STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world by JSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact firstname.lastname@example.org. ARTICLE XII. ON THE POSITION OF THE VAITAM-SUTRA LITERATURE OF THE ATHARVA-VEDA. By MAURICE BLOOMFIELD, I'KOFESSOE IN THE JOHNS HOPKINS UNIVKBSITT, HALTIMOBE, MB. Presented to the Society October 30th, 1884. The following pages contain an attempt to define more exactly than has as yet been done the position of the Vaitana- sutra in the literature of the Atharva-Veda. It is based upon a pretty thorough, though not exhaustive, comparison of the Vaitana-sutra with the Kaucika-siitra and its commentary by Darila. On several interesting points I was enlightened by the Atharvamya-paddhati, a second paddhati to the house-ritual of the Atharva-Veda, two modern copies of which, in the pos- session of the Royal Library at Berlin, I had the privilege of using for my forthcoming edition of the Kaucika. As this paddhati presents some special points of interest, I give here for the first time a short notice of it. It differs from the Daca karmani, of which I presented a short sketch at the meeting of the Society in October, 1883,* in that it not merely paraphrases the description of certain sections of the ritual as given in the Kaucika, but also comments upon them somewhat independently, occasionally differing from Darila. After a short introduction, it turns to the paribhasa-sutras at the beginning of the Kfiucika,f then continues with the djya- * See the Proceedings of that meeting — where xaaj be found also a short, notice of the commentary of Darila. f The much more important and interesting paribhasas which are contained in chapters 7 and 8 of the Kaucika it does not treat in any way. 376 M. Bloomfield, tantra (here called brhatkupandikd : cf. the word kupandika in Weber's catalogue of MSS., No. 1253), with its appendix the uttaratantra ; then it treats the following samskdras : garbhd- dhdna, pumsavana, slmantonnayana, jdtakarman, ndmakarana, nirnayana, annaprdpana, goddna, cuddkarana, upanayana (with vedavrata, kalpavrata, mrgdruvrata /, visdsahivrata*), samdvartana, vivdha with madhuparka inserted (as in the Daca karmani, Kathaka-grhya and Manava-grhya),f caturthlkarman with the madughamaniprdya$citta\ as appendix, laghupdld- karman, brhacchdldkarman, pdntyudaka, sampatkdma(-karman) , pustikdma(-karman), abhicdra(-karman), vrddhiprdddha, daha- navidhi (with asthisamcayana), pindadana, sodapopaedra, and vrsotsarga. The text quotes the usual Atharvan literature : Gopatha Brah- mana, Vaitana-sfitra, Naksatrakalpa, Angirasakalpa, and the paripistas ; also such stock-books as Manu and the Karmapra- dlpajs It also cites Darila, and two other commentators, Bhadra and Rudra, both of whom are elsewhere unknown. || It mentions further a paddhati-kdra by the name of Kecava; an dedrya Upavarsa as author of mlmdnsd;*H a work called the Pafica- patalika (cf. Bohtlingk's lexicon), and finally Paithinasi. The latter is cited frequently and familiarly by Darila ; and it seems possible that the smrti of Paithinasi may go back to a dharma- sutra belonging to the Atharva-Veda. Often as this text is cited, no MS. of it has as yet come to light, as I learn from Buhler and Jolly; it would be interesting to find the Sutra-ritual of the Atharva-Veda completed by a dharmasutra, or some smrti going back to a dharmasutra. The name of a teacher Mausaliputra Paithinasi occurs also in Ath. Pario. 4. 3 and 17. 13. Indian tradition is unanimous in presenting the ritual litera- ture immediately attaching itself to the Atharva-Veda as con- sisting of five kalpas. The Caranavyuha (Ath. Paric. 49) pre- sents them in the following order, and under the following names : naksatrakalpa, vitdnakalpa, samhitdkalpa, angirasakalpa, and pdntikalpa. The larger Caranavyuha, the fifth paripista of the White Yajur-veda, counts naksatrakalpa, vitdnakalpa, samhitd- kalpa (with the variant sarhhitdvidhi), vidhdnakalpa (withthe variant abhicdrakalpa), and pdntikalpa. The Devipurana (cited by Weber, Ind. Stud. iii. 279) counts naksatra, vditdna, samhitd- vidhi, dngirasa, and pdnti. The Visnupurana (iii. 6 ; vol. iii. p. 63 of Hall's edition of Wilson's translation)** knows these kalpas by the same name and in the same order as the Ath. Paricista. The * Cf. AY. xvii. 1-5. + See Proceedings A. 0. S., loo. cit. ; Jolly, Das Dharmasutra des Visnu und das KathakagrhyasQtra, Proceedings of the Munich Academy, June, 1879, p. 76. 1 Cf. Atharva-paricista 37. 9 (MS. or. fol. 973 in the Royal Library at Berlin) ; cf. Ind. Stud. v. 404. § Cited as often under the name of chandogaparifista of Katyayana. If darilo rudrabhadrau ca trayas te bhasyakarah. T Cf. Colebrooke, Miscellaneous Essays, ii. 322. "*Cf. Gopatha Brahmana, ed. by Rajendralalamitra, Introduction, p. 5. **/ The Vaitana-Sutra. 377 Atharvanlya-paddhati also follows the Ath. Paricista, but substi- tutes the name abhicdrakalpa for dngirasakalpa, and cites the guarantee of the mimansa-tea,cher Upavarsa that these are pruti, and that there are other kalpas which are stnrti* The Vayu- purana (61 ; p. 526 of the edition in the Bibliotheca Indica) reads : naksatrakalpo vditdnas trtlyah samhitavidhih : caturtho 'ngirasah (!) kalpah pdntikalpap ca pancamah. 