The Roles and Representations of Women in Pre-Modern Yoga Practice.’

The Roles and Representations of Women in Pre-Modern Yoga Practice.’

This dissertation is submitted in partial fulfillment of the requirements for the degree of MA in the Traditions of Yoga and Meditation of the School of Oriental and African Studies (University of London) on 11th September 2015.

Abstract

This paper examines how women’s roles in pre-modern yoga practices have been represented in tantra and hatha texts. Consideration is given to the shifting notion of yoga and the academic relevance of studying women within this. The texts examined include the Brahmayāmala, a śakta-śaiva scripture that details the participation of women in rituals and represents them in various divine forms; the hathayoga texts from the 12th-15th centuries, the Amanaska, Dattātreyayogaśāstra and

Hathapradīpikā. Themes explored include the limiting nature of sources written from and for the male perspective; sexuality and spiritual-sexual practices, specifically the practice of vajrolīmudra; the agency ascribed to women and to what extent representations align with the lived reality of women’s experiences.

Introduction

One need not be a statistician, ethnographer or ‘yoga’ enthusiast to know that the modern global yoga industry is one dominated by women.1 This has not always been the case. With this in mind, we may ask: historically, how have women been involved in ‘yoga’ and what sources reveal their roles? This question is inherently problematic. The term ‘yoga’ is not a singular, united concept or practice. Rather, it is a polyvalent conceptual category that has become a ‘floating signifier’ (Hatley 2013: 21; Smith 2005: 8987). It’s meaning has shifted over centuries, between texts and

sects to such an extent that origins for the term and certain associated practices can be ascribed to groups as diverse as Indian ascetic yogis, Muslim fakirs or shamans (Singleton 2010). Over the course of this paper I understand ‘yoga’ in its broadest sense to include some tantric traditions that include yoginīs, female deities, in their belief and ritual systems. If we take ‘yoga’ as a shifting notion we must also acknowledge that the relationship women have had with yoga has also changed. This becomes clear when considering the roles and representations of women in premodern tantric and hatha texts. I will question whether women are presented as

having independence, power and agency and what constraints were in place that limited the positions of women. Attempts will be made to understand to what extent textual representations of women align with the social reality of their lives but, equally, the difficulties and limitations to achieving this will also be made apparent. Regardless of the shifts in the meaning of ‘yoga’, certain themes within the texts consistently link male and female practitioners and women to yoga, namely sexual activity.

Before examining pre-modern sources I will outline the relevance and importance of studying the place of women within the yoga traditions in an academic context. Miranda Shaw, in Passionate Enlightenment, argues that western academics historically considered the male religious experience in South Asia as ‘normative and universally representative’ (2000: 5). Beatrix Hauser expands on this point when she

writes: ‘due to the lack of adequate sources, conservative scholars used to legitimize the previous inattention to women in the academic discipline of religious studies as a 1 Mark Singleton’s Yoga Body: The Origins of Modern Posture Practice (2010) traces how this came about.

consequence of women’s lesser degree of involvement in the religious domain’ (2012: 5). On the contrary, women have never been involved to a ‘lesser degree’ in the religious domain, as Hauser discovers first hand through ethnographic fieldwork. Rather, an andocentric view of South Asian religious traditions was constructed within the western academic domain that, by the mid-nineteenth century, allowed this to become the dominant view. This cannot be entirely ascribed to western scholars but came about through a process of ‘intercultural mimesis’ – a cultural interchange

between ‘the native and the Orientalist in the construction of knowledge’ about ‘the Orient’ (King 1999: 148). Typically, this interchange took place between western male scholars and, in the case of Hinduism, for example, male brahmanical pundits (149). In addition, by the mid-nineteenth century the ‘eastern’ traditions (namely Hinduism, Buddhism, yogic and tantric traditions) were understood in the west through textual study and translation (1999: 143). By locating these ‘eastern’

traditions in canonical texts and the male experience, women’s voices, lives and actions have been systematically excluded in the generation of knowledge. Whilst the ‘sameness of women’ across South Asia cannot be presumed this inattention to women and their religious lives has, on the whole, been a ‘collective phenomenon’ (Hauser 2012: 10). It is therefore no wonder that in his work regarding the textual

representations of women in the Brahmayāmala, Shaman Hatley begins by stating that ‘the prospects for meaningful recovery of women’s own voices seem particularly discouraging’ (Hatley 2015: 1). Whilst they might not offer us the voice of women, it is possible to reframe our reading of the texts in order to establish how women are represented and question what this tells us about the social position of women in the period of writing.

Women in Tantra Texts

There are of course many ‘yogas’, as touched upon in the introduction.2 Gerald James Larson goes to some lengths to differentiate the concepts of ‘yoga’ and ‘tantra’ in Sanskrit literary history. He finds that in classical Sanskrit literature the terms have discernibly separate meanings until approximately the 4th/5th century CE after which there is a distinct ‘sectarian turn’ (2009: 491). He notes that from the 6th century ‘in certain Buddhist and Śaiva [and Vaisnava] environments…the terms yoga and tantra come to have dramatically different’ and not such distinct meanings (491). In this new sense the ‘tāntrika’ (a person who practises tantra) is contrasted with the ‘vaidiaka’ (someone who follows the Vedas) (491). Goals of the yogi and the tāntrika,

from this point forwards, commonly include the attainment of liberation (jīvanmukti) and siddhis (extraordinary powers), meditative practices and the use of mantras (491- 492). In addition, current scholarship is in agreement that hatha yoga owes its origins to sexual rights, found in tantric traditions (Mallinson 2016: 2; Muñzo 2011: 125). In this way the categories ‘yoga’ and ‘tantra’ are closely aligned to such an extent that

examination of a tantric text can reasonably be made in this study under the label ‘pre-modern yoga’. It is often only in these sexual scenarios, as we will see below in the hathayoga texts, that women receive any textual representation. In the Brahmayāmala, however, women feature more broadly and frequently than merely in relation to sexual practices. It is for this reason I now turn to examine the representations and roles of women within this text.

Shaman Hatley’s pioneering work, ‘Representations of Women in the Brahmayāmala’ (2015) illuminates details regarding ‘female initiation, ritual sexuality and practices prescribed specifically for women’ (2015: 1). The Brahmayāmala is a tantric Śaiva scripture of twelve thousand verses that provides an ‘unusually detailed (as well as early) window in to women’s participation in tantric ritual’ (Hatley 2015:1). Hatley dates the text as being conceived, in some form, before

the 8th CE (2015: 2). At this time Goddess worship had been ‘progressively subsumed with the mainstream Śaivaism’ (Sanderson 2009: 45). Bhairava, the god who reveals the teachings of this text, was considered by devotees, as a ‘higher, more esoteric manifestation of Śiva’ (45). Whilst the appearance of coital rituals and the Larson notes that despite dramatic differences, it is possible to trace some origins of Hatha Yoga from Pātanjala Yoga in the sutras 2.49-51 and 3.26-34 (2009: 493).

involvement of women in these practices were not unusual in tantric texts in the premodern era, descriptions of purely women-only practices were less common. Additionally, the text ‘blurs boundaries’ between what the reader (at the time of writing and today) may understand to be a ‘woman’ and a ‘goddesses’ through

employment of the divine terms, namely śakti, yoginī and dūtī (2015: 1). Hatley briefly considers to what extent textual representations reflect the social reality of women’s lives, whilst recognising the substantial difficulty of uncovering any solid evidence. I shall analyse each aspect mentioned above in order to build a picture of the anticipated roles of women within this cult. Attempts shall be made to define how normative or subversive these representations were.

