Dakini Series 3.3: Is Dakini Hindu or Buddhist?
This is the third part of the third article in our Dakini series, drawn from the Dakini video series on the Siddha Tantra Arts YouTube channel. The previous two parts engaged with the early textual evidence and explained why advanced tantric texts prohibited even uttering the word Dakini. This article addresses the most persistent scholarly assumption of all: that the Dakini "evolved" over centuries from a malevolent folk spirit to a positive deity, and the related claim that the wisdom-Dakini of Buddhist tantra is essentially a Buddhist creation. Both are wrong, and the evidence to demonstrate it is not subtle. The full video is embedded below. And if something in this piece shifts something in you, or raises a question, we would genuinely like to hear from you. Leave a comment at the bottom.
Is Dakini Hindu or Buddhist?
Scholars trace the Ḍākinī from "malevolent in early texts" to "positive in Kaula texts" and read this as the Ḍākinī evolving over time. Some go further and argue that the Ḍākinī is essentially a derivative: that she emerged from the Mātṛkā cult, that she was originally just one of the fierce mothers in Rudra's retinue, and that the "positive" Ḍākinī was a later innovation, perhaps even a Buddhist reinterpretation imported back into Hindu tantra.
I want to address this directly. Because this "Ḍākinī-evolved-from-Mātṛkā" narrative is one of the most persistent scholarly assumptions, and it needs to be examined honestly.
What the Inscriptions and Stotras Actually Show
The evidence scholars cite for the evolution model is real. The Gaṅgdhār inscription of 424 CE, one of the earliest datable references, describes a shrine of the Mātṛkās that is, in the inscription's own words, ḍākinī-samprakīrṇam: permeated with Ḍākinīs. The shrine is described as exceedingly awe-inspiring, aty ugram. And it is explicitly said to be endowed with the power of the tantras, tantrodbhūta. So here you have Ḍākinīs appearing in a Mātṛkā context, in the fifth century, on stone.
Around the same period, the Ur-Skandapurāṇa contains a stotra (a hymn) addressed to Nandin by Agni, in which one of the epithets used is Ḍākini-rudra: the Rudra of the Ḍākinīs. This sits alongside epithets like Śmaśāna-rudra, Bhasma-rudra, Pramatha-rudra: the Rudra of the cremation ground, the Rudra of ashes, the Rudra of the wild ones. So Ḍākinī appears in the Rudra-Gaṇa complex, the retinue of Shiva, as an acknowledged category of fierce beings under Rudra's sovereignty.
Later, in the Mega-Skandapurāṇa, Ḍākinī and Śākinī appear in the retinue of Vīrabhadra during his assault on Dakṣa's yajña, alongside the nine Durgās, the Pramathas, Kubera's agents, and the circle of sixty-four Yoginīs.
And, this is important, even the Saiddhāntika tradition, the very Siddhānta stream that scholars treat as uniformly anti-Ḍākinī, has a reference in the Sūkṣmāgama where Ḍākinī is installed as a dvāradevatā: a door-guardian deity of the Śaiva temple sanctum. Worshipped on the left side. In sequence. With Mahālakṣmī above.
Sightings Are Not Birth Certificates
A scholar looks at all of this and reads a story of evolution. Ḍākinīs start as fierce mothers in the Mātṛkā cult. They appear in Rudra's retinue. Eventually they get "promoted," from pest to door-guardian, from fearsome spirit to object of sādhana. The positive Ḍākinī is a later development. The malevolent one is the original.
But this reading contains a fundamental assumption that I want to challenge. It assumes that the first text to mention something is the first time that thing existed. And it assumes that because the Ḍākinī appears alongside Mātṛkās in some contexts, she must be derived from them.
Neither follows.
The Gaṅgdhār inscription does not say the Ḍākinīs were created for this shrine. It says the shrine is permeated with them, samprakīrṇam, as though they were an already-existing force that infused the space. The Ur-Skandapurāṇa does not introduce Ḍākini-rudra as a new concept. It lists the epithet in a stotra alongside long-established categories, Śmaśāna-rudra, Pramatha-rudra, as though Ḍākini-rudra was already part of the known pantheon. These texts are not birth certificates. They are sightings. They tell you when the written record first noticed something, not when that something began.
