Rigveda

TRANSLATION AND NOTES


251


It hath filled heaven and earth and the sky. The sun is the self of all that stands
and moves.’ This I regard as the regular 8 Samhita as composed, thus says
Badhva. For the Bahvrcas consider him in the great hymn, the Adhvaryus
in the fire, the Chandogas in the Mahavrata rite. They see him in this earth,
in heaven, in the air, in the ether, in the wateis, in plants, in trees, in the moon,
in the constellations, in all beings. Him they call brah?nan. The self which
consists of sight, hearing, metre, mind, and speech, is like the year in number.
He, 7 who recites to another the self, which consists of sight, hearing, metre,
mind, and speech, and is like the year,

4 . To him the Vedas yield no milk; he has no part in what his teacher
has taught him. He knows not the path of viitue. A R.si says this also
(RV., X,.7i, 6), ‘He who forsakes the friend who knows his friends, 1 in
speech he has no part. What he hears, he hears in vain, he knows not the
path of virtue.’ This means that he has no part in what he has studied and that
he does not know the path of virtue. So a man who knows this should not*
lay the fire for another, nor sing the Samans of the Mahavrata for another, nor
recite the 6astras of that day for another. Only 3 may he recite for a father or
a teacher, for that is done for oneself. We have said 4 that this incorporeal
conscious self and that sun arc one and the same. Where these two are
separated, 5 the sun is seen like the moon, 6 its rays do not manifest themselves,

• All the above must be Badhva’s view, just as III, 2, 2 gave Kauntharavya’s views. The
following alludes to the fact that the Adhvaryu’s mystic speculations centre in the Agnicayana,
cf. Eggeling, S. B. E ., XLTTT, xxiv.

7 The section runs on in a way that cannot be early. V, I, 1 and 2 is precisely similar,
and the present section division must remain of doubtful (though early) date. The divisions of
the fsankhayana are similarly illogical. For the loc., cf. Dclbruck, A Hindi sc he Syntax , p. 205.

1 Sayana points out that Taittiilya Aranyaka, I, 3: II, 15, reads in this verse sakhividam ,
a point overlooked in Bloomfield, Vedu Comordancc , p. 7oo b . Sayan a’s reference docs tend
to show that he also wrote a Taiitirlya Aranyaka commentary, which on other grounds
might be deemed very doubtful (cf. Ill, 2, 3, n. 5).

3 i.e. act as Adhvaryu, Udgatf or Hotr priest. It is impossible to square the total pro¬
hibition here with V, 1, 5, which (see n. 5) contemplates a breach of the rule, but it agrees
with the opinion of‘some* (eke) in V, 3, 3, sec n. 1 on that passage.

3 A frequent exception. Cf. V, 3, 3, n. 1.

« III, 2, 3. The relevance of this passage is not obvious. Sayana takes it as a reflexion
induced by the idea of the attainment of brahman in the brief space of life, whence omens
as to the duration of life are inserted. The connexion of sun and self is elsewhere used to
give omens of death. In Brhadaranyaka Upanisad, V, 5, 2, the sun appears as white only
to the man about to die. The parallel passages in the Sankhayana are VIII, 7, and XT, 3 ; 4.

# This is not very logical, as there is no reason why the separation of the two should
be a sign of death. The rest of the signs are clearly old folkloie ideas pressed into service.
For the extensive literature on Vedic superstitions, see Hillebrandt, R itnaUI.it ter at nr,
pp. 167 sq., 1^3-185; Hatfield, Auianasddbhatani , J.A.O.S ., XV, 208, &c,; Bloomfield,



AITAREYA ARAN YAK A


III, 2, 4-


25 2

the sky is red like madder, the wind is not retained, his head smells like
a raven’s nest, and a man should know that his self 7 is gone and that he will
not have long to live. Let him do then whatever he considers must be done,
and recite seven verses beginning, ‘What is near, what is far ’ (RV., IX, 67, 21-27),
the single verse, ‘Of the ancient seed’ (RV., VIII, 6, 30), six verses beginning,
‘Where puiifying Brahman’ (RV., IX, 113, 6-11), and the single verse, ‘We
from the darkness’ (RV., I, 30, 10). Next when the sun is seen pierced, and
looks like the nave of a cart-wheel, or he sees his shadow pierced, let him know
that this is so. Next when he sees himself in a mirror or in the water with
a crooked head H or without a head, or when his pupils are seen inverted 9 or
crooked, let him know that this is so. Next let him cover his eyes and look ;
then threads 10 are seen as if falling together. If he sees them not, let him

Atharvaveda , pp. 82 sq.; Kau^ika Sutra, XIU, and Adbhuta Brahmana ; Aufrccht’s idea
(/. D. M. G., XXXII I, 573> that the passage is not in place is disproved by the parallel in the
Sfinkh^nna, VTII, 6 and 7 ; XT, 3 and 4.

8 1. e. its rays are pale and cold, h'dkakuldyagandhikam is probably an adj. as a quasi-
pred. For example's. cl. Dilbriick, A Itmdische Syntax, pp. 78, 79. Kulaya is a curious word:
in Manava Grliya Sutia, II, 14, 23, Jvnauer takes it (wiongly, I think) as = stall (cf. p. 55
of his edit.).

7 Annndatlrtha renders sampareto as \amnikrs[anigamah, Sayana as mrtah. In yat-
manyetei the opt. is probably indef. Tt may also be ‘attracted’, cf. Spcijer, Veiiisihe und
Samkr it-Syutax, § 281. The form in atuya is rare in the Brahrnnnas, cf. Delbruck, Altindisihe
Syntax, pp. 400, 401 ; Whitney, Sanskrit G> am mar, § 965. The u^e of man with participles
ol all sorts is cunous, cf. the use with the gerund, Whitney, § 994 e; Speijer, Vedisehc und
Sanskrit-Syntax, § 223; with the pics, part., Ill, 1, 4. With the past part., even in Bihaddevata,
c.g. VII, 125.

8 The leading of the test is supported by Sayana and also by Anandalirtha and is certain.
For water divination, cf. Farnell, Cults of the Greek States, IV, 230. For adaria (also in
the Brhadarauyaka and Kajha Upanisads), cf. Max Muller, S. />. E., XV, xxiv.

9 Sayana explains a white pupil m a black eyeball. It probably means only, upside down,
although the contrast of white and black in the eye is frequent, II. I, 5. &ank)iayana Aranyaka,
VIII, 7, suggests leading here jihme na vd , 1 or are not seen at all,’ and this may be right.

10 Sayana explains the opeiation thus, (aksusT nimllya netrasydpdngam avastabhya nctra-
samipam patyet ; Anandatiilha has, angulyd aksimfdam avastabhya. The batarakani fbardtakan
or vardtakan in Sank banana) arc, Sayana says, vartuldni suksmani Utklavar mint ketorypa-
ka iabdabh id It cyd ni, and he takes sampatantTva as satnyan net ran nirgachantiva. This is hardly
possible. For vardtakan, cf. &iihaisa, Khnndanakhandakhidya, p. 239, cited by Jacob, Lauki-
kanydydiljali, p. I. The construction is difficult, as til <2yathd is not properly m place. It may
be that yathd goes with batarakani and iva qualifies only sampatanti, and the sense is, things arc
seen like, &c., but it is also possible that tad yathd is practically = then it is that. This
use is of couise common in later Sanskrit, e.g. Bana, Kiidambari (p. 337, 12, ed. Peterson;
p. boo, ed. Nirnaya Siigara) : dgamesu sarvesu eva purd nard mdya nabhdratddisu samyag
anehaprakdrdh f dpavrttdh tad yathd, See. Cf. the Pali use of seyyathd. Bfhadaranyaka
Upanisad, IV, 3, 43 sq. has a series of tad yathd; so ibid., IV, 4, 4, 5, See.

Cases of conditional sentences without particles are of course very frequent in Vedic as



-Ill, 2, 4


TRANSLATION AND NOTES


253


know that this is so. Next let him cover his ears and listen, then there is
a sound as of a burning fire or of a chariot. If he does not hear that sound,
let him know that this is so. Next when the fire appears blue like the neck
of a peacock, 11 or when he sees lightning in a cloudless sky, or no lightning in
a cloudy sky, or in a great cloud secs bright rays as it were, let him know that
it is so. Next when he sees the ground as though burning, let him know
that this is so. So far as regards the visible signs. Then come the dreams. 1 ' 2
He sees a black man with black teeth, he kills him; a boar kills him ; a monkey
jumps on him; the wind carries him swiftly along; having swallowed gold
he spits it out; he eats honey; he chews stalks; he carries a single n lotus;

in other languages. C f. Speijer, Vedische u>td Sanskrit-Syntax, § 284; Sansktit Syntax , § 487 ;
Aufrccht, Aitareya Brahmana, p. 431; my note.y. R. A. S , 1909.

The Maitreya Upanisad (Max Muller, S.B.E., XV, xlvi) has a passage which may he
reminiscent of this text: agnir vaiivdnaro . . . iasyaigi gho^o bhavati yam (wrong reading
? yad) etat karrniv apidhdya irnoti sa y a doth' rarnisy an bhavati nainam ghoui m irnoti.

For upabdi , infra, which denotes literally the noise of going and is paiticulaily in place here,
cf. Aitareya Brahmana, IV, 9, 3; Jaiminiya Biahmana, I, 253; Jaimmiya Upanisad Brahmana, J,
37, 3, with Oertel’s note; R V., I, 74, 7, with Oldenbeig’s note (. 9 . B. E., XL\ I, 94); Schmidt, A'.
Z., XX V, 55. Schcftelowitz {Zur Slammlnldungin den indo-get manischen Sprat hen , § 9) compaies
R V., IX, 77, 4 : urub/d, which he considers as going back to Hi. pago , cf. (ireek ^77717. The con¬
struction above driyate and abhikhydyda in parallel uses, and below dt iyate-patym-na paiyen-
faiycta , arc decidedly curious (cf. Introd., p. 63). The temptation to amend to driyeta is very
strong, and on the whole I incline to think that it would be dangerous to insist on these examples.
The case of upeheta — driyante differs, for the two verbs are not parallel. The first is an
instruction, the second expicsses categorically the result (and driyante may have helped to
bring about the incorrect driyate). In TIT, 1, 4, where ugavadet and aha occur, the aha is
very strange, and one would like to take iaknosTty aha — hasyatity as two sentences both
dependent on bruydt. There is, however, the real difliculty that d—Vkd would be a strange
combination, and the di\ision of the sentences is also curious, though no more curious than
the aha. I suspect some corruption of the text. Say ana rendeis diffeiently. lie takes the
whole as one Mantra and supplies bhavdn as a subject for aha , and so in the next sentence
he interpolates bhavdn aha in sense. In the numerous passages in the Aitareya Brahmana
which are more or less parallel (sec the reff. cited in III, 1, 6, n. 5), no such aha occurs,
and hdsyati has no prefix. But probably d—hdsyati must go together. Aha might, of course,
be taken as a first person and made part of the quotation (cf. Speijer, § 178), but this is not
likely, and for the indef. opt., cf TII, 2, 1, n. r.

