Becoming a Pali Scholar

Explore the ancient language of the Tipitaka and Theravāda commentaries
Coyote
Posts: 845
Joined: Mon Sep 19, 2011 12:42 pm
Location: Wales - UK

Becoming a Pali Scholar

Post by Coyote »

I am now at the end of my third year of an undergraduate degree, Joint Greek and Classical Studies. Because of this I am thinking about what I want to do with my life, my future career and so on. I hope to take a year out teaching English in China which I am now applying for. Of course, a lot could change during that time, but one possibility would be to get back into academia after that. I am still unsure whether, if I decide to do that, I would continue my study of Greek and the Classical world. How feasible would it be to make the switch to Pali and early Buddhism, and how would I go about doing this? I have no understanding of Pali at the moment, but I imagine I would pick up the language well given my study of ancient Greek grammar as well as German, French and Welsh during high school.
Any help or comments are appreciated.
"If beings knew, as I know, the results of giving & sharing, they would not eat without having given, nor would the stain of miserliness overcome their minds. Even if it were their last bite, their last mouthful, they would not eat without having shared."
Iti 26
User avatar
Kare
Posts: 767
Joined: Sun Feb 01, 2009 10:58 am
Location: Norway
Contact:

Re: Becoming a Pali Scholar

Post by Kare »

Yes, your knowledge of classical Greek will be very useful if you start studying Pali. There are differences, of course, but there are also similarities which will make it much easier to understand the Pali grammar. Wishing you good luck with your studies! :twothumbsup:
Mettāya,
Kåre
Coyote
Posts: 845
Joined: Mon Sep 19, 2011 12:42 pm
Location: Wales - UK

Re: Becoming a Pali Scholar

Post by Coyote »

Thanks for the encouragement Kare!
Unfortunately it looks like the courses offering the Pali language are very scarce in the UK, which is not surprising. If anyone has any information about departments, scholars or courses that are highly regarded please let me know.
In the mean time I will probably try and get acquainted with the basics of Pali, something I have been meaning to do for a long time.
"If beings knew, as I know, the results of giving & sharing, they would not eat without having given, nor would the stain of miserliness overcome their minds. Even if it were their last bite, their last mouthful, they would not eat without having shared."
Iti 26
User avatar
waterchan
Posts: 699
Joined: Fri May 07, 2010 7:17 pm
Location: Kamaloka

Re: Becoming a Pali Scholar

Post by waterchan »

Coyote wrote:Thanks for the encouragement Kare!
In the mean time I will probably try and get acquainted with the basics of Pali, something I have been meaning to do for a long time.
If you read suttas at www.suttacentral.net and click on the Textual Information button, whenever you mouse over a Pali word, its meaning will pop up in English.

Sometimes you can connect the dots and glean insight into the meaning of a word while doing that. :)
quidquid Latine dictum sit altum videtur
(Anything in Latin sounds profound.)
User avatar
bharadwaja
Posts: 205
Joined: Wed Nov 13, 2013 1:20 pm
Location: London, UK

Re: Becoming a Pali Scholar

Post by bharadwaja »

Pali is not widely studied because it is a very peculiar language, if at all it can be called a language. It may have been a script (used to write the Theravada canon) rather than a language by itself.

Firstly outside of the Buddhist and more specifically Theravada world, it had virtually no existence in any literature. The only major non-Theravada literature that was known in it was Gunadhya's Brihat-katha, but even that is now lost. So Pali studies ipso facto means Theravada studies, specifically the canon and its commentaries.

Early Indian grammarians referred to Pali in unflattering terms - calling it Paisachi or bhuuta-bhaasha, both of which literally mean either "dead language" or "language of the dead" i.e. "ghoulish". The name "Pali" is a rather new name to this language. For more information on this "Indian" name for Pali, see http://menadoc.bibliothek.uni-halle.de/ ... view/56372

However, going back to your query, most serious Pali scholars, both today and formerly, start off with some study of Sanskrit which has had a significantly better literary presence in South Asia, and the knowledge of which greatly aids the understanding of the Pali forms of the words as well as its grammar (almost every Pali word form has its corresponding Sanskrit form).

All the best.
User avatar
Kare
Posts: 767
Joined: Sun Feb 01, 2009 10:58 am
Location: Norway
Contact:

Re: Becoming a Pali Scholar

Post by Kare »

arhat wrote:Pali is not widely studied because it is a very peculiar language, if at all it can be called a language. It may have been a script (used to write the Theravada canon) rather than a language by itself.
Of course Pali is a language, not a script.
Firstly outside of the Buddhist and more specifically Theravada world, it had virtually no existence in any literature. The only major non-Theravada literature that was known in it was Gunadhya's Brihat-katha, but even that is now lost. So Pali studies ipso facto means Theravada studies, specifically the canon and its commentaries.

Early Indian grammarians referred to Pali in unflattering terms - calling it Paisachi or bhuuta-bhaasha, both of which literally mean either "dead language" or "language of the dead" i.e. "ghoulish". The name "Pali" is a rather new name to this language. For more information on this "Indian" name for Pali, see http://menadoc.bibliothek.uni-halle.de/ ... view/56372
That Pali might be identical with Paisachi is one of several theories.
However, going back to your query, most serious Pali scholars, both today and formerly, start off with some study of Sanskrit which has had a significantly better literary presence in South Asia, and the knowledge of which greatly aids the understanding of the Pali forms of the words as well as its grammar (almost every Pali word form has its corresponding Sanskrit form).
Most serious French scholars start off with some study of Latin. True? No, this is just nonsense. The same goes for the study of Pali and Sanskrit. The reason why many Pali scholars start off with Sanskrit is simply that Sanskrit has higher prestige in the western academic world, and most universities demand that you start with Sanskrit. There are mainly two reasons for that prestige. One reason is that Sanskrit has a very high status as a divine language among brahmans in India, and western academics just inherited that arrogance. The other reason is that Sanskrit shows older forms than Pali, so it is more useful for Indo-european studies.

