⌊Cf . vs. 9.⌋ The adverbs translated 'behind' etc. might with at least equal propriety be rendered 'on the west,' 'on the east,' 'on the north,' and 'on the south.' Nearly all our mss. (not Bs.I.), and half of SPP's ⌊have the impossible⌋ accent jātávedas in c; and a few (including our O.R.) have at the end lokám. ⌊Although samyág enaṁ must pass for the true AV. reading, one cannot forbear to query whether it has not displaced an original samyák táṁ.⌋
12. Let the fires, kindled, take hold happily; let the Jātavedases, making done (çṛtá) here him that is Prajāpati's [and] sacrificial, not throw him down.
The verse (11 + 12: 12) lacks a syllable of counting as a full bṛhatī. Two of our mss. (O.R.) read in b jātavedaḥ (without accent). The comm. reads sam instead of çam at the beginning. ⌊Cf. vs. 13.⌋
⌊The verse reminds us strongly of 2. 4 (which see), where cikṣipas without ava has quite a different meaning. Here, cikṣipan with ava means 'let them [the fires, not] throw [any part of the dead man] down'; that is, Agni (in his kindly forms, çivā́s tanvàs) is to treat the dead man kindly and not let a foot, the head, or a hand fall off from the funeral pile, but is to consume him completely: cf. the comm., who aptly says, yathā niravaçeṣaṁ dahyate tathā. The importance, in Hindu belief, of having every member of the body carried by Agni to the other world for use in the next life is abundantly shown by the hymns: see especially AV. xviii. 3. 9 ab; 2. 24 c; 4. 64; iv. 34. 2: and Whitney, Oriental and Linguistic Studies, i. 56-57.—When, as often happens, the pile of wood is too short for the corpse, the feet will naturally overhang and drop off from the pyre. In my journal of a visit at Benares, under date of Feb. 25, 1889, I find the following: "Saw a cremation, at the Burning Ghat. One foot of the corpse fell off the pyre (which was none too long), and a man tried to put it back on the fire with a bamboo. But failing, he took it by the toe with thumb and fingers and chucked it back." An allusion to an occurrence of this kind is clearly made by the Chāndogya Upanishad at vii. 15. 3: atha yady apy enān utkrāntaprāṇāñ cchūlena [cf. the bamboo, above] samāsaṁ vyatisaṁdahet: nāi ’vāi ’nam brūyuḥ pitṛha ’sī ’ti etc.⌋
13. The sacrifice goes, extended, adapting itself (? kḷp), [taking] him who hath sacrificed, unto the heavenly (svargá) world; let the fires enjoy it, made a whole oblation; let the Jātavedases, making done here him that is Prajāpati's [and] sacrificial, not throw him down.
⌊Cf. vs. 12.⌋ 'Sacrifice' and 'whole oblation' ⌊and 'it,' that is tám in c,⌋ all refer, of course, to the deceased himself. ⌊Cf. Whitney, Oriental and Linguistic Studies, i. 56: "To burn the body of a deceased person was accordingly an act of solemn sacrifice, which made Agni its bearer to the other world, the future dwelling of its former possessor."—Cf. also Caland's most apposite citation from Bāudhāyana, jātasaṁskāreṇe ’maṁ lokam abhijayati; mṛtasaṁskāreṇā ’muṁ lokam, in his Todtengebräuche, pages 174, 178.⌋ The defective meter and incomplete construction of b make it altogether probable that the text is corrupt: ījānā́nām would help both. The comm. makes no difficulty of taking abhi..eti causatively, = abhigamayati. He paraphrases kalpámānas by iṣṭam pradeçam prāpayituṁ samarthaḥ. The mss. vary between kalpámānas, kálpamānas, and kalpamā́nas; all of ours that were collated before printing had kalpá-, which we accordingly admitted in our text; but we ought to have emended to kálpa-, which SPP. reads. Two of our mss. (O.R.), and one of SPP's, also have jātavedasaḥ at end of d. The verse counts just a full atiçakvarī (11 + 10: 11 + 12: 12 = 56).
