Homonymy in verbal roots in Sanskrit is a recognized phenomenon. Well-known examples of this are found in such cases as the following: ūh- ‘to remove’, ūh2 ‘to consider’; kṛt-1 ‘to cut’, Kṛt-2 ‘to spin’; gā-1 ‘to go’, gā-2 ‘to sing’; ci-1 ‘to gather’, ci-2 ‘to observe’; dhā-1 ‘to place’, dhā-2 ‘to suck’; bhuj-1 ‘to bend’, bhuj-2 ‘to enjoy’; vṛ-1 ‘to cover’, vṛ-2 ‘to choose’. In the case of vas- we find three homonymous roots (vas-1 ‘to shine’, vas-2 ‘to clothe’, vas-3 ‘to dwell’), and in the case of dā- no less than four (dā-1 ‘to give’, dā-2 ‘to divide’, dā-3 ‘to bind’, dā-4 ‘to clear’). In such cases the various meanings are so distinct from each other that the separation into two or more roots is done without any difficulty, and there is no doubt that we have here different roots, although they are identical in form. In practice, of course, the homonymy is considerably restricted by variation in inflection and derivation (vásati, váste, uccháti, etc.).
page 191 note 1 For a comprehensive discussion of this subject see Gonda, J., Zur Homonym ie im Altindischen), Acta Orientalia, 14 (1936), 161–202Google Scholar. Ancient Indian doctrine on this subject is summarized by Raja, K. Kunjunni in Adyar Library Bulletin, 19 (1955), pp. 193–222Google Scholar.
page 191 note 2 Kuiper, F. B. J., Acta Orientalia, 16 (1938), pp. 313ff.Google Scholar
page 191 note 3 Bailey, H. W., TPS., 1953, pp. 33ff., and 1955, pp. 59 ff.Google Scholar
page 191 note 4 See BSOAS., xvii (1955), pp. 326ff.Google Scholar
page 194 note 1 Oertel, H., JB Roots and Verb Forms, Journal of Vedic Studies, ii, 74Google Scholar.
page 196 note 1 Corrected by Bloomfield, (AJPh., xii, 422)Google Scholar from Bibl. Ind. Ed. kuśāh. His translation ‘smooth the kuśa- grass’ is not in accordance with general meaning of lubh- as seen from the other passages.
page 199 note 1 Cf. also Deśī luṭṭha- ‘fragments of brick’, < *luṣṭa-.
page 200 note 1 See Lüders, , Philologica Indica, p. 775Google Scholar.