If you study Pali, Sanskrit is useful for comparison, just as Latin is useful for comparison if you study French. But you can still study French on its own terms and go deeply into French literature without studying Latin. The same goes for Pali and Sanskrit.

I studied Pali for several years before I took up the study of Sanskrit. I did not need the Sanskrit to study Pali, but I took it up later from sheer curiosity, wanting to be able to read the Sanskrit buddhist literature as well as the Pali texts. It is true that Sanskrit literature has much of interest, especially authors like Kalidasa. But my Pali did not benefit much from the Sanskrit studies. Neither was it harmed in any way.

So you may start with Pali or you may start with Sanskrit, and you will enjoy your studies either way. But don't believe those who are stuck in old-fashioned European academic habits and say you have to start with Sanskrit. Choose your own way from what interests you most.
Mettāya,
Kåre
User avatar
bharadwaja
Posts: 205
Joined: Wed Nov 13, 2013 1:20 pm
Location: London, UK

Re: Becoming a Pali Scholar

Post by bharadwaja »

​Hi Kare, thanks for responding.
Of course Pali is a language, not a script.
That is arguable. In the Pali canon all the Brahmanical names and terms without exception are quoted in their middle-Indic form rather than in their original old-Indic form. All the characters are depicted as speaking the middle-Indic form of the language, even proper names that could have been in old-Indic are spelt in their middle-Indic form. Nobody is shown to be speaking a different language from the others. Therefore (going by the Suttas) the inference is that either everyone spoke Old-Indic without any visible linguistic distinctions between old & middle Indic, or everyone spoke Pali. We know for a fact that once everyone in that area must have spoken old-Indic (from which Pali is claimed to have evolved), if this was still true in the Buddha's time, then the pali form should have been the written non-phonetic form of the language rather than what they actually spoke. It becomes more difficult to establish that everyone without exception could on the other hand have spoken Pali when old-Indic was also in existence as a prestigious linguistic register. Thus it appears that what we call Pali today could have originally been the written form of Old-Indic, i.e. a non-phonetic script rather than a language by itself. Even the Pali commentators (Buddhaghosa et al), nearly a millenium after the Suttas were first recited, when they refer to Pali, they mean the written texts rather than an oral language.
That Pali might be identical with Paisachi is one of several theories.
Regardless of theories, both their texts have been compared and the linguistic forms have been found identical.

There is even an early Tibetan tradition which records that the Sthavira sect had their canon in Paisachi. The sthaviras are the theras and the Paisachi canon is therefore the Pali canon.
Most serious French scholars start off with some study of Latin. True? No, this is just nonsense.
Sanskrit & Pali are not comparable to Latin & French, for the following reasons:

1. There is no evidence of Pali's natural linguistic evolution from Old-Indic. Significant developments in Pali like the double consonants appear to be later artificial innovations rather than a natural feature of the language.

2. Pali was not spoken or used outside Theravada Buddhism, and has no native speakers.

3. Sanskrit is strictly speaking the grammar of Panini, it is not the entire Old-Indic language. Old-Indic existed both within and outside the sanskrit grammatical paradigm (i.e. there are many Vedic and Epic texts in Old-Indic that do not fit the rules of Sanskrit grammar, they were composed in vernacular Old-Indic not conforming to Sanskrit grammar). They cannot strictly be called Sanskrit as they don't obey Sanskrit grammar.

4. K.R.Norman, another Pali scholar, mentions that the Pali of the Canon has gone through numerous revisions in the course of its written history, it was once written with single consonants in words where we now find double consonants, see http://www.shin-ibs.edu/documents/bForu ... Norman.pdf
There is evidence that Pali was not a phonetic language, specially when the canon was first composed. The way it was written and the way it was pronounced were different. We cannot therefore form conclusions about the spoken language of the 1st millenium BCE (or even earlier) from the texts we have in front of us today.

5. Most of the Theravada literary activity in the last 2000 years took place not in Magadha or Kosala (kingdoms in northern India where the Buddha lived most of his life), but in Sri Lanka and Burma (the native language of most of the Sri Lankan people 2000 years ago was a Dravidian language close to Tamil, that of the Burmese was a Tibeto-Burman language). Therefore Pali has had no known native speakers at least for the last 2000 years.

6. Oskar von Hinüber, a Pali scholar, mentions (Kleine Schriften, Volume 1, 1985) that there are features of Pali which defy any linguistic explanation.

7. What exactly do we mean by Pali? Is it the language of the Buddha, the language of the original canon, the language of the written canon as it exists today, or the language of post-canonical commentaries etc. Each of these languages is different from the other.

8. If as you say Pali/Sanskrit have the same relationship as French/Latin, one would expect to see more Pali texts and speakers than Old-Indic at some point of time in history - there is no evidence that Pali ever had a greater literature or spoken population than Old-Indic at any point of time.