The revelation narrative of the Brahmayāmala, characterising the text as a whole, ‘explicitly includes women’ (Hatley 2015: 3). Bhairava prophesises in the first chapter: ‘In home after home, O Mahādevī, whether they be men fit for siddhi, or women fit for siddhi, [the tantra] shall spread to all of their homes’ (2015: 3). Siddhi

typically denotes success in, or the attainment of, supernatural powers arising ‘either directly or indirectly as a result of the practice of yoga or tantric rites’, through ingestion of herbs, asceticism or upon birth (Mallinson 2012: 327).3 The spreading of the word to ‘their homes’ suggests a householder audience. Significantly, two women are mentioned in the revelation narrative (2015: 3). One such woman is the Goddess who has been made ‘incarnate in the world in response to a curse’ (2015: 3). She is born to a village brahmin as a girl named Sattikā.

At birth she possesses auspicious marks, or laksanāvitā, and intellect, or buddhi. She does not receive any tantric initiation but attains perfection, i.e. siddhā, after ‘performing worship of the linga perpetually’ and ‘with great devotion’ (2015: 3). Following this, she ascends to the sky to re-join Bhairava, her consort, and regain her divine name: Aghorī. The narrative stage is set for Bhairava to then, once again, reveal the scriptures to her. This sets in motion its course for reaching the world (2015: 3). Already we see a

blurring of identities between a mortal woman and the divine Goddess. The story suggests that women are capable of achieving success and becoming divine but, notably, this occurs through devotional worship. Given the ‘abundant evidence in the text for female initiation’, Hatley suggests this ‘invites questions concerning the 3 This definition can also be found in the Yogasūtra 4.1.

nature of women’s roles in the religion’ (2015: 4). He does not, however, specify what these questions may be. Could the text serve to demonstrate that even uninitiated women have the potential to become Goddesses through devotion? Does this transformation teach that devotional worship is the right or highest practice for a woman? Or does this episode simply demonstrate that devotion was the most

common religious activity for women at this time? Hatley later suggests that Sattikā’s role in ‘transmitting scriptural revelation’ could serve as a model for female guruship (2015: 11). This must be seen as speculation given that the evidence examined below indicates the anticipated role of a woman was to facilitate the male practitioner’s

success, as a consort, not a guru. Equally, the goddess does not actively speak the revelation but is put in the position as a passive listener. The text implies there were many limits on female agency, relative to the potential agency of her male counterpart. Despite these limitations, representations of women in the Brahmayāmala could be seen as challenging, or subverting, wider cultural norms associated with women at the time it was written.

There is some evidence that the Brahmayāmala aims to subvert the dominant ideology in the revelation narrative. It ‘invokes a pan-South Asian topography’ by affiliating itself with individuals from Odradeśa (east), Sindh and the Swat Valley (northwest), Kashmir and Lampā (far north). These figures were male brahmins affiliated with specific Vedic schools, as well as people from a ‘spectrum of other castes’ (2015:2). The inclusion of other castes probably opened up new ways of

thinking about society, allowing the author(s) to represent, what Hatley terms, an ‘idealized community of practice’ (2015: 2). Hatley notes that, despite this inclusive attitude towards the diversity of participants in the Brahmayāmala cult, the text was probably produced and transmitted in a domain dominated by male brahmins (2015: 3). This indicates the complexity of the concept of ‘tantra’ and ‘brahmin’, often considered to have separate sites of ritual and religious practice. Unorthodox

practices, such as ritualised sexual activity involving multiple women, are not in keeping with orthodox attitudes associated with brahmanical society. This suggests that teachings in the Brahmayāmala can be viewed as dissenting from orthodoxy within their own domain. It is the attitude of dissent in the text that allows for unorthodox representations of women but perhaps this brahmanical domain ultimately

limits the roles prescribed to women. I shall briefly consider the norms associated with Brahmanical society.

The Laws of Manu (200 BCE – 200 CE) stipulate that a woman must not act independently. She should be ‘subjected to male control’ by her father, husband or son (4.147-49; 9.3; Hauser 2012: 4; Doniger 2009: 326). These moral guidelines may not represent the social reality of every woman’s life in pre-modern South Asia. They can, however, be used to ‘exemplify women’s loss of status’ that probably began before the Laws were committed to writing, and continued long after, throughout Indian culture (Hauser 2012: 4). As found in many texts – including the

Brahmayāmala – such attitudes were legitimised through a connection to the divine. The Laws state, for example, a widowed woman will join her husband in heaven if she remains chaste and does not remarry; in contrast, a widowed man should remarry (4.156-66; 4.167-69; Doniger 2009: 326). Marriage was considered the norm for

‘women’s religiosity and cosmological disposition’ according to the Laws and Brahmanical doctrine (Hauser 2012: 5). The latter can be seen as perpetuating a ‘theology of subordination’ (5). The expected married state of a woman, however, is not challenged in the Brahmayāmala (noted below). Scriptures such as the Brahmayāmala do, at the very least, suggest that models of femininity (based on piety and chastity) were not fixed.

Consideration of terms used to address women in the Brahmayāmala will further our understanding of how they were represented and, in turn, their expected or anticipated role(s). The vocabulary serving female practitioners is ‘entirely different’ to that ascribed to male ritualists or initiates (2015: 4). Each term has a double meaning: ‘on one hand denoting the status of female initiate, and on the other being a theological category designating divine beings’ (2015: 4). This suggests women had an element of autonomy from male practitioners, and the potential for power. The context in which the terms are used must be further examined to determine how much independence women were ascribed within the religious sphere. The primary terms are śakti, yoginī and dūtī.

Śakti is the most commonly used term in the text in reference to initiated women and is appended to ‘all female initiation names’ (2015: 5). The human female śakti always completes the male hero, vīra (2015: 5). This vīra ‘embodies the male

pole of the Godhead’ whilst the śakti is the female pole. The female consort is referred to as śakti during sexual rites. Similarly, dūtī is used in the context of sexual ritual. It signifies a female consort and can be used, in this context, interchangeably with śakti (2015: 6). These examples of śakti and dūtī existing as relational terms suggest women are, in certain circumstances, ‘subordinate’ and complementary to

male initiates (2015: 5). Yet, at the same time, śakti is the Goddess. She ‘pervades the cosmos as the myriad female deities who are her rays’ thus the term denotes the supreme deity’s power, specifically gendered as female (2015: 4-5). Hauser argues that the very concept of śakti recognises a woman’s potential: her ability to generate life (2012: 5). The use of this term in the text therefore suggests the acknowledgment of female power.

Through encounters with yoginīs, we find a more autonomous vision of the female practitioner. The term can simply denote a ‘female sādhaka’, that is, an initiated being who is able to participate in ritual and ascetic activities (2015: 4-5). The important point here, however, is that ‘yoginī’ often appears in non-sexual ritual

contexts, unlike the śakti and dūtī (2015: 5). Instead, the yoginī is an independent ritualist acting to attain her own objectives rather than ‘facilitating the aims of men’ (2015: 5). Hence, their place is not one of subordination but independence. In the Brahmayāmala a yoginī is also the goddess in her living form who can fly and shapeshift; who women can become and with whom male practitioners ‘sought visionary,

transactional, power-bestowing encounters’ (2015: 5). She is said to guard esoteric knowledge and live disguised amongst villagers (2015: 6). A yoginī has the ability to ‘bestow magical powers upon brothers of their own initiatory clans’ (2015: 6). Here, the living goddess is designated a power, not only independent from men, but over them. This representation of women as yoginīs collapses the boundaries between the

mortal and the divine (2015: 5-6). This divinisation provides legitimacy to the way in which women are used in rituals, as presented in this text. Unfortunately, despite positive representations, the generative śakti power is regarded as something that must be tapped and the magical yoginī power(s) must be controlled. This becomes increasingly apparent when examining the role and function of women in the ritual context.

In the Brahmayāmala the male, rather than female’s concerns dominate the ritual sphere, regardless of whether she is represented as a yoginī, śakti or dūtī.

