whose body (is covered with) a bodice, who has lotus-like eyes, beautiful arms, a very slender excellent waist, (wears) a golden crown, whose body is covered with golden and jewelled ornaments, with pure garments and fragrant substances (and) beautiful flowers; who stands, having placed (her) foot on the opening of a small vessel showering wealth, who (holds) a pomegranate in her left hand which rests on the knee of (her) bent foot (= leg) and (holds) in her other (= right) hand a lotus which contains a small vessel showering jewels, who is (so to say) a creeper growing out of the neck (of the vessel) which abounds in flows of riches issuing forth.”

TSS 22.25–27ab: (a�va�hayed : : : / 24a) bha�svatkan~culika�m. citravasana�m. maku.tojjvala�m / saumya�m uda�ra�m. hema�bha�m. sakala�kalpabhu�.sita�m // 25 vasuvar.sigha.tastha�n_ghrim. 10 va�maja�nva�ttapa�n. ina� / vahant�ım. da�d. imam. savyapa�n. ina�11 ca�run. otpalam // 26 ratnavars. igha.t�ıgarbha<m. > na�lanirgatavallar�ım / 27ab “(One should invoke [Vasudh�ar�a]) who (wears) a shining bodice, beautiful garments, a blazing crown, who is placid, generous, has a golden lustre, is adorned with all (kinds of) ornaments, whose foot rests on a vessel showering wealth, who carries a pomegranate with her (left) hand which is placed on her left knee and with her right hand a red lotus which contains a small vessel which showers jewels, who is (so to say) a creeper growing out of the neck (of the vessel).

According to MP 26.27+ and TSS 22.31–34ab, Vasudhara Laksmı is accompanied by Srı and Maha srı and the seven jewels (ratna). But in MP 26.48, which addresses another mantra of the goddess, Vasudh�ar�a is accompanied by Dhane�svara, the god of wealth. This is reminiscent of the Buddhist Tantric Vasudhara, who is often described as the consort of Jambhala, the god of wealth (e.g., SM, nos. 285 and 289). In MP 26.49+ the goddess's surrounding deities are listed as Dharma, Buddha, “Sahya” (corrupted for sam. gha), Sagaranirghos.a, Vajraraksa (v.l. Vajrayaksa), Tarun.in, A� ry�avaloke�svara, Prajna, Sarasvatı and “all Bodhisattvas.” The description of the worship ritual of Vasudh�ar�a in the two Hindu Tantric texts is clearly based on Buddhist material. The elements of the practitioner's visualization pattern summarized above are charac-teristic of Tantric Buddhist sa�dhanas. In addition, the names of the deities surrounding Vasudhara leave no doubt that the above ritual application of Vasudh�ar�a's mantras is taken from a Buddhist source.

The iconographic description of Vasudh�ar�a in the two Hindu texts most likely based on Buddhist sources as well. It is unlikely that the MP and the TSS would have replaced the iconographical description of the Buddhist goddess with a description of the earth goddess from their own tradition while adopting the Buddhist goddess's mantras and their ritual applications. In the Hindu tradition, the earth goddess is also known by the names Vasudh�a (S�T 15.138) and Vasundhar�a (cf. Agni-Pura�n. a 120.38–39); however, iconographic descriptions of Vasudha�/Vasundhar�a differ from the one found in the MP and the TSS.

However, the goddess's iconography does not correspond to that of Vasudha�r�a described in the rather brief sa�dhanas in SM, nos. 213–216, according to which the goddess holds a rice shoot (dha�nyaman~jarı�) and displays the wish-granting gesture. It also does not correspond to the iconography of a (six-armed) Buddhist Vasudha�r�a recorded in the Vasundharoddes�a whose text is discussed in both De Mallmann 1986, pp. 441–442 and in Bhattacharyya 1974, pp. 35–36. However, in one description found in the Vasundharoddes�a, the goddess is on a vessel of plenty (bhadragha.toparisthita�) and Loke�svara and Vajrap�an.i are among the surrounding deities of the goddess just as in the MP and the TSS. I am not aware of a textual source for the goddess's iconography. Such a source may be found once additional icono-graphic material has been made available from yet unedited Buddhist texts.

3. JAMBHALA, THE GOD OF WEALTH

The TSS and the MP describe rituals for Jambhala, the god of wealth, who is widely known as the Buddhist counterpart of Kuvera/Kubera.

The TSS presents Jambhala in chapter 26 (26.17–35) which mainly addresses Mr.tyum. jaya, a form of S�iva. Jambhala's mantra, icono-graphic description and connected rituals are immediately followed by a description of the mantras of Kubera. The MP presents Jambhala in chapter 32 (32.90–103) following the description of Candesvara, who is identified with Siva. As in the TSS, Jambhala's description is immediately followed by that of mantras and rituals for Kubera and Jambhala and Kubera share some of the same surrounding deities (MP 32.108cd–109ab).

Preceding Jambhala's iconographic description in the two texts are instructions for the worshipper's visualization. Jambhala is visualized on a lotus on which a hexagon-like man. d. ala is inscribed. From the seed (bıja) syllable hrım. a lotus appears; on top of the lotus, from

the syllable am. , a moon disc appears; then appears, on top of the lotus, from the syllable jam. (which is Jambhala's seed syllable) the deity Jambhala (MP 32.101; TSS 26.31cd–32ab). This visualization follows patterns found in Buddhist Tantric sadhana texts. Two Jam-bhala sadhanas in the SM (nos. 284 and 288) enjoin that the Yogin visualize an eight-petalled double lotus (visvapadma), on top of it a moon disc and on it – produced from the syllable jam. – the deity Jambhala.

The MP identifies Jambhala as the lord of the Yaks.as (32.100c), an epithet which is also found in his mantra which addresses him as the lord of the great army of the Yaks.as: man. ibhadramaha�senayak.sa�dhipataye jambhalaya jalendraya sva�ha� (MP 32.90–91ab). A similar mantra appears in SM, no. 295 (p. 570, 7–8): namo ratnatraya�ya namo ma�n. ibhadra�ya maha�yak.sasena�pataye om. jambhalajalendraya svaha. The deities immediately surrounding Jambhala are the Yaksas Manibhadra, Purnabhadra, Celimalin, Vikun.d.alin (most likely for Civikun.d. alin or, with TSS 26.16+, for S�ibikun.d.alin), Narendra and Carendra. These six names appear to have been taken from a Buddhist source listing eight names, such as SM, no. 284 (p. 561, 5–9): Manibhadra, Purnabhadra, Dhanada, Vaisravana, Kelimalin, Vicitrakun.d.alin, Mukhendra and Carendra (cf. also SM, no. 298, p. 566, 16–20). Siddhaikavıramahatantra, p. 158, 1–4 gives the same names, but has Varendra instead of Carendra.

NY, p. 63, 19–26 lists Purnabhadra, Manibhadra, Dhanada, Vaisravana, Civikun.d.alin, Kelimalin, Sukhendra and Calendra. A slab from Ratnagiri with a relief of Jambhala gives these names as Purnabhadra, Cilikun. d. alin, Vai�srama(for: �va�)n.a, Kelim�alin, Dhanada, Mukhendra, Manibhadra and Caran.endra (Mitra 1961, p. 40). The variants in the names are limited to Vicitrakun.d.alin (SM) for Civikun.d.alin (NY), Cilikun. d.alin (Ratnagiri); Mukhendra (SM, Ratnagiri) for Sukhendra (NY); and Carendra (SM) for Calendra (NY), Varendra (Siddhaikav�ıramah�atantra) or Caran.endra (Ratnagiri). According to NY, p. 63, 27 each of these Yaks.as holds identical attributes. These are Jambhala's characteristic attributes, the fruit of the citron tree (in the right hand) and the mongoose (in the left).

A xylograph from the Rin 'byun_, prepared by Mongol artists in circa 1810 (Chandra 1991, p. 310, no. 820; see Illustration 1), portrays Jambhala embracing a consort and surrounded by eight Yaks.as, each with a consort. The mantra inscribed below the picture invokes the eight male Yaks.as as follows:

o<m. > jambhalajale<n>draye sva�ha� / om. man. ibhadra�ya sva�ha� /

Illustration 1. 820 Many-deity Jambhala from a Sa¯dhana Ekav¯ıra sa¯dhana-vinirgata Bahudeva Jambhala T. Dpah.-bo chig sgrub-las byun_ -bah.i dzambhala lhaman_, NW. 74

om. pu�rn. abhadra�ya sva�ha� / om. dhanadaya (!) sva�ha� / om. bais�ravan. aya (!) sva�ha� / om. kilima�liniye (!) sva�ha� /312 om. picikun. d. alini (!) sva�ha� / om. mukhendra�ya sva�ha� / om. carendra�ya sva�ha� /

This group of Yaks.as is also seen in a sculpture in the round excavated at N�alanda� and preserved in the National Museum, New Delhi (acc. no. 47.62). The Yaks.as are positioned around Jambhala and constitute the surrounding deities of his man. d. ala (Mitra 1961, p. 41). As described in the NY, they hold the fruit of the citron tree in their right hand, the mongoose in their left and, in addition, place one foot on a vessel full of jewels.

