Sound Synthesis in Indo European Indo Iranian and Sanskrit

Author(s): Satya Swarup Mishra
Publisher: Ashutosh Prakashan Sansthan

Page 57

OLD INDO-AKYAN CONSONANT SYNTHESIS 39
59. m-tn > nn or mn in external sandhi; this is an OIA innovation; e. g. bhadram+nah > bhadran-naħ|bhadram -naħ. But in internal sandhi mt-n>mn; this is historical; e. g. šam-nā-li < IE kmm-na-ti, cp Gk kám-no < IE kmm-ni.
60. m>m before s, ş, s & h both in external & internal sandhi; e.g. tam-śiśum<tam-tśiśum, sam-saktaḥ< sam +-sal tah, sam-hatah<sam+hatah.
In IIr m>n before s; IIr ns > Skt ms; the other sandhi changes such as mś >ns > mś etc. have followed the change of ns > ms.
61. m>m before r always in external sandhi, excepting only one form sum rāt; e.g. sam-ramate < sam-ramate. But in internal sandhi m remains m before r; e.g. nam-ra.
Change of m>m before r may be a new formation in Skt; This may also be an IIr sandhi (vide 35) 62. m>m befo:e y, y & l in external sandhi. In vedic mn>y v ī (anasalized y y l) respectively before y v ! in external sandhi. But vedic manuscripts also show m instead; e.g. sam-yudhi or say-yudhi <sam-yudhi. yajñam vaștu < yajñam-vastu, sam-vartate < såm-vartate, samtāpaḥ < sam-tāpaḥ.
m>n before v in internal sandhi; e. g, jagan-vān< jagam-tvān. But m rewains m before y; e.g. gam-ya-te. Perhaps also m remains in before 1; e.g. am-la; this is of a doubtful root, no other suitable example is attested, for internal sandhi of m &l. The example apa-mlukta cited by Wackernagel & quoted by Macdonell (vide Vedic Grammar p. 68) is not an example of internal sandhi of m & I but that of conjuact ml,