Accounts of coital rituals provide, what Hatley terms, ‘a useful, though one-sided window into women’s roles’: descriptions of the dūtī are only found when she actually features in the ritual (2015: 6). Her religious life, external to this, is not elaborated on (6). In contrast, we receive a vast amount of detail regarding the male sādhaka, with whom she consorts. In chapter 45, for example, the dūtī only enters the scene after the two hundred verses dedicated to the preparatory rituals of the sādhaka (7). This provokes questions about the wider role of women who are expected to

participate in such practices: is the reader to presume that the female participant had no significant religious life beyond this context? If she has no part in the preparation, how well versed in ritual or ascetic life must she be? Does a woman’s designation as a dūtī apply only when she acts as a consort during rituals? Of course, not all of these questions can be comprehensively answered however one passage outlining the desirable characteristics of a consort does illuminate certain elements. It reads that a woman must be:

“Obtained by the command of the guru, lovely, possessing the marks of auspiciousness, who has given up comfortable seats (jitāsanā), possessing great spirit, purified by the true essence of the tantras, devoted to the guru, the deity, and her husband, unfatigued by hunger and thirst, ever steeped in nonduality, free of discriminative thoughts and lust, well-versed in samādhi, yoga, and the scriptural wisdom, steadfast in her vrata: after obtaining [a woman like] her, a man of great wisdom should practice what is taught in his ritual manual (kalpa).” (Brahmayāmala 45.186-189)

Clearly, as Hatley points out, the ideal female partner should be ‘an accomplished initiate’, have a capacity for ‘asceticism, meditation, devotion, and learning’ and an understanding of ‘ritual nondualism’ (2015: 7). This passage, however, also states she must be a woman who is devoted to her husband. The chapter dedicated to female initiation and practice also makes this point. Female initiates, yoginīs in this instance, are characterised by the goddess as being ‘weak in both intellect and fortitude

(sattva), yet dedicated to their husbands and full of devotion to the gurus’ (2015: 11). Whilst it is no surprise that a woman is expected to be married, the fact that it is a key feature of a dedicated yoginī drives home the point that there was very little room a woman to be a free agent, even in the idealised community presented in the Brahmayāmala. This ensures that a woman is always defined in relation a powerful

male – either a sādhaka in the ritual context, or her husband outside this context. It also indicates that whilst tantric texts are transgressive in some respects, they maintain certain social norms in others, namely that a woman must be defined in relation to a man, as specified in the Laws of Manu. Hugh Urban argues that this ‘male-female heterosexual binary’ and ‘heteronormativity’, fundamental to such Laws, are also the basis for most tantric texts and rituals (Urban 2010: 135). The Brahmayāmala is no exception.

During the coital ritual the dūtī plays a vital but barely active role implying she has little or no agency. She is instructed to be naked to allow her vulva to be worshipped by the sādhaka (2015: 8). The vulva is interchanged – as a site of worship – with the mandala and fire. This suggests she is little more than a ritual tool. She is told when and how to stand, sit, lie down, engage with the performance, coitus and

mantra incantation (2015: 8). She is not to rise from the bed whist the sādhaka performs the worship (yāga) or fire sacrifice (homa) (2015: 8). Hatley argues her role is ‘passive’ but as she is treated in a similar vein to the other objects required for the ritual, I would go further and suggest she is actively controlled and used in the way one would an object. This assertion is backed up details from other rituals: the dūtī is always ‘framed in terms of the sādhaka’s religious aspirations’ with little

consideration for her needs, desires or spiritual advancement (2015: 8). The rites undertaken for seeing his past lives (in order to identify obstacles to success) require him to carry out sexual performances with his own consort. With her, he may see up to three past lives, however, copulating with seven or eight women is preferable. This will conjure visions of many past lives (2015: 8). It is specifically the women’s

womb, or yoni, the site of past births, which is vital for this ritual as it acts as a kind of portal into the past. Further instructions stipulate that he must copulate with one woman each day during the daily rites (2015: 7-8). Here, it is apparent that the ritual, whilst including women, is not specifically for their benefit. Again, their sexual organs are like ritual tools and they are promised ‘no reward’ for their services (8).

The male practitioner, in contrast, has the potential to attain ‘mastery over all mantras and omniscient vision’ (8). Consequently, the female practitioner is represented as having little agency during the rituals, despite her central role. The chapter on ‘secret nectars’, or guhyāmrta, further supports the notion that women playing the roles of a dūtī, śakti or yoginī are objectified within the text (2015:

9). The primary concern here is with the woman’s ‘sexual and menstrual fluids’, for which the consort is milked like a cow (9). We also find details regarding the initiate’s (presumed male) sexual activity. Monogamy is given a higher status than polyamory but is not enforced: it ‘bestows rapid success’ yet is ‘arduous’ even for the

god Bhairava (2015: 9). The text therefore allows, and legitimises, male practitioners to take multiple female sexual partners by connecting the practice to Bhairava. In passages that discuss how a tālaka (‘pure’ sādhaka) may enter into relations with a consort various suggestions are made that implies each sādhaka has control and ownership over his consort (2015: 10). Once acquired, he should be able to give her over temporarily to distinguished guests ‘of his own lineage’ without ‘lust or

jealousy’ (10). The suggestion is that ownership of the consort may be transferred without the woman’s consent, which again places her in the position of an object and presents her as having little or no agency, power or control in the realm of these

tantric rituals (10). It is hard to say whether or not this was the reality for the majority of women in this religious context as this, and other, tantric texts offer little insight into the female experience. External evidence is also limited. One of the few firsthand accounts available regarding the experiences of a woman in a tantric community is from a Tibetan woman who lived 1892-1940. 4 This shall be discussed further below.

Placing women on a pedestal and ‘honoring their reproductive abilities’, is, as Urban argues, ‘one of the oldest and most recurring tropes in patriarchal discourse’ (136). Throughout this, and other tantric texts, locations of worship are placed on a woman’s womb, vulva or vagina, on her sexual or menstrual fluid or reproductive abilities. Clearly, only those aspects that serve men’s needs are praised (Urban 2010: 136). By making a woman’s power synonymous with biological organs and functions the female position is essentialized and limited. We see this concept inherited and perpetuated by the hathayoga texts.

The section on female initiation and practice – the vidyācakra of patala 14 – offers more positive representations of women. The Goddess asks Bhairava for an ‘easy means’ or sukhopāya, specifically designed to enable women to attain siddhi 4 Details can be found in Sarah Jacoby’s critical edition of Sera Khandro’s life story Love and Liberation (2014).

(2015: 11-12). In response she receives instructions as to how women can attain mantras or manipulate cakras; the latter in order to carry out magical acts typically associated with divine yoginīs, such as entering anthers body (2015: 12). The closing section of this chapter is arguably one of the most significant. It provides a ‘threefold classification of women who obtain the cakra’ and it is entirely different from the system used to categorize male sādhakas (2015: 12). Immediately, this signifies the potential for women to have independent religious experiences outside of male control. The full quote is as follows, translated by Hatley:

“[260cd] This is called the ‘Rite of the Clans’ (kaulika vidhi), [for it] gives delight to the clans of yoginīs. [261] She who obtains the Wheel of the Clans ([[[kula]]] cakra) will assuredly gain [the mantra know as] the kulavidyā, O queen of the gods. [Now] hear of her, being well-composed. [262] She who is [known as] ‘wisdom-in-the-womb’, the one ‘born of a clan’, and the ‘newcomer yoginī’ – [all of them] obtain [the kulavidyā], undoubtedly. [263] [One whose] mother and father both possess the

wisdom, [and whose] mother would bestow the kulavidyā to her while in the womb, is know as ‘wisdom-in-the-womb’. At [the age of] half thirty years she certainly obtains wisdom. [264-65] The mother fully possesses the wisdom, [but] the father has not entitlement: she is [one] ‘born in a clan’. Her mother would whisper the kulavidyā in her ear for six months when she is born. After twenty-four years, the wisdom arises in her. [266] She in whom the wisdom would arise through [consuming] the oblation gruel (caru), through the path of deity worship (yāga), or through consuming the [secret] nectars, is known as the ‘newcomer’ (Hatley 2015: 12-13).