11 Mayuragrivdh is perhaps intended by the reading of B, mayuragrivd ameghe (but
{sankhayana has mayuragrivd when it can be °vdh); and undoubtedly grtvdh is the form
alone recognized by Banini and usual in the earlier literature, J. R. A.S., 1906, pp. 916-919.
Probably the reading was originally mayiiragnvdmeghe by an incorrect Sandhi for maytira-
grivdh. For similar irregular Sandhi, cf. Buhler, S. B. E., II, xli (from Apastamba) ; Macdoncll,
Brhaddevatd , I, xxvii; and V, 3, 2, n. 9; III, 1, 3, n. 2. For the next portent, cf. Pischel,
Vedische Studien, I, 112.

l * The plural must be right. Cf. Markandeya rurdna, XLIII, 1 sq.; Ilillebrandt, op.cit., p. 184.

13 ‘Red’ in colour (Sayana) ; for red as unlucky, ,cf. Z. J). M. XL, 117.



254 AITAREYA ARANYAKA III, 2, 4-

he drives with a team of asses and 14 boars; wearing a wreath of red flowers,
he drives a black cow with a black calf towards the south. 16 If he sees any
of these, he should fast and cook a pot of milk, and offer it, reciting a verse
of the RatrT hymn (RV., X, 127,16) to each oblation, and having fed the Brahmins
with other food, 18 himself eat the oblation. Let him know that the person within
all beings who is not heard, 17 not reached, not thought, not subdued, not seen,
not understood, not classified, but who hears, thinks, sees, classifies, sounds,
understands, and knows is his own self. 18

5. Now comes this Upanisad of the whole speech. All these indeed are
Upanisads of the whole speech, but this they so call. The jnnutes are the
earth, the sibilants the sky, the vowels heaven. The mutes are fire, the sibilants
air, the vowels the sun. The mutes are the Rgveda, the sibilants the Yajurveda,
tlie vowels the Samaveda. The mutes arc the eye, the sibilants the ear, the
vowels the mind. The mutes are the up-breathing, the sibilants the down¬
breathing, the vowels the back-breathing. Then comes this divine lute. 1 The


14 'Or' (Sayana), which may be more correct.

18 The ten dreams are so taken by the commentator and by Max Muller whose note (p. 262)
is apparently wrong. Eieulm kimeid is noteworthy. The neut. of the pronoun is practically
nominal and is to be compared with the neut. in predication, TIT, 1, 2, n. 4. So in Latin,
e. g. Horace, Sat., i, 7 : Lydorum quiz-quid. The parallel passage in the 6ankhayana has corrected
the original kimeid of the MS., but the correspondence is conclusive.

18 Cooked in the house (Sayana). See Sankhliyana Gfhya Sutra, V, 5, 9, and my article,
J. R. A. S., 1907, p. 929 ; for sthalipdka , see Brhadaranyaka Upanisad, VI, 4,19; Gfhyasamgraha,
1 ,114 ; Oldcnberg, S. B.E, XXX, xvi, n. 4. For the causative with instr. and acc., cf. Pelbnick,
Altindische Syntax, pp. 224 sq.; Whitney, Sanskrit Grammar, §§ 277 a, 282 b; Speijer,
Vedische und Sanskrit*Syntax, § 21 ; Sanskrit-Syntax, § 49. According to Panini, I, 4, 52,
and the examples cited in the Kasika Vrtti, ad loc., here we should have two accusatives.

17 A tab is rendered by Sayana, asrnad dehendriyddisahghatad vilahana iti iesah, while
Anandatlrtha suggests adhikah.

18 This is the most advanced point in the definition of the Atman arrived at in the Aranyaka.
The Atman is not object, but subject only—as Sayana says, dtmd visayo na bhavati visayi tu
bhavaty eva. This occurs frequently later and with it the docliinc that the self cannot be
known. Sayana cites the antaryamibrdhmana , Brhadaranyaka Upanisad, III, 7, 13, the
aksarabrahmana, ibid., Ill, 8, 11; the Kausltaki Upanisad, I, 8; the Pra&na Upanisad, IV, 6;
and the Nfsirpbottaratapaniya Upanisad, II. See also Deussen, Philosophie der Upanishads,
PP- *33 S< 1 * > E.T., pp. 147 sq. Jaiminlya Upanisad Brahmana, IV, 18, is devoted to this topic
(= Kena Upanisad).

1 i. e. the human body. This metaphor explains Prafena Upanisad, II, 3, where vdna (V, 1, 4)
is equated to iarira , which Max Muller (S. B. E., XV, 374, n. 3) finds unintelligible. Connected
with Visnu is Anandatirtha’s explanation of the word daivt. Ambhana is a curious word.
I think it is from anu+*/bhan (as in Class. Sansk. for Vbhan, Wackemagel, Altindische
Grammatik, I, 194). Compare ambara for anu + vara and jdmbila for jdnu + bila (ibid., 59).
The omission %1 before v (common) led to omission before b and sporadically before bh. The
meaning would be ‘sounding-board’ (’). Cf. v. Schroeder, Ind. Lit., p. 755 *



-HI, 2, 5


TRANSLATION AND NOTES


255


human lute is an imitation of it. As there is a head of this, so there is a head
of that; as there is a stomach of this, so there is a cavity of that; as this has
a tongue, so that has a tongue; as this has fingers, 2 so that has strings; as
this has vowels, so that has tones; as this has consonants, so that has touches;
as this is endowed with sounds and firmly strung, so that is endowed with sounds
and firmly strung; as this is covered with a hairy skin, so that is covered with
a hairy skin. For in former times they covered lutes with a hairy skin. lie, who
knows this divine lute, is heard when he speaks, his fame fills the earth, and
wherever they 3 speak Aryan tongues, there is he known. Then comes the essence
of speech. When a man reciting 4 or speaking at an assembly gives not pleasure,
let him recite this verse, 1 May the she-ichneumon, that rules all speech, who is
covered as it were 6 by the lips, surrounded by teeth, the thunderbolt, cause me
to speak well here/ This is the essence of speech.

2 The words ah^ulayah and tantrayah seem to have been transposed in the original; they are
in conect order in Sankhayana Aranyaka, VIII, 7. Somewhat analogous is the tiansposition of
Sand jarayu in Satapatha Brahmana, VI, 6, 2, 15, on which see Fggehng’s note ( S.B.E
XLV 1 , 255'. Cf. also Brhadaranyaka Upanisad, III, 1, 4 with Max Muller’s note (6’. B. E ,,
XV, 122), and my Sankhayana Aranyaka , p. 55, n. 3.

3 The expression dr yd vdcah was not understood by the commentators (and in the Sankhayana
Aranyaka, VI IT, 9, we find that it has become dry a vdg vadati), who take drydh as nominative
and render it vedaiastrapdram gatdh. This is a clear sign of considerable antiquity, and the
expression may also be cited as an early piece of evidence for the existence of several
dialects of the early Indian language, which we know must have existed; see T, 5, 2, n. 19;
Oeitel, A. J. /*., XX, 447 on daivT, and Kathaka Sntpkita, XIV, 5. For the word arya, cf.
Zimmer, Allindisrhrs Lcben , p. 214; Pischcl, Z . ]). M. G ., XL, 125 ; Geldncr (I'cdische Studien ,
111 , 96, 97) insists that arya cannot mean ‘ the Aryan ’ which is represented by arya, Oldenberg
(see index to .S'. B. A’., XLVI) still adopts the equation Arya — Aryan.

* Sayana distinguishes between reciting at a conclave of priests, and speaking in a prince’s
hall. Virurucuseta is quite impossible as a form, and it is an easy eiror in view of the preceding
syllables, each having u. The middle of the opt. of the desiderative is not common. Cf.
Iloltzmann, Grammatisthcs aus deni Afakdhhdrata , p. 42.

5 Sayana gives an alternative rendering, na~ not, and pavih — clear, the subject being the
speaker’s defective speech. Anandatlrtha gives only the explanation as na «=» iva. The verse in
B occurs among the &anti veises of the so-called tliiid Adhyaya. For the metaphor, cf. Jaiminiya
Upanisad Brahmana, III, 19. In the version in the Ananda 4 rama ed., p. 2, nakult is printed
as a separate word. But rtakuIT can only mean a female ichneumon, and nakulTdantaih
is a phrase for which no parallel seems readily forthcoming. Sayana gives vajravaddham-
bhiitair autardlaihidrarahttair which does not help. In any case to join kulTdantaih makes
a curious though not unparalleled metre in an early verse such as this must be, and if a nom.
could be found in kuli the run of the veise would be much improved. The rendering of
the text by Max Muller ‘surrounded by birth, as if by spears’ is purely conjectural, and
I suspect the tradition. The parallel passages aie of little use. The Sama Mantra Brahmana,
I f 7, 15, has osthdpidhdnd nakult dantaparimitah pavih , while the Gobhila Grhya Sutra, III, 4,
29, gives osthdpidhdnd nakult only. Oldenberg (A. B.E., XXX, 84) renders ‘the she-ichneumon,
covered by the lips’, as does Knaucr in his translation. If this is to be made into sense, it



256 AITAREYA ARANYAKA III, 2, 6 -

6. Now Krsnaharita 1 proclaims this Brahmana 2 as it were regarding speech
to him. 5 Prajapati, the year, 4 after creating creatures, burst. He put himself
together by the metres. Because he put himself together by means of the metres,
therefore is it the Samhita. Of that Samhita the letter n is the strength, the letter
s the breath, the self. lie who knows the verses in the Samhita and 5 the

letters n and s, he knows the Samhita with its breath and its strength. Let him

know that this is lifegiving. 6 If he is in doubt 7 whether to say it with an n
or without an n, let him say it with an n. If he is in doubt whether to say it
with an s or without an .9, let him say it with an s. Hrasva Mandukeya says, ‘ If
we repeat the verses according to the Samhita, and if we say the teaching 8 of

must be taken that the she-ichncumon is a synonym for what is very piercing: the nearest
approximation to this idea is the passage in Atharvaveda, VT, 139, 5 (cited in Zimmer,
Altindischcs Leben , p. 86), which rcfcis to the ichneumon’s (m.) skill in chopping up and then
icstoring his work.

1 A son of I Tarda, who was dark in colour (Sayana), cf. TTiranyadant Vaida, II, 1, 5.

A Kumara Ilaiita (so, not Tfaiita) nppears in Brhadiiranyaka Upanisad, TT, 6, 3 ; IV, 6, 3 ; VI,

4, 4. Weber {Indian Titeratiue , p. 50) reads Ilarlla, and the lawyer is always so called (ibid.,

р. 269), even in Apastnmba Pharma Sfitra, I, 10, 29, 12; 16. On the other hand Varttika 8
on Pfinini, I, j, 73, recognizes Ilaritakata, and Panini, IV, 1, roo, Harita^ana as names, where
ITarita appears. Weber’s Ilarita here is therefore probably wrong, and Sankhayana Aranyaka,
VITT, ij, has krtsnahdnta .

a Brahmana here means secret doctrine like Upanisad. Iva seems to be used to indicate
the somewhat unusual sense; the Sankhayana version has eva ; cf. I, 1, 3 , n. 3; f. R. A. S.,
1908, p. 1193, n. I. Sayana in his commentary repeatedly has phiases like antary dmibrdh-
tnana , the secret doctrine of the antarydmin , sec III, 2, 4, n. 18, and cf. the name of
JIrhadaran)aka Upanisad, I, 4 ( purusavidhabrdh?nana ), Max Muller, S.B.E., XV, 25, and
the common toy ok tarn brdhmanam.

