The Mahāmāyātantra, named after its principal deity Mahāmāyā, belongs to the class of Yoginītantras. According to the post-tenth-century classification scheme of the Tibetan New Schools (gsar ma), the Mahāmāyātantra is categorized as a Mother tantra (ma rgyud) among Unexcelled Yoga tantras (bla na med pa’i rnal ’byor gyi rgyud, yoga­niruttara­tantra ). It earns this classification due both to the importance placed on female divinities in the tantra’s maṇḍala and to its inclusion of practices focused on the manipulation of the body’s subtle energies. In this tantra, Mahāmāyā presides over a maṇḍala populated primarily by yoginīs and ḍākinīs, those semi-divine female figures known throughout South Asian tantric traditions for the power they derive from being propitiated with blood, flesh, and sex. The practitioner engages the antinomian power of these beings through a threefold system of yoga involving the visualization of the maṇḍala deities, the recitation of their mantras, and the direct experience of absolute reality. The Mahāmāyātantra also incorporates the transgressive practices that are the hallmark of earlier tantric systems such as the Guhya­samāja­tantra,1 specifically the ingestion of sexual fluids and other polluting substances. The tantra promises the grace of Mahāmāyā in the form of mundane and transcendent spiritual attainments (siddhi) to those who approach it with diligence, courage, and devotion.

Though it is difficult to pinpoint precisely when the Mahāmāyātantra first began to circulate within Buddhist tantric communities, the work rose to prominence toward the end of the first millennium ᴄᴇ as part of an efflorescence of new tantric material associated with yoginīs, ḍākinīs, and other female divinities. The designation Yoginītantra was conferred on these texts precisely because of their incorporation of a more feminized vision of divinity and power. Yoginītantras vary in style and content, ranging from somewhat disordered and obscure works like the Cakra­saṃvara­tantra to refined and doctrinally coherent texts such as the Kāla­cakra­tantra. With its distinct narrative style, focus on the mundane spiritual attainments, and near total absence of clearly articulated Buddhist doctrine, the Mahāmāyātantra demonstrates a thematic and rhetorical similarity to the Cakra­saṃvara­tantra, one of the earliest Yoginītantras.2 Additionally, considering instances of intertextuality with the earlier Guhya­samāja­tantra,3 and a seeming lack of awareness of more doctrinally and structurally developed Yoginītantras such as the Hevajra­tantra and Kāla­cakra­tantra, it is reasonable to assume that the Mahā­māyā­tantra was among the earlier of the Yoginītantras, appearing in Buddhist tantric circles in the late ninth or early tenth centuries. It had certainly gained enough popularity by the eleventh century to draw the attention of the prominent monastic scholars of the period.

In India the Mahā­māyā­tantra probably circulated within both the major monastic institutions and the communities surrounding charismatic tantric masters. Foremost among the monastic scholars who commented on the text was Ratnākaraśānti,4 whose Guṇavatīṭīkā (A Commentary Endowed with Qualities) grounds the often enigmatic verses of the Mahā­māyā­tantra in mainstream Buddhist philosophy, especially that of Yogācāra. Kṛṣṇavajra’s commentary,5 the Mahā­māyā­tantrasya vṛtti smṛti (Recollection: A Commentary on the Mahāmāyā Tantra), frequently cites the oral tradition, pointing to the body of unwritten instructions that present the practical techniques almost completely absent in the Mahāmāyātantra itself. Contributions to the Mahāmāyātantra corpus came from outside the walls of the monastery as well. Kukkuripa and Kṛṣṇācārya, both of whom would eventually be counted among the eighty-four mahāsiddhas, composed practice liturgies (sādhana) for the tantra, while the siddhas Nāropa and Kaṇha are said to have taught the tantra to Marpa, thus facilitating its transmission to Tibet.6

The Mahāmāyātantra arrived in Tibet in the early eleventh century as part of the second wave of Buddhist teachings to reach the Land of Snows. It appears to have entered Tibet via two distinct lines of transmission, through Marpa Chökyi Lodrö and Gö Lhetsé. The lineage of Gö Lhetsé, a prolific translator and important teacher of the Guhya­samāja­tantra,7 appears to have died out, though the translation of the Mahāmāyātantra and the majority of its associated texts preserved in the Kangyur and Tengyur are his. The converse is true of Marpa; if he ever produced a translation of the Mahā­māyā­tantra, it has been lost, and yet it is his lineage, as passed through his disciple Ngoktön Chöku Dorjé (rngog ston chos sku rdo rje, 1036–1102), that continues to be transmitted to this day. The Mahāmāyātantra also received a great deal of attention from the Tibetan polymath Tāranātha (1575–1634), who composed two commentaries and a practice manual for it.8

