5
Untouchable women in some tantric texts
Mikael Aktor
Caṇḍālī – Chief Goddess, Lady of Hosts
The Hindu tantric text, the Yonitantra, evokes a yantra representing the yoni (vagina, womb), in which the central divinity is presented as ‘Cāṇḍālī’, a woman belonging to an untouchable caste (YT 3.14–20). This low-caste woman is nothing less than the ‘Chief Goddess, Lady of Hosts’, whose worship makes the worshipper equal to Śiva.1 Other tantric texts, most prominently the Buddhist and Vaiṣṇava tantric Sahajiyā literature, make use of a similar rhetoric in texts where the Cāṇḍālī or the Ḍombī washerwoman becomes a personification of the kuṇḍalinīśakti of the Hindu Tantra systems (Schoterman 1980: 24–25; Bhattacharyya 2005: 216).
These identifications raise a number of questions about the social reality and the ritual significance behind such caste labels. The general antinomian character of the tantric movements as well as the names of tantric masters can be read as a critique of Brahminical social exclusiveness by groups that were either non-Brahmin by birth or had left their Brahmin communities. In so far as the names of these masters indicate their caste background, many came from low castes like leather-workers (‘Camari’), washermen (‘Ḍombī’), hunters (‘Śabarī’), oil-grinders (‘Tailo’), weavers (‘Tānti’) etc. Among those who were born Brahmins, Saraha chose to become an Untouchable and married a low-caste woman (Bhattacharyya 2005: 263).
This impression of an anti-Brahmin attitude is obscured, however, by the idea that tantric practice was more an esoteric layer of personal religion than an identifiable feature of a particular recognisable group. Among what was probably a common Kaula stock of verses found in various tantric texts, Yonitantra 4.20 states:
Privately a Śākta, outwardly a Śaiva, among people a Vaiṣṇava – bearing various outward appearances, the followers of the Kula-system spread over the earth.2
According to this principle, tantric sādhakas might have belonged to various layers of society, and it would not always be easy to identify them.
The yonipūjā, the ritual that is the subject of the Yonitantra, involves sexual intercourse in order to extract the sexual fluids of the partner. These fluids, the yonitattva, consist both of the fluid produced in the vagina through the stimulation by the penis of the candidate during sexual intercourse, and the menstrual blood, which in these contexts – in contrast to the public Brahminical view – is regarded as an extremely powerful ritual remedy. In the Brahminical code menstruation and everything related to it is segregated from the ritual field, while in tantric texts it can occupy a significant aspect of the ritual practice. From the point of view of the Yonitantra, intercourse is ideally with a female partner during her menstruation, a practice abhorred in the texts of Brahminical law. The yonitattva is used in two ways: As a daily tilak (forehead mark) and in the worship of the female partner (Śakti) as Devī, the latter also involving the consumption of the yonitattva mixed with wine (Schoterman 1980: 28–29; White 2003: 109–110).
The social status of the female partner is flexible like that of the sādhaka, although somewhat ambiguous. Low-caste women are preferred, but not exclusively:
An actress, a kapālika woman, a prostitute, a washer woman, a barber’s daughter, a Brahmin woman, a daughter of either a Śūdra, a cowherd or a garland maker, these nine young women [kanyā] are recommended; or a woman from any caste who is experienced and wanton. Except for a mother, he may have intercourse with any woman between twelve and sixty years. … Only a yoni that has already been wounded is suitable for worship, never that of a virgin; by the worship of a virgin there will be destruction of powers every time, O Goddess.
(YT 2.3–5, 3.25–26)3
Other texts are more restrictive both in terms of social status and age. Some exclude any woman belonging to the three upper varṇas and limit the upper age at 25 (Schoterman 1980: 20). However, that they are referred to in the text as ‘kanyāḥ’ is, according to Schoterman, an indication that, although not virgins, they should still be unmarried (ibid.: 18). He also regards it as ‘a well-known fact that especially women living on the fringe of society were chosen as Śakti in tantric rituals’ (ibid.: 24).
Whatever the social facts are behind such caste labels in tantric texts, many other texts make it possible to understand the ritual significance of these labels. What emerges is a multifaceted image of ‘the Untouchable woman’ as a reservoir of contrasting forces that arouse contempt, horror and attraction at the same time.
