Classifying the Universe: The Ancient Indian Varna System and the Origins of Caste

by Frederick M. Smith
Classifying the Universe: The Ancient Indian Varna System and the Origins of Caste
Frederick M. Smith
Journal of the American Oriental Society
Start Page: 
End Page: 
Select license: 
Select License

 Reviewed work(s): Classifying the Universe: The Ancient Indian Varṇa System and the Origins of Caste by Brian K. Smith I began this book shortly after completing J. C. Heesterman's Broken World of Sacrifice: An Essay in Ancient Indian Ritual (Chicago: University of Chicago Press, 1993; see my review in Journal of Ritual Studies 9.2 [1996], as well as the preceding review, by C. Minkowski). At first it appeared that Brian Smith and Heesterman had read two entirely different Vedas, each with its own scholastic history and secondary literature. Heesterman's Veda is a historical document, though one that must be deciphered only by denying any poetic, literary, or theological sensibility that the authors of the Vedic texts might have possessed. Brian Smith's Veda, on the other hand, is a massive sociological tract largely decipherable through proper understanding of the Vedic authors' poetic, literary, and theological sensibilities. Heesterman reads Vedic myth as thinly veiled history, Brian Smith as thinly veiled sociology. Both studies are, however, strongly influenced by the idea that violence lies at the root of their subject: physical, sacrificial violence in Heesterman's case ("anthropophagy cannot be ruled out" [p. 176]), and the violence of word, ideology, and class oppression in Brian Smith's case ("the vis or 'masses' are regarded as the special delicacy of the Kshatriyas" [p. 47]). Heesterman's Veda seems guided by the intoxicating aroma of roast beef, Brian Smith's by the insatiable quest for status and power. Nevertheless, class and sacrifice commingle closely in the Veda, as both Heesterman and Brian Smith are fully aware. So there is a complementarity in their work, even if neither would claim bandhuta with the other. B.K.S.'s chapters revolve around classifications of society, the gods, space, time, flora, fauna, and revelation. For much of this he depends on thirteen extended cosmogonies presented in various ritual texts, including the Aitareya, Kausitaki, Satapatha, Pancavimsa, and Jaiminiya brahmanas, the Tattiriya, Maitrayani, and Kathaka samhitas, and the Maitrayani Upanisad. Classifying society means the assignation of abstract qualities to the various varnas and the establishment of a fixed order of authority. The gods also have their varnas, though they were not always fixed. For example, Agni, Brhaspati, Vac, and Mitra were brahmanas; Varuna, Rudra, Vayu, Yama, and Visnu were ksatriyas; multiple deities such as the Visve Devas, Adityas, Maruts, Vasus, and Rhus were vaisyas or sudras; Savitr and Soma could be either brahmanas or ksatriyas; Sarasvati could be either brahmana or vaisya; and Prajapati could be of any varna. It is important to note that vaisyas and sudras are almost never distinguished: vis includes both varnas. The classification of space encompasses the three characteristic regions: earth (brahmana), atmosphere (ksatriya), and sky (vaigya); as well as the four primary directions: east (brahmana), south (ksatriya), west (vaisya), and north (varying images, but tending towards sudra). For classificatory purposes, time was broken into the seasons and the parts of the day. Among the seasons, spring was a brahmana, summer a ksatriya, the rainy season and autumn were vaisyas, the winter and cool seasons were indeterminate, but tended toward sudra. Among the parts of the day, the morning was brahmana, the midday ksatriya, and the afternoon vaisya. Among flora, trees were especially subject to varna classification. For example, the palasa was brahmana, the banyan and khadira ksatriya, and the pipal (usually) and udumbara vaisya. Among fauna, large domesticated animals were subject to varna classification; for example, goat and cow were brahmana, horse and bull ksatriya, cattle, sheep, and dappled animals vaisya. By revelation, B.K.S. intends the vyahrtis, or syllables bhuh, bhuvah, svah, the Vedic meters, certain parts of the sacrifice, including the samans and the fireplaces, and even the Vedas themselves. For example, the gayatri meter, the rathantara saman, the ahavaniya fireplace, and the Rgveda would be brahmana; the tristubh meter, the brhat saman, the daksinagni fireplace, and the Yajurveda would be ksatriya; while the jagati meter, the vamadevya saman, the garhapatya fireplace, and the Samaveda would be vaisya. I should point out that these models are not identical in all texts, and B.K.S. rightly (and frequently) acknowledges fluidity, diversity, variation, and incompleteness (e.g., p. 196).  