---------- Forwarded message ----------
Date: Fri, 8 Sep 2006 11:18:32 -0400
From: Michael Witzel <witzel@fas.harvard.edu>
To: Indo-Eurasian_research@yahoogroups.com
Subject: Re: [Indo-Eurasia] ** Stratification

Dear ALL,

Here my 2 cents on Vedic tradition (1st part of 3) (NB: I will put off the [largely parallel] Avestan issue until a bit later).

Veda means "(sacred) knowledge" (cf. English "wit"). The Four Vedas are the oldest extent texts of India and contain religious and ritual poetry, ritual formulas, and the explanatory prose that interprets these very texts; additionally, in the late Vedic Upanisads there is some early philosophy.

[[According to post-Vedic and medieval Indian tradition, the Four Vedas are called S'ruti, that is 'something (revealed to and)  heard' by the 'primordial' sages (Rsi). By contrast,  the concept of Smrti, 'something learnt by heart', is restricted to the post-Upanisadic texts, such as the Sutras or Manu's law book,  all of which are believed to have been composed by human beings.]]

Internal evidence indicates that the Vedic texts were orally composed in northern India, at first in the Greater Punjab and later on also in more eastern areas, including northern Bihar, between c.1500 BCE and c.500-400 BCE.

The oldest text, the Rgveda, must have been more or less contemporary with the Mitanni  texts of N. Syria / Iraq (1450-1350 BCE); these mention certain Vedic gods (Varuna, Mitra, Indra, Nasatya), and some have forms of early Sanskrit that slightly predate the Rgveda (mazd[h]aa for the Vedic medhaa, vashana for the Vedic vaahana, etc.).

However, there still is no absolute dating of any Vedic text. Parameters include the first use of  iron (in a post-Rgvedic text, the Atharvaveda, at c.1000 BCE) and the life time of the Buddha (at 500 or perhaps rather 400 BCE) who postdates almost all Vedic texts. However, all Vedic  texts predate the grammatical commentary of Patanjali  (c.150 B.C.) and his predecessor Panini, who quote most of them.

The Vedic texts were orally composed and transmitted, without the use of script, in an unbroken line of transmission from teacher to student that has been formalized early on. This insured an impeccable textual transmission (for exceptions see below); it is something like a tape recording of c.1500-500 BCE.  Not just the actual words, but even the long-lost musical (tonal) pitch accent (as in old Greek or in Japanese) has been preserved up to the present.

DIVISION

According to Indian tradition, the Vedas are divided into four parts (Rg-, Sama-, Yajur- and Atharva-Veda). This division corresponds to that of the material as used in the post-Rgvedic ritual (see below); each Veda again is subdivided into four levels:

(1) the Samhita  of "Mantra" (= verse and prose ritual fomulae) collections",
(2) Brahmana '('theological'/ritual) commentary',
(3) Aranyaka 'wilderness [*not* 'forest'!] texts', and
(4) Upanisad '(secret philosophical) texts (of correlations and equivalences learned) sitting at the feet (of the teacher)'.

However, one has to add the ritual Sutras, which are regarded as belonging to the Smrti but are late Vedic in content and language.

These traditional divisions into four kinds of texts, however, actually represent five historical layers as indicated by the development of the Vedic language used:

(1) Rgvedic,
(2) of the Mantras,
(3) of Yajurveda expository prose,
(4) of the Brahmanas (incl. Aranyakas, Upanisads) and
(5) of the Sutras.

These five layers only partially overlap with the traditional divisions.

See the table at: <http://www.safarmer.com/Indo-Eurasian/VedaUp-Synopsis.pdf> It gives both the divisions of texts and the linguistic levels.

Obviously, these features are important for any stratification of the texts. Other means of leveling include internal evidence: quotations in later texts from older texts, development of ritual, thought, society, etc. etc. Especially useful are the Mantra quotations: Mantras and their variations across texts can be seen as functioning as the 'genes' (M. Fushimi) of Vedic developments (see further, below).

