Vedas and Brāhmaṇas text.
The Vedas constitute a corpus of orally composed religious texts that were produced by the priestly class of the earliest Sanskrit-speaking inhabitants of the South Asian subcontinent between approximately 1600 and 600 BCE. For centuries they were preserved by generation after generation through rote memorization before eventually being written down. The Vedas contain a diverse collection of material, including direct invocations of and appeal to divinities, ritual formulas designed to accompany the performance of rites, expositions of proper ritual procedure and its meaning, and metaphysical speculation. The Vedas also contain legendary and mythological narratives or allusions to such narratives, although material of this kind is generally subordinate to the main aim of presenting, justifying, or explaining ritual practice. Similarly, the speculative thought contained in the Vedas often grows out of and elaborates on ritual exegesis. Therefore it is more accurate to characterize the Vedas as being principally oriented to ritual practice ( yajña ) than to define them as the expression of religious belief per se.
The Vedas proper comprise three or four subdivisions, depending on what taxonomy one is following. Generally, the oldest references speak of three Vedas, whereas later ones sometimes speak of four. By the “Three Vedas” is meant the Ṛgveda, the Sāmaveda, and the Yajurveda. These three collections receive their names from the type of liturgical utterance with which each is primarily concerned, the Ṛgveda being concerned with knowledge regarding recited verses of praise (ṛc), the Sāmaveda with knowledge regarding verses sung to melodies (sāman), and the Yajurveda with knowledge regarding spoken ritual formulas (yajus). References to the “Four Vedas” intend the addition of the Atharvaveda, a heterogeneous collection closely akin to, but more easily discussed separately from, the others. Furthermore, each Veda is associated with the priest responsible for its particular type of liturgical contents: the Ṛgveda with the hotṛ, the Sāmaveda with the udgātṛ, the Yajurveda with the adhvaryu, and (probably secondarily) the Atharvaveda with the brāhmaṇa
According to the traditional scheme, each Veda is subject to a fourfold subdivision into Saṃhitā, Brāhmaṇa, Ᾱraṇyaka, and Upaniṣad. These subdivisions are partially based on the nature of the content they contain and in certain cases roughly correspond to the chronological layers of the texts. Nevertheless, an overly rigid attachment to this tradition scheme often obscures the structure of the corpus and the relation between individual texts. For example, material contained in the Brāhmaṇa of one vedic school may find close parallels in the Saṃhitā of another, and the Ᾱraṇyakas and the oldest Upaniṣads are included as part of the Brāhmaṇas of the schools to which they belong and do not necessarily indicate a chronologically later provenance.
A more analytically useful distinction regarding the constituent portions of the vedic corpus is between
From the point of view of relative chronology, the liturgical texts predate the commentarial texts, which necessarily presuppose them. Using linguistic development as a criteron, one may distinguish five broad chronological layers to the vedic corpus (including the Śrautasūtras), starting with the oldest:
The textual corpus of a given Veda often exists of different versions. The multiple versions of vedic material are associated with the multiple groups of Brahmans responsible for the preservation and transmission of the texts. Called “branches” (śākhās), each one of these groups passed down from teacher to pupil its own versions of the Veda to which it belonged and, in addition, composed its own ancillary literature. Because these groups were the primary social institution responsible for transmitting vedic oral literature, they are often referred to as “schools.” Although there is evidence in some cases for influence of one school upon another, each conceived of itself as a distinct organization identifiable by its particular recension of its Veda and by individual peculiarities, at the level of detail, in the performance of the vedic rituals. Thus, in order to identify precisely a particular text within the vedic corpus according to its traditional division, one must state the Veda under which it was classified (Ṛgveda, Yajurveda, Sāmaveda, Atharvaveda), the type of text (Saṃhitā, Brāhmaṇa, Ᾱraṇyaka, Upaniṣad), and the vedic school to which it belongs.
It will suffice here to give a brief catalogue of this vedic corpus. The extant rgvedic texts consist of a Saṃhitā in the Śākala recension, as well as two Brāhmaṇas, two Ᾱraṇyakas, and two Upaniṣads, one of each of these belonging to the Aitareya school and the other to the Kauṣītaki. The Saṃhitā of the Sāmaveda exists in two recensions, the Jaiminīya and the Kauthuma/Rāṇāyaṇīya. The Jaiminīyas also possess a Brāhmaṇa bearing their name, as well as the Jaiminīyopaniṣadbrāhmaṇa, which contains the Kenopaniṣad. Allied with the Kauthuma school are the Tāṇḍyamahābrāhmaṇa (1–25 is the Pañcaviṁśabrāhmaṇa; 26 is the Ṣaḍviṃśabrāhmaṇa) and the Chāndogyopaniṣad. Of yajurvedic texts there is a broad division between the schools of the Kṛṣṇayajurveda (Black Yajurveda) and the Śuklayajurveda (White Yajurveda). The Śuklayajurveda is represented by the Vājasaneya school in its Mādhyandina and Kāṇvīya recensions of the Vājasaneyisaṃhitā and the Śatapathabrāhmaṇa. The Bṛhadāraṇyakopaniṣad is contained in the Śatapathabrāhmaṇa (ŚBrM. 14.1–3; ŚBrK. 16.3). The Kṛṣṇayajurveda comprises the Taittirīya, Maitrāyaṇī, Kaṭha, and Kaṭha-Kapiṣṭḥala schools. Their texts may be most easily summarized in the following table:
The Atharvaveda traditions in both the Śaunaka and the Paippalāda form had only their respective Saṃhitās until quite late. The Gopathabrāhmaṇa represents a relatively late attempt by the Atharvavedins to develop a textual apparatus similar to that associated with the other Vedas, as do the so-called Atharvaveda Upaniṣads, such as the Praśnopaniṣad and the Māṇḍūkyopaniṣad.
The process according to which the material contained in the Vedas was composed, collected, and organized into a recognized religious “canon” spanned roughly a millennium. It is this epoch that scholars generally refer to as the “vedic period.” Though it is true that the transmission of vedic texts and ideas, the performance of vedic rituals, and the continuity of certain vedic social institutions continued well beyond the vedic period proper and, in some instances, persist even to the present day, it is the formative process of both ritual practice and textual canon that justifies the periodization. Nevertheless, it is difficult to be very precise about the dating of the vedic period. Content and language permit a reasonably accurate understanding of the relative chronology of the component texts of the Vedas (i.e. which came first and which came after), but establishing an absolute chronology linking the texts to specific dates has been much more difficult. Of fixed dates we know only that in the 14th century BCE, the leaders of the Hurrian kingdom of Mitanni made an alliance with the Hittite Empire in the name of deities identical to prominent gods of the vedic pantheon, and this Mitanni military elite spoke a dialect of early Sanskrit closely related to the Vedic language (Mayerhofer, 1966, 1974). At the other end, vedic texts and rituals were well known to Pāṇini in probably the 5th century BCE and to the Pali canon from around the 3rd century BCE (Thieme, 1935; Witzel, 1989, 1997a). A rough and ready estimation of the absolute dates of the vedic texts puts the rgvedic compositions 1600–1200 BCE, the yajurvedic mantra texts 1200–1100 BCE, the expository prose 1100–600 BCE, and the descriptive prose of most of the Sūtras 600–400 BCE.
Regarding the composition and redaction of the vedic texts, we have no outside sources. Judging from certain similarities in the form and content of the Ṛgveda and the Avesta, an Old Iranian liturgical collection, it is likely that the practice of orally collecting and preserving hieratic texts was already a feature of the culture of the Indo-Iranians, but this indicates only that the creators and collectors of the vedic materials did not invent an entirely new genre or initiate a previously unknown redactional procedure; it tells us nothing of the formation of the vedic corpus itself Similarly, it is not until well after the closure of the canon that any external evidence is available to help us understand the development of the Vedas or their place within a broader social, political, and religious context. Thus, any attempt to understand the creation of the vedic canon must rely on the internal evidence of the vedic texts.
In general terms one may say that the Ṛgvedasaṃhitā is a collection of the ritual poetry of specific priestly lineages, containing hieratic verse compositions (Sūktas) usually addressed to one or more deities. The majority of hymns are dedicated to the deified sacrificial fire, Agni, and to the warrior deity, Indra. Most of the liturgical material in the collection was designed for recitation during the performance of a particular religious rite in which the extracted juice of the soma plant was offered to Indra and other deities along with blood sacrifices and additional oblations. Rgvedic compositions are often referred to as “hymns,” though with a risk of mischaracterizing some of them; many do amount to praise and invocation, while others are clearly formulas intended to accompany ritual acts, or even cosmological speculation or magical charm.
The structure of the Ṛgvedasaṃhitā and the principles of its internal organization reveal much about the history and purpose of its redaction. The extant collection consists of ten “books,” or maṇḍalas. Books 2–7 constitute a collection of liturgical verse produced by a distinct priestly lineage and are, for that reason, often referred to as the “Family Books.” Thus, book 2 is credited to the gṛtsamadas, 3 to the viśvāmitras, 4 to the vāmadevas, 5 to the atris, 6 to the bharadvājas, and 7 to the vasiṣṭhas (Oldenberg, 1888a). Ṛgvedasaṃhitā 1 and 8, too, contain liturgies by more than one lineage, and these are grouped together into their own “family collections,” though they are not internally divided. In both books 1 and 8, two separate collections are discernable, and analysis shows both that the first 50 hymns of book 1 belong to a younger collection than the following ones (1.51–191) and that the first collection of book 1 is closely related to the first collection of book 8 (1–49). The second collection contained in book 8 consists of Ṛgvedasaṃhitā 67–103, while Ṛgvedasaṃhitā 8.49–59, the so-called Vālakhilya hymns, were inserted, it would seem, quite late into the extant rgvedic collection (Scheftelowitz, 1906). Book 9 differs fundamentally from 1–8, both because it contains only hymns dedicated to the deified plant, Soma, intended for recitation at a single ritual moment, and because it comprises compositions produced by a wide range of lineages, many of which are represented in other books. Book 10, the largest in the collection, is more amorphous but appears to consist of three separate collections (10.1–60; 10.61–84; 10.85–191), each one of which is subdivided into smaller groups (Bergaigne, 1886; Oldenberg, 1888b; Witzel, 1997c; Gonda, 1975).
The process by which the material contained in the separate maṇḍalas was brought together into a single collection may be inferred, in large measure, by its overall organization. First, it stands to reason that, since the texts themselves affirm that the art of producing liturgical poetry was lineage based, the first collections of liturgical poetry were made by individual lineages: all the material eventually contained in the grand collection of the extant Ṛgvedasaṃhitā would have once been preserved in smaller (oral) collections within the individual lineages. The “editorial” process by which the Ṛgvedasaṃhitā was eventually produced would have begun, then, with the collection and arrangement of the separate lineage collections. The core of the collection around which the rest was added is represented by the Family Books, 2–7. To this core the material now contained in books 1 and 8 was appended at the beginning and the end of the collection, probably in two stages of compilation corresponding to the two “minicollections” these two books contain. For reasons that are not clear, none of the lineage collections includes hymns for the important liturgical moment of purifying the pressed juices of the plant in the context of the soma ritual ; some of these, mostly associated with the lineages represented in books 1, 5, and 8, were brought together and appended as the ninth book. Finally, an assortment of late and thematically divergent material was organized in the massive collection that was added to the rest and became book 10 (Bergaigne, 1887; Oldenberg, 1888b; Witzel, 1997c; Proferes, 2003a).
