Pali language svara usage

[Satyan Sharma]

Namaskara,

My question is: Are udatta, anudatta and svarita (all three or at least one) used in Pali language?

Kindly guide.

Regards

Satyan

[Madhav Deshpande]

The Pali grammarians do not mention any of the Vedic svaras, although I have talked to several monks, some of whom were my classmates at the Pune University, and they told me that there were standardized ways of reciting the Pali sutras and they recited some for me.  From a historical perspective, the effects such as loss or reduction of unstressed syllables and expansion of the stressed syllables that are seen in forms like अस्ति/स्त:/सन्ति, नौमि/नुव:/नुम: have some survivals in Pali forms.  One sees reductions of api>pi, iti>ti, idānīm> dāni where accent-induced changes can be detected.  But there is no explicit instructional use of Vedic accents in relation to Pali, as far as I know.

Madhav M. Deshpande

Professor Emeritus

Sanskrit and Linguistics

University of Michigan

[Residence: Campbell, California]

--
You received this message because you are subscribed to the Google Groups "भारतीयविद्वत्परिषत्" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bvparishat+...@googlegroups.com.
To post to this group, send email to bvpar...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Ramakrishnan]

Satyan ji,

What exactly do you mean by Pali? The reason I am asking you this question is because I recognize at least 4 broad stages in what you call Pali, and different people use different definitions, so it is better to be more precise:

1. The original spoken language comprising the actual spoken dialects used in the earliest stages of Buddhism by the first one generation of Buddhists.
2. The pre-standardized written language of the earliest pre-Pali Tripitaka (which only contained the earliest texts of the current tripitaka).
3. The linguistically standardized written language of the full Pali canon as we have it today.
4. The late-Pali that we find in the written commentaries and subcommentaries to the Pali canon, and other historical chronicles, biographies & philosophical literature (such as the Dipavamsa, Visuddhimagga, Dhammasattha etc).
   

I do not view the first two stages above as Pali, they can be viewed as the ancestors of Pali.

As you can see from the above the Pali we have now is not (and has never been, as far as I know) a spoken language or vernacular dialect of any ethnicity. It is by definition the frozen literary register of the Pali tipitaka. However, Pali was based on an actual spoken language used by the first generation (or first few generations) of Buddhists.

I have read a story (I think from a vinaya text) where some monks were trying to chant some Buddhist texts with Vedic accents, this would probably fit the stage 1 above. It would appear from this story that the accents had become archaic but could still be used naturally in the spoken language (this would have been the case in Panini's lifetime).

Regards,
Ramakrishnan

[Abhinav Anand]

Please ignore the previous email.

Dear Satyan Ji,

Namaste

(i) According to the Pali Text Society, Pali-English dictionary edited by T. W. RHYS DAVIDS, and WILLIAM STEDE (1921), the Pali term Udatta (Sk. Udātta) is an adjective as elevated, high, lofty, clever in the post-canonical Pali literature, i.e., Netti-Pakarana: 7, 118, and 123. For more details see: DAVIDS, T. W. RHYS, and STEDE, W. (1921). Pali-English Dictionary. The Pali Text Society, p. 134.   

Sincerely yours,

Anand

On Mon, Mar 4, 2019 at 5:38 PM Abhinav Anand <anandabh...@gmail.com> wrote:

Dear Satyan Ji,

Namaste

(i) According to the Pali Text Society, Pali-English dictionary edited by T. W. RHYS DAVIDS, and WILLIAM STEDE (1921), the Pali term Udatta (Sk. Udātta) is an adjective, and used as elevated, high, lofty, clever in the post-canonical Pali literature, i.e., Netti-Pakarana: 7, 118, and 123. For more details see: DAVIDS, T. W. RHYS, and STEDE, W. (1921). Pali-English Dictionary. The Pali Text Society, p. 134.   