Jambhala is described twice in the MP and the TSS. According to the first description (MP 32.96–97) he is yellow, two-armed, seated on a white lotus, has three feet and is corpulent. The parallel description in the TSS (26.21) adds that Jambhala has three feet, three faces and (one) tawny eye. The deity's deformities correspond to those of Kubera in Hindu mythology (see Hopkins 1915, pp. 142, 147). Except for the red ornaments on the deity's body, no attributes held in his hands are described. The first iconographic description in the two texts is:

s�vetapadmasthitam. saumyam. p�ıta�bham. dvibhujam. prabhum / rakta�kalpasphuranmauliman. ikun. d. alaman. d. itam // 96 ha�rakeyu�raka.takaka.tisu�tra�dyalam. k.rtam / tripadam. tundilam. dhya�yet pu�ja�dau mantrasiddhaye // 97 “For the perfection of the mantra one should meditate at the beginning of the worship (pu�ja�) on the lord who is on a white lotus, is tranquil, has a yellow lustre, has two arms, is adorned with red ornaments, a shining crown and jewelled ear-rings, who is adorned with necklaces, armlets, bracelets of gold, a waistband, etc., who has three feet (and) is corpulent.”

s�vetapadmasthito ha�rapa.t.takeyu�rakun. d. al�ı / rakta�kalpapriyo devah. pin_ga�ks. as trimukhas13 tripa�t // “The god is on a white lotus, wears necklaces, a diadem, bracelets, ear-rings, is fond of red ornaments, has (one) tawny eye, three faces (and) three feet.

The second description is:

MP 32.102ab: (: : : smaret / 101b : : : jambhalam // 101d) b�ıjapu�ram. ca nakulam. dadha�nam. tam. caturbhujam / “(One should recall) the four-armed (Jambhala) who holds the fruit of the citron tree and the mongoose.”

TSS 26.32cd–33a: (: : : jambhalam / 32b) sab�ıjapu�ram. nakulam. dadha�nam. savara�bhayam // 32cd caturbhujam imam. dhya�tva� : : : / 33a “Having meditated on that four-armed (Jambhala) who holds the mongoose along with the fruit of the citron tree (and) who is endowed with the wish(-granting gesture and the gesture of) protection : : : ”

According to the second description (MP 32.102a), Jambhala is four-armed, but only two attributes are specified in the MP: the fruit of the citron tree (bı�ja�pu�ra) and the mongoose (nakula). This may be due to a confusion of the four-armed form of Jambhala with the previously described two-armed form. Since the two descriptions of the Hindu Jambhala appear in connection with two different rituals, it is likely that they were meant to describe two different iconographic types, but through the course of time were blended together. The parallel description in the TSS (26.32cd–33a) specifies the gestures of protection and wish-granting as the two other attributes, which makes for a total of four attributes. Since in the second description the number of Kubera's faces is unspecified, we have to assume that the deity has one face.

Buddhist sadhanas in the SM describe a form of the yellow Jambhala with two arms holding the fruit of the citron tree and the female mongoose (nakulı�), which is often said to spew forth precious stones, etc. However, the deity has only one face and not three faces as specified in the TSS's description. The Buddhist three-faced Jambhala, on the other hand, has six arms and represents a different iconographic type (cf. SM, no. 286).

Even though none of the Buddhist sadhana texts I examined give exactly the visualization pattern found in the two Hindu Tantric texts, and none of these texts give an identical description of Jambhala, the description of the deity and the rituals associated with his worship in the MP and the TSS are clearly of Buddhist origin. This is also substantiated by the fact that Jambhala is unknown in the Hindu Tantric pantheon, while Jambhala and

Kubera appear in Buddhist Tantric pantheons, often with similar characteristics.14 The adoption of Jambhala by the Hindu texts is based on material (textual or otherwise) that is yet unidentified. The following piece of information may be useful for identifying the source. The MP and the TSS state that the deity “was once for some reason injured by a wheel on his head” (the MP adds: by Hari). To alleviate his pain one should offer water libations on his head, whereby he will be pleased. This information could provide a clue as to the mythology connected with this deity.

4. YAMANTAKA

a) Introduction Dvivedi has drawn attention to the fact that the �IS�P (i.e., MP 47.11) cites a Yamantaka mantra which he believes is borrowed from the Kr. s. n. ayamaritantra. It must be added that the TSS (17.9cd–10ab) also records this mantra, which also appears in Laksmanadesika's S�T 24.18 and in texts citing the S�T, such as the S�r�ıvidya�rn. avatantra (S�VT). Moreover, it is only one of two mantras of Yamantaka found in the MP which are obviously borrowed from a Buddhist source. In the MP and the TSS the second mantra is identified as a mantra of Yama, not Yamantaka. These two texts seem to confuse Yama and Yamantaka as can be seen from the instances discussed below.

Ironically, the Buddhist mantras appear in the section of the MP which promotes the rites of black magic (abhicara) which are said to be revealed for the sake of the protection of the (Vedic) dharma (47.5a) from the enemies of the dharma and the Veda (47.1b), which include the Buddhists.

Both the TSS and the MP address the mantras and rituals for Yama/Yamantaka in the context of the abhicara rites (TSS 17.1– 29; MP 47.1–39). While the texts usually refer to a group of six acts (s. a.t karman. i), the following seven abhicara rites are listed in the TSS (cf. also Agni-Purana 306.1 with v.l) and the MP: (1) immobilization (stambha), (2) causing dissension (vidve.sa), (3) eradication (uccata), (4) liquidation (marana), (5) creating confusion or madness (bhra�nti, bhrama),16 (6) destruction (utsadana) and (7) creating illness (roga,17 vyadhi), especially fever. These rites are briefly defined in the anonymous commentary on the TSS.

The mantras of Yama/Yamantaka are to be inscribed in a yantra which is employed in black magic (abhicara). According to S�T 24.17d, this is a yantra of Pretaraja, i.e., Yama, the god of death. S�T 24.22cd specifies its use in the rite of liquidation (marana) (cf. also Raghavabhatta, p. 865, 20), while the anonymous commentary on the TSS, p. 238, 10 states that the yantra is perhaps to be used in the rite of causing dissension, since the TSS does not give precise information. According to sources from Bali which will be discussed below, the first of the two mantras is also inscribed in the squares of a yantra (Hooykaas 1973, drawing on p. 172, charts on pp. 204–205 and remarks on p. 233).BUDDHIST DEITIES AND MANTRAS IN THE HINDU TANTRAS 315 b) The Thirty-Two-Syllabled Mantra The first mantra is in Anus.t.ubh metre. Raghavabhatta's commentary, p. 866, 18 refers to it as the yamarajasloka. The version in MP 47.11 is:

ya ma ra� ja sa do me ya ya mo yo ru n. a yo da ya / da ya yo ni ra ya k.se ya ya k.se s�a s�ca ni ra� ma yah. // The version found in TSS 17.9cd–10ab is distorted and differs mainly in the first quarter of the verse: s�a ma .sa� ga la so me s�a ya me do ru n. a yo da ya // ya da yo ni ra ya k.se ya ya k.se pa n~ca ni ra� ma ya / The anonymous commentary on the TSS, p. 237, 12, the author of which is familiar with the MP, gives a version closer to the one in the MP:

ya ma ra� ja sa do me ya ya me do ru n. a yo da ya / ya da yo ni ra ya k.se ya ya k.se ya s�ca ni ra� ma ya // The mantra also appears in Laks.man.ade�sika's S�T 24.18 (last part of the tenth century or first half of the eleventh century). Its wording according to the three editions of the text is: ya ma ra� ja sa do me ya ya me do ru n. a yo da ya / ya di yo ni ra pa (ya S�T3) ks.e ya ya k.se ya va (pa k.se ya ca S�T2, 3) ni ra� ma ya //

S�VT, volume 2, p. 849, 8, quoting the S�T, gives the following reading of the mantra: ya ma ra� ja sa da� me ya ya me da� sa ja ra� ma ya / ya da yo ni ra pa ks.e pa pa ks. e pa ra ni yo da ya // The above mantra is known in Buddhist Tantrism as the mantra of Yamantaka, especially of his form Vajrabhairava, and continues to be recited in the Tibetan dGe lugs pa tradition up to the present.18 Decleer 1998, p. 296 reports that the Vajrabhairava cycle continues to be practiced under the name Mahis.asam. vara in Nepal. Contemporary ritual manuals based on older texts, such as Sharpa Tulku/R. Guard 1990, p. 66 and Sharpa Tulku/R. Guard 1991, p. 25 classify the mantra as Vajrabhairava's root mantra, a classification which is supported by two ancient Vajrabhairava texts cited below. As Dvivedi 1995, p. 184 states, the mantra appears in the (Sarvatathagatakayavakcitta-) Kr. s. n. ayamaritantra. This Tantra is referred to in Taranatha's History of Buddhism (Chattopadhyaya 1970, p. 243), along with the Trikalpa and the Saptakalpa (see below). Taranatha credits Lalitavajra (tenth century)

Illustration 2. 585 Sam_ ks.ipta Bhairava T. H. jigs-byed bsdus-pa with having brought these texts from the library of Dharmagan~ja in Ud.d. iy�ana. The mantra is referred to in K.r.sn. ayamaritantra 1.6–13 in an encoded form and appears in full in 6.13: ya ma ra� ja� sa do me ya ya me do ru n. a yo da ya / ya da yo ni ra ya k.se ya ya k.sa ya cca ni ra� ma ya

This mantra is also found in the Vimalaprabha commentary (cited in the introduction to the Kr. s. n. ayamaritantra, p. 19, note 1) on Kalachakra Tantra 4.118 with a variant in the fourth quarter, ya k.se ya cca instead of the above ya ks.a ya cca. The same version, with the variant ya ks. e yac cha, is inscribed on a xylograph from the Rin 'byun_ (Chandra 1991, p. 229, no. 585; see Illustration 2), where the deity is called Sam. ks.ipta-Bhairava.19 The reading ks. e better suits the parallelism in the verse and appears in several other texts cited below.