RGVEDA

The oldest Vedic text, the Rgveda (RV), is composed in archaic, highly stylized poetical Sanskrit. It contains verses of praise addressed to the Vedic gods and to some early contemporary chieftains. Most of the hymns, however, were intended to be recited at the yearly Soma ritual, celebrated at the time of New Year.

The RV contains 1028 hymns arranged in or books, actually 10 'circles' (mandala). It is generally agreed that Book 9 is a separate, fairly late collection containing the texts of Saman hymns to be sung during the Soma ritual. And, that book 10 and part of book 1 (1.1-1.50) are even later additions, forming a frame around older (and as such, further stratified) layers.

Internal evidence indicates that most hymns were composed over a span of just five generations, under the Puru and Bharata chieftains, notably the great Bharata king Sudas; they represent the middle Rgveda period, with such prominent poets as Vis'vamitra and the East Iranian(?) immigrant Vasistha. A few older hymns apparently come from other tribes, such as the Anu-Druhyu and Yadu-Turvas'a.

They were composed by members of  various clans of poets (among which 7 major ones, in RV 2-8); the hymns belonged to them and were transmitted by them as 'private property' which often was 'copyrighted' by including the names of the individual poets or clans or by typical refrains; most of the poets belonged to, or were attributed later on, to the Angirasa clans and also the  Kanva.

The names attributed to the authors of Rgvedic hymns seem to be partially correct, when corroborated by naming the poet or indirectly by certain poetic devices; however,  many of the names  recorded in the late Vedic Anukramani  ('list' of poets, deities, meters) are artificially derived from some key words in the hymns; these names often do not correspond to those given for the same stanzas by the Samavedic, Yajurvedic and Atharvavedic traditions.

ORDERING AND COLLECTION OF HYMNS

When the Samhita texts were collected, after the end of the RV period, under the new Kuru dynasty, they were ordered in a particular way: the RV is arranged  according to strict, mostly  numerical principles (Oldenberg 1888 (sic); Engl. transl. 2005).

Its first level of order is that of  author (family/clan), followed by that of  deity and meter.

In detail: inside each family collection the hymns are arranged according to deities. Agni and Indra come first, then other deities, depending on the number of hymns addressed to them (in decreasing order). Inside each deity collection, the longer hymns come first and the shortest last; in case of  equal length, a hymn with a longer meter (more syllables per verse) comes first.

This organization is well reflected in  the core ("family") books of the RV. All hymns that do not follow this order were added after the initial collection, as is also clear by their many late grammatical and other features.

However, the family books of the RV are arranged in *increasing* order, from short books (RV 2) to longer ones (RV 7); this is visible, of course, only when the additional hymns are excised. Thus, if  one knows -- as is still prescribed today before reciting a hymn -- its author, deity and meter, one can pinpoint its location in the RV family books accurately.  This 'numerical' arrangement was perfect for a society without script.

(In the margin: the chauvinistic amateur historian S. Talageri (2000), who is much praised by Hindutva adherents, does not know of these detailed principles of arrangement nor of Oldenberg's book, but has 'reconstructed' his own idiosyncratic 'unshakable' ordering -- of course, with surprising and historically unacceptable results for the early history of India)

RV: CONTENTS

Apart from the predominantly ritual contents of the Rgveda there are a few hymns of highly poetical value and of early philosophical speculation.  All of Rgvedic poetry is very complicated and enigmatic: it is based on the poetical norms of the preceding Indo-Iranian and Indo-European periods, it refers many  fragmentarily known myths, uses many archaic formulas and set phrases, and a vocabulary that was already archaic then, and  its expression in general is very elliptical.

There also are  stanzas  that praise the local chieftains, who where sponsors of Rgvedic ritual. The area of the Greater Panjab was inhabited by some 30 to 50 tribes and clans in whose service the transient RV poets composed ever "new hymns" in praise of the gods and chieftains.

