Religion in Human Evolution: From the Paleolithic to the Axial Age

by Robert N. Bellah

  Even if, in trying to answer our questions, we are largely confined to language-to surviving texts from the period-we are immediately confronted by the question of what language. Asoka’s inscriptions and those of other rulers for several centuries after him “were composed not in Sanskrit but in various Middle-Indic dialects, sometimes referred to as Prakrits. While closely related to Sanskrit, these dialects were considered entirely distinct from it by premodern Indian thinkers.“216 But a number of highly important Sanskrit texts have survived that we have strong internal reasons for believing date from the centuries just before and after the turn of the Common Era. The Prakrits presumably represented the spoken languages of their day, whereas Sanskrit was an elite literary language. The Prakrit inscriptions spoke to everyone, whereas Sanskrit was the language of a special group. Some Prakrits did develop at least for a time into literary languages, and Pali, a hybrid of several Prakrits in which the early Buddhist scriptures were written, survived for a long time in southern India and then up to the present in Sri Lanka and Southeast Asia as the primary literary language of Theravada Buddhism, one self-consciously not Sanskrit. It was, however, Sanskrit that would come to be the “cosmopolitan language,” to use Pollock’s term, in the following centuries (by the middle of the first millennium CE, even Buddhists in northern India began to write in Sanskrit), and some of the most influential Sanskrit texts were already written in Mauryan, or, more likely, immediately post-Mauryan times.

  In trying to understand why Sanskrit was so central in the surviving texts of our period, we can start with the fact that Sanskrit was the language of the Vedas and so had long been the language of the Brahmanical tradition, and the Sanskrit texts of our period were largely written by Brahmins. But there is one fundamental difference between the language of the Vedas and the new literary Sanskrit of post-Mauryan times: the former was oral, the latter was written. Writing provided the possibility that had remained only incipient in the oral tradition: second-order, critical thinking, beginning with language about language itself, in the new science of grammar.

  Panini’s Astddhydyi was a pivotal text. It was a treatise on the grammar of the Vedas that was part of a series of texts auxiliary to the Vedas called the veddnga that were developing to help students cope with a language increasingly remote from contemporary speech. As such it was part of the natural continuation of the Vedic tradition. But it was also the first of what would later be called sdstras, that is, systematic treatises concerning a particular subject, in this case grammar. Panini’s capacity to think rationally about language made his book a masterpiece in its genre-in advance of linguistic reflection anywhere in the world when it was written and still capable of stimulating the development of modern linguistics when it was translated into Western languages in the nineteenth century.

  Panini is usually dated in the fourth century BCE-on the cusp of Indian literacy-and there is an ongoing argument as to whether the Astadhyayr was originally oral or written. This is an argument about which I have no expert knowledge, but it seems to me likely that, though the book is a reflection on memorized speech, it was itself written and that writing was essential for the reflexive character of the book, that it was a critical reflection on language. Yet we also must remember that Panini’s book was not about language in general but about Vedic Sanskrit in particular, and that he undoubtedly believed the language he was studying was true language in itself, uniquely different from any other, which suggests the tension between the universal and the particular that exists in the Indian intellectual tradition from early on, and the special nature of Sanskrit as the “preeminent language of literature and systematic thought” long after its exclusively ritual use had been overcome.217 It is also worth noting that Buddhists did not use Sanskrit, at least not for many centuries. The Buddha is supposed to have told his disciples to preach in whatever language the people understood, at a time when no one spoke classical Sanskrit, so the decision not to use Sanskrit was deliberate. Pali developed out of one or more spoken languages but soon became a special language itself, and almost surely a written one fairly early, surviving in the Theravada canon and southern Buddhism generally until the present, itself becoming an elite language known almost exclusively to educated monks. The split in language was an expression of the profound split in culture that was developing between Buddhism and what would be known as Hinduism.