Sincerely yours,

Anand

--

You received this message because you are subscribed to the Google Groups "भारतीयविद्वत्परिषत्" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bvparishat+...@googlegroups.com.
To post to this group, send email to bvpar...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Regards,  

Abhinav Anand

Mob: +91 9654409939

--

Regards,  

Abhinav Anand

Mob: +91 9654409939

[rshuklaps]

प्राकृत भाषा के व्याकरणकरों ने इस प्रकार के शब्दों की प्रक्रिया पर  सिद्धहेमशब्दानुशासन और प्राकृत प्रकाश के व्याकरणकरों को देखा जा सकता है। कोई श्लोक/वाक्य है तो शेयर किया जा सकता है।

सादर

Dr. Ranjish Shukla

[Abhinav Anand]

Dear Satyan Ji,

Namaste

(i) According to the Pali Text Society, Pali-English dictionary edited by T. W. RHYS DAVIDS, and WILLIAM STEDE (1921), the Pali term Udatta (Sk. Udātta) is an adjective, and used as elevated, high, lofty, clever in the post-canonical Pali literature, i.e., Netti-Pakarana: 7, 118, and 123. For more details see: DAVIDS, T. W. RHYS, and STEDE, W. (1921). Pali-English Dictionary. The Pali Text Society, p. 134.   

Sincerely yours,

Anand

On Mon, Mar 4, 2019 at 4:45 PM Satyan Sharma <satyan...@gmail.com> wrote:

--

You received this message because you are subscribed to the Google Groups "भारतीयविद्वत्परिषत्" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bvparishat+...@googlegroups.com.
To post to this group, send email to bvpar...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

[rshuklaps]

प्राकृत भाषा में इस तरह के प्रयोग मिलते हैं। पालि से प्राचीन प्राकृत व्याकरण है।

Dr. Ranjish Shukla

--

[Ramakrishnan]

Satyan ji,

To add to what I said below yesterday, I have found the text from the pali canon which speaks about the first generation of buddhist monks reciting buddhist texts with the svaras.

This text is titled the Soṇa-sutta, and mentions that there was a monk (called Soṇa Kuṭikaṇṇa in Pali) who upon being asked by the Buddha, recited the oldest known text in the Pali canon (the aṭṭhakavagga) with svaras. Having heard the aṭṭhakavagga being recited clearly with svaras, the Buddha greatly applauded him for clearly pronouncing the text with a proper grasp of the meanings.

As I said, this scenario fits the first stage (spoken language of the first generation of Buddhists) of my four-stage classification of Pali's history when it was not only possible but natural to use the svaras in the spoken language during the Buddha's lifetime.

Regards,
Ramakrishnan

[Satyan Sharma]

Thanks a lot Sir.

Regards

--

Satyan Sharma

Research Scholar

Department of Sanskrit

Panjab University, Chandigarh, India.

Note:

This email (and any attachments) is confidential and may contain personal views. If you have received it in error, please delete it from your system. Do not use, copy or disclose the information in any way unless specified. Further communication will signify your consent to this.

[Satyan Sharma]

Thank you so much, for sharing this Sir. I still

have a doubt whether the word ‘sara’ (skt. svara) used here actually represents the three svaras (udatta etc.) or not. This intonation could also be just at the level of words and sentences and not at the level of syllables. As Panini mentions ‘rules’ of the non-vedic speech to specify the syllable on which the udatta accent should be placed (for eg. sutra 6.2.134), I am curious as to whether there is a textual evidence regarding the use of svaras within a word in Pali?

Regards

Satyan

--

You received this message because you are subscribed to the Google Groups "भारतीयविद्वत्परिषत्" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bvparishat+...@googlegroups.com.
To post to this group, send email to bvpar...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Satyan Sharma]

धन्यवाद महोदय।

सादर

[Madhav Deshpande]

I am not aware of any evidence of Vedic svaras for Pali texts.  Some Brahmin monks propose to recite the Pali texts (छन्दसो आरोपेम) either with Vedic accents or render them into Sanskrit of some form, and the Buddha is supposed to have forbidden this, and asked that his monks communicate his teachings in the local languages.  The wordings of these Pali passages have been disputed in research literature and there exist multiple interpretations.  The term सर in Pali grammars refers only to vowels, and not to Vedic accents.  The grammars of Pali [Moggalāna and Kaccāyana] are later than Buddhaghosa, who declares that Pali/Māgadhi is the original language of all beings [सब्बसत्तानं मूलभासा / सर्वसत्त्वानां मूलभाषा], and does not derive it from Sanskrit.  The Pali grammarians follow the style of the Sanskrit grammarians [more like Kātantra than Pāṇini], but describe Pali as an independent language, while the Prakrit grammars describe Prakrits always in relation to Sanskrit.  That is a significant difference between the Pali and the Prakrit grammatical traditions.  The Pali grammars were composed outside of India in places like Sri Lanka and Burma, while the Prakrit grammars were all produced in India.  The Pali that is available to us was already transported to Sri Lanka and codified there, and then transmitted further to Southeast Asia.  It was preserved and studied in relative separation from India and Sanskrit.

