Introduction 2: Development of the older Samavedic texts

[Full title: On the interrelation and the historic development of the older Samavedic texts]

The Samhita of the Samaveda, then, comprises: 1. Four song-books (gramegeya- and aranyegeyagana, uha- and uhyagana). 2. Three collections of verses: the text-books to these songbooks; the verses agree on the whole with the verses of the Rksamhita. The songs, chaunts, samans of the ganas are based or composed on the verses. From a verse (a rk) a saman is made by musical notation, by certain changes as stretching of vowels, and repetition of syllables, and by inserting different sounds and syllables, sometimes whole sentences or verses. These insertions are called stobhas. The relation between the purvarcika (and aranyakasamhita) and uttararcika on the one side and the first two ganas (grameand aranyegeyas) on the other side, is clear with each verse of the

purvarcik a corresponds a single chant, a single saman; each " " of these samans have a name, mostly derived from the name of that Rsi who is said to have " seen the saman and these samans are collected in the first two song-books. The arrangement of the yonis, i.e., the verses on which the samans are composed, in the purvarcika, is systematic: first come the verses addressed to Agni, then, those addressed to Indra, and then, those addressed to Soma pavamana. Each group is subdivided according to their metres The uttararcika on the other hand, rests on a wholly different principle and has a different aim. In this part of the Samhita single verses are not recorded, but its bulk consists of tristichs or pragathas (i.e. two verses: 1. either a brhati or a kakubh; 2. a satobrhati; in the praxis out of these two verses are made three, so that practically such a pragatha is equally a tristich.) We find likewise complexes of 4, 6, 7, 9 and 10 verses. The aim of this collection is purely practical: it contains the verses on which at the occasion of the several kinds of great sacrifices the lauds are chanted on the whole, each first verse of such a tristich occurs in the purvarcika, this means that the melody (the saman) which belongs to this verse in the purvarcika, is to be applied on the whole tristich of the uttararcika. One example will make this clear. The first verse of the pragatha in the uttaraarcika 1. 1. 13 (=vs. 35 of the continuous numeration), beginning tam vo dasmam rtisaham, is found in the purvarcika: 3. 1. 5. 4. (=vs. 236 of the continuous numeration). To this verse belong, according to the gramegeyagana, five melodies the last of which is the naudhasa (gram. VI. 1. 37, see Calcutta edition, Vol. I, page 487): On this melody, now, the two verses of the uttararcika 35 and 36 (out of which by repetition three verses are made) must be chanted in the praxis of the Soma-sacrifice: as third prsthalaud of the ordinary Agnistoma; these three verses as used in the praxis are given in the uhagana I. 1. 6 (see Calcutta edition, Vol. III, page 93). But it is a fact that there is a great number of verses in the uttararcika that have no correspondent verse in the purvarcika e.g., the first 18 verses of the uttararcika and many more. This is explained by the fact that these verses, which all belong to the morning service (the pratahsavana), are chanted on the gayatra-melody which is composed on the well known verse addressed to Savity: tat savitur varenyam, etc. This melody has been recorded in the devatadhyayabrahmana; it is given as an appendix to the editions of the