  There are sdstras somewhat later than Panini’s Astadhyayi that deal with the relation of religion and politics more immediately related to our present discussion, but before considering them we may just say a little about what the term sdstra refers to. Pollock sometimes translates it as “science” or “systematic thought.“218 He even calls it “theory,” which, unlike the other terms he uses for it, he puts in quotes. In speaking of the proliferation of sdstras in later centuries, he goes so far as to say that “nothing in old India was untheorized.” However he makes it clear that by “theory” he mainly means “detailed inventory and taxonomy.“219 Consequently he emphasizes the empiricism of Sanskrit systematic thought, the belief that “nothing is beyond the reach of Sanskrit sdstra; everything, everywhere, however intimate [he is talking here about the Kamasutra], is knowable and has become known.“22° Clearly inventory and taxonomy, expressed often in sdstras by endless lists, are the beginning and foundation of scientific theory; but systematic thought becomes theory without quotation marks only when it reaches the level of generalizations that can be put to the test. It is interesting that the earliest sdstra, Panini’s Astadhyayi, was the most scientific in this sense.

  We noted above the early Indian idea of the three (or four) ends of lifedharma (duty), artha (success), and kama (pleasure), also moksa (salvation, liberation)-and there were sdstras devoted to each of them, notably Manu’s Dharmasdstra, Kautaliya’s Arthasdstra, the Kamasutra, and the Yoga Sutras concerning moksa, though each of them is a composite text difficult to date precisely. The Arthasastra, a treatise on economics and politics, used by Romila Thapar to provide some sense of what Mauryan rule was like in the absence of more direct data‘221 gives a nod to the preeminence of dharma in rulership, but is mainly devoted to the practical exigencies of rule without much sensitivity to moral or religious issues. Wendy Doniger writes, “The Artha-shastra is a compendium of advice for a king, and though it is often said to be Machiavellian, Kautilya makes Machiavelli look like Mother Teresa.“222 The Arthasastra, like most sastras, consists largely of a list of rules, rules that can be read as descriptive but that are primarily prescriptive, so we cannot use them to tell us exactly how things were. Although at times seeming to describe a big state, the Arthasastra at other times seems to be describing a group of small states maneuvering for hegemony between enemies and allies. Undoubtedly the text, parts of which were written at different times by different authors, was concerned with both, though it is often not clear how. The great emphasis on a complex system of spies suggests precarious rule lacking in deep legitimacy. Because the Arthasastra was concerned with artha and not primarily with dharma, it is not particularly helpful with respect to the relation of religion and politics.

  Manu’s Dharmasastra, on the other hand, though it overlaps in part with the Arthasdstra, gives much more attention to the relation of religion and politics. Written by and for Brahmins, Manu’s Dharmasastra is first of all a kind of handbook for proper Brahmin behavior but secondly it is concerned with how kingship relates to dharma and more specifically to the Brahmins. It was preceded by and draws from the Dharmasutras, which are classified among the veddnga, that is, works auxiliary to and explanatory of the ritual and linguistic complexities of the Vedas. As we saw in the case of Panini’s Grammar, these were systematizing works moving in the direction of the later sdstras. The four major Dharmasutras are tentatively said by their translator, Patrick Olivelle, to date from the early third to the late second centuries BCE; if these dates are accurate, they may have been written In any case they consist largely of lists of ritual rules and rules of condu
ct for Brahmins through the course of the life cycle (many of these rules also apply to the other two twice-born castes), with only 6 to 12 percent of the texts devoted to statecraft.224 Though Manu develops the same subject matter, adding a much longer treatment of rulership, his book is an organized treatise, beginning and ending with cosmological and religious reflections: it is thus a true sdstra.