3 To his pupil or son (Anandatirtha and Sayana).

4 The rending of B, satnvatsaram (see Introd., p. 3), must be a correction to improve the
sense. But it could never have been corrupted into samvat sarah. Piajapati as the year is a
Brahmnnic commonplace (for its deeper significance, see Fggeling, S. B. E. y XI.III, xx sq.),

с. g. Aitarcya Brahmana, II, 17, 2; VT, 19, 7; Maitrayani Samhita, I, 10, 8; Kausltaki
Brahmana, VI, 15; Sankhayana Aranyaka, T, I, See. The phrase Prajdpatih prajdh sfsfvd
vyasramsata is fiequent in fsatapatha Brahmana, Vl-X, not in I-V ; Weber, Ind. Stud., XIII,
268 ; and for a similar case cf. II, 4, 3, n. 14. One might translate, ‘he is the year.’ Cf.,
however, Satapatha Brahmana, X, 1, 1, 1 and 2. The confusion of vyasramsadd and °sata is
another example of the confusion of surd and sonant so common in £arada MSS. Cf. Lanman
in Whitney’s Translation of the Atharvaveda , pp 57 j 10 45 > J* Hertcl, Tantrdkhyayikd ,
p. xvi; Roth, Z. D. M. G., XLVIII, 106-m.

5 This is the literal rendering. Sayana takes it, * Who recites the verses thinking of the n and s
which accompany the Sarphita.’

8 To the Samhita (Sayana), or peihaps to the reciter, if not to both.

7 Sayana takes it, ‘ If a pupil ask his teacher,’ but this is unnecessary. The question is, he
says, whether the reflection on the Samhita is to take the differences of n and s into account or not.

8 Sayana refers this to Suravlra’s doctrine, TII, 1,1. Por updptau, cf. Kausltaki Brahmana,
XIV, 5 ; Sankhajana Aran)aka, I, 6, \ihcrc Ur. Friedlander renders ‘ hinrcichend, gemigend



-Ill, 2, 6


TRANSLATION AND NOTES


257


Mandukeya, then the letters n and s are obtained for us.' Sthavira Sakai) a'*
says, ‘ If we icpeat the verses according to the Samhita, and if we say the teaching
of JVIandukeya, then the letters n and s are obtained foi us/ Then the seers,
the Kavaseyas, knowing this, 10 say, ‘To what end shall we repeat the Veda,
to what end shall we sacrifice ? For we sacrifice breath in speech, 11 or in breath
speech. For what is the beginning, that is the end.’ These Samhitas let no
one 18 tell to one who is not a resident pupil, who has not been with the teacher
for one year, and who is not himself to become a teacher. Thus say the
teachers. 13

,J The sayings are identical, and apparently this is intended to denote that the doctrine
received universal acceptance. The passage may indicate (cf. also Saiikhayana Srauta Sutra,
IV. *0, 3, where Sakalya is younger apparently than Mandnkeya) that the Mandukeya Sakha had
its Sniyihita text before Sakalya produced the Pada lM(ha, which is cpiite likely.

10 This is a clear proof that the holders of the Aranyaka doctrine icjected sacrifices or
recitations as means of knowledge, cf. Brhadaraiiyaka Upanisad, I, 5, 23 ; Kausitaki Upanisad,
II, 5 ; Chandogya Upanisad, V, 11-24 ; Taittiriya Upanisad, II, 5 ; Deussen, Phil. d. Upanishads ,
p. 63. A Tura Kavaseya purohita of Janamejaya occurs in Khila, I, 9, 6, and in—as already
noted by Colebrooke, Essays, I, 72; see Oldenbeig, Z.D.M. CL, XL 1 I, 239 stj.—the Aitaieya
Brahmana, IV, 27 ; VII, 39 ; VIII, 21. For the spelling cf. Scheftelowitz, Pie Apokryphcn dcs
Rgi'tda, Addenda, p. 190; Wackcrnagel, Altindische (Jrammatik, I, 239. Wintcrmtz (6 'esi/i. dcr
indisc/i. Litt. % I, 199) uses the stoiy of Kavasa as the son of a non-brahmin (Aitarcya Brahmana,
II, 19) as a piece of evidence in favour of the theory of the attribution to the Ksatriyas of
philosophic speculation over the origin of the doctrine of transmigration (cf. Introd., pp. 50,
51; Garbe, Licit rage zur indiuhen Kulturgcsihichie , pp. 1 s<p). He argues that the Brahmins
merely accepted and made these doctrines their own by adopting them along with the doctrine
of the four Asiamas. This all seems very doubtful. That among the piicsls none should
rise superior to the sacrificial cultus is contiary to all religious history. That hermits, ike.,
weie originally not of the priestly caste is a mere theory and not a piobable one. Winternitz*
view leads him (p. 202, n. 1) to adopt the improbable theory of Aranyaka as a text to be
studied by Vanaprasthas, for which he cpiotes the (late) Aruneya Upanisad (Deussen, Sei/izig
Upanishads, p. 693) and Ramanuja (Thibaut, XLV 11 I, 645). Cf. Intiod., p. 16.

It must always be remembeicd that the Brahmanas contain alieady in germ all the ideas
which make up the fundamental doctrine of the Upamsads; even the doctrine of trans¬
migration is presaged in the doctiine of repeated deaths in the other woild. It is impossible
to explain why the Brahmins became so completely the bearers of the at man doctrine if it
was not theirs ex initio. Professor Macdoncll has told me that he concurs in this view, which
thus gains gieat weight, and sec my notes, J. K. A. .S'., 1908, pp. 838, 868, 1142. The Kavase¬
yas are cited by Sankara on Svctasvatara Upanisad (ed. Roer, p. 257) as opposed to works,
Weber, Ind. Stud., TT, 418.

"^Cf. JaimniTya Upanisad Brahmana, I, 2, 2, 6.

13 Cf. V, 3, 3 ; Weber, Indian Literature, p. 49, n. 35.

13 Mahidasa, See. (Anandatiitha). Cf. 1 , 1, 1, n. 5 ; II, 3, 5, n. 4. Probably the plural is
only maid tat is.


KEITH


S



258


AITAREYA AR ANY AKA


IV-


ARANYAKA IV

ASvalayana (Srauta Sutra, VII, 12, 10) gives the following account of the
purpose of the Mahanamni verses. On the fifth day of the prslhya six day
ceremony, at the midday pressing of the Soma, corresponding to the Ni^kevalya
6astra, the Udgatrs sing sometimes the &akvara Saman as one of the Prstha
Slotras, 1 and then 2 use the Mahanamni vcrj.es as the basis of the Saman. These
number nine, but for the purposes of the Saman they arc made into three, each
consisting of three verses. These verses are recited adhyardhakdram , that is,
first one and a half verses are recited, then comes a pause, then the remaining
one and a half, followed by the syllable orn. Then are recited the nine putisa-
paddni , additional verses. These may either be recited simply straight on as
they stand in the text, or the first five may be made into two sets of five syllables
each, thus:

Eva hi eva I tvd hi Agnd 3 u \ the hi being taken without Sandhi, the last four
purhapaddni being repeated without a pause in the middle. See also §ankhayana
Srauta Sutra, X, 6, 10, and comm.

The MahanamnI verses occur in the Aranya Samhita, and in the Naigeya Sakha
at the end of the Purvarcika of the Samaveda, and as one of the Khilas of the
Rgveda, see Peterson, Second Report , p. 97, Scheftelowitz, Die Apokryphen des
Rgveda, pp. 134-136. They are referred to in the Brhaddevata, VIII, 100,
iknkhayana Srauta Sutra, X, 6, io, Rgvidhana, IV, 25, and Sankhayana Grhya
Sutra, II, 11, 12, &c. From these sources, and from Baudhayana, cited in
Oldcnberg, Prolegomena , p. 509, n., it appears that they followed directly upon
the verse tnc chantyor, which, according to the Saiikhayana Grhya Sutra, IV, 5, 9,
is the end of the Rgveda Samhitfi (in the Bnskala recension), and, according
to Narayana on Asvalayana Grhya Sutra, III, ,9, 9, is the end of the Baskala
recension. 3 It is not, however, quite clear what this means, since lac chant yor
occurs as the last verse of two Khila«, V, 1 and 3, in Scheftclowitz’s edition,
viz. the satnjhdnam and prddhvardndm Khilas, and the three Khilas, V, 1—3,
the second being the nairhasiyam y have S + veises. 1 he view of

1 For these, see especially Eggeling, S. B. E. , XLI, xx sq.

1 The Silk vara is normally based on Samaveda, II, 1151-1153 (Sayana and Mahulhara cited
by Egfiehng, p. xx, n. 2).

5 Cf. also Oldenberg’s note on Sankliayana Grhya Sutra, IV, 5, 9, and Ind, Stud 1 , XV, 150.



-IV


TRANSLATION AND NOTES


2 59


Oldenberg, who had not 4 5 * the evidence of the MS. of the Khilas before him,
was (.Prolegomena , p. 502) that the Samhita ended with the first tac cham yor ,
i. e. with Khila, V, 1, and Scheftelowitz (pp. n, 132) holds that this is correct.
Oldenberg, however, held (p. 509) that the Mahanamni verses followed directly
after tac cham yor, and (p. 501) expressed the view that the following ten verses
were some of them modern. But of the direct evidence for the immediate
sequence of the Mahanamni verses, cited by Oldenberg, the Rgvidhana alone
fully bears him out, for the Khila MS. has the Mahanamni verses after the
pradhvarandm Khila, and this is probably the meaning of Brhaddcvata, VIII, 94,
as interpreted by Prof. Macdonell. It is an easy conjecture that the Rgvidhana,
which has other coincidences with the Brhaddevata®, followed that work, but
misunderstood the word caturtham , which most probably must mean ‘ the fourth
of the hymns after X, 190'. This fact weakens greatly the force of Oldenberg's
argument from the modern character of the last ten verses, and in point of
fact it is difficult to deny that the verse tac cham yor is modern in appearance,
and that it need net be separated in time fiom the last seven verses. For the
second tac cham yor being the end of the Samhita in the Baskala recension,
we have the clear evidence of the commentator on the Caranavyuha/ who
actually cites the verses. Dr. Scneftelow'itz considers that the commentator is
untrustwoithy, and later than Sayana, but this appears very doubtful. We know,
he argues, that the commentator explains the eight extra hymns attiibuted to
the Ba ; kala Sakha by the AnuvakanukiamanI as being seven of the Valakhilyas
and the samjitdnam hymn of fifteen verses, but the number should be ten, as
the samjhdna?n hymn is really composed of three hymns. But it is difficult
to maintain that it is impossible that the fifteen verses, despite their difference
of contents, were not regarded in early days as one hymn, for several of the
Rgvcdic hymns are notoriously patchwork, and this applies moie strongly still
to later SamhiLis.

