Is one’s own interpretation in fact the common vernacular? restoring janapadanirutti with reference to Chinese parallels

Continuing on from previous discussions here:

I have gone back to check Brian Levman’s paper Sakāya Niruttiyā Revisited, which is the spark for the modern re-evaluation of these passages, and I saw that he has in fact discussed the Chinese Vinayas, including three texts we have not noticed. It’s obviously been too long since I read it! It’s linked in this previous discussion.

I’ll upload the relevant portions here.

What is to be drawn from this? Well, we can say with some confidence that the passage is pre-sectarian, as both the Sarvastivada and Vibhajjavada branches have it. It is missing, apparently, from the Mahasanghika, but this is not unexpected, since their Khandhaka was massively rewritten.

I’ve checked Frauwallner on this, and he does not mention this passage. However, in his discussion of the relevant chapter (p. 127), he says, “This chapter was more exposed than any other to amplifications and additions. And accordingly these are found in large quantities.” This agrees with my argument in the original post that this passage was added around the time of the Second Council.

Back to the Vinaya passages, clearly they differ quite a bit and it is difficult to say exactly what common core, if any, they may have shared. Levman himself does not attempt this, but rather continues discussing the Pali on its own terms.

But let me sum up the final item, that which is allowed by the Buddha. I’ll use Levman’s renderings for consistency.

  • Mahiśāsaka: “reading and recitation in the sound of the country”
  • Dharmaguptaka: “the sounds and common language of the country”
  • Vinayamātṛka: “act according to the country”
    • This is T 1463, usually said to be either Sarvāstivāda or Haimavata (another Vibhajjavada school)
  • Sarvāstivāda: apparently no allowance, just a prohibition against using the “chant of the heretics”.
  • Mūlasarvāstivāda: singing style only allowed if it is the “regional character of the voice”.
    • This is the Chinese text; Levman does not mention the Tibetan or Sanskrit.

Obviously the consensus of what is allowed in Chinese Vinayas must have been janapadanirutti. The Sarvāstivāda appears to be an exception, but given that it is mentioned in the Mūlasarvāstivāda, I assume there may have been some loss or corruption in the Sarvāstivāda text. Leaving aside discussion of exactly what janapadanirutti means for now, let us consider what the implication is for the Pali.

Now, generally speaking I think the Pali Vinaya is the earliest and most reliable of the Vinayas. When comparing different Vinaya passages, it is usually the case that the text of other Vinayas is more elaborated and developed. But such general impressions are of little use when considering individual passages, as there are early and late features in every Vinaya.

In this case, there are indeed some elaborated features in the Chinese passages. The (Mūla-)Sarvāstivada includes more backstory about the two brahmins, as does the Mahiśāsaka and to a lesser degree the Dharmaguptaka. Mahiśāsaka also mentions “reading”, although this might be a translation artefact.

These are fairly minor matters, however, and I don’t think they really impact the question of the final allowance.

What we have, then, is a consensus that the allowance is for janapadanirutti, the exception to this being the Pali.

In terms of textual transmission, it is readily apparent how a term used previously in the passage can come to replace a later term by means of a simple slip of memory or, if it was later, in copying the manuscript. There are other examples of this kind of thing in the Pali. On the other hand, it would be more effort to introduce a different concept into this short passage and ensure it was propagated in all Vinayas.

Now, let us add to this the evident problem with interpreting the Pali, for which this and other threads serve ample witness. Norman, Levman, Ven Brahmali, and others have discussed the dual use of sakāya nirutti in both the original problem and the final solution, and concluded that they have different senses. I have argued another interpretation, that they have the same meaning.

It would seem we are all just farting in the wind. The original allowance was most likely janapadanirutti, not sakāya nirutti, and the Pali is wrong.

Apart from this detail, the Pali appears to be the simplest and cleanest, so let us assume that in other respects it is our best witness, bearing in mind that a closer reading of the Chinese might disprove this. Here, then, would be the relevant portions of the text reconstructed:

sakāya niruttiyā buddhavacanaṁ dūsenti. buddhavacanaṁ chandaso āropema
janapadaniruttiyā buddhavacanaṁ pariyāpuṇituṁ

“They ruin the Buddha’s word with their own interpretations of terms. Let us elevate the Buddha’s word with pleasing poetry.”
“I allow you to memorize the Buddha’s word in the janapadanirutti.”

This gives us a much more straightforward meaning. The Buddha (bearing in mind that this is in fact the voice of someone(s) around the time of the Second Council) rejects both personal interpretations and elevating with chandas. Instead, he reiterates the advice already given in the Araṇavibhaṅgasutta, to use janapadanirutti. This is another point in favor of this correction: it means that Sutta and Vinaya are saying the same thing.

What, then, is the Buddha allowing? In its use of the word chandas, the Pali is suggesting the use of Vedic metre. That the passage concerns the introduction of non-Buddhist features is reinforced by the fact that the next passage in the Pali concerns monks learning cosmology. The Chinese texts also reinforce this impression. In different ways they are rejecting the elevated, elite speech of the Vedas (whether Vedic language is in fact elevated is not the issue; it was perceived as such.) Levman points out that in several cases the issue is the chanting style, and the passage is connected to the well-known rejection of “long drawn-out singing”, which here appears to imply a Vedic recitation style.