gramegeyagana (Ed. Calcutta, Vol. V, page 601). On the other hand, many yonis (and samans composed on these) are found in the purvarcika which have no correspondent verse (and chant) in the uttararcika. This is explained by the fact, that samans are chanted not only at the great sacrifices of soma, but likewise on many other occasions at the establishing of the sacred fires (the agnyadheya); at the foundation of the high altar of bricks (the agnicayana); at the pravargya ceremony and on other occasions. These samans were chanted as solo (parisaman), mostly by the Prastotr (Laty. I. 5. 8 sqq., I. 6. 1 sqq., etc.). About the hagana and the hyagana the following remarks may be made. Each of these ganas is divided in 7 parvans: 1. dasaratra, 2. samvatsara, 3 ekaha, 4. ahina, 5. sattra, 6. prayascitta, 7. ksudra. As to their aim, these two ganas precisely correspond with the uttararcika, they serve the praxis of the soma-sacrifices: the ubagana runs parallel with the uttararcika and the grainegeya, the hyagana with the aranyakasamhita and the aranyegeyagana The names themselves are clear: uhati means "to adapt "; the uhagana contains the melodies of the gramegeya but adapting them and working them out so as they are to be chanted in the praxis; the uhyagana (which word is abbreviated from uharahasyagana, rahasya being another name for aranyaka), adapting the melodies of the aranyegeyagana to the verses on which they must be chanted in the praxis. It can be proved with certainty that these two ganas belong to the later strata of the Samavedic literature: that they are later than the grame- and aranyegeyaganas, later than the Pancavimsabrahmana, later than the Arseyaand Ksudrakalpa, later than the sutra of Latyayana-Drahyayana, later even than the Puspasutra. That they are later than and, even in their sequence, based on the Arseyakalpa and the Ksudrasutra, can be proved. The Arseyakalpa describes the gavamayana, the ekalas, the ahinas and the sattras; all the samans to be used on these occasions occur in the 2 nd-5 th parvans of the uha- and uhyaganas in precisely the same sequence as in the Arseyakalpa. For the chants of the tenday-rite, which had been treated fully in the Brahmana. the uha- and uhyaganas lean on this text. The Ksudrasutra describes the prayascitta and kimya rites: with these run parallel the last two parvans of the uha and the uhya. That both texts, Arseyakalpa and Ksudrasutra, are prior to the uha- and uhyaganas, may be proved by one

example from each. In the Arseyakalpa IX.: 9. b. we read ausunasyarksu vaisvajyotisam antyam yat pra gayateti: "the closing chant is the vaisvajyotisa, chanted on the verses of the ausana, as indicated by the words pra gayata". The verses of the ausana are SV. I. 523=II.27 (28) pra tu drava etc.; on these must be chanted the vaisvajyotisa belonging to SV I. 534 (pra gayata). But, as there are likewise vai- svajyotisa-melodies on other verses, the addition was necessary. Now, the uhagana XIX. 1.7 (Samaveda, ed. Calcutta, Vol. III, page 82) gives the vaisvajyotisa on these verses pra tu drava. If the author of the Arseyakalpa had been acquainted with the uhagana, his indication : "(the vaisvajyotisa) as indicated by the words pra gayata" would have been wholly superfluous. To prove that the Ksudrasutra also is earlier than the uhagana, the following example may suffice. Ksudrasutra I. 6 (No. 30) we read: annadyakamasyendra yahi dhiyesita iti madhyamam trcasya prabhrtim krtva tasu kaleyam, i.e., "for one who is desirous of obtaining food, he should take the middle verse of the tristich beginning: indra yahi dhiyesitah as first and chant on these (so arranged) verses the kaleya." The tristich here mentioned is SV. II. 496-498, its middle verse is the one intended; on these verses, but arranged 497, 496, 498, he should chant the kaleya (gramegeyagana VI. 2. 7, ed. Calc. Vol. I, page 491, composed on SV. I. 237). Now, we find in the uhagana XXI. 2. 4 (ed. Calc. Vol. V, page 638) the chant precisely as the author of the Ksudrasutra had prescribed it. Had the author of the sutra been acquainted with the uhagana, his indication "taking the middle verse at the beginning" would have been wholly superfluous. To prove that neither was the Sutrakara acquainted with the uuhaand hyaganas, the following proof may be adduced. In the Pancavimsabrahmana it is prescribed at V. 4. 5-6 that the chanters at the mahavrata-rite must chant the saman called vasisthasya nihava over the catvala, and that this saman must be chanted on a verse addressed to the "All-gods". As there are many verses addressed to this deity in the collection of the Samaveda, it is uncertain which verse is to be taken. The Sutrakara (Laty. III. 9. 12, Drahy. X. I. 11) makes the following precision, visve deva iti vasisthasya nihavam uhet, i.e.: "he should adapt the vasisthasya nihava to the verse SV. ar. samh. III. 9"; the saman, composed on SV. I. 313: asavi devam, is given by gramegeyagana VIII. 1. 36 (ed. Calc. Vol. I, page 629); the melody on the verse addressed to the "All-gods" is found in the uhagana IX. 3. 12 (ed.