Untouchable women are powerful, first because female sexuality in general is an expression of śakti; in the Law Books, a much more freely flowing sexuality is attributed to women than to men; according to these texts women are driven by a constant itch of sexuality (MDhŚ 9.14–15; Leslie 1989: 260). Second, there are reasons to believe that this sexuality is manifested even more strongly in low-caste women, because it remains untamed by marriage to men from the four varṇas. In general, marriage is the remedy for unrestrained female sexuality, and, in addition, Caṇḍālas are often associated with a savage life beyond the norms of civilised culture. These unmarried but sexually active low-caste women possess an untamed sexuality and as such they are highly potent agents of ritual transformation in an antinomian tantric context. Their position in society without any role in the kinship system of the varṇas and as an underprivileged group with poor access to protection from outside also makes them easily available. Apart from the general secrecy it involved, the use of low-caste women could therefore have been a way of extricating the tantric ritual from the constricting public norms of family and marriage.
The Caṇḍāla woman
In classical South Asian texts belonging to a wide range of genres, The Caṇḍāla represents a category of untouchable castes more than it denotes one single caste (for an overview, see Aktor 2010). It is, as Dumont has expressed it, ‘the old prototype of the Untouchable’ (Dumont 1980: 52). In the later layers of the vedic literature we already find clear expressions of contempt for Caṇḍālas, and it is interesting in the context of the Yonitantra that one of the earliest formulations (in Chāndogyaupaniṣad) is explicitly with regard to the Caṇḍāla-yoni, which is far from being a source of liberation or uplift. In conjunction with the new doctrine of karman and rebirth we can read:
But people of foul behaviour can expect to enter a foul womb [yoni], like that of a dog, a pig, or an outcaste [caṇḍāla] woman.
(ChU 5.10.7)4
We do not know exactly what was understood by ‘caṇḍāla’ in this early text. But during the centuries of the composition of the Mahābhārata and of the early Law Books two different images had emerged. One was the domesticated caste of Caṇḍāla, that is, low caste groups living at the borders of villages and cities but still performing duties within them, typically related to the impurities of death, excrement or criminality. This image is dominant in the Law Books, whose very purpose is to record the customs of ‘good people’ and on that basis prescribe social norms. The very nature of these Caṇḍālas is a transgression of kinship rules, in that, as is explained in this literature, they are regarded as the outcome of illegitimate sexual relations between Brahmin women and Śūdra men (e.g. GDhS 4.18; MDhŚ 10.12). The other type is the savage Caṇḍāla representing a tribe rather than a caste. These are mostly depicted in the epic and other narrative literature as ferocious hunters living in the jungle. An example is the well-known episode from Mahābhārata about Viśvāmitra, who steals dog’s meat from a Caṇḍāla hamlet during a period of starvation (MBh 12.139.27–30). In contrast to the Law Books, the notion of mixed classes (varṇasaṃkara) is absent in such stories, probably because in this narrative literature these tribal Caṇḍālas represent a contrast to Āryan civilisation and as such are depicted as being unrelated to its ideal kinship system, the four varṇas (Aktor 2008: 87–104).
Later narrative literature preserves this tribal stereotype. In the novel Kādambarī by Bāṇabhaṭṭa and his son dating from about 600 CE, the hero, magically transformed into a parrot, is abducted by ferocious, savage Caṇḍālas. Shocked by the scene of the Caṇḍāla hamlet in the jungle, Vaiśampāyana, the parrot, gives us a long description of this terrible place. The description is clearly an expression of all the standard stereotypes of the Caṇḍāla as representing the antithesis to Brahmin values:
Here and there the village was hidden in a thick forest of bamboo and could be inferred only by smoke, which had the stench of meat. Everywhere there were enclosures of fences made mostly of skulls, heaps of feces alongside roads that were made mostly of bones, and courtyards sloshing with the slop of chopped meat, fat, marrow, and blood. The people’s livelihood consisted mainly of the hunt; their food was mainly meat; their oil was mainly bone marrow; their clothes were mainly of raw silk; their beds were mainly of hides; their retinue consisted mainly of dogs; their mounts were mainly cows, their objects of human pursuit [puruṣārtha] were mainly women and wine; their offerings to gods were mainly blood; and their deeds of duty [dharmakriyā] were mainly sacrifices of beasts. It was the shape of the Hell of Torment. It was like the cause of all evils. It was like the place of all cremation grounds. It was like a city of all sins. It was like a temple of all hellish punishments. … The men were more shameless in all their business dealings than were the hearts of all men. The behaviour was no different in children, young people, or the aged. Intercourse was had with one’s own and others’ women. It was a market for only foul deeds.
