History of Religions

427

that he or she has carefully observed in the light of texts belonging to the sect under study. Even if the aim is to study a single Tantric author, he should be put into proper perspective with reference to his predecessors in the tradition and his contemporary social context. However, I should like to express my appreciation of this book. Its thorough understanding of the contemporary Traipurakas of South India is quite valuable. If read with considerable caution, it can be used profitably by those who have some previous experience in Tantrism.

SANJUKTA GUPTA Oxford, England (Emeritus Utrecht)

Criminal Gods and Demon Devotees: Essays on the Guardians of Popular HinAlbany: State University of New York duism. Edited by ALF HILTEBEITEL. Press, 1989. Pp. xii+491, 45 plates. $74.50 (cloth); $24.50 (paper). In the not so distant past, for many Western scholars, and perhaps Indian scholars too, the word "Hinduism" was primarily associated with the idealistic monism of Advaita Vedanta philosophy, vegetarianism and nonviolence, and the aesthetic refinement of temple architecture and Brahmanic ritual. We knew, of course, that periodically the gods must gird themselves to battle the demonic forces that threaten the divine order, but even so, in our constructions of the Hindu world we tended to relegate the forces of chaos, impurity, danger, and. violence to the peripheries, if we acknowledged them at all. All that is fast changing: recently a number of scholars have argued that behind its composed exterior, Hinduism harbors a heart of darkness.' Evidence of Hinduism's darker elements can be found in a number of places. First, there are the so-called popular and folk cults of deities who forcefully possess their devotees and who demand offerings of flesh and blood. And if one were to argue that such cults are peripheral and secondary to the mainstream classical Hindu tradition, many scholars would now respond that the boundary between classical and folk is not as clearcut as we had once assumed, and further, that the so-called mainstream classical tradition shares many elements with its folk "peripheries." This is a position with which many of the contributors to this volume would agree. The topical rubric that draws these essays together is indicated in the title chosen by the volume's editor, Alf Hiltebeitel, who organized the conference where many of these essays were first presented. In his introduction Hiltebeitel offers a succinct explanation of the categories "criminal gods" and "demon devotees":

Among the works that come to mind are J. C. Heesterman, The Inner ConJicr of Tradition: Essays in Indian Ritual, Kingship and Society (Chicago: University of Chicago Press, 1985); and David Dean Shulman, The King and the Clown in South Indian Myth and Poetry (Princeton, N.J.: Princeton University Press, 1985). Other relevant sources may be found in the selected bibliography of this volume.

'

428

Book Reviews

"Criminal godsn-if we can take the term criminal metaphorically. . . -are perhaps a worldwide phenomenon: gods who violate the sacred codes and boundaries by which other gods, and humans, would seek to live. What is specific to the "criminal gods" of Hinduism is the specifically Indian codes they violate-societal, sexual, theological, culinary, sacrificial-and the ways they violate them. "Demon devotees," on the other hand, are perhaps uniquely Indian (though not necessarily uniquely Hindu), for their mythologies are shaped by a theology of bhakti, or devotion, in which the gods repeatedly convert their demon adversaries-sometimes by defeating them, but more often by killing them (implying the principle of reincarnation)-into their devotees. [P. 11