9. Old-Indic has the largest literary corpus among all ancient languages as well as a very strong grammatical tradition, these are the reasons why it is popular in classical linguistics, apart from its relative linguistic age vis-a-vis Pali. Prestige and inherited arrogance, on the other hand are probably the bottom two reasons, not the top two as you've claimed, for Old-Indic being more recognized.
I studied Pali for several years before I took up the study of Sanskrit. I did not need the Sanskrit to study Pali
I am sure that is a valid approach, however most people, including the earliest Pali commentators and grammarians did go from Sanskrit to Pali rather than just study Pali, because Pali forms are not as much grammatically and etymologically analysable as Old-Indic forms are. There are words in the canon that have been traditionally misinterpreted and misunderstood since the most ancient times.. one of them is the meaning of Isipatana, another is the word sutta, these can only happen when there were no native speakers for the language. For Buddhaghosa's use of sanskrit grammar in interpreting the Pali canon, see http://www.ocbs.org/ojs/index.php/jocbs ... wnload/5/5 . Even Kacchayana's earliest grammar of Pali came a full 1 millenium after the Buddha's time, where he refers to no earlier Pali grammar and evidently modelled it on Panini's Sanskrit grammar. Why would he do so? Did the natives of Sri Lank not learn Pali grammar at all until Kacchayana? Or did they use Sanskrit grammar like Buddhaghosa did?
Last edited by bharadwaja on Tue Mar 25, 2014 11:08 pm, edited 1 time in total.
Coyote
Posts: 845
Joined: Mon Sep 19, 2011 12:42 pm
Location: Wales - UK

Re: Becoming a Pali Scholar

Post by Coyote »

waterchan wrote:
If you read suttas at http://www.suttacentral.net and click on the Textual Information button, whenever you mouse over a Pali word, its meaning will pop up in English.

Sometimes you can connect the dots and glean insight into the meaning of a word while doing that. :)
Thanks, that looks really useful. :anjali:
"If beings knew, as I know, the results of giving & sharing, they would not eat without having given, nor would the stain of miserliness overcome their minds. Even if it were their last bite, their last mouthful, they would not eat without having shared."
Iti 26
User avatar
Kare
Posts: 767
Joined: Sun Feb 01, 2009 10:58 am
Location: Norway
Contact:

Re: Becoming a Pali Scholar

Post by Kare »

arhat wrote:
Of course Pali is a language, not a script.
That is arguable. In the Pali canon all the Brahmanical names and terms without exception are quoted in their middle-Indic form rather than in their original old-Indic form. All the characters are depicted as speaking the middle-Indic form of the language, even proper names that could have been in old-Indic are spelt in their middle-Indic form. Nobody is shown to be speaking a different language from the others. Therefore (going by the Suttas) the inference is that either everyone spoke Old-Indic without any visible linguistic distinctions between old & middle Indic, or everyone spoke Pali. We know for a fact that once everyone in that area must have spoken old-Indic (from which Pali is claimed to have evolved), if this was still true in the Buddha's time, then the pali form should have been the written non-phonetic form of the language rather than what they actually spoke. It becomes more difficult to establish that everyone without exception could on the other hand have spoken Pali when old-Indic was also in existence as a prestigious linguistic register. Thus it appears that what we call Pali today could have originally been the written form of Old-Indic, i.e. a non-phonetic script rather than a language by itself. Even the Pali commentators (Buddhaghosa et al), nearly a millenium after the Suttas were first recited, when they refer to Pali, they mean the written texts rather than an oral language.
You bring up some interesting points. But the fact that the brahmanical names are quoted in their middle-indic form, speaks against your theory. A vernacular language naturally modify names to conform with its own phonetics. And some Pali texts in fact mention that different regions have different names for this or that, so you are wrong saying that the inference is that everyone spoke Pali. More important, however, is your disregard of a dialectal continuum. There were several dialects, as can be seen from the Asoka inscriptions and from the Gandhara manuscripts, although these were a little later than the Buddha's time. But from what we can see, the dialects in Northern India were probably mutually intelligible. So when one group of monks retell the words of the Buddha, it is just natural that they do so in their own dialect. This was not a translation, but a natural use of one of the dialects. The Pali texts were probably normalized and polished, some think this may have happened during the Polonnaruwa renaissance in Sri Lanka. But even this is also a normal feature of a language.
That Pali might be identical with Paisachi is one of several theories.
Regardless of theories, both their texts have been compared and the linguistic forms have been found identical.

There is even an early Tibetan tradition which records that the Sthavira sect had their canon in Paisachi. The sthaviras are the theras and the Paisachi canon is therefore the Pali canon.
Most serious French scholars start off with some study of Latin. True? No, this is just nonsense.
Sanskrit & Pali are not comparable to Latin & French, for the following reasons:
OK, I admit that the Latin/French analogy has its drawbacks.
1. There is no evidence of Pali's natural linguistic evolution from Old-Indic. Significant developments in Pali like the double consonants appear to be later artificial innovations rather than a natural feature of the language.
Pali falls naturally into the dialectal range of middle Indic. It is not derived directly from Sanskrit, since it has some features that only are found in Vedic. So it is derived from Old Indic the same way as Sanskrit and the other Prakrit dialects. It is, however, possible that Pali combined features from several dialects.
2. Pali was not spoken or used outside Theravada Buddhism, and has no native speakers.
We do not know how widely it was spoken. It is preserved only through Theravada, like Vedic is preserved only through the Vedas, Ardha-Magadhi through the Jains, etc.

4. K.R.Norman, another Pali scholar, mentions that the Pali of the Canon has gone through numerous revisions in the course of its written history, it was once written with single consonants in words where we now find double consonants, see http://www.shin-ibs.edu/documents/bForu ... Norman.pdf
There is evidence that Pali was not a phonetic language, specially when the canon was first composed. The way it was written and the way it was pronounced were different. We cannot therefore form conclusions about the spoken language of the 1st millenium BCE (or even earlier) from the texts we have in front of us today.
If we cannot form conclusions about the spoken language, how then can you form your conclusions about Pali not being a phonetic language?
5. Most of the Theravada literary activity in the last 2000 years took place not in Magadha or Kosala (kingdoms in northern India where the Buddha lived most of his life), but in Sri Lanka and Burma (the native language of most of the Sri Lankan people 2000 years ago was a Dravidian language close to Tamil, that of the Burmese was a Tibeto-Burman language). Therefore Pali has had no known native speakers at least for the last 2000 years.
There was some Dravidian in Sri Lanka, but the majority of the population probably spoke early Srilankan, a language from northern India. That is, however, irrelevant. Pali was imported from India together with the Pali texts. Therefore we should not expect native speakers of Pali in Sri Lanka.

