I don’t agree with some of Stefan Karpik’s claims either that there was some kind of Oral-Transmission of EBTs in the first 3-4 centuries of Buddhism before they were finally written down. There is (in my knowledge of the facts) no evidence for any kind of early Oral-Transmission of the first 4 Nikayas of the Sutta Pitaka (in any language - Pali, Gandhari, or Sanskrit). Therefore the vast majority or entirety of the prose EBTs (excluding perhaps the verse suttas of the Atthakavagga, Parayanavagga etc whose oral-transmission is less clear) were in writing from the very beginning.
The idea that the Sakyas spoke Dravidian natively is similar to the Oral-Transmission hypotheses - it is entirely unsubstantiated and the facts speak against it. As a native speaker of Tamil (a Dravidian language with a textual tradition as old as that of Pali), I find the idea quite bizarre.
In fact, most Old-Tamil texts (written texts from BCE) speak of only one “வடமொழி” (literally meaning “Northern language” in Tamil), and that word refers to Sanskrit. The early Tamils to the best of my knowledge didn’t recognize the existence of (or name/identify) any other languages in Northern India in BCE. In early-Tamil texts, the association of Northern & Central India with Sanskrit speaking Indo-Aryans is firmly established - and there are thousands and thousands of Indo-Aryan loanwords still in use in daily speech from Sanskrit in all Dravidian languages (and in Dravidian literature from the late Mauryan era onwards). It is much harder to prove that the majority of Indo-Aryan loanwords in Tamil originate from Pali or a language similar to Pali - because the phonetic forms in which they are attested in Tamil do not betray such an ancestry.
Even Canonical Pali has no linguistic recognition in Tamil literature despite the fact that Pali has spent most of its existence in Tamil speaking areas. The conclusion that appears to me inescapable is that the early Tamils who studied Pali and practiced early-Buddhism must have thought of Pali as being a written-register of Sanskrit.
Do the scholars you follow know or read early-Dravidian texts; or consider/know all these stuff? I doubt it. The number of Pali and Sanskrit scholars who are also competent in Dravidian is probably very low.
Not really. Have you tried reading them yourself, or is that hearsay?
Here is Ashoka’s Rock Edict 1 from Girnar -
- the word forms that display uniquely Sanskritic phonetic features i.e. which cannot be phonetically Middle-Indo-Aryan, are in bold
- the ones not in bold also contain Sanskritic words but they are not unique to Sanskrit - such as rājā, na, jīvam, samājo, bahukam, sādhu, etc
- the rest of the words are not all necessarily identical in form to Pali either:
The edict
(A) iy[aṃ] dhaṃma-lipī Devānaṃpriyena Priyadasinā rāña lekh[ā]pitā
(B) [i]dha na kiṃci jīvaṃ ārabhitpā prajūhitavyaṃ
(C) na ca samājo katavyo
(D) bahukaṃ hi dosaṃ samājamhi pasati Devānaṃpriyo Priyadasi rājā
(E) asti pi tu ekacā samājā sādhu-matā Devānaṃpriyasa Priyadasino rāño
(F) pura mahānas[amhi] Devānaṃpriyasa Priy[a]dasino rāño anudivasaṃ bahūni prāṇa-sata-sahasrāni ārabhisu sūpāthāya
(G) se aja yadā ayaṃ dha[ṃ]ma-lip[ī] likhitā tī eva prāṇā ārabhare sūpāthāya dvo morā eko mago so pi mago na dhruvo
(H) ete pi trī prāṇā pachā na ārabhisare
The above is from the Girnar edicts, which have been suggested by several Pali scholars to be linguistically the closest to Canonical Pali. Even here (leaving alone the other locations), the evidence does not suit the theory (of Ashoka writing in only middle-Indo-Aryan).
How did so many words that display uniquely Old-Indo-Aryan forms (leaving alone other words that are common phonetically to both Sanskrit and Middle-Indo-Aryan) get in each line of Aśoka’s edicts? There are hundreds of such uniquely Sanskritic word forms if we study the entirety of the Ashokan textual corpus. Such uniquely Sanskritic word-forms are even more abundant in his Gandhari edicts. How can we brush these under the carpet and claim that Ashokan edicts are in Middle-Indo-Aryan across the board?