And the Mātṛkā-origin theory has a deeper problem. Siddha Dharma is unambiguous: Ḍākinī is a distinct yoni, a separate class of being, not a subcategory of Mātṛkā, not a derivative of Yoginī, not a regional variant of the fierce mothers. Siddhapedia is explicit: Ḍākinīs are "a distinct class or yoni, separate from other celestial entities such as Devīs and Apsarās." The fact that Ḍākinīs appear alongside Mātṛkās does not mean they evolved from Mātṛkās, any more than the fact that Gaṇeśa appears alongside Śiva means Gaṇeśa evolved from Śiva.
Ḍākinīs appear in Mātṛkā contexts because powerful śakti forces converge in powerful sacred spaces. They appear in Rudra's retinue because Rudra is sovereign over all fierce beings. They appear in the Vīrabhadra narrative because all categories of power mobilize when cosmic order is at stake. Their presence in these contexts is expected. Their origin from these contexts is an unwarranted inference.
Why No Tradition Could Place Her
And here is something I want to say about why Ḍākinīs have been so difficult for every tradition to place.
Siddhapedia records something remarkable about Ma Kurukullā, the supreme Ḍākinī goddess. It says that when all the traditions of Sattva, Rajas, and Tamas did not accept Ḍākinīs, when no established system would give them a place, it was Kurukullā who created the Kula traditions to house them.
Think about what that means. The Sattvic traditions did not accept Ḍākinīs: too fierce, too raw, too undomesticated. The Rajasic traditions did not accept them: too unpredictable, too independent, too difficult to harness for worldly purposes. Even the Tamasic traditions did not fully accept them: too sovereign, too uncontrollable, not reducible to mere instruments of power. No existing box could contain them.
And this is not a failure of those traditions. This is the Ḍākinī being what she is. She is scandalous. She is not your goodie-two-shoes deity. She does not fit neatly into established frameworks of purity or pollution, of benevolent or malevolent, of high or low. The Ucchiṣṭa Ḍākinīs (those associated with impurity, with what has been touched, with what has been tasted) have ritual requirements that are fundamentally different from anything in the Sattvic worship framework. A tradition built around ritual purity simply has no category for a being whose sādhana begins where purity ends.
And this is precisely what makes the Ḍākinī so interesting. She is not evil. She is ungovernable. She does not comply with the taxonomies that other traditions use to organize the divine feminine. And when a being does not comply with your taxonomy, you have two options: you can call her dangerous and warn people away, or you can recognize that your taxonomy is incomplete.
The Bhūtasrotas chose the first option. The Vidyāpīṭha chose the first option. Even the Siddhānta tradition, which eventually installed Ḍākinī as dvāradevatā in some temple contexts, was encountering fragments of what the oral tradition always held in full. Perhaps they received wisdom over time. Perhaps their surviving record is simply more incomplete than the reality was. Either way, their shifting representations are not the Ḍākinī changing. They are the record catching up.
The Himalayan Siddhas chose the second option. They did not try to box the Ḍākinī. They did not place her in the camp of "good" or "bad." They recognized her as a separate yoni with her own nature, her own protocols, her own terms of engagement. And they built the Ḍākinī Kula Maṇḍala, the complete system, to hold her in all her fullness: Deva Ḍākinīs to Nikṛṣṭha Ḍākinīs, sattvic to tamasic, from flower offerings to cremation-ground sādhana. The complete spectrum, held in a single living tradition.
The Buddhist Question
For those who need more than metaphor, here is something concrete.
The most careful scholars themselves acknowledge that there must have been a class of Hindu texts, the Ḍākinī-tantras, that predated the Buddhist adoption of Ḍākinī worship. The Buddhist philosopher Dharmakīrti, in the seventh century CE, polemically mentions "Ḍākinī-tantras" alongside "Bhaginī-tantras," clearly Hindu tantric texts.
Now, let me show you what the Buddhist tradition itself was doing with Ḍākinīs at this same period. Because this is where the argument becomes impossible to dismiss.
The Mañjuśrīya Mūlakalpa is classified as the earliest surviving Indian Buddhist tantra. Its core dates to the sixth century. Its full compiled form was complete by around 775 CE. It is a massive text, originally in Sanskrit, translated into Tibetan and Chinese, and it is the foundational Kriyā-tantra of the Buddhist system. This is not an obscure text. It is canonical. It is in the Kangyur. The 84000 translation project published its English rendering in 2020.