Regarding the relative chronology of the various parts of the Ṛgveda, it may be safer to speak in terms of a given book having a greater proportion of older material than another book than to declare flatly that one book is older than another. Given this caveat, one may say that the youngest material in the Ṛgveda is contained in book 10 and in parts of book 1 and 8, whereas the oldest material is represented in the Family Books, especially, perhaps, books 2, 4, and 6.
The creation of the Ṛgvedasaṃhitā was an extended process by which the liturgical traditions of many closely related but distinct groups were gradually consolidated. From what can be gleaned from the available information, it was attended by significant changes of orientation to ritual poetry, the rituals in which it was performed, and the institutions that were responsible for it. Throughout the period in which the material collected in the Ṛgvedasaṃhitā was composed, vedic ritual culture placed a premium on the creation of new praise poetry. Though the liturgies of each new ritual performance closely followed a preset pattern and employed a stock of common formulas, their composers often boasted that they were made new. By contrast, after the rgvedic period, new liturgies were no longer produced, and instead older ones were recited from memory. Only a few examples of post-rgvedic verse liturgies exist, notably in the Ṛgvedakhilas and here and there in the Yajurvedasaṃhitās (Scheftelowitz, 1906; Witzel, 1997b; Proferes, 2003a). Thus, one key element in the transition from rgvedic-period ritual culture is the cessation of the production of new liturgical poetry and its replacement by the memorization and recitation of older creations. Furthermore, the amount of material that now had to be mastered was much greater, since priests now memorized and performed the hieratic poetry not only of their own lineage but also from all the lineages included in the massive, multilineage collection, the Ṛgvedasaṃhitā. The need to internalize this enormous amount of material led to the perfection of mnemonic techniques that served to ensure the oral preservation of the text for nearly two millennia with astonishing accuracy.
Another aspect of the transition from the rgvedic period to the later vedic period was the standardization of ritual performance. Throughout the rgvedic period, the different priestly lineages both produced their own religious poetry and followed slightly different customs as regards the detail of their religious rites (Bergaigne, 1889; Gonda, 1979; Proferes, 2003a). Sometime around the completion of the rgvedic collection, however, a more or less standard program for the performance of the individual rites was adopted – the differences that exist in the ritual prescriptions of the later vedic schools were the result of subsequent divergence. At some point during or immediately following the final stages of its compilation, the Ṛgvedasaṃhitā became the exclusive source of verse liturgies for the most important religious rites of the vedic people (the Śrauta rites). Thus, the closure of the Ṛgvedasaṃhitā not only marked the first stage of the process of canonization of the Vedas, but it also coincided with a restructuring of the older ritual system and the establishment of a definitive set of liturgies for it.
Finally, the process leading to the creation of the Ṛgvedasaṃhitā and the standardized ritual must have entailed a fundamental transformation of the institutions in which custodianship of the vedic religious heritage was centered. During the rgvedic period, orthopraxy and the production and preservation of ritual poetry was in the hands of the separate lineage groups, the members of each tracing their patrilinear affiliation back to a common eponymous ancestor, such as Viśvāmitra, Vasiṣṭha, Bharadvāja, and so on, who were revered as great seers ( ṛṣis) and creators of ritual poetry (Oldenberg, 1888a). Many of the lineages discernable in the rgvedic poems were continued in the gotra (lineage) and pravara (a series of ancestors) lists preserved in the Śrautasūtras (Brough, 1946, 1947, 1953), but what changed after the rgvedic period was that the lineages ceased to be the basic social unit responsible for maintaining the vedic religious heritage. It is at around this same time that the classical, fourfold varṇa system was developed, and, although the ability to trace one’s ancestry to one or another of the older lineages is what defined one as a member of the elite Brahman class, this new class consciousness meant that class identity rather than lineage identity became the determining factor in the preservation of the vedic traditions.
Given the primary importance of religious poetry and ritual in defining the identity of the vedic peoples, this process of consolidation must be seen as more than just a move to organize the existing liturgical and ritual traditions. The integration of the diverse rgvedic-period lineage traditions into a monolithic vedic tradition constituted a decisive break with the institutions and practices of the past. Such a move would have necessarily had a social and political impact and could only have been achieved under the impetus of some powerful centralizing force. It is for this reason that scholars have hypothesized that prominent tribal groups of the period, such as the Puru, the Bharata, and, at the end of the period in which the Ṛgvedasaṃhitā took shape, the Kuru, were instrumental in promoting these changes (Witzel, 1997a). Such a move would have been equivalent to unifying the culture and religious practice of the diverse groups over which a politically dominant group held sway and thus would have served to promote the integration of its rule. It is very likely that the creation of the Ṛgvedasaṃhitā and the redefinitions of orthopraxy that went along with it were crucial steps in the gradual transition from tribal organization to the development of state structures across the northern part of the subcontinent at the beginning of the Iron Age.
An early redaction of the Ṛgvedasaṃhitā became the primary source for the verse mantras contained in the Sāmaveda, which are, except for about 76 verses, borrowed from a preexisting collection of the Ṛgvedasaṃhitā (mostly from the tṛca [a strophe consisting of three verses] and pragātha [a kind of stanza] material from Ṛgvedasaṃhitā 8 and 9; Eggeling, 1982; Gonda, 1975). Where discrepancies exist between the rgvedic and the samavedic texts, it is demonstrable on linquistic, metrical, and/or grammatical grounds that the readings of the Ṛgvedasaṃhitā are older and reflect a more accurate transmission (Oldenberg, 1888b; Brune, 1909). In essence, the Sāmaveda was originally a distillation of those rgvedic verses intended to be sung to melodies (Oldenberg, 1884; cf. Bergaigne, 1889, 15). It was made, no doubt, to facilitate the work of those priests responsible for the performance of the sāmans in ritual, the udgātṛ and his assistants, for whom it was not necessary to know the entire Ṛgvedasaṃhitā, which became, in its entirety, the preserve of the hotṛ. Because inculcation of the knowledge of the melodies was a major undertaking in its own right, the samavedic priests produced what amounted to an abbreviated and slightly reorganized version of the Ṛgvedasaṃhitā tailored to their purposes. Thus, in both its Kauthuma and its Jaiminīya recensions, the Saṃhitā of the Sāmaveda is arranged in three main sections. The first, the Pūrvārcika, is a collection of individual stanzas, organized according to deity and meter, each associated with a particular melody (sāman) upon which it was to be sung and thus clearly intended for teaching the singing of the melodies. In contrast, the second section, the Uttarārcika, consists of tristichs arranged in accordance with their liturgical application. The third section, subdivided into four groups (gānas), gives the melodies for the verses in the preceding two sections. In short, the arrangement of the samavedic Saṃhitā demonstrates how certain verses of the Ṛgveda were systematically reorganized into a collection better serving the pedagogical and ritual needs of the udgātṛ and his assistants.
The mantra portions of the Yajurvedasaṃhitā also drew upon the Ṛgvedasaṃhitā for their verse selections. In such cases it is often clear that the verses have been secondarily applied to new ritual circumstance (Proferes, 2003b). No doubt it was the great prestige of the rgvedic text that led the Yajurvedins to incorporate verses from it into their own liturgical parts. The few verse liturgies original to the Yajurveda have linguistic affinities with the latest portions of the tenth book of the Ṛgvedasaṃhitā, suggesting that immediately after the closure of this canonical compilation, some new liturgical verse was still being composed, though probably only for newly devised rites unknown to the older ritual system (Proferes, 2003c). The majority of the mantras in the Yajurvedasaṃhitā, however, are composed in prose and are generally intended to accompany the ritual actions performed by the adhvaryu priest. The mantra portions of the Yajurvedasaṃhitā are arranged according to their sequence in the rite to which they belong; thus, for the adhvaryu priest (and his subordinates), memorization of the mantras of his Veda was equivalent to memorizing his part in the ritual performance. From the way in which the rites were grouped within the yajurvedic Saṃhitās, it appears that the taxonomical structures of post-rgvedic ritual thinking had already been refined: the broad division of the rites into havis (an oblation or burnt offering) and soma (juice from the soma plant offered in libations; see also intoxication) types and the derivation of other rites from the paradigmatic new and full-moon rite (darśapūrṇamāsa), on the one hand, and the agniṣṭoma, on the other, is reflected in the fact that the mantras for these two rites are the first to appear in the mantra collections of all the yajurvedic schools, and when a mantra is repeated later in the collection, only its initial part (pratīka) is recorded (for a comprehensive overview of the vedic rites, see Hillebrandt, 1887)
As in the case of the Ṛgvedasaṃhitā, when the samavedic and yajurvedic mantras were first collected, there would have been a single, original compilation, an Ur-samavedic mantra collection and an Ur-yajurvedic mantra collection. The eventual generation of multiple versions of the mantra texts was the result of differences in the way in which different teachers passed on the mantras, owing largely, one may imagine, to regional differences. It was these differences in vedic recitation that led to the recognition of the separate vedic Schools (śākhās), which based their unique identities, first and foremost, on their recitational and ritual peculiarities (Renou, 1947). Thus, just as in the case of the compilation of the Ṛgvedasaṃhitā, it appears likely that the first collection of samavedic and yajurvedic mantras was the result of a highly organized plan to produce a definitive compendium of mantras related to the most sacred ritual traditions of the period.
The atharvavedic collections represent a more diverse range of concerns than do the collections belonging to the other three Vedas (although one may point to the diversity of book 10 of the Ṛgvedasaṃhitā as a precedent). Unlike the other Vedas, the atharvavedic mantras do not consist primarily of liturgies fashioned for the Śrauta rites. Many are intended to accompany rites promoting health, fecundity, and success in various undertakings or, conversely, to thwart the designs of rivals or cast sickness upon them. Marriage and funerary rites are represented, as are royal rites. Other atharvavedic compositions deal with the mystical import of certain aspects of the sacrifice and related cosmogonic speculations and have affinities with some of the late hymns contained in the tenth book of the Ṛgveda: their content seems appropriate to a period of increasing theological reflection. The Atharvaveda also preserves occasional examples of some of the earliest extant prose.
The two extant recensions of the Atharvavedasaṃhitā, belonging to the Śaunaka and the Paippalāda śākhās, respectively, differ structurally to a far greater degree than do the various school recensions of the Saṃhitās of the Yajurveda. Nevertheless, the redactional principles underlying the differences between them are well enough understood to conclude that, originally, there was a single recension of the Atharvaveda just as in the case of the Yajurveda (Witzel, 1985; Insler, 1998; Griffiths, 2003). As the linguistic form of the atharvedic mantras suggests a chronology for the initial redaction around the same time as that of the verse mantras of the Yajurveda and the Sāmaveda as well as the Yajurveda prose mantras but subsequent to the redaction of the mantras contained in the Ṛgvedasaṃhitā, it is reasonable to assume that the atharvavedic collection was part of the general move to preserve and (re)define the heritage of hieratic poetry, which, as we have seen above, heralded a new phase of vedic ritual institutions.
