1. What that is defective (chidrá) of my mind, and what of my voice hath found (? jagā́ma) Sarasvatī enraged, let Bṛihaspati, in concord with all the gods, mend (sam-dhā) that.
The meaning of b is extremely doubtful. SPP. reads sárasvatī against the large majority of his authorities and all of ours, which have -tīm. Our hárasvantam was a conjecture, and perhaps not a particularly successful one. The translation given (tentatively) above implies -tīm manyumatī́m, while all the mss., the comm., and SPP., have -mántam. Ppp. reads sarasvatī: manvavittaṁ jagāma. Hardly a ms. gives an accent to jagāma (one of SPP's, probably by accident, and another, p.m.). Ppp. reads in d sandadātu. There are corresponding verses in VS. (xxxvi. 2) and Āp. (xiv. 16. 1), but they cast no light on b: VS. reads yán me chidráṁ cákṣuṣo hṛ́dayasya mánaso vā́ ’titṛṇṇam bṛ́haspátir me tád dadhātu; and Āp., yan me manasaç chidraṁ yad vāco yac ca me hṛdaḥ: ayaṁ devo bṛhaspatiḥ saṁ tat siñcatu rādhasā.
2. Do not ye, O waters, devastate (pra-math) our wisdom nor our bráhman; come ye flowing with easy flow, being invoked; [be] I of good wisdom, having splendor.
All the mss. accent ā́pas in a, and SPP. refuses to follow us in the obvious emendation to āpas, although the comm. also takes the word as vocative. At the end of b, the comm., followed by three of SPP's authorities, has mathiṣṭa naḥ. In c, the mss. read çuṣyadā́ (p. the same), and the comm. understands it as çuṣyat: ā; SPP. emends by conjecture to suṣyadā́s (p. su॰syadā́ḥ), which is decidedly more successful than our çuṣmadā́s. The translation, however, ventures to emend yet further, to suṣyádā́ (p. su॰syádā: ā́). Most of the mss. give syannadhvam; ⌊W's O. and SPP's carefully corrected Dc., s.m.,⌋ join with the comm. in syanda-, which stands in both printed texts. The translation, finally, implies our reading úpahūtās, against the mss., SPP., and the comm., which give -tas (-to ‘ham); against, also, the Anukr.,* but making much better meter than if c is ended with syandadhvam. The mss. and SPP. have sumédhās (the p.-mss. wrongly su॰médhā); ⌊in the edition⌋ we ⌊should have⌋ rectified the accent ⌊so as to read⌋ sumedhā́s. *⌊The Anukr. would scan as 6 + 8: 8 + 12; the Berlin text as 6 + 8: 12 + 8.⌋
3. Do not ye injure our wisdom, nor our consecration, nor what ardor (tápas) is ours; be they propitious to us in order to [prolonged] life-time; let them become [our] propitious mothers.
The mss. have in b hiṅsiṣṭam, and SPP. does not follow our obviously acceptable emendation to -ṣṭa. The comm., too, has -ṣṭam, and explains it by calling ⌊into service⌋ that everlasting pair dyāvāpṛthivī; but, as alternative, he apparently goes on to anticipate the pair of Açvins from vs. 4; there is, however, a lacuna in the sole manuscript, cutting off this explanation almost at the beginning. In c, the mss. all have sáṁsvanta ā́y- (p. sám॰svantaḥ!), but the comm. çaṅsantu, explaining it as = stuvantu; ⌊SPP. reports that the text of the comm. is çaṁsantu and that the sole ms. of the commentary has actually santu;⌋ SPP. takes this, and alters it to çáṁ santu; our emendation to simple santu, suiting both sense and meter, is obviously to be accepted. The pada-mss. in general give çivā́ in c; ⌊but one of W's and one of SPP's, each s.m., have çivā́ḥ;⌋ they all ⌊except perhaps W's L.⌋ have çivā́ḥ in d. There is no good reason why the Anukr. should call the verse bṛhatīgarbhā, since c can be read as eight syllables in either form.
