Ppp. has devās for pūrve in a, tābhyām for divam in c, and at the end -bhiç çivebhiḥ. TS. reads gṛhṇanti in a; and TS. K. MS. (ii. 5. 10 c, d) invert the order of c and d, and give the better reading óṣadhīṣu for dívaṁ gacha ⌊cf. RV. x. 16. 3⌋; MS. also has hutás for svargam. The comm. makes pūrve mean "the gods previously stationed in the atmosphere"; perhaps it is 'before the demons get hold of it.'
35. To expiate errors in the sacrifice: to Viçvakarman.
[An̄giras.-—vāiçvakarmaṇam. trāiṣṭubham: 1. bṛhatīgarbhā; 4, 5. bhurij.]
Found (except vs. 5, and in the verse-order 2, 3, 1, 4) in Pāipp. i. The same four verses are found in TS. (iii. 2. 81-3: in the order 2, 4, 3, 1), and the first three in MS. (ii. 3. 8; in the order 1, 3, 2). The hymn is used by Kāuç. (38. 22) in a rite intended, according to the comm., to prevent faults of vision (dṛṣṭidoṣanivāraṇāya; Keçava says "to prevent rain," vṛṣṭinivāraṇāya; perhaps his text is corrupt), accompanying the eating of something in an assembly. Its employment (59. 21) with the hymn next preceding was noticed under the latter. The comm. (differing in his reading and division of the rules from the edited text of Kāuç.) declares it to be used in all the sava sacrifices, to accompany the purastād homas (59. 23-4: uttareṇa savapurastāddhomān); and vs. 5 is used (3. 16) with a purastād homa in the parvan sacrifices. In Vāit. the hymn appears (9. 7) in the cāturmāsya sacrifice, with two oblations to Mahendra and Viçvakarman respectively; and again (29. 22) in the agnicayana. In all these applications there is nothing that suits the real character of the hymn.
Translated: Weber, xiii. 211; Ludwig, p. 302 (vss. 1-4); Griffith, i. 76.
1. They who, partaking [of soma] (bhakṣ), did not prosper (ṛdh) in good things, whom the fires of the sacrificial hearth were distressed about (anu-tapya-)—what was the expiation (avayá) of their ill-sacrifice, may Viçvakarman ('the all-worker') make that for us a good sacrifice.
The translation implies emendation of dúriṣṭis in c to -ṭes, and of tā́n in d to tā́m; tā́m is read by the comm., as well as by TS. and MS., and SPP. even admits it into his text, though nearly all his mss., as well as ours, read tā́n. Our P. and M. read āvṛdhús at end of a; TS. has ānṛhús, MS. ānaçús. TS. elides the a of anu in b; it begins c with iyám for yā́, and ends it with dúriṣṭyāi, thus supporting our emendation. Both TS. and MS. give kṛṇotu in d, and MS. puts it after viçvákarmā. The pada-mss. read in c ava॰yā́, but SPP. alters his pada-text to ava-yā́ḥ, on the authority of the comm.; it is a matter of indifference, as the concluding element, in spite of the native grammarians, is doubtless the root yā. Ppp. gives duriṣṭā sviṣṭam in c, d. The various readings, here and in the following verses, are in good part of the kind which show that the text-makers were fumbling over matter which they did not understand. The comm. is no better off. Here, in a, he is uncertain whether to take ná as 'as if' or 'not,' and to make vásūni object of bhakṣáyantas or of ānṛdhús (= vardhitavantas, which is not bad). ⌊The fires, pāda b, are personified in like fashion at AGS. iv. 1. 2, 3.⌋ The verse (12 + 12: 9 + 11 = 44) is much more irregular than the definition of the Anukr. admits.
2. The seers declare the master (-páti) of the sacrifice by reason of sin disportioned, distressed about [his] offspring. What honeyed drops he offended in (? apa-rādh), with them let Viçvakarman unite (sam-sṛj) us.