The Mahāmāyātantra describes, in verses both terse and enigmatic, the practices and attainments associated with the deity Mahāmāyā and her retinue of yoginīs. Yoginīs, and their close counterparts the ḍākinīs, are renowned for their ability to grant mastery of temporal and transcendent spiritual attainments to devoted and courageous practitioners who are willing to brave an encounter with these often ferocious beings. In this tantra Mahāmāyā is invoked as the queen of ḍākinīs, the queen of the yoginīs, and the supreme secret of these secret goddesses, making her the paramount source of spiritual attainment. The tantra promises the accomplishment of such powers through the visualization of its deities and their maṇḍalas, the manipulation of the body’s subtle energies and the cultivation of the power of transgression through the ingestion of impure substances. And yet these practices are only hinted at within the tantra itself; the specifics are reserved for initiates and are to be spoken only by a lineage holder. Thus, as is the case with most tantras, this text is meant to be practiced only after receiving initiation and instruction from a qualified master.

The Mahāmāyātantra’s three chapters and eighty-one verses depart from the traditional narrative structure of Buddhist scripture. Buddhist tantras typically follow the sūtra tradition by beginning with a formulaic introduction meant to establish the time, place, audience, and above all the authority of a given scripture. This formula, which begins “Thus I have heard...” (evaṃ mayā śrutaṃ), is absent in the Mahāmāyātantra, which begins instead with a perfunctory “Now...” (ataḥ).9 This deviation is in part explained by the fact that this is not a discourse of the Buddha, but rather a dialogue between two deities associated with the tantra. However, the lack of a formal introduction leaves it unclear precisely who these deities are. The interlocutor appears to be none other than Mahāmāyā herself; verse 1.25 invokes her by name as the recipient of the tantra’s instruction. The speaker, however, is never explicitly identified within the text. Kṛṣṇavajra ventures a guess, suggesting Vajraḍākinī as the source of the teaching.10 Ratnākaraśānti takes a different approach and, instead of concerning himself with identifying figures not explicitly named in the text, considers the Mahāmāyātantra to be derived from a much larger mythical compendium of tantras (which he refrains from identifying).11 It is in that collection, he declares, that one may find the traditional introduction that establishes the important details of the Mahāmāyātantra’s setting.

From this abrupt beginning, the first chapter continues with several verses invoking Mahāmāyā’s qualities and describing the powers acquired through her successful propitiation. Tantras centered upon yoginīs and ḍākinīs, whether Buddhist or non-Buddhist, typically emphasize the attainment of mundane powers over the transcendent; in this the Mahāmāyātantra is no different. The yogī who masters the practices of Mahāmāyā will, the text promises, be able to magically influence and attack others, acquire whatever he desires, fly through the sky, travel through other realms, become invisible, possess the bodies of others, and become immortal. Compared to this comprehensive and repetitive list of mundane attainments, the tantra makes only passing reference to the more altruistic and transcendent attainments. Only once does the text mention that its attainments can be used to benefit others (1.28), and only once does it explicitly announce that its practice can result in buddhahood (3.25).

Following this exhaustive account of the attainments, the first chapter closes with a description of a short sequence of visualization. The reader may be puzzled at this point to find that Mahāmāyā, who had, up to this point, been invoked using explicitly feminine epithets, is suddenly referred to using undeniably masculine terms. Though some suggest this is the tantra of a male deity with a feminine name,12 Mahāmāyā is a female deity, as the verses of invocation make clear—she is the great Queen Mahāmāyā, the mother of the guhyakas, and the queen of yoginīs. She is consistently addressed using a specifically feminine epithet, vidyā, that simultaneously invokes her status as the embodiment of knowledge and as the female deity presiding over a maṇḍala. And yet here, and in the third chapter where her iconography is fully described, she becomes the male Heruka, the Virile One (vīra) embracing the consort Buddhaḍākinī.

