origins of Pali

Michael Witzel witzel at FAS.HARVARD.EDU
Wed Jan 12 04:21:39 UTC 2000

>I am afraid we have to start from Adam and Eve again, which I am not
>really willing nor do I have the time to do:
>* Asoka prefers to write in the  local language/dialect (Greek /Aramaic in
>Afghanistan, or NW MIA dialect in NW Pakistan, with typical NW metathesis:
>dhrama for dharma, etc. etc.)
>* where he does not,this has to be explained, e.g. in S.India, where he
>unfortunately did not write in Dravidian but in MIA... (NB: I do not know
>of a cogent explanation for this last fact -- except maybe that Dravidian
>was not yet WRITTEN at that time ---; interestingly, as far as I remember
>here at home, these southern inscr. are in "Patna" chancellary dialect)
>* If one wants to deny the above two items, and go beyond circular
>arguments, one has to link the inscriptions with other well known facts --
>(not directly with the "modern" and medieval form of Pali; even the oldest
>MS, from Nepal, is only mid 1st mill.++, in Gupta type script)--
>-- but with  the (much) later gramm. descriptions or with the preceding
>Vedic dialects -- or even better, with near contemporary (early
>post-Asoka) inscriptions.

>We all know of course that Pali is a literary Koine with many dialect
>features (and Sanskritization even throughout the Middle Ages, standard
>examples : -sattva instead of -satta, brAhmaNa instead of bahmaNa etc.),
>---  but its basic dialect  has (hackneyed) western  nom. -o, not -e, has
>r not l, etc . etc.
>For some of the above features / discussions see C. Caillat and Witzel in:
>Dialectes dans les litt. indiennes, Paris 1989 (ed. C. Caillat), and see
>Caillat in: Inside the texts- beyond the texts,(ed. Witzel)  Cambridge
>1997 ( (
>); also  note Caillat's insistence on historical changes between late
>Vedic, early MIA, early Pali, and Asokan inscr., -- not to forget the
>complicated problem of Asoka's "chancellary" language as opposed to
>typical local dialect features of Maagadhii.

<more -- details -- in next message>

-------------- next part --------------
A non-text attachment was scrubbed...
Name: attachment.bin
Type: text/enriched
Size: 2038 bytes
Desc: not available
URL: <>

More information about the INDOLOGY mailing list