The Feminine Sacred in South Asia

South Asia is the only major region where the “Great Goddess” is still a living reality for beievers- yet its society remains male-dominated. Drawing their examples from ritual practice, myth, and sacred texts the contributors to this volume discuss the place of the feminine within the sacred sphere of South Asian religion. The theme is full of contradictions, for the impurity of woman must be held against the powers she incarnates, and the religious status of these powers is an old theme of debate among Hindu and Buddhist thinkers. Finally, the feminine pole in religious thought cannot simply be equated with human womanhood…. Yet the very presence of feminity in the sacred sphere contrasts with its exclusion from scriptural Islam or from Protestantism, and offers, perhaps, to women a mode of religious expression in an idiom where gender is a central paradigm of thought.

This volume then, contributing to the debate on feminity in South Asian religion, should also be of interest to scholars dealing with gender in a broader perspective.

Harald Tambs-Lyche, a social anthropologist, is professor of ethnology at the University of Picardie- Jules Verne, Amiens (France). He has worked on Gujaratis at home and abroad (London Paridars, Routledge, 1980; Power, Profit and Poetry, Manohar, 1997). He is currently doing fieldwork in Karnataka.

This book started as a session on the Sacredness of Women in South Asia, part of the 13th European Conference on Modern South Asian Studies, Toulouse, August-September 1994. Other than me seven scholars brought in their contributions: Theo Damsteegt (Amsterdam), Sanjukta Gupta (Oxford), Bert van den Hoek. (Leiden), McKim Marriott (Chicago), Andre Padoux (Paris Balgopal Shrestha (Kathmandu), and Catherine Weinberger-Thomas (Paris). I thank them all, as well as several others who participated in the discussions.

Damsteegt and Weinberger- Thomas preferred to abstain from publishing their papers, as these formed part of books which have since been published. Due to several other pressures, Marriott, too, finally decided to abstain. Finally, different versions of van den Hoek and Shrestha's papers did appear in the Nepali research journal Kailash, though they both agreed to publish their present versions here.

Marine Carrin, had, like myself, participated in an earlier session on a similar theme two years back, at the 12th Conference in Berlin, a contribution which had remained unpublished. The Berlin session, in fact, was an amalgam of several proposals among which mine had been one, so that Carrin's paper had been written very much in the spirit of the session as realized at Toulouse. It therefore seemed natural to include an abridged version of it in the present volume.

These alterations have unhappily caused some delay in producing the book, for which I apologize both to the authors and to the CNRS (French National Research Council) whose grant of a subsidy made the publication possible. I hope that, in a field becoming more crowded, ours will still be a useful contribution to the present-day debate.

Since Bachofen (1861), female divinities have fascinated scholars and matriarchy, female dominance, and fertility cults are themes that have had a checkered history. No sooner did we assume the idea of matriarchy to be dead and buried, than it was resurrected by certain feminist scholars and writers of the seventies. They were met, on their own grounds, by Bamberger, who saw as the issue not .

whether women did or did not hold positions of political importance at some point in prehistory ... but that there are myths claiming women did these things, which they now no longer do. This mythological status of primitive matriarchies poses as interesting a problem as any generated in the nineteenth century about the credibility or viability of matriarchy as a social system. (1974: 266-7).

Thus the myths forged by scholars, present or past, join those of the people they studied-making us wonder about the import of this kind of thinking, rather than about the existence of its object (Wood 1996).

Advances in gender studies during the last twenty-five years, have displaced matriarchy from a central position in scientific or ideological discourse. Thus Eisler, retracing Bachofen's path without referring to him or matriarchy, interprets the archaeological findings of the last hundred and thirty years, arguing that an age of partnership and equality of the sexes preceded a change to male dominance some five thousand years ago (1990: xvi-xvii). The cult of the mother goddess, in the subsequent period, is seen as 'memories of a lost age' , when women were equal to men (op. cit.: 59-77).