This gender ambiguity is clarified when we consider that the tantra describes Mahāmāyā as a feminized form of absolute reality. She is said to pervade everything in the three worlds, to be the source of all the gods, and to create, sustain, and destroy the universe. Most importantly, she is exactly what her name suggests, the “great illusion” that constitutes apparent reality. When she takes embodiment, she can do so in any form necessary, which in the context of this tantra is Heruka, the male deity most frequently found at the center of tantric maṇḍalas. Ratnākaraśānti makes explicit the ontological primacy of the feminine Mahāmāyā; in his commentary he equates her with Vajradhara, the embodiment of absolute reality, who is typically male, and identifies her as “she who has the form of Heruka.”13

The second chapter is the tantra’s shortest and most challenging. Its verses only hint at the techniques and visualizations to which they refer. The communities in which the Mahāmāyātantra circulated would have guarded their teachings and techniques closely, making complete explanations accessible only to an intimate circle of initiates. The tantra’s verses—especially those that appear to refer to specific techniques—are not meant to be edifying, only allusive. It is the role of the tantric master to unpack each verse and convey its practical content to those prepared to receive it. Thus in this chapter we are offered only the barest of introductions to what was likely a complex sequence of techniques.

The fifth verse of the second chapter introduces the primary structure of the tantra’s practices, one the later Tibetan commentators would identify as a framework for the development stage and completion stage practices unique to the Mahā­māyā­tantra.14 Verse 2.5 states:

Kṛṣṇavajra provides the necessary elaboration:

The visualizations, recitations, and sequences of practice that follow in the text are to be understood and employed in terms of this threefold structure.

The remainder of the chapter presents a series of practical techniques for attaining mastery over the phenomenal world and generating bodily energies and fluids for use in ritual practice. Each verse from 2.7 to 2.14 alludes to a specific visualization centered on a single mantric syllable and its associated deity, the practice of which leads to the stated outcome. Yet it is impossible to determine, based on the verses alone, what practices are being described. It is left to the commentators and the oral tradition to fill in the detail. Consider verse 2.7:

Ratnākaraśānti comments:

It thus becomes apparent that the words of the tantra itself provide merely an outline, a shorthand version for tantric practitioners already well versed in its practices. Likewise, each of the subsequent verses of this chapter points to complex meditation techniques, a type of knowledge that is, as verse 2.17 indicates, “secret, obscure, and unwritten.”

The third and final chapter brings the tantra to a close with a description of the rites for preparing and consuming impure substances, followed by a complete presentation of the iconography of Mahāmāyā and her retinue. The cultivation of the power of transgression through deliberate consumption of impure, polluting substances is an important aspect of Buddhist tantra. Notions of purity and pollution have been formative elements of South Asian identity since Vedic times. Such an identity is destabilized through contact with and ingestion of impure substances and bodily fluids. Indian Buddhist tantras, especially those of the later Yoga tantra class such as the Guhya­samāja­tantra,17 positioned the consumption of sexual fluids at the climax of the ritual process so as to harness the force of the transgressive act into a powerful soteriological moment. The same holds true for the ingestion of illicit meats: their ritual consumption negated a social identity formed through the observation of dietary and behavioral proscriptions. Through the eating of taboo substances, practitioners’ bonds with mundane society are fully severed and their acceptance into the community of spiritual adepts is confirmed.

Yoginītantras, including the Mahā­māyā­tantra, carried on the transgressive practices that developed in tantras such as the Guhya­samāja­tantra and the Guhya­garbha­tantra. In the Mahā­māyā­tantra, the consumption of sexual fluids is presented at the end of the second chapter and the ingestion of illicit meats in the early verses of the third. The final two verses of the second chapter describe, in a typically occluded fashion, the ingestion of sexual fluids that marks the culmination of the meditation sequence. The verses instruct the yogī to keep his mind free of concepts and take the “wish-granting jewel” between his thumb and ring finger and place it in his mouth. Kṛṣṇavajra explains that this refers to the ingestion of “relative bodhicitta” (kun rdzob kyi byang chub sems, i.e. semen) after its prolonged retention in the tip of the penis.18 In the parlance of the later tradition, this marks the completion stage section of the practice. In the words of the tantra itself, the ingestion of sexual fluids triggers “everlasting spiritual attainment.”

Moving into the third chapter, the tantra introduces a sequence of verses outlining the rites associated with the ingestion of sacramental substances. The yogī is instructed to gather the five illicit meats (elephant, horse, cow, dog, and human), roll them individually into pellets, and store them for seven days. Next they are mixed together and stored in the cavity of a rotting jackal for an additional seven days. Finally they are taken out and consumed, resulting in an experience of “the great fruitions” (mahāphalāni, 3.6), which Ratnākaraśānti identifies as “the state of wisdom.”19 Though the five illicit meats are common enough in tantric literature, the especially polluting addition of a jackal corpse appears to be a unique contribution of the Mahā­māyā­tantra. We can understand this distinctly unpalatable process as intended to collapse the binaries of pure and impure, precipitating a more complete rejection of dualistic concepts and, as verse 3.1 states, leading to “omniscience.”