(Layne 1991: 341–342; Kād pp. 505–506)5
But, while this place may be terrifying in every decent respect, the wild and unspoiled Caṇḍāla women are also disturbingly attractive. At the start of the novel a young woman from this Caṇḍāla tribe brings the wonderful parrot to Śūdraka, the noble king of the country. The text describes vividly the sight of the woman and the king’s reaction:
She, like a swoon, rendered men unconscious. … As if insubstantial, she avoided touch. She, like a painting, was enjoyment only for the eye. … She was in her early adolescence, and her figure was supremely lovely. Śūdraka (gazed at her with unblinking eyes), was struck with amazement, and turned these thoughts over in his mind: ‘Aho! What a wasted effort by the Creator, pouring such beauty into such an unsuitable vessel. For if she has been so created that her beauty mocks all other beauty, then why give her a caste with which the pleasures of sexual intimacy and touch are forbidden? It seems to me that the Lord of Creatures may have produced her without even handling her, out of fear of the stain of a Mātaṅga’s [i.e. a Caṇḍāla’s] touch, else where comes this beauty of such perfection; for (such beauty as this does not come about for limbs spoiled by the touch of hands). Fie! fie! upon the Creator who brings about (such an improper combination)! For this maiden whose shape is so very enchanting vexes one, since love-making with her will always be condemned because of her low caste.’
(Layne 1991: 14, with additions/alternatives in brackets from Smith 2009: 47; Kād pp. 24–25)6
Śūdraka’s thoughts reveal ambiguous attitudes to untouchability. He seems to turn its rationale on its head. That the Caṇḍāla girl is untouchable protects her from the touch of others more than it protects others from her touch. The fear of her touch aroused in others keeps her in a state of perfect beauty, unsullied by the touch of lustful hands. Untouchability is a social fact that the king has to obey, however much he may want to touch her. But below or beside this level of social norm and public censure, there is another level at which the Untouchable embodies a pure beauty that is protected from the impure touch of the Touchables.
The level of society
The rules preventing king Śūdraka from having sexual relations with the beautiful Caṇḍāla girl were formulated in increasing detail in the medieval dharmaśāstra texts. While untouchables were considered ritually polluting, they also made up an essential unskilled labour force. Their upward mobility was, therefore, impeded by rules restricting many different types of contact, especially sexual contact and economic transactions. The Parāśaramādhavīya, a huge mid-fourteenth-century commentary on the Parāśarasmṛti, contains long chapters on Untouchability rules. Judging from the amount of text that is expended, for a man from any of the four varṇas to have sex with a Caṇḍāla woman is a very serious transgression. A vast number of rules are cited, and since they prescribe different penances to repair the same transgression, each penance has to be explained in relation to a specific circumstance. The penances range from a Prajāpati Penance, which is a fast lasting twelve days, to penances that involve the suicide of the sinner. The criteria for distinguishing different degrees of seriousness are many, typical ones being intention or knowledge of the character of the act, frequency with which the sin was repeated, completion or interruption of the sexual intercourse, and fertilisation of the woman (Aktor 2008: 119–129).
But what is the nature of the Caṇḍāla’s devastating touch. Are Caṇḍālas impure? Yes, they are more impure than any other caste, but this is not simply a matter of being the lowest in a continuum of impurity. The nature of the Caṇḍāla’s impurity is different. We get a very precise definition of this special quality in one of the earliest sources to the Untouchability complex. This is Patañjali’s commentary on Pāṇini’s rule (2.4.10) regarding the use of names of certain groups of Śūdras in the Sanskrit copulative compound (dvandva). Pāṇini prescribes that names of Śūdras ‘who are not excluded’ (aniravasita) can be used in the group of dvandva compounds that have the value of neuter singular collectives (samāhāradvandva). In his commentary Patañjali, who like other grammarians and like many lexicographers classified Caṇḍālas and similar groups as low categories of Śūdras, explains who the ‘excluded’ are and what the criteria are for their exclusion. Following common practice, Patañjali casts his argument as a debate, first presenting the prima facie view that segregation is a matter of exclusion from a geographic territory, the country known as Āryāvārta. But this has to be rejected because some dvandva compounds formed by names of foreign people and known from common usage do not conform to this demarcation (that is, these names can be members of a samāhāradvandva even though the people that they designate live outside Āryāvārta). Hence, it is suggested that it is a matter of being excluded not from Āryāvārta but from Aryan settlements within this country:
In that case, what is meant is rather compounds of ‘Śūdras who are not excluded from an Aryan dwelling-place’. – But what is an Aryan dwelling-place? – It is a village, a cattle farm, a city, or a market-place. – But that means that when Caṇḍālas and Mṛtapas live inside these large habitations, there the compound ‘caṇḍālamṛtapāḥ’ [plural] would also be wrong.7