Madhav M. Deshpande

Professor Emeritus

Sanskrit and Linguistics

University of Michigan

[Residence: Campbell, California]

[Ramakrishnan]

Madhav ji,

Many thanks for your observations on this. You have said "The term सर in Pali grammars refers only to vowels, and not to Vedic accents", and I agree.

However, this does not mean that the use of the word sara (svara) in the existing Pali grammars (which as you have said are all dated more than a 1000 years after the lifetime of the first generation of Buddhists) must ipso facto always agree in meaning with the use of the same word in the Pali tripitaka.

If we look at the text cited by me (the relevant extract from which I have quoted below), the phrase 'sarena abhaṇi' ("recited with svaras") cannot mean "recited with vowels" (because that option does not exist to recite a text without vowels); therefore in this case it can only mean 'recited with the tonal accents. This must also mean that it was possible to omit the accents, and therefore there was a real option available to the reciter to include or omit them.

āyasmā soṇo bhagavato paṭissutvā soḷasa aṭṭhakavaggikāni sabbāneva sarena abhaṇi. Atha kho bhagavā āyasmato soṇassa sarabhaññapariyosāne abbhanumodi: “sādhu sādhu, bhikkhu, suggahitāni te, bhikkhu, soḷasa aṭṭhakavaggikāni sumanasikatāni sūpadhāritāni, kalyāṇiyāsi vācāya samannāgato vissaṭṭhāya anelagaḷāya atthassa viññāpaniyā

To be sure, Pali does not use (and has never used in my understanding) tonal accents so there is no disagreement about that. I do not postulate that in this case the bhikṣu was speaking any kind of Pali. 

However, it is equally clear that the language that the first generation monks spoke had the accents, and those who recited thus were appreciated by the Buddha himself for reciting well.

-------------
Sakkāya niruttiyā:

I am aware of the 'Sakkāya niruttiyā' controversy, the controversy exists because the text is misinterpreted by some scholars. The issue in there is not a distinction or preference between sanskrit and prakrit. Let me explain:

Two bhikṣus approach the Buddha and tell him "etarahi, bhante, bhikkhū nānānāmā nānāgottā nānājaccā nānākulā pabbajitā. Te sakāya niruttiyā buddhavacanaṃ dūsenti Handa mayaṃ, bhante, buddhavacanaṃ chandaso āropemā".

In this quote above, clearly the problem mentioned by the bhikṣus is not the existence of multiple languages or dialects. The problem posed is that the societal diversity of bhikṣus (with various names, gotras, jātis and kulas) is indicative of people with differing levels of education/understanding, and when they become parivrājakas, their different levels of understanding leads them to (mis)interpret the buddhavacana in their own peculiar ways. This is what is mentioned in the text, the text is not speaking of multiple languages at all.

The buddhavacana is evidently already available in a specific dialect/language, and the problem posed by these nānāvidha bhikṣus is that due to the differences in their social backgrounds, they understand the already existing buddhavacana (in an already existing dialect) in their own peculiar ways. Now the solution these two enterprising bhikṣus propose is if they redact the buddhavacanam in chandasaḥ (metric poetry), that would promote greater standardization in understanding (much like the texts used by the various vedic śākhas) as that would subject it to similar rigorous pedagogical standards and methods in their dissemination. The Buddha here, concerned that this solution would prevent wider dissemination of his teachings, objects to the idea and allows the nānāvidha bhikṣus to understand it in whichever way they could (their social and educational backgrounds notwithstanding).

This does not conflict with my earlier example where the Buddha praises the bhikṣu who utters a text with svaras for reciting it well. 

Regards,
Ramakrishnan

[Ramakrishnan]

Satyan ji,

Apologies if I was not clear. I do not postulate that the monk was speaking Pali. As I told you, the spoken language of the first generation of buddhists that had the pitch accents was certainly not any kind of Pali.

Pali was never meant to be (or was) a 'spoken' language of any ethnicity. It is by definition the literary (purely written) language of the Pali canon (and derivative literature). There has never been a standalone spoken-Pali outside the Pali canon. So the language spoken by the first generation of buddhists in the given example was not Pali but an earlier language that had the tonal accents (and where it was optional to include or exclude them, as the quoted example makes clear).

Please see my fourfold chronological classification of the language of the Buddhist Tripitaka. There is no Pali in the first two stages, Pali only takes birth in the 3rd stage and is almost extinct in the fourth stage. If you disagree with this fourfold classification, I would appreciate to know the reasons why.

