Etymon: paTTaN, pattan, patan

[Dominique.Thillaud]

At 1:36 +0100 15/11/97, DR.S.KALYANARAMAN wrote:
>
>Close to the Sarasvati River, we have names of ancient settlements such as
>Prabhas Patan (SaurASTra), Pattan Minara (Bahawalpur Province) (cf.
>Mughal, Ancient Cholistan, 1997 where a 'pre-Islamic' shrine is identified
>on the banks of the River).
>
>paTTaNa is a common suffix in names of ports such as chennappaTTaNam,
>macilIpaTTaNam, vizAkhapaTTaNam.
>
>Is the etymon paTTaN related to an ancient port of call? Is it related to
>paTTi, a street? If so, Pattan Minara could have been a terminal port town
>on the desiccated river?
>

        In an article of Language, 12, 133, R.A.Hall jr. suggest to link
paTTana <? *partana with the Latin portus "passage, door, haven", a
well-known eurindien word (see engl.ford, &c.). In Indo-Iranian, we have
the av. peretuS "ford" but the root is considered as unknown in Sanskrit.
This root is in fact linked to *per- "to cross" and (except perhaps the old
alternance r/n) without direct rapport with path, panthAs, pontus, pontos.
        But your examples are very interresting ones; do you have name
alternances with tIrtha ? or saras ?
        We have perhaps a borrowing with pattana "town (MBh)" but, indeed,
the evolution 'rt' > 'TT' seems a bit strange (but phonetically very
acceptable). Do someone have an other example of it ?
        Regards,
Dominique

Dominique THILLAUD
Universite' de Nice Sophia-Antipolis, France