Note the long a at the end of the compound yamaraja in the versions of the two texts, which is ungrammatical unless it is supposed to be combined with an a at the beginning of the next word. This mantra with the same variant in the fourth quarter also appears in three Vajrabhairava texts. In chapter 3 of the Vajra Maha Bhairava Tantra, identified with the Saptakalpa by Siklos 1996, p. 9 and Decleer 1998, p. 290, the mantra is classified as the root mantra of the buffalo-headed Vajramahabhairava and appears in an encoded form. Siklos 1996, p. 36 does not attempt to assemble the syllables of the mantra, which is to be extracted from the alphabet syllable by syllable. In his translation of the relevant passage he also omits syllables.20 This mantra also appears in a text referred to as The Myth in Siklos 1996 and as

The Mythological Antecedents under one heading in Decleer 1998, p. 291, but with a short a at the end of the compound yamaraja, with the variant yac ccha (for yac ca) and the reading ni ra ma� ya (for ni ra� ma ya) in the fourth quarter of the verse.21 Lastly, the mantra is found in chapter 1 of the Three Chapter Tantra of Vajrabhairava, which is identified with the Trikalpa22 by Siklo�s 1996, p. 9 and Decleer 1998, p. 293.23 In this text we also find the short a at the end of the compound yamaraja and the spellings ya ccha (for yac ca) and ni ra ma ya (for ni ra� ma ya) in the fourth quarter of the verse. This text confirms that the mantra is the deity's root mantra.

The above mantra verse appears in Buddhist texts from Bali, which are mostly hymns of praise which the editors have named Yamara�jastavas.

They are preserved in Archipelago Sanskrit. Different versions are recorded in Hooykaas 1964, p. 63 and p. 66; Goudriaan/Hooykaas 1971, no. 815, verse 10 and no. 941 and Hooykaas 1973, p. 210 (part of a ritual application). The versions gathered by these two scholars can be listed as follows: 1ab) ya ma ra� ja sa do me ya ya me do ro da yo da ya / 2ab) ya ma ra� ja sa do me ya ya me do ra da yo da ya / 3ab) ya ma ra� ja sa do me ya ya me do ro da yo da ya / 4ab) ya ma ra� ja sa do me ya ya me do ro da yo da ya / 5ab) ya ma ra� ja sa do me ya ya me do ro da yo da ya /

6ab) ya ma ra� ja sa do me ya ya me ro do da yo da ya / 7ab) ya ma ra� ja sa do me ya ya me ro do da yo da ya / 8ab) ya ma ra� ja sa do me ya ya me no do da so da ya / 1cd) ya da yo ni ra ra k.si ya ya k.si sa nti n. �ı ra� ma ya // 2cd) ya da yo ni ra ya k.si ya ya k.si sa nta n. ı� ra� ma ya // 3cd) ya da yo ni ra ya k.si ya ya k.si sa n~ca n. ı� ra� ma ya // 4cd) ya da ya ni ja ra ma ya ya : : : sa n~ca n. �ı ra� ma ya // 5cd) ya da yo ni ja ra ma ya ya k.si sa n~ca n. ı� ra� ma ya // 6cd) ya da yo ni si ra pi ya ya pi ra si ni ra ya ma // 7cd) ya da yo ni si ra pi ya ya ksi pa n~ca ni ra ma ya // 8cd) ya da yo ni si ra pi ya ya ksi pa n~ca ni ra ma ya //

Another variant of this mantra from Bali is found in a hymn entitled Yamastava and published in L�evi 1933, p. 51, 2–3: om. ya ma ra� ja sa do me ya ya me du ru da yo da ya / ya da yo n�ı ra ya k.s�ı ya yak.si sa n~ca na ra ma ya // The first quarter of the verse, yamara�jasadomeya, also appears independently in a number of other Balinese hymns. One version reads yamara�jasa�rameya (Goudriaan/Hooykaas 1971, no. 942, verse 1a). Since Sarameya is the name of Yama's dog in Indian mytho-logy, Goudriaan/Hooykaas 1971, p. 544 and Hooykaas 1973, p. 203 (note) assume that Sadomeya is a variant of the name Sarameya. Accordingly, a wood sculpture of a dog is identified as Sadomeya in Goudriaan/Hooykaas 1971, photograph, p. 24+. In his earlier publica-tion, Hooykaas 1964, p. 67, the author suggests, even less convincingly, a derivation of sadomeya from sadodyama, “one who always exerts himself.” In these different versions of the stanza we can identify the epithet yamaraja, “king Yama,” expressions such as yak.sesas ca, “and the lord of the yaks.as,” and – in some of the Balinese versions – the word yaksi. The last word niramaya in inverted order of syllables reads yamarani (for yamaran. ı), which is the female reversal of yamaraja (Hooykaas 1964, p. 67) and appears clearly at the end of stanza 2 of hymn no. 941 in Goudriaan/Hooykaas:

ya ma ra� ja sa do me ya ya me ro do da yo da ya / ya da yo ni si ra pi ya ya k.si pa n~ca ni ra� ma ya // ya me do sa ya me do ro ya da yo da ya da yo ni / yu ra ks. i ya ya k.si ya k.si ya ma ra� ja ya ma ra� n. i // Of the versions listed above the one from the S�T cited in the S�VT is peculiar. Not only does it have some readings that are not found in the other versions (�sada�meya), but it also arranges the syllables in such a way that the second quarter of each half of the verse is the exact reversal

of the first quarter. This arrangement is termed pratilomayamaka in Dan.d.in's Ka�vya�da�rs�a 3.73: ya ma ra� ja sa da� me ya (=) ya me da� sa ja ra� ma ya / ya da yo ni ra pa ks.e pa (=) pa ks. e pa ra ni yo da ya // The inversion of syllables in a mantra in a somewhat similar fashion is attested to in Hindu Tantric texts, such as the Prapan~casa�ra (PS) (most likely tenth century) and the S�T. PS 13.55 and S�T 24.24 give the following mantra of K�al�ı and PS 13.56 and S�T 24.25 provide the following mantra of Yama. Each verse quarter consists of four syllables in regular order and the same four syllables in inverted order.

In the S�T these two mantras appear in the same chapter as the above thirty-two-syllabled mantra: Mantra of K�al�ı ka� l�ı ma� ra (=) ra ma� l�ı ka� l�ı na mo k.sa (=) k.sa mo na l�ı / ma� mo de ta (=) ta de mo ma� ra ks.a ta ttva (=) ttva ta k.sa ra // Mantra of Yama ya ma� pa� .ta (=) .ta pa� ma� ya ma� .ta mo .ta (=) .ta mo .ta ma� / pa� mo bhu� ri (=) ri bhu� mo pa� .ta .ta r�ı stva (=) stva r�ı .ta .ta //

S�T1 reads va� mo (for pa� mo) and mo va� (for mo pa�) in the third quarter of the verse. All versions of the mantra verse listed above show the repetition of similar syllables, which is known as yamaka in Indian poetics. Leaving aside the distortions in the different versions, the underlying scheme seems to be that the last three syllables of a verse quarter are repeated in inverted order at the beginning of the next verse quarter. A yamaka in which the last syllables of a verse quarter are repeated (in regular order) at the beginning of the next quarter is termed sam. da.s.tayamaka in Dan.d. in's Kavyadarsa 3.51. In the scheme of this verse, however, the syllables are repeated in reverse (pratiloma) order, which resembles the pratiloma yamaka illustrated above. Each verse quarter contains two syllables which are not repeated. In the following scheme they are represented with the symbol :

abc def fed ghi / ihg jkl lkj cba // Based on this scheme the following reading of the verse is suggested. An avagraha has been inserted before meya in the first quarter of the verse:

yamarajasado'meya yame doradayodaya / yad ayonirayaks. eya yak.seya ca niramaya // Considering that the choice of syllables in the mantra is obviously dominated by sound-effects at the expense of grammar and sense, the following meaning could perhaps be extracted from this reading of the verse:

“O you immeasurable (i.e., undefeatable) (ameya) by the assembly (= troups) (sadas) of king Yama! Having arms (dor) in which there is no (a-) rise (= production) (udaya) of mercy (daya) towards Yama! Since (you) can destroy (= stop falling into) the iron hell (ayo-niraya-k.seya), I would desire to worship (you), O disease-killer (nir-a�maya)!”