Contributions in this vein were not particularly well received in the world of scholarship. The work of the late Marija Gimbutas, however- on which Eisler and others lean strongly-must be taken seriously, especially in its original version (1974). Gimbutas located her theories of the Goddess within a framework defending the indigenous origin of much of European culture, as opposed to diffusion from Asia. Her stress on a specific 'European tradition of Goddess-worship distinguishes her from theorists whose claims concern a universal golden age, ruled by the Goddess. Yet in her later work (1989, 1991), she comes close to the latter position, which Wood characterizes as the 'universalist approach'.

Gimbutas, says Wood, here engages in 'the construction of a periphery which embodies all the positive factors of the "lost paradise" of the Goddess-an insistence that it did survive on the margins of patriarchal culture. In this way a context is set up from which the lost paradise can be revitalized as the centre' (1996: 21). Briefly, critics like Wood feel that present interest in a Great Goddess of the past is predicated on current gender discourse and lacks substantial evidence for are-evaluation of the past. Yet, as Hawley puts it in a recent contribution, (1996: 1): 'Could she not be worshipped again? Indeed, she is!'

This brings us to India, where one need neither turn to the past nor to the periphery but observe Goddess cults at first hand (cf. Hawley 1996: 2). Some (Bhattacharya 1977) do indeed see, in this flourishing devotion to the Goddess, a link to a pre-Aryan matriarchal past (we have ample evidence of cults of female divinities dating back to the civilization of the Indus). Muted somewhat in the Vedic texts, the Goddess then 're-: emerges' according to Mackenzie Brown (1990: 1), with the gradual growth of Hindu tradition. But was this the same phenomenon? Certainly, the contributors to this volume are critical of universalist Goddess theories. If gender discourse, quite generally, has found it necessary to distinguish between the biologically determined binary opposition of sex, and the realm of gender, as it appears variously through cultural representation, we must be even more conscious of this distinction when dealing with religion. For the study of religion, by necessity and tradition, is, par excellence, the study of representations.

There is, indeed, as Obeyesekere has shown (1981, 1984), a vast field open to us in investigating the psychoanalytic basis of the symbolic representations in Goddess cults. These may well be of wider significance-O'Flaherty (1980: 113) notes their similarity across the Indo- European world-and I, for one, am convinced about the fertility of this line of thought. But in this volume, only Carrin, to some extent, approaches the problem from this angle, and a detailed consideration of it would greatly transcend the limits of this introduction.

We are dealing, therefore, not with a general principle of feminine sacredness but with its representation in South Asian religious thought and practice. No great religious system of the present makes more of the Goddess than Hinduism does. Her sanctuaries outnumber those to male Gods in most Indian localities, and the variety and number of goddesses is 'simply overwhelming', as Kinsley puts it in his invaluable sourcebook (1988). Yet there is considerable weight behind the argument that they are all variants of a single conception of feminine divinity: Kinsley aptly subtitles his book 'Visions of the Divine Feminine'. Among those who try to define more closely the common essence of the Hindu Goddess is Madeleine Biardeau (1981). The problem of singular or plural, as Hawley (1996: 8) puts it, remains and both Indianists and Indians discuss whether there is one Goddess or many.

In the same vein, we may well argue that the Goddess is generally of lower status than the male God, only to find our reference countered by others, where the Goddess is given preference. Thus, in Tamil Nadu, Daniel identifies three alternative 'models' of the male-female relationship, embodied in Parvati's relation to her husband Siva. In the first, Siva predominates and the human wife is controlled by her husband; the second represents the husband as entirely dependent on the powers (sakti) of the wife, whereas the third refers to androgynous icons where Siva and Parvati form the right and left halves of the same body, and enjoin the equality of the spouses. Indologists (O'Flaherty 1982) as well as anthropologists (Marglin 1982) are familiar with all three possibilities, which would seem mutually exclusive as references for role behaviour: Daniel's point, however, is that villagers do not 'make an exclusive, consistently held choice' but 'make use of whatever model seems meaningful ... in any given context' (1980: 85).