And in its very first chapter, the Mañjuśrīya Mūlakalpa lists the beings summoned by the Yamāntaka-mantra to attend Mañjuśrī's teaching. It is a demonological catalogue. And here is where Ḍākinī appears:
"...bhaginīs, great bhaginīs, ḍākinīs, great ḍākinīs, cūṣakas, great cūṣakas... rogas, great rogas, apasmāras, great apasmāras, grahas, great grahas, ākāśamātṛs, great ākāśamātṛs..."
Listen to the company. Cūṣakas: essence-suckers. Rogas: disease-spirits. Apasmāras: epilepsy-demons. Grahas: seizure-spirits. And right there between the bhaginīs and the essence-suckers: ḍākinīs.
Notice also: the Ḍākinī appears immediately alongside the Bhaginī, the "sister-spirit." The same pairing that Dharmakīrti used when he named the lost Hindu texts: ḍākinī-bhaginī-tantrādiṣu. The Buddhist canonical text and the Buddhist philosopher are drawing on the same Hindu source vocabulary.
And the Mañjuśrīya Mūlakalpa does not stop at listing them. It contains prayogas, ritual procedures, specifically designed to counter Ḍākinī attacks. Counter-Ḍākinī mantras. Apotropaic technology to neutralize their afflictions. Exactly the same functional category as the Uddāmareshvara's Sarva-bhūta-ḍākinī-damana-mantra. Exactly the same as the Jayākhya-saṃhitā's Mantra-rāja deployed for "curtailing all Ḍākinīs."
This is not a Hindu text. This is the Buddhist tradition's own foundational tantra. And in it, Ḍākinīs are listed between essence-sucking spirits and epilepsy-demons. They are objects of subjugation by Yamāntaka's power, not agents of wisdom. The Buddhists themselves did not yet know the positive Ḍākinī.
The Transplant
Now hold that in one hand. And in the other hand, hold what was happening on the Hindu side at the exact same period.
By the time the Mañjuśrīya Mūlakalpa is being compiled, eighth century, Hindu sources have already had centuries of positive Ḍākinī engagement. The Gaṅgdhār inscription, 424 CE, three hundred and fifty years earlier, already shows Ḍākinīs with cultic prominence in a Mātṛkā shrine. Dharmakīrti, a hundred and fifty years before the MMK's compiled form, already names Ḍākinī-tantras as an existing Hindu genre. The Brahmayāmala, roughly contemporary, already treats Śākinīs and Ḍākinī-class beings as transmitters of the highest Kula knowledge. Matsyendranath's Kaula-jñāna-nirṇaya lists Ḍākinī-siddhi among the attainments of advanced practice.
So at the moment the earliest Buddhist tantra is writing counter-Ḍākinī mantras, the Hindu Kaula tradition has already been worshipping them for centuries. The Buddhist "wisdom Ḍākinī," the sky-dancer, the enlightened feminine, did not emerge from Buddhist soil. It was transplanted.
And the record of how the transplant happened survives.
In the Sarva-tathāgata-tattva-saṃgraha, a Buddhist tantra from around the eighth century, the Hindu goddess Cāmuṇḍā, also known as Vajrakālī, is "vajrified," given a Buddhist identity. Her new name: Vajraḍākinī. That is the first time a Buddhist text assigns a Ḍākinī a positive, "vajra" identity. Not through organic Buddhist development. Through a naming ceremony performed on a borrowed Hindu deity.
And here is something extraordinary. An eighth-century Chinese commentary on the Mahāvairocana-sūtra, likely by the master Śubhākarasiṃha or his disciple Yixing, makes the association of Ḍākinīs and their practices with Śiva and Śaivism explicit. The Buddhist commentators themselves were still transparent about where this material came from. They had not yet erased the source. The concealment came later.
The full Buddhist Ḍākinī-as-wisdom-being did not arrive until the Yoginī-tantras (the Cakrasaṃvara, the Hevajra) of the eighth to tenth centuries. That is when Ḍākinīs finally became central to Buddhist practice. And even a scholar sympathetic to the Buddhist tradition, Shaman Hatley, notes that at the STTS stage, there is "as yet little indication of their transformation into the wild and ambivalent, yet supremely powerful and potentially beneficent, sky-wanderers of the Yoginī-tantras." The conversion was still incomplete.
