First written Buddhist Canon?

So I have a number of scholastically naive questions. My questions focus on the written suttas/sutras. When did they first appear and in which script and language? Also, if the first written sutras were not in Pali, then the same questions for the Pali suttas; when were they first written down, by whom or which monastery, and was it in the existing devanagari script we have today?
Any there any good articles/pdfs on the history and transmission of the written suttas/sutras?
When were the suttas first written down and in what language and what script?
Were the pali suttas translated from other written suttas or written down from recitations?
Is there any one person or place or institution that claim to be the first to have writen down the words of the Buddha or the sutras or the canon?

Thanks in advance for any answers.

2 Likes

As far as I understand:

The monks who left to India and to the hilly areas of Sri Lanka, during the difficult period, returned to Anuradhapura, and decided to transcribe the Tripiṭaka (philosophical doctrines of Buddhism) for the preservation and for the use of future generations. The monks selected Aluvihare Rock temple in Matale as the most suitable and secured place to carry out this important event. This transcription was carried out due to the fear that the doctrine would be lost during the upheaval caused by repeated South Indian invasions.[5] It is said that 500 scholarly monks congregated at Aluvihare Rock temple to perform the difficult task of first reciting the doctrines and agreeing on an acceptable version before transcription. The entire transcription was done in books made of ola leaves, locally known as puskola poth. These books were made up from thick strips created from the leaves of either the palmyra or talipot palm and the doctrines were written down in Pali language. A metal stylus was used to inscribe the characters on the ola leaves.

The old library atof Aluvihare Rock Temple, which had safely housed the volumes of this transcribed manuscripts for so many centuries, was totally destroyed during the Matale Rebellion in 1848.[6] Many parts of the temple complex was destroyed too by this incident. The consequences of this disaster are still evident today at the temple premises. It took a long a time for the recompilation, as few generations of monks had to transcribe the Tripiṭaka again. The recompilation and transcription took a long time as only few monks were engaged in this painstaking task, and the first of the three “baskets of the law” was only completed in 1982.

4 Likes

Matale is only 30km from my hometown.
My birth certificate was written in Ola Leaves by the astrologer in my village.
Unfortunately, I lost it.
This technique is still alive in Sri Lanka.

5 Likes

I think there are two smaller, but possibly more precise, questions hidden within the OP, namely:

What is the temporal and geographical difference between the first postulated written Buddhist Canon (which would now be dust, presumably) and 1) the earliest sutta fragments that have survived, and 2) the earliest “complete” tripitakas we have.

I don’t know the answers offhand, but I think that might help focus the OP, since it directly intersects with the above.


1st EDIT: I wasn’t sure this was all correct, so I wanted to make sure. I am now reasonably thinking this is correct, if not so, I am sure someone will correct me.

So there is a famous archaeological find (not just famous for Buddhist archaeology) that was underreported in popular media even when it was first discovered. It was the Bāmiyān find, in Afghanistan, and it contained Buddhist scriptures in Gāndhārī from between the 1st and 2nd centuries AD.

Interestingly enough, as I recall, it is also the site of the location of the oldest Mahāyānasūtra, the Aṣṭasāhasrikāprajñāpāramitāsūtra (found in the same find?), but I might be confusing this find with another one in Afghanistan.

Since this interests me and I have some spare time, I’m doing some research on the oldest “complete” tripitakas, Mahāyāna tripitakas included since they are likely to contain āgamāḥ, this is not to say that I am indirectly arguing that “Mahāyāna tripitakas are older” by including them in search, but rather, I am thinking it is very possible the oldest that have survived in tact might have did so in East Asia, and would have also contained Mahāyānasūtrāṇi.


2nd EDIT (Oldest Pāli Canon/Tripitaka):The Pāli Canon, as a “complete” entity, in “complete” manuscript, is something of a hard thing to find, historically, it seems, for my cursory searches.

I am looking for actual better sources, but this is on wikipedia:[quote]The climate of Theravāda countries is not conducive to the survival of manuscripts. Apart from brief quotations in inscriptions and a two-page fragment from the eighth or ninth century found in Nepal, the oldest manuscripts known are from late in the fifteenth century,[53] and there is not very much from before the eighteenth.[54][/quote]And the footnotes link to von Hinüber’s 2000 publication: A Handbook of Pāli Literature, pages 4-5, and [quote] “Pali Text Society Home Page”. Palitext.com. Retrieved 2012-10-14.[/quote], which is an out-of-date reference.

