Atharva-Veda Samhita/Book IV/Hymn 21

1324812Atharva-Veda SamhitaBook IV, Hymn 21William Dwight Whitney

21. Praise of the kine.

[Brahman.—gavyam. trāiṣṭubham. 2-4. jagatī.]

This hymn is not found in Pāipp., but it occurs in the Rig-Veda (vi. 28. 1-7; vs. 8, in a different meter, is perhaps a later addition), and also in TB. (ii. 8. 811-12). It is used by Kāuç. (19. 1), with i. 4-6 and others in a rite for ailing kine, and also (21. 8 ff.) in one for the prosperity of kine, vs. 7 being specifically mentioned as repeated when they go forth to pasture; vs. 7 appears further to be quoted at 19. 14, in a rite for the cow-stall; but the comm. declares two verses to be intended, and, if so, they must be vii. 75. 1, 2, since there is here no following verse. In Vāit. (21. 24), in the agniṣṭoma, the cows intended as sacrificial gifts are greeted with this hymn. The schol. (Kāuç. 16. 8) reckons vs. 4 to the abhaya gaṇa. The comm. ⌊and Keçava's scholion to Kāuç. 27. 34⌋ declare hymns 21-30 to be mṛgāra-hymns (Kāuç. 27. 34; 9. 1), but the name would seem properly to belong only to hymns 23-29, which form a related group, and are by the Anukr. ascribed to Mṛgāra as author.

Translated: by RV. translators; and Griffith, i. 161; Weber, xviii. 87.


1. The kine have come, and have done what is excellent; let them stay (sad) in the stall (goṣṭhá); let them take pleasure with us; may they be rich in progeny here, many-formed, milking for Indra many dawns.

The other texts have no variants for this verse. The comm., after his wont, turns the two aorists in a into imperatives; he renders raṇayantu alternatively by ramayantu and ramantām; and he takes "dawns" as equivalent to "days" (divasān). ⌊'Full many a morning yielding milk for Indra.'⌋


2. To the sacrificer and singer, to the helpful one (?), Indra verily gives further, steals not what is his; increasing more and more the wealth of him, he sets the godly man (devayú) in an undivided domain (? khilyá).

The other texts have in a the decidedly better reading pṛṇaté ca çikṣati of which ours is simply a corruption; the comm., heedless of the accent, takes our çíkṣate as a verb (= gāḥ prayacchati). In d they have the better accent ábhinne; and TB. reads khillé; most of our mss. could be better understood as khilpé than as khilyé; the comm. defines khila as aprahataṁ sthānam, and khilya as tatrabhava; R. conjectures "stonewall" for khilya. All our mss., and part of SPP's, read mukhāyati in b.


3. They shall not be lost; no thief shall harm [them]; no hostile [person] shall dare attack their track (?); with whom he both sacrifices to the gods and gives, long verily with them does the kine-lord go in company.

Both the other texts* accent vyáthis in b, as does one of our mss. (O.), and one of SPP's. Before this word TB. has nāí ’nā amitró. The comm. explains vyathis as vyathājanakam āyudham. The pāda is very obscure as it stands. ⌊An earlier draft of the translator's ms. reads: "Naçanti, by its association, and its difference from naçyanti, must be meant as subjunctive (aor.), notwithstanding its ending." I am tempted to suggest ná tā́ naçan; tā́ (acc. pl. fem.) ná dabhāti táskaras.—BR., vi. 1438, take vyáthis as 'unbemerkt von,' with genitive, āsām. But see Geldner's discussion of the combinations of vyáthis with ā-dhṛṣ, Ved. Stud. ii. 29.—Note that TB's ămitró (both ed's read so in the text and both have ă- in the comm.) is neither amítro nor āmitró. *⌊In TB., the pratīkas of vss. 3 and 4 stand in RV. order at ii. 8. 811; but the vss. are given in full at ii. 4. 69.⌋


4. No dust-raising horseman (? árvan) reaches them; not unto the slaughter-house (?) do they go; those kine of that sacrificing mortal roam over wide-going fearlessness.

RV. differs only by retaining the a of açnute in a, as do one or two of our mss. (O.K.) and half of SPP's; and its pada-text divides saṁskṛta॰tra in b, while the AV. pada-mss. (except our Op.) leave the word undivided (by an oversight, the AV. Index Verborum gives the RV. form). The comm. explains arvā by hiṅsako vyāghrādiḥ, and -kakāṭa by udbhedaka; also saṁskṛtatra by māṅsapācaka (because viçasitaṁ trāyate pālayati), quoting from an unknown source the line saṁskṛtaḥ syād viçasitaḥ saṁskṛtatraç ca pācakaḥ. The comment to Prāt. ⌊iv. 58⌋ makes the word come from the root kṛ. TB. ⌊also retains the a of açnute and it⌋ has in d mártyasya. In our printed text, the upper accent-mark in reṇúkakāṭo is over the wrong k.


5. The kine [are] Bhaga; Indra has seemed to me the kine; the kine [are] the draught of first soma; these kine—that, O people, [is] Indra; with whatever heart [and] mind I seek Indra.

The translation implies in a the RV. reading achān, of which our ichāt seems merely an unintelligent and unintelligible corruption; TB. has instead acchāt, and our O.K. give the same. Both the other texts add íd after ichā́mi in d. The comm. translates in a "may Indra desire that there be kine for me." ⌊The latter part of c is of course the well-known refrain of RV. ii. 12.⌋


6. Ye, O kine, fatten whoever is lean; the unlovely (açrīrá) one ye make of good aspect; ye make the house excellent, O ye of excellent voice; great is your vigor (váyas) called in the assemblies (sabhā́).

The RV. version agrees at all points with ours; TB. accents kṛ́çam in a and has açlīlā́m in b (its kṛṇuthāt is a misprint, as its commentary shows). The comm. reads kṛṇuta in c; sabhāsu in d he paraphrases with janasamūheṣu.


7. Rich in progeny, shining in good pasture, drinking clear waters at a good watering-place—let not the thief master you, nor the evil-plotter; let Rudra's weapon avoid you.

The translation of a follows our text, though the false accent ruçántīs (TB. has the same reading) shows that the word is only a corruption of the RV. reading riçántīs 'cropping, grazing.' The comm., though reading ruçantīs, renders it tṛṇam bhakṣayantīs. ⌊The TB. comm. in both ed's reads riçantīs.⌋ Both the other texts have in a sūyávasam, and at the end hetī́ rudrásya vṛjyāḥ (TB. vṛñjyāt). With our c, d compare also TS. i. 1. 1 (differing only in the order of words in drudrásya hetíḥ pári vo vṛṇaktu, which is metrically much better than our AV. order, albeit the RV. order is as good as that of TS. if we pronounce rudṛ-ásya⌋). The comm. supplies to aghaçaṅsas in c vyāghrādir duṣṭamṛgaḥ. ⌊For īçata, see Skt. Gram. §615.⌋