While Babb' s twofold typology (1975) implies that married goddesses are benign and single ones dangerous, Marglin, among others, has criticized this view (1982, 1985). She redefines the relation of sexuality and marriage: the former is not dangerous in itself but when the goddess is celibate her powers are, so to speak, free to oscillate (Hawley 1996: 14) between nurturing and destroying; it is the 'power of life and death' (Marglin 1985: 55). Thus female dominance is not necessarily negative, nor singleness wholly dangerous. We have to conclude, that there is no single or correct way to array this range of possibilities (Hawley 1996: 9), which form, as it were, a semantic field of discourse rather than a structured scheme. What is implied by Daniel's three 'codes' (1980) as well as Marglin's (1980,1985) critique, however, is the mutual dependence of male and female in each case.

The second element these interpretations have in common is the fundamental importance of sakti, the ubiquitous concept of power which is always feminine and in some way or other present in or through females. While the goddess sometimes stands 'supreme' (Hawley 1996: 9), this clearly does not mean that the masculine element is absent from the world she dominates. When the Goddess lacks male counterparts, however, the male element is clearly secularized and subsumed to the female (Coburn 1996: 31-48). The debate, then, is less concerned with the importance of these female powers than with the question of how and by whom they are controlled. For Daniel the question is whether men should control women-and thus their sakti-or whether these powers should be mastered by women themselves. The androgynous alternative abolishes the question: when the couple becomes one, the powers and their control accrue to the same entity.

With regard to the cross-cultural theme of the Goddess, we make here three points. The ritual preponderance of the goddess does not exclude the existence of male dominance in secular society, even at one and the same time and place. Second, the Goddess is not alone but forms part, in India, of an interdependence of male and female elements. Third, in spite of all this Indian women are associated with certain powers that are central to the understanding of Hindu thought.

The feminine dimension of the divine, then, is a living reality in South Asian religion. Sometimes, the Goddess is praised in terms that evoke the myth of the Great Goddess. But even when she is relegated to the periphery, her exclusion has to be argued strongly and persistently. Central or peripheral, ranked above or below her male counterparts, she is constantly present. 'Men know', says Hawley, 'that women possess a power that exceeds the structures they would like to impose, structures that have collectively been called dharma: the way things should be, at least according to the dominant men' (1996: 11). Yet, in some cases (Carrin and Tambs-Lyche 1993) the Goddess is precisely the guardian of dharma, with the king (Tambs-Lyche 1991) but the secular and highly gullible representative of that moral order. In Rajasthan and Gujarat, at least, the Goddess imposes her dharma on reluctant men.

What should we make of this feminine presence in Indian religious thinking? A number of different approaches present themselves: the Goddess and her cult, the exaltation of the feminine principle, the lives of Hindu women.

I had proposed to discuss 'the sacredness of women in South Asia'. Yet sacredness is not the prerogative of Indian women, as my colleagues quickly pointed out, nor do human females dominate the pole of the sacred which is seen as feminine. Indeed, as Andre Padoux notes in his contribution to this volume, the conceptions we discuss are, to a large extent, not those of women but of men. This is certainly one reason for the mystification of the feminine that is implicit in its sacredness, and much of the long-standing discourse on the theme must be seen in this perspective.

While the authors agree that the feminine sacred is a concept of considerable importance in South Asian religion, McKim Marriott quickly criticized my use of the term 'central' in an early communication. He is obviously right: it is difficult to conceive of a thought system where the feminine is central since it is always one of a pair of gendered poles- even if the feminine pole is stressed to a greater extent in South Asia than in most other regions of the world.

This volume deals with the feminine element in cosmology, ritual, myth and religious discourse. We are concerned with the feminine sacred as a dimension of a wider universe of discourse pervading South Asian religion. In trying to define this dimension, philosophers and scholars, Indian and Western, converge on an object of discourse, the feminine sacred.

We leave unexplored the 'status of women' in society.' as well as the deeper levels of language and philosophy, where, as Padoux notes, speech itself is a Goddess. We are concerned, rather, with the way ideas of the feminine sacred inspire action, and with the interpretation of action in terms of such images.