7. What exactly do we mean by Pali? Is it the language of the Buddha, the language of the original canon, the language of the written canon as it exists today, or the language of post-canonical commentaries etc. Each of these languages is different from the other.
The fact is that we do not know exactly what dialect the Buddha spoke, and we do not know exactly the dialect of the original canon. Therefore you can not assert that it was NOT Pali. What we can say, however, is that it must have been a North Indian dialect belonging to the same dialect continuum as Pali does. So we can regard Pali as a good representative of that dialect continuum.

I do not know if you live in a country with a dialect continuum. I do. In Norway there are several dialects that are different, but mutually intelligible. We can even count Danish and Swedish as parts of that same continuum. If a person with a dialects different from mine tells me something of importance and I want to convey his ideas to my friends, I do not try to imitate his dialect. I will naturally convey his ideas in my own dialect. And nobody would react or find it strange. I think this is a good analogy of the language situation in Northern India at the time of the Buddha.

So what do we mean by Pali? The language of the Pali texts.
8. If as you say Pali/Sanskrit have the same relationship as French/Latin, one would expect to see more Pali texts and speakers than Old-Indic at some point of time in history - there is no evidence that Pali ever had a greater literature or spoken population than Old-Indic at any point of time.
No analogy is perfect.

If you wish to continue this dialogue, may I please suggest that you limit the length of your postings - just a little bit? :anjali:
Mettāya,
Kåre
User avatar
waterchan
Posts: 699
Joined: Fri May 07, 2010 7:17 pm
Location: Kamaloka

Re: Becoming a Pali Scholar

Post by waterchan »

This is a very informative discussion. Please carry on, you two. I hope others will join the discussion too. We don't get threads like this very often.

*sits down with a bag of popcorn*
quidquid Latine dictum sit altum videtur
(Anything in Latin sounds profound.)
User avatar
bharadwaja
Posts: 205
Joined: Wed Nov 13, 2013 1:20 pm
Location: London, UK

Re: Becoming a Pali Scholar

Post by bharadwaja »

Hi Kare, thanks again for your response. Apologies for the long posts but I cant seem to make them any shorter than they are, I have a lot more to say though, which I will reserve for later discussion. :smile:
But the fact that the brahmanical names are quoted in their middle-indic form, speaks against your theory. A vernacular language naturally modify names to conform with its own phonetics.
Sure, but:
1. Pali and Old-Indic did not have a radically different phonemic inventory to start with, most of the Old-Indic sounds are known to be present in Pali as well.

2. In the Suttas, we find no old-Indic vs. middle Indic distinction even in verbatim "iti" :quote: quotes, one would expect verbatim quotes to be verbatim, not transliterated or translated (specially if there was mutual intelligibility as you mentioned, there would be no need to translate to Pali).

3. Old-Indic Proper names of brahmanical rishis like Viswamitra/Vasishta are very unlikely to be transliterated into their middle-Indic forms (vessamitta/vasettha) every single time, unless there was no way of expressing them in Old-Indic at all.

I do therefore believe that indeed there was no way of writing Old-Indic at all except in their equivalent middle-Indic form (the earliest script used to preserve the canon must have been incapable of representing old-Indic conjuncts, as it was evidently a foreign script). So the conjuncts (double-consonants) must have been simplified in writing. A word like dharma would have been written as dhama, bhiksu as bhiku etc. We have evidence that this was indeed the case, see Norman's paper that I've linked to in my last post -- http://www.shin-ibs.edu/documents/bForu ... Norman.pdf

Later editors of the canon must have then restored double consonants and made it dhamma, bhikkhu etc (still not entirely approximating the original sounds), but approaching what we now consider Pali. Then as the script evolved, and full phonetic old-Indic writing became possible, we see that Old-Indic sounds are represented faithfully in the later Buddhist Hybrid Sanskrit stage.

The move from middle-indic back to old-Indic forms represents therefore the evolution of phonetic writing which has ever since been characteristic of sanskrit. The theravada by this time must have migrated to Sri Lanka with their written canon, and did not therefore participate in the evolution of the script towards Old-Indic that the Indian sects participated in.

It is quite likely that the Theravada canon was the earliest complete written literature of India, and it must have been written down in India at a very early date much before it even reached Lanka. Homeless early buddhist monks memorizing thousands of prose suttas all day sitting in the wilderness throughout the year and passing them down almost intact over hundreds of years in an evolving spoken language is unthinkable, memorizing huge prose texts was not suitable to the wandering minimalistic lifestyle of early bhikkus, the predominantly prose canon would not have survived almost intact if that were so, it would have been in several languages with hundreds of rescensions. Even Brahmanical literature by comparison was predominantly in verse and they had a known mechanism of preserving it (see http://en.wikipedia.org/wiki/Patha ), they did that all their lives, there is no comparable evidence for Bhikkus memorizing texts, or converting it all in their minds, without using writing, to Sanskrit later on (as the Sarvastivadins etc did).

Writing (using an imperfect foreign script) was therefore evidently used first by pre-sectarian Bhikkus to record the canon, probably because that was the only available way of preserving the canon for posterity. Therefore I again come back to my original assertion that Pali referred originally to the script and the written texts rather than the language.