'In the Mahabha- rata, xii. 13258, the five kalpas of the Atharva-Veda are men- tioned, but their names are not given. In the Bhagavata-purana xii. 7. l,f the kalpas occur as : naksatrakalpah pantip ca ka$ya- pdngirasddayah (!). In the Mahabhasya, there occur together several times the kalpas of Kacyapa and Kaucika (Ind. Stud. xiii. 417, 419, 436, 445, 455). Weber (ibid. 436 and 445) is disposed to see in this Kaucika and in the derivative Kancikin the repre- sentatives of a Yaj us-school ; but this passage from the Bhaga- vata-purana shows clearly that the Kaucika of the Atharva-Veda is meant. . But who is Kacyapa among the kalpas? Can there be any nexus between him and the otherwise authorless Vaitana- sutra?J The commentary to the Atharva-Veda by Sftyana, which Shankar Pandurang Pandit announced in the London Academy of June 5th, 1880, in the Introduction refers the ritual of the Atharva-Veda to five kalpas: kaucika, vaitana, naksatra, dngi- rasa, and pdnti. It adds the interesting statement that these ritual books belong to four of the nine fdkhds or bhedas into which tradition unanimously divides the Atharva-Veda : namely, the Caunakiyas, the Aksalas, the Jaladas, and the Brahmavadas. I find essentially the same statement in the introduction to the Atharva-paddhati : atharvavedasya nava bheda bhavanti : tatra catasrsu pdkhdsu pdunakddisu kdufiko 'yam samhitavidhih . . . A negative corroboration of this statement is to be derived from the relation of the Kaucika and Vaitana sutras to the one other branch of the Atharva-Veda which has come to light, the Kacmira- recension, which calls itself the Piiippalada; we know that the sutras do not belong to this pdkhd, from the technical fact that they cite the mantras of the Piiippalada in lull instead of citing the pratlkas.% And once the.Kaucika implicitly confesses itself as fdunaklya, in 85. 6, 7, where the opinions of the Cimnakins and Devadarcins about measurements are confronted ; the opinion of the Cfiunakins is given last, and is therefore, in accordance with the usual method of the Kaucika, to be considered as the * mimansayarh smrtipdde kalpasutradhikarane naksatrakalpo vitanakalpas trtlyah samhitakalpat caturtha angirasakalpah fantikalpas iu pancamah : ete kalpa veda- tulya hi Hi bhagavan (!) upavarsucaryena pratipaditam : anye kalpah smrtitulvah fCf. Gop. Br., ibid. p. 7. ' ' »*-..». X The word vaitana itself is occasionally personified: see Gop. Br., Introduc- tion, p. 4: in the Ath. Paddh. to Kauc. 1. 6, the Vaitana is cited as cdunakiya- sutra. § So Kauc. 72, 91, 107, 115; Vflit. 10. 17 ; 14. 1 ; 24. -1 : cf. Roth, Der Atharva- veda in Kaschmir; and Garbe, in the Introduction to the Vaitana (text), p. vii. VOL. XI. 48 378 M. Bloomjield, accepted one.* The prevalence of this canon of five kalpas is furthermore attested by the compound pancakalpah in the Maha- bhasya ; pancaJcalpah is probably not to be understood (with Weber, Ind. Stud. xiii. 455) as one studying five different kalpas, i. e. prauta-sutras, but means an Atharvavedin who is familiar with these five kalpas. The gist of these traditions is therefore as follows : There are five ritual books regarded as pruti, and they are (if we disregard the crooked order in which they are usually presented) : I. The Kaucika-sutra, known also under the name of Samhita- kalpa, or Samhita-vidhi. II. The Vitana-kalpa, or Vaitana-sutra. III. The Naksatrakalpa. TV. The Canti-kalpa. V. The Angirasa-kalpa, known also as Abhicara-kalpa or Vi- dhana-kalpa. Accordingly the statement of Weber, Indische Literaturge- schichte, p. 169, is to be corrected. He says, speaking of the Kauc^ka-sutra : " Zu diesem sutra gehoren noch fttnf sogenannte halpa : naksatra, panti, vitana [which he does not identify with the Vaitana, mentioned on the preceding page], samhitakalpa [which is itself the Kaucika], uud abhioarafcalpa." So much for the Hindu juxtaposition of the kalpas. To the western student of Indian literature, these coordinated five texts are of very different value. To begin with, the Angirasa- or Abhicarakalpa has, so far as I know, not as yet turned up in Europe. From the character of the references t'o it, one can safely guess that it is essentially nothing more than a paripistd of the Atharva-Veda, treating mainly abhicara, or witchcraft ; and it would probably be of some value in elucidating the sixth book- of the Kaugika. That the Naksatrakalpa and Qantikalpa are pa- ricistas appears from their titles : e. g. Ch. 110 (Weber's Cata- logue, p. 89), brahmavedaparipistam naksatrakalpabhidhanam. In a modern codex of the Ath. Paricistas, which Btlhler presented *A still more incisive difference between the various pakhas of the Atharva- Veda is reported in Atharvaparicista 2. 