These "preliminary and minimal definitions" account either directly or secondarily for most, if perhaps not quite all, of the subjects which engage the attention of the fifteen contributors. Included is, for instance, the Tamil criminal god and violator of codes par excellence Kattavarayan, who makes an appearance in the pieces contributed by Madeleine Biardeau, David Shulman, and Eveline Masilamani-Meyer. In the demon devotee category (and Hiltebeitel rightly notes that there is considerable overlap between the two categories), we find such figures as POttu Raja and Muttal Ravuttan, guardians found in Draupadi temples in Tamilnadu; Mani and Malla, demons defeated by Khandobii, a form of Siva popular in western India; and Bhairo, demonic guardian at the temple of Vaisno Devi in Jammu. (These figures are discussed, respectively, in the essays of Hiltebeitel, John Stanley, and Kathleen Erndl.) Also treated are those "criminal" manifestations of Siva within the classical tradition, Virabhadra and Bhairava, discussed, respectively, in the essays by David Knipe (in relation to a cult of deified dead children found in Andhra) and ElizabethChalier Visuvalingam. Somewhat removed from the two categories that at least nominally provide the common ground for these essays but which nevertheless share many similar concerns are the contributions by Gunther Sontheimer on Khandoba and his alter egos Mallanna and Mailiira; by Velcheru Narayana Rao on the Telugu epic hero Kiitamaraju; by Diane Coccari on the Bir Babas (a category of deified dead) of Banares; by David Lorenzen on new data concerning the KBpBlikas, a sect of ~ a i v i t e ascetics who took Bhairava as their model; by Dennis Hudson on those nciyanmcir (model devotees of Siva in Tamil ~aivism) who expressed their love for Siva in particularly violent ways; and by Joanne Punzo Waghorne on the Tontaimans, rulers of the Tamil kingdom of Pudukkottai, in whom Waghorne finds an analogue of the demon devotee of mythology. Finally, Sunthar Visuvalingam develops a theoretical model that he locates paradigmatically in the conceptual underpinnings of the preclassical Vedic diksci-a model that he and others (most notably, E.-C. Visuvalingam) refer to as "transgressive sacralityW-and proceeds to elaborate his model in reference to the material treated by the other contributors. Hiltebeitel indicates in his introduction that "the terms ["criminal gods" and "demon devotees"] are not employed in this volume to fix these types, or even further to define them, but to draw our contributors into a common discussion of the problematics that surround such figures in the Hindu tradition" (p. 1). He subsequently identifies these problematics, as he perceives them, by introducing each of the volume's component papers, which he has arranged to follow "a sort of counterclockwise tour of the subcontinent," beginning and

History of Religions

429

ending in Tamilnadu. There is little reason for me to recapitulate here what Hiltebeitel has already done so well in his introduction. Therefore, rather than attempting comprehensively to survey the major ethnographic, textual-analytic, or theoretical points made in each of the fifteen essays, I will confine my observations to what for me are some of the more striking and far-reaching issues that emerge from these essays. Some of these issues are very far-reaching indeed. Several of the essays, though they do not explicitly attempt to define what is meant by or properly belongs to the realms of popular, folk, or classical Hinduism, nevertheless provoke close scrutiny of the degree to which these currents are autonomous or are intermingled and mutually informing. To those who are familiar with the recent work of Biardeau, it will come as no surprise that she finds numerous echoes of the Vedic horse sacrifice in the festival ritual of the Tamil village goddess Miiriyamman or that she takes the position that in the particular context of the village goddess's festival "the whole of Hinduism is present in its superior and inferior forms" (p. 23). Biardeau's principal points of reference are the Mahcibhdrata and the Vedic sacrifice, and she proceeds to decode the initially puzzling ritual elements of the folk cult by identifying what she refers to as "classical references" (p. 20). While her method would appear to privilege the classical over the folk, her explicit aim seems to be, rather, to demonstrate that Hinduism as a whole is held together by a set of core themes (or "deep structures," if you will) that appear in varied textual, iconographic, and ritual forms ("surface structures"), which, transformed in various ways, may be thought of as multiforms of one another.' Nevertheless, one has the impression with Biardeau that one has somehow explained the "real" meaning of the folk surface structure by referring it to its classical counterpart. Biardeau is not the only contributor who is concerned with the status of the several currents within Hinduism relative to one another and indeed with whether particular mythic themes, ritual performances, etc., are most profitably interpreted in reference to a larger whole, be it a body of myth, a repertoire of ritual variants, or even all of Hinduism. Many of the contributors employ structuralist or quasi-structuralist methodologies and demonstrate an inclination to approach the vast and variegated Hindu universe as a repertoire of core mythic motifs, iconographic forms, and ritual practices that are transformed, inverted, transposed, linked, conflated, assimilated, projected, encompassed, prefigured, symbolized, and/or analogically related in various ways.3 (I quickly lost track of the number of times the word "multiform" appears in this collection.) Elizabeth-Chalier Visuvalingam, for one, candidly argues that the project of interpreting the mythology of Bhairava and related sectarian practices is best served by a structural approach, as opposed to a historical or philological
The terns "deep structure" and "surface structure," borrowed from linguistics, are not actually employed by any of the contributors, but the ideas these terms signify are relevant. To string these terms together in this way perhaps is to parody the language actually employed by the authors of these essays, but all these terms or their grammatical variants do appear at various points in the book. Of course, some of the contributors are more inclined to employ such language than are others.

