It has been known for some time that the last layer of bauddha tantra composition, yoginI tantra collection, heavily plagiarized material from the shaiva bhairava-srotas and its kaula evolutes. However, while the term “yoginI” or “yogeshvarI” is the most commonly used for the goddesses of the corresponding shaiva system, the bauddha-s use the term DAkinI for their equivalents.
It is notable that in the early extant shaiva tantra-s, the terms like DAkinI and shAkinI are primarily used for malevolent deities, and they are not the focus of yAga-s of those systems. They might have a primarily negative connotation in these texts: e.g., the Damara-tantra-s (uDDAmareshvara, like affiliated with the bhUtasrotas) invoke chaNDeshvara to suppress the DAkinI-s and ghosts. We have comparable counter-DAkinI spells invoking unmatta-bhairava in the eponymous bhairava-tantra. However, DAkinI and shAkinI also appear as the names of kula-s yoginI-s in the vidyApITha class of tantras in the bhairava srotas, but even there, they retain a malevolent character and are not the most prominent group of yoginI-s receiving high worship.
But in the early kaula tantra-s like the yoga-pITha-kramodaya, DAkinI and shAkinI are notable yoginI-s of the dhAtunAthA class, but not classes of their own. Yet, despite maintaining their malevolent nature, the DAkinI-s appear early in the context of mAtR^ikA worship in the famous ga~NgdhAr inscript of 424 CE of mayUrAkShaka the minister of vishvavarman. It has a the statement in a lacunose line reading: “… idaM dAkinI-samprakIrNaM veshma+aty ugraM nR^ipati-sachivo .akArayat puNyahetoH ||’’ It talks of the establishment of a shrine of the mAtR^i-s which is endowed with the mighty power of the tantra-s and permeated with DAkinI-s who evidently make the shrine awe-inspiring (aty ugram). This hints that there was a lineage of shaiva-shAkta tradition where the DAkinI-s while still fierce or malevolent had a cultic prominence.
Against this background, it is curious that the bauddha scholar dharmakIrti in his attack on Hindu ritual systems uses the phrase: “…DAkinI-bhaginI-tantrAdishu darshanAt”.
He is believed to have lived close to Adi shaMkarAcharya in time 600-660 CE. At this time, the bauddha tantra-s surviving in Sanskrit or Chinese translation do not yet show any signs of adopting DAkinI-s as central deities of their tAntrika system. In fact, they even assign negative roles to them as in the manjushrIya mUlakalpa.
However, we have evidence from the Gilgit manuscripts that the shaiva vAma-srotas tantras – the bhaginI tantra-s were already in place by then. Hence, together with the context of dharmakIrti’s reference to them, we should assume that there was also a non-bauddha (Hindu) class of tantra-s in place by then known as the DAkinI-tantra-s. Based on the ga~NgdhAr inscription we may infer that they likely had a shaiva-shAkta orientation, with an emphasis on the second element of that dyad. Hence, we suspect that in these apparently lost DAkinI tantra-s, the DAkinI-s played overlapping roles with the yoginI-s of the more standard bhairava-srotas and their kaula evolutes.
Hence, it seems that although the bauddha yoginI tantra-s borrowed heavily from the bhairava/kaula material, they inherited the class name “DAkinI’’ of their cultic female deities from the posited DAkinI-tantra-s. We see the first glimmer of this in the sarva-tathAgata-tattva-saMgraha where the mantra of the “vajrified’’ chAmuNDA, i.e., vajrakAlI, calls her vajraDAkinI.
On the H side, we believe that apart from incorporation into some of the kulAmnAya-s, some of the early DAkinI material persisted in some form in the poorly studied tantra-s with a heavy shAkta orientation like hAhArAva, kAlAnala and kaNkAlamAlinI and also in some pratyaNgirA related material.