3, 4. The passage describes the rules for choosing the purohita. Not only is an adherent of the three remaining Vedas excluded, but even within the Atharvan itself those' belonging to the Mauda and Jalada pdkhds are rejected, while Paippaladas and (Jauuaklyas are considered desirable: na havih pratigrhnanii devatah pitaro dvijah : tasya bhumipater yasya grhe nd 'tharvavid gwuh. samahitangapratyangavidyacaragundnvitam : paippald- dam gurum kurytlt frlrastrarogyavardhanam : tatha (aunakinam Ca 'pi devamantra- vipafdtam : . . . . bahvrco hanti vai rdstram adhvaryw ndfayet mtan : chandogo dhanarh ndcayet tasmdd atharoano gwuh : ajnandd vd pramaddd va yasya syad bahvrco guruh : defarastrapuramdtyandfas tasya na samfayah. yadi vd "dhvarya- varh ' raja niyunakti purohitam : pastrena badhyate ksipram pariksindrfhavdhanali. yathai 'va pangur adhvdnam apdksl ca 'ndabhojanam : evam cfiandogagurund raja vrddhim na gacchati. purodha jalado yasya maudo vd syat katham cana : abddd dacdbhyo masebhyo rastrabhrancaih sa gacchati. Of. also Yajfiavalkya i. 312. On the (akhds of the Atharva-Veda c£.,in general Indische Studien, i. 296; iii. 277-8 ; Weber, Omina nnd Portenta, p. 413; Roth, Der Atharva-Veda in Kaschmir, p. 24 fg; Weber, Indische Literaturgeschichte, p. 166 ; Rajendralalamitra in the In- troduction to the Gopatha Brahmana, p. 5. The Vaitana-Sutra. 379 to the Royal Library at Berlin (MS. Or. fol. 973),theNaksatrakalpa actually stands at the head, and is counted as the first paripista. Of the Berlin codex of the Qfmtikalpa I possess a copy, but there is to be found in it nothing of great intrinsic interest ; nor does the text add anything to the elucidation of either the Kaucika or the Vaitana. The Naksatrakalpa in some of its last sections bears upon a few points in the Kaucika. A short summary of both texts is to be found in Weber's Naksatra II. p. 392 fg. The two remaining texts contain ritual sutras, but their relation to one another and to the Atharva-samhitii is in many respects different from that of corresponding texts in the pdkhas of the other Vedas. The doubtful eanonicity of the Atharva- Veda (which gains expression in the prevailing designation of the Vedas as a trdividyam or trayl vidyd), or at least the patent fact that the hymns of the Atharva-Veda are not well-fitted for employment in the prai/ta-ritual, has left its stamp on the prduta- literature. This is both secondary and scarce ; its only brahmana, the Gopatha Brahmana, and its prduta-sutra, the Vaitfma, no doubt belong to the latest products of their respective kinds of litera- ture. They are perhaps merely imitations of the literary condi- tions in vogue in the caranas of the other Vedas. On the other hand, the private life and the private performances of one who adhered to the Veda of incantation and exorcisms naturally were very extensive, bringing in much which was unknown in other Vedic schools ; and to this the largest and most important ritual text of the Atharva-Veda, the Kaucikastitra, owes its peculiarly prominent position among the Atharvan books. As a rule, the grhya-sutras are dependent upon the prduta- sutras; the former refer to the latter familiarly, and do not de- scribe a second time performances which have been treated in the prduta-sutra. So Pfirask. i. 1. 4 and i. 18. 1 refers to Kilty. Cr., with the expression purvavat, ' as above.'* In the collection of the sutra-carana of Apastamba, in which all religious sutras are united into one work, the prduta-sutra precedes the grhya. Stenzler. in the preface to his edition of the Acv. Gr., p. iii., re- marks : " The chief obstacle in the way of a correct understand- ing of the grhya-sutras lies in the fact that they appear as ap- pendices to the prduta-ru]es, and presuppose an acquaintance with the prrmta-performances."t Nothing corresponding to this is to be found in the correlation of the Kaucika and the Vaitana. There is no point in which the Kaucika depends upon the Vaitana ; on the other hand, the de- pendence of the Vaitana upon the Kaucika is apparent at almost every step, where the difference of the subject-matter, and the difference between Vedic ceremonial and house ceremonial, allow it. _ The position of the Vaitana may therefore perhaps be de- scribed as follows : it is not the product of practices in Vedic ceremonies which have slowly and gradually developed in a cer- * Paraskara's sutra is called outright Katiya-sfitra, or Paraskara's Kfttiya-siitra. f Cf. e. g. Acv. Gr. i. 5. 1 ; iv. 1. 16 ; 4. 6. 380 M. Bloomfield, tain school, but probably a somewhat conscious product, made at a time when Atharvavedins, in the course of their polemics with the priests of the other Vedas, began to feel the need of a manual for Vedic practices which should be distinctly Atharvanic. That the Atharvan is poorly fitted for serving as foundation to a Vedic ritual can be seen from the fact that very little that dis- tinctly belongs to its proper sphere (carmina, devotiones), and at the same time is not borrowed, or does not at least occur in the Kaucika, is found in the Vaitana. On the other hand, it contains numerous verses and formulas from the Yajus-samhitas; and in the description of the ritual it follows very closely Katyfiyana's Crauta-siitra. I n Vait. i. 1.8, devata havir daksind yajurveddt, ' the authority for the divinities, the sacrificial material, and the sacrificial reward, is from the Yajur-Veda,' we seem to find formal recognition of this fact. And perhaps it may be taken as a sign of intimacy