In the case of the rgvedic, samavedic, and atharvavedic traditions, the Saṃhitā contains exclusively mantra material (with a few exceptions of exegetical prose included in the Atharvaveda). In these traditions, commentary and exegesis is reserved for the Brāhmaṇas (and for the Ᾱraṇyakas and Upaniṣads where they exist). In the case of the Yajurveda, by contrast, the situation is considerably more complicated. In the schools belonging to the Kṛṣṇayajurveda (White Yajurveda), the Saṃhitā text comprises sections containing the mantras for individual rites as well as separate sections containing the commentary. The “Saṃhitā prose” is not easily distinguished from “Brāhmaṇa prose”; indeed, much of the prose commentary contained in the Taittirīyabrāhmaṇa has closely corresponding counterparts in the Saṃhitā collections of the other Kṛṣṇayajurveda schools. At the same time, the schools of the Śuklayajurveda (Black Yajurveda) blend mantra and prose commentary in their Brāhmaṇas, while their Samhitā contains exclusively liturgical mantras. In this case, rather than the Saṃhitā collection being older than the Brāhmaṇa collection, it may even be that the mantras of the Saṃhitā were taken over from Brāhmaṇa in an attempt to make the Vājasaneya corpus resemble more closely those of the other vedic schools (Caland, 1932). Thus it can be seen that the traditional subdivision of the Vedas into Saṃhitā, Brāhmaṇa, Ᾱraṇyaka, and Upaniṣad and the chronological judgments sometimes made on the basis of them are not entirely consistent and can often obscure more than they clarify.
One interesting development that may be noted regarding developments in the genre of vedic exegetical prose has to do with the increasing interest in the personality of the ritualist upon whose authority specific practices or interpretations are made. In the earliest examples of vedic prose, such as the Black yajurvedic texts and the first five books of the Aitareyabrāhmaṇa, no significant attention is given to the religious figures who are responsible for shaping the content of the texts. By contrast, in texts like the Jaiminīyabrāhmaṇa and, especially, the Śatapathabrāhmaṇa, we are often treated to the mise en scène of ritual debates. In addition to offering us a glimpse into the institutional forums in which ritual argumentation was practiced, these texts suggest shared points of orientation with the early Buddhist oral literature, in which accounts of the figure of the great teacher trumping in debate supporters of an opposing point of view figure prominently. However that may be, it is true that the interest taken in recounting stories about teaching as opposed to simply the teaching itself is a tendency that is in marked contrast to the strict exegesis of most vedic prose texts.
While the yajurvedic mantras provide evidence as to the structure of the post-rgvedic rites, the exegetical prose texts offer the first glimpse of the social and political contexts in which they were practiced (Rau, 1957). It is fairly clear that, while the rgvedic-period ritual functioned within, through, and in support of clan and tribal structures (Oberlies, 1998; Proferes, 2007), the later ritual scheme was adapted to a very different social and political context (on the geographical provenance of the individual vedic texts, see Mylius, 1965; Witzel, 1987). One of the most notable differences is that in the rgvedic period the main sponsor of soma rites was invariably a political leader, a chieftain or king, whereas in the developed rite, any member of the twice-born class could sponsor a soma rite.
Over the course of the vedic period, the geographical center of the vedic cultural and political sphere slowly spread eastward. The composers of the rgvedic hymns were concentrated primarily in the Punjab around the tributaries of the Indus River, although some speak of rivers in eastern Afghanistan to the west and the Yamunā and the Gaṅgā to the east. Throughout the period in which the yajurvedic mantras and the early exegetical literature took shape, the area north of Delhi in modern Haryana, known as Kurukṣetra, acquired importance and eventually came to be regarded as the legendary center of vedic orthopraxy; this was the area in which the post-rgvedic Śrauta system took shape under the influence of the Kuru kings somewhere around 1200–1100 BCE. Slightly later, the Kuru-Pañcāla lands to the east in modern Uttar Pradesh became important regions of vedic culture. Later still vedic culture was imported into the areas as far east as eastern Uttar Pradesh and Bihar, where the kings of Kosala and Videha became enthusiastic patrons of the vedic Brahmans, as the figures of the famous king Janaka of Videha and the Brahman Yājñavālkya attest (for details regarding the association among specific texts, regions, and polities, see Witzel, 1987, 1995, 1997c; on the dialectical differences among the vedic texts, see Scharfe, 1996; Witzel, 1989a). The process by which vedic practice (together with the Sanskrit language) spread eastward must be seen, at least in part, as a gradual adoption of a prestige culture by local leaders wishing to assert their importance (for a recent and provocative thesis regarding the indigenous culture of the eastern areas of “greater Magadha,” see Bronkhorst, 2007).
The vedic exegetical prose texts of the Saṃhitās and Brāhmaṇas already contain much information about how the rituals were structured and performed at the time their commentaries were composed. Nevertheless, they do not provide detailed, step-by-step guides to the performance of a kind that would be useful in the transmission of the knowledge of ritual practice from generation to generation. For this purpose the genre of ritual Sūtra was developed. Each of these texts provides instructions for the priests of its respective Veda about how to perform their ritual roles, and by collating them we can today reconstruct in great detail the entire vedic system of Śrauta ritual.
Altogether there are nine extant Śrautasūtras belonging to the Yajurveda (Baudhāyanaśrautasūtra, Vādhūlaśrautasūtra, Bhāradvājaśrautasūtra, Ᾱpastambaśrautasūtra, Satyāṣādhahiranyakeśinśrautasūtra, and Vaikhānasaśrautasūtra following the Taittirīya recension of the Yajurveda; Mānavaśrautasūtra and Vārāhaśrautasūtra following the Maitrāyaṇī recension; and Kātyāyanaśrautasūtra following the Vājasaneyi recension), two belonging to the Ṛgveda (Ᾱśvalāyanaśrautasūtra, related to the Aitareyabrāhmaṇa, and Śāṅkhāyanaśrautasūtra, related to the Kauṣītakibrāhmaṇa), and three belonging to the Sāmaveda (Lāṭyāyanaśrautasūtra and Drāhyāyaṇaśrautasūtra following the Kauthuma recension and Jaiminīyaśrautasūtra following the recension of that name). Finally, the Vaitānaśrautasūtra primarily follows the Śaunaka recension of the Atharvaveda (Gonda, 1977). The earliest extant examples of the Sūtra form are the Baudhāyanaśrautasūtra (Caland, 1903) and the Vādhūlaśrautasūtra. These two texts are composed, for the most part, in a straightforward, descriptive style intended to lay out in as detailed a manner as possible the sequence of ritual actions to be performed by the adhvaryu priest, his underlings, and the sacrificer, as well as the mantras that are to be recited in accompaniment to these actions. However, both the Baudhāyanaśrautasūtra and the Vādhūlaśrautasūtra also incorporate passages in the style of the exegetical prose texts, and on the basis of both style and language may be dated approximately to the same period. Thus, it is misleading to speak too definitively of a “Sūtra period” following chronologically a “Brāhmaṇa period,” as is sometimes done. Rather, the Sūtra form represents a development and elaboration of those portions of the exegetical prose texts describing or prescribing ritual action (vidhi). Over time this form was progressively refined and older formulations abandoned, so that most of the extant examples do, indeed, come from the very end of the vedic period. In fact, in its phrasing and argumentation, the Kātyāyanaśrautasūtra appears to link the older Sūtra tradition with that of the Mīmāṃsakas( other texts in Sūtra form are also important: the Pitṛmedhasūtras are passed down as part of the Śrautasūtras in some schools, independently in others, and as part of the Gṛhyasūtras in still others; the Śulbasūtras were developed to record and pass down the knowledge of practical geometry necessary for constructing the altars; the Gṛhyasūtras deal with domestic rites and rites of passage [ saṃskāras]).
While the process of the codification of ritual science can be traced more or less from the period of exegetical prose onwards, the extant astronomical and astrological treatises were all composed well after the vedic period (Pingree, 1981). Nevertheless, the vedic exegetical prose texts and even certain mantras reveal an interest in the constellations, and it is not difficult to imagine that these were considered important for defining the ritual season (Falk, 1987).
More clear, however, is the importance of the science of language for the vedic tradition. From an early period, the study of phonetics (śikṣā) was important to the oral preservation of the vedic texts, and by the time of the Ᾱraṇyakas and early Upaniṣads, the system was already highly developed (Scharfe, 1977). The codification of the pronunciation of the vedic texts was continued in the Prātiśākhyas. These were manuals developed within the individual śākhās to describe the phonetic and accentuation details peculiar to their respective pronunciations of the Vedic language and providing, in addition, rudimentary morphological analysis. The phonemic description of the Sanskrit language recorded in the Prātiśākhyas would have had more than mere descriptive or theoretical value – specifying the position of the respective organs of the mouth necessary to produce specific sounds would have greatly facilitated the pedagogical challenge of preserving stable pronunciation of the texts over many generations.
Both phonological as well as grammatical insight is reflected in the development of the “word-for-word recitation” of the vedic texts, the Padapāṭhas. The earliest recension of the Ṛgvedasaṃhitā was in the form of a Saṃhitāpada, which is to say a text that preserves saṃdhi (the words joined together according to phonetic rules). At a date sometime between the establishment of the Saṃhitāpada and the life of the great grammarian Pāṇini (Bronkhorst, 1981, 1982, 1991; Cardona, 1997), an analytical version of the text was produced, known as the Padapāṭha and credited to one Śākalya (other versions were known in the past but were lost; see Renou, 1947). By providing certain analytical operations on the text, such as rendering the words as they would appear in isolation, unaffected by the rules of saṃdhi, indicating compounds, and in certain cases marking the distinction between stem and suffix, the Padapāṭha provided a rudimentary grammatical analysis that would have helped in the understanding of the text’s difficult and increasingly archaic language. In addition to the Saṃhitāpada and the Padapāṭha, various other techniques were developed for reciting the vedic texts in order to facilitate their memorization, although positive evidence for them is somewhat later (Scharfe, 2002).
Little more than supposition is possible in regard to the history of the development of the science of grammar (vyākaraṇa) in India before Pāṇini. Nevertheless, it does seem likely that concern with the preservation and understanding of the vedic texts played an important role in the early stages. Time and again the composers of the vedic hymns express their interest in the mysteries of language and speculate on its nature and source. More concretely, the compositional exigencies of oral poetry, which demand that the poet has a repertoire of linguistic forms to suit various metrical conditions, would have, no doubt, led to reflections on the relationship among prefixes, stems, and endings, for example (Proferes, 2000). The desire to understand the relationships between similar-sounding words in order to interpret the increasingly unfamiliar language of the mantras would have provided an incentive for the pursuit of what would become etymology (nirukta); indeed, the oldest extant treatise on Sanskrit derivation is Yaska’s Nirukta, a commentary on the glossary of obscure Vedic terms called the Nighaṇṭu. Throughout the period of exegetical prose, etymologizing was a common hermeneutic technique, and there can be little doubt that, despite its rudimentary and unscientific form, it served as the basis for later advances.