In Japan, the tradition even preserves a literalized version of this conversion story. The Ḍākinīs were "subjugated and converted to Buddhism by the Buddha Vairocana under the guise of Mahākāla." Subjugated. Converted. The language is transparent. This is not organic development. This is theological capture.
So here is the picture, stated plainly.
A Hindu Ḍākinī-tantra tradition existed before the seventh century: Dharmakīrti names it. The Buddhist tradition at this same period still treated Ḍākinīs as hostile entities to be fought: the Mañjuśrīya Mūlakalpa proves it. The Hindu Kaula tradition had already developed positive Ḍākinī worship centuries before the Buddhists adopted it. The Buddhists then borrowed the Ḍākinī, performed a "vajrification" ceremony on a Hindu goddess, acknowledged the Śaiva source in their own commentaries, and eventually elevated her to a central wisdom-being in their Yoginī-tantras.
The Buddhist "wisdom Ḍākinī" is not a Buddhist discovery. It is a Hindu Ḍākinī in Buddhist robes.
And I want to be clear: I am not saying this to attack Buddhism or Buddhist practitioners. Buddhist sādhakas who practice with the Ḍākinī with sincerity and devotion are practicing with a real force. The Ḍākinī's power does not diminish because the lineage that transmitted her was not acknowledged. She is still herself, in any setting, by any name. But the history matters. Because when someone claims the Ḍākinī originates from Buddhist tantra, they are erasing the Hindu tradition that held her for centuries before Buddhism encountered her. And that erasure is something I must correct.
The "Lost" Texts Problem
Now, the scholars note that no manuscripts of the original Hindu Ḍākinī-tantras survive. They call these texts "lost."
And then, this is crucial, they reconstruct what they think was in them. They posit that these lost texts probably overlapped with the Yoginī and Mātṛkā material of the Bhairava-srotas. They suspect that the Buddhist Yoginī-tantras borrowed the class name "Ḍākinī" from these posited texts while borrowing the content from Bhairava-Kaula sources. The language is revealing: "we suspect," "we believe," "posited," "it remains to be seen."
I want you to hear what is happening here. The entire scholarly model of Ḍākinī evolution rests on texts that the scholar has never read. They are reconstructing the contents of manuscripts they do not possess, and then building a historical narrative on those reconstructions. This is not dishonest. It is the best they can do with what they have. But it is also not fact. It is conjecture shaped by the assumption that the written record is all there is.
And consider what "lost" actually means in the context of Indian manuscript history. In 1931, a chance discovery in Gilgit, in the same northwestern Himalayan cultural zone as Kashmir and the Kaulantak region, revealed a massive cache of texts that completely reshaped our understanding of what traditions were active in that area and how early they were developed. Without that single accidental find, we would have far less evidence for how sophisticated goddess-centered tantra was in that exact region. One landslide, one flood, one raid, one fire, and centuries of textual tradition simply vanish from the scholarly record. Every time a lucky discovery surfaces from this corridor (Gilgit, Kashmir, the Nepalese archives) the picture gets pushed back earlier and becomes more elaborate than anyone expected.
So when we say no manuscripts of the Ḍākinī-tantras survive, that tells us something about the fragility of manuscripts. It tells us nothing about the fragility of lineages.
From the Siddha Dharma perspective, these texts were never lost. They are the knowledge that was, and continues to be, held within the Himalayan tradition of Kaulantak Peeth. The manuscripts may have perished. The lineage did not. The scholar must posit what was in the Ḍākinī-tantras because they do not have access to the living tradition that held them. Two models fit the evidence: either the Kaula tradition gradually "adopted" Ḍākinīs from lost external sources, or the Kaula tradition was the source from which fragments radiated outward into other streams. Both models account for the data. But only one requires inventing phantom texts.
What the Surviving Evidence Actually Shows
The surviving evidence itself favours the second model. Let me show you why.
The Brahmayāmala, also known as the Picumata, is one of the oldest Vidyāpīṭha texts. It is early. It is serious. And look at what it says about the Śākinīs, the Mātṛkās, and the Yoginīs:
"Mātṛ-yoginikā yāni śākinīnāṃ kulāni tu / sidhyanti sādhakendrasya yogenānena suvrate / kathayanti ca sadbhāvaṃ kulajaṃ jñānam uttamam"
"The Kulas of Mātṛs, Yoginīs, and Śākinīs confer siddhi on the chief of the sādhakas through this yoga, O lady of good vows, and they teach the true essence, which is the highest knowledge arising from the Kula."