430

Book Reviews

one (pp. 159, 166-67). Structuralist inclinations are also clearly evident in Shulman's elegant analysis of the story of Kiittavariiyan. He concludes his essay by observing that the story of Kiittavariiyan may be characterized "as a kind of anti-purnna securely rooted in the social and cultural world of the lower castes" and that "in this case, as in others, the contrast [between the mythology of Kgttavarayan and that of Murukan whose mythology contains many similar but inverted motifs] is based on specific thematic and semantic transformations between texts which express the variation in context" (p. 63; emphasis added). Similarly, Hiltebeitel writes, "I will approach the links between them [Draupadi's guardians POttu R2ja and Mutt21 RBvuttan] . . . from a structural perspective that is not made explicit in the myths themselves, but is implicit within the cult's mythology a s a whole" (p. 348; emphasis added). Thus, these and other contributors are united in their commitment to an approach that attempts to get at deep structures that can be brought into focus only by relating the particular data at hand to larger wholes of one sort or another, which may range greatly in scope, culminating with Biardeau's understanding of "the whole of Hinduism." In the quest to identify a deep structure that undergirds the vast landscape of surface structures associated with Hinduism, it is S. Visuvalingam and E.-C. Visuvalingam who are by far the most ambitious. This is not the place to attempt to explicate the Visuvalingams' theory of "transgressive sacrality" (or "transgressive sacrificial embryogony," as E.-C. Visuvalingam refers to this theory at one point) and its numerous analytic applications, many of which are very intricate indeed. E.-C. Visuvalingam's frame of reference is the mythology of ~ i v a ' s"criminal" manifestation, Bhairava, in its numerous variants and related sectarian practices. As mentioned above, S. Visuvalingam establishes the transgressive paradigm in the context of the Vedic diksd (hence the introduction of "embryogony," since the diksd involves an "embryogonic" death and rebirth of the diksita) and subsequently assimilates virtually all the topics treated by the other contributors to this paradigm. It is the Visuvalingams among the contributors to this volume who appear to be most intent on revealing the Hindu heart of darkness ("the transgressive center of the nevertheless Brahmanical stage" [S. Visuvalingam, p. 4331) as the ultimate goal of their scholarly endeavors. The Visuvalingams' essays are indeed ambitious, and I, for one, find them to be problematic in their attempt to unearth a pattern that universally pervades Hinduism. The Visuvalingams are certainly adept at demonstrating how the transgressive pattern can be found in all sorts of material, but as is perhaps inevitable when a scholar becomes attached to an intellectual construct, the construct can all too easily be seen as a kind of master key suited to unlock every door in its path. On a purely intellectual level, the key may appear to fit very well indeed, but by pursuing such an approach without restraint there is always the danger of flattening the object of study and subordinating it to a theoretical construct. This is a common criticism of structuralist and poststructuralist scholarship, and I see this as a problem in the Visuvalingams' promotion of "transgressive sacrality" as a master key. I would argue that we should be wary of phrases such as "the single all-embracing symbolic universe of the