The contributions to this volume share a focus but differ widely in their approach. Andre Padoux works from Tantric texts, while I use what may be characterized as oral tradition confined in writing; Bert van den Hoek, Sanjukta Gupta and Balgopal Shrestha deal, each in their own way, with ritual; Marine Carrin analyses how particular women acquire sacred powers. On the level of literature, says Padoux, the sacredness of the feminine pole is predominantly a male construction, even in the Tantric tradition with its bias towards the feminine sacred. The rejection of females by the Buddha, referred to by van den Hoek, may be seen to represent an opposite view within the same male-dominated universe of discourse.

Yet, when femininity is excluded from the ritual scene, it 'pops up' in parallel ceremonies; even the Buddha could not dispense with the feminine sacred, as van den Hoek shows. Its presence is also fraught with danger, as Shrestha shows clearly: the two contributions from Nepal show the masculine and feminine poles of sacredness to be conflicting as well as contrastive; their mediation is not a foregone conclusion. The world of Indian religion, as Padoux puts it, 'might be a masculine religious world where one cannot avoid the feminine'. Moreover, if the female sacred is a male construction, it is one which, once established, refuses to comply with male dominance.

And yet there is indeed a mediation, as Shrestha shows. Not only does the public at Sankhu refuse to acknowledge that the goddess really kills the demon, this refusal also implies that there is affinity between the adversaries, even if only at the invisible level of the ritual.

This indicates that female and male, in South Asian religious discourse, are not simple poles on a common gender 'variable' but include a number of different contrasts which must be conceived as co-existing within a multidimensional semantic space. 'Hindus,' Sanjukta Gupta reminds us, 'also believe that every person is essentially divine .... '

Gupta goes on to show how women, through their ritual activity, control or manage their inherent powers, appropriating to themselves the agency of their sakti. Women are controllers just as men are. This implies taking a stand with regard to the male-female opposition: if, as some interpretations have it, sakti is female and needs male control, then woman herself, insofar as she controls her powers, would seem to incorporate the masculine element: but should we accept that women are androgynous?

Such a conclusion might seem to fit the goddesses with which I deal, for they certainly involve themselves in the male sphere of politics and war. Though their images are not androgynous, their powers might be; indeed the weapons carried by the Goddess as a warrior have a masculine aspect. This, again, would explain why goddesses generally have no need to marry, and also why the marriages I discuss are means of linking them to this world rather than to the male sex.

In Carrin's paper, however, it is clear that the masculine element in female sacredness has its counterpart in the feminization of the male ojha or priest. In other words, it may be the Hindu concept of sacredness itself that is ambiguous in terms of gender. The trident or trisida is a case in point. Weapon of Siva, as well as of the Goddess, the trident is apparently feminine when, as in western India, it is commonly described as Siva's sakti, or, as we might choose to say, the symbol of this feminine power. Yet Carrin tells us that when her priestesses, on the boundary of low- caste Bengali Hinduism and tribal tradition, assume the right or strength to carry the trident, they lose part of their femininity-to the point that they may cease to menstruate.

The Bengali priestess, the western Indian Goddess and, indeed, the ordinary housewife acting as a ritual agent share one important feature: they mix 'male' and 'female' elements, and they do, in different degrees, become sexually ambiguous.

But if male and female share the same qualities, albeit in different proportion,' how are we to deal with those arguments and rituals which emphasize gender opposition? It appears that we must accept that gender difference, too, is sometimes emphasized, sometimes denied, in South Asian religious discourse. At one level, male and female are opposites; at another they merge in common divinity, as they do in common humanity. But what we may see, from the outside, as 'levels', dissolve at closer range into multiple positions and arguments, often confusingly polysemic. This is not just what Daniel (1980) calls the 'toolbox approach' of Tamils to religious issues but as O'Flaherty underlines 'there is an old pedigree for Indian eclecticism' (1980: 6). We must, she says, try to 'hear as many as possible of a myth's voices' (1980: 12).

Male and female is a theme, and an important theme, in Hindu religious discourse, but even the importance of this theme may not be beyond discussion in every particular case. What makes the study of the feminine sacred possible for South Asia is that texts, myths, and popular discourse constantly return to it, and that there are in spite of opposition and disagreement, certain concepts which have long remained central to this discourse.