The authority of these cosmogonies for the authors of the Vedic texts can hardly be questioned, as B.K.S. presents considerable supporting textual evidence. The nature of the cosmogonies is formulaic: items are linked serially with each other. A simple example would be from his "Cosmogony II": the mantra bhuh is linked with this world, the brahman power, and the Self; the mantra bhuvah with the atmosphere, the ksatra power, and humans; and the mantra svah with the sky, the power of the vis, and animals (p. 331, SB This cosmogony mentions what B.K.S. labels the "elemental qualities," namely the primary metaphysical powers brahman, ksatra, and vis. These constitute one of the two important categorical schemes that guide the cosmogonical identifications. The other consists of what B.K.S. calls "essential powers," which are secondary metaphysical forces, such as brahmavarcasa and tejas associated with brahmanas; virya, ojas, and bala associated with ksatriyas; and pusti, urj, and anna associated with vaisyas. Such cosmogonies were obviously expressions of efforts to make sense of the universe. They were also, B.K.S. contends, efforts, often insidious, by the brahmana authors to define and strengthen their own social position. "Vedic classification according to varna, like all other religiously oriented taxonomies, is rooted and legitimated in cosmogony" (p. 59). This is indeed the guiding principle of the book, most of which is occupied with unpacking cosmogonic details and lineages, and highlighting their relationship with varna. He discusses strategies employed by the brahmana authors: "Cosmogonic myths comment on the present: they legitimize the present state of affairs and ensure that any challenges to the status quo be cast as deviations from the norm; or they can decry the degeneracy of the times by positing a past, and lost, ideal situation." Furthermore, "[p]ortraying the varna system as coeval with creation was one of the primary means by which the Brahmin authors of the Veda represented their worldview as authoritative" (p. 59). He thus concludes that "the authors purposefully insinuated their own class interests into the cosmogonies they invented and disseminated" (p. 60). Though the evidence for this is substantial, it is often presented in the texts as much more marginal than B.K.S., with a hypothesis to prove, would have us believe. The brahman. a varna is not always prior to or productive of everything else. In fact the brahmana and other varnas are often very incidental to certain cosmologies, occasionally appearing as an afterthought simply to complete a formulaic paradigm, in other cases not at all. The book is admittedly repetitious, though I should quickly add that it is also fairly easy to read. The subject, at least as presented here, does not necessitate deep philological investigation. This, combined with the repetitiveness of the evidence and the interesting introductions to each chapter, contribute to the book's readability, at least for Vedic scholarship. Smith justifies the book's length and tediousness by claiming that the mere weight of citations is necessary to overturn the entire history of scholarship on the subject up to this point, which has consistently contended that Vedic thought is disorderly, unsystematic, "comprised of poetic flights of fancy, mystical esotericism, and/or priestly conceits (take your pick), and it is distinguished by its proclivity toward drawing equivalences or identities ... that are variously described as inconsistent, fanciful, and absurd" (p. 13). One of his primary missions is to demonstrate that "there is, after all, a consistency and logic to the Vedic connections others have dismissed" (p. 13). This he has easily accomplished. For this reviewer, the most enjoyable chapter of the book deals with classification of fauna. Besides the classification of animals according to their varna, B.K.S. discusses the classifications of sacrificial animals according to dental and foot structure, mode of procreation, and edibility. The problem, of course, is the human who is exempt because of being two-footed, "five-clawed," and with incisors in both jaws. Nevertheless, he is first in enumeration of pasus and is "generally proclaimed in the texts to be the highest of all possible sacrificial victims" (p. 251). This, B.K.S. would agree (contra Heesterman), is an expression of categorical panditry, rather than actual practice. Though this book teaches us much about Vedic mythology and sociology, we should not mistake it for a textbook on either subject. It casts a certain sociological light on mythology that future researchers will find useful and demonstrates beyond doubt that one or more classificatory systems may be detected in the construction of bandhu, the identifications which pervade the Vedic ritualistic universe. B.K.S. proves forcefully that there was a method to the madness of the Vedic ritualistic theologians, demonstrating that previous scholars were indeed very shortsighted in their cavalier dismissal of this important subject. Several theories, distinct but not unrelated, depicting a fundamental organizing principle of the Vedas, have been presented in the past generation: Heesterman asserts that it is sacrificial violence, Wendy Doniger O'Flaherty sees it as sex (or sex and violence), Brian Smith views it as varna. All of these ideas, it may be noted, are contrary to both the more spiritualized views of most Indian scholarship, as well as the anthropological or mythological orientation of most previous Western indological study. Perhaps partially driven by reflections on this, Brian Smith closes his book with an impassioned justification of his project, which in the end he sees as breaking down the contentions of Vedic religious discourse. This, possibly combined with a chance encounter in a health food store with an advocate of transcendental and ahistorical Veda imbedded with impersonal truth (p. 324), may have pushed him to the dramatic and scholastically righteous, but ultimately very pedestrian, conclusion that religion and scholarship on religion are "fundamentally opposed enterprises" (p. 325). FREDERICK M. SMITH UNIVERSITY OF IOWA COPYRIGHT 1996 American Oriental Society 

  • Recommend Us