between the Vaitana and Katyayana, that the com- mentary to the latter cites the Vaitana quite frequently, under the names of Vaitana-sutra, Atharvana, and Atharva-sutra (cf. Garbe, preface to text, p. vi.). On the other hand, the relation of the Vaitana to the Kaucika may be described by stating that the Vaitana treats the Kaucika almost as though it were another samhita, taking for granted that the ritual of the Kaucjika, and the mantras which it quotes from other sources than the vulgata, are understood and known by its readers. The Vaitana-sutra we possess in an excellent edition and trans- lation by Richard Garbe (text, London, 1878 ; translation, Strass- burg, 1878). Before bringing detailed evidence for the state- ments made, we may collect here a few addenda and corrigenda, most of which come from sources which were inaccessible to Garbe. Vait. 1. 19: fivabhir is translated by ' rait Jiwwasser.' Darila to Kauc. 3. 4 explains jlvdbhir by jiva sthe Hi catasrbhih : i. e. 'with thejlwa-verses' (AV. xix. 69. 1-4). Viiit. 10. 5 : gandhapravddabhir is translated ' mit Stoffen welche als wolriechend zu bezeichnen sind.' According to the Daca kar- mflni to KauC/. 54. 4, the gandhapravddd rcas are AV. xii. 1. 23- 25, all beginning with the words yas te gandhah. Vait. 24. 3 : apam suktdir. The hymns to the waters are given as follows by Darila to Kauc. 7. 14: ambayo yanti, pambhumayo- bhu, hiranyavarnddayah, krsnath niyanam, sasruslr, himavatah pra sravanti, vayoh puta ity apam suktani (AV. i. 4. 1 ; 5. 1 ; 6. 1 ; 33. 1 ; vi. 22. 1 ; '23. 1 ; 24. 1 ; 51. 1). Vait. 37.23: utthapambhir, 'mit den utthdpani-V eraen.'' Ac- cording to the Antyestikarman (a paddhati on the burial-cere- monies) to Kauc. 81. 31, these verses are AV. xviii. 3. 8, 9 ; 2. 48; 1. 61 ; 2. 53 ; 4. 44. Vait. 37. 24 : harinlbhih, ' mit den /wM^Vu-Versen.' According to the Antyestikarman (ibid.), they are AV. xviii. 2. 11-18. The passage Vait. 5. 10, treating of the preparation of the pdntyudaka, the only one in the text which Garbe finds himself The Vaitdna-Sutra. 381 compelled to leave untranslated, will be explained below, in an- other connection. Several of the pratlkas cited, whose origin was unknown to Garbe,* can be found in the vulgata by the aid of Whitney's In- dex Verborum : thus, Vait. 3. 17, indra glrbhih, is AV. vii. 110. 3'; Vait. 9. 4, purnd darve, is AV. iii. 10. T ; Vait. 8. 6, indragni asmdn, is AV. x. 1. 2V' d ; Vait. 28. 32, krte yondu, is AV. .iii. 17. 2*. The pratika in Vait. 16. 1, purnam adhvaryo pra bhara, and the following verse, which is referred to in sutra 2, are prob- ably a modification of AV. iii. 12. 8, 9 : purnam ndri pra bhara.] The quantity of material in the Vaitana which is characteristi- cally Atharvanic is quite small, if we exclude what is shared with it by the Kaucika. In Vait. 1. 1, the demand is made that the super- vising priest, the brahman, shall be a brahmavedavid. The expres- sion brahmaneda does not occur in the Kaucika; in its place Kauc. 63. 3 and 94. 3, and the Vaitana itself in 1. 5, have the more archaic bhrgvangirovid ; and Vait. 11. 2 still further exhibits the expres- sion atharvangirovidam brahmdnam. In the Gopatha Brahmana and the par ipistas, also, the expression bhrgvangirovid prevails ;J brahmavedavid^ probably belongs to a later period. On several occasions, the Vaitana takes it upon itself directly or indirectly to express the superiority of the Atharva-Veda and its priests. In 11. 2, it does so by giving precedence to the atharvdngirovid who shall be chosen as brahman (in the technical sense) over the udgatar, hotar, and adhvaryu. This precedence of the Atharvan appears in a more valuable form, because impliedly, in themantra- passage Vait. 6. 1 : the fire compared to a steed is addressed with tarn tvd ''dadhur brahmane bhagam agre atharvdnah samavedo yajunsi : rgbhih putam prajdpatir atharvane '■g.vam prathamam nindya. In the mantra-passage Vait. 37. 2, the brahman carries on a petty polemic against the other priests participating in the sacrifice, saying : " You are not a guardian of the sacrifice higher than I ; you are not better, more excellent, do not stand above me ; you do indeed speak instructively, but you must not place yourself on a level with me." This accords again with the polem- ical tone of the later Atharvan-literature, especially the paripistas (cf. the extract from Ath. Paric. 2. 3, 4 above), in which all possi- ble effort, and sometimes virulent language, is expended upon the *See text, p. 81. 1 1 have noted the following errata : in the preface to the text, p. vi., read abhicaresv etc. 2. 10, instead of . . . 2. 12; text 1. 18, vag for vag; 2. 1, brah- mann for brahyann ; 25. 1, osadhirtor osaslr; 28.11, phanta for panto. ; 28.12, bhagali for bhagali. In the translation, 34. il, uccairghosa for -gosa. % See Gop. Br. i. 2. 24 ; 3. 1 ; and, for the parifistas,' Weber, Omina und Por- tenta, p. 346. § It is indeed quite obvious that the original brahman in Vedic sacrifices was not connected with the Atharva-Veda at all, but was supposed to know the three other Vedas, in distinction from the hotar, udgatar, and adhvaryu, who each knew only one. Thus the brahman is fitted for his office of supervisor of the sacrifice, in which he would receive but little aid from a knowledge of the Atharvan: cf! Haug, Brahma und die Brahmanen, p. 10; Miiller, in the Journal of the German Oriental Society, vol. ix., p. xlvii. 