What distinguishes Tantra?

Many people see David White's Kiss of the Yoginī as a definitive book on Tantra in its original South Asian context (as its subtitle claims), so here I'm presenting a section from the conclusion of my PhD dissertation (Berkeley, 2014), which argues both that White's book is mistaken on key points, and, perhaps more importantly, presents a definition of Tantra that premodern South Asian tāntrikas themselves would not have agreed with.

I would have the reader note that this is not just my opinion, but rather the result of careful research in the primary sources and consultation with other scholars. On the latter point, I want to acknowledge the pioneering work of Shaman Hatley, who agrees with my criticism of White and who is doing excellent scholarly work that will both correct and supplant White's (e.g., on sexual rituals in original Tantra). Hatley is preparing a new edition and translation of the Kaula-jñāna-nirṇaya, the Sanskrit work that White wrongly reads sexual rites into. May it soon see the light of day!  And now, on with this week's post . . . 

The real nature of Tantra

What distinguishes Tantra from other streams of Asian religion? I use the term Tantra to denote the teachings and practices taught in the Śaiva (and Śākta-Śaiva) tantras and āgamas and their commentaries—teachings and practices that were available only to those who had taken Tantric initiation as taught in those scriptures. In a wider sense, we can take Tantra to denote the esoteric traditions of all the South Asian religions that were influenced by these innovations within Śaivism: thus Tantra collectively can denote Tantric Śaivism, Tantric Buddhism, Tantric Vaiṣṇavism, and Tantric Jainism. What gave Tantra (so defined) its specificity in the medieval period was:

a) the liberative initiation ritual (called nirvāṇa-dīkṣa in Śaivism) that utilized an innovative technology featuring the entrainment of mantra, breath, ritual gesture (mudrā), and visualization in elaborate sequences that operated in terms of both theologically defined micro-macrocosmic correlations and in terms of complex cosmologies that were mapped onto initiatory diagrams called maṇḍalas;

b) yogic practices for daily sādhana that utilized the same ritual technologies and recapitulated the actions of the initiation rite, gradually displacing the socio-culturally constructed self in favor of identity with the Deity; and

c) specific yogic and charismatic techniques that triggered altered states of consciousness in initiates, including but not limited to experiences construed as being the penetration of the layers of one’s being by the energy of the Deity and/or the dissolution of limited selfhood into the Deity itself.

These are the central features that define Tantra and that distinguish it from other forms of Indian religion, in sum: a) liberating initiation, b) innovative yogic technologies, and c) samāveśa. It is simply not that case that, as David White has it, “sexualized ritual practice is the sole truly distinctive feature of South Asian Tantric traditions” (2003: 13). The above three features are distinctive, especially if one realizes that these specific technologies (maṇḍalas, mudrās, uccāra, bīja-mantras, etc.) are only found in non-tantric religious spheres due to Tantric influence. White does state that sexual practices “never constituted the mainstream of Tantric or Kaula practice,” but he contends that the mainstream practice in question was “satisfying multiple and petulant divinities by feeding them” (2003: 14). Certainly gratifying one’s mantra-deity with offerings was central to daily Tantric practice, but the idea of gratifying “multiple petulant divinities” can only apply to the Kaula Yoginī cult, which was not in fact part of the mainstream of the tradition (Sanderson’s distinction of the Kulamārga from the Mantramārga is relevant here). But there are deeper misunderstandings at work here, for White effectively takes this Yoginī cult to be the whole of the tradition. He posits a distinction between “hard-core” Tantra (which is original, in his view) and “soft-core” Tantra (which is a bowdlerized, aestheticized, and “semanticized” version of the “hard core”) and then asserts that:

In both cases, the female Yoginī ‘seizes’ or ‘possesses’ her counterpart. However, whereas in the former case [the “soft core” of the Tantric mainstream], she simply preys upon her human victim (paśu), in the latter [the Kaula “hard core”], the male partner takes an active role, inducing a sort of ‘mutual possession’ (samāveśa) in a sexual mode. (2003: 14).