Listen to what the Brahmayāmala is saying. The Śākinīs (kin of the Ḍākinīs) are not being treated as pests to repel. They are not being managed or countered. They are conferring siddhi. They are teaching. They are transmitting "the highest knowledge arising from the Kula." This is exactly, exactly, what Siddha Dharma describes as the Ḍākinī's fundamental function. Haraṇa. Transmission. The flame passing from wick to wick.
This is not a text "evolving toward" a positive Ḍākinī view. This is a text that already holds the understanding Siddha Dharma has always held: that Ḍākinī-class beings are transmitters of the deepest knowledge. The scholars read this as "an early version of something developing." The tradition reads it as the complete understanding surfacing in writing.
And the same Brahmayāmala maps different Ḍākinī types onto different tattva levels: Ḍākinīs emanated by the Kāla-tattva, Rudra-Ḍākinīs by the Māyā-tattva, Yoginīs by the Puruṣa-tattva, Lākinīs emanated by Ucchuṣma-bhairava. This is not a random assignment. This is a structured hierarchical mapping of different Ḍākinī categories across cosmic levels, the same structural principle as the Siddha Dharma Kula Maṇḍala's mapping of seven Ḍākinīs across seven lokas. The system is already there, in one of the oldest Vidyāpīṭha texts. It is not being invented. It is being recorded.
The System Is Already Complete
And it does not stop there. When we look at the earliest surviving Kaula texts, the texts that scholars themselves date to the base of the Kaula tradition, the complete system is already present. Not gradually assembling. Already there.
The Yogapīṭha-kramodaya, one of the earliest Kaula texts, probably predating the branching of the tradition into the four āmnāyas, already contains seven Yoginīs at the centre of its practice. Kusumamālinī, who is Hākinī. Surrounded by Yākinī, Śākinī, Kākinī, Lākinī, Rākinī, and Ḍākinī. Seven. These are the Dhātunāthās, the "tissue-Yoginīs," each presiding over one of the seven dhātus of the body: semen, bone, sinew, muscle, marrow, blood, and skin. Stationed in the cakras along the Kulapatha, the Suṣumṇā.
Seven Ḍākinī-class beings. Mapped to the seven dhātus. Stationed in the cakras along the central channel. This is the Siddha Dharma Kula Maṇḍala. Seven Ḍākinīs across seven lokas, governing the complete spectrum of embodied experience. The structural correspondence is exact. And it appears at the very earliest stratum of the Kaula written record.
Matsyendranath's Kaula-jñāna-nirṇaya, another foundational Kaula text, contains the same system: Ḍākinī-siddhi, Rākinī-siddhi, Lākinī-siddhi, Kusumamālinī-siddhi, Yoginī-siddhi. The same Dhātunāthās. And in the context of the Gurupaṅkti, the guru lineage of the Yoginī-Kaula system, the same names reappear: Lākinī, Ḍākinī, Śākinī, Kākinī, Yākinī. These are not abstract entities to be feared. They are part of the guru lineage itself. They are part of the transmission chain.
Then the Kaṅkālamālinī-tantra of the Dakṣiṇāmnāya names sixteen Ḍākinīs corresponding to the sixteen svaras of the Sanskrit varṇamālā. Sixteen forms. Siddha Dharma teaches that Ḍākinī has sixteen forms. Exact match. And the Tantrarāja-tantra places Ḍākinī specifically in the Viśuddha-cakra, the throat centre, in a sixteen-petalled lotus with the svaras. The svara-Ḍākinī connection runs through the entire Dakṣiṇāmnāya stream.
And this is not confined to one branch. The Śakti-sūtra of the Paścimāmnāya provides the pādukā-tarpaṇa of the seven Dhātunāthās. The Kālānala-tantra of the Paścimāmnāya repeatedly identifies its goddess with the Ḍākinī. The Gandharva-tantra of the Dakṣiṇāmnāya gives the full Ḍākinī mantra with her svara connections. The Jñānārṇava-tantra gives six Dhātunāthās starting with Ḍākinī in the maṇḍala of Kāmeśvarī, described as having forms resembling Kāmeśvarī herself. Across the Dakṣiṇāmnāya, the Paścimāmnāya, the Uttarāmnāya, everywhere you look in the Kaula tradition, the same complete Ḍākinī system appears.