Regarding your comment that the svaras with which the text was being recited could be something else other than the tonal accents, in my opinion that would be extremely unlikely as the already known meanings of svara (or sara in Pali) are our only alternatives. 'Sarena  abhaṇi' cannot mean recited 'with vowels', therefore in my understanding it means 'recited with the pitch accents'.

Thanks,
Ramakrishnan

[Madhav Deshpande]

Dear Shri Ramkrishnan,

     I understand your interpretation of sarena [=Skt. svareṇa] in the passage you quote.  I am not sure I agree with you that this refers to recitation with Vedic svaras.  This, in my opinion, refers to the beautiful voice of the reciter, which is later appreciated with the words: kalyāṇiyāsi vācāya samannāgato vissaṭṭhāya anelagaḷāya atthassa viññāpaniyā.  Be that as it may, whatever the original form of Pali, it shows loss of syllables in forms like api>pi, iti>ti etc., which look like loss of unaccented syllables, and it is likely that that original language may have had accents.  However, no such notion of recitational accent has survived in the Pali tradition.  Effects of accents on the morphology of Sanskrit continue long after the accents themselves cease to be used in the pronunciation of Sanskrit.  Something analogous may have happened in the case of Pali.  While the accents of Sanskrit have been preserved in the Vedic literature, and hence we have a full knowledge of that accent system, the accents of Pali and Prakrits have not been preserved, and we can make only some conjectures based on what we notice in the morphologies of those languages.  With best regards,

Madhav M. Deshpande

Professor Emeritus

Sanskrit and Linguistics

University of Michigan

[Residence: Campbell, California]

[Ramakrishnan]

Dear Madhav ji,

Thanks once again for your thoughts. You say the word sarena refers to the beautiful voice of the reciter, rather than the (optional) employment of svara (tone/pitch) in recitation.

I am not sure if you mean to say svara would mean "good voice" or just "voice". If svara is supposed to specifically mean good voice, or specifically show the quality of the voice, then I havent come across such a sense anywhere (whether in Pali or Sanskrit), and I would be grateful if you point out such a usage.

If sarena were to simply mean "with a voice" that would be even more odd.

If it were a good voice, he would not have an option to use a good voice (as opposed to deliberately using a bad voice). Here the sense of the word is that he had an option to recite it with "sara" or not, and he did, that is what makes it worthy of mention. It appears that the Buddha was not impressed simply by the voice but by the manner of recitation, and it is that manner of recitation that is meant by the use of the word sarena.

There are other texts in the Pali tripitaka that mention people with beautiful voices, but this word (sara) has not been used in such contexts. Good voice is usually called kalyāṇa-vāk- in the tripitaka, not 'sarena'.

Therefore it appears to me that sarena has a different sense than 'good voice'.

Regards,
Ramakrishnan

[Madhav Deshpande]

Dear Shri Ramakrishnan,

     My description of sarena in Pali as referring to beautiful voice was not intended to give the literal meaning of the word sarena, but what this refers to in the context of the appreciation that follows: kalyāṇiyāsi vācāya samannāgato vissaṭṭhāya anelagaḷāya atthassa viññāpaniyā.  That appreciation seems to expressive qualities of the vācā of the monk.  The none of the adjectives kalyāṇi, visiṭṭhā, anelagalā, atthassaviññāpanī refer to anything that we can call Vedic accents, but qualities like beautiful, distinctive, not slurring, and clearly communicating meaning.  In the context of these adjectives, I described sarena as beautiful voice.  In any case, as I mentioned in my previous email, whatever the spoken language it was that formed the basis of Pali, it could have had accents in its spoken form, and some of this can be inferred from reductions like api>pi, iti>ti, iva>va etc.  With best regards,

Madhav M. Deshpande

Professor Emeritus

Sanskrit and Linguistics

University of Michigan

[Residence: Campbell, California]

[Madhav Deshpande]

Sorry for my sloppy English.  Not yet fully awake.

Madhav M. Deshpande

Professor Emeritus

Sanskrit and Linguistics

University of Michigan

[Residence: Campbell, California]

[Ramakrishnan]

Dear Madhav ji,

Thanks again for your response.