The language of the canon that we call Pali today must have been one or more dialects of Old-Indic. Look at it another way - Panini standardized the grammar of Old-Indic sometime around the 5th-4th centuries BC, that means there were some or many vernacular Old-Indic dialects that must have existed for him to standardize the grammar for. This can only be the case if the pre-Sanskrit dialect(s) of the canon were in Old-Indic, but written in the "Pali" script.
If we cannot form conclusions about the spoken language, how then can you form your conclusions about Pali not being a phonetic language?
There are several ways.

First Norman in the article cited above makes it clear that the double consonants of Pali words were once written with single consonants. But that would destroy the meter in Pali poetry. So the single consonants (which represent conjunct consonants in Old-Indic) were doubled in a later stage of the language. Kalpa in Old-Indic was originally written as kapa, then this became kappa, as we now have it.

Second even after this restoration, some poetry still refuses to fit metrical considerations. This is because of sarabhatti vowels, see http://www.ancient-buddhist-texts.net/T ... ine-06.htm. Thus these vowels were not supposed to be pronounced in the spoken language, but they are written.

Third when we examine the dhatus of words, we find that the root of dhamma is dhr, not dhm, because this root also gives rise to words like dhareti. Therefore even though the word is written as dhamma, we know it has to be pronounced as dharma, i.e. going back to my assertion that these so-called middle-Indic innovations that we find in Pali are merely restricted to the written form of the language.

There are other ways as well but this post has already become longer than my previous one so I will stop.
There was some Dravidian in Sri Lanka, but the majority of the population probably spoke early Srilankan, a language from northern India. That is, however, irrelevant. Pali was imported from India together with the Pali texts. Therefore we should not expect native speakers of Pali in Sri Lanka.
No, there is no textual or archaeological evidence for any pre-pali early north-Indian languages having existed in Lanka. Besides Ashoka's edicts make it clear that his empire's southern boundaries were all Dravidian people. He even mentions the names of the dravidian kingdoms.
The fact is that we do not know exactly what dialect the Buddha spoke, and we do not know exactly the dialect of the original canon. Therefore you can not assert that it was NOT Pali.
We do know that the Buddha did not speak Pali because
1. Pali was not to be spoken as it is written, i.e. it's not phonetic, see above.
2. We also know some of the linguistic changes it has gone through before reaching its current state (again see above, Norman's paper)
3. Modern Pali scholars claim that the Buddha lived in Eastern India, while Pali is most definitely a western Indian dialect. While I accept the claim that Pali is indeed a western dialect, I doubt if the Buddha indeed lived in Eastern India for there are a lot of inconsistencies in that argument. But if he did live in Eastern India, he could not have spoken Pali.
User avatar
Kare
Posts: 767
Joined: Sun Feb 01, 2009 10:58 am
Location: Norway
Contact:

Re: Becoming a Pali Scholar

Post by Kare »

arhat wrote:Hi Kare, thanks again for your response. Apologies for the long posts but I cant seem to make them any shorter than they are, I have a lot more to say though, which I will reserve for later discussion. :smile:
Yes, it is tempting to write long dissertations on themes we really are interested in. But too long posts easily lead to even longer replies, and before you know it, the posts get so long that nobody bothers to read them.
But the fact that the brahmanical names are quoted in their middle-indic form, speaks against your theory. A vernacular language naturally modify names to conform with its own phonetics.
Sure, but:
1. Pali and Old-Indic did not have a radically different phonemic inventory to start with, most of the Old-Indic sounds are known to be present in Pali as well.
Yes, they were parts of a dialect continuum. Pali simplified the diphtongs ai and au into e and o, and the liquid consonants were turned into vowels. Hence Gautama > Gotama, etc.
2. In the Suttas, we find no old-Indic vs. middle Indic distinction even in verbatim "iti" :quote: quotes, one would expect verbatim quotes to be verbatim, not transliterated or translated (specially if there was mutual intelligibility as you mentioned, there would be no need to translate to Pali).
No, you are mistaken. When dialects are mutually intelligible, names and quotes are naturally given in the dialect of the speaker/writer, not in the dialect of the original. This is the normal procedure, and that is exactly what we see in Pali.
3. Old-Indic Proper names of brahmanical rishis like Viswamitra/Vasishta are very unlikely to be transliterated into their middle-Indic forms (vessamitta/vasettha) every single time, unless there was no way of expressing them in Old-Indic at all.
As I said above, what happened here is just what normally happens in such situations.
I do therefore believe that indeed there was no way of writing Old-Indic at all except in their equivalent middle-Indic form (the earliest script used to preserve the canon must have been incapable of representing old-Indic conjuncts, as it was evidently a foreign script). So the conjuncts (double-consonants) must have been simplified in writing. A word like dharma would have been written as dhama, bhiksu as bhiku etc. We have evidence that this was indeed the case, see Norman's paper that I've linked to in my last post -- http://www.shin-ibs.edu/documents/bForu ... Norman.pdf

Later editors of the canon must have then restored double consonants and made it dhamma, bhikkhu etc (still not entirely approximating the original sounds), but approaching what we now consider Pali. Then as the script evolved, and full phonetic old-Indic writing became possible, we see that Old-Indic sounds are represented faithfully in the later Buddhist Hybrid Sanskrit stage.

The move from middle-indic back to old-Indic forms represents therefore the evolution of phonetic writing which has ever since been characteristic of sanskrit. The theravada by this time must have migrated to Sri Lanka with their written canon, and did not therefore participate in the evolution of the script towards Old-Indic that the Indian sects participated in.
This is an interesting theory. It stands or falls with your assertion that the early scripts were incapable of representing old-Indic conjuncts. The Pali texts were written down in the 1st century BC. We have not preserved any original Pali manuscripts from that time. Therefore we have to look at other texts of the same age and see if your theory is right, see if the scripts of that time were incapable of representing old-Indic conjuncts. In the Gandhara area there are found several manuscripts and manuscript fragments that are dated from the 1st century BC till the second century AD. So they are roughly the same age. You can find more about these manuscripts in Richard Salomon: "Ancient Buddhist Scrolls from Gandhara".