And here is something the scholars themselves note but do not seem to fully absorb. The Hāhārāva-tantra of the Uttarāmnāya, a text teaching the worship of Guhyakālī, contains both counter-Ḍākinī prayogas and incorporates the six Ḍākinīs as positive maṇḍala deities. Both. In the same text. Not a contradiction. A complete picture. The same tradition that knows how to counter Nikṛṣṭha Ḍākinīs also knows how to worship Deva Ḍākinīs. This is exactly the full-spectrum understanding Siddha Dharma describes: not malevolent or benevolent, but a complete yoni with a complete range. One text holding both truths simultaneously.
And after presenting all of this evidence (the seven-fold system at the earliest stratum, the sixteen forms, the cakra mapping, the cross-āmnāya presence, the dual nature in a single text) the scholar's conclusion is: "we posit that they were part of the original Ḍākinī pantheon that might have been inherited from the Ḍākinī-tantras."
Posit. Might have been.
They see a complete, coherent system appearing fully formed at the very base of the Kaula tradition, present across every branch, and they have to guess where it came from. They have to invent phantom source-texts to explain why this system exists. The tradition does not need to guess. The tradition knows where it came from. It came from the living lineage that held it before anyone wrote it down.
A Surviving Fragment From Jammu
One more thing. The Rudra-yāmala, a later text, but doctrinally significant, contains a verse that scholars often cite to show Ḍākinī negativity:
"ḍākinī taṃ bhakṣayati dīkṣā-mantrārṇa-hīnakam"
"The Ḍākinī devours him who is devoid of initiation in the mantra-bījas."
But read this again. Does it say the Ḍākinī is evil? No. It says the Ḍākinī devours the uninitiated. The one without dīkṣā. The one without the proper mantra container. This is not a statement about the Ḍākinī's nature. It is a statement about the practitioner's readiness. It is exactly the raw-transmission argument from the previous article: without the initiatory container, without the guru's preparation, the Ḍākinī's power overwhelms rather than illuminates. The tradition itself distinguishes between the prepared and the unprepared. The Ḍākinī is the same in both cases. The result differs because the receiver differs.
And we have a surviving text that demonstrates this principle, not once, but fifteen times over.
Among the manuscripts of the Dharmartha Trust at Raghunath Temple in Jammu (the Himalayan foothills, the very region of Kaulantak Peeth) there is a text called the Tithi-Ḍākinī-kalpa. The scholar who analysed it believes it "probably contains rare surviving material from the old Ḍākinī-tantras," the very texts everyone else calls "lost." And where does this rare fragment surface? Not in Varanasi. Not in Tamil Nadu. Not in a European archive. In Jammu. In the tradition's own backyard.
The text itself has fifteen short sections, one for each tithi. Each section follows the same structure: a named Ḍākinī seizes the victim with specific symptoms (fever, trembling, head pain, delirium). Then the text prescribes a bali offering (meat, alcohol, fish) at a specific time and place: crossroads, cremation grounds, riverbanks, at sandhyā-kāla. And each section ends with a mantra to the same Ḍākinī, to relieve the affliction. Yena saṃpadyate sukhaṃ: by which comfort is attained.
The same Ḍākinī who causes the affliction provides the cure. Not a different deity intervening. Not a higher power overriding her. The same Ḍākinī, approached with the correct mantra, the correct bali, the correct protocol. Without the mantra: disease. With the mantra: sukhaṃ. Fifteen sections. Fifteen demonstrations of the same principle. The prepared and the unprepared, receiving different results from the same source of power.
The scholar calls this "a possible bridge between their two expressions in tradition," between the malevolent and the protective Ḍākinī. A bridge. As if these are two separate things slowly being connected. From the Siddha Dharma perspective, this is not a bridge. This is the original understanding. The Ḍākinī has always been both. Not because she changed, not because the tradition evolved, but because that is her nature. She is the fire that burns the unprepared hand and cooks the prepared meal. Same fire. Different relationship.
And notice the bali offerings in this text that the scholar considers the most archaic surviving Ḍākinī-tantra material: madya, māṃsa, matsya. Alcohol, meat, fish. Offered at cremation grounds and crossroads. This is vāmācāra. This is ucchiṣṭa practice. In what may be the oldest surviving fragment of the Ḍākinī-tantra tradition, the ritual framework is already exactly what Siddha Dharma describes as natural to Ḍākinī worship. Not an innovation. Not a later development. Already there.