A number of hymns are in dialogue form; these  have hardly been used in later ritual; however, they belong to the most beautiful and poetical pieces of the RV.  The hymns dealing with early philosophical speculation have usually been understood as presenting contemporary developments, but many of the topics, such as that of the primordial giant (Purusa) go back to Indo-European (i.e. the Old Norse Ymir) and even to a preceding Eurasian period (i.e. the Chinese/Miao Pangu). After the end of the Rgveda, this kind of speculative poetry was continued in the AV (S'aunaka AV books 8-12) which still were composed by the brahman poets, now turned  priests; and later on, in the Yajurvedic Gathas and S'lokas, down to the Upanisad period.



Here is the 2nd part of my Veda report:
VEDA (2)

Samaveda (SV)

While the Rgveda contains original compositions,  the Samaveda was extracted, except for 75 verses, from RV books 9 and 8.  These stanzas are sung, mostly during the Soma ritual, in a very elaborate fashion, including much coloratura and the often nonsensical stobhas (such as haa o haa o haa haayi or bha bhi bhu). They are  the earliest preserved music of India.

The SV  is divided into two main sections, the Arcika containing the actual text used, and the Gana containing the melodies themselves. These are designated by the *text* of well known melodies, somewhat in the following fashion: one should sing a certain text according to the melody "God save the Queen", which is also applied to the American song "America it is thee,"  to the imperial hymn of Germany, the royal one of Norway.

Atharvaveda (AV)

In stark contrast to the other Veda texts, the Atharvaveda contains, in its oldest sections (of equal age as the SV and Yajurveda Mantras), magical poems used for healing and  for all sorts of magic, including destructive sorcery (AV 1-7). To these sometimes very  old  texts (but in post-RV language), a large number of speculative hymns (AV 8-12) and other hymns have been added.  The latter  deal with the most important life cycle rituals (AV 13-18);  they are followed by two appendixes (AV 19-20)

The AV,  is ordered, most clearly in its  Paippalada version, in clear opposition to the arrangement of the Rgveda:  it  is starts with a book that is composed entirely of *short* hymns of just 4 stanzas and increases to one that have 19. To this nucleus of sorcery stanzas (PS 1-15), the speculative (PS 16-17) and Grhya (domestic ritual)  type hymns (PS 18) as well as the appendixes (PS 19-20) were added.

Yajurveda (YV)

The Yajurveda, however, mainly contains prose Mantras (yajus)  that are used as offering formulas; they must accompany each individual action in ritual  (yajna) carried out by the Adhvaryu priest who mumbles them as he proceeds:  for example "you are heaven, you are earth", "move through the interspace!"

These prose Mantras have not been recorded in the Rgveda, though the yajus genre is mentioned; the extent YV Mantras are younger in form and grammar than the RV. Originally, they consisted only of simple, though rhythmical  prose; but already in the first collections (Maitrayani-Samhita, Katha-S, Taittiriya-S), verses from the RV have been added in a linguistically later form that is often slightly degraded by perseveration.

Once the YV  Samhitas were collected according to the traditions of the diverse recensions (s'aakhaa), however, the form of the YV  Mantras did no longer change and they were transmitted faithfully *in this form* to this very day.

This non-Rgvedic, degraged form of Rgvedic mantras was fixed early on, in the YV Samhitas. Thus, again, they can serve as new markers in the 'haplogroups' of the changing 'genetic' code of  Vedic texts... The lineages are clear as in any stemma, cladistic arrangement. In other words, clear stratification is easily established.

To these Mantras, large sections of brahmana style expository prose have been added during the subsequent YV prose period (see below). Both of them combined constitute the texts of the Black Yajurveda; however, in the White YV the explanatory prose (S'B) is separated from the Mantras (Vajasaneyi-S.); this Samhita was only secondarily extracted from the late Vedic S'B.

Arrangement:

The YV Mantras have not been arranged numerically as in the RV, SV and AV  but in the exact order they are used in S'rauta ritual: they form small, individual Mantra collections meant for  each ritual.