The achievement of ‘wisdom-in-the-womb’ and the whispering in the ear of infant girls highlights the possibility of a matriarchal, mother-to-daughter, lineage. Transmission of ‘genetic’ or ‘lineage’ teachings is outside male control. This is clear when it says ‘the father has no entitlement’ (13). These yoginīs, therefore, hold tantric wisdom and inherent knowledge normally unavailable to men. Such representations of divine women may explain the ‘logic of practices oriented towards attaining

visionary encounters with yoginīs’. Male practitioners want to gain these lineage teachings, or sampradāya, otherwise unavailable to them (2015: 13). One must wonder whether this chapter acts as a justification as to why yoginīs/women must be harnessed and used by male tantrikas to further their own spiritual endeavours. Moreover, pontificating that women can become powerfully divine through no effort 15 supports that judgment that they need to be controlled, within the ritual and wider social spheres. Evidence for this lies in a number of passages. One teaches ksaranaprayoga, a method of sexually enslaving a woman (2015: 13; Brahmayāmala 14.230-35). Another lengthy section on hathamelaka instructs how to draw down

form the skies and master ‘dangerous, divine’ yoginīs (2015: 14; 14.204ff). On closer examination it can be concluded that this chapter, despite being about women’s initiation and ritual, is not for women. Instead it offers warnings and instruction to men about how to master a divine woman. This is in keeping with the rest of the text – male concerns dominate and women feature as ‘vehicles for the sādhaka’s perfection’ (14).

It is clear that the Brahmayāmala does not offer us direct access to the voices of women engaged in tantric practices. When analysing representations we inevitably ‘address the propagation of discourse’ and the discourse is bound up, in this instance, in patriarchy (Biernacki 2007: 8). It includes ‘ways in which women are talked about’ and how this ‘talk functions as a means for constructing identities’, a ‘network of social relations’ encoded as social power (2007: 8). This ‘talk’ is a shadow, contour, or reflected image of women ‘configured through a lens’ (8). All is not lost: even

though it is one distorted lens, it is still a valuable insight (8). The Brahmayāmala challenges the normative view by providing ‘glimpses’ into an esoteric community, in which women did participate in multiple rituals (Hatley 2015: 14). The Vedic context, from which the Brahmayāmala and other Tantras actively broke away, did not encourage the participation of women in rituals at all.5 The sexual nature of the rituals suggests they are shaped by male desire. Sexuality and sexual acts are then used as an ‘operation of power’ (Urban 2010: 24). Deleuze argues that modern psychoanalytic discourse that pervades western consciousness ‘reduces sexuality to sex’ and therefore limits its value and potential (Urban 2010: 22). What we see in the

Brahmayāmala (and in many other tantric texts) is the potential for desire to alter a 5 Here I draw on Alexis Sanderson’s definition: ‘The scriptural revelations of the Śaiva mainstream are called Tantras, and those that act in accordance with their prescriptions are consequently termed Tantrics (tāntrika). The term tantra means simply a system of ritual or essential instruction; but when it is applied in this special context it serves to differentiate itself from the traditions that derive their authority from the Vedas.’ (‘Śaivism and the Tantric Traditions’, 1988: 660).

mortal’s state and enable them to become a god or goddess. For a male practitioner this requires the use of a woman but there are hints that women may achieve this through her own, independent means.

This study of the Brahmayāmala can be taken as an example of how one can reframe textual analysis to consider the role(s) of women. It should be highlighted that tantric literature is ‘by no means monolithic in its views, particularly in its views of women’ (Biernacki 2007: 6). Loriliai Biernacki argues that increased ‘attention to this diversity may be fruitful in advancing our understanding…about women’s

various roles in Tantra’ and would indeed build a larger, more complex impression (2007: 6). She focuses her own study on the literature of north-east India from the 15th-18th centuries but this principle may also be applied to earlier sources. Shaw contests the assumption of male authorship by claiming gender neutral or genderinclusive case endings having been lost in translation, which led to women being overlooked as authors and subjects of Sanskrit texts (2000: 169). This gynocentric vision attempts to redress the balance of andocentric academia however, as Biernacki points out; proving definitive female authorship is difficult (2007: 226). Given the evidence from the Brahmayāmala, Shaw’s view is a moot point as the content clearly champions male power and objectifies women. In addition, if a female practitioner authored a scripture in keeping with internalised notions of male dominance, her gender would have little impact on how she represented women within the text. Patriarchy is the limiting factor when trying to build a picture of the lives of women in tantric communities. What we are left with is representation, not reality. The two may intersect but one should not be mistaken for the other.

Women in Hatha Yoga Texts

In this section I will discuss the roles of women as represented in hathayoga texts. Mallinson writes that ‘modern scholarship on yoga is in widespread agreement that hatha yoga owes its origins to sexual rights, in particular those of certain Kaula Śaiva tantric traditions’ (2016: 2). Adrián Muñzo supports this claim when he writes:

‘sexual practices had [probably] always been an integral element of hatha yoga, on account of the tantric origins of this system’ (2011: 125; Mallinson 2016: 2). Moving the enquiry from early tantric to hatha representations is therefore a logical step. As in the Brahmayāmala, references to women in hathayoga texts are often limited to practices of a sexual nature for which they should either be utilised or shunned. Jason Birch’s paper on the meaning of hatha and hathayoga in pre-modern

texts demonstrates the cross-pollination of terminology between Buddhist, Śiva and Yoga traditions (2011: 527). Hathayoga can literally be defined as ‘the Yoga of force’. This ‘force’ denotes the effort, strain and even pain the practices may cause (527). Alternatively this can be interpreted in more esoteric terms: the process and

practice of hathayoga leads to the ‘union of the sun (ha) and moon (tha) in the body’ (527). Birch notes the absence of the term from the early Śaiva tantras. This is considered surprising given that these tantras are ‘replete with much of the terminology of the Hathayoga corpus’, including mudrā, vedha, bindu and āsana (2011:527; Mallinson 2016: 16). It can be observed that this tantric terminology is reworked in the hatha context, a process arguably down to Śaivism being the

‘dominant religion’ and therefore India’s principle religious idiom during the period 600-1300 CE (2016: 16). Birch argues that defining a ‘corpus of ‘early Hathayoga texts’ is somewhat arbitrary’ as some teach āsana, mudrās and prānāyāma – practices that later come to characterize hathayoga – but do not name it as such (529). After the 11th century the term hathayoga almost always appeared ‘in conjunction with rājayoga’ (527). Often they were rival practices: rājayoga was the superior practice (527). The 13th century Dattātreyayogaśāstra reconciled this by codifying four

aspects of yoga – Mantra, Laya, Hatha and Rājayoga (2011: 527). It is clear from this introduction to the concept of hathayoga that ‘yoga’ a shifting and multivalent category, as are many of the associated terms. Śaiva and Yoga traditions shared terms, practices and ideas but neither are wholly fixed or unified notions. If we consider yoga’s modern global incarnation we see that it continues to undergo such

transformations. Similarly, attitudes towards women are not fixed. There are both continuities and disparities between the various hatha texts, as well as between hatha and tantric manifestations of yoga.