I, personally, would be (and am) suspicious of this small wikipedia quote, because wikipedia, IMO and from observation, tends toward sectarian polemics fuelled by Buddhist infighting, and as such, often “interprets” facts or chooses “certain phrasings” to suit agendas (such as making either Mahāyāna or Theravāda “older” or “newer” to collect occasionally pseudo-historical “argument points”, rather than participating in legitimate historical inquiry).

5 Likes

After doing some research, I tentatively think that the 팔만 대장경 (Palman Daejanggyeong, or the “Goryeo (Koreana) Tripiṭaka”) is the oldest extant complete tripiṭaka. It is a Mahāyāna tripiṭaka, and as such, contains a lot of Mahāyāna material, which makes sense, given that it is a historical preserved slice of 13th century Korea.

I can’t find a complete exhaustive list of all of the EBTs contained within it, but I know that at the very least, the tripiṭaka contains the Sarvāstivāda Saṃyuktāgama which is called SA on SuttaCentral (I believe that it is the same translation) as well as the Smaller Saṃyuktāgama (BZA, called SA-2 on SuttaCentral) (source: Bingenheimer, Studies in Āgama Literature, available on the internet, search for “koreana” in that document). It is likely that more āgamāḥ than this are preserved in the Palman Daejanggyeong.

4 Likes

How with the Asoka rock written. Asoka write 4 Nikayas only at several rocks :thinking:

Thanks everyone for your contributions! My 2 cents:

According to the Pali commentary to MN 140, certain texts were written down during the life of the Buddha. There is no evidence to support this, but it is certainly not impossible. The texts written down (likkhati) include an extensive description of practice (cūḷasīlamajjhimasīlamahāsīlādīni ekadesena likhitvā chadvārasaṃvaraṃ satisampajaññaṃ catupaccayasantosaṃ navavidhaṃ senāsanaṃ, nīvaraṇappahānaṃ parikammaṃ jhānābhiññā aṭṭhatiṃsa kammaṭṭhānāni yāva āsavakkhayā ekadesena likhi, soḷasavidhaṃ ānāpānassatikammaṭṭhānaṃ vitthāreneva likhitvā). They were said to be inscribed on gold plates (suvaṇṇapaṭṭa), so barring the accidents of history, they could even survive.

Most likely this is just a story, which would make the first written texts the Ashokan pillars. These, however, contain the edicts of Ashoka, not Buddhist texts. However, they do refer by name to a number of texts, and in addition refer to a number of terms and phrases found in the EBTs. These texts are written in Brahmi script, which is the ultimate ancestor of most of the scripts used today in India, as well as south and south-east Asia.

The language of the edicts is Magadhi:

This is, more or less, the same language as used in the Jain texts. Certain details of Magadhi are apparent in Pali, which is usually taken as a sign that Magadhi, or something very like it, was the earliest language of the Buddhist texts.

However, a little caution is due here. The differences between these dialects are very small, and in almost all cases do not create any difficulties for understanding. In many ways they are more like regional accents than distinct languages. From the earliest times, we know that there were mendicants from different regions who were memorizing the texts, and undoubtedly there would have been some variation in pronunciation. More controversially, I think it is likely that the Buddha himself used different dialects. In any case, there is no need to assume that there ever was a single original language/dialect.

As detailed by Sarath, the first recorded written full canon was the Pali canon at Matale. This would presumably have been written in an early version of the Sinhala script, which was derived from Brahmi have been written in Brahmi. The Sinhalese script was not in use until much later.

Presumably there were texts being written down in this period in India, but we don’t have any concrete historical references. In the Mahayana sutras, however, which were composed from this time, there is frequent reference to writing, so it must have become standard practice for Buddhist scriptures.

The earliest extant manuscripts, as mentioned by Coemgenu, are the Gandhari manuscripts from Afghanistan. These cover a wide range of the different kinds of Buddhist texts that existed at the time, including EBTs. They are written in the Gandhari language:

And use the Kharosthi script:

There is a small but dedicated team of scholars working on this field, whose website is here:

https://gandhari.org/

In addition to these texts, there is a large number of other texts preserved from ancient times in manuscript form. Some of these have been published, and we have collected as many on SC as we can. Other manuscripts still await editing and publication: it is a slow, difficult job.

As mentioned by Coemgenu, the Tripitaka Koreana is the oldest complete canon. This is the basis for most modern editions of the Chinese canon, including the Taisho canon, which was digitized by CBETA and forms the basis for the Chinese texts on SC. Despite being created in Korea, the texts are Chinese. It contains a huge number of texts, including the ancient Agama and Vinaya texts.