The word sara is again used in the same text in the phrase 'soṇassa sarabhaññapariyosāne abbhanumodi' (śravaṇasya svara-bhāṇa?-paryavasāna- abhyanumud-) here again we see that it is not primarily his inherent voice that is being referred to but his paryavasāna of the 'sarabhañña' recitation.

The meaning of the word 'sarabhañña' is listed in the Concise Pali-English Dictionary of A.P. Buddhadatta Mahathera as "[nt．] intoning； a particular mode of recitation". 

Similarly, the Pali dictionary published by the Vipassana Research Institute also mentions sarabhaññaṃ as "A particular mode of reciting or intoning sacred compositions"

I agree when you say the adjectives you mentioned reflect the qualities of the voice and the reciter (and not the accent), but they are simply a stock phrase (or a canned phrase) used over and over again in other suttas of the pali canon (even where the voice is not described as 'sara' in the other cases).

For example in a text called the Ovādasutta, we find this canned phrase again with the same adjectives (kalyāṇavāco hoti kalyāṇavākkaraṇo, poriyā vācāya samannāgato vissaṭṭhāya anelagaḷāya atthassa viññāpaniyā....) where the voice or recitation is not described with the words sarabhañña or sarena.

In another text called the Alaṃsutta, we again find the same stock phrase (dhammānudhammappaṭipanno ca hoti; kalyāṇavāco ca hoti kalyāṇavākkaraṇo, poriyā vācāya samannāgato vissaṭṭhāya anelagaḷāya atthassa viññāpaniyā...) where again the recitation or voice is not described as sarabhañña.

Therefore, I do not believe these adjectives describe the 'sara' in this case but are just a stock phrase inserted into the text as in other cases where the Buddha praises a recitation (and apply regardless of whether the recitation was sarena or not).

Regards,
Ramakrishnan

[Satyan Sharma]

Thanks to Ramakrishnan ji and Madhav ji for such informative discussion. As I had mentioned earlier, ‘sara’ here may also refer to sentence stress. Putting stress on the wrong words in a sentence may misrepresent the

meaning of it. In that sense, the Buddha’s appreciation might equally apply to the use of sentence stress correctly.

Regards

Satyan

[Ramakrishnan]

Namaste Satyan ji,

Would you kindly share the source (or logic) of your interpretation of the word 'svara' as 'sentence stress', either in Pali or Sanskrit? I have not heard this interpretation anywhere else so it is a bit hard to understand how it could mean sentence stress.

I do not believe it means any kind of stress, not even word stress, let alone sentence stress, the most straightforward meaning (of the cited words sara & sarabhañña in this context) is  'intonation' i.e. the pitch/tone accent (of Old-Indic?). Svara literally means vowel, and the intonation (pitch accent) applies to vowels. It does not apply to words/sentences, and it does not mean 'stress' as far as I know.

This tone/pitch accent only exists in Vedic as a mandatory/native feature (and it gradually becomes optional/scholarly in Classical Sanskrit). When I say Classical Sanskrit here, I am not restricting it purely to the Paninian Grammatical/Phonetic Standard dialect (i.e. the language that complies entirely with the Ashtadhyayi), but I am including all the contemporary spoken Old-Indic dialects which directly derive from late-Vedic historically (evidences of the existence of which are found all over in BCE texts), i.e. all the Old-Indic dialects that Panini must have used to standardize their grammar (one of which evidently was the dialect in which this intoned recital praised by the historical Buddha occured).

My point here is only to emphasize that this recital was in Old-Indic; not in any kind of middle-Indic or Pali. 

Let me take this line of argument a bit deeper than what I have so far stated (for whatever it may be worth):

In the conventional understanding, one may ask the question "why hasn't the Tripitaka made it clear that the recital was in a Sanskrit dialect (i.e. a dialect that could have accents optionally), rather than in Pali, if the bhikshu suddenly shifted to Sanskrit from Pali (or whichever middle Indic dialect that he was previously speaking to the Buddha in)".

I would instead invert this question and say "Why does the Tripitaka, which narrates so many dialogues, have no sanskrit vs prakrit alternations in those dialogues; à la the prakrit dialogues in Sanskrit dramas)

The reason why the Tripitaka, or any other co-eval early literature for that matter (including Sanskrit texts), do not make this distinction clear (between old-Indic and middle-Indic speech as narrated in the suttas) is because no such distinction appears to have existed until the early centuries CE and all historical speech narrated in the Pali canon was in Old-Indic, although it has been ultimately redacted/written in middle Indic (Pali).

