[Buddha-l] Ariyapariyesana

Dan Lusthaus vasubandhu at earthlink.net
Thu Nov 10 02:42:33 MST 2011

Neither brief nor long-winded:

>> Maybe I read too much Yogacara and Mahayana, but āgantuka-kleśa is the
>> term of choice.

> I think that perhaps you do. I cannot find the term in any immediately
> accessible source in any non-Mahāyāna text preserved in Sanskrit. Of
> course, the notion that upakkilesā are adventitious is found in Pali,
> but never the compound āgantukakilesa.

I haven't time to hunt down references at the moment, but what first popped 
to mind is Sthiramati, in his Triṃśikā-vijñapti-bhāṣya, in the bhāṣya to 
4ab: "...āgantukair upakleśair..." - broken up in typical commentarial style 
(granted, this too is upakleśa). The Chinese 客塵煩惱 kechen fannao is an 
attested equivalent for āgantuka-kleśa, āgantukatā, āgantuka-mala, 
āgantukair upakleśaiḥ, and āgantu-kleśa. One finds that Chinese compound 
roughly 350 times in the Taisho ed. of the canon (and reversing the terms --  
煩惱客塵 fannao kechen -- which is another version, occurs 30 times). The 
term 雜染 zaran occurs over 3500 times. Depending on the translator, it 
indicates saṃkleśa, saṃkliṣṭa in Skt, or āgantuka-kleśa. Many Chinese 
discussants understood these terms as synonyms, i.e., they understood zaran 
as "adventitious defilements" (雜 za can mean both "mixture, heterogenous 
mixture, mixed up" and "gathered together, combined").

>> Then the cloth sutta, [...] Can we decide which are earlier and which 
>> later
>> on the basis of whether 'disturbed' or 'infiltrated' better fits?

> I will have to see that to comment.

I described it in a previous message and provided some of Ven. Analayo's 
discussion. Sorry if I was unclear about which sutta I was referring to:
MN 7; PTS: M i 36
Vatthupama Sutta: The Simile of the Cloth

>> Sthiramati's commentary on Vasubandhu's Pancaskandha-prakarana preserved 
>> in

> The recently found Sanskrit text is being edited here in Oxford at the
> moment by Dr Kramer from Vienna, but I don't know whether it will have
> any bearing on this issue.

We are eagerly awaiting this, along with some of the other recent finds. The 
sole Chinese version, translated by Divakara, is extremely truncated, 
compared to what the Tibetan contains. Their explanations often overlap, but 
the explanation of prasAda re: rUpa-prasAda is one of the places where, even 
though the Ch and TIb are running parallel their 'examples' diverge, each 
using a different pool analogy. I'm not sure the Skt will solve the question 
of what Divakara was (or wasn't doing), but it will be good to have it.

> Actually, Ānanda doesn't propose visiting the brahmin at all. He
> suggests going to the brahmin's ashrama (assamo) and points out that it
> is an attractive spot. This is a place provided by the brahmin for
> mendicants, perhaps specifically for Buddhist monks. There is no
> expectation that the brahmin will actually be there.

I understand your point. You take it as Ananda implicitly alerting Buddha 
that he is being expected to 'teach' at the hermitage. That is a possible 
reading. As I noted (relying on Analayo), the entire setup up to the actual 
talk is found in the Anguttara. From a fn in Analayo's book (p. 170 n. 136):

"The same introductory narration – covering the Buddha’s stay at Jeta’s 
Grove, his begging alms in Savatthi, his going with Ananda to the Hall of 
Migara’s Mother for the day’s abiding, and his approaching the Eastern 
Bathing Place to take a bath in the evening – recurs as the introduction to 
another discourse, AN 6:43 at AN III 344,18. The remainder of this discourse 
proceeds differently, as it records how the Buddha explained to Udayi what 
constitutes a real naga."

When one adds in the ingredient of the monks complaining that they don't get 
to hear Buddha speak much and not at all recently (not included in the AN 
version), prompting Ananda to set up what I consider an ambush (you don't 
think it's an ambush; you consider Buddha fully warned), I see a subtext 
there that humanizes the whole situation. It also gives some credit for 
ingenuity to Ananda. Similarly, Buddha's politeness, waiting for a lull and 
announcing himself by clearing his throat and knocking on the door, is a 
charming detail, much easier to relate to than a Buddha who announces he is 
about to speak by zapping the infinite realms in the 10 directions with 
multicolored laser beams streaming from his noggin.