In this attempt at translation, which takes into account that the text compromises in grammar and meaning of words, the form k.seya is taken as a gerundive from the root k.si (to destroy) and yak.seya as the first person singular, optative, Atmanepada of the desiderative (without reduplication) from the root yaj (to worship). The iron hell (ayo-niraya) would refer to one of the many hells described in texts (cf. Bhagavata-Purana 5.26.7 for the hell named ayah. pana). The word ca is left untranslated and is considered an expletive particle to fill in the metre.

In this interpretation, the mantra praises not Yama – as the editors of the Balinese texts cited above assume –, but the enemy of Yama, who could be identified with either Siva in his manifestation as Kalari (for the Hindus) or with Vajrabhairava (for the Buddhists).

In a graphic form, the text of the mantra could be inscribed on the sides of a square and read in clockwise direction beginning from the upper left corner. The syllables in bold font are repeated in the scheme and the syllable ya in each corner must be read each time one begins a new line. ya ma ra� ja sa do me ya ma me ra� do ni ra ca da ya yo ks.e da ya ks.e ya ra ni yo da ya c) The Twelve-, Fourteen- or Ten-Syllabled Mantra The thirty-two-syllabled mantra is immediately followed in the MP by this twelve-syllabled mantra:

om. hr�ım. vikr.ta�nana�ya hum. pha.t sva�ha�. In the MP, this mantra is identified as a mantra of Yama (47.12ab). It is followed by a description of the six limbs (an_gas) of Yama�ntaka. The mantra is not found in the TSS, but appears in the commentary on TSS, p. 238, 11–12 as the mantra of Yama in the following form:

om. s..tr�ım. vikr.ta�nana�ya sva�ha�. The mantra appears in S�T 24.20 in an encoded form. The verse in S�T1 (with readings in S�T2, 3 and S�VT, volume 2, p. 849, 9–10) is:

pran. avah. (pran. avam. S�T3) s.t.r�ım. (.s.thr�ım. S�T1, s�r�ı S�T2, pran. avo `gram. S�VT) tato dam. .s.tra� tatparam. vikr. tam. tatah. (vikr. ta�ntatah. S�T 1, 3) / a�nana�ya vadhu�r vahner mantro `yam. dva�das�a�k.sarah. // The individual syllables of the mantra in the code are given as: pran. ava (= om. ) s..tr�ım. dam. .s.tra� vikr. ta+a�nana�ya (= vik.rta�nana�ya) vadhu�r vahner (= sva�ha�), which yield the mantra: om. .s.tr�ım. dam. .s.tra�vikr. ta�nana�ya sva�ha�.

R�aghavabhat.t.a's commentary specifies this mantra as the Yam �antaka mantra (p. 866, 21). R�aghavabhat.t.a adds that instead of the above version of the mantra, some authorities recite pran. ava (= om. ) ma�ya� (= hr�ım. ) s..tr�ım. vik.rta�nana hum. pha.t sva�ha�. This version resembles the Buddhist version of the mantra given below. This mantra appears frequently with small variants in the Buddhist K.rs. n. ayama�ritantra, e.g. 1.14, in which it has fourteen syllables and is specified as the mantra of Yam�ari:

om. hr�ıh. s..tr�ıh. vikr. ta�nana hu�m. hu�m. pha.t pha.t sva�ha�. For hu�m. the variant hum. (with the short vowel) is also found, and before hu�m. /hum. other words, usually imperative forms of verbs, can be inserted (e.g., 4.23+). Variants of the mantra employed in abhica�ra rites described in the Tantra are recorded in the list of mantras, p. 165, appended to the edition of the Kr. .sn. ayama�ritantra. This mantra is also found in sa�dhanas of the red and the dark Yam�ari in SM, nos. 268–272, nos. 274–276 and nos. 278–280, in which it is occasionally classified as the deity's root ( mu�la) mantra. It also appears in Guhyasama�jatantra 15.18+ and in Abhaya�karagupta's NY, p. 37, 14.

According to NY, p. 37, 14–15, the mantra is the deity's heart ( h.rdaya) mantra as well as the mantra for all actions (sarvakarmikamantra). This latter term implies that all ritual applications can be performed with it (cf. NY, p. 31, 13–14). The mantra is inscribed on a xylograph of the

Illustration 3. Red Yamari Raktari/Rakta Yamari T. Gs´ed dmar red Yamantaka from the Rin `byun_ (Chandra 1991, p. 229, no. 584; see Illustration 3) with the spelling kri instead of kr. in vik.rtanana.

The mantra appears in a ten-syllabled form (in code) in chapter 3 of the Vajramahabhairavatantra: hr�ıh. s..tr�ıh. vik.rta�nana hu�m. hu�m. pha.t.25 The mantra is spelt out in full with minor variants in chapter 1 of the above-mentioned Three Chapter Tantra of Vajrabhairava.26 It appears in

somewhat corrupted form with seed syllables having short vowels and, as above, the spelling kri for k.r in vik.rta�nana: om. hrih. .s.trih. vikritanana hu�m. pha.t.

In the Vajramahabhairavatantra and in the Three Chapter Tantra of Vajrabhairava the mantra is classified as the action mantra. Contemporary ritual manuals of the dGe lugs pas, such as Sharpa Tulku/R.

Guard 1990, p. 66 and Sharpa Tulku/R. Guard 1991, p. 25, follow this classification. Man~jusrımulakalpa, p. 29, 11–12 gives the mantra as om. hr�ım. h. jn~ı�h. vik.rta�nana hum. : : : pha.t pha.t svaha. The occurrence of an anusva�ra (for originally anunasika) to which a visarga is added (= m. h. ) in this version is occasionally attested to for other mantras in Tantric texts.27 The only translatable word in this mantra is the vocative vik.rta�nana, “O you of deformed face.” The syllables hr�ıh. (hrı�m. ) and .s.tr�ıh. (s..tr�ım. ) are seed (b�ıja) syllables. Hu�m. (hum. ), pha.t and svaha are each classified in Hindu Tantras as closing words (jati) of mantras. Pha.t imitates the sound of bursting or breaking and sva�ha� is an exclamation already known from Vedic times, which accompanies an offering in the fire. The S�T's version of the mantra, : : : dam. s..tra�vikr. ta�nana�ya : : : sva�ha�, translates as “: : : sva�ha� to the one whose face is deformed by fangs (dam. s..tra�).” While the Hindu versions of the mantra have some seed syllables end in the nasal (hr�ım. s..tr�ım. ), the corresponding seed syllables in the Buddhist texts end in the visarga (hr�ıh. .s.tr�ıh. ). A string of seed syllables similar to those in the above mantra appears in Manjusrımulakalpa, p. 574, 20 in a mantra invoking the Yaks.in.�ı Pramoda; one of

the seed syllables ends in an anusvara followed by a visarga: om. s..thr�ım. hr�ım. h. mahanagni hu�m. pha.t sva�ha�. d) The Eight-Syllabled Mantra (of Yama) In the preface (p. 4) to his and Dvivedi's edition of the K.rs. n. ayama�ritantra, Samdhong Rinpoche states that the “importance of this work [i.e., the K.r.sn. ayamaritantra] can be judged from the fact that a Saivite Tantra called Isanasivagurudevapaddhati has borrowed some of its materials from the Kr.s. n. ayama�ri Tantra.” Samdhong Rinpoche refers here to the inclusion of the thirty-two-syllabled mantra in the MP. In his statement Samdhong Rinpoche echoes Dvivedi 1995, p. 184, who believes that the author of the �IS�P borrowed the thirty-two-syllabled mantra from the Kr. .sn. ayamaritantra. Dvivedi does not address the second mantra. The occurrence of the thirty-two-syllabled mantra in the K.rs. n. ayamaritantra cannot be taken as proof that this Tantra constitutes

the source from which the TSS and the MP have borrowed, especially since no entire passage (except for the two mantras) from the Tantra can be identified in the TSS or the MP. This mantra, as well as the second one which Dvivedi does not address, also appears in three Vajra-bhairava texts. S�T 24.19 provides a third mantra, dhu�ma�ndhaka�ra�ya sva�ha�, which is to be inscribed in the same yantra. This mantra is also referred to in MP 47.19c as the eight-syllabled mantra of Yama. It is not included in the TSS, but referred to in its commentary, p. 238, 15–16, also as the mantra of Yama. This mantra is not found in the Buddhist texts examined above. Therefore, it seems that the TSS and the MP (as well as the S�T) did not borrow the two mantras directly from the K.r.sn. ayamaritantra or from the Vajrabhairava texts cited above but rather from another unidentified source, in which a third mantra was included.

e) On the Iconography We do not find iconographic descriptions of Yama/Yamantaka in the TSS or the MP. Raghavabhat.t.a, p. 866, 25, commenting on the thirty-two-syllabled mantra, states that the visualization of the deity should be learnt from one's preceptor. The mantras of the deity's limbs ( an_ga) found in MP 47.11+ refer to the deity's deformed face, his dark ( kr. .sn. a) colour, speak of his nine faces and reddish-brown hair mass. These references most likely have contributed to Dvivedi 1992, p. 42 labelling the section describing the deity's yantra, which he extracts and reprints from the MP, as the sadhana of the dark Yamari (Krsnayamari), as opposed to that of the red Yamari (Raktayamari). The Kr.s .n. ayamaritantra, which Dvivedi 1995, p. 184 considers the source for the thirty-two-syllabled mantra of Yamantaka in the MP, does not describe a nine-faced form of Yamari. This nine-faced dark Yamantaka (cf. also MP 47.23a+) can be identified as a form of Yamantaka called Vajrabhairava (cf. De Mallmann 1986, pp. 400–401). That deity is described with nine faces, sixteen legs and thirty-four arms in chapter 4 of the above-mentioned Vajramahabhairavatantra (Siklos 1996, pp. 38–41).28