Even Sanskrit is mutually intelligible with Pali to pretty much the same extent as the Gandhari & Ashokan registers. Most people who know Classical Sanskrit well enough can comprehend Pali (and vice-versa). Most Pali words exist in either the exact same form or with slightly modified spellings in Sanskrit - so I don’t see comprehensibility as a relevant factor.
Pali is not alone in not having a name or an established linguistic identity before the common era. No other middle-Indo-Aryan language/dialect/register had such an established linguistic identity. None of the others too (Ardhamagadhi, Gandhari, Ashokan, etc) were recognized as vernaculars, or as dialects, or as languages by any name whatsoever - in the early-Jain sources, or in the early-Buddhist sources, or in the Ashokan edicts, or in early-Sanskrit texts. The idea that they were spoken languages thus has no underlying evidence. They are linguistic registers, sure, and they have texts written in them, but the existence of written texts does not prove that the spoken language was phonetically identical to how it is written in the texts.
Those were the first few centuries of writing en-masse in India - and it is not hard to imagine that the spoken language was simplified (using varying conventions) in order to write them conveniently/quickly. So some of the early Buddhists simplified it in a way that approximates what came to be called Pali later, while the early Jains simplified it in a way that came to be called Ardhamagadhi. Both of them were likely following the linguistic conventions and precedents set by the early-Mauryan (Ashokan etc) scribes - which is why they have common linguistic features. For the early-Buddhists, the ability to quickly copy mountains of canonical manuscripts over and over was of paramount importance in their quest to spread Buddhism all over, so they were driven by necessity to use the artifice of the phonetically and grammatically simplified language first introduced in early-Mauryan India for writing Brahmi texts on stone.
I don’t know of anyone else - and you are right, it is I myself who claims it - it’s my own original claim. Perhaps someone else has said it before, I’ll mention it if and when I come to know of it.
However the native texts from BCE are quite clear and forthright about it, and I find they make sense. I don’t see why the eye-witness accounts should not be trusted. The Sanskrit grammarian Patañjali, writing around 150 BCE mentions an excuse that people of his time offered as they refused to spend time learning Sanskrit grammar “vedān no vaidikāḥ śabdāḥ siddhā lokācca laukikāḥ, anarthakam vyākaraṇam” (“From studying the Vedas, we comprehend the Vedic language, and from speaking everyday with everyone, we know the classical sanskrit - hence studying grammar is useless either to know the Vedic or the classical Sanskrit”). So one century after Ashoka, people commonly felt studying a grammar to know their native day-to-day spoken Sanskrit was an overkill. The same source gives numerous such anecdotes of everyday Spoken sanskrit across the social landscape. So Sanskrit was spoken natively by all of Indo-Aryan India, not just by Brahmins, but they (the Brahmins) were certainly its most conservative exemplars.
Some questions are in order:
- What are those influences, and who has proved that they are not traceable to any other currently known language?
- Who says so, and how pervasive are such words/influences in Pali?
- Are those features unique to Pali?
- Did those features survive in later dialects?
- If not where did they go?
- Where are such variant dialects found attested before Buddhism?
I have read such claims and I find them greatly exaggerated. Those words that are cited as evolved from variant pre-Pali dialects are less than 1% of the Pali vocabulary - and they may very well be Iranic (Old-Persian / Median / Avestan / Scythian) loanwords - if at all they are of Indo-European origin. I dont even think such words exist in the later Pali texts or commentaries. So they are very likely Iranic loanwords from a time when the Achaemenid Empire ruled parts of North-Western India (which I believe was the homeland of Canonical Pali & Gandhari) from Persia. The period of Achaemenid suzerainty was when the Buddha lived and died.
They vast majority of Pali word forms are phonetically simplified forms of classical sanskrit words and are therefore capable of being etymologized from classical or vedic sanskrit. Can any scholar prove that this is not so?