Calc. Vol. V, page 602). It is not possible that the Sutrakara would have expressed himself as he does, if at his time the uhagana existed. Dhanvin, the commentator of Drahyayana, remarks on this last men- " tioned passage: 66 Is not the expression: 'he should adapt' meaningless, as its (i.e. the saman's) reading is known (viz. from the uhagana)? That is true, but they say :-"the uha (i.e. the uhagana) uha is made after the time of this expression (of the Sutrakara's) " ¹.-We have now to prove our assertion that even the Puspasutra is older than uha- and uhyaganas, an assertion not accepted by all scholars². The aim of this text, in my opinion, is to give the rules for adapting the samans, as they are handed down in the grame- and aranyegeyaganas, for practical use at the sacrifices. This is clearly proved by the last words of the part, which originally ended the text (VIII. 234): evam sarvesam esa vikaravidhir, etena pradesenohyah samaganah kalpayitavyah. Professor Simon, the learned editor and translator of the sutra, renders these words as follows:-"Hiermit sind nun die Veranderungen in allen (samans) ordnungsgemasz angefuhrt. Nach dieser Darlegung ist es moglich, die Gesammtheit der zum uhagana gehorenden Samans in der Praxis zu gebrauchen." According to my opinion, the last sentence should be translated: < by means of this indication (this set of rules) the group of samans (as they are handed down in the grame- and aranyegeyaganas) must be adapted (and) made ready (for practical purposes)." It is highly probable that amongst the Samavedic Brahmins in early times certain rules were established and handed down by oral tradition for the adaptation (the uha) of the samans in the grame- and aranyegeyaganas, that these rules were at last collected and arranged in a book (our Puspasutra), and that afterwards for convenience's sake these rules were all brought into action; in this manner the two last books of songs, the two last ganas, arose. So, when we read in the Nidanasutra (II. 1): rsikrtah svid uha 3 anrsikrta iti vai khalv ahuh......... rsikrta ity aparam. and in the Jaimininyayamala vistara (IX. 2. 1-2): uhagrantho 'pauruseyah pauruseyo 'tha vagrimah | vedasamasamana-tvad vidhisarthatvato 'ntimah || 1 The text runs: nanu ca uhed iti vacanam anarthakam, tasyadhyanasiddhatvat, satyam etad, vacanottarakalam uhah krta ity ahuh. 2 Regarding this controversy, see the author's criticism of Simon's Puspasutra in "Deutsche Literaturzeitung" 1909, No 30; Simon's paper in the Journal of the German Or. Soc. vol. 63, page 731, and the author's paper ib. vol. 64, page 347.