So what do we find in those manuscripts? They are not in Pali, they are in the Gandhari dialect, which belonged to the same dialect continuum - which means that the language was close to Pali, but not identical. The Gandhari dialect preserved more old-Indic conjuncts, something that was typical of the western Prakrits. In Gandhari we find for instance "mahasamudro" (the great ocean) - Pali "mahasamuddo". The conjunct is preserved in the Gandhari, not in Pali. Another example: Gandhari: "trae kasavastue" (three topic for discussion) - Pali "tayo kathavatthu". Again the conjuncts are written.

And please what happens here: vastu > vatthu. If your theory were correct, the development would have been vastu > vatu > vattu. From a purely scriptual perspective it would have been meaningless to add the aspirate th. But in a phonetic development the breathing in s (st) is preserved when the conjunct is simplyfied, but moves over into the aspirate consonant th. This is a very revealing feature that proves phonetic development.

So with reference to preserved texts from the period when the Pali texts were written down, your theory just falls. Indian scripts of that age were able to write conjuncts. The simplification of conjuncts in Pali therefore are not a result of defect script, but are a natural phonetic development of central North Indian dialects. What we can observe, is that consonant clusters were better preserved in the west. In the east not only consonant clusters, but many intervocalic consonant are elided. Pali seems to avoid both extremes, so it may possible have a more central North Indian origin, or it may preserve features from several dialect, as a kind of North Indian lingua franca. Here some theories differ on this point, so I have no fixed opinion.

It is quite likely that the Theravada canon was the earliest complete written literature of India, and it must have been written down in India at a very early date much before it even reached Lanka.
That is a theory that needs substantial evidence to be accepted.
Third when we examine the dhatus of words, we find that the root of dhamma is dhr, not dhm, because this root also gives rise to words like dhareti. Therefore even though the word is written as dhamma, we know it has to be pronounced as dharma, i.e. going back to my assertion that these so-called middle-Indic innovations that we find in Pali are merely restricted to the written form of the language.
No, we do not know that it was pronounced as dharma. What your example proves, is that the early Pali grammarians were able to think, and that they modeled their work on Sanskrit grammars.
There was some Dravidian in Sri Lanka, but the majority of the population probably spoke early Srilankan, a language from northern India. That is, however, irrelevant. Pali was imported from India together with the Pali texts. Therefore we should not expect native speakers of Pali in Sri Lanka.
No, there is no textual or archaeological evidence for any pre-pali early north-Indian languages having existed in Lanka. Besides Ashoka's edicts make it clear that his empire's southern boundaries were all Dravidian people. He even mentions the names of the dravidian kingdoms.
It is believed that about the 6th century BCE, settlers from North-Eastern India reached the island of Sri Lanka. (http://en.wikipedia.org/wiki/Sinhalese_language)
We do know that the Buddha did not speak Pali because
1. Pali was not to be spoken as it is written, i.e. it's not phonetic, see above.
2. We also know some of the linguistic changes it has gone through before reaching its current state (again see above, Norman's paper)
3. Modern Pali scholars claim that the Buddha lived in Eastern India, while Pali is most definitely a western Indian dialect. While I accept the claim that Pali is indeed a western dialect, I doubt if the Buddha indeed lived in Eastern India for there are a lot of inconsistencies in that argument. But if he did live in Eastern India, he could not have spoken Pali.
1. Wrong. See above.
2. Correct. There are good reasons to regard the Pali of the Pali texts as different from the dialect of the Buddha. The interesting question is just HOW different. From what we know of middle Indic dialects, I would guess that the differences can not have been very great. Probably just a question of minor nuances.
3. The Buddha lived in central and eastern North India. He was born in Kapilavatthu, a little to the east of the center. So he should be expected to have spoken a central dialect containing none of the extreme conjuncts of the western dialects and none of the extreme vowel elisions of the eastern dialects. Pali fits this description quite well.

I admit that we probably never will know for sure, but my guess, based on available evidence, is that the Buddha spoke a dialect very close to Pali. Pali is probably based on this dialect, even though grammarians may have doctored and polished the language a little in later times.
Mettāya,
Kåre
Jeffrey
Posts: 153
Joined: Fri Sep 28, 2012 6:08 am

Re: Becoming a Pali Scholar

Post by Jeffrey »

It seems no one has yet chimed in on courses. Apologies if I am repeating. Here are two summer intensives. The first in the UK:


http://www.ocbs.org/events-ocbsmain-129 ... chool-2014



The second in the US:

http://www.pariyatti.org/P257;li/P257;l ... fault.aspx
User avatar
bharadwaja
Posts: 205
Joined: Wed Nov 13, 2013 1:20 pm
Location: London, UK

Re: Becoming a Pali Scholar

Post by bharadwaja »

Yes, they were parts of a dialect continuum. Pali simplified the diphtongs ai and au into e and o, and the liquid consonants were turned into vowels. Hence Gautama > Gotama, etc.
I think both these assertions may be wrong, for the following reasons:

1. Usually when a language has evolved, people don't continue to speak the mother dialects (even if the mother dialects have well recognized texts in them). We do not speak Shakespearean/Chaucerian English today irrespective of the earlier dialects having prestigious literature in it. Nor do we claim Shakespearean english somehow forms a dialect continuum with modern english.