This dissertation has surveyed virtually all the relevant Śaiva literature for the first five centuries of the documented existence of Tantra and did not turn up a single piece of evidence that corresponds to what White describes here. First, we never saw any mutual possession involving Yoginīs, let alone “mutual possession in a sexual mode.”[1] Second, samāveśa almost never involved sexual intercourse of any kind in our sources. Third, samāveśa means (lit.) thorough or complete entry (samyag āveśana), not mutual entry. Fourth, possession forms no part of the daily practice of the non-Kaula Tantrics (= White’s “soft core”). Fifth, we never saw the use of the term graha(ṇa) (“seize”) in connection with salutary possession. Sixth and seventh (the most egregious errors), Yoginīs are not invoked by non-Kaulas, nor do they “prey on” the Tantric practitioner, who by definition is not a paśu, since that term can only refer to non-initiates. In the latter assertation, White is confusing Tantric propitiation of Yoginīs with occasional assertions in the literature that Yoginīs suck the vital essences from paśus (non-tāntrikas) (e.g., Netratantra ch. 20), an extraction said to be ritually imitated by certain extreme worshippers of Bhairava and Kālī described in the far-left scriptures (e.g., the Jayadratha-yāmala and the Brahma-yāmala),[2] accounts which he seems to conflate with fictional fantasy stories of Yoginīs such as those in the Kathāsaritsāgara and Mālatī-mādhava. Furthermore, even White’s categories of “hard core” and “soft core” are confused, for he associates Abhinavagupta’s lineage with the latter category (2003: 14-15), even though Abhinavagupta himself strove mightily to defend and endorse the Kaula sexual practices, and did not bowdlerize them in the slightest, as can be seen in the rather explicit account at Tantrāloka 29, which is the only Śaiva exegetical source to teach the sexual rite or kulayāga. Thus Abhinava’s practices did not “shade into those of orthoprax brahmanic ritual” (2003: 15) in any way. Finally, White claims that the “soft-core” practice of the “high-caste Hindu practitioners” is greatly outnumbered by the “Tantric mainstream” (Ibid.), a statement based on the erroneous (and completely unsupported) notion that the “cults of (predominantly female) village deities whose worship was often conducted by the socially and culturally marginalized” were, by that very definition, Tantric cults (2003: 5). This obviously neglects the central feature of the Tantric tradition, that of initiation without regard to caste into specifically circumscribed communities of spiritual elites who performed a rigorous daily ritual and yogic practice intended to bring about their spiritual liberation. Conflating Tantra with the worship of village deities not taught in Tantric texts allows one to say virtually anything one likes about “Tantra,” including that it was the “predominant religious paradigm . . . of the great majority of the inhabitants” of India (2003: 3), which it was not. There are countless other errors in White’s book that we could unpack, but suffice to say that we agree with Hugh Urban in his characterization of White’s view of Tantra as “Kaulo-centric” and his view of Kaulism as “sexo-centric” (2006: 283).[3]

Criticizing other scholars’ misunderstandings is not something I take any pleasure in. The purpose of it here is to point out how far the discipline of Religious Studies has to go in understanding Tantra (for White is considered a foremost scholar of the Hindu Tantra by many in that discipline, at least in the American academy), and how well served they would be by putting aside their assumption about what Tantra is and examining the careful philological work being done on Tantric texts and inscriptions. Such work, while painstaking, slowly builds up a picture of things that constitutes a significantly closer approach to whatever degree of historical truth can be determined from textual materials.

(Feel free to email me for a copy of my dissertation if you want to learn more!)


[1] The closest approach to this might be the Vijñāna-bhairava verse (69) in which śaktyāveśa denotes sexual intercourse with a consort, but this is with a human woman not identified as a Yoginī in the text or its commentary.

[2] See Sanderson 1985: n89.

[3] However, we should also note that White gave a much more salient and historically grounded account of Tantra in 2000, when he defined it as “that body of beliefs and practices which, working from the principle that the material universe is nothing other than the concrete manifestation of the divine energy of the godhead that creates and maintains that universe, seeks to appropriate and channel that energy, within the human microcosm, in creative and emancipatory ways” (2000: xxiii). The main problem with this definition, however, is that the key clause beginning “the material universe” excludes the Śaiva Siddhānta, which as noted is not only Tantric, but constitutes the broad base of the tradition (sāmānya-śāstra), establishing the ritual forms, yogas, and even doctrines which other lineage-groupings nuance or deviate from in varying degrees.