The evolution from Vedic to pre-standardized Middle-Indic (i.e. early Gandhari) and later to standardized middle-Indic (late Gandhari & Pali) is more or less a set of sudden/rapid (predominantly orthographic in the early stages) innovations rather than a gradual natural evolution of the spoken language. No evidences (either literary or grammatical) of any intermediary evolutionary stages between Old-Indic and middle-Indic have been attested in literature (because there was no such gradual evolution). Nor are such gradual evolutionary shifts from Old to Middle Indic likely to be found in the future if my understanding is correct. My surmise is that middle-Indic innovations largely arose in the 4th century BCE due to middle-Iranic influences on Old-Indic, and the arrival of writing in Aramaic/Kharosthi/Brahmi (from Persia proper to the newly Persianized NorthWest-India in the mid first millenium BCE) and attendant orthographical innovations.

It would have been too obvious in the late centuries BCE that middle-Indic had no natural spoken existence (too obvious to even be worthy of a mention), although the evolution of scripts/writing were slow to catch up with the siksha treatises (leading to the emergence of artificial 'ghost'/paisaca languages like Gandhari, Pali and Ardha-Magadhi in the intervening period that are attested in writing but were not spoken phonetically). 

I think this incident (where the aṣṭakavarga/aṭṭhakavagga, being chronologically the earliest known text in the Pali canon, is chanted with intonation) is historical because it is very unlikely to have been inserted/redacted as such if it actually did not happen.

Ramakrishnan

[Satyan Sharma]

Namaste Ramakrishnan ji

My interpretation of 'sara' as sentence stress is based only on a possibility that it could mean things other than the 3 kind of svaras. There is an explanation of sarena bhaṇanaṃ where it is said "Sarabhaññe kira taraṅgavattadhotakavattagalitavattādīni dvattiṃsa vattāni atthi"  (link https://tipitaka.org/romn/cscd/vin02a3.att4.xml). Here there is a mention of vattas (vrttas) like taranga, dhotaka and galita, which are 32 in total. I haven't yet found any such vrttas/vattas elsewhere. This might provide some insight into what has been meant by 'sara' in the sutta being discussed.

I am not sure if sudden or rapid innovations can lead to such a major change in a language. Also how do we establish that Gandhari, Pali, Ardhamagadhi languages weren't spoken phonetically? 

Regards

Satyan

[Ramakrishnan]

Dear Satyan ji,

Thanks for the quote from the commentary for Sarabhañña: "Sarabhaññe kira taraṅgavattadhotakavattagalitavattādīni dvattiṃsa vattāni atthi" - it does not seem to make much sense to me. 

Please share your understanding of it. What is a taraṅga (wave?), dhautaka (rinsing/washing?), galita (oozing/trickling?) got to do with sarabhañña?

The use of the word 'kira' (=kila), especially in the commentary, appears to indicate it is based on hearsay (kirasaddo anussavane), so I am not sure what to make of this. Perhaps the commentator is commenting based on hearsay as he himself is not sure of what it originally meant?

If you look at another commentator (from the Majjhima Nikaya Tika), he says "sarena suttassa uccāraṇaṃ sarabhaññaṃ" which to me makes better sense but is not adding anything beyond the obvious.

In another instance, the commentary says "Sarena abhāsīti suttussāraṇasarena abhāsi, sarabhaññavasena kathesīti attho. sarabhaññapariyosāneti ussāraṇāvasāne" which does not make much sense to me again. Is this ussāraṇa = uccāraṇa, or is it something else?

I think based on the above, using the pali commentaries (that are nearly one millenium younger than the mula texts and dont appear to belong to a continuous commentarial tradition) to understand the mula text is best avoided as they appear more likely to mislead than elucidate.

-----------

About speaking Gandhari, Pali, Ardha Magadhi phonetically, first we need to establish that they were spoken languages before we establish that they were spoken/written phonetically or non-phonetically.

Is it not interesting that these names (Gandhari, Pali, etc) themselves are modern names, there are no references in BCE literature to any such languages being spoken anywhere, there are no grammars, no dictionaries, no quotes, nothing. Paisaci is mentioned (and to me it approximately means gandhari) but even the name Paisaci is not found in BCE texts as the name of a spoken language, it is only named in CE literature as the name of a prakrit.

If we compare the edicts of Ashoka, he has inscribed the same identical texts (in obviously the same language) at multiple geographical locations, but the spellings of the words differ. How do we establish which one was the standard spelling and which one was the non-standard spelling? That would probably lead us to infer that the spoken language was not represented phoneticaly in writing.  