  Manu’s Dharmasastra is a work of great importance, with continuing influence to the present day, because it attempted to do three things, each central to the continuity of the tradition but difficult to synthesize: (1) codify and absolutize the Vedic tradition as it was understood in the author’s day; (2) affirm the new strand of Brahmanic spirituality as represented by the Upanisads; and (3) respond to the challenge of the ethical universalism of Buddhism and of Asoka’s Dhamma. With respect to the last point Olivelle writes, “The very creation of a Brahmanical genre of literature dedicated to Law (dharma) [of which Manu was the culmination] was possibly due to the elevation of this word to the level of imperial ideology by Asoka.“225

  By giving the book the name of the first man and/or the first king in Indic mythology, Manu, the claim to virtual canonical status is clear. But this claim in turn is based on the status of the Vedas as uncreated scripture, on which the book of Manu claims to be based. Thus the Vedic injunctions with all their ritual and other forms of particularity are placed beyond question. Yet the renouncer tradition of the Upanisads, with its radical rejection of the fire sacrifice at the heart of the Vedas, is also affirmed. In fact there is much more emphasis on nonviolence and nonkilling of animals than there is on Vedic sacrifice, yet the latter is never rejected. Manu resorts to language that other religions have at times found convenient: “Killing in a sacrifice is not killing … The violence to those that move and those that do not move which is sanctioned by the Veda-that is known as non-violence” (Manu 5.39, 44).226 Doniger points out that at a time in history when Vedic sacrifice had become “largely irrelevant and to some extent embarrassing,” Manu must still defend it, even using the terms of its critics to do so.227

  With respect to ethical universalism, Manu uses two tactics, each with its own problems. First, there are several places where Manu describes what Doniger calls general dharma as opposed to particular dharma. Doniger contrasts the particular dharma arising from caste, svadharma, to universal dharma, “sometimes called perpetual dharma [sanatana dharma] or dharma held in common [sadharana dharma]” that involved general moral precepts applicable to all classes and castes.228 She cites one verse: “Nonviolence, truth, not stealing, purification, and the suppression of the sensory powers are the dharma of the four classes [varnas], in a nutshell” (10.63), and an overlapping one to the same effect (12.83-93), as well as one (6.93) somewhat different (lacking nonviolence) for the top three (twice-born) varnas.229 Perhaps particularly significant because coming very near the end of the book in a discussion of the “highest good” is the list at 12:83: “The recitation of the Vedas, inner heat, knowledge, the repression of the sensory powers, non-violence and serving the guru bring about the supreme good.“230 However, Manu goes on to qualify what looks to be a rather heterogeneous list, in which only nonviolence seems to count as an ethical universal, by saying:

  One should understand that acts prescribed by the Veda are always a more effective means of securing the highest good both here and in the hereafter than the above six activities. All these activities without exception are included within the scheme of the acts prescribed by the Veda, each in proper order within the rules of a corresponding act. (12.86-87)231

  Doniger draws the implication of such a passage when she writes, “The particular rule generally overrides the general rule; sva-dharma trumps general dharma.“232

  Another strategy for affirming a kind of universalism while defending particularism was Manu’s portrayal of the Brahmin as being the most perfect and complete manifestation of human flourishing, the universal man. The other varnas are then defined in a kind of subtraction theory as being like the Brahmins but lacking certain of their qualities, such as the competence to perform sacrifices, ending with the Sudras (and of course the outcastes), who cannot hear or understand the Veda. If the Brahmin ideal were a model for everyone, so that anyone who acted like a Brahmin could be considered one (an idea suggested in Buddhist scripture), it would indeed imply ethical universalism. Yet Manu relentlessly reaffirms the impossibility of this idea. In his cosmological introduction he sets down what has been and always will be:

  In the beginning through the words of the Veda alone, [the Lord] fashioned for all of them specific names and activities, also specific stations … As they are brought forth again and again, each creature follows on its own the very activity assigned to it in the beginning by the Lord. Violence or non-violence, gentleness or cruelty, righteousness (dharma) or unrighteousness (adharma), truthfulness or untruthfulness-whichever he assigned to each at the time of creation, it stuck automatically to that creature. As at the change of seasons each season automatically adopts its own distinctive marks, so do embodied beings adopt their own distinctive acts. (1.21, 28-30)233

  Doniger quotes Ramanujan commenting on Manu’s “extraordinary lack of universality” in making dharma so “context-sensitive” that adding all the stations, classes, and stages of life to any particular ethical injunction means that “each addition is really a subtraction from any universal law. There is not much left of an absolute or common (sadharana) dharma which the texts speak of, if at all, as a last and not as a first resort.“234