It can be seen that the branches of linguistic study later classified as Vedāṅgas were closely linked to the study of the Veda, even if not always in their fully developed form. Something similar may be said for the study of prosody (chandas). Not surprisingly, the composers of the Ṛgvedasaṃhitā are already well familiar with the meters that they employ; they associate them with different qualities and base certain speculations on them. A codification of the meters of that text can be found already in the Sarvānukramaṇī, a component of the apparatus developed for and passed down along with the Ṛgvedasaṃhitā. Prepared probably in the late vedic period or even after and ascribed to Kātyāyana, the Sarvānukramaṇī records the deity or deities to which each hymn is addressed, the prosodic meter in which it is composed, and the name of its purported composer. Identification of the basic meters was well known to those who provided exegesis of the verse mantras in the vedic prose texts, but the Sarvānukramaṇī codifies this analysis for all the hymns of the Ṛgveda. As such, it may be considered to reflect a systematic concern with the topic of prosody, reflected also in texts like the Nidānasūtra of the Sāmaveda and later developed in the postvedic chandas compendiums.
Thus, although the extant ancillary literature rarely dates back to the vedic period, there is good reason to believe that it represents a development and systematization of fields long felt to be important to the understanding of the vedic texts. In this way much of Indian science, from linguistics and poetics to mathematics and astronomy, can be seen to have grown from vedic roots even if, more often than not, the direct evidence is limited. Furthermore, it serves as a reminder that the full extent of the oral traditions associated with topics related to but not developed within the vedic corpus are no longer extant not because they never existed, but because they were partially integrated within or completely replaced by compositions reflecting later advances.
The earliest extant texts from this school of vedic hermeneutics are the Sūtras attributed to Jaimini, which relate to the proper conduct of ritual, and the Sūtras attributed to Bādarāyana, concerned with the proper understanding of the older Upaniṣads. Each of these Sūtras became the foundational text for two distinct and often antagonistic schools of philosophical thought, the Pūrvamīmāṃsā ("Early Mīmāṃsā") and the Uttaramīmāṃsā ("Later Mīmāṃsā"), respectively. However, it is most likely that the differences between these schools grew out of the different specializations of the subject matter (Parpola, 1981, 1994). Before there were Pūrvamīmāṃsākas concerned only with ritual and Uttaramīmāṃsākas concerned only with the liberation of the self, there were simply Mīmāṃsākas, all of them concerned simply with developing a rationally based interpretive framework in which to properly understand the vedic texts as well as to defend them against the criticisms of those who did not accept as axiomatic their value and import. Debate between ritual experts was already a prominent feature of the exegetical prose texts. In addition to the esoteric, brahmodya-style (question and answer riddles) discussions most famously depicted in the older Upaniṣads, disagreements over more concrete matters of ritual procedure are also presented in the exegetical literature. It was the gradual development of a logic-based method for analyzing and resolving such controversial issues that led eventually to the elaborate hermeneutical system of the Mīmāṃsā, which in turn significantly informed theory of jurisprudence in ancient and medieval India (Lingat, 1973; Menski, 2003). Added incentive was given to the development of the system of Mīmāṃsā by the attacks against ritualism by heterodox groups such as the Buddhists and the resulting need for a persuasive apologetics (Clooney, 1990).
From the point of view of the inheritance of the Vedas in the religious life of later Hinduism, a crucial step was the Mīmāṃsā move to overcome śākhā divisions in order to treat the different recensions of the same Veda as equally valid. It defined the vedic corpus, theoretically at least, as a unified canon presenting no internal discrepancies and therefore defensible and justifiable on rational grounds. It is largely because of the work of the Mīmāṃsāka thinkers that one can speak of “the Veda” despite the plethora of different texts and different schools.
Commentarial interpretation of vedic texts began within the vedic tradition itself. The exegetical prose texts often provide glosses for or explanations of the mantras upon which they are commenting (Gonda, 1988). Ancillary literature such as the Nirukta and the Bṛhaddevatā also reflect efforts of interpretation (Patton, 1996). From about the 7th century, full-fledged commentaries on certain vedic texts are well attested, such as those on the Upaniṣads by Śaṅkara and Skandasvāmin’s rgvedic commentary, and it is certain that these commentators had predecessors whose works have not survived. In the 14th century, there was a resurgence of interest in the study of the vedic ritual texts in the Vijayanagara Empire, and major commentaries on many of the Saṃhitās, Brāhmaṇas, and Ᾱraṇyakas were produced by or under the name of Sāyaṇa.
Nevertheless, affirmation of the authority of the Veda is often considered the sine qua non of Hindu orthodoxy even by those who do not study the Veda or practice its rites, or who are excluded from doing so because of their gender or social station (Renou, 1960; Halbfass, 1991). To deny the legitimacy of the Veda is tantamount, for its upholders, to denying the legitimacy of the cosmic as well as the social order and is grounds for ostracism from the ranks of upright men (e.g. MaSm. 2.10–11). This is because, however each understands the message of the Veda, all of the supporters of the Veda accept its language to be the truest expression – or even the ultimate source – of reality. The Veda (Saṃhitās, Brāhmaṇas, Ᾱraṇyakas, Upaniṣads collectively) is Śruti, divine revelation communicated through language, and as such it is the source of dharma . Thus, the Dharmasūtras and the Dharmaśāstras declare the Veda to be the ultimate source of their own authority (Renou, 1960; Lingat, 1973). Even in cases of nonvedic texts rated as being on par with or even surpassing the Veda in sacredness or authority, the Veda remains the paradigm for and measure of revealed truth. For example, the Mahābhārata can be considered equal to the Vedas in sanctity and even called the “Fifth Veda,” and the Śrīvaiṣṇavas of South India can refer to the Tiruvāymol̠i, composed in the Tamil language, as the “Tamil Veda” (Carman & Narayanan, 1989). The sectarian Ᾱgamas may occasionally criticize the Vedas, but they also present themselves as different expressions of the same truth, or as an access to that truth for those without the privilege of vedic study. Such examples demonstrate that, like it or leave it, the Veda defines scriptural authority within all branches of Hinduism.
If one may not, generally speaking, reject the Veda and remain within the orthodox fold, one may present a great range of views as orthodox by arguing that they are commensurate with or sanctioned by the Veda. Thus it is that many different strategies for confronting the Veda can be found, ranging from interpretive maneuvers designed to properly understand the texts to an emphasis on the sheer power of its sound when recited. For example, for the Mīmāṃsā schools, the Veda is eternal and without any author, either human or divine; its authority is self-sufficient and self-validating, whereas for the Nyāya and Vaiśeṣika thinkers, the Veda is the word of God (Īśvara), and its authority is rooted in the reliability of its author. Again, for the Pūrvamīmāṃsākas, the content of the revelation of the Veda is dharma, that is, ritual duty, and all readings of the Veda must be coordinated around its injunctions to perform rites, other portions of the texts being understood as meaningless apart from their relation to these enjoined actions. By contrast, the various (and often conflicting) schools of the Uttaramīmāṃsā privilege the upanishadic corpus and argue that the ultimate purpose of the Veda is the revelation of the true nature of reality. Philosophers of language like Bhartṛhari understand the vedic word itself to be the ultimate source and cause of the cosmos, society, and the individual thoughts of human beings (Halbfass, 1991), while adherents of the Sāṃkhya school once claimed that the vedic texts prove their view that the differentiation of the world arises from prakṛti (the Sāṃkhya arguments are preserved only in Uttaramīmāṃsā refutations; see Proferes, forthcoming).
Part of the reason why such a wide variety of interpretations are possible is, of course, because support for them could plausibly be found within the vedic texts themselves. What is interesting, though, is that the unity of social and religious identity that existed among those who accepted the Veda was greater than the divisions created by disagreements regarding dogma. Those who accepted the Veda were, first and foremost, embracing a heritage they viewed as tracing its roots back to a direct vision of the very nature of reality, and so quibbling over the details of dogma could not seriously undermine the strength of their mutual bond. However one might interpret its content, acceptance of the Veda amounts to the acceptance of the idea that the blueprint for society and the cosmos has been revealed to humankind and that knowledge of this revealed design constitutes the highest wisdom.
The Vedas proper comprise three or four subdivisions, depending on what taxonomy one is following. Generally, the oldest references speak of three Vedas, whereas later ones sometimes speak of four. By the “Three Vedas” is meant the Ṛgveda, the Sāmaveda, and the Yajurveda. These three collections receive their names from the type of liturgical utterance with which each is primarily concerned, the Ṛgveda being concerned with knowledge regarding recited verses of praise (ṛc), the Sāmaveda with knowledge regarding verses sung to melodies (sāman), and the Yajurveda with knowledge regarding spoken ritual formulas (yajus). References to the “Four Vedas” intend the addition of the Atharvaveda, a heterogeneous collection closely akin to, but more easily discussed separately from, the others. Furthermore, each Veda is associated with the priest responsible for its particular type of liturgical contents: the Ṛgveda with the hotṛ, the Sāmaveda with the udgātṛ, the Yajurveda with the adhvaryu, and (probably secondarily) the Atharvaveda with the brāhmaṇa
According to the traditional scheme, each Veda is subject to a fourfold subdivision into Saṃhitā, Brāhmaṇa, Ᾱraṇyaka, and Upaniṣad. These subdivisions are partially based on the nature of the content they contain and in certain cases roughly correspond to the chronological layers of the texts. Nevertheless, an overly rigid attachment to this tradition scheme often obscures the structure of the corpus and the relation between individual texts. For example, material contained in the Brāhmaṇa of one vedic school may find close parallels in the Saṃhitā of another, and the Ᾱraṇyakas and the oldest Upaniṣads are included as part of the Brāhmaṇas of the schools to which they belong and do not necessarily indicate a chronologically later provenance.
A more analytically useful distinction regarding the constituent portions of the vedic corpus is between
- 1.
- the liturgical material intended for use within the immediate context of ritual performance; and
- 2.
- the commentarial material providing exegesis of the rites and speculations of various sorts regarding them.
From the point of view of relative chronology, the liturgical texts predate the commentarial texts, which necessarily presuppose them. Using linguistic development as a criteron, one may distinguish five broad chronological layers to the vedic corpus (including the Śrautasūtras), starting with the oldest:
- -
- the Ṛgveda, excluding the youngest parts of book 10;
- -
- All other mantra texts, including parts of book 10, the Ṛgvedakhila, the Atharvaveda,the Sāmaveda, and the mantras of the Yajurveda;
- -
- the prose portions of the Yajurvedasaṃhitās;
- -
- the prose of the Brāhmaṇa texts, the oldest Śrautasūtras, and the older sections of Ᾱraṇyakas and Upaniṣads; and
- -
- the majority of the Śrautasūtras.
The textual corpus of a given Veda often exists of different versions. The multiple versions of vedic material are associated with the multiple groups of Brahmans responsible for the preservation and transmission of the texts. Called “branches” (śākhās), each one of these groups passed down from teacher to pupil its own versions of the Veda to which it belonged and, in addition, composed its own ancillary literature. Because these groups were the primary social institution responsible for transmitting vedic oral literature, they are often referred to as “schools.” Although there is evidence in some cases for influence of one school upon another, each conceived of itself as a distinct organization identifiable by its particular recension of its Veda and by individual peculiarities, at the level of detail, in the performance of the vedic rituals. Thus, in order to identify precisely a particular text within the vedic corpus according to its traditional division, one must state the Veda under which it was classified (Ṛgveda, Yajurveda, Sāmaveda, Atharvaveda), the type of text (Saṃhitā, Brāhmaṇa, Ᾱraṇyaka, Upaniṣad), and the vedic school to which it belongs.