In other words, in common parlance it is correct to say ‘caṇḍālamṛtapāḥ’, that is ‘Caṇḍālas and Mṛtapas’ (the latter probably being guardians of the cremation ground) joined in one dvandva compound inflected in the plural, but not ‘caṇḍālamṛtapam’ inflected as neuter singular. That being the case, these people must, in fact, be excluded from something, since only names of Śūdras who were not excluded could figure as members in this latter type of dvandva compound. So the job is to define the criterion of exclusion. The next suggestion is exclusion from sacrificial rituals. Only Twice-born men are allowed to perform vedic sacrifices. But this does not work either, since it is also correct to speak of ‘Carpenters and Blacksmiths’ (takṣāyaskāram) and ‘Washermen and Weavers’ (rajakatantuvāyam) in the neuter singular despite the fact that these artisans, like any other Śūdra, are definitely excluded from performing the sacrifice. Finally we get the solution:
In that case, what is meant is compounds of ‘Śūdras who are not excluded from the exchange of food vessels’. When food vessels are regarded as being purified after they have been used and cleaned correctly, then the people that have used them are the ‘not excluded’. When food vessels are not regarded as purified after they have been used, even though they have been cleaned correctly, then the people that have used them are the ‘excluded’.8
That is, if food can be eaten from a food vessel that has been used by another and cleaned properly afterwards, then this other person belongs to those who are not excluded. This is why it is correct to construct compounds out of names of Śūdra artisans inflected in the neuter singular. Although they cannot perform sacrifices, they are not excluded from the exchange of food vessels that have been properly cleaned. This also explains why it is wrong to construct samādhāradvandva compounds with names such as ‘Caṇḍāla’ and ‘Mṛtapa’. These people carry a kind of pollution that cannot be removed by ordinary cleaning. Food vessels that have been used by them cannot under any circumstances be used by others, and the only way their names can be combined is in dvandva compounds inflected in the plural.
In the Buddhist Jātaka stories we find other early texts that confirm the same idea. The Satadhammajātaka (Jāt 2.82–85) is narrated as a warning against monks who make their living unlawfully. Such illicit ways are compared to eating the food of Caṇḍālas. The story tells us about a Brahmin, who wanders along the road but who has not brought any food. He meets a kind and polite Caṇḍāla who offers him food, but the arrogant Brahmin rejects this offer. However, when he starts getting hungry, he secretly takes some of the Caṇḍāla’s food thinking that if he only removes the surface layer of food, polluted by the Caṇḍāla when he had his meal, it would be acceptable. But no. Even the inner part of the food, which has not been in direct contact with the Caṇḍāla while he partook of his meal, is severely polluting. The Brahmin immediately gets sick and dies alone in the jungle.
Another story, the Setaketujātaka (Jāt 3.232–237), describes a young Brahmin who commands a nearby Caṇḍāla to move away, because he feared that, after touching the Caṇḍāla’s body, the wind might touch his own. The idea is made explicit in Cittasambhūtajātaka (Jāt 4.390–401) where young women who happen to see two Caṇḍālas in the street exclaim: ‘Seeing this is an evil omen!’ (4.390).
All these examples express the idea that Caṇḍālas carry with them an inborn quality that affects their surroundings. This quality is ‘magic’ in the sense that only ritual means can remove its influence. This also implies that a transformation of this quality can be potentially powerful, as are other inauspicious elements used in tantric ritual.
Menstruation
Where the Yonitantra placed the Cāṇḍālī at the centre of the yoni, another text, the Bṛhadyonitantra, identifies the same as puṣparūpiṇī, that is ‘flowering’, a general euphemism for a menstruating woman (Schoterman 1980: 24). And indeed the term ‘cāṇḍālī’ is frequently used in law texts and elsewhere to designate a woman on her first day of menstruation, that is, the most polluting day. The relation between Caṇḍāla women and menstruation in general is therefore well established. But she is not alone:
She is declared to be a Caṇḍāla woman on the first day [of menstruation], a Brahmin-killer on the second day and a Washerwoman on the third day, while she is purified on the fourth day.9
Mādhavācārya, the commentator of this verse from the Parāśarasmṛti, admits that this is metaphorical language but maintains that it should be understood in practical terms:
The consequences of having sexual intercourse with a menstruating woman on these days decrease to the same extent as those of have sexual intercourse with a Caṇḍāla woman and the other women mentioned. It is with regard to this that these names are used.10
The relation between Caṇḍālas and menstruation can be traced back to the earliest texts that point out those considered to be untouchable. Apart from Caṇḍālas these include menstruating women and women who have recently given birth as well as grievious sinners and people who have touched a corpse (GDhS 14.30). Indeed, death, grievious sin, female sexuality and menstruation are all connected in a mythical aetiology:
When Indra kills Viśvarūpa, he is condemned by the three worlds as a brahmin-killer. To escape the effects of this most heinous crime, he persuades the earth, trees and women to assume one-third of his guilt each. In return, he grants each a boon: the earth, when dug, will heal within one year; trees, when cut, will grow again; and women, unlike other animals, may enjoy sexual intercourse at any time (Taittirīyasaṃhitā II.5.1ff), even in advanced pregnancy …. Indra’s guilt took the form of fissures in the earth, sap in trees – hence one should avoid ‘red secretions’ and resin from cut trees (Manu V.6; cf. Tait.Saṃ. II.5.4) – and menstrual blood in women.
