Atharva-Veda Samhita/Book XIX/Hymn 56
56. To sleep (or dream).
[Yama.—ṣaṭkam. dāuṣvapnyam. trāiṣṭubham.]
Found also in Pāipp. iii. The comm. quotes no authority as to the viniyoga, but points out that the hymn is shown by its content to belong to the ceremony for getting rid of duḥsvapna 'evil-dreaming.' He holds, namely, throughout the hymn, that svápna means duḥsvapna (in the Atharvan always duṣvápnya); and the language is too obscure to show clearly whether he is right or not; the probability is certainly against him, because elsewhere (e.g. in the next hymn), when evil-dreaming is intended, its own name is freely used, and in xvi. 5 svapna is contrasted with duṣvapnya. ⌊As in the case of hymns 53 and 54, the Anukr. suggests that the hymns 56 and 57 are only two divided parts of one group of 11 verses; and the suggestion is reinforced by the juxtaposition in the RV. text (viii. 47. 15 and 17) of matter corresponding to our 56. 4 and 57. 1 (see under those verses); and hymns 56 and 57 are translated together by Ludwig.⌋
Translated: Ludwig, p. 467; Griffith, ii. 313.
1. Out of Yama's world hast thou come hither (ā-bhū); with mirth (?) dost thou, wise, make use of (pra-yuj) mortals; knowing, thou goest in alliance (sarátham) with the solitary one, fashioning (mā) sleep in the lair (yóni) of the Asura.
If prámadā (p. prá॰madā) is to be rendered as above, it must have its accent changed to pramádā; the comm., against the pada-text, understands it as pramadās = striyas, joint object with mártyān of prá yunakṣi. One or two mss. read mártān in b. The comm. renders dhīras by dhṛṣṭas. ⌊Apart from Ws P.M.W., which have svā́pnam,⌋ all the mss., the comm., and SPP., read svápnam in d, and the translation follows this rather than our svápna, willing, in so obscure a matter, to stick as closely to the authorities as possible. Ppp., to be sure, gives svapna mi-, but this counts for very little. The comm. understands the verse to be addressed to the demon of ill-dreaming (he duḥsvapnābhimānin krūra piçāca); but his explanations through the whole hymn are worthless, being only the etymologizing guesses of one to whom the real sense is precisely as obscure as it is to us: asura is prāṇavant ātman; the 'lonely one' is the man who is dying of the effect of evil-dreaming, having abandoned son, wife, relatives, etc.; yāsi means yantalokam prāpayasi; and so on, and so on. ⌊Griffith cites "Death and his brother Sleep" of Shelley's Queen Mab (it is found also at Iliad xiv. 231) and "the twins, Sleep and Death," II. xvi. 682.⌋
2. The all-vigorous bond saw thee in the beginning, in the one day before the birth of night; from thence, O sleep, didst thou come (ā-bhū) hither, hiding thy form from the physicians.
In this verse also, for the reason given above, the translation follows the mss. more closely than does our text. Nearly all authorities have at the beginning bandhás; a couple ⌊the reciters, K. and V.⌋, with the comm., bandhús (wrong accent ⌊as in vs. 5⌋); Ppp. reads baṁbhas. All have viçvácayās (p. viçvá॰cayāḥ), though in some of them the c could be read as v; the comm. is able to make a sense for it: sarvasya cetā, saṁceta, sraṣṭā; the translation implies viçvāvayās, as the smallest possible intelligible change; Ppp. presents viçvavathāv ⌊and apaçyan⌋. The pada-mss. divide in b rā́tryā: jánitaḥ: réke; the comm. understands, with us, rā́tryāḥ: jánitoḥ: éke; and SPP. substitutes this in his pada-text. Eke ahni ⌊which Ppp. combines to eke ‘hni⌋ might, of course, also be understood as locative absolute. Our tátas at beginning of c was an emendation for táva of the mss.; the comm. has it (also Ppp.), and SPP. accordingly also adopts it in his text. The whole pāda reads in Ppp.: tatas svapnena madhyā ca bhāyatha. In d all the mss. have bhiṣágbhya r-, and the pada-mss. bhiṣágbhya॰rūpam (!); only one or two give an accent to rūpám; the comm. understands bhiṣágbhyo rūpám, and SPP. reads this; the translation follows it. There is much discordance as to the accent of apagūhamānas. Ppp. reads bhiṣajña rūpam apigūh-. The comm. is a grammarian of such sort that he does not accept éke as used here for the more regular ékasmin; but he does accept ahni as used, by the ordinary license to put one case in place of another, for ahnas, coördinate in construction with rātryās; and eke means mānasaprajāpatyādayas, and is subject of apaçyan understood! The evil-dreaming hid itself away from the medicine-men, says the comm., lest they should meet it with an efficacious remedy; and something like that is possibly the real meaning.