Regards,

Ramakrishnan

[Madhav Deshpande]

I can accept that sara in the Pali passage under discussion can contextually refer to "intonation" in a general sense, but I do not accept that it specifically refers to Vedic accents.  Already, from the three accent system of the early Vedic texts, we come to two accent system of the Śatapatha-Brāhmaṇa [the so called bhāṣika-svara of the Bhāṣikasūtra] and then Pāṇini saying that mantras are recited in ekaśruti, except in Japa, Nyūṅkha and Sāman recitals [yajñakarmaṇi ajapanyūṅkhasāmasu].  Buddha's  time coming somewhere between Pāṇini and Aśoka, the gradual loss of accentuation in Indo-Aryan, including Sanskrit is evident.  So, the chance that the language spoken by the Buddha in eastern India will have retained the three-accent system of the R̥gveda is almost nill.  I also do not accept that the ancient Prakrits did not have a basis in spoken languages or that they were merely a result of the arrival of writing systems.  Fully accurately or partially, the Ashokan inscriptions give us a dialect geography of Prakrits, which to a large extent is confirmed by the trends in the Prakrits found in literatures and Prakrit grammars.  I do agree that writing systems often give us an approximation of the spoken language, but that is the best we have for all pre-modern periods.

Madhav M. Deshpande

Professor Emeritus

Sanskrit and Linguistics

University of Michigan

[Residence: Campbell, California]

[Ramakrishnan]

Dear Madhav ji,

Many thanks for sharing your thoughts.

Given that the language that the first generation of Buddhists spoke is very likely one or more dialects of late-Vedic, or derivative dialects of late-Vedic, I don't know what to make of your opinion that it need not refer to some sort of a vedic pitch/tone accent.

I do not claim that the first generation of Buddhists spoke the Rgvedic mantra dialects (or followed their archaic system of accentuation). I am simply saying that their dialect was very likely an Old-Indic dialect, in which some form of historical Old-Indic tone/pitch accent was still possible (although probably not invariably used in common speech except in chanting the sacred texts during the lifetime of the first generation of Buddhists).

You have said the Buddha's time comes between Pāṇini and Aśoka which seems to slightly differ from my understanding - in my understanding the first generation of Buddhists precede Pāṇini (very likely by less than a century), or perhaps Pāṇini was a younger contemporary of the historical Buddha. The reasons why I believe this to be true are as follows:

• The word nirvāṇa as far as I am aware is not attested in non-buddhist literature before the time of first generation Buddhists (the reason why I keep mentioning first generation Buddhists, instead of the Buddha, is because I have no idea of the historical Buddha's corporeal identity, and to therefore speak about the Buddha as a person without knowing anything about him from independent sources is not probably very productive. We can however talk of the first generation of Buddhists as a group without knowing their individual identities).
• Since Pāṇini  (in sūtra 8.2.50) gives an irregular derivation for 'nirvāṇa' (nirvāṇaḥ avāte), and hints that it means death (or rather the absence of breath - noting avāte is in the locative case), it is my understanding that he primarily had the death of the historical Buddha in mind, and this must have probably happened in his time or slightly before.
  

You have also mentioned that the Buddha was from eastern India, this is entirely at variance with my understanding (that the Buddha lived primarily in the Punjab region, and also probably in Rajasthan & Gujarat) for the following reasons:

• Most or all of the early Buddhist reliquary sites are based in the Punjab region as far as I am aware.
  
• Gandhari is a western language, Pali is a western language, Sanskrit without doubt is originally a north-western language. These are the languages in which early buddhist texts are found in.
• There are many brahmins named 'Todeyya' (Skt: taudeya) mentioned in the tripitaka who live in the area of Kosala where the Buddha usually lives. One of them is mentioned in the tripitaka as the purohita of the Kosalan monarch Prasenajit (Pali: Pasenadi). Pāṇini not only knows the existence of this tūdī grāma but gives a rule (in sūtra 4.3.94) to explain that 'taudeya' is a person from this place. This tūdī is mentioned with other north-western-Indian locations like śalātura, varmatī & kūcavāra.
  
• As per the tripiṭaka, the historical buddha was initially a disciple of Ārāḍa kālāpa (Pāli: Āḷāra Kālāma). The kālāpas are usually mentioned together with the kaṭhas, and Pāṇini has some sūtras mentioning them. The kātantra-vyākaraṇa is also called the kālāpa vyākaraṇa. So the historical buddha was educated by yajurvedins (if we go by the Pali canon) and he was evidently particularly close to the kaṭhas & kālāpas early on. 
  