It will suffice here to give a brief catalogue of this vedic corpus. The extant rgvedic texts consist of a Saṃhitā in the Śākala recension, as well as two Brāhmaṇas, two Ᾱraṇyakas, and two Upaniṣads, one of each of these belonging to the Aitareya school and the other to the Kauṣītaki. The Saṃhitā of the Sāmaveda exists in two recensions, the Jaiminīya and the Kauthuma/Rāṇāyaṇīya. The Jaiminīyas also possess a Brāhmaṇa bearing their name, as well as the Jaiminīyopaniṣadbrāhmaṇa, which contains the Kenopaniṣad. Allied with the Kauthuma school are the Tāṇḍyamahābrāhmaṇa (1–25 is the Pañcaviṁśabrāhmaṇa; 26 is the Ṣaḍviṃśabrāhmaṇa) and the Chāndogyopaniṣad. Of yajurvedic texts there is a broad division between the schools of the Kṛṣṇayajurveda (Black Yajurveda) and the Śuklayajurveda (White Yajurveda). The Śuklayajurveda is represented by the Vājasaneya school in its Mādhyandina and Kāṇvīya recensions of the Vājasaneyisaṃhitā and the Śatapathabrāhmaṇa. The Bṛhadāraṇyakopaniṣad is contained in the Śatapathabrāhmaṇa (ŚBrM. 14.1–3; ŚBrK. 16.3). The Kṛṣṇayajurveda comprises the Taittirīya, Maitrāyaṇī, Kaṭha, and Kaṭha-Kapiṣṭḥala schools. Their texts may be most easily summarized in the following table:
| Saṃhitā | Brāhmaṇa | Ᾱraṇyaka | Upaniṣad |
| Taittirīya | Taittirīya | Taittirīya | Taittirīya (= TaiᾹ. 7-9) |
| Maitrāyaṇī | — | — | — |
| Kaṭha | Kaṭha (fragments) | Kaṭha | — Kaṭhaśikṣā |
| Kapiṣṭhala | Kapiṣṭhala (fragments) | — | — |
The process according to which the material contained in the Vedas was composed, collected, and organized into a recognized religious “canon” spanned roughly a millennium. It is this epoch that scholars generally refer to as the “vedic period.” Though it is true that the transmission of vedic texts and ideas, the performance of vedic rituals, and the continuity of certain vedic social institutions continued well beyond the vedic period proper and, in some instances, persist even to the present day, it is the formative process of both ritual practice and textual canon that justifies the periodization. Nevertheless, it is difficult to be very precise about the dating of the vedic period. Content and language permit a reasonably accurate understanding of the relative chronology of the component texts of the Vedas (i.e. which came first and which came after), but establishing an absolute chronology linking the texts to specific dates has been much more difficult. Of fixed dates we know only that in the 14th century BCE, the leaders of the Hurrian kingdom of Mitanni made an alliance with the Hittite Empire in the name of deities identical to prominent gods of the vedic pantheon, and this Mitanni military elite spoke a dialect of early Sanskrit closely related to the Vedic language (Mayerhofer, 1966, 1974). At the other end, vedic texts and rituals were well known to Pāṇini in probably the 5th century BCE and to the Pali canon from around the 3rd century BCE (Thieme, 1935; Witzel, 1989, 1997a). A rough and ready estimation of the absolute dates of the vedic texts puts the rgvedic compositions 1600–1200 BCE, the yajurvedic mantra texts 1200–1100 BCE, the expository prose 1100–600 BCE, and the descriptive prose of most of the Sūtras 600–400 BCE.
Regarding the composition and redaction of the vedic texts, we have no outside sources. Judging from certain similarities in the form and content of the Ṛgveda and the Avesta, an Old Iranian liturgical collection, it is likely that the practice of orally collecting and preserving hieratic texts was already a feature of the culture of the Indo-Iranians, but this indicates only that the creators and collectors of the vedic materials did not invent an entirely new genre or initiate a previously unknown redactional procedure; it tells us nothing of the formation of the vedic corpus itself Similarly, it is not until well after the closure of the canon that any external evidence is available to help us understand the development of the Vedas or their place within a broader social, political, and religious context. Thus, any attempt to understand the creation of the vedic canon must rely on the internal evidence of the vedic texts.
The First Stages of Canonization and the Standardization of Ritual Practice
In discussing the chronology of the Vedas (or any other textual canon, for that matter), one must distinguish between the time when textual material was collected, organized, standardized, and fixed for posterity in the form of the traditionally recognized parts of the corpus, and the time when individual portions of that textual material were produced. (On the definitions of “canon,” the social and political factors contributing to canon formation, and the function of inclusion/exclusion in defining canonical traditions, see the contributions in Assmann & Assmann, 1987). Put another way, recitations to the vedic gods were made long before they were brought together in the form of a Ṛgvedasaṃhitā, and the ones collected there were memorized and recited long after the composition of new ones ceased. This implies that there are at least three major phases requiring analysis in order to approach an understanding of a defined textual canon like the Vedas:- -
- the period before the canon is defined, during which the material that will be included within it is being produced;
- -
- the period (sometimes extended) in which the process of canon formation unfolds; and
- -
- the period following the closure of the canon extending for as long as it is still treated as a living tradition by one or more social, religious, or political communities – in the case of the Vedas, this third period extends to the present day.
In general terms one may say that the Ṛgvedasaṃhitā is a collection of the ritual poetry of specific priestly lineages, containing hieratic verse compositions (Sūktas) usually addressed to one or more deities. The majority of hymns are dedicated to the deified sacrificial fire, Agni, and to the warrior deity, Indra. Most of the liturgical material in the collection was designed for recitation during the performance of a particular religious rite in which the extracted juice of the soma plant was offered to Indra and other deities along with blood sacrifices and additional oblations. Rgvedic compositions are often referred to as “hymns,” though with a risk of mischaracterizing some of them; many do amount to praise and invocation, while others are clearly formulas intended to accompany ritual acts, or even cosmological speculation or magical charm.
The structure of the Ṛgvedasaṃhitā and the principles of its internal organization reveal much about the history and purpose of its redaction. The extant collection consists of ten “books,” or maṇḍalas. Books 2–7 constitute a collection of liturgical verse produced by a distinct priestly lineage and are, for that reason, often referred to as the “Family Books.” Thus, book 2 is credited to the gṛtsamadas, 3 to the viśvāmitras, 4 to the vāmadevas, 5 to the atris, 6 to the bharadvājas, and 7 to the vasiṣṭhas (Oldenberg, 1888a). Ṛgvedasaṃhitā 1 and 8, too, contain liturgies by more than one lineage, and these are grouped together into their own “family collections,” though they are not internally divided. In both books 1 and 8, two separate collections are discernable, and analysis shows both that the first 50 hymns of book 1 belong to a younger collection than the following ones (1.51–191) and that the first collection of book 1 is closely related to the first collection of book 8 (1–49). The second collection contained in book 8 consists of Ṛgvedasaṃhitā 67–103, while Ṛgvedasaṃhitā 8.49–59, the so-called Vālakhilya hymns, were inserted, it would seem, quite late into the extant rgvedic collection (Scheftelowitz, 1906). Book 9 differs fundamentally from 1–8, both because it contains only hymns dedicated to the deified plant, Soma, intended for recitation at a single ritual moment, and because it comprises compositions produced by a wide range of lineages, many of which are represented in other books. Book 10, the largest in the collection, is more amorphous but appears to consist of three separate collections (10.1–60; 10.61–84; 10.85–191), each one of which is subdivided into smaller groups (Bergaigne, 1886; Oldenberg, 1888b; Witzel, 1997c; Gonda, 1975).
The process by which the material contained in the separate maṇḍalas was brought together into a single collection may be inferred, in large measure, by its overall organization. First, it stands to reason that, since the texts themselves affirm that the art of producing liturgical poetry was lineage based, the first collections of liturgical poetry were made by individual lineages: all the material eventually contained in the grand collection of the extant Ṛgvedasaṃhitā would have once been preserved in smaller (oral) collections within the individual lineages. The “editorial” process by which the Ṛgvedasaṃhitā was eventually produced would have begun, then, with the collection and arrangement of the separate lineage collections. The core of the collection around which the rest was added is represented by the Family Books, 2–7. To this core the material now contained in books 1 and 8 was appended at the beginning and the end of the collection, probably in two stages of compilation corresponding to the two “minicollections” these two books contain. For reasons that are not clear, none of the lineage collections includes hymns for the important liturgical moment of purifying the pressed juices of the plant in the context of the soma ritual ; some of these, mostly associated with the lineages represented in books 1, 5, and 8, were brought together and appended as the ninth book. Finally, an assortment of late and thematically divergent material was organized in the massive collection that was added to the rest and became book 10 (Bergaigne, 1887; Oldenberg, 1888b; Witzel, 1997c; Proferes, 2003a).
Regarding the relative chronology of the various parts of the Ṛgveda, it may be safer to speak in terms of a given book having a greater proportion of older material than another book than to declare flatly that one book is older than another. Given this caveat, one may say that the youngest material in the Ṛgveda is contained in book 10 and in parts of book 1 and 8, whereas the oldest material is represented in the Family Books, especially, perhaps, books 2, 4, and 6.
The creation of the Ṛgvedasaṃhitā was an extended process by which the liturgical traditions of many closely related but distinct groups were gradually consolidated. From what can be gleaned from the available information, it was attended by significant changes of orientation to ritual poetry, the rituals in which it was performed, and the institutions that were responsible for it. Throughout the period in which the material collected in the Ṛgvedasaṃhitā was composed, vedic ritual culture placed a premium on the creation of new praise poetry. Though the liturgies of each new ritual performance closely followed a preset pattern and employed a stock of common formulas, their composers often boasted that they were made new. By contrast, after the rgvedic period, new liturgies were no longer produced, and instead older ones were recited from memory. Only a few examples of post-rgvedic verse liturgies exist, notably in the Ṛgvedakhilas and here and there in the Yajurvedasaṃhitās (Scheftelowitz, 1906; Witzel, 1997b; Proferes, 2003a). Thus, one key element in the transition from rgvedic-period ritual culture is the cessation of the production of new liturgical poetry and its replacement by the memorization and recitation of older creations. Furthermore, the amount of material that now had to be mastered was much greater, since priests now memorized and performed the hieratic poetry not only of their own lineage but also from all the lineages included in the massive, multilineage collection, the Ṛgvedasaṃhitā. The need to internalize this enormous amount of material led to the perfection of mnemonic techniques that served to ensure the oral preservation of the text for nearly two millennia with astonishing accuracy.
Another aspect of the transition from the rgvedic period to the later vedic period was the standardization of ritual performance. Throughout the rgvedic period, the different priestly lineages both produced their own religious poetry and followed slightly different customs as regards the detail of their religious rites (Bergaigne, 1889; Gonda, 1979; Proferes, 2003a). Sometime around the completion of the rgvedic collection, however, a more or less standard program for the performance of the individual rites was adopted – the differences that exist in the ritual prescriptions of the later vedic schools were the result of subsequent divergence. At some point during or immediately following the final stages of its compilation, the Ṛgvedasaṃhitā became the exclusive source of verse liturgies for the most important religious rites of the vedic people (the Śrauta rites). Thus, the closure of the Ṛgvedasaṃhitā not only marked the first stage of the process of canonization of the Vedas, but it also coincided with a restructuring of the older ritual system and the establishment of a definitive set of liturgies for it.