5. APPENDIX: OTHER MANTRAS OF BUDDHIST ORIGIN

In addition to the above mantras, the TSS and the MP include a few other mantras of Buddhist origin. These mantras include fragments of typically Buddhist Tantric offering mantras. Invocations such as namo ratnatraya�ya, “salutation to the three jewels” (i.e., to the Buddha, the dharma and the sam. gha), as well as epithets employing the prefixBUDDHIST DEITIES AND MANTRAS IN THE

vajra- indicate their Buddhist origin. In the following, I refrain from a detailed discussion of each mantra for reasons of space. a) The can.d.a�sidh�ar�a-mantra for the destruction of evil demons (graha) which attack children The MP inserts the following long mantra between 43.52ab and cd. Its name, can. d. asidhara (cf. also MP 43.52c), means “the edge of the fierce sword.” The mantra includes the epithet can. d. a�sidha�ra�dhipati, “overlord of the edge of the fierce sword.” In the MP and the TSS this mantra is followed by another mantra which addresses Khad.gar�avan.a.

This Khad.gar�avan.a, who is known as a form of S�iva, is also addressed as Candesvara, Rudra and “the lord of the edge of the fierce sword” (can. d. a�sidha�ra�pati). The mantra invokes Can.d. avajrap�an. i, a fierce form of the Yaks.a Vajrapan.i. namo bhagavate ratnatraya�ya </> namas� can. d. avajrapan. aye maha�sattvasena�pataye (for mahayaksa) </> namas� can. d. as�r.n_khala�ya prad�ıpta�ya prajvalita�rpitad�ıptakes�a�ya n�ılakan..tha�ya cintita�ya ru�pa�ya (?) lambodara�ya maha�jn~a�navaktra�ya bhruku.tika�ma�ya caturdam. s..tra�ya kara�la�ya maha�vik.rtaru�pa�ya vajragarbha�ya ehy ehi ka�yam anupravis�ya s�irasi gr. hn. a cak.sus. �ı ca�laya hari (v.l. bhiri) kim. cira�yasi siddhadevada�navagandharvayaks. ara�ks. asapretana�gapis�a�ca�m. s tra�saya kampaya samayam anusmara hana jaha paca matha vidhvam. saya can. d. a�sidha�ra�dhipatir a�jn~a�payati hum. pha.t sva�ha� / The version in TSS 13.44ab+ is:

namo ratnatraya�ya </> namas� can. d. avajras�.rn_khala�ya prad�ıpta�ya prajvalitahasta�ya prajvalita�rcitad�ıptakes�a�ya n�ılakan. .tha�ya k.rta�ntaru�pa�ya lambodara�ya maha�jn~a�navaktra�ya bhruku.t�ımukha�ya caturdam. .s.tra�kara�la�ya maha�vik.rtaru�pa�ya vajragarbha�ya / ehy ehi ka�yam anupravis�ya s�irasi gr.hn. a / cak.sus. �ı ca�laya / hiri bhiri kim. cirayasi / devada�navagandharvayaks. ara�k.sasabhu�tabh�ı.san. a pretana�gapis�a�ca�pasma�ra�n tra�saya kampaya samayam anusmara hana daha paca matha vidhvam. saya can. d. a�sidha�ra�dhipatirudro jn~a�payati hum. pha.t sva�ha� /

In addition to Candavajrapani, the TSS's version of the mantra invokes Vajrasrnkhala, who may be the male counterpart of Vajrasrnkhala who is known as a Yaksinı in the Jain pantheon (Misra 1981, p. 128, p. 174). She also figures as a gate keeper in deity man. d. alas described in Abhayakaragupta's NY. Compare parts of this mantra to parts of the following dharanı from SM, no. 205, p. 404, 5–16:

namo ratnatraya�ya / namas� can. d. avajrapa�n. aye maha�yak.sasena�pataye / namo bhagavati maha�vajraga�ndha�ri anekas�atasahasraprajvalitad�ıptateja�yai ugrabh�ımabhaya�naka�yai yogin�ıyai bh�ıs.mabhagin�ıyai dva�das�abhuja�yai vik�ırn. akes�ı�yai anekaru�pavividhaves�adha�rin. ı�yai / ehy ehi bhagavati maha�vajraga�ndha�ri traya�n. a�m. ratna�na�m. satyena a�ka.ta a�ka.ta baladeva�dikam. ye ca�nye samaye na ti.s.thanti ta�n a�varttayis.ya�mi / s�ı�ghram. gr. hn. a gr. hn. a gr.hn. a om. ala ala ala ala hulu hulu mulu mulu culu culu dhama dhama rak.sa rak.sa rak.sa�paya rak.sa�paya pu�raya pu�raya a�vis�a

a�vis�a bhagavati mahavajragandhari siddhacan. d. avajrapa�n. ir a�jn~a�payati hr�ıh. hah. hum. pha.t sva�ha� / The invocation namas� can. d. avajrapan. aye maha�yak.sasena�pataye, which is found in the can. d. a�sidha�ra�-mantra as well as in the initial part of the dha�ran. ı�from the SM, also appears in texts such as the Satasahasrika Prajnaparamita (p. 2, 2). It is found several times as part of a dharanı of Mahabala in the Chinese version of the Aryamahabala(nama-mahayana-)sutra as reproduced in Bischoff 1956, pp. 81, 85 – 86, 100: namo ratnatrayaya namas can. d. avajrapan. aye mahayak.sasenapataye. It also found in the Balinese Buddhaveda (Levi 1933, p. 80, 24): namo ratnatrayaya namas can. d. avajrapan. imahayak.sasenapati. The phrase kim. cira�yasi samayam anusmara svaha is part of a mantra of gate keeper Abhimukha in Mahavairocanasutra (Yamamoto 1990), p. 56 and p. 114 and the phrase kim. cirayasi only appears in a mantra of the surrounding deities in Mahavairocanasutra, p. 48. The expres-sion samayam anusmara is found several times in texts such as the Sarvatathagatatattvasam. graha, p. 171, 2; p. 199, 5–6, p. 260, 3 and p. 266, 3.

b) The vajragandharı-mantra for protection from Pisacas, evil demons (graha) and fever (MP 43.72+)

om. ra.s.tidehim. coktajika�dha om. ka�r�ım. (?) ka�tya�yan�ım. (?) nair.rtya�m. ka�l�ım. maha�ka�l�ım. vajraka�l�ım. yas�asvin�ım. suka�l�ım a�gneya�m. va�yavya�m. ka�lika�m. pan_ktis�aktim. s�a�nta�k.s�ım indra�n. �ım. yak.sakauber�ım. ma�hes�var�ım. vai.sn. av�ım. ca�mun. d. �ım. raudr�ım. va�ra�h�ım. kauber�ım. ya�s� ca�nya� mama samaye tis..thanti tanna�ma�vartayis. ya�mi / s��ıghram. gr.hn. a / om. lala culu pu�raya dhara a�naya subhage / a�vis�a bhagavati / maha�vajraga�ndha�ri siddhacandravajrapa�n. ir a�jn~a�payati hr�ım. hah. ha�m. ha�m. ha�m. hum. pha.t sva�ha� /

In this mantra we encounter the name Vajrag�andh�ar�ı, who figures as a Yaks.in.�ı in Jainism (see Misra 1981, pp. 128, 174, 175). Parts of this mantra are similar to the dha�ran. ı� cited from the SM in a). That dha�ran. ı� is part of SM, no. 205, which is the only sa�dhana in the SM, which is dedicated to Vajrag�andh�ar�ı. c) Mantra against fever (jvara) (MP 45.55+) The presiding deity of the following mantra is the Buddha (buddha�dhidaivato mantro). namo s.aratnatra (for ratnatraya�ya?) s�a�s�am. (for jvara?) h.rdayam a�vartayis. ya�mi / bho bho jvara s�r.n. u hana garda charda sarvajvara ca.ta vajrapa�n. ir a�jn~a�payati / s�iro mun~ca kan..tham. mun~ca uro mun~ca h.rdayam. mun~ca udaram. mun~ca ka.tim. mun~ca jan_ghe mun~ca pa�dau mun~ca can. d. apa�n. ir a�jn~a�payati hum. pha.t sva�ha� / TSS 15.2+ gives a similar mantra:

namo ratnatraya�ya </> jvarahr. dayam a�vr. tayi.sya�mi bho jvara s�.rn. u hana garja charda s�arvajvara ava.ta vajrapa�n. ir a�jn~a�payati mama s�iro mun~ca kan. .tham. mun~ca ba�hum. mun~ca udaram. mun~ca ka.tim. mun~ca guhyam. mun~ca u�rum. mun~ca ja�num. mun~ca jan_ghe mun~ca pa�dau mun~ca can. d. apa�n. ir a�jn~a�payati hum. pha.t sva�ha� / d) Mantra accompanying a bali offering (TSS 30.37+)

iti miti timi ka�katun. d. ini sva�ha� / namo ratnatraya�ya pin_gala�ya anale kunale g.rhn. a pin. d. am. pis�a�cini sva�ha� //