• The pali tripiṭaka mentions the buddha meeting pauṣkarasādi (pokkharasāti), a very famous brahmin living in Kosala who had received a town (called ukkaṭṭha in Pāli) as brahmadeya from the emperor (Prasenajit). This (pauṣkarasādi) is a personal name that seems quite popular in late-vedic India as it is mentioned in the śāñkhāyana āraṇyaka, taittirīya prātiśākhya etc. It is hard to assume that the loci of these texts were in eastern UP or Bihar in the time of the first generation buddhists, rather than in the (yajur)vedic heartland - the Punjab.
• Coming to Rajasthan, Aśoka's 2 edicts at Bairat saluting the saṅgha there (and not anywhere else as far as I know) probably means that he expected his edict to be seen there by a very large number of buddhist monks (probably more than anywhere else). Perhaps Rajasthan was a buddhist stronghold in his time, more than any other site in eastern India.
• If the understanding that Jains were living in the same area as early Buddhists is correct, Gujarat and Rajasthan were the strongholds of Jainism from its very inception, so we must give credence to the inference that Asoka saluting the sangha & talking about buddhism twice in his Bairat edicts rather than anywhere else is a pretry strong indication of it being a major buddhist stronghold before his time. I am not aware of Jains being particularly strong in the eastern UP or Bihar at any point of time.
• Most of Asoka's most popular major rock edicts are in western India (Kalsi, Mansehra, Girnar, Shahbazgarhi). The majority of his (and his predecessors') empire therefore was based in western India, Kalinga being a noteworthy exception.
  
• In my understanding Girnar is Girivraja (because its topography appears to uniquely satisfy the meaning of the word girivraja), and Pali is apparently based on a language that was spoken in that region. Therefore Magadha was originally in my understanding somewhere close to (or in) southern Rajasthan and Gujarat. Kosala & Kekaya were located in the Punjab region (and may have been neighbouring kingdoms). Kaikeyi prakrit is also based in western punjab.
• In edict 5 of Aśoka's major rock edicts, only the Girnar version mentions 'Pāṭalipute' while all other copies elsewhere replace it with 'hida'. I take this to mean that only people in the vicinity of Girnar (i.e. Girivraja as above) would actually know where Pāṭaliputa was (as it was proximal to this site) so it was relevant only there.
• In my understanding prakritic innovations first arose not in eastern India, but in north-western India (Old-Gāndhārī, as evidenced in the Aśokan edicts, is the earliest known 'prakrit', and Kharoṣṭhī the earliest known script that it used). The earliest attestations of Prakritic innovations I'm aware of are in Pāṇini (cf. the words naṭa and maireya in the Aṣṭādhyāyī).
• The buddha lived in the pre-writing era, and all prakritic innovations occur immediately after the introduction of writing, first in Kharoṣṭhī and later in Brāhmī. There is no evidence I know of prākṛtic innovations (as gradual evolutionary sound shifts unconnected to writing) before the introduction of writing. Please can you share if you are aware of any evidence of 'evolutionary' prakritic sound shifts spread over centuries leading to gradual prakritization of the dialects leading upto Pāṇini and beyond?
  
• The Aśokan edicts are all in the same language/dialect, and as far as I am aware, do not show any regional dialectal features. If we take one small example - consider the word kṛta, it is invariably represented as kaṭa/kata or kiṭa/kita in the aśokan edicts (sometimes both forms are found within the same text). This would, in my mind, imply that the underlying spoken dialect had kṛta and it was alternatively written as kiṭa or kita or kaṭa or kata (and it didnt matter because they were all equally lossy given the non-phonetic script being used). Phonetic representation of Old-Indic was evidently impossible in the early centuries of writing.

Best regards,
Ramakrishnan.

[Madhav Deshpande]

Dear Shri Ramakrishnan,

     Thank you so much for such a long discourse.  You have given me much to think about.  I appreciate your patience with me, though I may not always agree with all the opinions expressed.  With best regards,

Madhav M. Deshpande

Professor Emeritus

Sanskrit and Linguistics

University of Michigan

[Residence: Campbell, California]

[Ramakrishnan]

Madhav ji,

Many thanks for your time and responses so far, and apologies if I have taken too much of your time.

Regards,
Ram