Finally, the process leading to the creation of the Ṛgvedasaṃhitā and the standardized ritual must have entailed a fundamental transformation of the institutions in which custodianship of the vedic religious heritage was centered. During the rgvedic period, orthopraxy and the production and preservation of ritual poetry was in the hands of the separate lineage groups, the members of each tracing their patrilinear affiliation back to a common eponymous ancestor, such as Viśvāmitra, Vasiṣṭha, Bharadvāja, and so on, who were revered as great seers ( ṛṣis) and creators of ritual poetry (Oldenberg, 1888a). Many of the lineages discernable in the rgvedic poems were continued in the gotra (lineage) and pravara (a series of ancestors) lists preserved in the Śrautasūtras (Brough, 1946, 1947, 1953), but what changed after the rgvedic period was that the lineages ceased to be the basic social unit responsible for maintaining the vedic religious heritage. It is at around this same time that the classical, fourfold varṇa system was developed, and, although the ability to trace one’s ancestry to one or another of the older lineages is what defined one as a member of the elite Brahman class, this new class consciousness meant that class identity rather than lineage identity became the determining factor in the preservation of the vedic traditions.
Given the primary importance of religious poetry and ritual in defining the identity of the vedic peoples, this process of consolidation must be seen as more than just a move to organize the existing liturgical and ritual traditions. The integration of the diverse rgvedic-period lineage traditions into a monolithic vedic tradition constituted a decisive break with the institutions and practices of the past. Such a move would have necessarily had a social and political impact and could only have been achieved under the impetus of some powerful centralizing force. It is for this reason that scholars have hypothesized that prominent tribal groups of the period, such as the Puru, the Bharata, and, at the end of the period in which the Ṛgvedasaṃhitā took shape, the Kuru, were instrumental in promoting these changes (Witzel, 1997a). Such a move would have been equivalent to unifying the culture and religious practice of the diverse groups over which a politically dominant group held sway and thus would have served to promote the integration of its rule. It is very likely that the creation of the Ṛgvedasaṃhitā and the redefinitions of orthopraxy that went along with it were crucial steps in the gradual transition from tribal organization to the development of state structures across the northern part of the subcontinent at the beginning of the Iron Age.
The Compilation of the Mantra Portions of the Yajurveda and the Sāmaveda
If the processes that led to the creation of the Ṛgvedasaṃhitā began gradually and went through several phases, its conclusion was definitive. The hieratic poetry that the collection included was safely preserved forever after, while whatever other material was available at the time was quickly lost and forgotten. (Mantras relating to domestic rites and rites of sorcery that were likely of great antiquity were preserved in the Atharvavedasaṃhitā, though in a form linguistically younger than all but the latest portions of the Ṛgvedasaṃhitā.) Moreover, though the tradition of verse composition continued in limited fashion and its output was collected in the Ṛgvedakhilas, the Atharvavedasaṃhitā, and, to a much lesser degree, the Yajurvedasaṃhitā, it was now the exception rather than the rule.An early redaction of the Ṛgvedasaṃhitā became the primary source for the verse mantras contained in the Sāmaveda, which are, except for about 76 verses, borrowed from a preexisting collection of the Ṛgvedasaṃhitā (mostly from the tṛca [a strophe consisting of three verses] and pragātha [a kind of stanza] material from Ṛgvedasaṃhitā 8 and 9; Eggeling, 1982; Gonda, 1975). Where discrepancies exist between the rgvedic and the samavedic texts, it is demonstrable on linquistic, metrical, and/or grammatical grounds that the readings of the Ṛgvedasaṃhitā are older and reflect a more accurate transmission (Oldenberg, 1888b; Brune, 1909). In essence, the Sāmaveda was originally a distillation of those rgvedic verses intended to be sung to melodies (Oldenberg, 1884; cf. Bergaigne, 1889, 15). It was made, no doubt, to facilitate the work of those priests responsible for the performance of the sāmans in ritual, the udgātṛ and his assistants, for whom it was not necessary to know the entire Ṛgvedasaṃhitā, which became, in its entirety, the preserve of the hotṛ. Because inculcation of the knowledge of the melodies was a major undertaking in its own right, the samavedic priests produced what amounted to an abbreviated and slightly reorganized version of the Ṛgvedasaṃhitā tailored to their purposes. Thus, in both its Kauthuma and its Jaiminīya recensions, the Saṃhitā of the Sāmaveda is arranged in three main sections. The first, the Pūrvārcika, is a collection of individual stanzas, organized according to deity and meter, each associated with a particular melody (sāman) upon which it was to be sung and thus clearly intended for teaching the singing of the melodies. In contrast, the second section, the Uttarārcika, consists of tristichs arranged in accordance with their liturgical application. The third section, subdivided into four groups (gānas), gives the melodies for the verses in the preceding two sections. In short, the arrangement of the samavedic Saṃhitā demonstrates how certain verses of the Ṛgveda were systematically reorganized into a collection better serving the pedagogical and ritual needs of the udgātṛ and his assistants.
The mantra portions of the Yajurvedasaṃhitā also drew upon the Ṛgvedasaṃhitā for their verse selections. In such cases it is often clear that the verses have been secondarily applied to new ritual circumstance (Proferes, 2003b). No doubt it was the great prestige of the rgvedic text that led the Yajurvedins to incorporate verses from it into their own liturgical parts. The few verse liturgies original to the Yajurveda have linguistic affinities with the latest portions of the tenth book of the Ṛgvedasaṃhitā, suggesting that immediately after the closure of this canonical compilation, some new liturgical verse was still being composed, though probably only for newly devised rites unknown to the older ritual system (Proferes, 2003c). The majority of the mantras in the Yajurvedasaṃhitā, however, are composed in prose and are generally intended to accompany the ritual actions performed by the adhvaryu priest. The mantra portions of the Yajurvedasaṃhitā are arranged according to their sequence in the rite to which they belong; thus, for the adhvaryu priest (and his subordinates), memorization of the mantras of his Veda was equivalent to memorizing his part in the ritual performance. From the way in which the rites were grouped within the yajurvedic Saṃhitās, it appears that the taxonomical structures of post-rgvedic ritual thinking had already been refined: the broad division of the rites into havis (an oblation or burnt offering) and soma (juice from the soma plant offered in libations; see also intoxication) types and the derivation of other rites from the paradigmatic new and full-moon rite (darśapūrṇamāsa), on the one hand, and the agniṣṭoma, on the other, is reflected in the fact that the mantras for these two rites are the first to appear in the mantra collections of all the yajurvedic schools, and when a mantra is repeated later in the collection, only its initial part (pratīka) is recorded (for a comprehensive overview of the vedic rites, see Hillebrandt, 1887)
As in the case of the Ṛgvedasaṃhitā, when the samavedic and yajurvedic mantras were first collected, there would have been a single, original compilation, an Ur-samavedic mantra collection and an Ur-yajurvedic mantra collection. The eventual generation of multiple versions of the mantra texts was the result of differences in the way in which different teachers passed on the mantras, owing largely, one may imagine, to regional differences. It was these differences in vedic recitation that led to the recognition of the separate vedic Schools (śākhās), which based their unique identities, first and foremost, on their recitational and ritual peculiarities (Renou, 1947). Thus, just as in the case of the compilation of the Ṛgvedasaṃhitā, it appears likely that the first collection of samavedic and yajurvedic mantras was the result of a highly organized plan to produce a definitive compendium of mantras related to the most sacred ritual traditions of the period.
The Atharvaveda
Like the mantra portions of the Yajurveda and the Sāmaveda, the Atharvavedasaṃhitā also shares many verses and even whole hymns with the Ṛgveda. Here, too, the precedence of the Ṛgvedasaṃhitā is confirmed. Most of the material, however, is unique to the Atharvavedasaṃhitā and belongs linguistically to a historical phase closely following the very latest parts of the Ṛgvedasaṃhitā. It is likely, though not subject to firm proof, that the mantras for domestic rites and sorcery rites reflect much older practice in a linguistically younger form.The atharvavedic collections represent a more diverse range of concerns than do the collections belonging to the other three Vedas (although one may point to the diversity of book 10 of the Ṛgvedasaṃhitā as a precedent). Unlike the other Vedas, the atharvavedic mantras do not consist primarily of liturgies fashioned for the Śrauta rites. Many are intended to accompany rites promoting health, fecundity, and success in various undertakings or, conversely, to thwart the designs of rivals or cast sickness upon them. Marriage and funerary rites are represented, as are royal rites. Other atharvavedic compositions deal with the mystical import of certain aspects of the sacrifice and related cosmogonic speculations and have affinities with some of the late hymns contained in the tenth book of the Ṛgveda: their content seems appropriate to a period of increasing theological reflection. The Atharvaveda also preserves occasional examples of some of the earliest extant prose.
The two extant recensions of the Atharvavedasaṃhitā, belonging to the Śaunaka and the Paippalāda śākhās, respectively, differ structurally to a far greater degree than do the various school recensions of the Saṃhitās of the Yajurveda. Nevertheless, the redactional principles underlying the differences between them are well enough understood to conclude that, originally, there was a single recension of the Atharvaveda just as in the case of the Yajurveda (Witzel, 1985; Insler, 1998; Griffiths, 2003). As the linguistic form of the atharvedic mantras suggests a chronology for the initial redaction around the same time as that of the verse mantras of the Yajurveda and the Sāmaveda as well as the Yajurveda prose mantras but subsequent to the redaction of the mantras contained in the Ṛgvedasaṃhitā, it is reasonable to assume that the atharvavedic collection was part of the general move to preserve and (re)define the heritage of hieratic poetry, which, as we have seen above, heralded a new phase of vedic ritual institutions.