The Chinese translators agree with modern scholars that janapada here means “regional, local”.

But I have already pointed out the problem with this: we have two Indic lists of words that exemplify such janapadanirutti, one being a list of terms for “bowl”, another a list of terms for “person”. But not a single one of those terms has been shown to be a localized or regional term. They are all regular Pali/Sanskrit, either attested in various texts or in a couple of obscure cases, traced by etymology. Words such as patta or puggala are very widely found in all strata of relevant literature, and clearly cannot be considered “local”. Other terms might be found more rarely, but even though our texts mention hundreds of different localities, it is not possible, so far as I am aware, to associate any of them with a particular location.

The logic of the “local” interpretation requires that all the terms be local. The lists are given as examples of terms whose characteristic is that they are janapadanirutti. Given this, the presence of even a single term that is not regional would refute this interpretation. If I say, “Here is a list of fruit: meat, oranges, apples, pears”, we would be immediately thinking there is some problem. But it is not just a single term, rather, we have two lists, neither of which have any local terms.

As far as I can tell, modern scholars have simply assumed that these must be regional and have not attempted to adduce evidence for this fact. Levman argues that several of the terms for bowl might have non-Indo-Aryan roots, but his suggestions are not convincing. In any case, regardless of whether some items with unknown etymologies might ultimately be traceable to non-Indo-Aryan roots, it remains the case that all of the terms for bowl are either attested in Pali/Sanskrit, or can be explained with straightforward Indo-Aryan etymologies.

I haven’t been able to review all the previous research on this, but so far as I am aware, previous scholars had not realized this and it is an original contribution by myself. I’m not blowing my own horn here, I just want you to know that we build on previous work, and sometimes we know things that people previously did not. I’m not disagreeing with the consensus to be stubborn, but because the evidence leads that way. As far as I’m concerned, unless there is actual evidence that any of these terms are actually localized, this reading must be rejected.

So what does the passage mean? Janapada can, as I pointed out in my previous essay, have the sense “the people”. This is less common and somewhat linguistically ambiguous, but the sense is well attested in both Pali and Sanskrit. It is often used in this way when “the people” are contrasted with the elite, thus having the sense, “the commonfolk”.

This fits the overall argument well. The Buddha is rejecting both “individual interpretations of terminology” and “elevated with Vedic verse” and reiterating the position of the Araṇavibhaṅgasutta that we should stick to the “terms used among the people”, i.e. the “common vernacular”.

Returning to the Aranavibhaṅga for a moment, the passage there is:

janapadaniruttiyā ca abhiniveso hoti samaññāya ca atisāro

Now, this is presented as a single item, and the explanation also treats it as a single item. This implies that, according to the extremely common Pali idiom, the two phrases are synonymous. That is to say, janapadanirutti is samaññā, which according to DPD is “common language; normal usage; general meaning”. The explanation makes it clear that one is to use “common language” without “insisting” (abhinivesa) or “overstepping” (atisāra). The language itself is not the problem, it is our attitudes to it.

Thus the context supports my proposal that janapadanirutti has the sense “common vernacular”.

To return to the Vinaya passage once more, my argument leaves us in a tricky situation. If it is accepted, then the Pali is erroneous in a non-trivial way that distorts the meaning. What is a translator who accepts this situation to do? They might opt to render a reconstructed version. But if they choose to render the Pali as-is, then the two occurrences of sakāya niruttiyā should be rendered similarly, even if it is clumsy, to show that it is an artificial repetition.

It’s been a long road on just a few words, and congratulations on anyone with the patience to make it thus far! But for myself, I do feel some headway has been made.


Moving on from the original ending of this essay, there is one more point that I think the Chinese parallels alert us to. So far I have focused on the conclusion of these passages, where there is a clear unanimity on the word janapadanirutti. The earlier portion of the passages, however, is less clear and it is not obvious how it relates to the Pali. Namely: what exactly were the two brahmins complaining about?

  • Mahīśāsaka: They heard recitations “lacking integrity” (不正) and confusing the pronunciation.
    • the Chinese character 正 frequently stands for Indic words from the root sat.
  • Dharmaguptaka: He said monks are ruining the Buddhavacana and should be fixed using “the fine speech of the world” (世間好言論). Levman says this means Sanskrit.
    • The word for “fine” here is 好 (Skt. sādhu, ruci, abhirata; vyañjana; adhimukta, anurakta, anuvyañjana, arthika, ārāma, iṣṭa, udāra, kāma, kṣama, guruka, caukṣa, bhalla, mana-āpa, mano-jña, vara, vedanīyatā, śobhā, suvarṇa, hita, hitatva; Tib. dga’ ba) [Charles Muller; source(s): Hirakawa, YBh-Ind]
  • Vinayamātṛka: They ask for Buddhavacana to be put in chandas, the Buddha says in the sasana there is no concern for “fine language”.
    • Fine here is 美 ((Skt. varṇa, abhirūpa, praṇīta, madhura, accha, abhirūpayā prasādikayā darśanīyayā, iṣṭa, audārika, kalyāṇa, *kānta, gamya, guṇa, cāru, darśanīya, peśala, prasādika, bandhura, bhavya, mano-jña, mādhurya, mṛṣṭa, lāḍita, sat, su-, sudhā, *sundara, suvarṇa, svādu; Tib. bsngags pa, dang ba, mnyen pa, snyan pa). 〔瑜伽論 T 1579.30.282c29〕 [Charles Muller; source(s): Nakamura, Hirakawa]
  • Sarvāstivāda: brahmins recite “four Vedas”
  • Mūlasarvāstivāda: apparently the same.