(Leslie 1989: 251, see also Leslie 1994)
The level of mystical experience: the Ḍombī of the Caryāgīti
This perception of the Untouchable as a polluting force both in the home (as menstruating women) and in society (as Caṇḍālas, Ḍombas and other low castes) was rooted in the social norms of family and class, according to which the purity of lineage and caste was the precondition of social respect and success. At the same time these were the very same norms that centuries of world-renouncing ideologies had portrayed as a lethal trap. The attraction of women, sexual pleasure, home and family form only a thin covering over a pit of suffering, old age and death. But unlike other such ideologies the tantric movements insisted on both yoga and bhoga. For this reason, rather than simply rejecting the enjoyments of the senses (yoga), they inverted social norms in such a way that experiences considered impure according to the public social norms were attributed ritual power, allowing their enjoyment (bhoga) to become a means of attaining yogic states of consciousness.11
This inversion of norms is also articulated in relation to untouchability. One of the early expressions of this is seen in the Buddhist Sahajiyā tantric texts known as the Caryāgīti (Kværne 2010).12 This is a collection of songs in a seemingly simple lyrical style, in which a complex tantric cosmology and practice is encoded. Composed in the eleventh century or earlier (ibid.: 7) in a language characterised as Old Bengali or Old Maithili (ibid.: 3), these songs are attributed to a group of nineteen tantric siddhas (ibid.: 4–5), whose names also appear in other medieval lists (White 1996: 78–86). The text of the songs is embedded in a thirteenth-century exegetical commentary by Munidatta, a scholar well-trained in the Buddhist Tantras as well as in a range of older Mahāyāna works (Kværne 2010: 2, 19).
In relation to the present subject the first two verses of the tenth song are especially interesting. Per Kværne (2010) has edited the song together with Munidatta’s commentary and a Tibetan translation. His English translation of the songs reproduces the original verses with the essence of Munidatta’s exegetical gloss in brackets. Omitting these brackets, the first two verses of song 10 read:
Outside the town, o Ḍombī, is your hut; the shaven-headed brahmin (boy) goes constantly touching (you).
Ho Ḍombī! I shall associate with you, (I,) Kāṇha, a kāpāli-yogin, shameless and naked.
(Ibid.: 113)13
According to Munidatta, these verses apply a language of implicit meanings (sandhyābhāṣā, Kværne 2010: 116) in which the social level of the narrative refers to a different level of mystical experience. The two levels are opposite, and everything is turned around: In society the Ḍombī occupies the lowest status as a polluting Untouchable, but here she represents the purified avadhūti, the central vein or psychic channel, through which the bodhicitta moves upwards towards the Cakra of Great Bliss in the head to be transformed in the supreme bliss of non-duality. The Brahmin, whose vedic knowledge and observance of purity rules secures him the highest rank in the social-religious hierarchy, is here ridiculed as a youth constantly touching the Untouchable without ever attaining the embrace of the illuminating Wisdom embodied by the Ḍombī. Only the socially despised Skull-Bearer, Kāṇha, enjoys her fully. The narrative transcends the level of social hierarchy to indicate a process of transcending the sphere of the sense objects. Regardless of the explicit antinomian character of tantric texts we may here recognise an old theme in Indian literature, the inversion of social hierarchy as an expression of imparting esoteric knowledge. The fickle Brahmin boy14 of these verses is one other Śvetaketu/Setaketu as known from the Upaniṣads and the Pali Canon (ChU 5.3–10; Jātaka 3.232–33) whose knowledge is proven inadequate when confronted with true wisdom, always represented by a person of lower status.