Old-Indic could not have therefore formed a dialect continuum with middle-Indic, if middle-Indic represents a much evolved later stage of the spoken language (as to differ both phonetically and morphologically from the earlier dialects).

2. In old-Indic, Gautama is the vrddhi of Gotama i.e. lengthened grade of the ablaut vowel gradation. Gautama in old-Indic is a patronymic (i.e. gotra or surname) meaning a descendant of Gotama (which is a given name). In Pali, both Gautama and Gotama are written as Gotama. A person speaking Pali would not be able to distinguish between gotama and gautama if both were to be written as gotama.

Thus spoken phonetic Pali cannot have been mutually intelligible with Old-Indic, even if they were spoken side-by-side (which cannot have been the case either).
No, you are mistaken. When dialects are mutually intelligible, names and quotes are naturally given in the dialect of the speaker/writer, not in the dialect of the original. This is the normal procedure, and that is exactly what we see in Pali.
There is no evidence for the supposed mutual intelligibility of Old-Indic and middle-Indic that you keep referring to, indeed the evidence points the other way. Also there are no references to indicate the co-eval existence of spoken old-Indic and spoken Pali in the Buddha's time either in the Pali canon or in old-Indic literature.

There is evidence that Pali forms were not fully intelligible even to ancient theras whose native language we consider it to be. I mentioned about the routine and traditional misinterpretation of words like Isipatana (which Prof. Dr. Colette Caillat has in her 1969 paper in the Journal Asiatique conclusively shown to be *ṛsyavṛjana, i.e. a synonym of migadāya, which it usually appears together with). If Pali was the native language of the earliest Theravada monks, isipatana would not have been transliterated into BHS incorrectly as ṛsipatana.
This is an interesting theory. It stands or falls with your assertion that the early scripts were incapable of representing old-Indic conjuncts.
There were no written texts in Old-Indic until the script became fully/nearly phonetic towards the start of the common era (which is when we find Buddhist adoptions of sanskrit and hybrid sanskrit, as well as non-buddhist written texts in sanskrit). If Old-Indic could be written phonetically right from the beginning we would find large prose texts in Sanskrit like the Pali canon from the time of the Buddha. This is why the Pali canon was incapable (even occasionally) of expressing old-Indic forms of words and names phonetically.

This is also probably why for more than a millenium after the Buddha's time, there were no grammars for Pali nor are there any major texts in it except the canon. There are more books (Buddhist and non-Buddhist) in Old-Indic than in Pali, right until the present day. I believe Pali was therefore a temporary written form which died out in a few centuries once the imported script was phonetically perfected.
The Pali texts were written down in the 1st century BC.
That's just the traditional belief. There is no evidence that they were not written earlier, or even that they were written in the 1st century. Writing first originated in north-western India, and it spread to other parts (including Sri Lanka) with the spread of Buddhism. As I said, there is no conceivable way for the thousands of pages of prose to be transmitted nearly verbatim for hundreds of years by a monastic community in the pre-christian era without the use of writing.

There is a story of Ananda hearing a young monk recite a verse from the Dhammapada (verse 113) with the words *udaya-vyayam (arising and cessation) as udaka-bakam (water heron). Ananda corrects him saying he did not hear the Buddha say udaka-bakam (water heron) but rather *udaya-vyayam, but the monk's teacher (senior monk) advises the young monk to ignore Ananda's correction as he was an old-fool, and recite as he had learned it.

Now the closest to this udaka-baka spelling (which distorts the meaning of the original) is found in the so-called gandhari version of the dhammapada. From this, two things are clear - assuming Gandhari was a language (which is arguable), the so-called mutual-intelligibility is simply not evident, and secondly if the story has a grain of truth, the young monk was reciting from a written text in Pali using an early form of the gandhari script where udaya was written as udaka and vyaya (Pali baya) was written as baka, so writing was in use by Buddhists as early as Ananda's lifetime to preserve the Buddhavacana. Moreover, most of the suttas are claimed by tradition to be from the mouth of Ananda, so the inference is that when he recited it (presumably just once), it was written down regardless of the script being incapable of phonetic representation.
In the Gandhara area there are found several manuscripts and manuscript fragments that are dated from the 1st century BC till the second century AD. So they are roughly the same age. You can find more about these manuscripts in Richard Salomon: "Ancient Buddhist Scrolls from Gandhara".
Yes i have read several of the gandhari manuscripts. They represent a different chronological layer of the written language (Pali) i.e. with single consonants where Pali has double consonants (and old-Indic has conjunct consonants), lack of long vowels etc. Moreover, most of these early writing had usually something to do with Buddhism i.e. there is evidence that Buddhists were the pioneers in adopting writing in India. Non-Buddhist texts in the earliest period of writing are relatively rare. Ashoka in one of the Girnar inscriptions even mentions dhamma-lipi where lipi means script (i.e. he apparently associates the script/writing with Buddhism).
So what do we find in those manuscripts? They are not in Pali, they are in the Gandhari dialect, which belonged to the same dialect continuum - which means that the language was close to Pali, but not identical. The Gandhari dialect preserved more old-Indic conjuncts, something that was typical of the western Prakrits.