However, the order of these individual  Mantra collections inside the two dozen extent YV Samhitas followed a fixed order already by the time the first YV Samhita collection was made; this order is maintained, with minor variations, throughout the post-YV Samhita texts down to the Sutras.

The YV  starts with two small collections, that of the vegetarian New and Full Moon offerings (haviryajna) and of that of the all important Soma ritual, both of which form the paradigm (prakrti)  of (most) other S'rauta rituals; even the animal sacrifices (pas'ubandha) are technically considered as haviryajnas.

BRAHMANAS

All aspects of the solemn S'rauta ritual  have been discussed at length in the so-called brahmana texts.  The oldest texts, in a stark expository style, are contained in the YV Samhitas of the Black Yajurveda. The linguistically younger brahmana style texts are independent texts, the Brahmanas proper, which are attached to each of the four Vedas-Saahitas (see table)

The Brahmana style text stress correct knowledge ('he who knows thus', ya evam veda) of the hidden meanings of the ritual and the correlations (homologies) on which it is based. This so-called  'identification' technique  correlates  certain items in the three spheres of  microcosm (humans, society), mesocosm (yajna, i.e. ritual), and macrocosm (gods, universe (cf. Farmer et al. BFMEA 2000). This procedure led to a complex, amorphous web of 'hidden' cosmic and mundane interrelations that was known only to the ritual specialists who used it to obtain certain desired effects; the system is  the predecessors of Upanisadic correlations (brahman = atman etc.) .

The Brahmana style prose texts  thus are  the oldest explanations, in fact native commentaries,  of the  literal meaning of the Mantras and of ritual actions.

ARANYAKAS

Aranyaka (Ar.) should have been translated, for nearly a century (Oldenberg 1915), as 'wilderness (texts)',  not as frequently still met with, as 'forest texts'. These texts are not texts meant for ascetics but are regular brahmana style texts which discuss the more secret and dangerous rituals.  Therefore, they have been prescribed to be learned and recited outside the settlement "from where one cannot see the roofs of the settlement".

Because of their special position as additional texts, the Ar. have become an open category where one could add all sorts of later Vedic texts, such as many Upanisads and even one early Sutra. Many extraneous items have also been added to the nucleus of dangerous S'rauta rituals, including  even post-Vedic Upanisads.

(The frequently  maintained connection of the Ar. texts with the post-Vedic life stage of the vanaprastha is only a  medieval fiction).



Here is Veda part  (3)

UPANISADS

The Upanisads (Up.) contain the secret teaching, by a variety of late Vedic teachers, of early philosophical speculation about  the nature of the world and of humans  and their fate after death, as well as the earliest discussion of the workings of rebirth and karma. Various small heterogeneous  sections have subsequently been added, such as some last admonitions of the teacher to his 'graduating', departing students and even a detailed hands-on recipe of how to procreate a 'yellow-eyed son'.

As they represent the (alleged) collected dicta of various teachers they are multilayered.

The texts were traditionally often called Rahasya 'secret', as they were supposed to be learned only by specially selected students, which explains their often less well preserved  state of transmission. Tradition, indeed, sees the Up.s as the 'end of the Veda' (vedanta), that is at the end of the four assumed 'historical' levels of the Samhitas, Br.s, Ar.s, and Up.s, while in fact, the late Vedic Sutras (see below) still are an integral part of the Vedic canon.

It is from the background of the Brahmana style texts that the thinking of the early Upanisads emerges (Brhadaranyaka-Up., Chandogya-Up., Jaiminiya Upanisadbrahmana, etc., see table). It involves a thorough rethinking of the existing correlative premises, in part influenced by late Vedic social conditions of the eastern territories of North India (Kosala, Videha).

There,  a thorough re-organization of the brahmana style texts (in S'atapatha Br.) took place, including a rethinking of many of the
earlier "theological" positions. Further, the increasing Sanskritization of the area along western (Kuru) models brought about the formation of canonical texts, a general ordering of S'rauta procedure, and new deliberations of its inherent meaning.