Much more important is the question of their antiquity. Oldenberg makes
the Mahanamni verses an exception to his general view, that the Khilas arc
on the whole of later origin, and holds that they are coeval with the Rgveda,
and were merely omitted because of some reason of ritual teaching from the ten
Mandalas. Dr. Scheftelowitz, who disputes Oldenberg’s general position, and
accepts Hillebrandt’s theory of the purer ritual tradition, assigns the verses
(p. 3) to the end of the Rgvcdic period. Further, Oldenberg 7 has suggested

4 He takes no notice of the new evidence in his review of Scheftelowitz, Gott. gel. Attz
1907, p. 227, for which and for other valuable papers I am indebted to his kindness.

5 Macdonell, Brhaddcvata , I, 147. * Oldenberg, Piolegomena, pp. 495, 501, 502.

7 XXIX, 156.

S %



26 o


AITAREYA ARANYAKA


IV-


that the verses are alluded to as the SakvarT verses in Rgveda, VII, 33, 4; X, 71,
11, and this suggestion is at least plausible. They are apparently referred to
as Mahanamnis in the Atharvavcda and Yajurveda (see below). It is borne out
to sonic extent at least by the character of the language, which shows the rare
forms dnuiamsisah , s/use, vide , tse, rnjdse, and samnyase. The metre is also
of an archaic t}pe in so far as resolutions are frequently necessary to restore
it. The Khila AnukramanI gives the following note: viJd dasa pCidas ca pafica
Visvamitra Indro va Prajdpaiir Aindram pnvanam dnustubham purisapaddny
Agneyavaiinavaindrapawnadaivdni vairajani dvifyapancamydv usnihau caturthi
nyahkusdrini s apt ami purastddbrhati navamyantye pahkfi . As a matter of fact,
as both Weber 8 and Oldenberg recognize, the verses are not preserved in their
primitive form, but only as modified to suit their supposed sacred character.
In verses 2, 5, and 8, which were apparently originally anuxtubhs , the fourth pada
has been omitted for the insertion of a sort of refiain. Verses 1, 3, and 6 are
in anmhibh. Verse 4 appears to be 8 + 12 + 8 + 8; verse 7, 12 + <8 + 8 + 8 ; verse 9,
8 + 8 4-8+8 + 8. The rest is in no regular metre. Oldenberg (p. 33) considers
that originally the metre consisted of seven and five sets of eight syllables
respectively, but this seems hardly borne out by the facts. It should be noted
that the Khila text manufactures the last four of the nine punsapaddni into one
verse ('), and in this respect is certainly not old, for the punsapaddni cannot
reasonably be held to have ever made up a verse. They are referred to, however,
as five in the KauMtaki Biahmana, XXIII, 2, and connected with Prajapati, Agni,
Indra, Pusan, and Dcvah, and in the Brhaddevata, VIII, 102, they are connected
with the same deities, save that Visnu is substituted for the Dcvah (so the A
version; the B version omits Prajapati, while Mitra’s text includes both Prajapati
and the Dcvah, see Macdonell’s note). They are also mentioned in the Panca-
vimsa Brahman a, XIII, 4, 12, where claboiate directions arc given as to their
selection to make up the Sahara sdman, Latyayana Sutra, IV, 10, 18, Sankhayana
Srauta Sutra, X, 6, 13, &c., and in the Aitarcya Brahmana, IV, 4; V, 7; VI, 24;
Atharvavcda, XI, 7, 6; Vajasaneyi Samhita, XXIII, 35: Kathaka Sarahita,
X, 10; TaittirTya Samhita, V, 2, 11, i. 9

The verses contain several phrases reminiscent of the Rgveda, perhaps bor¬
rowed from earlier hymns; at least they tend to convey an impression of second¬
hand use: jet a ram apardjitam — RV .> I, 11, 2; sd nah par sad dti— RV., X, 187, 1;
Indram dhdnasya sditdye is the last pada of RV., VIII, 3, 5' 1 (this 1 owe to Bloom-

8 Ind. Stud., VIII, 68.

9 For the last four reff. I am indebted to Bloomfield, Vcdic Concordance, p. 696*, who
gives other passages; cf. also Weber, Ind . Stud., XVII, 358; Fggeling, S. B, E ., XLI, xx;
XLIV, 380, n. 2.



-IV


TRANSLATION AND NOTTS


261


field, Vedic Concordance , p. 210^); sdm anye'su bravavahai— RV., I, 30, 6; sdkhCi
susdvo ddvayah — RV., I, 187, 3*1; lav is t ha vajrinn rhjdse — RV., I, 80, i c (with
ojdsd). These last two cases seem to me strongly in favour of the later date
of these verses, for bravdvahai is not unnatural in RV., 1 , 30, 6, uhcre it seems to
refer to Indra and the speaker who are to agree in other battles, the previous
half verse referring to a conflict, but it is distinctly awkward here where the
first half verse has no reference to a fight or other occasion of association. This
only, however, proves that the MahfinamnI verses are not among the earliest parts
of the Rgveda. 1

The last four purisapadani are made out of the preceding verses, evti In sakrd ,
from v. 2 ; vast hi sakrd, from v. 5 ; vdsan dnu , from v. 4. The Asvalayana Srauta
Siitra, VI, 2, 9, shows that other padas of the verses were used independently
in the ritual: prace/ana praietaydyahi piba maisva I kra/us chanda rtam brhat
sumna a dhehi no vasav iiy anustup I Ibid., 12, has: ud yad bradhnasya vistapam
iti pandhamyCi I evd hy evaivd hiudra 3 | era hi sakro vast hi sakra Hi japitvd I apah
purvqdtn harivah sntdndm i/iyajati \ and again the purisapadani in VI, 3, 26.

hor the question of the ‘authorship 7 of this Aranyaka by Asvalayana, cf.
Introd., pp. 18 sq. For the view that this forms a sort of Asvalayana Samhita
may be compared the fact that there is an ApastambTya Mantiapatha, a collection
of Grhya verses and formulae, to accompany the Apastamba Grhya Sutra. So
too, as Oldenberg (S.B.T., XXX, 3-11) has conclusively 10 shown, the Mantra
Rrahmana was prepared to accompany Gobhila’s Grhya Sutia, though it is not
apparently ascribed to Gobhila, just as IV is not attributed to Asvalayana in
the Aranyaka itself. Winternitz (Gesch. der indisch. Li//., I, 232) merely repeats
Max Mtiller (Aneient Sanskrit Literaitire, pp, 3T4sq., 339).


O generous one, show 1 us a path, proclaim the regions, guide us, lord of
many mights, wealthy one 11 1 n

With these aids of thine, wise one, make us wise, for glory and for stiength,
Indra. For thine is strength 11 2 11

For wealth, for might, thunderer, most powerful, bearer of the bolt, thou


10 1 d< ^ n0t consi(1cr Winternitz {Mantrapat ha, I, xxxi sq.) to have refuted Oldenberg.

1 vida is rendered vet si by Sayan a, and S takes it as a Vedic form of vida, i. e. imner.
of the aor. of y/vid (Whitney, Sanskrit Grammar, § 851). Possibly this is correct (cf. vide in
ver. 5), and it is from y/vid in the sense ‘find*, for which see the examples in Bloomfield
Vedic Concordance , pp. 866 b , 867V But it may perhaps be really viddh the subj. of the aor. of
V vld (Whitney, § 849) or an injunctive from vi + Vdd. The accent would then, however,
probably have been viddh, but exceptions are not unknown. The same question arises in RV *
IX » 4 °» 3 • vidafy sahasrlnir Isah. For the accent, pCu-vbidm, cf. Whitney, Sanskrit Grammar,
§ 319- *or tad, of. Macdonell, Vcdtc Mythology, pp. 58, 122 ; Pischel, Vcdisthc Studien, 11
i,n.; Oldenberg, Religion dcs Veda, p. 239, n. 6.



262 AITAREYA AR ANY AKA IV-

movest. 2 Thou movest, most generous, bearer of the bolt. Come hither, drink,
and be glad ii 3 ll

Grant us wealth with good heroes. Thou art 3 the lord of might according
to thy will. Thou movest, most generous, bearer of the bolt, who art the most
powerful of heroes ll 4 II

Most generous of givers, wise one, guide us aright. Indra finds 4 all. Him
I praise. For he has will and strength ll 5 ll

Ilim we summon to our aid, the conqueror, unconquered. May he convoy
us 6 beyond our foes. He is strength, resolve, and mighty order ll 6 ll

Indra we summon for the winning of wealth, the conqueror, unconquered.
May he convoy us beyond our foes. May he convoy us beyond our enemies 0 ll 7 II

2 riljase may be regarded ns the second singular pres, indie, of a sixth class root rflj,
as Whitney (Sanskrit Grammar, § 758 a) takes it here. The exact sense is doubtful. It may
conceivably = ‘ thou ait piaised’, but the sense ‘move* is possible, if the root is akin to the
Greek bpiyu. Cf. Delbriiek, Altindisihcs Verbum, p. 181; Barlholomae, Indog. Forsch., II,
281; Neisser, Bczz. Bcitr , XX, 39; Olden beig, S.B.E. , XLVI, 39b, 436 (‘press on, strive
forwaid’); lhschel ( Vediuhe Studieu , I, 109), however, compares saraj with bpiyu, and Geldncr
(ibid.. Ill, 29 scj.) postulates a V'rj-iubh: dlptau , either transitive or intransitive. lie does
not, unhappily, quote or explain this passage. In KV., VIII, 9, 17 be renders vimi tvd
Fit uni rfijasl as ‘ I desire to adorn thee’, and possibly the form rfijasl might be an infin. ■=
an impuative (cf. Delbruck, AUindische Syntax, p. 412; Neisser, Bezz. Bcitr., XX, 59;
Hopkins, A.J. P., XIII, 21 sq.; Speijer, Vcdische und Sansknt-Syntax, § 216 d). The
accentuation piba mdtsva seems most probable, cf. tardnir ij jayati ksjii pusydti in RV.,
VII, 32,9, and other examples given in Delbmek, Altindischc Syntax , pp. 36 sip; Whitney,
Sanskrit Grammar, § 594 b; Sptijcr, Vcdische und Sanskrit-Syntax, p. 80; Macdonell, l ’edit
Gi ammar, p. 105. mdlsva is irregularly accented, but there are many parallels, \\ hitney, § 62S;
Macdonell, p. 99 (foot).