The mantra contains a salutation to the three jewels. The text following the invocation resembles the following mantra inscribed on a xylograph entitled “the three sisters (bhagin�ıtraya) dByug gu ma” from the Rin `byun_ (Chandra 1991, p. 317, no. 842). I have made no attempt to correct the text of the mantra:

om. anale kun. d. ale maha�pis�aciniye sva�ha� / om. gr. hna maha�bha�n. d. a pis�aciniye sva�ha� / e) Mantra to protect cattle (TSS 30.7cd+; MP 49.6ab+) namo bhagavate vajrahum. ka�radars�ana�ya (vajramuka�� MP) om. cuku (vila MP) mili meli siddhi gomari vajrin. i hum. pha.t / asmin gra�me gokulasya rak.sa�m. kuru s�a�ntim. kuru svaha /

Vajrahum. k�ara, whose name appears in the invocation part of this mantra, is known as a deity of the Buddhist Tantric pantheon (cf. SM, no. 257). The vocative gomari may refer to a form of (Maha)marı, the goddess of pestilence, whose mantra is taught in chapter 137 of the Agni-Purana and other texts. A mantra similar to the above one appears in Agni-Purana 302.29–30: om. namo bhagavate tryambaka�yopas�amayopas�amaya culu culu mili mili bhidi bhidi gomanini cakrin. i hrum. pha.t / asmin grame gokulasya raksam. kuru santim. kuru kuru kuru svaha.29 It is important to note that in the Agni-Purana the name Vajrahumkara has been replaced with that of Tryambaka, i.e., Siva, and the vocative vajrin. i with cakrin. i. In addition, the Agni-Purana's version of the mantra shows corrupt forms of words.

6. CONCLUSION

Both the MP inserted into the �IS�P and the TSS incorporate descriptions of Vasudhara and Jambhala, originally Buddhist deities, along with the procedures for their ritual worship which include typically Buddhist Tantric elements. The two texts must have been extant before 1493, the year in which Raghavabhat.t.a completed his Padarthadars�a commentary328 GUDRUN BU¨HNEMANN on the S�T, in which they are cited. The relationship between the two texts is briefly addressed in the introduction to this paper. While this issue requires further examination, the sections of the two texts studied in this paper seem to confirm Goudriaan's hypothesis that chapters 15–38 of the MP are earlier than the TSS, while chapters 39–52 of the MP are based on the TSS.

The worship procedures for Vasudhara and Jambhala described in these texts clearly show Tantric Buddhist elements. The iconography of the two deities is likely to be Buddhist as well, but their sources have not yet been identified in Buddhist texts. An identification of these deity descriptions may be possible when additional text material is made available in edited form. Vasudhara, who is classified as a form of Laksmı, is yellow, holds a pomegranate in her left hand and a red lotus with a vessel showering jewels in its interior in her right and has her foot placed on a vessel from which wealth flows. This description does not correspond to that of Vasudhara frequently found in Buddhist texts, where the goddess holds a rice shoot and makes the wish-granting gesture.

Jambhala is described twice. According to the first description he is yellow, is seated on a white lotus, has three feet and is corpulent.

The attributes held in his two hands are not given. While the MP does not mention the number of his faces, the TSS specifies three faces. In the second description, the deity is four-armed and holds the fruit of the citron tree and the mongoose and makes the gestures of protection and wish-granting. The number of his faces is not specified and must therefore be taken as one. Neither description has so far been identified in Buddhist texts, even though the two attributes, the fruit of the citron tree and the mongoose, are common attributes of the two-armed Jambhala.

The mantras of Yamantaka appear in connection with a yantra of Yama used in the rites of black magic (abhicara), most likely the rite of liquidation (marana). In the ritual applications of the MP and the TSS both the names Yama and Yamantaka (elsewhere known as Yamari) appear. I would, however, not conclude that an observation made by O'Flaherty 1976, p. 232, is applicable here, according to which Yamantaka and Kalantaka, “death, the ender,” were originally epithets of Yama, which were then transferred to Siva and reinterpreted as “the ender of death.” The cause of this confusion is that mantras of the Buddhist Yamantaka were incorporated into a yantra of Yama. The first mantra is thirty-two-syllabled and the second twelve-, fourteen- or ten-syllabled. While the texts of the Yamantaka cycle of the Tibetan Buddhist tradition

employ both of these mantras as mantras of Yamantaka/Yamari, the Hindu Tantric texts examined in this paper identify the second mantra as a mantra of Yama. The wording of the two mantras, which continue to be recited by Tibetan Buddhists up to the present, does not indicate a connection to Tantric Buddhism. The first one seems to be in praise of the enemy of Yama, who could be identified either as Siva in his manifestation as Kalari (for the Hindus) or as Vajrabhairava (for the Buddhists). The second mantra addresses the (deity) with a face deformed (by fangs). The main texts of the Yamantaka cycle in which these two mantras appear, are said to have originated in Uddiyana.

Uddiyana/Oddiyana is normally identified with a province in the Swat Valley in the north-west of the subcontinent, present-day Pakistan, where Tantrism once flourished. According to the Hindu Tantras, the two mantras are said to be inscribed in the yantra along with a third eight-syllabled mantra which cannot be identified in the Buddhist texts examined in this paper. The third mantra is identified as a mantra of Yama. It appears as though texts such as the MP and the TSS did not borrow the three mantras directly from Buddhist Tantric texts, for example the K.r.sn. ayamaritantra, but rather from another source which included the third mantra.

The mantras of Yamantaka's limbs ( anga) listed in MP 47.11+ address a dark deity with nine faces and reddish-brown hair. This description suggests a nine-faced form of the dark Yamantaka (cf. also the references to his nine faces in MP 47.23a+) who is identified as Vajrabhairava. This nine-faced form of Yamantaka is not described in the Kr. s. n. ayamaritantra but in chapter 4 of the Vajramahabhairavatantra.

The question that arises is what attitudes the compilers of the MP and the TSS had toward the Buddhist material they included. The compilers of the MP and the TSS seem to have had an ambivalent attitude. On the one hand, they describe the rites of black magic (abhicara) for use against the enemies of the (Vedic) dharma and the Veda. On the other hand, they incorporate mantras from these very enemies. Unlike other groups in Hinduism who included the Buddha among Visnu's avataras, the compilers of these two texts made a distinction between their own tradition and that of the Buddhists. The two Yamantaka mantras are inscribed in yantras. Since they were transmitted as part of a ritual procedure which included the drawing of a powerful yantra, they could not easily be omitted. In the case of the other mantras, which were transmitted as part of a series of mantras to cure diseases, the compilers apparently did not want to exclude popular mantras, which were believed to be powerful, even though they carried traces of the

Buddhist context from which they were taken. Other mantras were inserted between descriptions of ritual procedures for similar Hindu deities for the sake of completeness. The description of Vasudhara, for example, precedes that of different forms of Durga and is directly followed by the presentation of the mantras of the traditional Hindu earth goddess Bhudevı. The description of Jambhala is followed by that of Kubera. In the above discussed texts the Buddhist deities do not occupy the positions of major deities.

Jambhala, Vasudhara and Yama are all associated with the Yaks.a cult as well as Vajragandharı, Vajrapani and possibly Vajrasrnkhala (if he is the male counterpart of Vajrasrnkhala), whose names are invoked in some of the mantras of Buddhist origin listed in the appendix of this paper. Some of the mantras explicitly invoke the lord of the Yaks.as. In their subordinate positions they were apparently not felt to interfere with the compilers' sectarian affiliations.

NOTES

1
See Bhattacharyya 1930, 1932, pp. 147ff. and his introduction to his edition of the Sa�dhanama�la�, volume 2, pp. cxxxvff.
2
This section of the TSS shares identical passages with the Kasyapa-Sam. hita� (Garud. apan~ca�k.sar�ıkalpa). For some information, see Aiyangar's preface to his edition of the TSS, p. 8.
3
In the following I identify a number of citations from the Narayan. �ıya in R�aghavabhat.t.a's commentary on the S�a�rada�tilaka (S�T1) that are from the TSS: Raghavabhat.t.a, p. 589, 12 = TSS 25.29cd; p. 589, 20 = TSS 25.23ab; p. 590, 13–14 = TSS 25.24cd–25ab; p. 591, 24–26 = TSS 25.14–15ab; p. 592, 5 = TSS 25.3d; p. 592, 25 = TSS 25.7ab; p. 592, 27–28 = TSS 25.7cd–8ab; p. 593, 15 = TSS 25.9a; p. 593, 18–19 = TSS 25.6; p. 748, 18 = TSS 23.39ab; p. 749, 15 = TSS 23.44cd; and p. 750, 19–20 = TSS 23.51a–c.
4
Both the preface to the edition of the TSS, p. 1 and Goudriaan in Goudriaan/Gupta 1981, p. 128 give the date as the fifteenth or the sixteenth century.
5
See V.V. Dvivedi's remarks in connection with the Prapancasara in the introduc-tion to his edition of the Nitya�s.od. as�ika�rn. ava (Varanaseya Sanskrit Vishvavidyalaya, Varanası 1968), p. 41. Isanasiva of the Mattamayura lineage was the teacher of Vairocana who wrote the Pratisthalaksanasarasamuccaya. The oldest manuscript of the text dates from 1168 CE (see The Hindu Deities Illustrated according to the Pratis..thalaksanasarasamuccaya. Compiled by G. Bu¨hnemann and M. Tachi-kawa (The Centre for East Asian Cultural Studies, Tokyo 1990), part 1: The Prati.s.tha�lak.san. asarasamuccaya and Its Illustrations by G. Bu¨hnemann, p. 12).
6
The MP erroneously reads Amitagha.
7
MP signals variant readings in the text, but it is not entirely clear which reading they are replacing: “ka�ntimat�ı, tanv�ıti kecit.”
8
The edition of the MP takes verses 20a and 20b as one stanza, which is numbered as 20. The metre of both verses is M�atr�asamaka (16 ma�tras per quarter).
9 �d
. a�d. ima�m. MP.
10
Suggested emendation, �gha.tasya�n_ghrim. TSS.BUDDHIST DEITIES AND MANTRAS IN THE HINDU TANTRAS 331
11
Suggested emendation, vamapanina TSS.
12
The MP erroneously reads ksayadhipataye.
13
tris�iras v.l. TSS.

