¹ Renou's view that all third singular and plural middle presents without dental are due to influence of the perfect does not directly concern us here, since, although such forms are common (see my note 2), endings equivalent to specifically perfect middle endings do not occur elsewhere in the present paradigm except in the case of śṛṇviṣé. Renou goes on to say that the passive value of some of these forms without dental supports the view that they resulted from influence of the perfect. However, the most important consideration for êrnvisé is, as we shall note (§2), that śṛṇviṣé, -iré function not only as passives but also as statives.

² Other presents forming their third singular and plural middle without dental are (in the order of present classes): joṣe, mahe, stave, cité, duhé, bruve, vidé (vid ‘find‘), duhré, sunvê, hinvé, invire, ṛṇvire, pinvire, sunvire, hinviré, tanvire (Paip., see Renou, loc.cit.), vṛñnjé, gṛṇe; see Macdonell, Ved. gramm. §§422,451, 464, 470,476. Sève, classed by Macdonell as a third singular (§422), is a first singular in its only occurrence in the RV (10.95.8). Dadhé, given by him as a third singular (§458), is, in its two occurrences in the AV (8.5.6, 17.29), a first singular; see Whitney-Lanman ad loc. Macdonell classes , śáye, śére (AV) as presents (§451), but these are old perfects; see Thieme, loc.cit. Huvé, which Macdonell assigns to the sixth class (§430), is more probably a root form; see Renou, Mél. Vendryès 310 (1925), and Tedesco, Lg. 20.215 (1944). Tośé, considered a first-class present by Macdonell (§422), is interpreted by Oldenberg, Noten ad 4.38.1, as an unreduplicated perfect, so that it and tośate can be interpreted like , īśate; see Thieme, op.cit. 45. Whitney, Gramm. §699b, considers pinvire, hinviré possible unreduplicated perfects.

³ Renou, Etudes véd. et pāṇ. 7.51: ‘type de variation morphologique pure.‘

⁴ Although it is not from a root having Zustandsbedeutung, śrutá- ‘renowned’ does not have preterital meaning. It is thus a member of the class, set up by Pāṇini as including verbs of desiring, knowing, honoring (3.2.188: matibuddhipūjārthebhyaśca), which make an exception to the general rule (3.2.102) ascribing preterital meaning to -tá- forms. On these forms see also Wackernagel-Debrunner, Ai. Gramm. 2.2.578.

⁵ The passive of śru, common in later Sanskrit, is extremely rare in Vedic, attested only once in the RV (10.22.1), not at all in the AV. An Avestan equivalent of śṛṇnvé ‘is renowned’ is not attested but the aorist middle occurs in the Avesta in this usage: Y 32.3 ... yāiš asrūdūm ‘by which ye are known’; see Humbach, Gāthās 2.33.

⁶ 8.66.9cd kéno nú kam śrómatena ná śuśruve janúṣaḥ pári ‘through what fame is the slayer of Vṛtra not renowned since birth?‘

⁷ See Renou, La valeur du parfait dan les hymnes védiques, ch. 8 passim. The active perfect śuśráva etc. is generally transitive.

⁸ It is not necessary, for our discussion, to consider the remaining occurrences. For some the passive meaning is clear, while for others one might argue for either passive or stative value.

⁹ Thus Geldner, Rigveda, and Renou, Etudes 5.16.

¹⁰ No doubt referring to Agni; cf. 1.96.8.

¹¹ On the meanings of dúvas see, most recently, Renou, Etudes 4.14.

¹² Interpreting vā as equaling vām; see Wackernagel-Debrunner 3.477 and Renou, Etudes 7.75.

¹³ On this verse and its relation to 4.8.6 see Bloomfield, RV Rep. 217.

¹⁴ This offers an example of the subordination of metrical to expressional considerations. Compare Thiemel statement, with regard to RV 3.59.2d (Mitra and Aryaman 44 note 31a) : ‘Seven instead of five syllables before the caesura. I should not consider correcting the line by a conjecture ..., but rather think that the poet by intentionally crowding his line wants to picture aṃhas.’

¹⁵ However, Oldenberg immediately says, ‘Aber wie entstand die Korruptel?’