■ bhAvail is according to Whitney ( Sanskrit Grammar, § 83 b, c; cf. Delbruck, l.c.,
p. 144) either an injunctive of an unaugmented a aorist, or a subjunctive of the root aoiist.
But in sense it may be an indicative. vAiaTi Ann may perhaps be ‘according to our will*.
raydh suvtiyam is cuiiotis, but the variant ray I is merely an easy correction. Cf. ray As
pouim, KV., IV, 40, 4. The TaittirTya Samhita, III, 1, 9, 4 has: vider gaupatyam rayas posam
suviryam samvatsaiinam svasiim , where the conjunction of rayas and suviryam is different,
but where vider supports the derivation of vida from *>/vid. Cf. V, 1, 6, n. 3.

* vide must be 3rd sing, like Tie, and may mean ‘knows’, cf. Hopkins, J. A. 0 . S., XV,
376, n. Sayana renders it as a 2nd sing. For stun see Whitney, Sanskrit Grammar , § 894 d;
Delbruck, l.c., p. 181. If stusj is read, the accent is somewhat irregular. But irregular
accents in quasi-subordinate clauses are numerous, cf. Whitney, Sanskrit Grammar , §§ 595—
598; Delhiuck, AUindische Syntax, p. 43; KV., I, 189, 3; III, 1, 1, with Oldenberg’s notes
{S. B. E., XLVI, 182, 223); Z.D.M.G, LX, 735 sq.

6 Sayana lakes ati par sad as ‘ let him destroy’, and the last pada as meaning, ‘the sacrifice,
the metre used, the fruits of the offering, and all great.’ The words aic clearly not in place
here, and make little sense.

8 sridhah Sayana explains as those whom wc should hate, although they do not hate us.
The meaning is perhaps ‘beyond all failures’; cf. Ati sridhah in this sense in KV., I, 36, 7 ; 111 ,
9 > 4 i 10. 7 -



-IV TRANSLATION AND NOTES 263

Place us in thy favour, ancient one, lord of the thunder, bright one. Most
powerful, thy rewards are extolled. For the strong god bears rnle 11 8 II

Lord of man, slayer of Vrtra, this new hymn 7 I offer now to thee. Among
others let us two converse together. The hero who fares for the cows is a
kind and guileless friend 11 9 11

Thus, 8 thus, O Agni. Thus, thus, O Indra. Thus, thus, O Visnu. Thus, thus,
O Pusan. Thus, thus, O Gods. For he is strong. For he has strength and will,
according to his will. On all sides 9 come hither. Show, generous one, show.

7 This is doubtful, mduyase , the variant of the other texts save SV., is icmarkable as being

accented, and does not help. It looks like an obvious error or correction for sdmnyase, which
becomes sdnnyase , SV., Naigcyn Sakha, and then by haplography sdnyase , SV., Arnnya Saiphita,
and then mduyase through the frequent mistake of s for m in Saiadfi MSS. mdnyase makes no
good sense, but sanmyase also is very difficult (even if taken as Oldenberg (.S’. B. E., Xl.VI,
404) would take it in RV., V, 17, 2, as a first person). It comes appaiently frpm as .
Dr. Scheftelo\Nitz now ngiees with this view (cf. my remark 111 J. 1 \. A. S., 1907, p. 224).
For tarn tan (i. c. tad) can be lead (supply siiktam or, with tarn, man tram) but navy am may
be fiom meaning piaiseworthy. The dual bravdvahai in the oiiginal context refers to

the singer and Indra who are in other (contests) to be united. Ileic it must (cf. 11. 7 on I,
1, 2) mean something of the same soit, but anye\u has no longer any direct antecedent.
SV. aryd\u is merely a facile correction like so many SV. leadings. For the loc., gd}U, cf.
Dclbiiick, Altindisihe Syntax, p. 122 ; Speijer, Vedhche und Sanskrit-Syniax, § 81 b; Whitney,
Sanskrit Grammar , §§ 301, 304 ; A.J. XIII, 284. Sdnyase as a dat. hardly makes sense.

8 Saynna takes evd as from \li and d. The sentence is practically a mere exclamation
and cannot be translated. The words a, yd , &c., yield no sense as they stand. Say ana renders,
* He who comes to think what is to be thought of for our weal, let him come to think what
is to be thought of/ The variant dyo is no help, though it might mean * Come to the man
who deserves favour’, cf. Taittirlya Samhitd, II, 1, 3, 2. For the pluti, a3 z, see Wackemngel,
Aliindischc Gram mat ik, I, 298.

9 Cf. &ankhayana &rauta Sutia, XVII, 12, where the sentence runs: city evd hindropehi
viivatha vidd maghavan vidd tit , from which it may be legitimate to assume that chi should
be supplied in the purlsapaddni. The last vidd may point to viddh being the form, vilvadhd
in RV. means cither (1) everywhere, I, 14T, 6; (2) always, V, 8, 4.

The Taittirlya Aranyaka, I, 20, has : evd hy tva I evd hy Ague I evd hi Vdyo I evd hhtdra I evd
hi Pusan \ era hi devdh 1 when Sayana rendcis eva as ayanaUldditya and era as ctavydh
prdptavydh kdmd/t, and supplies asi, ‘Thou art desires:' hiUibdcnddttyasya sarvakdm a h etu tva -
prasiddhir ncyate. Ibid., 23, has : evd hy evt'ti I . . . evd hy Agtta Ui I. . . evd hi Vdyo Iti I . . .
evd hindreti I . . . evd hi PiUann Iti 1 evd hi dev a Iti I The accents are those of the
Anandasrama text (I, 88, 89), and may be wrong. In the Maitrayanl Snmhita, II, 3, 18
(a reference which I owe to Bloomfield’s Vedic Com01 dance, p. 305*) all the MSS. have evd
(or evd) hy Ague. The Kausllaki Brfihmnna, XXIII, 2, gives two accounts of the Mahanamnls
or Sakvaris, and gives a3 the five purtsapaddni: evd hy eva I evd hy Ague I evd hJndra I evd
hi Pitman I evd hi devdh 1

It is by no means obvious how these verses came to be considered as an especially fruitful
rain-spell. As such they are clearly recognized in the Gobhila Grhya Sutra, III, 2, and the
Khadira Grhya Sfitra, II, 5, 22 sq., where the Sakvanvrata is clearly a rite of sympathetic magic
to produce plentiful rain (see Oldenberg, Religion des Veda , pp. 420-422, with whose remarks
I fully coiicui).



264


AITAREYA A RAN YAK A


V, 1, 1-


Aranyaka V

Adhyaya 1.

In the Mahavrata ceremony there are twenty-five verses to accompany the
kindling of the fire. 1 In the twenty-one 2 verses (used in the Visuvant) four
are inserted before the second last, beginning, ‘With fuel Agni ’ (RV., VIII, 44, 1).
A buU is to be offered to Visvakarman 3 accompanied by muttering the verses.
I he Ajya and Praiiga Sastras are taken from the Visvajit. 4 The Sastras of

1 Sayan a explains that although the Samidhcni verses are not part of the Soma sacrifice
itself, yet they arc used in the animal sacnlice which forms a part of it and so are in place
here. He quotes Mim.imsa Sutra, III, 1, 18, 9: dnarthakydt tad aiigesti, They aic to be said
after tlte anointing of the animal by the Adhvaryn, according to Apastamba. Cf. also his
Vajuaparibh.isa, 2 and 3 (S. />’. E., XXX, 319, 345). For the gen., cf. Caland, Altindisches
7 aubi-yritital , p. iS, n. 2 ; Satapatha Brahmana, X, 1, 5, 4; TTI, 1, 1, n. 3.

a I here arc in the Dar&ipfiinnmasesJ:i, see Ilillrbrandt, A T eu - imd Vollmondsopfer , pp. 74 sq.,
lilteen verses beginning with RV., Ill, 27, 1 (cf. Oldenburg, S.E.E., XLVI, 299; Bergaigne,
hcihcuhcs stir Vlustoire dc la hturgtc vdditjue, p. 19); see Taittiilya Brahmana, III, 5, 2, 1.
lhcrc are only eleven separate verses, blit the first and last are each thrice repeated. In the
Visuvant the fifteen are extended into twenty-one by the interpolation of six verses beginning with
RV., Ill, 27, 5. These are inserted before the second last verse, RV., V, 28, 5. Then four
more verses, beginning with RV., VIII, 44, 1, aie added before this vcisc to make up the twenty-
five. 1 he Sankhdyana here ignores these verses. Aitarcya Brahmnna, I, 1, 14, gives the
number as 17. See a list in Asvalayana Srauta Sutra, I, 2, 7. The construction acc. for nom. is
lemarkablc and is not a mark of late or careless style, for these 11 regularities and the use of
numerals aie found in the Mantras (e.g. saptd rynam, iatdm purbhth, cited by Whitney,
Sanskrit Grammar , § 486 c) and in the Aitareya Brahmana, III, 48, 9 : catuhsadim kavaana
dstik, while in VII, 2, 7, panialarah sasfis ti ini ca hat any it hi tya occurs (sec Aufrecht, p. 428).
Above, TT, 2,4; 3» 8, occurs sattiimhatam sahasrdni , while Aitareya Brahmana, VII, 1 has
satin nth atom ckapaddh , which examples all appear to l>c transfers of accusative for nominative,
though the possibility of their being new stems in a cannot be denied (especially as the
Aitareya Brahmana actually has trayastrimUtyd, a trnnsfer to the * declension). Cf. Introd.,
p. 56. The idiom has hardly been adequately noticed in Delbruck, AltindiscJie Syntax , p. 82.

3 The Sankhayarm Aranyaka, I, 1, prescribes a bull for Indra and a goat for 1 ‘rajapati.
The Srauta Sutra, XVII, 7, 7, mentions also a savaniya paiu, see Ilillebrandt, Ritual-Litteratur,
pp. 125, 136. Cf. also Katyayana Srauta Sutra, XIII, 2, 17. UpamUt means not in silence but
so as not to be overheard, see Sayana’s quotation, karanavad aiabdam manahprayogam , and
Apastamba Yajnapaiibhasfi, 9, 11 and 113 (S.E.E., XXX, 319 and 345), where the Samidhenis
are not upinnUt but antard (see note on 11).

i For the Ajya sec I, 1, 1. The Prauga consists of seven liras t I, 1, 3-4, preceded by the
pur onus, Vdyur agregd yaj/iaprfr, &c., Sankhilyana Srauta Sutia, VII, 10,9. The purorucs
are also given in SchcftelowiU, Die Apokiyphen dcs Rgvcda } as Khila, V, 6.