Discerning the role of women in hathayoga practices using pre-modern textual sources is not simple: they offer few direct or detailed references. This is probably because the texts were largely written by and for men, both ascetics and householders. Birch’s locus for defining hathayoga is the 15th century Hathapradīpikā; from this text he looks backwards to delineate the development of practices that came to, and continue to, be associated with it. Retrospectively, these are āsanas, mudrās,

prānāyāma and bhanda (2011: 548). Mallinson, however, begins his analysis with earlier texts and moves forwards in time to demonstrate the relationship between hathayoga and the practice of vajrolīmudrā (2016). This is a more useful stance for the purposes of this study. Using Mallinson’s translations and interpretations I shall consider various hathayoga texts in order to highlight evidence regarding the representations and expected roles of women in pre-modern yoga practice. The early texts teaching hathayoga were composed between the 11th – 15th

centuries CE (Mallinson 2013: 1). They consign great emphasis on physical practices known as mudrās (2013:1). Mallinson demonstrates how the history of vajrolīmudrā in particular ‘epitomises the history of hathayoga as whole’ through ethnographic, textual, and anatomical evidence (Mallinson, forthcoming 2016: 2). The technicalities of vajrolī can be summarised as a ‘method of drawing liquids up the urethra’ to prevent the loss of bindu, or semen (2016: 1-2). This practice is known as

bindudhārana (2016:11). The purpose of vajrolīmudrā is to ‘manipulate the vital energies’ in order to, ultimately, achieve rājayoga, or ‘royal yoga’ and defeat death (2016: 1-2). It is likely that mudrās had been practiced in this way by members of Indian ascetic traditions for hundreds of years before they were committed to writing and, as we see from Mallinson’s contemporary ethnographic work, they remain a

relevant topic for today’s genuine yogi (2013: 1). When considering the mechanics of vajrolīmudrā Mallinson notes that ‘several texts’ state it is ‘possible for women to practise’ however these texts (either original or commentarial literature) do not provide explicit information. Equally, no historical or contemporary ethnographic reports offering women instructions regarding vajrolīmudrā have been brought to light (2016: 3). This is in vast contrast to the details addressed to male practitioners in

pre-modern yoga texts as to how and why they may perform the technique. Reports of contemporary male yogis in India carrying out the practice have also been recorded, the authenticity of which Mallinson can ‘personally verify’ (2016: 3). The majority of ascetics were (and continue to be) men who reject society, family life and therefore women. Despite this, descriptions of vajrolīmudrā do suggest that the presence and participation of a woman was sometimes necessary to carry out the practice if

instructions state that it must be done either during or after sexual intercourse (2016: 11). Vajrolīmudrā’s application and interpretation varies between texts and commentaries. Sometimes it is intimately related to sexual rituals. In these instances women feature more heavily, compared to the texts that do not link the practice with sexual rituals. Vajrolīmudrā’s relationship with sex and women alike is, therefore, not straightforward (2016: 2).

From the earliest sources the participation of women in hathayoga is implied. The 12th-century text, the Amanaska, dismisses those who “take upwards’ (ūrdhvam nayanti) ‘semen that is falling into/from a young woman’s vagina’ (yuvatibhagapatadbindum)’ (2016: 5). Mallinson clarifies this verse by suggesting the yogis use a pipe to ‘draw upwards semen as it falls from young women’s vaginas’

rather than ‘once it has fallen into a vagina’ (2016: 5). The implication is that a young woman, or at least her vagina, is required to participate in the practice. There is no suggestion as to whether this method is preferable to the practice of retaining bindu, or semen, either through celibacy or during sexual intercourse. Additionally, there is no specific regard as to whether the required woman should be well versed in yogic practices. This is in contrast to the Brahmayāmala and the Dattātreyayogaśāstra.

The 13th-century Dattātreyayogaśāstra is the first to teach both practices, hathayoga and vajrolīmudrā, and name them as such.6 Vajrolī is taught as one of the methods of hatha, the others being mantrayoga, layayoga, rājayoga. By attributing the teachings to Dattātreya the text aligns itself with the muni tradition. It is a Vaiṣṇava text as opposed to a Śaiva text. It does however present itself as being accessible to those beyond the Vaiṣṇava traditions when it reads, ‘(41-42) Whether Brahmin, ascetic, Buddhist, Jain, Skull-bearer or materialist, the wise man endowed 6 Translation of the text taken from James Mallinson’s critical edition of Dattātreyayogaśāstra, 2012.

with conviction who is constantly devoted to his practice obtains complete success’. This ‘lack of emphasis on sectarianism’ suggests that yogic techniques were freely exchanged amongst the ‘ascetic milieu’ (Mallinson 2015: 2). This anticipated audience does not explicitly include women (unlike the Brahmayāmala that directly addresses women) but the brief mentions they receive do offer some insight. The

exclusion of women is implied in teachings about ‘Breath control (prānāyama)’, section 3.1.3. The yogi is told that success in his practice could result in him looking like the god of love (84). Consequently women will want to have sex with him (85), which is considered a negative result:

‘If he has sex, his semen is lost. Through loss of semen (bindu), [his] lifespan is diminished and he becomes weak. (86) Therefore he should carefully carry out his practice and not have sex with women…(87-88) so the yogi should make every effort to preserve his semen’. Here the text clearly addresses the male yogi thus excludes lay women and female practitioners alike. As with the Amanaska, however, the presence of a woman for a male yogi to practice vajrolīmudrā is implied, if not deemed necessary. It reads: ‘(125) the yogi who knows Vajroli is worthy of success, even if he behaves selfindulgently’. The implication here is that an accomplished male yogi may indulge in

sexual activity but he will know how to redeem himself through the performance of vajroli. The two substances required for this success are ‘ksīra and āngirasa’ (154). The former literally translates as ‘milk’ but the fact that it is referred to as being available (153) implies the author means ‘semen’ (Mallinson 2016: 8). The latter, also used in an obscure context and described as ‘hard for men to get’, can be interpreted as ‘female generative fluid’ or ‘menstrual blood’ (2016: 8; Doniger 1980: 33-39). This is the woman’s rajas. Men must ‘use some stratagem to procure it from women.’ The teaching continues:

‘(155-156) A man should strive to find a woman devoted to the practice of yoga. Either a man or woman can obtain success if they have no regard for one another’s gender and practise with only their own ends in mind. If the semen moves then [the yogi] should draw it upwards and preserve it’…(158-160) All yogis achieve success through the preservation of semen.’

The implication is that for a male adept to gain the āngirasa he must engage in sexual activity whilst conserving his own bindu/semen (2016: 8). Equally, the female practitioner should conserve her own fluids. This representation of vajrolī suggests that women did practise yoga and were capable of success through the technique. It is unlikely that this text would mention the involvement of women if it did not reflect

reality to some extent. The tension between the initial rejection of women in 3.1.3 and the requirement of a female adept in 3.5.9 perhaps suggests a discernment on the authors part between lay women (who will want to engage in sex with an attractive male yogi for indulgent purposes) and an accomplished female yogi (who will engage in sexual activity for more spiritual purposes).

This seemingly paradoxical attitude towards women and sexual activity, as taught by Dattātreya, can also be accounted for if we consider the purpose of the hatha yogic vajrolīmudrā. It most probably originated in a ‘non-tantric celibate ascetic milieu’ (2016: 20). The purpose of the composition of the hatha texts, however, was to bring the yogic techniques of these ascetic traditions to a ‘noncelibate householder audience’ (20). Vajrolīmudrā then became a ‘method for householders to remain sexually active while not losing the benefits of their yoga practice’ (20). The role of women, in this context, was presumably that of a wife as

opposed to an accomplished female yogi. It is not clear, however, whether these two roles were mutually exclusive.