382 M. Moomfield, task of showing the dignity of the brahman-\>riest and his Veda. Curiously enough, this effort is not restricted to Atharvan-writings; e. g. in the Mahabhasya the Atharvan consistently receives the first position among the Vedas (Ind. Stud. xiii. 433 fg.) ; Yajiia- valkya, i. 312, demands that the purohita shall be acquainted with the Atharvdngirasas ; and even in the grhya-sutras of the Rig- Veda (Cankh. iv. 10. 3 ; Cambavyagrhya, Ind. Stud. xv. 153 ; Acv. iii. 4, 4), on the occasion of the tarpanarn, the Atharva-m Sumantu, whom tradition unanimously designates as the first source of Atharvan-lore,* is honored with the first place in the list of sages. In addition, there is very little Atharvan-material which belongs especially to the Vaitana : thus, 2. 10, dsdditesu havihsu ''Man purastdddhomdn juhoti : abhicdresv dbhicdrikdn, and 43. 25, paunakayajno l bhicarakamasya, bring in abhicdra in an independent way — moreover, the paunaka-s&crifice is unknown elsewhere. The passage describing the pdntyudaka, which will be translated below, also contains a special contribu- tion of the Vaitana to Atharvan-practices ; and Vait. 1. 3 con- tains an interesting paribhasa-sutra, in which are given the dic- tates of several Atharvan teachers concerning the expedients which are to be resorted to when no mantra is prescribed for an oblation. Further, the Vaitana quotes independently and in full three hymns of the Kashmir-branch, the Paippalada (in 10.17; 14. 1 ; 24. 1), just as the Kaucika and Gopatha Brahmana quote hymns and verses of the same version : see Roth, Der Atharva Veda in Kaschmir, p. 23. That the hymns of the vulgata are cited largely, and only with their pratikas, proves nothing for the archaic character of the work ; the author of a prduta-sutra to any Vedic collection, at any period whatever, would observe this practice and the technicalities attached thereto. As indicated above, the few points mentioned are far from ex- hausting the materials in the Vaitana which are characteristically Atharvanic. But those remaining cannot be adduced as testi- mony in favor of the originality and age of the text, because the Vaitana shares them with the Kaucika. The points of contact between these two texts in general are very numerous. In the matter of external form, it may be worth noting that the Vaitana-sutra proper, which is divided into eight adhydyas, is frequently found extended to fourteen adhydyas^ by the addition of prdyapcitta-sutras. So in one of the manuscripts which Garbe used in his edition, and in a codex presented to the Royal Library at Berlin by Prof. Eggeling (MS. or. oct. 343). This may have been done in deference to the fourteen adhydyas of the Kaucika, The Vaitana as well as the Kaucika frequently begins a chapter with a long wa«£r«-passage which belongs to the action of the preceding chapter : so Vait. 6, 12,f 14, 24 ; and Kauc. 2, 3, 4, 5, 70, 88, etc. etc. I have not noticed in the Vaitana the *Visnupurana iii. 6; Vayup. 60; Bhagavatap. xii. 7.1; cf. Roth, Der Atharva- Veda in Kaschmir, p. 26. f So also Gop. Br. i. 3. 22. The Vaitana-Subra. 383 practice, common in the Kaucika, of disregarding hiatus produced by sandhi: so Kauc. 6. 17, patnydnjalau = patnyd(h) anjaldu; 6.34, ddivateti = ddivata(y) iti; 17. 3, talpdrsabham = talpa(y) drsabharn ; cf. also the seven cases mentioned by Weber in his Omina und Portenta, p. 390, note 3. But both texts (Vait. 7. 15 ; 28. 26 : Kaug. 87. 23 ; 88. 29, etc.) have a case of peculiar sandhi in the words pitryupavita and pitryupavltin.* Each text has once a nominative plur. in -ayas employed as accusative: Vait. 11. 24, angusthaprabhrtayas tisra ucchrayet, 'he shall lift up the first three fingers beginning with the thumb;' Kauc. 8. 19, trayo- dapyddayas tisro dadhimadhuni vdsayitvd badhndti, ' he ties on the amulet after having kept it in a mixture of sour milk and honey during the three nights beginning with the night of the thirteenth.' The parallelism between -prabhrtayas tisras and -ddayas tisras is hardly accidental. Both texts frequently intro- duce pfo^a-passages describing or complementing in metrical form what has been previously described or indicated in sutras. So Vait. 4. 23 and 9. 12 are introduced by tad api plokdu vadatah ; 19. 20 and 20. 4 and 1 1 by tatra plokah; 31. 15 by tad etac chloko l bhivadati ; and Kauc. 6. 34 by athd ''pi plokdu bhavatah; 68. 35 by tatra plokdu; 74. 10 by tad api ploko vadati. Both texts occasionally refer to the Brahmana with the phrase brdhmanok- tam or iti brdhmanam ; the Gopath a Brahmana is not the text referred to. So Vait. 7. 25, brdhmanoktam agnyupasthdnam, 'the approach of the fire is described in the Brahmana;' 17. 11, brdhmanoktdn ity anubrdhmaninah, which Garbe translates, hesi- tatingly, ' the Anubrahmanin employ the things prescribed in the Brahmana;' 31. 