The Exegetical Prose Texts
Soon after the collection of mantras and the standardization of the ritual system, the vedic ritualists began to produce exegetical compositions, and we know that the extant examples are not the earliest (Hoffmann, 1975–1976, 509; one may also compare certain of the occasional prose sections of the Atharvaveda, such as the Vrātya book, AVŚ. 15.). This genre is represented in the commentarial prose sections of the Black yajurvedic Saṃhitās, the Brāhmaṇas, and the Ᾱraṇyakas of all the vedic schools. The composers of these texts had various aims, such as explaining the functional and symbolic import and the etiology of the rites or of a particular component thereof, often by linking the human ritual with a divine narrative, glossing mantras and demonstrating why they are appropriate to the ritual procedures they accompany, lauding or giving reasons for the efficacy of the ritual, and defining proper ritual procedure in cases where the practices of other schools differ in detail. The precise character of the exegesis differs somewhat, depending upon which Veda the text is associated with, as the ritual and recitational concerns and the functions of the priests are slightly different in the case of each Veda. In general these texts follow the structure of the individual rites upon which they provide exegesis and are more suitably approached as a running collection of “footnotes” on the liturgy than as integrated discourse. It can often be a challenge to understand the inner logic of these texts, but in general this may be characterized as an attempt to define the hidden interconnections between the cosmos and the ritual and natural life, and to demonstrate the importance of ritual practice as intermediary between the macrocosmic and microcosmic worlds (Oldenberg, 1919; Gonda, 1965; Witzel, 1979; Wezler, 1996). The exegetical texts present a theology and theory of ritual that is more coherent and systematic than a cursory glance at them might suggest (Lévi, 1898). The Ᾱraṇyakas are to be distinguished from the rest of the exegetical prose genre only because they treat rites of a special nature, which were only to be taught in seclusion in the “forest,” that is, away from the village and from cultivated land. Nevertheless, one also finds in the exegetical prose texts the seeds of the kind of speculative elaboration that characterizes the textual genre that has become known as “Upaniṣad.”In the case of the rgvedic, samavedic, and atharvavedic traditions, the Saṃhitā contains exclusively mantra material (with a few exceptions of exegetical prose included in the Atharvaveda). In these traditions, commentary and exegesis is reserved for the Brāhmaṇas (and for the Ᾱraṇyakas and Upaniṣads where they exist). In the case of the Yajurveda, by contrast, the situation is considerably more complicated. In the schools belonging to the Kṛṣṇayajurveda (White Yajurveda), the Saṃhitā text comprises sections containing the mantras for individual rites as well as separate sections containing the commentary. The “Saṃhitā prose” is not easily distinguished from “Brāhmaṇa prose”; indeed, much of the prose commentary contained in the Taittirīyabrāhmaṇa has closely corresponding counterparts in the Saṃhitā collections of the other Kṛṣṇayajurveda schools. At the same time, the schools of the Śuklayajurveda (Black Yajurveda) blend mantra and prose commentary in their Brāhmaṇas, while their Samhitā contains exclusively liturgical mantras. In this case, rather than the Saṃhitā collection being older than the Brāhmaṇa collection, it may even be that the mantras of the Saṃhitā were taken over from Brāhmaṇa in an attempt to make the Vājasaneya corpus resemble more closely those of the other vedic schools (Caland, 1932). Thus it can be seen that the traditional subdivision of the Vedas into Saṃhitā, Brāhmaṇa, Ᾱraṇyaka, and Upaniṣad and the chronological judgments sometimes made on the basis of them are not entirely consistent and can often obscure more than they clarify.
One interesting development that may be noted regarding developments in the genre of vedic exegetical prose has to do with the increasing interest in the personality of the ritualist upon whose authority specific practices or interpretations are made. In the earliest examples of vedic prose, such as the Black yajurvedic texts and the first five books of the Aitareyabrāhmaṇa, no significant attention is given to the religious figures who are responsible for shaping the content of the texts. By contrast, in texts like the Jaiminīyabrāhmaṇa and, especially, the Śatapathabrāhmaṇa, we are often treated to the mise en scène of ritual debates. In addition to offering us a glimpse into the institutional forums in which ritual argumentation was practiced, these texts suggest shared points of orientation with the early Buddhist oral literature, in which accounts of the figure of the great teacher trumping in debate supporters of an opposing point of view figure prominently. However that may be, it is true that the interest taken in recounting stories about teaching as opposed to simply the teaching itself is a tendency that is in marked contrast to the strict exegesis of most vedic prose texts.
While the yajurvedic mantras provide evidence as to the structure of the post-rgvedic rites, the exegetical prose texts offer the first glimpse of the social and political contexts in which they were practiced (Rau, 1957). It is fairly clear that, while the rgvedic-period ritual functioned within, through, and in support of clan and tribal structures (Oberlies, 1998; Proferes, 2007), the later ritual scheme was adapted to a very different social and political context (on the geographical provenance of the individual vedic texts, see Mylius, 1965; Witzel, 1987). One of the most notable differences is that in the rgvedic period the main sponsor of soma rites was invariably a political leader, a chieftain or king, whereas in the developed rite, any member of the twice-born class could sponsor a soma rite.
Over the course of the vedic period, the geographical center of the vedic cultural and political sphere slowly spread eastward. The composers of the rgvedic hymns were concentrated primarily in the Punjab around the tributaries of the Indus River, although some speak of rivers in eastern Afghanistan to the west and the Yamunā and the Gaṅgā to the east. Throughout the period in which the yajurvedic mantras and the early exegetical literature took shape, the area north of Delhi in modern Haryana, known as Kurukṣetra, acquired importance and eventually came to be regarded as the legendary center of vedic orthopraxy; this was the area in which the post-rgvedic Śrauta system took shape under the influence of the Kuru kings somewhere around 1200–1100 BCE. Slightly later, the Kuru-Pañcāla lands to the east in modern Uttar Pradesh became important regions of vedic culture. Later still vedic culture was imported into the areas as far east as eastern Uttar Pradesh and Bihar, where the kings of Kosala and Videha became enthusiastic patrons of the vedic Brahmans, as the figures of the famous king Janaka of Videha and the Brahman Yājñavālkya attest (for details regarding the association among specific texts, regions, and polities, see Witzel, 1987, 1995, 1997c; on the dialectical differences among the vedic texts, see Scharfe, 1996; Witzel, 1989a). The process by which vedic practice (together with the Sanskrit language) spread eastward must be seen, at least in part, as a gradual adoption of a prestige culture by local leaders wishing to assert their importance (for a recent and provocative thesis regarding the indigenous culture of the eastern areas of “greater Magadha,” see Bronkhorst, 2007).
Ancillary Literature
The vedic ritual was the catalyst for much early Indian intellectual investigation. In the postvedic period, the study of six subjects was classified as subsidiary to and in the service of vedic study. These subjects are the so-called Vedāṅgas, “Limbs of the Vedas,” consisting of the study of ritual (kalpa), astrology/astronomy (jyotiṣa), phonetics (śikṣā), prosody (chandas), etymology (nirukta), and grammar (vyākaraṇa). Although the extant codifications of these subjects, with the exception of the oldest Kalpasūtras, postdate the vedic period and often have little actual relevance to vedic study, there can be little doubt that they represent advancements in fields that were of interest and concern to the vedic ritualists from a much earlier date. To greater and lesser degrees, occupation with the subjects of each of the Vedāṅgas can be traced back to the vedic period.The vedic exegetical prose texts of the Saṃhitās and Brāhmaṇas already contain much information about how the rituals were structured and performed at the time their commentaries were composed. Nevertheless, they do not provide detailed, step-by-step guides to the performance of a kind that would be useful in the transmission of the knowledge of ritual practice from generation to generation. For this purpose the genre of ritual Sūtra was developed. Each of these texts provides instructions for the priests of its respective Veda about how to perform their ritual roles, and by collating them we can today reconstruct in great detail the entire vedic system of Śrauta ritual.
Altogether there are nine extant Śrautasūtras belonging to the Yajurveda (Baudhāyanaśrautasūtra, Vādhūlaśrautasūtra, Bhāradvājaśrautasūtra, Ᾱpastambaśrautasūtra, Satyāṣādhahiranyakeśinśrautasūtra, and Vaikhānasaśrautasūtra following the Taittirīya recension of the Yajurveda; Mānavaśrautasūtra and Vārāhaśrautasūtra following the Maitrāyaṇī recension; and Kātyāyanaśrautasūtra following the Vājasaneyi recension), two belonging to the Ṛgveda (Ᾱśvalāyanaśrautasūtra, related to the Aitareyabrāhmaṇa, and Śāṅkhāyanaśrautasūtra, related to the Kauṣītakibrāhmaṇa), and three belonging to the Sāmaveda (Lāṭyāyanaśrautasūtra and Drāhyāyaṇaśrautasūtra following the Kauthuma recension and Jaiminīyaśrautasūtra following the recension of that name). Finally, the Vaitānaśrautasūtra primarily follows the Śaunaka recension of the Atharvaveda (Gonda, 1977). The earliest extant examples of the Sūtra form are the Baudhāyanaśrautasūtra (Caland, 1903) and the Vādhūlaśrautasūtra. These two texts are composed, for the most part, in a straightforward, descriptive style intended to lay out in as detailed a manner as possible the sequence of ritual actions to be performed by the adhvaryu priest, his underlings, and the sacrificer, as well as the mantras that are to be recited in accompaniment to these actions. However, both the Baudhāyanaśrautasūtra and the Vādhūlaśrautasūtra also incorporate passages in the style of the exegetical prose texts, and on the basis of both style and language may be dated approximately to the same period. Thus, it is misleading to speak too definitively of a “Sūtra period” following chronologically a “Brāhmaṇa period,” as is sometimes done. Rather, the Sūtra form represents a development and elaboration of those portions of the exegetical prose texts describing or prescribing ritual action (vidhi). Over time this form was progressively refined and older formulations abandoned, so that most of the extant examples do, indeed, come from the very end of the vedic period. In fact, in its phrasing and argumentation, the Kātyāyanaśrautasūtra appears to link the older Sūtra tradition with that of the Mīmāṃsakas( other texts in Sūtra form are also important: the Pitṛmedhasūtras are passed down as part of the Śrautasūtras in some schools, independently in others, and as part of the Gṛhyasūtras in still others; the Śulbasūtras were developed to record and pass down the knowledge of practical geometry necessary for constructing the altars; the Gṛhyasūtras deal with domestic rites and rites of passage [ saṃskāras]).
While the process of the codification of ritual science can be traced more or less from the period of exegetical prose onwards, the extant astronomical and astrological treatises were all composed well after the vedic period (Pingree, 1981). Nevertheless, the vedic exegetical prose texts and even certain mantras reveal an interest in the constellations, and it is not difficult to imagine that these were considered important for defining the ritual season (Falk, 1987).
More clear, however, is the importance of the science of language for the vedic tradition. From an early period, the study of phonetics (śikṣā) was important to the oral preservation of the vedic texts, and by the time of the Ᾱraṇyakas and early Upaniṣads, the system was already highly developed (Scharfe, 1977). The codification of the pronunciation of the vedic texts was continued in the Prātiśākhyas. These were manuals developed within the individual śākhās to describe the phonetic and accentuation details peculiar to their respective pronunciations of the Vedic language and providing, in addition, rudimentary morphological analysis. The phonemic description of the Sanskrit language recorded in the Prātiśākhyas would have had more than mere descriptive or theoretical value – specifying the position of the respective organs of the mouth necessary to produce specific sounds would have greatly facilitated the pedagogical challenge of preserving stable pronunciation of the texts over many generations.
Both phonological as well as grammatical insight is reflected in the development of the “word-for-word recitation” of the vedic texts, the Padapāṭhas. The earliest recension of the Ṛgvedasaṃhitā was in the form of a Saṃhitāpada, which is to say a text that preserves saṃdhi (the words joined together according to phonetic rules). At a date sometime between the establishment of the Saṃhitāpada and the life of the great grammarian Pāṇini (Bronkhorst, 1981, 1982, 1991; Cardona, 1997), an analytical version of the text was produced, known as the Padapāṭha and credited to one Śākalya (other versions were known in the past but were lost; see Renou, 1947). By providing certain analytical operations on the text, such as rendering the words as they would appear in isolation, unaffected by the rules of saṃdhi, indicating compounds, and in certain cases marking the distinction between stem and suffix, the Padapāṭha provided a rudimentary grammatical analysis that would have helped in the understanding of the text’s difficult and increasingly archaic language. In addition to the Saṃhitāpada and the Padapāṭha, various other techniques were developed for reciting the vedic texts in order to facilitate their memorization, although positive evidence for them is somewhat later (Scharfe, 2002).