the answer must be even as said in the last part of this sloka: " The uhagana is made by the Rsi's, so far as it contains the melodies" seen by them, but, on the other hand, it is made by men, so far as it has been adapted by men." With the fact, now, as it seems, clearly established that these last two ganas are of younger date, the Samavidhanabrahmana also harmonises, where samans are cited from these two ganas, but only of the first two. The names of these ganas are now all clear. The gramegeyagana contains those samans that were studied in the community, the aranyegeyagana, on the other hand, hands down those samans that, because they were so very sacred, or, what is nearly the same, from a primitive standpoint, so very dangerous, were studied in the forest, outside the community (see especially the exhaustive paper of H. Oldenberg on Aranyaka in Nachrichten der kgl. Ges. der Wiss. zu Gottingen, Phil.-hist. kl. 1915, page 382 and for the Aranyaka texts of the Samaveda page 392 sqq.). We now come to a very difficult question, and one which is of the highest importance not only for a true insight into the historic development of the older Samavedic texts, but also of great weight for the explanation of many a passage in the Brahmana and the kindred texts. This question is: "was the purvarcika or was the uttararcika the older part?" Scholars are at variance. I myself maintained that the uttararcika must be regarded as prior to the purvarcika, chiefly on the argument that a collection of verses on which the Samans had to be chanted (as is the uttararcika) must have been a priori older than a collection of verses that served to register the melodies on which these verses had to be chanted (as is the purvarcika). Oldenberg, on the other hand, has made it appear that the purvarcika (together with the aranyaka part) was the older part, because this part only is mentioned in the vratas, and, moreover, the uttararcika is nowhere quoted in the Samavidhanabrahmana. I add to this that even so late a work as the Atharvaparisista mentions (46. 3, 6) as last verse of the Samaveda the last but one of the purvarcika (viz. Sv. I. 584). Convinced by Oldenberg's strong arguments, I thereupon proposed to formulate the facts thus: that from the oldest times on the chanters must have had at their disposal a certain collection of tristichs and pragathas, that served them at the soma-rites for chanting after their melodies; that this collection might have been the fore-runner of the uttararcika as it is known to us now-a-days. Oldenberg himself,

who wrote: "moglich ist es naturlich, dasz, als das Purvarcika redigiert wurde, ein Kanon der Agnistoma-liturgien schon fixirt war," did not materially differ from my view. The study of the Brahmanas and the kindred texts has now suggested to me a hypothesis, that perhaps may throw light on this difficult question. To state it directly at the beginning of my argumentation, this is my hypothesis: the author of the Brahmana was not acquainted with our uttararcika, it did not exist at his time, but the chanters drew the verses they wanted, directly from the Rksamhita, and the uttararcika was composed in later times, in order to have at hand, in the regular order of the sacrifices, the verses that were wanted. The facts, on which this thesis is based, are the following. 1. It is expressly stated (see the translation of Pancavimsabrahmana IV. 4. 1, sqq. with my notes) that in a certain case a great number of verses had to be taken directly from the Rksamhita. That the word used here by the three sutras: dasatayisu (Sayana periphrases dasatayyam) means "in the ten books of the Rksamhita," is proved by a sentence in the Nidanasutra (II. 11, see note 4 on Panc. br. VIII. 8. 26) dasatayenadhyayena as contrasted with ib. II. 2: chandasenadhyayena, which must signify: "a chapter of the Samavedasamhita." This presupposes for the chanters the acquaintance with the Rksamhita or at least with a part of it, from which their verses were to be taken directly. 2. The expression sambharya to denote a complex of verses to be taken from different parts of the Veda occurs thrice in the Brahmana: XI. 1. 5, XVI. 5. 11 and XVIII. 8. 8. This expression is simply incomprehensible from a Samavedistic standpoint, because in the uttararcika they are given as a whole, all after one another, but from a Rgvedistic standpoint they are truly sambharya's. 3. From the passage in the Brahmana IV. 2. 19, where a jarabodhiya-saman is mentioned, to be chanted on SV. I. 25 (=II. 733-735) it seems right to infer, that the uttararcika was later than the Brahmana. If this is true, the Brahmana cannot but have taken his verses directly from the Rksamhita. The compilator of the uttararcika may have inserted this triplet, because it occurred in the Brahmana. 4. See also the notes on the Brahmana XI. 7. 3, XIV. 1. 9. There can, however, be made several objections to this hypothesis.