1, . . . sattram upayanto brdhmanoktena dikseran, ' those entering upon a sattra shall consecrate themselves in the way described in the Brahmana ;' so also the word brahmana in 43. 45. In the same way Kauc. 6. 22, nd 'daksinam havih hur- vlta: yah kurute krtydm dtmanah kuruta iti brdhmanam ; 80. 2, dahananidhdnadepe parivrksdni nidhdnakdla iti brdhmanok- tam.f So also brdhrnanohtam in 58. 3. The Vaitana lapses into the broad brdhmana-style at the end (43. 46) : ya imdu halpdv adhlte ya u edi 'vam veda tena sarvdih hratubhir istam bhavati sarvdnp ea kdmdn dpnoti. This occurs quite frequently in the Kaucika : e. g. the long passage 6. 23-28, 67. 3; and especially 73. 28. Both texts employ very frequently the expression man- trokta, 'the person or thing mentioned in the mantra whose pratlka is cited.' So Vait. 1. 14, sinlvdli prthustuka iti mantrok- tdm, ' with the verse " O Siniviill with the broad braids " he addresses the divinity mentioned in the mantra;' 5. 7, agnim apvatthdd iti . . mantrokte aranl grhnantam vdcayati. So also 11.15; 30.27; 31.4. Very much more frequently this occurs in the Kaucika: e. g. yad yat krsna iti mantroktam, ' with the verse " whatever is black " he does what is stated in *SoalsoGop. Br. i. 3. 12. f Atharvaulya-paddhati : dahanastkime vrksavarjite deft nidhanakala iti brah- manoktam. 384 M. Bloom-field, the mantra :' i. e. he polishes the mortar and pestle. So also Kauc, 21.11; 23. 14 ; 26. 14 ; 31. 17, 21 ; 32. 5, etc. Worthy of note also is the frequent employment of the solemn exclamation janat in addition to bhus, bhuvas, svar. According to the Go- patha Brahmana i. 1. 8, this is symbolic of the angirasa, i. e. the Atharva-Veda.' So Vait. 1. 3, 18; 2. 1 ; 8. 3; 17. 4; 18. 17; 30. 15 : Kauc. 3. 4 ; 55. 1 ; 69. 22 ; 70. 6 ; 90. 18 ; 91. 9, etc. Further- more, the two texts share many technical terms, which are en- tirely restricted to the Atharvan-ritual or occur preponderatingly in it. So the purastdddhoma and samsthitahoma, an introduc- tory and a final oblation of melted butter, are peculiar to the Atharva-Veda. They occur also in the Gopatha Brahmana. The terms sarupavatsd, ' a cow with a calf of the same color as herself,' and its derivative sdrupavatsam, ' milk from such a cow,' occur very frequently in the Kaucika; the latter occurs also once in Vait. 12. 14. The terms sampdta, 'dregs or residue of an offering,' and sampatavant, though not restricted to the Atharva- Veda ritual — they occur also in Gobhila, and are explained in the Grhyasamgraha i. 113 — occur with especial frequency in the Kaucika; in the Vaitana, 12. 14, we have once sampatavant; sampdta occurs twice : 31.25; 32.7.* The term dkrtilosta, ' nat- ural lump of mud,'f occurs once in Vait. 5. 12; it occurs very frequently in the Kaucika, as will be seen below. The pdntyu- daka occurs once (5. 10) in the Vaitana, but very frequently in the Kaucika; the latter gives in the 9th chapter a full descrip- tion of its preparation, which is expanded still further by the paddhatis.\ The peculiar abstract rasaprdpam, ' eating of broth,' occurs twice in the Vaitana (21.20; 30.6); once also in the Kaucika (21. 19). The word puroddpa-samvarta in the sense of puroddpa-pinda occurs once in each text (Vait. 22. 22: I£auc. 30. 17). ' Each text has once the dvandva ydmasdrasvata (Vait. 37. 26: Kauc. 83. 16). Further, the teachers mentioned in the Vai- tana occur in the Kaucika, and the designations of (/anas, ' strings of hymns,' as also of single hymns of prominent character and wide application, .and groups of verses, are essentially the same: the cdtanani, aparh suktdni, pambhumayobhu, and dfdpdllya hymns, the gandhapravdda, .jlva, utthapanl verses occur in both texts : see below, p. 386. Finally, there are about sixty passages in the Vaitana, in which the ritual described shows a more or less close resemblance to performances in the Kaucika; the resem- blance sometimes amounts to absolute identity, and from that shades off to mere similarity in the arrangement of certain details, in the employment of hymns, etc.§ * Of. Gop. Br. ii. 6. 1. fg. ; Weber, Omina und Portenta, p. 371; Garbe trans- lates sampatavant in Tait. 1 2. 14 incorrectly by ' welche er (in der Milch von einer eben gescbilderten Kuh) zubereitet hat.' f Darilafol. 42", 1. 10 (to Kauc. 8. 16): akrtilvstah Icsetralostah. \ Of. also Gop. Br. i. 2. 18. § The following is a concordance of these passages: The Vaitana- Sutra. 385 It would certainly be going too far to suppose that the Vaitana has drawn upon the Kaugika for all these numerous correspond- ences ; it is very probable that many of the Atharvan specialties of both texts were simply current in Atharvan-schools, in such a way that they would be at the bidding of the compiler of a religious manual at any time. So there are at present, as far as I know, no means of deciding whether or not the Gopatha Brah- mana is anterior to the Vaitana, in spite of the many correspond- ences, which in a number of places rise to absolute identity : Vait. 2. 15 = Gop. Br. i. 5. 21 ; Vait. 3. 10 = Gop. Br. ii. 1. 2; Vait. 3. 12: Gop. Br. ii. 1. 3; Vait. 3. 20: Gop. Br. ii. 1. 7; Vait. 5. 10: Gop. Br. i. 2. 18; Vait. 5. 18: Gop. Br. i. 2. 18; Vait. 11. 1: Gop. Br. ii. 1. 