Little more than supposition is possible in regard to the history of the development of the science of grammar (vyākaraṇa) in India before Pāṇini. Nevertheless, it does seem likely that concern with the preservation and understanding of the vedic texts played an important role in the early stages. Time and again the composers of the vedic hymns express their interest in the mysteries of language and speculate on its nature and source. More concretely, the compositional exigencies of oral poetry, which demand that the poet has a repertoire of linguistic forms to suit various metrical conditions, would have, no doubt, led to reflections on the relationship among prefixes, stems, and endings, for example (Proferes, 2000). The desire to understand the relationships between similar-sounding words in order to interpret the increasingly unfamiliar language of the mantras would have provided an incentive for the pursuit of what would become etymology (nirukta); indeed, the oldest extant treatise on Sanskrit derivation is Yaska’s Nirukta, a commentary on the glossary of obscure Vedic terms called the Nighaṇṭu. Throughout the period of exegetical prose, etymologizing was a common hermeneutic technique, and there can be little doubt that, despite its rudimentary and unscientific form, it served as the basis for later advances.
It can be seen that the branches of linguistic study later classified as Vedāṅgas were closely linked to the study of the Veda, even if not always in their fully developed form. Something similar may be said for the study of prosody (chandas). Not surprisingly, the composers of the Ṛgvedasaṃhitā are already well familiar with the meters that they employ; they associate them with different qualities and base certain speculations on them. A codification of the meters of that text can be found already in the Sarvānukramaṇī, a component of the apparatus developed for and passed down along with the Ṛgvedasaṃhitā. Prepared probably in the late vedic period or even after and ascribed to Kātyāyana, the Sarvānukramaṇī records the deity or deities to which each hymn is addressed, the prosodic meter in which it is composed, and the name of its purported composer. Identification of the basic meters was well known to those who provided exegesis of the verse mantras in the vedic prose texts, but the Sarvānukramaṇī codifies this analysis for all the hymns of the Ṛgveda. As such, it may be considered to reflect a systematic concern with the topic of prosody, reflected also in texts like the Nidānasūtra of the Sāmaveda and later developed in the postvedic chandas compendiums.
Thus, although the extant ancillary literature rarely dates back to the vedic period, there is good reason to believe that it represents a development and systematization of fields long felt to be important to the understanding of the vedic texts. In this way much of Indian science, from linguistics and poetics to mathematics and astronomy, can be seen to have grown from vedic roots even if, more often than not, the direct evidence is limited. Furthermore, it serves as a reminder that the full extent of the oral traditions associated with topics related to but not developed within the vedic corpus are no longer extant not because they never existed, but because they were partially integrated within or completely replaced by compositions reflecting later advances.
Hermeneutics and Apologetics
The reception of the Vedas within Brahmanical tradition over the course of its first millennia and a half or more can be viewed as glacial alternations between varying forms of parochialism and ecumenicalism. Initially the diverse lineage collections were brought together in the Ṛgveda, and a single, standardized ritual was formalized and reflected in the yajurvedic and samavedic mantras. Then, on the basis of growing differences in the transmission of the texts and the details of ritual practice, different śākhās, “schools,” of each individual Veda were established, and divergent lines of transmission perpetuated. Finally, at some point following the conclusion of the vedic period, another move toward unity was made. This time there was no attempt to unify divergent schools or to standardize their ritual practices and textual heritages. Instead, an effort was made to assert the unity of all the various lines of vedic tradition by demonstrating their internal coherence on the one hand and their underlying identity on the other. The resultant systematization of vedic exegesis became known as the Mīmāṃsā.The earliest extant texts from this school of vedic hermeneutics are the Sūtras attributed to Jaimini, which relate to the proper conduct of ritual, and the Sūtras attributed to Bādarāyana, concerned with the proper understanding of the older Upaniṣads. Each of these Sūtras became the foundational text for two distinct and often antagonistic schools of philosophical thought, the Pūrvamīmāṃsā ("Early Mīmāṃsā") and the Uttaramīmāṃsā ("Later Mīmāṃsā"), respectively. However, it is most likely that the differences between these schools grew out of the different specializations of the subject matter (Parpola, 1981, 1994). Before there were Pūrvamīmāṃsākas concerned only with ritual and Uttaramīmāṃsākas concerned only with the liberation of the self, there were simply Mīmāṃsākas, all of them concerned simply with developing a rationally based interpretive framework in which to properly understand the vedic texts as well as to defend them against the criticisms of those who did not accept as axiomatic their value and import. Debate between ritual experts was already a prominent feature of the exegetical prose texts. In addition to the esoteric, brahmodya-style (question and answer riddles) discussions most famously depicted in the older Upaniṣads, disagreements over more concrete matters of ritual procedure are also presented in the exegetical literature. It was the gradual development of a logic-based method for analyzing and resolving such controversial issues that led eventually to the elaborate hermeneutical system of the Mīmāṃsā, which in turn significantly informed theory of jurisprudence in ancient and medieval India (Lingat, 1973; Menski, 2003). Added incentive was given to the development of the system of Mīmāṃsā by the attacks against ritualism by heterodox groups such as the Buddhists and the resulting need for a persuasive apologetics (Clooney, 1990).
From the point of view of the inheritance of the Vedas in the religious life of later Hinduism, a crucial step was the Mīmāṃsā move to overcome śākhā divisions in order to treat the different recensions of the same Veda as equally valid. It defined the vedic corpus, theoretically at least, as a unified canon presenting no internal discrepancies and therefore defensible and justifiable on rational grounds. It is largely because of the work of the Mīmāṃsāka thinkers that one can speak of “the Veda” despite the plethora of different texts and different schools.
Commentarial interpretation of vedic texts began within the vedic tradition itself. The exegetical prose texts often provide glosses for or explanations of the mantras upon which they are commenting (Gonda, 1988). Ancillary literature such as the Nirukta and the Bṛhaddevatā also reflect efforts of interpretation (Patton, 1996). From about the 7th century, full-fledged commentaries on certain vedic texts are well attested, such as those on the Upaniṣads by Śaṅkara and Skandasvāmin’s rgvedic commentary, and it is certain that these commentators had predecessors whose works have not survived. In the 14th century, there was a resurgence of interest in the study of the vedic ritual texts in the Vijayanagara Empire, and major commentaries on many of the Saṃhitās, Brāhmaṇas, and Ᾱraṇyakas were produced by or under the name of Sāyaṇa.
The Authority of the Veda
It is obvious from what has been said above that the Veda stood at the center of Brahmanical culture and thought, serving both as the focal point for intellectual discovery and as the defining feature of the social and religious identity of those groups that valued it and secured its unrivalled place as a marker of prestige. Despite the general acceptance of the dogma of the unity of the Veda, however, the various schools of interpretation have long disagreed about the ultimate message of the Veda and even about what the precise source of its message is. The inscription of the properly ordered syllables into the memory by means of daily oral rehearsal was no less an act of ritual practice than the performance of the rituals themselves (Malamoud, 1997), and thus, for those who spent their lives internalizing the texts just as much as for those who participated in the rituals, the Veda was more a set of duties to be carried out than a set of beliefs to be adhered to. What was uncontestable was not dogma or creed, but the importance of accurate memorization and correct ritual performance; there was great scope for freedom in the interpretation of meaning. For this reason, the vedic tradition is often said to be more concerned with orthopraxy, “correct practice,” than with orthodoxy, “correct belief.”Nevertheless, affirmation of the authority of the Veda is often considered the sine qua non of Hindu orthodoxy even by those who do not study the Veda or practice its rites, or who are excluded from doing so because of their gender or social station (Renou, 1960; Halbfass, 1991). To deny the legitimacy of the Veda is tantamount, for its upholders, to denying the legitimacy of the cosmic as well as the social order and is grounds for ostracism from the ranks of upright men (e.g. MaSm. 2.10–11). This is because, however each understands the message of the Veda, all of the supporters of the Veda accept its language to be the truest expression – or even the ultimate source – of reality. The Veda (Saṃhitās, Brāhmaṇas, Ᾱraṇyakas, Upaniṣads collectively) is Śruti, divine revelation communicated through language, and as such it is the source of dharma . Thus, the Dharmasūtras and the Dharmaśāstras declare the Veda to be the ultimate source of their own authority (Renou, 1960; Lingat, 1973). Even in cases of nonvedic texts rated as being on par with or even surpassing the Veda in sacredness or authority, the Veda remains the paradigm for and measure of revealed truth. For example, the Mahābhārata can be considered equal to the Vedas in sanctity and even called the “Fifth Veda,” and the Śrīvaiṣṇavas of South India can refer to the Tiruvāymol̠i, composed in the Tamil language, as the “Tamil Veda” (Carman & Narayanan, 1989). The sectarian Ᾱgamas may occasionally criticize the Vedas, but they also present themselves as different expressions of the same truth, or as an access to that truth for those without the privilege of vedic study. Such examples demonstrate that, like it or leave it, the Veda defines scriptural authority within all branches of Hinduism.
If one may not, generally speaking, reject the Veda and remain within the orthodox fold, one may present a great range of views as orthodox by arguing that they are commensurate with or sanctioned by the Veda. Thus it is that many different strategies for confronting the Veda can be found, ranging from interpretive maneuvers designed to properly understand the texts to an emphasis on the sheer power of its sound when recited. For example, for the Mīmāṃsā schools, the Veda is eternal and without any author, either human or divine; its authority is self-sufficient and self-validating, whereas for the Nyāya and Vaiśeṣika thinkers, the Veda is the word of God (Īśvara), and its authority is rooted in the reliability of its author. Again, for the Pūrvamīmāṃsākas, the content of the revelation of the Veda is dharma, that is, ritual duty, and all readings of the Veda must be coordinated around its injunctions to perform rites, other portions of the texts being understood as meaningless apart from their relation to these enjoined actions. By contrast, the various (and often conflicting) schools of the Uttaramīmāṃsā privilege the upanishadic corpus and argue that the ultimate purpose of the Veda is the revelation of the true nature of reality. Philosophers of language like Bhartṛhari understand the vedic word itself to be the ultimate source and cause of the cosmos, society, and the individual thoughts of human beings (Halbfass, 1991), while adherents of the Sāṃkhya school once claimed that the vedic texts prove their view that the differentiation of the world arises from prakṛti (the Sāṃkhya arguments are preserved only in Uttaramīmāṃsā refutations; see Proferes, forthcoming).
Part of the reason why such a wide variety of interpretations are possible is, of course, because support for them could plausibly be found within the vedic texts themselves. What is interesting, though, is that the unity of social and religious identity that existed among those who accepted the Veda was greater than the divisions created by disagreements regarding dogma. Those who accepted the Veda were, first and foremost, embracing a heritage they viewed as tracing its roots back to a direct vision of the very nature of reality, and so quibbling over the details of dogma could not seriously undermine the strength of their mutual bond. However one might interpret its content, acceptance of the Veda amounts to the acceptance of the idea that the blueprint for society and the cosmos has been revealed to humankind and that knowledge of this revealed design constitutes the highest wisdom.
Comments
Post a Comment