In contrast, the Hathapradīpikā teachings are more explicit therefore the role of women is relatively clear (3.82-99). This 15th-century text borrows extracts from many earlier texts.7 Instructions on vajrolī are lifted from the Dattātreyayogaśāstra (Mallinson 2016: 9). The Hathapradīpikā does not specifically identify Dattātreya probably because, on the whole, it aligns itself with the tantric yoga as opposed to the

older yogic paradigm based on tapas, or asceticism, as the Dattātreyayogaśāstra does (Mallinson 2013: 1). This may account for the subtle differences in attitudes towards women. In the Hathapradīpikā the male adept is instructed to draw up the semen as it falls into the vagina therefore sexual intercourse and a woman’s participation is necessary. Additionally, a woman becomes a yoginī when she uses vajrolī to preserve 7 It borrows ‘over 40 verses from texts associated with orders of yogis’ and 20 from Dattātreya (‘Siddhi and Mahāsiddhi in Early Hathayoga’, Mallinson: 2012 from Jacobsen (2012: 327-344)).

her rajas. This is to be performed after sexual intercourse and in this case rajas is her sexual rather than menstrual fluid (Mallinson 2016: 9). Not only does this closely align vajrolī with sexual activity but vajrolī with women. Here, they are presented, not as a wife or consort but as being capable of achieving success through the same

means as a male practitioner. Urban argues that, whilst tantric texts and rituals discursively construct women and their sexuality in ‘essentialized and heteronormative’ terms, the fact that they are also identified with the power of the goddess opens space to ‘renegotiate the dynamics of power’ (2010: 128). The Hathapradīpikā breaks away from earlier hatha texts and harks back to these tantric ideas when it suggests a woman can become a yogini through yogic practices (Hatley 2012; 2013).

Following this brief survey of some, but by no means all, hathayoga texts it is apparent that even within this corpus the roles and representations of women are not consistent. Some commonalities, however, can be deduced. The Amanaska, Dattātreyayogaśāstra and Hathapradīpikā identify women through the female sexual organs and fluids, similarly to the earlier tantric Brahmayāmala. Women rarely

receive a mention under other circumstances in both the tantric and hatha corpus. The special requirement of a woman’s ‘vagina’ or rajas (menstrual or sexual fluids) to practice vajrolī suggests that other practices, asana for example, do not require or involve women. Equally, instructions about women are addressed to men; therefore

the assumed reader is male and the female practitioner is ascribed no agency or power. This identification of gender roles with biological function offers an essentialist and binary view of gender categories that allows and perpetuates the

separation of male and female realms in the religious contexts.8 Tantric and hatha practices therefore participate in a ‘larger cultural framework’ that determines how sexuality is constructed (Urban 2010: 135). Tantric texts legitimise these gender roles through ritual divinisation, as mentioned earlier. Hathayoga texts also reinforced gender norms, rather than challenging them, by consistently presenting practices from the male perspective and representing women in relation to men, rather than as independent beings.

8 Judith Butler was a pioneer in the unmasking of ‘binarized gender categories’ that supported ‘gender hierarchies’ and paved the way in studies regarding gender theory and religion (Encyclopedia of Religion 2005: 3314).

Nuances arising between the texts, in relation to the representations and roles of women, can be attributed to the differing lineages and traditions with which each text claims to be associated. The Dattātreyayogaśāstra, for instance, does not give explicit instructions for the yogi or yoginī to have sex. This is in keeping with Dattātreya’s position as a custodian of an ascetic celibate lineage and consequently the preservation of semen is valued over intercourse (Mallinson 2016: 22). In this text

Kapila is said to have been the first to teach the hathayoga mudrās. He, too, is an ancient sage ‘associated with asceticism, celibacy and yoga’ (22). Certain deviances regarding instructions about attaining female generative fluid suggest that it is not so easy to distinguish between puritanical ascetic practices and other traditions that lack such sexual moral restraint (22). It suggests that the text draws from multiple paradigms. The Hathapradīpikā, on the other hand, aligns itself more closely with tantric traditions. A reference is even made to Bhairava, who reveals the scriptures of the Brahmayāmala. Therefore it advocates sexual activity more explicitly than the Dattātreyayogaśāstra.

The Hathapradīpikā is part of a corpus that appropriated and remodelled vajrolīmudrā (Mallinson 2016: 1). This is an example of how yogic practices and their association with women changed, not just between lineages but also over time. The re-conception of vajrolī occurred in light of two tantric concepts: firstly ‘an early notion of sexual fluids being the ultimate offering in ritual’ and secondly (as an interiorisation of the former) ‘the visualisation of the combined products of sex being drawn up the central channel’ (2016: 23). The latter allows tantric traditions to claim that vajorlī enables ‘one to absorb one’s partner’s sexual fluids during intercourse’

(23). Mallinson’s investigations into the biological mechanics of the technique have shown that this is near physically impossible (2016). It does however open the door to the use of sexual acts as a means to liberation, which leads the appropriation of male/female power dynamics within sexual terms. This concept can be found in Śaiva (most famously in Paścimānāya Kaula) and Buddhist sources.9 Here, the male and female principles unite within the body of the yogi as ‘the goddess Kundalinī rises from the base of the spine to unite with Śiva in the head’ (17). To clarify, pre-hatha 9 Serentiy Young refers to vajrolī in Buddhist terms in her book Courtesans and Tantric Consorts (2004).

tantric and Buddhist texts do not teach ‘vajrolī-like physical techniques’ that became a quintessential feature of hathayoga (17).

Women in Non-textual and Buddhist Sources

Whilst the texts invariably align women and sexual practices, artistic representations offer a different picture. In light of the textual representations of women in pre-modern yoga it is important to remember that women do no necessarily

‘always share the male view’ or ‘follow traditions blindly’ (Hauser 2012: 8). In various ways women may ‘perceive, enact and legitimize gender stereotypes’ and can either ‘modify or reject these images’ (10). From this point of view ‘gender is a matter of ‘doing’ rather than passive being – an ongoing achievement’ (11). The texts represent women as having little or no agency within their own practice, however, images found in the Windsor Castle collection suggest that women had independent and active lives as yogic practitioners, outside of the sexualised ways in which they are presented in the texts. The images depict women sitting cross-legged under

trees.10 They are probably part of the large collection of early Mughal paintings that ‘bear witness’ to ascetic archetypes (Mallinson 2015: 2). Various signifiers indicate the women belong to a yogic sect – the earrings, forehead markings, the beads they hold and the way in which they hold them (Mallinson 2015). This not only verifies

the existence of female yogic practitioners, alluded to by the hatha texts, but also indicates that women cultivated a practice just as male yogis did. This suggests that it would have been possible for a man to find a women who was ‘devoted to the

practice of yoga’, as instructed in the Dattātreyayogaśāstra (155-56). Interestingly, the paintings show women in groups without men, or set slightly apart from them, which implies they did practice independently from men. It is likely that the variance between the textual representations and painted images of women can be accounted for by the position of those producing each depiction. The Mughal painters observed

members the yogic traditions as outsiders. They recorded what they encountered without the same agenda as those writing within either the hatha or tantric contexts. The latter produced texts to maintain certain ideals, instructions and norms. They will have anticipated certain assumptions held by their readership; a very different set of assumptions held by the Mughal’s. Resultantly, we find different representations of women.

10 Images passed onto me by James Mallinson (without specific dates).

What neither of these modes of production offer, however, are selfrepresentations of women. The lack of sources from this perspective is verified by Hugh Urban and Ronald Davidson: ‘virtually all of the known Hindu and Buddhist Tantric texts – including those Buddhist tantras ostensibly written by women – describe the sexual ritual from the male perspective and focus exclusively on the male sexual position’ (Urban 2010: 134). Urban continues: ‘out of the hundreds of Tantric manuals and lineages, living and dead, we have uncovered ‘not a single text or

lineage that preserves instructions about yogic or sexual practices that relate to women’s position (134; Davidson 2002: 95). Consequently, there is little available within the textual sources to indicate the agency of female practitioners or the reality of women’s roles. This, unfortunately, is the major limiting factor when considering sources from the pre-modern period.