So it seems that none of the five Chinese parallels agree with the sense of the Pali sakāya in the sense “one’s own”. In three of the cases, we have a word for “fine”, which is plausibly attributed to a root sat. In the remaining two cases the “four Vedas” are specified.

The obvious step here would be to assume there has been a confusion between sat and saka. Let’s see where this leads us.

Turning to the Pali commentary, it has a rather curious feature. Normally the commentary, as one would expect, comments on words in sequence. Here, however, it first glosses chandaso āropema and then sakāya niruttiyā, despite the fact that this reverses the sequence of the actual text. Given all we have seen, it is no great leap to assume there has been corruption in this text, and this may be one of the evidences.

Further, the gloss on chandaso says it means “esteemed speech like the Veda” (vedaṃ viya sakkatabhāsāya). Note the use of sakkata here, which resolves to sat + kata “made good”, “esteemed”), using the same root that we have hypothesized underlies three Chinese passages, next to the mention of the Veda, which in explicitly mentioned in the remaining two Chinese passages. The Sanskrit form of this is satkṛta, which is in fact very close to the word Sanskrit itself (saṃskṛta). Perhaps the Pali is, in fact, a variant of this word. Regardless, it clearly means something similar, “esteemed speech” as used in the Veda.

Here is the Pali text of the two brahmins’ request:

Te sakāya niruttiyā buddhavacanaṁ dūsenti. Handa mayaṁ, bhante, buddhavacanaṁ chandaso āropema
They ruin the Buddha’s word with their own terminology. Please sir, let us elevate the Buddha’s word with pleasing poetry.

Now, if we compare with the Dharmaguptaka, the first sentence would be simply that they are ruining the Buddha’s word; and the second part has the portion on “fine speech”. Perhaps, then, we should reconstruct the Pali thus:

Te buddhavacanaṁ dūsenti. Handa mayaṁ, bhante, buddhavacanaṁ chandaso āropema sakkatāya niruttiyā
They are ruining the Buddha’s word. Please sir, let us elevate the Buddha’s word with pleasing poetry in esteemed terms.

This would bring the Pali in line with the Dharmaguptaka, and closer to the other Vinayas. And it gives the same word order as in the commentary.

This is a less confident reconstruction than the former part, as we have less clear agreement among the Chinese texts, and it requires both a change in vocabulary and in word order. Perhaps a closer reading of the Chinese would help, as well as an examination of the Tibetan Vinaya, which has so far not been considered at all. Nonetheless this does give us a path towards resolving three apparently disparate features of the texts:

  • partial agreement among the Chinese
  • a ready hypothesis unifying the Chinese with the Pali (sat/sakāya)
  • the curious phrasing of the Pali commentary.

Putting both these reconstructions together, the key passages would then be:

“They are ruining the Buddha’s word. Let us elevate the Buddha’s word with pleasing poetry in esteemed terms.” …
“I allow you to memorize the Buddha’s word in the common vernacular.”

If this, or something like it, turns out to have merit, then it would seem the hunt for the meaning of the phrase sakāya niruttiyā is a wild goose chase, as it is merely a scribal error.

11 Likes

Which Pali text you refer to?

Does this Pali text have any corresponding counterparts (such as Chinese versions)?

From the Khuddakavatthu.

https://suttacentral.net/pli-tv-kd15/en/brahmali?lang=en&layout=sidebyside&reference=main&notes=sidenotes&highlight=false&script=latin#33.1.1

Yes, the passages discussed above.

1 Like

The usual way is to publish your original research in an academic article, translate according to the Pāḷi as it is and add a note (foot or end) explaining that you believe this to be a textual corruption, with your preferred reading and in parentheses a citation to your article on the subject.

2 Likes

Good suggestion! Shall we just renderer it as “one’s own dialect/language/interpretation/expression”.

The second occurrence in the Pali text should not present as “You should learn the word of the Buddha using its own expressions”.

I’ve just updated the article, with a new section on sakāya niruttiyā, which I now suspect may in fact be a ghost word.

1 Like

I consider this is not the Pali text’s issue, but the Buddhaghosa’s interpretation problem (based on his own sectarian tradition). Buddhaghosa explains that sakāya niruttiyā means the Māghadhan language, which is Pali, spoken by the Buddha. Buddhaghosa ignores the textual context, although being the same words, sakāya niruttiyā.

So, the second occurrence of the same words, sakāya niruttiyā, in the Pali text should be translated as “You should learn the word of the Buddha using one’s own expressions/languages ”.