Abhayakaragupta's NY gives different descriptions of Kubera. According to NY, p. 61, 12, Kubera holds a mace and a noose, while NY, p. 73, 19–20 describes him as holding a mace and a female mongoose. According to NY, p. 89, 3–4, the deity holds a mace, a jewel, a mongoose and a lotus. NY, p. 89, 3–4 and p. 93, 1 use the epithet Jambhala as a synonym of Kubera. Chandra 1991, p. 172 (no. 388) includes a block-print of the Buddhist Kubera from the Pantheon of the Mongolian Kanjur, in which Kubera holds Jambhala's attributes. According to Blom 1989, pp. 31–32, who addresses depictions of Kubera from Nepalese sketchbooks, Kubera holds, among other attributes, Jambhala's characteristic mongoose and the fruit of the citron tree.

Among the different iconographic forms of the Buddhist Jambhala, one form even steps on Kubera to demonstrate his superiority (cf. SM, no. 292). In texts of the Hindu tradition, the gestures of wish-granting and protection as attributes of Kubera are also recorded in the Am. subhedagama (cited in Rao 1914–1916, volume 2, p. 263).
The fruit of the citron tree appears as an attribute of Kubera in Rupaman. d. ana 2.37 as well. An icon of Kubera from Mathura, described in Misra 1981, p. 69 (see also Figure 9), represents a six-armed Kubera, holding the citron and the mongoose, displaying the gestures of wish-granting and protection and holding two other objects.
15