—• V ,i,i TRANSLATION AND NOTES 265

the Hotrakas are taken from the Caturvimsa rite. 6 In the morning pressing
the Ihahmanacchamsin should add the verses, beginning, ‘The busy moving
ones’ (RV., X, 153, 1), and at the midday pressing the verses, ‘Of this strong
youthful one drink* (RV., X, 160, r). 8 The tristich which forms the strophe
begins, ‘The buffalo in the bowls, the barley-mixed* (RV., II, 22, 1), the tristich
forming the antistrophe consists of the three verses, ‘ Indra, come hither to us
from far away* (RV., I, 130, 1), ‘For to Indra heaven, the wise one, bowed*
(RV., X, 127, 1), and, ‘To him a song excelling* (RV., X, 133, i). 7 The Maru-
tvatlya Sastra is taken over from the Catuivimsa and extended by the hymns,
‘Fair has been my effort, singer* (RV., X, 27, 1),‘Drink the Soma for which
in anger thou breakest’ (RV., VI, 17, 1), ‘With what splendour’ (RV., I, 165, 1),
and, ‘Indra, with the Maruts ’ (RV., Ill, 45, i). g The Marutvatiya Sastra ends
with the hymn, ‘Thou art born, tcirible, for strength, for energy’ (RV., X, 53, 1).
At the end of the Marutvatiya Sastra, the Ilotr, leaving his place by the incomplete
route, 9 offers three oblations in the Agmdh’s fire with a ladle of udumbara wood
(accompanying them with the verses):—

8 The Hotrakas are the Maitravarunn, Bralunniucchamsin, and Achnv.lka. In the Agnistoma
their Sastias begin with RV., 1 TI, 62, 16; VIIT, 17, 1; 111 , 12, 1, respectively. T11 the Calur-
vim&a they begin with RV., V, 68, 3 ; I, 4, 1 ; VIII, 72, 13, icspectively.

8 The MahJivrata differs in these points even from the Caturvimsa. S.iyana leaves it
undecided whether the passages extend to live verses, or only to one verse by the paribhdsdx ,
ream pddagrahanc, for which sec Asvallyana Srauta Sutia, I, 1, 17.

7 These verses are apparently to precede the Sastra of the Bilhmanaccharpsin at the midday
pressing. The word stotriya is used because the verses correspond to those used in the Saman
corresponding to the Sastra, cf. Ilillcbrandt, Ritual-Litteratur, p. 103. The Saiikhayana
Sakha ignores the 6astras of the Hotrakas. The reference to the midday pressing is out of
order.

8 For the Marutvatiya Sastra of the Ilotr at the midday pressing, sec I, 2, 1 and 2. In the
Agnistoma it begins with RV., VIII, 68, 1-3, and VIII, 2, 1-3. The Caturvimsa contains
alterations, and the Mahavrata adds the hymns enumerated. Atdnah (found in VS., TS., &c.)
must mean vista rah as S.iyana has it here. Cf. Aitareya Brahmana, V, 4, 12, where S.iyana
renders tastraklptih . Friedlander, on Sahkhayana Aranyaka, I, 3, suggests the sense ‘scheme’
for it. In RV., II, 1, 10, dtdnih ‘ expander ’; cf. my Sdnkhdyana Aranyaka , p. 3, n. 6.

9 S.iyana here (cf. Anartlya on fsankh.iyana Srauta Siitra, VI, 13, 7; VII, 7, 4; Asvallyana
{srauta Sutra, V, 19, 8 ; VT, 5, 1, and comm.) explains that the samsthitasamcarah is when, after
the completion of the pressing, the Ilotr departs from the sadas by the west, the visamsthitd J
is when, before the pressing is 'finished, he leaves by the eastern side. The Sahkhayana
Siauta Siitra, XVII, 12, gives eight oblations on the dgnldhriya, instead of three there and
ten in the mdrjdliya. The Mantras arc quite different. See XVII, 12, 1-4. The first is a
long prose Mantra; the second to the seventh svdhd Mantras, and the eighth consists of
a couple of verses, the first an anustubh , the second a gdyatrt in strongly marked iambic
metre of an archaic type, neither of which veises has, according to Bloomfield’s Vcdu
Concordance, any parallel. After leciting the verses, he puts down the ladle yathdyatannm ,
depaits by the way he came, and in front of the sadas to the north of the sruli , facing the



266 AITAREYA ARANY AKA V, i, j-

‘ Indra, Brhaspati, Soma, and the goddess, Vac, have aided me. 10 May Mitra
and Varuna, Heaven and Earth, aid me when first I call || i n

‘ May the Adityas, the all-gods, and the seven anointed Kings, 11 Vayu, Pusan,
Varuna, Soma, Agni, Surya, with the constellations, may they help me n 2 11

‘ May the fathers protect me, and all this universe, and the children of Prsni,
the Mai uts, with their splendour, ye who have Agni as your tongue and are worthy
of sacrifice, may ye gods, hearing our cry, protect us n 3 n'

He offers ten oblations on the mdrjdliya altar 12 to the south, the last of
which he first divides into four and deposits to the north of the fire. In the
middle of the day, after the carrying forth of the fire, the mdrjdliya fire is made

cast, he mutters the panmdddh japdh, vdg dyur vilvdyur viivarn dyur ehy tvd hindropehi
viivatha vidd maghavan vidd ill (cf. above, p. 263), after which he adores the several members
of the fire altar conceived in human form (XVI I, 12, 6—13, 6). For the Parimads themselves,
cf. my Sdhkhdyana Aumyaka, p. 4; Eggcling, S. B. £., XLI, 288, n. 2, and for the meaning of

mad, Lanman in Whitney’s Translation of Atharvavcda, p. 138. The Ilotr rocs north to
the Agnidh’s fire. (For Agnidh, cf. Oldenbcrg, S.B.E., XLVI, 189, and Macdonell, Vedic
Grammar , p. 18, n. 6.)

10 Oi ‘may they aid me’, as Saynna takes it. lie thinks purvahutau is an epithet of
Dyavaprthivi or Mitt dvarunau.

11 Saynna explains this by the list in Taittiiiya Aranyaka, T, 7, dr ego bhrdjah patarah
pataiigah j svaniaiojyotidman vibhdsah \ ie asmai sarve divam dtapanti I This may be l ight,
otherwise one might expect it to mean the seven Adityas. No doubt the seven Adityas set
the model to the later theory of seven suns, whose names are variously given (cf. seven Rsis,
seven Ilotrs, seven sounds, &c., Oldenbcrg, S.B.E., XLVI, 225); sec Visnu Puiana, Vl’,’2 j
Ilopkins, Great Epic of India , p. 475. Rajendralala reads in the text met nit, which is cci taiiily
wrongly accented and seems not quite as likely as mdnu in view of the ami elsewhere used.
The 1 aittiiiya Brahmana, II, 5, 8, 2 has: dim tvendio madatv dnu Brhaspdtih I dnu Sdmo
dnv Agnir dvit \ dnu tvd vlive deva avantn I dnu sap/d rajdno yd utdbhisfktdJ.i I dnu tvd
Mitravdrundv ihdvatdm \ dnu dyavaprthivi vitvdtainbhu I suryo dhobhir dnu tvdvatu I can -
drdmd ndksatiair dnu tvdvatu I Note the different reading utd abhisiktdh. The text
appears from Bloomfield, Vedic Concordance , p. 973*, to occur in Kafhaka Samhita, XXXVIT,

9 d, which has (9 c) suryo 'hobhir anu tvdvatu , eonfiiming mdnu against Mitra’s md nu
(which is followed in the Concordance , p. i028 b ), and (9 b) anu Somo anv Agnir avit, ami
(9 a) anu tvendro madatv anu Brhaspatih, thus presenting only one line as against the two
lines of the Aranyaka and the Brahmana. In the next verse yd agnijihvd utd vd ydjatrdh
is a tag found in RV., VI, 52, 13 c, and in the other Saiphitas (Bloomfield, p. 795**) ; t h e
other three pddas seem as yet unparalleled. The scries of prose Mantras below is also (sec
Index II) unique.

12 In the middle of the sadas and the havirdhdnas there is a space from north to south.
The dgnidhriya altar is at the_ north, the mdrjdliya at the south. With caturgrhitam\
djyarn must be understood, see Apastamba, Yajfiaparibhasfi, 195 ( S.B.E ., XXX, 341); c f!
caturgrhitena juhoti, Taittiriya Aranyaka, V, 2; catmgrhitds thru djyahutir , Aitareya Brahmana*
VIII > 10 > 9 , 0 nr bit am, VII, 21, but the construction is very awkward. Throughout the terms
dakuna and uttara are ambiguous. For the sadas the priests’ tent, cf. £atapatha Brahmana

5 > 3 > 5 > an d Fggcling’s note.



TRANSLATION AND NOTES


-V, i, i


267


to kindle. 13 (The offering is made in ii) when it is covered up, and either to the
east, the north, or the north-east side. (The verses used are as follows):—

4 May I become unassailable like fire; may I become firmly rooted like
the earth 111 11

‘May I become unapproachable 14 like the sky; may I become unassailable
like the heaven 11 2 11

‘ May I become without a superior like the sun; may I become renewed like
the moon 11 3 11

‘May I become renewed like mind; may I be multiplied like the wind 16 11 4 II
‘ May I become one’s own like the day 1,1 ; and dear like night 11 5 11
< May I become born again like kine ; may I become glorious 17 like a pair ll 6 II
‘ Mine be the flavour of water and the form of plants u 7 ll
‘ May I become widespreading 18 like food, and lordly like the sacrifice ll 8 11
‘May I become like the Brahmin in the world, and like the Ksalriya for
prospeiity ll 9 ll

‘ When, O Agni, this assembly is gathered (RV., X, 11, 8) 19 1110 11'

1S The idea seems to be that the lire is kept in from the time it is lighted on the mdrjdlJya
altar but is now ‘wakened’, prabhrti in this use is first found in the Srauta Sutras, Speijer,
Vcdische und Samkrit-Sjnitax, § H2.

14 The attraction of andpyam is cuiions, but is paralleled in RV., I, 65, 5 : putfir mi rctnva
ksitir mi firthvi gi>lr mi bhujmd (Oldenbeig, S.B.E., XLVI, 56), and below, viana ivdpurvam ,
atinain iva vibhu, gdva ivet putiarbhuvafj , and in the case of the verb, RV., V, 25, 8 :
dyumanto arcayo grdvtvcuyatc hr hat , Oldenbcrg, S. B. E., XLVI, 417. Cf. also Taittirlya
Aranyaka, VITT, 6; Weber, Ind. Stud , IT, 221, n. For a scries of woids with bhuydsam , cf.
Jaimmlya Upnnisad Brahmana, HI, 20 and 21.

lft yathd tnana uttarottaram abhivrddhikdnfoayd prayatamdnam sat iatiatphalapriiptyd
nut an am rtipam pratipailyate . . . yathd vdyur asddhddtmdse samudi athddidde vd svayarn
uttarottardbhivrddhyd sahgharupo bhavati (Sayana).

18 Sayana renders svam as wealth. The day gives wealth by permitting mercantile operations.
Emendation to svar is easy but impiobable. Cf. the curious svah in RV., I, 77, 5 (Oldenbcrg,
S. B.E., XLVI, 88), yaksam iva, Gobhila Grhya Sutra, 111 , 4 , 28; Geldncr, Vcdisihe Studicn ,
III, 140. Night gives rest to the weary (Sayana); note priyo not priyd.