Sarah Jacoby’s book Love and Liberation: Autobiographical Writings of the Tibetan Buddhist Visionary Sera Khandro analyses the writings of the Tibetan Buddhist woman who lived from 1892-1940 in the Golok region of eastern Tibet. It is

a useful source when considering to what extent male-authored textual representations of women acting as tantric consorts align with their genuine experience. Khandro was a renowned Treasure revealer during her lifetime and continues to have a following in Tibet (Jacoby 2014: 2) Given the desperate tone in which Davidson and Urban lament

the lack of sources regarding women’s perspectives, I anticipate this text will become invaluable in the study of women in tantric and Buddhist traditions, and has the potential to become influential in wider studies regarding women’s writing and selfrepresentation. Examination of the autobiography must be preceded by several

caveats. Firstly, Khandro’s experience is situated within the ‘context of early twentieth-century religious encampments in the nomadic highlands of eastern Tibet’ therefore it is not necessarily universally representative (Jacoby 2014: 191). It does however take the female experience out of the esoteric, symbolic and even idealised frameworks of the ‘prescriptive doctrinal and ritual Tantric texts’, encountered in the

Brahmayāmala (191). Additionally, this is not strictly a pre-modern source but, just as contemporary ethnographic work has proved useful to historical investigations for Urban, Mallinson, Hauser and Biernacki, it provides useful insights into the role of women in traditional tantric and yoga practices. It must also be acknowledge that Khandro lived within the Vajrayana Buddhist tradition, which, whilst it inherited and

adopted many elements of the Śaiva tantric tradition (some of these are even referred to as ‘yoga’), does not replicate it whole-heartedly. For these reason it has not been the central feature of this investigation but acts supporting evidence. Sera Khandro represents the possibility of her own liberation by re-fashioning

visions about the flow and manipulation of vital energies through sexual practices, in her own terms. Tantric literature typically leaves this experience for women ‘unexplored’ and language regarding this process focuses on the ‘male partner’s spiritual advancement’ (204). Tantric Buddhists believed that vital energies naturally flow downward and outward in the form of semen, causing aging, loss of vitality and

death (203). Celibacy therefore slows the aging process and retains the vitality required for yogic practices. It was believed that drawing up ‘vital energy’ into ‘successively higher energy centres of the body’ (i.e. cakras) could actually reverse aging but it required one to break the vow of celibacy. Ultimately, this practice led to ‘the complete realization of buddhahood’ (204). This description is noticeably similar to re-figured practice of varjorlī as a visualisation, as mentioned in Mallinson’s work above (2016: 23). The majority of literature about this in Tantric manuals, such as the 14th century Nyingma master Longchenpa’s commentary on the ‘Guhyagarbha Tantra’ titled ‘Dispelling the Darkness of the Ten Directions’, explains how the vital

nuclei moves from the crown of the male practitioner’s head, to the ‘throat, heart, and navel along with the arising of successive stages of primordial wisdom’ (Jacoby 2014: 203). The seminal fluid and the female ‘vital nuclei’ (i.e. female sexual fluids) are then drawn back up into the male’s body and he gains spiritual insight. This is typical of the heterosexual male-centric descriptions found in Tantric literature. Khandro, on the other hand, subverts this when she describes ‘the man’s cloud of semen raining into the woman’s secret centre [vagina] and pervading upward’ to arouse the vital nuclei and energy centres of the body (204). Evidently she participates in these sexual-based activities for her own liberation.

Khandro does not only re-imagine the energetic and physical practices in her own terms but also destabilizes social relations through constant negotiation of her position as a woman. This ultimately leads to her social and spiritual transformation. Not only does she serve as a consort for male lamas but she explicitly utilises male consorts for her own spiritual benefit (190). Accounts of sexual yoga from her female standpoint flips the ‘familiar Buddhist paradigm of male subjectivity and female

subordination on its head’. (190). Her firsthand descriptions exemplify how women could ‘gain religious realization from participation’ in ‘contemplative practices that include sexual liaisons with multiple male consorts’ (190). There are many instances in which she negotiates her position as a female consort in order to exercise power in this role but I will consider two key occasions to exemplify the process.

In the first instance she resists requests from the monk Gara Gyeltsen, who wants to take her as a consort, on the basis that she does not want to break his vows of celibacy. Khandro is steadfast in her position, even though she is given permission

from her husband. Instead she preaches the dharma and manifests ‘the insight of a wisdom dākinī’, a celestial female being (228-230). As a much sought-after Tantric consort, Sera Khandro often found herself on the ‘fault lines of alternative interpretations of Vajrayāna Buddhist sexual conduct’ and the onus fell on her to determine how to proceed (222). This suggests she held agency and the power to control certain scenarios presented to her.

The second example is when Drimé Özer, her root consort, persuades her to visit the Lama Andzom Rinpoché (214). Again, she attempts to resist this suggestion, as she has ‘no thought of needing to live with a consort’ and does not want to live ‘under others’ sway’ (215). Drimé Özer urges her: ‘if you could be a consort endowed with auspicious connections to prolong Lama Andzom’s life…this would benefit the Buddhist teaching and sentient beings would be happy. In particular, it would be a

resource for you and others to accumulate a vast store of merit’ (215). Here, Özer clearly uses external religious structures to influence her. She does, eventually, submit. When living with Andzom as his consort, Khandro is cured of her debilitating arthritic leg pain. He also recognises her as an incarnation of the dākinī, Künga Buma (215). These occurrences suggest that Khandro engaged in a mutual exchange. Khandro subverts the power the Lamas assume they hold by benefiting from her consort practices. Consequently, there is little sense of submission.

Hatley references Khandro’s experiences to support what he finds in the Brahmayāmala. He asserts that Khandro is ‘transferred’ between Lamas ‘without consultation’ in an attempt to exemplify that women lack agency and power in tantric communities (2015: 10). Patricia Jeffery argues that agency is a matter of perspective: ‘how women’s agency is read…depends on who is reading’ (2001: 466). Whilst

Khandro is urged by male practitioners to act as a consort to other men, she is never ‘transferred’. Instead, if we read Khandro’s situation from her point of view, we find a nuanced, complex set of power relations between male and female practitioners that she constantly negotiates, enacts and modifies. Here, there is a palpable chasm between the ways in which women represent themselves, compared to how they are

represented by doctrinal scriptures in relation to spiritual practices. The insight offered by Khandro’s autobiography prompts certain questions regarding the representations and roles of women in pre-modern yoga practice. Did female hatha yoga practitioners also negotiate their position; did they wield power as the possessors of sexual fluids required by male yogis; were they able to yield power in non-sexual yogic contexts?

Conclusion

To conclude, I will return to the initial question: what roles do women play and how have these roles been represented in the context of yoga practice? In the hatha and tantra scriptural evidence women are consistently represented as sexualized beings. There is an underlying tension between this and the assumption they are connected to the divine, as well as further tension between these converse

representations and the expected roles of women. As demonstrated throughout the essay, nuances arise within and between texts but sexuality is the central theme. In the Dattātreyayogaśāstra (3.1.3: 84-88), women are initially eschewed because they pose a sexual threat to men engaged in hathayoga, yet they hold the key to transcendence: the practice of vajroli defeat’s death and, whilst intercourse is not explicitly

suggested, it is implied that access to an accomplished woman is necessary for the yogi to achieve success (3.5.9). In the Brahmayāmala women are identified with their reproductive organs – yoni, vulva, womb – that in turn serve as a ‘principle locus for worship of the deities’, or as portals to divine experiences during tantric yoga practices (Hatley 2015: 8). Despite these revered representations, women’s roles are

passive, limited and controlled. The yoginī is required to be a vehicle for the sādhaka’s perfection (2015: 14). Regardless of these severe limitations, Hugh Urban argues that the identification of women as celestial beings – such as yoginī, śakti or dūtī, as they are in the Brahmayāmala – opens up opportunities for women to assume power and become ‘spiritual authorities’ and even lead Tantric lineages (2010: 143). This is alluded to in the ‘classification of women who obtain the cakra’ in the Brahmayāmala but is overshadowed by what follows: methods of harnessing this

divine feminine power for the benefit of the male practitioner (2015: 12-13). The potential for a woman to become an authority manifests in the life of Sera Khandro. Khandro is an exception but in the absence of other sources, she enables the possibility of a ‘micro-study’ about how women create self-representations of their roles within tantric and ‘sexual yoga’ contexts (Jacoby 2014: 14).

