Yarvāṇastarvāṇaḥ and Rishis located in southern India
Recently I came across a phonetic feature in Dravidian that reminded me of Patañjali's remark in the Paspaśāhnika of the Mahābhāṣya of two ṛṣis (remembered/named yarvāṇastarvāṇaḥ for their peculiar pronunciation) who supposedly pronounced 'yad vā naḥ tad vā naḥ' in sandhi as yarvāṇastarvāṇaḥ in common speech, while pronoucing those words correctly while performing yajña karmas where correct pronunciation was expected.
The comment by Patañjali goes like this:
“एवं हि श्रूयते- यर्वाणस्तर्वाणो नाम ऋषयो बभूवुः प्रत्यक्षधर्माणः परापरज्ञा विदितवेदितव्या अधिगतयाथातथ्याः। ते तत्रभवन्तो “यद्वा नस्, तद्वा न” इति प्रयोक्तव्ये “यर्वाणस् तर्वाण” इति प्रयुञ्ञ्जते, याज्ञे पुनः कर्मणि नापभाषन्ते।”
Some modern academic 'writers', writing about Sanskrit in western academia use anecdotal examples like this for pushing their point of view that these constitute evidence of prakrit speaking rishis.
I never came across any prakrit where this sound shift is either attested or is usual nor is Patañjali even talking about any non-Sanskrit speakers. That set me thinking about what this pronunciation could be originating from.
I then came across the name Sarguṇarāja in Tamil, which is the way Tamils normally pronounce Sadguṇarāja, and it seems Malayalis do a similar thing as well, pronouncing the sanskrit bhagavatpāda & utsava as bhagavalpāda and ulsava etc - so an Indo-Aryan dental (t / d)changes to (r / l) in certain phonetic environments in Dravidian.
So yad vā naḥ tad vā naḥ > yarvāṇastarvāṇaḥ is not prakrit but the Sanskrit pronounced like that in a Dravidian phonetic environment.