>I wouldn't doubt that the linguistic
> form of the Pali texts is closer to the form in which the texts were
> originally written down than the highly Sanskritized versions of the
> so-called Mūlasarvāstivāda. The issue is more complex with the Recension
> One texts from Central Asia, not to mention the material in Gāndhārī. I
> would also suspect that the content of some of the other texts (e.g. the
> Ekottarika) is later, as in this case.

There are many indications that much of the Agama material translated in 
Chinese was coming from prakrits of various types -- many translation 
'errors' have been identified as misreading based on prakrit forms. (Analayo 
collects some of them; others, such as Jan Nattier, have documented numerous 
other examples). There are still some debates about the affiliation of each 
of the Agamas preserved in Chinese (they do not come from the same source). 
There are also debates about the affiliation of some of the translators, 
which may have had some influence on how they read/rendered the texts they 
were translating.

While the Sanskritized Agamas of the Mūlasarvāstivāda that are extant 
represent a *linguistically* later form (or so we may presently presume), 
their *content* sometimes preserves an earlier passage or variant, as 
Schopen and others have shown.

So, in general, one should not really generalize about one entire corpus vs 
another corpus re: content, but take every passage, sentence, and term on a 
case by case basis, with as many witness texts as possible.

> In regard to Analayo's earlier work, I tended to think he over-favours
> the Chinese. He seems to use a principle of evaluation which I would
> call 'primacy of the plebeian' — if there is anything complex or subtle
> in the Pali, it must be a corruption or later addition. Since I suspect
> that the Chinese translators also tended to simplify, I suspect that
> distorts the situation.

That is not my impression of his work at all. He is paying attention to the 
Chinese texts because, (in my opinion) (i) they have been largely ignored 
and marginalized in discussions of the Nikaya-Agama lit. until recently, or 
treated in crude block generalities rather than careful reading of their 
contents, (ii) they shed a great deal of light on the degree of variation --  
in passages, in redaction placement, in doctrinal lists, in doctrinal 
interpretations, in the names to whom various actions and sayings are 
attributed, etc. -- which in turn (iii) sheds a great deal of light on how 
the Nikaya-Agama project worked, mutated, and was used. Nor do I see him 
biased for or against the Pali, Ch., etc. On the contrary, I find him very 
restrained about drawing conclusions and implications. Since he tends to 
discuss the texts using paraphrases with occasional translated passages, 
rather than putting up full and detailed translations of each of the 
versions side by side (which would be a monumental undertaking -- and his 
Comparative MN book, 2 vols, is already monumental), the full impact of the 
degree of variance between the Pali and the Chinese trs. may not always be 
fully evident in the English discussion (he's hitting highlights, not 
detailing every shred of variant minutiae). That is much clearer if one 
reads the Pali and Chinese versions side by side.

As for suspicions of possible distortions introduced by translators -- the 
same could be said for transmitters of the Pali (hence the need for 
Buddhahosa to clean up the mess). But more importantly, while we only have 
very late variant editions of the Pali (Sinhala vs Burmese, etc.)(the 
Buddhaghosa project successfully eliminated variants), we have for many 
Agamas multiple translations of the same sutta, by different translators, 
most probably from different root texts, so -- unless one wants to blanket 
accuse every single translator of making exactly the same sorts of errors a 
priori, which would be an odd position to hold without having read a single 
word of their works -- comparisons between the different Chinese versions, 
in additional comparison to Pali and (when available) Skt versions, provides 
a check against that sort of criticism. In fact, sometimes the Ch. has less 
detail than the Pali, sometimes a great deal more.

Analayo does occasionally speculate about primacy on the basis of 
'coherency' -- when a passage fits the context, when it doesn't, when its 
occurrence is awkward, etc., in one version but not in another, or when 
something odd doesn't appear at all in other versions, or in some but not 
all, and so on. One might argue that coherency can just as much be a sign of 
a later revisionist or redactor cleaning/clearing up a unclear passage as it 
might be a sign of an original 'coherent' statement that gets garbled in 
transmission. There is no a priori for this. One has to, again, take all 
such instances on a case by case basis, and one is free to agree or disagree 
with Analayo's impressions and conclusions. Either way, he has done a 
tremendous service by putting the evidence easily within reach after having 
organized and collated it for us.

Once you have had an opportunity to examine Analayo's book (more thorough 
than his previously published articles on some of the suttas -- articles 
which I also find useful and thoughtful), perhaps you might revise your 


More information about the buddha-l mailing list