And one more detail. The fifteen Ḍākinīs in this text are named after women from every level of society. Brāhmaṇī. Kṣatriyā. Carmakārī, the leatherworker. Kumbhakārī, the potter. Cāṇḍālī. Saucakī, the washerwoman. From Brahmin to Cāṇḍāla. The Ḍākinī tradition does not recognize the hierarchy that every other ritual system is built upon. This is the Kaula dissolution of purity boundaries, not as philosophy, but as embedded structure in the oldest surviving Ḍākinī ritual text.
Primarily a Kaula Phenomenon
And finally, the scholars themselves observe that the full, positive incorporation of the Ḍākinī class is "primarily a Kaula phenomenon." Think about what this means. The complete understanding of Ḍākinīs appears primarily in Kaula contexts. Kaulantak Peeth is the Kaula source. Saying "the complete Ḍākinī tradition appears in Kaula contexts" is like saying "the most complete understanding of Viṣṇu appears in Vaiṣṇava contexts." Of course it does. That is where it lives. That is where it has always lived.
This is not evidence of Kaula traditions gradually "adopting" Ḍākinīs from elsewhere. It is evidence that the Kaula tradition always held the complete Ḍākinī knowledge, and other traditions, to varying degrees, caught glimpses of it. Fragments surfaced in the Brahmayāmala. Fragments surfaced at Gaṅgdhār. Fragments surfaced in the Sūkṣmāgama temple liturgy. The Kaula tradition held the whole.
And consider Matsyendranath's Kaula-jñāna-nirṇaya, one of the earliest surviving Kaula texts, explicitly identified with the original Kaula teaching. Matsyendranath is one of the most important Siddhas within the Siddha Dharma tradition, with a direct connection to Kaulantak Peeth. His text contains exactly the kind of positive, complete Ḍākinī engagement that the mainstream Śaiva texts lacked, not because it was newly invented, but because it was finally written down.
The knowledge did not evolve into existence. It radiated outward from its source. And the further downstream you go, the more fragmentary it appears.
None of this should intimidate a sincere practitioner. The fact that Ḍākinīs were misunderstood by traditions that lacked the full framework, the fact that they were feared even by advanced tāntrikas, should not stop anyone who feels genuinely called to this path. It should simply tell you: approach with respect, approach with guidance, approach through the living guru who holds the complete Ḍākinī Kula Maṇḍala. The Ḍākinī is not dangerous because she is evil. She is powerful because she is real, more real than the categories people have tried to fit her into. And the Siddha tradition exists precisely to provide the container that makes her power accessible, intelligible, and transformative.
So to summarize across these three articles: the texts are not wrong. They are partial. The folk encounters are real but limited to Nikṛṣṭha Ḍākinīs. The Vidyāpīṭha prohibitions are valid within their system but reflect a communication protocol mismatch, not the Ḍākinī's nature. The Mātṛkā association is contextual presence, not evolutionary origin. The epigraphic and textual record (Gaṅgdhār, Ur-Skandapurāṇa, Sūkṣmāgama) shows sightings of something that was always there, not birth of something new. The Brahmayāmala itself records the Ḍākinī class as transmitters of the highest Kula knowledge: exactly the function Siddha Dharma has always described. And the scholarly model of gradual evolution rests on reconstructions of texts they have never read, when the living tradition that held those texts still exists, still transmits, and still holds the complete Ḍākinī Kula Maṇḍala.
The texts got the symptoms right. Siddha Dharma holds the diagnosis.
All teachings in this series are drawn from the authoritative knowledge corpus of Kaulantak Peeth, and the Siddha Dharma tradition as revealed by Kaulantak Peethadhishwara Mahasiddha Ishaputra.
If you wish to reference this article:
APA: Nath, A. (2026). Is Dakini Hindu or Buddhist? Siddha Tantra Arts. https://www.siddhatantraarts.com/articles/dakini-hindu-or-buddhist
Chicago: Mahayogi Ashutosh Nath. "Is Dakini Hindu or Buddhist?" Siddha Tantra Arts, 2026. https://www.siddhatantraarts.com/articles/dakini-hindu-or-buddhist