16 ; Vait. 12. 1 : Gop. Br. i. 3. 22; Vait. 16. 15: Gop. Br. ii. 2. 1 2, etc. But in looking over the parallels between the Vai- tana and the Kaugika, it is found that the Vaitana frequently exhibits a certain fact or series of facts in a fragmentary way, merely presenting sundry features of a group of facts, which the Kaugika exhibits apparently in full. This occurs in so marked a manner that the lesser bulk of the Vaitana does not suffice to explain it away. The Vaitana, for example, mentions teachers here and there : Kaugika, Yuvan Kaugika, Bhagali, Mathara, Qaunaka. There occurs in addition only the expression acaryah (1. 3; 5. 13; 1. 16), which Gaibe, following a suggestion of Btihler's, regards Vaitana. Kaugika . Vaitana. Kaugika. 1.1,5 64. 3 ; 94. 3 10.3 8. 13 1.11 1.32 10.4 88.16 1.13 1.33; 73 13 10.5 13.12 54.4 1. 16 5.5 11.2 64.3: 94.3 1. 19 3.4 11. 13 82. li 1.20 3.5; 137 33 J1.14 24.28 2.2 2.6 11.24 8. 19 2.4 6.23 12.7 57.5 2.5 137.11,15 15.15 3.4 2.6 76.6 16.6 23.1 3.5 5.12 16.15 57.5 3.12 9.2 16.17 6. 1 3.20 6.23 17.4 108.2 4.7 6.9 18.2 62.1: 88.17 4.11 76.26 20.9 88.29- 89.1 4.13 6.4 21.9 68.6 4.22 6.19; 42. 5; 68.33 21.20 21.19 5.10 8.16 22.22 30. 17 5.15 70.1 24.3 7.14; 140. 1 7 5.18 7.21 24.7 6.11 6.11 1.31 24.14 40. 10 6.12 1.35; 82. 44 .28. 30 20.1 7.4 3.4 34.11 16.1 7.6 3.9 34. 12 16.7 7. 15 87.23; 88. 29 36.7 17.1 7.20 6.20 37.23 82.31; 83.23 8.6 5.2 37.24 82.31; 83 23 9.4 138. 11 37. 26 83.16 9.8 87.7 38.1 97.7 9.20 84. 19 38.4 88.29; 89. I 10.2 8.12 43.7 4.18 VOL. XI. 49 386 M. Bloomfield, as pluralis mqjestaticus, and translates by ' my teacber.' The Kaucika has all of these excepting the dedryds, and in addition knows the following worthies : Gargya, Parthacravasa, Kanka- yana, Paribabhrava, Jatikayana, Kaurupathi, Isuphali and Deva- dar9a. The same superiority of the Kaucika is exhibited in the employment of ganas (hymn-lists) and verse-lists. The Vaitana knows the following list of designations, which is approximately exhaustive: the cdtana, mdtrndma, and vdstospatlya ganas, the apam suktdni, the sampdta-hymns ; the pambhumayobhu-hymns, the madhu-sukta, the dpdpdliya-hymn, the sahasrabdhu-hymn ; ihejivd, the gandhapravddd, the utthdparil, and the harini-v ernes. The Kaucika has all of these, excepting the designations madhu- sukta, sampata-, and sahasrabdhu-sukta ; and in addition the brhat and laghu-pdnti-ganas, the salila-gana, the abhaya-gana, the mrgdra-suktdni, the sdmmanasydni, samgrdmikani, and vdr- casydni suktdni; the paihtdti-hymn, the pumbham-hymn, the sarh- nati-vere.es, the ahholingd-xerses, the two verses called dig-yukte, etc. etc. The paribhdsd-sutras for the Kaucika proper are con- tained in chapters 7 and 8 ; the first six chapters, treating the darpapurnamdsa-ceremomes, I think can be proved to be of later origin : or, at least, their conjunction with the parts following has taken place after the bulk of the text was fixed. These paribhd- sds find but little application in the Vaitana; but we have two of the rules practically employed in 10. 2 and 3 : ardtiyor iti yuparh vrpcyamdnam anumantrayate (Kauc. 8. 12 : ardtiyor iti taksati) ; yat tva pikva iti praksdlyamdnam (Kauc. 8. 13 : yat tva pikva iti praksalayati). The term dkrtilosta occurs but once in the Vai- tana; in the Kaucika it occurs frequently (8. 16; 21.2; 25. 7; 37. 8; 60. 15 ; 69. 11, etc.), and plays a prominent part in its prac- tices, occurring frequently in connection with valmlkalosta or val- mlkavapd, 'lump of mud with ants:' both of these are mentioned in the paribhasd-sutra 8. 16 as pdnta, 'holy.' The pdntyudaka, ' holy water,' occurs but once (5. 10) in the Vaitana, in a passage which in part refers to the Kaucika ; the latter devotes the ninth chapter to a description of its preparation, and its application is very common in the performances of the personal sacraments (the saniskdras). Again, both texts have the upavatsyadbhakta, ' fast-day-food, eaten on the day preceding solemn sacrifices ;' the Kaucika, 1. 32, defines it as 'diet from which honey, salt, meat, and beans are excluded.' And the sporadic occurrence in the Vaitana of the terms sdrupavatsa and sampata, which are so characteristic of the Kaucika, testifies to the fact that they are taken from the ritual of that text. There is, however, stronger evidence than this for the depend- ence of the Vaitana upon the Kaucika. In several passages, one of which is especially important, the Vaitana refers to ritual described in detail in the Kaucika, indicating merely the first and last traits of the performance : thus, Vait. 1. 19, jvodbhir dcamye ' tyddiprapadandntam, 'he performs the rite at the beginning of which he rinses his mouth, accompanying the ' act with utterance The Vaitana- Sutra. 387 of the jivd-verses, and which ends with the pr apad-i or mula;' Kaucika 3. 4, jivabhir avamyo ''potthdya vedaprapadbhih prapa- dyata om prapadye bhuh prapadye bhuvah prapadye svah prapa- dye janat prapadya iti* Again, Vait. 11. 14 : daksinena , ynim JcaQipvetyadiviksanantam, ' south of the fire he performs the rite which commences with the spreading of the cushion and ends with contemplation.' This refers to quite a long passage from the Kaucika, 24. 