The term Ḍākinī is already seen in the 4-5th centuries B.C.E. in the works of the Sanskrit grammarian, Pāṇini. There, the term refers to a type of flesh-eating female deity that appears in the retinue of the goddess Kālī.
Over the following centuries, Ḍākinīs continued to be a part of the Indian pantheon, though only as relatively minor figures.
In the 8th century C.E., however, as Buddhist Tantra was taking shape, Ḍākinīs began to acquire a greater importance.
Initially it seems that the term was used to refer to human women who gathered around sacred sites and rituals:
Portrayed as typically low caste—prostitutes, washerwomen, and the like—these women would serve as consorts for the male tantric practitioners.
These socially liminal women were held to have a mysterious and dangerous power, and before long Ḍākinīs were cast as Enlightened Beings in their own right, Vajrayoginī, Vajravārāhī, and Ekajāti being some better known examples.
The Yoga-rātna-mālā (Garland of Jewel-like Yogas), an Indian commentary on the Hevajra Tantra, derives the term Ḍākinī from the Sanskrit root, dai, meaning “to fly.”
The accuracy of this derivation has been debated by Western scholars, but it was clearly accepted by Tibetans when they chose to translate the term as mkha’ ’gro (sky dancer).
The Ḍākinī thus described is often understood as able to move freely through the space of reality, the Dharma-dhātu.
In Tibet, Ḍākinī can refer either:
a) to a living woman Buddhist teacher or
b) to a spirit of ambivalent nature.
Regarding the latter type, the idea has persisted that Ḍākinīs are attracted by Buddhist practitioners, drawn in swarms to powerful meditators like mosquitoes to blood.
Tibetans further distinguish 2 kinds of Ḍākinīs:
1) gnostic (ye shes), also called “otherworldly” and
2) flesh-eating (sha za), also called “worldly”
—the first being helpful for one’s progress along the Buddhist Path, and the second harmful.
Telling one type from the other is famously difficult, so that, just as was the case in 8th century India, Ḍākinīs in Tibet continued to hold a dangerous power:
The Buddhist practitioner’s difficulty in judging them is made worse by a tendency for each type to blur into the other, so that a gnostic Ḍākinī can suddenly become dangerous, and a flesh-eating Ḍākinī can provide assistance.
Ultimately, the meditator is advised not to fall victim to either dualistic conceptualization of these gossamer beings.
The Ḍākinīs enigmatic nature has helped it to serve a mercurial role in Tibetan Buddhism, slipping easily between the human realm and those of the Buddhas.
For followers of the Nyingma school, this role has placed Ḍākinīs at the centre of the “treasure” (terma) revelation process:
A Ḍākinī often guides the treasure revealer to the discovery site, and then the treasure teachings themselves are typically received in the condensed language of the Ḍākinīs (mkha’ ’gro skad):
Like the Ḍākinī herself, the symbolic syllables (mkha’ ’gro brda yig) of her language are polyvalent, their significance difficult to determine.
The process for decoding these encrypted teachings is a mysterious one, involving the revealer opening his body’s chakras to allow the treasury of the Buddhist teachings to flow forth unimpeded.
Thus the Ḍākinīs language suggests a shimmering field of possibilities rather than a single determinate meaning.