It is clear that women did engage in non-sexual yoga practices in the hatha tradition. We see this in the Mughal paintings and the brief reference to the requirement of a ‘woman devoted to the practice of yoga’ in the Dattātreyayogaśāstra and the Hathapradīpikā. Dattātreya’s hathayoga addresses ‘the wise man’ (12; 38) and gender difference is clearly an important feature of yoga practice therefore one

should be wary of assuming that devoted female yogic practitioners engaged in the same practices as male yogis. The doctrinal hatha texts I have examined offer little further insight due to the male-dominated system in which they were produced: this is the major limiting factor. From these textual representations we can only reconstruct, not the actual roles, but the idealized roles of women. Regardless of this idealization, it is clear that the ‘yoga’ women participate in today is very different to the yoga practices they were expected to engage in, in the pre-modern period.

Bibliography

Biernacki, Loriliai (2007). Renowned Goddess of Desire: Women, Sex and Speech in Tantra. Oxford: Oxford University Press. Birch, Jason (2011). pp 527-554, ‘The Meaning of haṭha in Early Haṭhayoga’, The Journal of the American Oriental Society, 131.4. Castelli, Elizabeth A. (2001). Women, Gender, Religion: A Reader. New York: Palgrave.

Clémenitn-Ojha, Catherine (2011). ‘Female Ascetics’, pp. 60-67 in Brill’s Encclopedia of Hinduism vol. 3. Ed. Jacobse, Knut A. Boston: Brill. Doniger, Wendy (2009). The Hindus: An Alternative History. New York: The Penguin Press.

Erndl, Kathleen M. (2000). ‘Is Shakit Empowering for Women? Reflections on Feminism and the Hindu Goddess’, pp. 91-103 in Is the Goddess a Feminist? The Politics of South Asian Goddesses. Ed. Hiltebeitel, Alf and Erndl, Kathleen M. New York: New York University Press.

Flood, Gavin (2006). The Tantric Body: The Secret Traditon of Hindu Religion. London: I.B. Tauris.

Grimshaw, Anna (1992). Servants of the Buddha: Winter in a Himalayan Convent London: Open Letters.

Hatley, Shaman (2012). ‘From Mātr to Yoginī: Continuity and Transformation in the South Asian Cults of the Mother Goddess’, pp. 99-129 in Transformations and Transfer of Tantra in Asian and Beyond. Ed. Keul, István. Berlin: Walter de Gruyter.

——— (2013). ‘What is a Yoginī? Towards a Polythetic Definition’, pp. 21-31, in ‘Yoginī’ in South Asia: Interdisciplinary Approaches. Ed. István Keul London: Routledge.

——— (2015). ‘Representations of Women in the Brahmayāmala’ presented at the Symposium: Tantric Communities in Context: Sacred Secrets and Public Rituals in Vienna.

Hauser, Beatrix (2012). Promising Rituals: Gender and Performativity in Eastern India. London: Routledge.

Hawthorne, Sîan (2005). ‘Gynocentrism’, pp 3719-3721, in Encyclopedia of Religion, 2nd edition, vol. 5, Ed. Jones, Lindsay. London: Thomson Gale. Jacoby, Sarah H. (2009) Buddhism Beyond the Monastery: Tantric Practices and their Performers in Tibet and the Himalayas. Boston: Brill.

——— (2014). Love and Liberation: Autobiographical Writings of the Tibetan Buddhist Visionary Sera Khandro. New York: Columbia University Press. Jeffery, Patricia (2001). ‘Agency, Activism, and Agendas’, in Women, Gender, Religion: A Reader. Ed. Castelli, Elizabeth A. New York: Palgrave.

Keul, István (2013). ‘Tracing yoginīs – religious polysemy in cultural contexts’, pp.1- 17 in ‘Yoginī’ in South Asia: Interdisciplinary Approaches. Ed. István Keul. London: Routledge.

King, Richard (1999). Orientalism and Religion: Post-colonial Theory, India and the Mystic East. London: Routledge.

Larson, Gerard James (2009). ‘Differentiating the Concepts of ‘Yoga’ and ‘Tantra’ in Sanskrit Literary History’, pp. 487-498, in the Journal of the American Oriental Society, vol. 129 no. 3.

Mallinson, James (2012). ‘Siddhi and Mahāsiddhi in Early HathaYoga’, pp. 327-344 in Yoga Powers ed. Jacobsen, Knut A. Boston: Brill. ——— (2012) Translation of ‘Dattātreya’s Discourse on Yoga’ ——— (2013). ‘Śāktism and Haṭhayoga.’ In: The Śākta Traditions. Oxford: Oxford University Press.

——— (2015). ‘Yogic Identities: Traditions and Transformations’. Freer and Sackler Galleries. ——— (2016) Forthcoming: ‘Yoga and Sex: What is the Purpose of Vajrolīmudrā?’ in Yoga in Transformation. Vienna: V&R unipress (in the series ‘Wiener Forum für Theologie und Religionswissenschaft’).

Samuel, Geoffrey (2008). The Origins of Yoga and Tantra: Indic Religions to the Thirteenth Century. Cambridge: Cambridge University Press. Ray, Reginald A. (1989). ‘Accomplished Women in Tantric Buddhism of Medieval India and Tibet’ in Women’s Religious Lives ed. Falk, Nancy Auer and Gross, Rita M. California: Wadsworth Publishing Company.

Sanderson, Alexis (1988). “Saivism and the Tantric Traditions.”, pp 660-704 in The World’s Religions, ed. S. Sutherland, et al.. London: Routledge. ——— (2009). ‘The Sáiva Age’ in Genius and Development of Tantrism, pp. 41-348. Ed. Einoo, Shingo, Tokyo: Institute of Oriental Culture, University of Tokyo. Shaw, Miranda (1994). Passionate Enlightenment. New Jersey: Princeton University Press.

——— (2000) ‘Is Vajrayogini a Feminist? A Tantric Buddhist Case Study’ pp. 166- 180 in Is the Goddess a Feminist? The Politics of South Asian Goddesses. Ed. Hiltebeitel, Alf and Erndl, Kathleen M. New York: New York University Press. Singleton, Mark (2010). Yoga Body: The Origins of Modern Posture Practice. Oxford: Oxford University Press.

Sing, Pancham and Rai Bahadur Srisa Chandra Vasu (2009). The Forceful Yoga: Translations of the Hathayoga-Pradīpikā, Gheranda-Samhitā and Siva-Samhitā’ ed. Bhatt, G.P. Delhi: Motilal Banarsidass Publishers.

Urban, Hugh B. (2010). The Power of Tantra: Religion, Sexuality and the Politics of South Asian Studies. London: I.B Tauris.

White, David Gordon (2005). ‘Tantrism’ in in Encyclopedia of Religion, 2nd edition, pp. 8984-8994. Ed. Jones, Lindsay. London: Thomson Gale. Young, Serinity (2004). Courtesans and Tantric Consorts: Sexualities in Buddhist Narrative, Iconography, and Ritual. London: Routledge.