3. He of great kine (?) turned unto the gods away from the Asuras, seeking greatness; to that sleep the three-and-thirty ones, having attained the sky, imparted over-lordship.
At the beginning, the saṁhitā-mss. in general read bṛhád gā́vā (p. bṛhát: gā́vā or grā́vā); Ppp. has vṛhaṁgrāvā ⌊combining -vāsurebhyo⌋; the comm. gives bṛhadgāvā, as nom. of -gāvan, deriving it from gā 'go'; and SPP. accepts this (bṛhadgā́vā, p. bṛhat॰gā́vā), while at the same time suggesting that -gavó may be meant, by such a confusion of the two modes of writing o as we have already more than once ⌊cf. 55. 5 b and the note⌋ had occasion to conjecture. Ppp. reads ‘bhi instead of ‘dhi. There is discordance among the mss. as to the accent of upā ’vartata. Ppp. reads at end of b ṛcchan. The majority of mss. have in d tráyastriṅçāsá sv-; and part of the pada-mss. divide tráyaḥ॰triṅçāḥ: sáḥ: sv-; SPP. gives trayastriṅçā́saḥ sv-, as do we.
4. Not the Fathers, and not the gods, know it, whose (pl.) murmur goes about within here; in Trita Āptya did the men (nṛ́), the Ādityas, taught by Varuṇa, set sleep.
The mss. read at the beginning nāí ’tā́m ⌊and so SPP.⌋, the etā́m 'it' apparently being viewed as relating to jálpis; the comm. makes the relation plainer by giving in b yāi ’ṣām; but this latter SPP. rejects. At end of b, the mss. vary between antarétám and antaré ’dám (p. antarā́: idám); the comm. gives the latter, and SPP. accepts it; Ppp. also has it; ⌊and it is implied in the translation⌋. The mss., as always, vary between tṛté and trité; the great majority here give the former. Váruṇena in d was our emendation for ár-; it is read also by the comm., and by one of SPP's mss., and is found in Ppp.; SPP. also adopts it in his text. The mss. all accent ā́dityāsas, and SPP. does not emend, as we do, to ādityā́sas. Ppp. reads in b jalpyaç c-, and has for c trite svapnam arididṛhā prate narā (ādit-). With c, compare RV. viii. 47. 13-17 and especially 15 c, d. The comm. tells a tale of how duḥsvapna, having received overlordship from the gods, waxed topping, and seized on the Ādityas; the latter applied for relief to Varuṇa, and, duly instructed by him, put off the duḥsvapna upon the great seer (maharṣi) Trita, son of the waters: this is pretty plainly no tradition, but an account devised by the comm. to fit the immediate case.
5. Of whom the evil-doers shared ⌊bhaj⌋ the cruelty, the well-doers, by non-sleep, [shared] the pure (púṇya) life-time—thou revelest in the sky (svàr) with the highest relative; thou wast born out of the mind of one practising fervor.
6. We know all thine attendants (?) in front; we know, O sleep, who is thine over-ruler here; protect us here with the glory of the glorious one; go thou away far off with poisons.
In b, the pada-mss. give blunderingly svapna॰yáḥ. Ppp. reads yo ‘dhipā hyo te. Of course, yaçasvinas in c may be accus. pl., 'us who are glorious'; the comm. takes it so, and explains that the glory comes from our wonderful knowledge as set forth in the first half-verse. ⌊Ppp. reads yaçaso hi for yáçase ’há. In d, the saṁhitā-mss. (also Ppp.) give ārā́dviṣ; which is equivalent to ārā́ddviṣ-; and the pada-mss. assume the latter, dividing ārā́t: dviṣébhiḥ; since a derivative dviṣá is as good as unknown, and of a very unusual formation, we preferred to read viṣébhis; ⌊the comm., text and exposition, has dviṣobhis;⌋ SPP. has dviṣébhis. In a the translation of parijā́s is that of the comm. (= parijanān), which seems more probable than the conjecture of the Pet. Lexx., "perhaps places of origin."