While the Up.s are often treated as the beginning of philosophical tradition in India (or as a precursor to early Buddhist and Jain thought) they are in fact the almost inevitable outcome of the intellectual development of the Brahmana period, when such questioning was prominent both inside and between the Vedic schools (s'aakhaa).

However, it was expressed differently, not in Upanisadic dialogue form, but by statements such as  "some say..." or by the frequent quotations of divergent views in the brahmana type texts, especially in S'B where various "solutions" to a problem are habitually discussed and still presented as authoritative, positive statements of truths. The Up.s, however, contain discussions in the form of real dialogues, involving severe questioning and reluctant admission of innocence [=ignorance] or boastful claims of knowledge.

Several factors come together and lead to a qualitative breakthrough, which results in the new karmic rebirth idea and, based on increasing use of higher levels of correlations, in the assertion of the identity of the human soul (atman) with that of brahman (neuter) in such famous sentences as tat tvam asi (Chandogya Up. 6.10.3).

Later Vedic thought quite naturally led to this stage, and to a whole range of more or less contemporary and quite diverse points of view, as discussed in the Pali canon (Dighanikaya 2.)

It is important to note that  these developments took place precisely at this moment, c. 500 CE, and in eastern Northern India (Kosala, E. Uttar Pradesh, and Videha, N. Bihar).

The Kosala-Videha area was one of great mixture of peoples due to various movements of tribes and individuals, and consequently also of ideas. It also was a part of the spread-zone of the western, Kuru type Vedic orthopraxy. Some late- or post-Vedic immigrants such as the Malla, or especially the S'akya, (originally maybe an Iranian tribe who may have transmitted (para-)Zoroastrian influence). Further, there was admixture of local Munda peoples, of older, eastern Indo-Aryan settlers, and of contemporaneous immigrants including many western Brahmins.

A comparison of the late Vedic and early Buddhist texts indicates admixture of the older, para-Vedic Indo-Aryan religion of the East with the orthodoxy and orthopraxy of the 'missionary' Kuru-Pancala Brahmins of the West, who were invited by such kings as Mahakosala and (Maha)-Janaka of the the emerging large kingdoms of Kosala and Magadha.

Finally, there was the social ferment created by the contemporaneously emerging cities (of the so-called second urbanization, after the Indus civilization). The Vedic texts hardly, if ever, speak about towns; yet, by the time of the Pali texts, cities are fully established, with rich merchants carrying out a long distance overland trade (of the luxury article, Northern Black Polished Ware), and brahmins living in the formerly off-limits lands of Magadha and Anga.

The so-called Middle Up.s (Is'a, Katha, Kena, Prana, Mandukya, S'vetas'vatara, Mahanarayana, etc.) are no longer composed in prose and dialogue form but in verse, and are heavily influenced by the post-Vedic (Epic) language. Many of them show a tendency towards the sectarian worship of a particular deity.

The Sannyasa Up.s, composed around 300 BCE, discuss the newly introduced life stage of the renouncer (sannyasin).

The Bhagavadgita of the Mahabharata (book 6) is sometimes regarded as an Up. as well. Sectarian Up.s (in Epic/Classical Sanskrit) have been composed well into the Middle Ages. In the interpretation of the Upanisads the c. 8th century monistic philosopher S'ankara has played an important but generally overrated role. We still need  a detailed philological edition and discussion of the important older Upanisads (see table).

======

SUTRA

The Vedic canon concludes with the late Vedic  Sutras ("thread, guideline", or Kalpasutra "ritual guidelines'') which form the true end of the Vedic period and its texts, though the classical/medieval tradition assigns them to a separate category, the Smrti texts.  Indeed, the older ones among them (Baudhayana Srautasutra, Vadhula S'S, etc.) are still composed in late Br. language. The Sutras are descriptive and prescriptive  texts that deal systematically, in the proper order of ritual procedure, with the solemn ritual (S"rauta Sutra), with the domestic rituals  (Grhya Sutra), and with the rules of proper behavior as a Veda student or as married householder (Dharma Sutra).  (There are various later additions to all Vedic texts, Paris'ista).