In the first place, the verses as cited in the Brahmana diverge in not a few cases, more or less, from the reading presented in our Rksamhita. But it is a fact that, when the author of the Brahmana quotes from the Rksamhita in general, his citation mostly contains one or more corruptions (see the Brahmana I. 2. 9, I. 7. 6, IX. 2. 22, XXIV. 1. 9). In one case a change in a verse of the Rksamhita can be proved to be caused by the ritual (see the note on XI. 5. 1). Elsewhere a divergence from the Rkverse may have been caused by its saman-reading (see XIV. 6. 4). So the variants may be caused by slovenness, but in part they could be explained by the surmise that our Brahmana presupposes a recension of the Rksamhita different from the one known to us (see note 1 on VI. 10. 14). That some verses are found in the arcika but not in our Rksamhita might be explained by the conjecture that our Rksamhita once contained more verses than it now does in its present recension (see note 2 on VIII. 1. 1, note 1 (end) on XII. 13. 22). But I am not able to remove all the difficulties that seem to move against my hypothesis. I draw the reader's attention to the puzzle contained in VIII. 8. 26, SV. II. 62 as against RS. VIII. 98. 9 (here perhaps we might guess that the compiler of the uttararcika changed the Rkverse so as to bring it in accordance with the Brahmana.) A weighty argument against the hypothesis might be drawn from XII. 1. 9, 10, XII. 7. 9, 10, XIII. 1. 10-12, XIII. 7. 8-9, where the terms pentastichs, decastichs and so on, are used. We may ask here: "how could we know which verses are intended, unless it were supposed that the uttararcika existed at the time when the Brahmana was composed?" But perhaps the author of our Brahmana allowed here a free choice, whilst in later times the compiler of the uttararcika fixed these verses probably in accordance with the Jaiminiyabrahmana, which, as far as I have ascertained, states by citing their opening words, which verses had to be employed. Another difficulty is presented by the passage XIII. 1. 1 as compared with ib. 5. Although, then, all is far from certain, I suppose that now we may with some measure of probability propose the following development of the oldest Samavedic literature. The chanters were acquainted with the Rksamhita, at least with certain parts of it. From this text they took their verses 1 and on these 1 Cp. Oldenberg in the Journal of the German Or. Soc., Vol. XXXVIII, page 441 : " der Rigveda ist zugleich der alteste Samaveda," but read in the context this assertion is not equivalent to my hypothesis.

they chanted the lauds at the sacrifices of soma. In order to facilitate the study and memorization of this material, a double register was composed, one of the melodies and one of the verses, on which the melodies were chanted. The first register they arranged so, that each time only the first verse of a tristich or a pragatha was taken up, and this verse was considered as the bearer of the inelody. To this collection were added samans of a different kind, that had to be chanted by a single Chanter at other occasions. In this manner two books of songs, two ganas, arose the one to be studied inside the community: the gramegeyagana, and another to be studied in the forest, outside the community: the aranyegeyagana; this book comprised those samans that were the most holy (dangerous). Besides these song-books arose at the same time the second register: the arcika (the later purvarcika) and the aranyaka. Thereupon a Brahmana originated. Based on this Brahmana the Arseyakalpa was composed by Masaka, who described minutely the sacrifices of soma, but omitted the ten-day-rite, because its detailed description was already given in the chapters 10-15 of the Brahmana. Based again on these two books the srautasutra was composed (that of LatyayanaDrahyayana). Then only the uttararcika was compiled, which contained the verses of the Rksamhita, arranged in their regular order, so as they were to be used in the sacrifices. Now, if a Chanter knew by heart his two ganas and if he knew which verses ought to be adhibited in any rite, he could bring about all the changes that were necessary for adapting a melody to a given triplet or pragatha. These rules for adaptation were then fixed and systematically arranged in a special book the Puspasutra. But, in order to have at hand for immediate use the samans so as they were to be adapted according to the rules of the Puspasutra, two more books were composed: the adapted songbooks 1. the uhagana and 2. the uharahasyagana (called by abbreviation uhyagana). The first contains ready made for sacrificial use the samans in their regular order resting on the gramegeyagana, the second those that rest on the aranyegeyagana. This is according to my view the history of the oldest Samavedic texts¹. : 1 The material on which rest "for the greater part the above given considera · tions and conclusions is found in the paper of Oldenberg in the Journal of the German Or. Soc. (Vol. XXXVIII); in a paper: "De Wording van den Samaveda (Versl. en Meded. der. Kon. Ak. v. Wetensch. Afd. Lett. 4 e R., Deel IX); in my "