The TSS uses the term k.sudra as a synonym of abhica�ra. Perhaps the seven k.sudras (also referred to in TSS 18.4c) were considered to be a special group of abhicara rites.
16
Here I follow the explanation given in the commentary on TSS, p. 235, 9–10, which gives the synonym unma�da for bhrama. Goudriaan 1978, p. 356 opts for the meaning “wandering, causing to wander.”
17
The MP erroneously reads yoga instead of roga.
18
For the importance of the Vajrabhairava cycle in the dGe lugs pa tradition, see Siklo�s 1996b, pp. 186–187.
19
In the block-print, the mantra is prefixed with the syllable om. and appended with the syllables hu�m. hu�m. pha.t pha.t sva�ha�. The mantra hr�ıh. .s.tr�ıh. vikrita�nana (!) hu�m. pha.t follows, which is a variant of the mantra addressed in section 4c of this paper.
20
Cf. the edited Tibetan text, Siklo�s 1996, pp. 91–92.
21
For the Tibetan text, see Siklo�s 1996, p. 138 and for the Mongolian text, which reads ni ra ma ya (for ni ra� ma ya), see see Siklo�s 1996, p. 210; cf. also the translation section, Siklo�s 1996, p. 65.
22
In the Blue Annals (Roerich 1976, p. 375) the title Trikalpa also appears as part of the title of another text, the K.rs. n. ayama�ritantrara�ja-Trikalpa, which is distinguished from the above cited (Sarvatatha�gataka�yava�kcitta-) Kr.s.n. ayama�ritantra; cf. also Siklo�s 1996, p. 18, note 24 and the discussion in Decleer 1998, pp. 293–294.
23
See the Tibetan text in Siklo�s 1996, p. 145 and the Mongolian text in Siklo�s 1996, p. 217; see also the translation section, Siklo�s 1996, p. 70.
24
See Rao 1914–1916, volume 2, pp. 156ff. for the story of M�arkan.d.eya and S�iva's manifestation K�al�ari according to several A�gamas.
25
See the Tibetan text in Siklo�s 1996, pp. 92–93 and the translation, p. 37.
26
See Siklo�s 1996, p. 145 for the Tibetan text, p. 217 for the Mongolian text and p. 70 for the translation.
27
Cf. above cited mantra of the Yaks.in.�ı Pramod�a from the Man~jus�r�ımu�lakalpa, the seed syllable gam. h. for Gan.e�sa in S�VT, volume 2, p. 668, 8 (interpreting TSS 24.31) and the seed syllable hr�ım. h. in NY, p. 65, 20.332 GUDRUN BU¨HNEMANN
28
A discussion of the deity's iconographic forms is beyond the scope of this article.
Sometimes Vajrabhairava also appears in a subordinate position. Thus SM, no. 312 (p. 598, 20–21) describes him under the feet of a sixteen-armed Mah�ak�ala.
29
The Agni-Pura�n. a reads .tha .tha .tha. The syllables .tha .tha correspond to sva�ha� as is widely attested. Since the third syllable .tha is redundant, it has been omitted here.
30
The so-called “first edition” of 1992, published by the Chaukhamb�a Sam. skr.t Pratis.t.h�an, Delhi, is only a photomechanical reprint of the edition from Madras.
REFERENCES
Agni-Pura�n. a, Agnipura�n. am. General editor H.N. A�pt.e (A�nand�a�srama, Pun.e 1957) (second edition).
Banerjea, J. N.: 1956, The Development of Hindu Iconography (University Press, Calcutta) (second edition, revised and enlarged).
Bhattacharyya, B.: 1930, `Buddhist Deities in Hindu Garb'. Proceedings and Trans-actions of the Fifth Indian Oriental Conference. 2 volumes (University of the Panjab, Lahore), pp. 1277–1298.
Bhattacharyya, B.: 1932, An Introduction to Buddhist Esoterism (Oxford University Press, London).
Bhattacharyya, B.: 1933, `The Cult of Bhu�tad.�amara'. Proceedings and Transactions of the Sixth All-India Oriental Conference (Bihar and Orissa Research Society, Patna), pp. 349–370.
Bhattacharyya, B.: 1958, The Indian Buddhist Iconography Based on the Sadhanamala and other cognate Tantric Texts of Rituals (Firma K.L. Mukhopadhyay, Calcutta) (second edition, revised and enlarged).
Bhat.t.acharyya, D. C.: 1974, Tantric Buddhist Iconographic Sources (Munshiram Manoharlal Publishers Pvt. Ltd., New Delhi).
Bischoff, F. A.: 1956, Contribution a� l'e�tude des divinite�s mineures du bouddhisme tantrique: A�rya Maha�bala-Na�ma-Maha�ya�nasu�tra Tibe�tain (Mss. de Touen-Houang) et Chinois (Paul Geuthner, Paris).
Blom, M. L. B.: 1989, Depicted Deities: Painters' Model Books in Nepal (Egbert Forsten, Groningen).
Bu¨hnemann, G.: 1996, `The Goddess Mah�ac�ınakrama-T�ar�a (Ugra-T�ar�a) in Buddhist and Hindu Tantrism'. Bulletin of the School of Oriental and African Studies 59: 472–493.
Bu¨hnemann, G.: 1999, The Iconography of Hindu Tantric Deities. Volume One: The Pantheon of the Mantramahodadhi (Egbert Forsten, Groningen).
Chandra, L.: 1991, Buddhist Iconography, begun by the late Prof. Raghu Vira.
Compact edition (International Academy of Indian Culture/Aditya Prakashan, New Delhi).
Chattopadhyaya, D. (ed.): 1970, Ta�rana�tha's History of Buddhism in India. Translated from the Tibetan by Lama Chimpa/Alaka Chattopadhyaya (Indian Institute of Advanced Study, Simla).
De Mallmann, M.-Th.: 1963, Les Enseignements Iconographiques de l'Agni-Purana (Presses universitaires de France, Paris).
De Mallmann, M.-Th.: 1964, Divinit�es hindoues dans le tantrisme bouddhique. Arts Asiatiques 10: 67–116.
De Mallmann, M.-Th.: 1968, Hindu Deities in Tantric Buddhism. Zentralasiatische Studien 2: 41–53.
De Mallmann, M.-Th.: 1986, Introduction a� l'iconographie du ta^ntrisme bouddhique (Maisonneuve, Paris).
Decleer, H.: 1998, Review of Siklo�s 1996. Indo-Iranian Journal 41: 290–301.BUDDHIST DEITIES AND MANTRAS IN THE HINDU TANTRAS 333
Dvivedi, V. V.: 1992, `Vasudh�ar�a evam. kr.s.n.ayam�ari s�adhana'. Dh�ıh. 13: 33–44 (Hindi).
Dvivedi, V. V.: 1995, `A�c�arya�san_kar�ıyah. prapan~cas�arah.'. In: Nigam �agam�ıyam. sam. skr.tidar�sanam <by V.V. Dvivedi> (S�aivabh�arat�ı-S�odhapratis.t.h�anam, V�ar�an.as�ı), pp. 182–199 (Sanskrit).
Getty, A.: 1928, The Gods of Northern Buddhism: Their History, Iconography and Progressive Evolution Through the Northern Buddhist Countries (Clarendon Press, Oxford).
Goudriaan, T.: 1977, `Khad.ga-R�avan.a and his Worship in Balinese and Indian Tantric Sources'. Wiener Zeitschrift fu¨r die Kunde Su¨dasiens 21: 143–169.
Goudriaan, T.: 1978, Ma�ya� Divine and Human (Motilal Banarsidass, Delhi).
Goudriaan, T./S. Gupta: 1981, Hindu Tantric and S�a�kta Literature (Otto Harrassowitz, Wiesbaden).
Goudriaan, T./C. Hooykaas: 1971, STUTI and STAVA (Bauddha, S�aiva and Vai.sn. ava) of Balinese Brahman Priests (North-Holland Publishing Company, Amsterdam).
Guhyasama�jatantra, The Guhyasama�ja Tantra. Edited by Y. Matsunaga (Toho Shuppan, Osaka 1978).
Hooykaas, C.: 1964, A�gama T�ırtha: Five Studies in Hindu-Balinese Religion (North-Holland Publishing Company, Amsterdam).
Hooykaas, C.: 1973, Balinese Bauddha Brahmans (North-Holland Publishing Company, Amsterdam).
Hopkins, E. W.: 1915, Epic Mythology (Karl J. Tru¨bner, Strassburg). �IS�P I�s�a�nas�ivagurudevapaddhati, The ^ısa^nasivagurudevapaddhati by ^ısa^nas�ivagurudevamisra. Edited by T. Gan.apati S^astr^ı. 4 parts (Trivandrum University Press, Trivandrum 1920–1925).
K.rs. n. ayama�ritantra, Kr.s.n. ayama�ritantra With Ratna�val�ı Pan~jika� of Kuma�racandra.
Edited by Samdhong Rinpoche and V.V. Dvivedi (Central Institute of Higher Tibetan Studies, Sarnath, Varanasi 1992).
L�evi, S. (ed.): 1933, Sanskrit Texts from Bali (Oriental Institute, Baroda).
Man~jus�r�ımu�lakalpa, The A�ryaman~jus�r�ımu�lakalpa. Edited by T. Ganapati Sastri. 4 parts (Sri Satguru Publications, Delhi 1989) (reprint of the edition of Trivandrum 1925).
Misra, R. N.: 1981, Yaksha Cult and Iconography (Munshiram Manoharlal Publishers Pvt. Ltd., New Delhi).
Mitra, D.: 1961, `Jambhala-Man. d.alas in Sculpture'. Journal of the Asiatic Society (Calcutta) 3/1: 39–41.
Miyata, T. (ed.): 1988, Ritual Directions for the S�a�ntika Homa Offered to Acala (Koyasan Shingon Foreign Mission, Koyasan).
MP Mantrapa�da of the �IS�P see �IS�P.
NY Ni.spannayoga�val�ı, Ni.spannayoga�val�ı of Maha�pan. d. ita Abhaya�karagupta. Edited by B. Bhattacharyya (Oriental Institute, Baroda 1949).
O'Flaherty, W. D.: 1976, The Origins of Evil in Hindu Mythology (University of California, Berkeley).
Pal, P.: 1981, Hindu Religion and Iconology According to the Tantrasa�ra (Vichitra Press, Los Angeles).
PS Prapan~casa�ra, Prapan~casa�ra Tantra of S�an_kara�ca�rya. With the Commentary Vivaran. a by Padmapa�da�ca�rya and Prayogakramad�ıpika�, a V.rtti on the Vivaran. a.
Revised and documented with exhaustive Introduction by A. Avalon And Edited By A. Sarasvat�ı. Parts I, II in One Volume (Motilal Banarsidass, Delhi 1981) (reprint).
Rao, T. A. G.: 1914–1916, Elements of Hindu Iconography. 2 volumes (Law Printing House, Madras).
Roerich, G. (transl.): 1976, The Blue Annals. Parts I & II (Bound in One) (Motilal Banarsidass, Delhi, second edition).
Ru�paman. d. ana In: Devat�amu�rtiprakaran.am and Ru�paman.d.anam. Edited by U. M.
Sankhyatirtha (Metropolitan Printing and Publishing House, Calcutta 1936).334 GUDRUN BU¨HNEMANN
Sanderson, A.: 1988, `S�aivism and the Tantric Tradition'. In: The World's Religions.
Edited by S. Sutherland/P. Clarke/F. Hardy (Routledge, London), pp. 128–172.
Sanderson, A.: 1994, Vajray�ana: Origin and Function. In: Buddhism into the Year 2000. International Conference Proceedings (Dhammakaya Foundation, Bangkok/Los Angeles), pp. 87–102.
Sarvatatha�gatatattvasam. graha, Sarva-Tatha�gata-Tattva-San_graha na�ma Maha�ya�na-Su�tra. A critical edition based on a Sanskrit manuscript and Chinese and Tibetan translations. Edited by I. Yamada (Jayyed Press, New Delhi 1981).
S�atasa�hasrika� Prajn~a�pa�ramita�, C¸ atasa�hasrika�-Prajn~a�pa�ramita�: A Theological and Philosophical Discourse of Buddha with his Disciples (in a hundred-thousand stanzas) (Baptist Mission Press, Calcutta 1902–1914).
Sharpa Tulku/R. Guard: 1990, Meditation on Vajrabhairava, by Phabongkha Dechen Nyingpo. Compiled and translated by Sharpa Tulku with R. Guard (Library of Tibetan Works and Archives, Dharamsala).
Sharpa Tulku/R. Guard: 1991, Self-Initiation of Vajrabhairava. Compiled by K.
Phabongkha Rinpoche, translated by Sharpa Tulku with R. Guard (Library of Tibetan Works and Archives, Dharamsala).
Sharpa Tulku/M. Perrott: 1987, A Manual of Ritual Fire Offerings. Translated and compiled by Sharpa Tulku and M. Perrott (Library of Tibetan Works and Archives, Dharamsala).
Siddhaikav�ıramaha�tantra: 1997, `Siddhaikav�ıramah�atantram'. Dh�ıh. 23: 143–166.
Siklo�s, B.: 1996, The Vajrabhairava Tantras: Tibetan and Mongolian Versions, English Translation and Annotations (The Institute of Buddhist Studies, Tring, U.K.).
Siklo�s, B.: 1996b, The Evolution of the Buddhist Yama. In: The Buddhist Forum IV. Seminar Papers 1994–1996. Edited by T. Skorupski (School of Oriental and African Studies, London), pp. 165–189.
SM Sa�dhanama�la�, Sa�dhanama�la�. Edited by B. Bhattacharya. 2 volumes (Oriental Institute, Baroda 1925–1928).
S�T1 S�a�rada�tilaka, S�a�rada�-Tilaka Tantra. Text <with Ra�ghavabha.t.ta's Pada�rtha�dars�a> with Introduction. Edited by A. Avalon (Motilal Banarsidass, Delhi 1982) (reprint).
S�T2 S�a�rada�tilaka, S�a�rada�tilakam of S�r�ı Laks.man. ades�ikendra with Pada�rtha�dars�a Commentary by S�r�ımad Ra�ghava Bha.t.ta. Edited by Mukund Jha Bkashi (Chaukhambha Sanskrit Sansthan, Varanasi 1986) (third edition).
S�T3 S�a�rada�tilaka, S�a�rada�tilaka-gu�d. ha�rthad�ıpika�-sahita (Gan.e�s Prabh�akar Press, V�ar�an.as�ı 1884).
S�VT S�r�ıvidya�rn. avatantra, Shrividyarnava Tantra. Edited by R.C. Kak/Harabhatta Shastri. 2 volumes (Kashmir Mercantile Electric Press, Srinagar 1932–1937).
TSS Tantrasa�rasam. graha, Tantra Sa�ra San_graha (with Commentary). A treatise teaching formulae and rites for the attainment of health and happiness and even of super-human power of Na�ra�yan. a (Ta�ntric) of S�ivapuram. Edited : : : by M.D.
Aiyangar (Government Oriental Manuscripts Library, Madras 1950).30 Unni, N. P.: 1987, Tantrapaddhati (Bharatiya Vidya Prakashan, Delhi).
Yamamoto, C. (transl.): 1990, Mahavairocana-Sutra (International Academy of Indian Culture and Aditya Prakashan, New Delhi).
University of Wisconsin-Madison Department of Languages and Cultures of Asia 1240 Van Hise Hall, 1220 Linden Drive Madison, Wisconsin 53706, USA

Source