There are no historical references to affirm the existence any language that we now call Gandhari. The name Gandhari was invented in 1946 for the language of the texts written in Kharosthi script and found in the extreme north-western part the subcontinent. Like Pali (which was one unknown script), therefore Gandhari is just another name for the Kharosthi script as far as I can see. Everything written in Kharosthi is now called the Gandhari language.
In Gandhari we find for instance "mahasamudro" (the great ocean) - Pali "mahasamuddo". The conjunct is preserved in the Gandhari, not in Pali. Another example: Gandhari: "trae kasavastue" (three topic for discussion) - Pali "tayo kathavatthu". Again the conjuncts are written.
That only proves that the Pali forms that dont preserve the conjuncts were reduced to writing earlier than these so-called "Gandhari" forms where the conjuncts appear properly. Else are we claiming that Gandhari was an old-Indic dialect rather than a middle-Indic dialect? By the way, there are no early Pali canonical texts found anywhere in the Gangetic plains which people consider to be the home of the Buddha. Indeed there are more archaelogical finds of pre-christian era Buddhist artifacts in the north-western region than in any other region of India.
And please what happens here: vastu > vatthu. If your theory were correct, the development would have been vastu > vatu > vattu. From a purely scriptual perspective it would have been meaningless to add the aspirate th. But in a phonetic development the breathing in s (st) is preserved when the conjunct is simplyfied, but moves over into the aspirate consonant th. This is a very revealing feature that proves phonetic development.
The Pali canon has been edited over and over again and the form of Pali we now see in the canon is a heavily phonetically standardized close approximation of the old-Indic spoken forms. It was not a mechanical replacement of syllables, I don't claim it was. Even the Buddhist Hybrid Sanskrit (BHS) forms were not mechanical transliterations from the Pali forms back to old-Indic forms, there was a lot of thought that went into the BHS transliterations from Pali.

When I say there is no evidence of phonetic development, I mean independent of the canon, the spoken language of the pre-Buddhist era was not showing a phonetic evolution towards Pali; on the other hand, Vedic of the Brahmanical mantras (hymns) shows a definite evolution to the language of the Brahmanas (i.e. http://en.wikipedia.org/wiki/Brahmana) , and further into the language of the Upanishads, and then on to the vernacular old-Indic dialects of the mid-1st millenium BCE of which sanskrit was the grammatized standard. It was not evolving in the direction of middle-Indic at all.
So with reference to preserved texts from the period when the Pali texts were written down, your theory just falls. Indian scripts of that age were able to write conjuncts.
No you are apparently comparing texts of different periods, without establishing that the Pali forms and the so-called Gandhari forms belong to the same decade or even century. I've mentioned that the scripts in which the language was written evolved more and more towards phonetic representation. Early writing was simply not phonetic. I mentioned sarabhatti and provided a link, don't know if you've read it. I also quoted from Hinüber's paper about some Pali forms defying linguistic explanation.
No, we do not know that it was pronounced as dharma. What your example proves, is that the early Pali grammarians were able to think, and that they modeled their work on Sanskrit grammars.
There were no "Pali" grammarians until Kacchayana (6th century CE) i.e for a full 1000 years since the canon came into being.

What I'm saying is if dhr- was the root of both dhareti and dhamma (which is the case), dhamma should have to be *dharma, even if it is written non-phonetically as dhamma, if not it defies linguistic explanation.
It is believed that about the 6th century BCE, settlers from North-Eastern India reached the island of Sri Lanka. (http://en.wikipedia.org/wiki/Sinhalese_language)
Again this is personal belief, I don't know of any independent evidence for the existence of pre-Buddhist north Indian speakers in Sri Lanka.
2. Correct. There are good reasons to regard the Pali of the Pali texts as different from the dialect of the Buddha. The interesting question is just HOW different. From what we know of middle Indic dialects, I would guess that the differences can not have been very great. Probably just a question of minor nuances.
The differences were as great as the udaka-bakam story above illustrates.
3. The Buddha lived in central and eastern North India. He was born in Kapilavatthu, a little to the east of the center. So he should be expected to have spoken a central dialect containing none of the extreme conjuncts of the western dialects and none of the extreme vowel elisions of the eastern dialects. Pali fits this description quite well.
The linguists don't agree. Pali is more similar to the language of Ashoka's Girnar rock edicts (extreme west of India), and interestingly Girnar is not far from where the Paisachi texts that have been compared and found to be identical with Pali have been found. None of the central (Hindi) or eastern Indian languages (Bengali, Oriya, Bhojpuri etc) of today display a clear linguistic evolution from Pali.
I admit that we probably never will know for sure, but my guess, based on available evidence, is that the Buddha spoke a dialect very close to Pali. Pali is probably based on this dialect, even though grammarians may have doctored and polished the language a little in later times.
Yes, the canon is based on what the Buddha actually spoke (or rather when the suttas were first recited after his parinibbana), but the written form of Pali we have today is significantly different from that original recitation. We have a really good idea about the linguistic changes that the canon has gone through, see below.

Please read Norman's paper, from the point (pages 83-86) where he starts with the deficiencies of the original "Pali" script:

"There was another type of error, however, which was due to deficiencies in the early Indian writing system. In the earliest form of the Brāhmī script, double consonants were not written, and the marks for long vowels were frequently omitted."

Please go through pages 83-86 in this paper before replying.
User avatar
Kare
Posts: 767
Joined: Sun Feb 01, 2009 10:58 am
Location: Norway
Contact:

Re: Becoming a Pali Scholar

Post by Kare »

I am reading your postings with interest, although I have to admit that I am rather puzzled by the discrepancy between your detailed knowledge of features of Pali and your obvious ignorance of elementary features of linguistics.

But now it seems we are reaching the stage where we are to starting repeating ourselves and losing ourselves in irrelevant details. So let's be brief and get down to basics. The center point of your theories seem to be this:
The Pali texts were written down in the 1st century BC.
That's just the traditional belief. There is no evidence that they were not written earlier, or even that they were written in the 1st century.
You need to substantiate this assertion. If you can prove that the Pali texts were written earlier, then let us continue this dialogue. If not, I see no point in continuing. Then I'll just thank you for an interesting discussion.

:anjali:
Mettāya,
Kåre
Post Reply