  If Manu is only a collection of particular injunctions, how can we consider it to be a sdstra, that is, a work of “systematic thought” or “science”? Unlike the Dharmasutras, Manu does not just collect particular injunctions; he writes in full awareness of the issue of universality and attempts to defend particularism theoretically, so to speak-to use universal arguments to defend particularism.215 Only rarely do some tensions in his intellectual structure show through. Toward the end of chapter 6, which deals with the third and fourth stages of (the Brahmin’s) life, the Forest Hermit and the Wandering Ascetic, which seem to be closer to the ultimate aim of life than the earlier stages, he reaffirms the superiority of the householder:

  Student, householder, forest hermit, and ascetic: these four distinct orders have their origin in the householder. All of these, when they are undertaken in their proper sequence as spelled out in the sacred texts, lead a Brahmin who acts in the prescribed manner to the highest state. Among all of them, however, according to the dictates of Vedic scripture, the householder is said to be the best, for he supports the other three. As all rivers and rivulets end up in the ocean, so do all the orders end up in the householder. (6.87-90)236

  Yet chapter 6 ends by saying that the brahmin who has “cast off the inherent evil of rites by retiring from all ritual activities … erases his sins and attains the highest state” (6.95-96).237

  Even more remarkable is a passage at the very end of the book that seems to stand alone but raises more than one question in the reader’s mind:

  A man who knows the Vedic treatise is entitled to become the chief of the army, the king, the arbiter of punishment, and the ruler of the whole world. As a fire, when it has picked up strength, burns up even green trees, so a man who knows the Veda burns up his taints resulting from action. A man who knows the true meaning of the Vedic treatise, in whatever order of life he may live, becomes fit for becoming Brahman while still living in the world. (12.100-102)238

  It is clear from the passage immediately following this one and discussions like those at the end of chapter 6 that knowing the Veda involves much more than just hearing it, that it entails ascetic and yogic practices that, according to Manu, the Veda entails. This kind of knowledge and practice characterizes the renouncer ideal of the fourth stage of life and involves a universal experience that transcends ritual and other particularistic obligations. It is for this reason that, in spite of his sense that this stage is the highest good, in
chapter 6 Manu still held the householder stage to be best. But in the passage above, the renouncer ideal is wedded to the notion of rule, even world rule; this has parallels in some Buddhist teachings and is much more fully worked out in the epics. It is remarkable to find it in Manu. But what are we to make of the fact that one who truly knows the Veda “in whatever order of life he lives” can become one with Brahman in this very world? Becoming one with Brahman is equivalent to the Buddhist notion of attaining nirvana in this life. This apparently universal claim must be qualified because sudras and those beneath them can have no knowledge of the Veda, but for a moment in this most particularistic book, the very standard for Hindu particularism for all succeeding time, we seem to have a glimpse of the universal.239

  Olivelle holds that Manu was probably written in the period of confusion and uncertainty after the fall of the Mauryas as an effort to defend Brahmanical privilege, but, more generally, as an effort to reaffirm the Brahmanical understanding of society, while responding to the substantive challenges posed by the ascetic sects, especially Buddhism, and the teaching of Asoka in particular.240 He believes that the Mahabharata (and I would think the Ramayana as well) arose at roughly the same time and place as Manu and with the same concern with the reaffirmation of traditional dharma. However, though both epics are centrally concerned with dharma, both the Mahabharata and the Ramayana move far beyond the simple pieties of Manu in dealing with the deep tension between dharma as righteousness and the mass of particular obligations related to status that dharma means in the Brahmanic tradition. In grappling with this tension, they do not solve it, but they widen the horizon of all subsequent Indic culture, of which they can probably be said to be the formative texts. Each of these epics is immense-the Mahabharata is said to be seven or eight times as long as the Iliad and the Odyssey combinedand enormously complex in both narrative and ethical reflection. Here I can only suggest why they are such formative texts, not only for the Indic imagination but for the human imagination in general.