Both of these levels are contained in the notion of touch. At the social level of the Law Books the Ḍombī is ‘untouchable’, expressed by the prescriptive participle aspṛśya, but here, at the level of wisdom, she is instead ‘not associated with touch’, expressed by the descriptive adjective aspṛśa/asparśa,15 that is, she embodies a state beyond any contact with sense objects and is therefore ‘outside the town’.16 Therefore, ‘touch’ in the Caryāgīti songs is overlaid with several layers of psychological and esoteric meanings and Munidatta’s commentary even more so. Still, the logic is similar to that of King Śūdraka in Kādambarī: What at the social level is considered a protection for the Touchables against the polluting touch of the Untouchables becomes at the level of wisdom an expression of the purity of something that in the narrative assumes the form of an Untouchable. Thus, as in Bāṇabhaṭṭa’s drama, Untouchability is turned on its head. It preserves the purity of the ‘Untouchable’ against the touch of the Touchables.
A tentative conclusion
Untouchability has here been illustrated by texts in which the meaning of touch and physical contact is understood in different degrees. In Kādambarī touch is physical and is connected with intimacy and love-making; the untouchability of the young Caṇḍāla woman both explains and preserves her beauty. In the Yonitantra touch is understood ritually; sexual contact is a means of producing the ritually potent yonitattva, but at the same time the Cāṇḍālī is reinterpreted as a goddess and is thereby clearly referring to a different level of meaning than that of social hierarchy. In the Caryāgīti song, and especially in Munidatta’s interpretation of it, the psychological and esoteric meanings of touch and the Ḍombī seem to dominate. ‘Purity’ is ambivalent; the untouchable Ḍombī is an adequate expression of a notion of purity beyond any purity/impurity dualism.
All three texts assume a two-levelled world. On the surface there is the world of the respectable life according to the expectations of family and class. Here kinship is the leading principle, and sexuality is seen in the context of preserving and protecting the family line. Accordingly, there are good wombs and bad wombs, and Untouchability preserves this distinction. The foul yoni of the Cāṇḍālī in the Chāndogyaupaniṣad is the womb of a mother. But there is another secret level where sexuality is free from associations with mothers, children and family obligations. This is the level of King Śūdraka’s dreams of lovemaking with the beautiful, unspoiled Caṇḍāla woman, and this is the level of the Siddhas’ experiments with sexual energy and transcendent states of consciousness. Rejecting the values of family and social hierarchy while insisting on enjoyment means that relations are turned upside down. Seen from this level, observance of Untouchability rules represents the unripe, unenlightened state of the public social world. At the same time it preserves the unspoiled śakti, which is embodied by young, untouchable and ever childless women, whose yonibecomes at once the source of sexual joy and, allegedly, the opening to blissful states of consciousness beyond any contact with the world of sense experience.
Notes
1 yonimadhye pradhānā ca cāṇḍālī gaṇanāyikā / tasyāḥ pūjanamātreṇa mama tulyo na saṃcayaḥ // (YT 3.20).
2 antaḥ śāktā bahiḥ śaivāḥ sabhāyāṃ vaiṣṇavā matāḥ / nānārūpadharāḥ kaulā vicaranti mahītale // (Schoterman 1980: 16; YT 4.20).
3 naṭī kāpālikā veśyā rajakī nāpitāṅganā / brāhmaṇī śūdrakanyā ca tathā gopālakanyakā // mālākārasya kanyā ca nava kanyāḥ prakīrtitāḥ // athavā sarvajātīyā vidagdhā lolalocanā // mātṛyoniṃ parityajya sarvayoniṃ ca tāḍayet // dvādaśābdādhikāyoniṃ yāvat ṣaṣṭiṃ samāpayet // … kṣatayoniḥ pūjitavyā akṣatāṃ naiva pūjayet // akṣatāpūjanād devi siddhihāniḥ pade pade // (YT 2.3–5, 3.25cd-26).
4 atha ya iha kapūyacaraṇā abhyāśo ha yat te kapūyāṃ yonim āpadyerañ śvayoniṃ vā sūkarayoniṃ vā caṇḍālayoniṃ vā. (ChU 5.10.7).