17 This must be the sense though the expression marlcayah , * glories/ is curious. Kinc
have offspring yearly, and pairs (e.g. Uma and Mahe^vaia, Laksmi and Narayana) are glorious
(Sayana).

1H The reading vibhu is certain, but both Rajendralala and the Anandasrama edition lead
in the commentary vibhuh , and "Sayana may have so lead, but this is not necessary, tor a
converse case, cf. V, 2, 1, when Rajendralala reads vastt for vasuh. The next Mantra offers
considerable difficulty. Sayana renders as the Brahmin in the world and ksatram rdjyam
gaj (1 tvddiiriydm adhipatih , apparently taking triydm as a genitive (cf. Whitney, Sansktit
Grammar, §§ 349,351). But the parallelism of the sentence calls urgently for a locative
which gives fair sense, ‘in point of wealth.’ 3 he speaker desires (a) renown, (b) wealth.
Only the exact force of the locative varies in the twq cases.

18 The last oblation is accompanied by a RV. verse.



2 68 A1TAREYA ARANY.AKA V, j,

(In this stanza) the three words (lira, vibhajCitha, and vi/ha are not in accord-
ance with the Rgveda text. 20

Standing there he worships the sun, 21 turning so as to keep his right side
towards it as it turns, with these verses, omitting the cries of svCihdp and with the
verse, ‘Come hither, this is sweet, this is sweet. Drink this bitter draught. This
is sweet, this is sweet/ He then instructs the maidservants, 23 who carry full
pitchers, six in front, three behind, (saying), ‘ Walk three times from left to right
round this altar and this pitcher of water, smiling your right thighs with your
right hands, and saying, “ Come hither, this is sweet, this is sweet/''


This must mean that in the rite the RV. verse is to be altered by reading J n pdda 3 ratnd
«xlra vibhajatha svadhdvah for ratnd 01 yad vibhajdsi, and in pdda 4, bhdgam no atm
vasumantam vftha for vJtat. Saynna adds that these alterations are improper, just as the
alteration vidhch for vrdhatu in Hrhaspatir no havhd vrdhdtn, Tnittirlya Snmhita, 1 2 2 1 •

l 2 \ 3 j Snmhita, J, 2, 2; ITT, 6, 4. The v.l. is not in Bloomfield. Bu’t ihis is

not implied in the Aranyaka. The verse occurs in Atharvaveda, XVIII, 1, 26, and MaitraynnI
Snrphita, IV, 14, 15, but in neither place so altcied. Bloomfield ( Vedic Concordance , pp 43*
749 ) also can meicly quote Sayaiia’s view. Perhaps the Baskala Sakha is meant. A different
case occurs in IV: Indram dhdnasya sdtdye havdmahe when havdmahe is added (as in
Maha, aiaynna Upanisad, 7, cited by Bloomfield, Vedh Concordame, p. 2io<h to the first three
words which are found in RV., VIII, 3, 5 d. But the MahanamnI verses aic not part of the
K V - 1 . Ir occurrcncc is not , parallel to this remaikable case.

^ I his is done later in the Sankhayana Aranyaka, I, 5, where the words arc almost identical
atranm lidhann adityam npatidhate, The Mantra is quite different, see Srauta Sutra XVIf’
I0 ‘ hor the following, sec my Sankhayana Aranyaka , pp. 76 sq.

" ofTerings are accompanied as usual by the cry svdhd. These are omitted. For the
rule, cf. Apastamba, Yajfmpaiibhasa, 87 (.S'. />’. E., XXX, 339).

ri ht T S ft"^ a S f auta Sr ‘ tra - ^ VI /» l *> where apparently deliberately the direction is from
ght to left (apradaksinam), though the words said arc alike, hai maha 3 idam madhu idam

T, , f ‘iTr ^ t an . C ? 1S ° Car,y a rain and ve f? cta tion spell, cf. Famell, Cults of the Greek
States, III, 103. These and the other ceremonies are all mentioned in the other parallel
passages, Latyayana Srauta Sutra, III, 10-12; IV, 1-3; Tiindya Brahmnna, V, 5, 6; Kathnka,
XXXIV, 5; katyayana Srauta Sntrn, XIII, 3; Taittirlya Samhita, VII, 5, 9 and 10; Taiitirlya
Brahmnna, I, 2, 6, 7. These versions differ in many details; the most impoitant rite which
i> mentioned in neither of the Rgvcdic works is the struggle of an Arya and a Sndia for a
round skin which represents the sun (cf. Oldenbc.g, Religion des Veda, pp. 444, , 0 6 • Uscner
Auhw f A'eltgtonswtssensthaft, 1904, pp. 297 sq.). It is noteworthy that in Utyayana IV
3, 18, where the words repeated aie like those in Sankhayana the form vculatyah also occuis’
So I)rahyayn ja ; I nittiiiya Samhita, VII, 5, 10, has gdyantyah. The direction there is also
la^tnam. ^After the eight djya libations in the agmdhrtya fire, according to the Sankhayana
Aranyaka, I, 4, come the pari mads. They arc twenty-five in number and are followed* bv
seven stotnyas named dhgirasa sdman, bhutcchaddm sdman, krofa, anukroia, payas, arka and
tnkaputfa The Satapatlia Brahinana, X, 1, 2, 8; 9, contains a somewhat parallel version
/ C - ,,'? KC ing> XLIII, 288, n. 2, and thus again (cf. Introd., p. 36) agrees with the'

‘ ‘"J U)ana aga,nst lhe Aitarcya. These sdntans are called devachanddmsi, Sankhayana I c
am are followed by japas. r \ hen comes an adoiation of the members of the fire (see here
V, i, 2), and of the sun, and the IIot r declares that the ‘great one has united with the great



-V, I, 2


TRANSLATION AND NOTES


269


2. ‘ When the singing of the stotra has been requested, then do ye cast down
the water in three places, on the northern altar, on the marjafiya altar, and
the rest within the enclosure.’ 1 Having gone away so as to keep the mdrjaUya
fire on his right, 2 he stands before the sacrificial post in front of the fire, with
face to the west, and worships the head of the fire with the words, ‘ Honour to
the Gayatra which is thy head: ’ then, returning by the way he came, 3 with
face to the north, he worships the light side of the fire with the words, ‘ Honour
to the Rathantara which is thy right side/ Then passing to the west of the
tail of the fire, 4 with face to the east, he worships the left side of the fire with
the words, ‘ Honour to the Brhat which is thy left side/ Then on the west s

one’, i.e. Agni with PrthivI, ‘the god with the goddess,’ i. e. Vayu with Antaiiksa, 4 Brahman
(neut.) with Brahmanl ’ (see Introd., p. 68, n. 1), i. e. Aditya with Dyaus. On this follows
(1, 6) a Vi&vamitra legend (cf. Aitareya, II, 2, 3) to explain these identifications. For the
use of ufa + Vithd, cf. the famous passage in the Mahabhasya, I, 3, 25 (Weber, Ind. Shut,
XIII, 4^0> where an ape upattjhati to warm himself, but a man upatist hate in reverence.

1 For antarvedi, cf. Aitareya Brahmana, VII, 33, 1; antahparidhi, Brhaddevata, VII, 98;
Wackernagel, Altindische Gramma tik, I, 312. This belongs of course to the end of the
preceding Khanda, and it is dtfiicult to see why it has been separated in Sayana’s text, uttare
mdrjdllya means the dgnuihrlya fire, which was used for the same purpose.

* This describes the worship of the fire altar in its simplest bird shape, head, two wings,
tad, and body. In Sankhayana Srauta Sutra, XVII, 13, the sdmans and the order differ,
being (1) pdrvdtdha with Gayatra, (2) right side with Rathantara, (3) left side with Brhat,

(4) madhya with Vamadcvya, (5) tail with Yajhavajuiya. Cf. Satapatha Brahmana, JX, 1,
2, 35 and 39; X, 1,2, 8, and Kggeling’s summary (based on this passage and Sankhayana), S.B. E.,
XLI1I, 283, n.; Llityayana Srauta Sutra, III, 11, 3, wheic as here the body is placed last, but
which agrees as to the sdmans with Sankhayana and also with Drahyaynna, and in which
the sprinkling of water in three parts also occurs. The Simians referred to will be found as
follows, gdyatram in trivrt stoma, Samaveda, II, 146-148; 263*265; 800-802 (or JI, 8, 4, see
S. h\ E., XL1II, 178); rathantauim in paiiiadaia stoma , ibid., II, 30, 31 ; brhat in sapladaUi
stoma , ibid., II, 159, 160 ; rdjanam in paHcavttnia stoma, ibid., JI, 833-835 ; bhadra m chart mia
stoma , ibid., II, 460-462. For the banians cf. II, 3, 4. For a drawing of the agnik^etra see
Weber, Ind. Sind., XIII, 235.

9 He had gone from the mdrjdllya in the south to the cast side of the citydgni and he
now returns to the south. Rathantara is unusual, but it is supported by all the MSS.
Latyayana and Sankhayana have rathantardya .

4 It is not clear why he should not go round to the north, but all that is done is to go
to the end of the west or tail side, when looking east, along the left side, he litters the
Mantra.

6 palcdt may simply mean ‘next’, or, as Sayana takes it, refer to the place where the
Ilotr stands. Apparently the difference between this and his foimcr position is that lie stands
directly behind the tail, instead of going past it. This account of his movements coiresponds on
the whole with that of the ceremony of the Satanulriya, which has analogies to the Mahiivrata
(Satapatha Brahmana, IX, 1, 1, 44). In it, according to the Satapatha, IX, 1, 2, 35 sq.,
the Simians, (1) gdyatram , (2 ) ra/hantaram, (3) h hat, (4) Vamadcvyam, (5) yajildyajfliyam ,
and (6) Prajdpatihrdaya, correspond to (1) head-, (2) right wing, (3) left wing, (4) body,

(5) ta fi> (6) heart; according to Latyayana, I, 5, 11, which very closely follows the order of



270


AITAREYA ARAN YAK A


V, 1, 2-


of the fire, with face to the east, he worships the tail with the words, ‘ Honour
to the Bhadra which is thy tail and thy support.’ Then on the south of the
tail he worships the body with the words, ‘Honour to the Rajana which is
thy body.’