S'RAUTA SUTRA

The older Sutras (especially BS'S, VadhS'S)  explain the complicated ritual step-by-step, more or less in the order of the various  Mantra collections in the Yajurveda, (another useful indication of stratification).  They do so at great length, in clear prose and by
quoting the Mantras in extenso. Even if a ritual that is described later in the text is built out of ritual blocks described earlier, these older Sutras, for the most part, still describe such complex rituals in extenso. Later Sutras make increasing use of the referring technique which points back to earlier parts of the text by quotation ('as said earlier') and of using just the initial words of a Mantra
(pratika). The later texts use shorter and shorter (nominal) clauses, a technique  seen in its apogee in Panini's  grammatical Sutras, the Astadhyayi.

The contents of the S'rautasutras follow, by and large, the scheme first set out in the Mantra collection of the Yajurveda Samhitas, and the individual rules those of the Brahmana style texts.

GRHYA SUTRA

The Grhya Sutras (GS) often form part of, or actually are, an appendix to the extent S'rauta Sutras, and some of them actually refer back to ritual details described earlier in the same text or even in the S'rauta Sutra. Their contents, however, often are very old. Some of the rites of passage (samskara), such as burial and marriage, have been described already in the RV and AV, and some of the details may in fact go back even to the Indo-European period, for example the offering of three meat balls in the anniversary rituals (s'raddha) for one's three immediate ancestors, or the cult of the fire, or the marriage ceremony; other items, such as the initiation of the student by a girdle, seem to be of  Indo-Iranian  age.

By and large, the GS deal with the rites of passage form birth to death, or rather, from one's conception to one's dissolution in the
vague group of ancestors. The GS thus are a cyclical set of rituals variously arranged as starting with  marriage, with initiation to Veda study (upanayana), or  even with pregnancy.

Just as the AV Samhitas, the 'Grhya Sutra' of the AV (Kaus'ika Sutra) contains many facets of early Indian life that would otherwise escape us. It provides many usages for the AV Samhita spells; it is a virtual handbook of customs and beliefs, of  common white and black sorcery, of healing procedures, of omina and portenta. Many such details can be followed up later on in the AV Paris'ista and in the medieval books on dreams.

DHARMA SUTRA

These Sutras deal with dharma 'proper behavior',  beginning with that of a Veda students, and moving to that of a married man (grhastha), his daily and seasonal ritual duties, family life, to the death rituals and ancestor worship and  inheritance; some also include the duties of a king and his jurisprudence, the four stages in life, and long sections on atonements for wrong behavior. These rules have provided the basis for medieval and modern Anglo-Indian Hindu law.

Again, all these texts quote from the respectively earlier layers of the Vedic corpus, directly or anonymously, and the Mantras used in such quotations provide clear indications from which kind of Vedic text they come, and often, from which level of these texts.

Finally, there are a number of appendixes to the Vedic texts, of various ages, such as the Rgvedakhila, or the AV-Paris'ista, some of which are composed already in the style of the Epic and Puranas, (there even is an Allah Upanisad). Manu's 'lawbook', the Manu Smrti, is one such early post-Vedic text that sums up much of what is said in the Grhya- and Dharmasutras.  In this sense, the Vedic canon has been open, and some nowadays even want to "rewrite" the Manu Smriti, so as to 'update' it.

________________________________________________
Michael Witzel
Department of Sanskrit and Indian Studies, Harvard University
<http://www.fas.harvard.edu/~witzel/mwpage.htm><
<http://groups.yahoo.com/group/compmyth>
<http://groups.yahoo.com/group/Indo-Eurasian_research/>
< http://users.primushost.com/~india/ejvs/>