5 ahaṃ [Vaiśampāyana] tu … itastato visragandhidhūmodgamānumīyamānasāndravaṃśavanāntaritaveśmasaṃniveśaṃ, sarvataḥ karaṅkaprāyavṛtivāṭam, asthiprāyarathyāvakarakūṭam, utkṛttamāṃsamedovasāsṛkkardamaprāyakuṭīrājiram, ākh eṭakaprāyājīvaṃ, piśitaprāyāśanaṃ, vasāprāyasnehaṃ, kauśeyaprāyaparidhānaṃ, carmaprāyāstaraṇaṃ, sārameyaprāyaparivāraṃ, dhavalīprāyavāhanaṃ, strīmadyaprāyapuruṣārtham, asṛkprāyadevatābalipūjaṃ, paśūpahāraprāyadharmakriyam, ākaram iva sarvanarakāṇāṃ, kāraṇam iva sarvākuśalānāṃ, saṃniveśam iva sarvaśmaśānānāṃ, pattanam iva sarvapāpānām, āyatanam iva sarvayātanānāṃ … lokahṛdayebhyo ’pi nirghṛṇatarasarvasaṃvyavahārasamastapuruṣm, aviśeṣācārabālayuvasthaviram, avyavasthitagamyāgamyāṅganopabhogam, apuṇyakarmaikāpaṇaṃ pakkaṇam apaśyam / (Kād, pp. 504–506).
6 mūrcham iva manohāriṇīm … amūrtam iva sparśavarjitām, ālekhyagatām iva darśanamātraphalām … aciroparūḍhayauvanām, atiśayarūpākṛtim, animeṣalocano dadarśa / samupajātavismayasya cābhūn manasi mahīpateḥ / aho vidhātur asthāne rūpaniṣpādanaprayatnaḥ / tathā hi / yadi nāmeyam ātmarūpopahasitāśeṣarūpasaṃpad utpāditā kim artham apagatasparśasaṃbhogasukhe kṛtaṃ kule janma / manye ca mātaṅgajātisparśadoṣabhayād aspṛśateyam utpāditā prajāpatinā / anyathā katham iyam akliṣṭatā lāvanyasya / na hi karatalasparśakleṣitānām avayavānām īdṛśī bhavati kāntiḥ / sarvathā dhig vidhātāram asadṛśasaṃyogakāriṇam atimanoharākṛtir api krūrajātitayā yeneyam asuraśrīr iva satataninditasuratā ramaṇīyāpy udvejayati / (Kād, pp. 24–25).
7 evaṃ tarhy āryanivāsād aniravasitānām / kaḥ punar āryanivāsaḥ / grāmo ghoṣo nagaraṃ saṃvāha iti / evam api ya ete mahāntaḥ saṃstyāyās teṣv abhyantarāś caṇḍālā mṛtapāś ca vasanti tatra caṇḍālamṛtapā iti na siddhyati // (MBhāṣ 1.475.4–7).
8 evaṃ tarhi pātrād aniravasitānām / yair bhukte pātraṃ saṃskāreṇa śuddhyati te ‘niravasitāḥ / yair bhukte pātraṃ saṃskāreṇāpi na śuddhyati te niravasitāḥ // (MBāṣ 1.475.8–10).
9 prathame ’hani caṇḍālī dvitīye brahmaghātinī / tṛtīye rajakī proktā caturthe ’hani śudhyati // (PS 2.7.18c-19b).
10 caṇḍālyādigamane yāvān pratyavāyas tāvān udakyāgamana ity abhipretya tair nāmabhir vyavahāraḥ / (PM 2.7.18c-19b, p. 168).
11 Admittedly, I use the yoga-bhoga dichotomy here as a broadly accepted trope of ‘Tantra’ rather than with precise historical reference. This general use goes back to, among others, Louis Dumont in his famous essay on world renunciation (Dumont 1980: 279) where he quotes Kulārṇavatantra 2.23: yogī cen naiva bhogī syād bhogī cen naiva yogavit / bhogayogātmakaṃ kaulaṃ tasmāt sarvādhikam priye // Yonitantra 3.18c-19b is a variant of the same verse: yogī cen naiva bhogī syād bhogī ca na tu yogavān / yogabhogātmakaṃ kaulaṃ yadi yoniprapūjakaḥ // Both variants express the idea that the essence of Kaula doctrine is the union of what on the surface seems a contradiction, i.e. yoga and bhoga. The YT variant emphasises the connection to the yonipūjā.
12 I want to thank Dr Hartmut Buescher for his keen and critical comments to a previous draft of this section on the Caryāgīti.
13 nagara bāhiri re ḍombi tohori kuṛiā / choi choi yāi bāhmaṇa nāṛiyā // ālo ḍombī toe sama karibe ma sāṅga / nighiṇa Kāṇha kāpāli joi lāga // (Kværne 2010: 113).