26-3 J. And in the same manner Vait. 24.3, apdrh suktdir ityddyupasparpandntam, ' he performs the cere- mony which begins with bathing, accompanied by the recitation of the hymns to the waters, and which ends with the touching of the water,' refers to Kauc. 7. 14 and 140. 17. And Vait. 24. 7, vi rmcncdmityddimdrjandhtam, ' he performs the act which begins with the recital of the mantra vi mwhcdmi and ends with the act of drying (the face),' refers to Kauc. 6. 11— lS.f Sound as this evidence seems to be at first sight, I confess that I cannot regard it as a final proof that the Vaitana has in such cases actually borrowed from the Kaucika. It might well be possible that we have here ritualistic acts common and familiar in the Atharvan schools reported independently by the two texts, and that it is due merely to difference of style and method that the Kaucika reports them in full, while the Vaitana only sketches their outline. And it is worth noting that the Vaitana never men- tions the Kaucika as the source from which these outlines may be filled in. The much stronger evidence which the passage describ- ing the fdntyudaka (5. 10) furnishes is therefore welcome. The passage reads : usasi pdntyudakam karoti cityddibhir atharvanl- bhih kapurviparvdrodakdvrkkdvatindddtiirdahantibhir angirasi- bhip ca. G-arbe finds himself unable to translate it, and remarks in the critical notes : " For this incurably corrupt passage the unanimous reading of all MSS. is given, as no explanation can be obtained from the AtharvanukramanI, out of which I have pre- pared an index of rsis and devatds." Garbe is misled by the sup- position that the words cityddibhir dtharvanibhih, and the corres- ponding passage ending with dngiraslbhih, refer to certain kinds of hymns. The meaning of the passage becomes clear in the light of the paribhdsd-sutra Kauc. 8. 1 6 : citi-prdya$citti-%$aml- * In the same manner the Vaitana refers to practices which have been de- scribed in its own preceding chapters. So 15. 15, dcamanadwiksanantam., 'he performs the rite which begins with the rinsing of the mouth and ends with the act of contemplation.' This refers back to Vait. 1. 19. f Noteworthy are three other passages, in which ritualistic practices are sketched in the same manner, by giving their initial and final traits; neither Garbe nor myself have found anything in either Vaitana or Kaugika which fills in the outline. They are Vait. 5. 12, akrtilostetyadyupasthanantam ; 10. 14, pafav anayai Hamityady anjanantam ; 16. 15, havir upiivahrta ityadivaifvdnaro 'gnistoma Uyantdbhih. Were there ever other ritualistic sa*ra-collections in the Atharvan schools? X Darila comments upon these words as follows : citih prayafcitih svacitih pra- yafcitih dvarhdvah tiibhywm evamvidhdbhydm varanena siddhibhumisthavartulapra- siddhe : metra prayafcittivifesah : trlni trlrji sarvani parvani ubhe tatamejakare. I give the text without emending even the most obvious errors. 388 M. Bloomfield. pamakd-savanpd-pdmyavdkd-taldpa-paMpa-vdpd-pinpapdrpimbala- sipuna-darbhd-'pamdrgd^krtttosta-valmlkavapd-durvdprdnta-vrlr hi-yavdh pdntdh. This is a list of articles, largely plants, which are considered as holy, and are therefore employed in the prepa- ration of the pdntyudaka, ' the holy water.' In the Vaitana they are called dtharvana, and are contrasted with the list following, which is called dngirasa. This does not occur in the Kaucika. In this case there can be no doubt that the Vaitana, in abbrevi- ating the first list, which does occur in the Kaucika, so as to say ' with the articles citi etc.,' and in giving in full the second list, which does not occur in the Kaucika,* confesses itself directly dependent on and later than the Kaucika. We have finally a technical proof. Both texts follow the usual practice of citing the hymns belonging to the canon of their own school by their pratlkas. The Kaucika moreover follows the current method of giving in full any hymns or formulae solennes which come from another pdkhd of the same Veda, as well as from the other Vedas. The Vaitana in general follows the same practice, with one very noticeable exception. Any hymn or for- mula which occurs in the Kaucika as well as in the Vaitana is cited with the pratlka only, it matters not whether the hymn in question occur also in some other samhitd, or is, as seems often the case, the special property of the Kaucika, not to be found in any of the existing samhitds. An example or two may illustrate this statement. The Taittiriya-samhita, iii. 2. 4. 4, has the formula : ahe ddidhi- savyo W atas tistha 'nyasya sadane sida yo i smat pdkatarah. This is cited in full in the frduta-sictra of the White Yajur Veda, Katy. ii. 1. 22. So also Kauc. 3. 5; but Vait. 1. 20 has only the pratlka : ahe ddidhisavya. Kauc. 6. 11 has a mantra, which Darila designates as kalpajd, and which I have not been able to trace in any samhitd: vi mun- cdmi hrahmand jdtavedasam agnirh hotdram ajaram rathasmrtam, etc. Vait. 24. 7 cites only the pratlka: vi muficdmi. * One may venture to state that the angirasakalpa would bring some explana- tion of these obscure terms. Compare also Gop. Br. i. 2. 18.