The final section of the tantra concerns the method of practice, the sādhana, for the deity, including the details of the visualization of Mahāmāyā and her maṇḍala. The stages by which a tantra is to be practiced are frequently concealed within the text—the order will be scrambled and the details elided.20 The act of organizing these elements is left to tantric teachers and lineage masters, who composed independent ancillary texts to codify the proper sequence of practice. These texts incorporated practices drawn from the oral tradition and embedded the basic outline provided by the tantra within an established sādhana framework that normally included common Buddhist elements such as refuge in the Three Jewels, the generation of the aspiration for awakening, the offering of confession, and so forth. Thus while the core material for a sādhana is found within the tantra itself, each sādhana is unique, reflecting styles and interpretations that are as diverse as their authors. Sādhanas composed on the Mahā­māyā­tantra by the Indian authors Ratnākaraśānti, Kukkuripa, Ratnavajra, and Kumé Dorjé (sku med rdo rje), as well as a number composed by Tibetan authors, have been preserved in Sanskrit and Tibetan.

Within the Mahā­māyā­tantra, although verse 3.7 announces the commencement of the sādhana, the majority of practices have already been introduced in the preceding two chapters. But it is only here, at the end of the tantra, that we are finally given the complete iconographic description required for the self-visualization of the deity that precedes all other practices.21 The visualization begins by imagining a red lotus flower with four petals, in the middle of which sits a single subtle sphere. From this sphere, four additional subtle spheres emerge and come to rest on each of the four petals. These four spheres then transform into four yoginīs, each distinct in color and wielding the specific implements described in the tantra. They sing a song of invocation in the language of the yoginīs 22 to the central subtle sphere, summoning Mahāmāyā and consort from their essential state, first in the form of mantra syllables, then in full iconographic detail. Mahāmāyā takes the dark blue form of Heruka embracing the red Buddhaḍākinī. (S)he has four faces—blue in the east, yellow in the south, red in the west, and blue-green in the north. (S)he has four hands, each holding a different implement: a skull cup, a khaṭvāṅga, a bow, and an arrow.

Once this five-deity maṇḍala has been fully imagined, the practitioner is prepared to apply any of the diverse practices that have been described throughout the work: the visualization of mantra syllables and additional maṇḍala deities, the manipulation of subtle energies, or the consumption of sexual fluids and illicit meats. Through identification with Mahāmāyā and engagement in these potent practices, the diligent practitioner will be rewarded with mundane powers, reach “highest attainment of suchness” (3.25), and, as we are told in the tantra’s final verse, be forever protected by Mahāmāyā, “the queen of the yoginīs.”

This translation is based upon a comparative edition, made by the translators, of the Tibetan recensions of the Mahā­māyā­tantra drawn from the Degé (sde dge), Lhasa, Narthang (snar thang), Peking, and Stok Palace (stog pho brang) editions of the Tibetan Kangyur. Though the tantra itself no longer exists in Sanskrit, Ratnākaraśānti’s important commentary, the Guṇavatīṭīkā, is preserved in two Sanskrit manuscripts held in Nepal’s Royal Archive, which have been edited by the Central Institute of Higher Tibetan Studies. Because the Guṇavatīṭīkā attests to the majority of the Sanskrit terms, it allows access to much of the Mahā­māyā­tantra in its original language. We have therefore given careful consideration to both the Sanskrit edition of this commentary and the manuscripts upon which it is based. Where the Sanskrit term is available, we have privileged the semantic range of the Sanskrit over that of the Tibetan, allowing us to recapture something of the flavor of the original Sanskrit. The commentaries of Ratnākaraśānti and Kṛṣṇavajra were consulted throughout the translation process and greatly influenced our reading of the root text.

The Mahā­māyā­tantra is a challenging text. It is thematically inconsistent and disjunctive in places, and at times the referent of a given verse is unclear. Though this is certainly due in part to attempts by its authors to conceal the meaning or proper sequence of the tantra, it is clear that a grammatically flawed work is hiding behind the Tibetan translation. This is common among Buddhist tantras, as their authors did not always possess a solid command of the Sanskrit language. Where possible we have clarified the text to make it more readable, but there remain passages that are enigmatic and difficult to follow.

In the translation, the numbers within square brackets preceded by F. indicate the folio numbers of the Degé edition of the Tibetan; the numbers without letters indicate the page numbers of the dpe bsdur ma (comparative) edition of the Tibetan (see bibliography).