14 According to Munidatta, the Brahmin with his childish and fickle mind represents the Bodhicitta of the uninstructed yogin: brāhmaṇa iti … capalayogatvāt cittabaṭukaṃ / asampradāyayogināṃ bodhicittaṃ (Kværne 2010: 116).
15 aspṛśayogatvāt ḍombī iti (Munidatta in Kværne 2010: 116); asparśā bhagavatī yasmāt tasmād ḍombī prakathyate (HT 1.5.18; Snellgrove 2010, part 2: 18).
16 The full passage about the Ḍombī and the Brāhmaṇa reads: aspṛśa-yogatvāt ḍombī iti pariśuddha-avadhūtī nairātmā boddhavyā / brāhmaṇa iti brahma-hūṃ-kāra-bīja-jātaṃ capala-yogatvāt citta-baṭukaṃ / asampradāya-yogināṃ bodhicittaṃ saṃvṛti-śukra-rūpaṃ maṇimūlād viramānandaṃ spṛṣṭvā spṛṣṭvā gacchati / (Kværne 2010: 116, his orthography).
References
Aktor, Mikael. 2008. Ritualisation and Segregation: The Untouchability Complex in Indian Dharma Literature with Special Reference to Parāśarasmṛti and Parāśaramādhavīya, Corpus Iuris Sanscriticum et fontes iuris Asiae Meridianae et Centralis, ed. Oscar Botto, vol. 9. Turin: Comito ‘Corpus Iuris Sanscriticum’/CESMEO.
Aktor, Mikael. 2010. “Untouchability”. In Brill’s Encyclopedia of Hinduism, vol. 2, ed. Knut A. Jacobsen, Helene Basu, Angelika Malinar and Vasudha Narayanan, pp. 876–881. Leiden: Brill.
Bhattacharyya, N.N. 2005 [1999, 1982]. History of the Tantric Religion: An Historical, Ritualistic and Philosophical Study. 2nd rev. edn. New Delhi: Manohar.
BORI. 1971–1976. The Mahābhārata: Text as Constituted in Its Critical Edition. 5 vols. Pune: Bhandarkar Oriental Research Institute.
Dumont, Louis. 1980. Homo Hierarchicus: The Caste System and Its Implications. Rev. edn. Chicago, IL: University of Chicago Press.
Kāle, Moreśvara Rāmacandra. 1895–1896. Kādambarī by Bāṇabhaṭṭa and His Son. 2 vols. Mumbai: Shāradākrīdan Press.
Kielhorn, F. 1962–1972 [1880–1885]. The Vyākaraṇa-Mahābhāṣya of Patañjali. 3rd edn, 3 vols. Pune: Bhandarkar Oriental Research Institute.
Kværne, Per. 2010 [1977]. An Anthology of Buddhist Tantric Songs: A Study of the Caryāgīti. 3rd edn. Bangkok: Orchid Press.
Layne, Gwendolyn (transl.). 1991. Kādambarī: A Classic Sanskrit Story of Magical Transformations. New York: Garland Publishing.
Leslie, I. Julia. 1989. The Perfect Wife (Strīdharmapaddhati). Oxford: Oxford University Press.
Leslie, I. Julia. 1994. “Some Traditional Indian Views on Menstruation and Female Sexuality”. In Sexual Knowledge, Sexual Science: The History of Attitudes to Sexuality, ed. Roy Porter and Mikuláš Teich, pp. 63–81. Cambridge: Cambridge University Press.
Olivelle, Patrick. 1998. The Early Upaniṣads: Annotated Text and Translation. New York: Oxford University Press.
Olivelle, Patrick. 2000. Dharmasūtras: The Law Codes of Āpastamba, Gautama, Baudhāyana, and Vasiṣṭha. Delhi: Motilal Banarsidass.
Olivelle, Patrick. 2005. Manu’s Code of Law: A Critical Edition and Translation of the Mānava-Dharmaśāstra. New York: Oxford University Press.
Schoterman, J.A. 1980. The Yonitantra: Critically Edited with an Introduction by J.A. Schoterman. New Delhi: Manohar.
Smith, David. 2009. Princess Kādambarī. vol. 1. The Clay Sanskrit Library. New York: New York University Press.
Snellgrove, David. 2010 [1959]. The Hevajra Tantra: A Critical Study. 2nd edn. Bangkok: Orchid Press.
White, David Gordon. 1996. The Alchemical Body: Siddha Traditions in Medieval India. Chicago, IL: University of Chicago Press.
White, David Gordon. 2003. Kiss of the Yoginī: ‘Tantric Sex’ in Its South Asian Contexts. Chicago, IL: University of Chicago Press.