And finally, to confirm that apart from occasional suttas, the oldest physical manuscript of the complete Pali canon would be quite recent. I am not aware that any “oldest” manuscript has been identified, but it would not be more than a few centuries old at most.

Unfortunately, yes. And in my experience, the winner of the the argument is the one best versed in wikipedia “law”, not in Dhamma. I don’t think it has improved much in the past, what, nearly a decade now. Still, it is better as a basic resource than pretty much any other freely available site.

12 Likes

I posted today some Wat Umong photos on another post, but here’s an example of Brahmi text:

3 Likes

Thanks Michael, that’s interesting!

Just in case anyone’s unclear, these signs in a Thai Wat are not “original” objects. They are 20th century artifacts, produced as one example of Ajahan Buddhadasa’s efforts to get Thais to understand and appreciate their heritage from India. He had a number of Ashokan objects copied, including pillars.

4 Likes

Bhante what is stone inscription (Sellipi) in Sri Lanka?

1 Like

Its interesting how much Brahmi looks like runes! Obviously there is no connection, but it is interesting nonetheless.

1 Like

It’s Brahmi. I can’t identify all the characters with confidence, but most of them are clear enough. If the text below doesn’t show, download and install Noto Brahmi from here:

https://www.google.com/get/noto/#sans-brah

𑀕𑀢𑀘𑀢𑀼 𑀤𑀰
gatacatudasha

If this is somewhere in the region of correct, and if it is Pali or something like it, then 𑀘𑀢𑀼 𑀤𑀰 probably means “fourteen”, and the previous characters look like a suffix from a trunctated word.

Did I mention that I don’t know what I’m doing? But I have fun doing it!

The connection is probably that they were both carved in stone, which would strongly favor simple shapes.

6 Likes

Regarding Brahmi script, the Majjhima Nikaya is available in Brahmi/Pali. An Amazon search will come up with it. I got a copy to see if I could get familiar with it.

Thanks, I didn’t know this. If anyone’s interested, here’s the page:

https://www.amazon.com/Brahmi-Majjhima-Nikaya-Middle-Sayings/dp/1478360755

And this is what it looks like:

And just FYI, http://sanskritdictionary.com/ allows you to see terms in Brahmi. We’d like to have a Brahmi script converter for SC, too, but have never got around to writing it.

2 Likes

Love the idea! Is it something easy to do? I mean, can someone as not techsavvy as me be of any help towards that goal? Or is it something we could pay anyone to do at a reasonable cost? :sweat_smile:

2 Likes

Sure. We just need to find a developer familiar with ancient Indic scripts. :wink:

2 Likes

A good thing is that standardized brahmi script is reasonably simple. Nothing compared to the super complex Sinhala!

#Vowels and vowel diacritics
http://www.omniglot.com/images/writing/brahmi_vwl.gif

#Consonants
image

Source: http://www.omniglot.com/writing/brahmi.htm

Now I have no idea of where to look for this someone.
Do you have any idea of where we should start the search?
:anjal:

1 Like

Are letters always spelled as in the tables, or does the spelling differ based on context? If it differs, are there definite rules for that, or is it as random as English pronunciation? :wink:

If it would boil down to simple substitution I could write some simple python script for that.

Hi Tuvok, thanks for that! :slight_smile:

As this is an abugida scrypt, the diacritics are added to the consonants following the example or how those would combine with the first consonsant ‘k’, as seen in the vowels depicting figures .

You could get a sense of the combination rules by checking the guts of the font available at this link:
https://sites.google.com/site/brahmiscript/

In this link you have an example of how the taking of five precepts would be spelled in brahmi:

http://www.virtualvinodh.com/wp/five-precepts-brahmi/

1 Like

I was joking, for the record! But if you’re serious, as you say the limited nature of the script would probably make it not too hard. Probably any programmer could do it with a bit of effort. Maybe an Indian or Sinhalese might have a little edge in being familiar with the idea of an abugida. They’d have to start by looking at the code that we use transform pali, and figuring out how to extend it to Brahmi. If there is someone who could do it, we could take the chance to extend to other scripts, too, maybe Cyrillic for example.

How to find someone? Ask here and at other Buddhist forums? Facebook? You could try the online freelancing sites like Upwork. We’d be happy to offer advice and tech support, but apart from that we don’t have extra resources at the moment.

3 Likes