3 . lie returns to the seat as he went. 1 The swing has already been made
ready.* Having cleansed the two posts, the ropes, and the cross-beam, and
having taken them by the road called firtha , 3 having gone round to the left
the Agnidh’s altar, 4 (having brought them within) the seat by the east door
(he places the implements 5 ) to the left of all the altars. The planks of the swing
are made of udumbara or of paldsa , or of both. There should be three planks
worked on both sides, or two, and a like number of sharp-pointed sticks. The

movements in this Aitareya passage, the (1) gdyatram , (2) rathantaram, (3) hrhat, (4) yajfid-
yajtiiyam , (5) Vdmadevya, and (6) Prajdpatihrdaya , correspond to (1) head, (2) right side,
(3) left side, (4) tail, (5) right arm-pit, and (6) left arm-pit. Cf. also the elaborate ceremonial
of the parimddah at the Mahavrata as described in Jsatapatha, X, 1, 2, 9 ; Jsankhayana Aranyaka,
II, 4 (with Friedlander’s note, p. tf ); and the similar use after the beginning of the prdha stotra
of the parimadah (prana, apdna , vratapaksau , Pi aja pa ter hrdaya, VasUthasya mhava,
Sattrasyardhi , tloka and anuUoka , ydma , ay us, navastobha , ryasya sat nan) in the worship
of the paits of the altar in Tfindya Brahmana, V, 4, 1-13; L.ltyilyana 6rauta Sutra, III, 9,

1 st [.; Taittiriya Brahmana, I, 2, 6,5. In the Mahavrata Saman the parts of the bird aie
head, right wing, left wing, tail, and trunk only (Fggeling, S.B.E ., XLIII, xxvii). The
whole conception is clearly borrowed (cf. Introd., p. 50) from the altar in the Agnicayana
which gave origin to the mystic doctrines of the Adhvaryus (see especially Satapatha
Binhmana, VI-X), and of which the Mahavrata is an adaptation by the Ilotrs. In Vajasaneyi
Samhita, XII, 4, the irivrt is the head, the gdyatram the eyes, hrhat and rathantaram the
wings, the hymn the soul, the yajumsi the name, the metres the limbs, the Vamadczyam the
body, the yajildyajilTyam the tail. For the relation of saman and words, cf. Oklenberg,
Z. D. Af. XXXVIII, 439 sqq., 464 sq. ; Wintemitz, Gesc/i. dcr indisch. Litt., I, 143 sq., and
see Fggeling, S. />’. E., XLIII, 180, n. 2 ; Weber, Jnd. Stud., XIII, 276 sq. The Vamadevya is
based on Samaveda, II, 32, 33; the Yajnayajiaya on Samaveda, IT, 53, 54.

1 lie comes back to the seat near the mdrjdlJya fire, which he left to worship the city a
altar. The expression occurs several times in the Srauta Sutra. For the eight altars see
Fggeling, S.B.E ., XXVI, 148, n. 4 and the plan on p. 475, followed by Caland and Ilenry,
L'Agmdoma ; Ilillebrandt, Neu- und Vollmondsopfer , p. 191.

3 By the Adhvaryus. Cf. Aitareya Brahmana, VII, 32.

3 This is the name of the passage between the utkara and cdtvala , £ankhayana Srauta Sutia,
V, 15, 3, &c.; Maitrayam Sainhita, III, 8, 10. The action is rendered intelligible by a glance
at the plan in Fggeling.

* The pari of parivrajya must refer to circumambulation. The meaning of the phrase is
probably given by Sankhayana Siauta Sutra, XVII, 11, 4, purvayd dvdrdgmdhram prapa-
dyottarendgnid/iriyatn dhBtiyam paryetya , though the put vayd dz>drd here is otheiwise applied.
The idea is, he goes round the altar from right to left, probably. Cf. also ibid., V, 14.
The sentence is so elliptical as to be unintelligible without Saynna’s pravetya. Sankhayana,
XVII, 7, 11, is much more simple.

a The verb must be gathered from atyddadhati below; strictly speaking the next sentences
arc parenthetical and this sentence is continuous with dakdnoltare sthunc nikhdya.



-v, I, 3


TRANSLATION AND NOTES


271


swing should be a yard in size from east to west, its cross breadth should be
a yard less a hand; the points of its (planks) should be to the north, and they
should be fastened together by sticks with their points east. Having inserted the
posts in the earth to the north and south, around the seat of the Hotr, he spreads
the cross-beam over them so that it is on a level with the worker’s face. 8 Holes
are (bored) in the corners of the planks of the swing. He fastens the planks
above by means of the ropes, the right one on the south, the left on the north. 7
The ropes should be of darlha grass, and with three strands, 8 one rope to

. • In the Sahkhiiyana Srauta Sutra, XVII, 10, 7 and 8, the height is measured by the head of
the Hotr, or if he is small his outstretched arms. Ibid., 4, 6, shows that both the planks and
the cross-beam have the points north. For the construction with kartn/i dependent on
rhya°, cf. Whitney, Sansk/ it Grammar , § 1316. Speijer ( Vc disc he. und Sanskrit-Syntax, § 113)
gives many classical examples. For abhiiah with accus., cf. Delbruok, Altindische Syntax ,
p. 183. It is found in Mantra, but more often in Brahmana, Speijer, Vedischc und Sanskrit -
Syntax , § 88. For uttarena with accus, cf. Gaedicke, Der Accusativ in Veda , pp. 207 sq. ; sec
Liebich, Bezz. Beitr. y XI, 284. Delbiuck and Gaedicke seem right in explaining the use as
derived from the accus. with antdr and antara. Whitney, Sanskrit Grammar , § 273, offers
no explanation. In V, 1, 1, we find uttamto 'gnch\ in V, 1, 2, dahinatah puchasya with the
more natuial adnominal genitive. But in V, I, 2, aparena has the accus. In Srinkhiiyana
Aranyaka, VTT, 3, antarena has the gen.; in the Sutra, the acc. The measures are dubious, see
Hopkins,/. A. 0 . S., XXIII, 141.

7 The &ahkhayana fsrauta Sutra, XVII, 10, 14, 15, explains that the right rope is tied to
the north of the south post, the left to the south of the north post, i. e. inside the posts,
just as in a modern swing. The point of view is of course facing east, with the south on
the right and north on the left.

8 The use of tribune and dvigtoie with different senses of guna is awkward, but appears
clearly so meant. Sfiyana points out that the rope as doubled would be 2^ fathoms in length,
of which only a yaid would be used by the rope passing under the plank (above iuimdtrah
prdh prchkhaJi). There would thus be plenty of lope available for the tying, as the top
was only a man’s height or less. Sayana takes savyadafcinc as 1 inclining to the left and
right’, i.e. the ropes should not go stiaight up. The only obscuie point in this description of
the lying on of the seat of the swing to the cross-beam is pradakdnam , since it is not at first
sight obvious how this applies to the act of fastening ropes. It apparently must mean that
after the rope has been passed under the scat of the swing the one end is rolled round the
cross-beam slanting to the right, the other (on the opposite side, of course a ) also slanting to
the right and the ends then are tied across. Provided there was sufficient friction to keep
the ropes fiom slipping this would seem to give a substantial knot (cf. mstarkya). If this
is so, we cannot accept Sayana’s theory of savyadakunc and must fall back cither 011 the
view that the word means metcly left (hand) rope and right (hand) rope, or take the epithet


a It is very unlikely that both ends of the rope should have been brought to the same
side of the cross-beam. In that case pradalsinam would be rather less than more m point.
Speijer ( Vedische und Sanskrit-Syntax, § 106, n.) points out that adjective dvandvas are not
unknown even in Sanskrit (cf. his Sanskut-Syntax , § 208), and (p. 32, n. 1) argues fiom
l’anini, VI, 2, 38, when ckadaUi is given as a dvandva that the grammarians recognized such
types. He (§ 107) gives classical examples of distributive dvandvas.



272


AITAREYA A RAN YAK A


V, 1, 3-


the left, one to the right, and five fathoms long, and should be folded double.
Then folding (each end) thrice (to the right) round the cross-beam he makes
a knot on the top, which can only be untied by twisting. They support the
posts so as to be steady by means of branches and brushwood. 9 The swing
should be four fingers or a hand distant from the ground. 10 On the right it
may be somewhat higher or level. It should be a foot from the altar.

4. When 1 the swing has been put in position, the Hotr taking a lute of
udumbara wood, with a hundred strings, in both hands, strikes it, 2 beginning
from the lower side, as one docs an ordinary lute. 3 The different notes of the
lute he should produce in turn by the seven metres, 4 each with four (syllables)

as applying to each rope and as meaning, 1 with strands coiled from left to right.’ Cf. perhaps
the equally obscure passage, Apastamba Yajfmparibhasa, 60, 61 (S. li. E., XXX, 331, where
Max Muller says, ‘The exact piocess here intended is not quite clear. The ropes seem to
have l>een made of vegetable fibres. See Katy., I, 3, 15-17’). If savyadaksine = left and
right, cf. for the use of the dvandva , Wackernagel, A It indisc he Grammatik, II, i, 160, who cites
Atharvaveda, XII, I, 28: padbhyant daksinasavyabhydm ; Taittirlya Brahmana, I, 5, io, 1:
suvarnarajatahhya m hdibhydm. The different order of woids, savyadakdna , is in accordance
with the usual rule as to number of syllables determining the order of the numbeis of their
compounds, Wackernagel, II, i, 166.

’ Sayan a explains that they fill up the holes in which are placed the feet of the posts

with dust, which is not thrown in by hand but by branches and firsts. This, however, is

quite unnecessary. Ihushwood would be a much better material for strengthening the hold
of a post. He defines brsT as trnavallitdlapatravenudalddiohir nirmitd alpakatavitesdh. The
swing was obviously shaped like this |\_/J.

10 The distance according to Sfinkhayana should be a prddeta , XVII, 10, 13. Ibid., XVII, 1,
discusses the planks; 2, the ropes and dsandi\ 3, the lute; 4, the drums; 5, 6, 7, the other
accessories and the preliminary steps, in great order and detail. Cf. Latyayana Srauta Sutra,
111,12.

1 There are similar passages in the Tandya Brahmana, V, 5, 4 sq., and Latyayana Srauta

Sutra, III, 12, 8 ; IV, 1, besides in the 6ankh«lyana Srauta Sutra, XVII, 3; 15, 10 sq. Sayana

points out that the Ilotr is now seated to the west of the swing. The exact words as to
the lute do not occur in 6ankhayana, but it is elaborately described, XVII, 3.

2 Siiynna renders merely, ‘ he should hold it on his left side like a lute.’ But the idea is
perhaps rather that he strikes one stiing after nnother, ascending in the scale, beginning
from below and ascending, uttaratah , cf. urdhvam below and Agnisvamin on Latyayana
Siauta Sutra, IV, I, 4.

3 So Sayana on RV., I, 85, 10, where he similarly explains the phrase vdnam dhamanlah
used of the Maruts, cf. Ill, 2, 5, n. 1 ; Benfey ( Sdmaveda , G/ossar, p. 169) takes vdna there as
flute, and Zimmer ( Altindisches J.ebcn , p. 289) follows him. Max Muller (Afarut Hymns ,
pp. 120, 1 21) preferred to see in it meiely ‘voice’. For uduhami , cf. Wackernagel, Allindhche
Grammatik , I, 92, who considers u here an ablaut of u. Panini lestricts its use to Atmancpada,
but Katyayana allows Parasmaipada with a prefix as here (Licbich, Panini , p. 84).

4 i.c. he plays notes corresponding to verses composed in these meties. The four more
are, Sayana says, virdj, dvipadd , at ichan das, and chando ’ntararn . If this last be omitted ten
are got. But despite its use elsewhere, e. g., 6atapatha Brahmana, X, 1, 2, 8, it must surely



-V, r, 4