Introduction

In the present chapter I continue studying the topic of discourse merger in the Ekottarika-āgama, broached in the previous chapter. In what follows I take up one more such instance, namely an Ekottarika-āgama discourse that corresponds to two separate discourses in the Majjhima-nikāya and the Madhyama-āgama. Inasmuch as discourse merger studied here and in the previous chapter has considerable bearing on our understanding of the nature and transmission history of the Ekottarika-āgama collection preserved in Chinese translation, in this chapter I also critically examine suggestions made in a recent study by Antonello Palumbo regarding the circumstances of the translation of this Ekottarika-āgama collection.

I begin with a study of the seventh discourse in chapter fortynine of the Ekottarika-āgama, which combines material found in the Majjhima-nikāya in the Bhaddāli-sutta (MN 65) and the Laṭukikopama-sutta (MN 66), as well as in their Madhyama-āgama parallels.[n.1] [2] In Anālayo 2008a: 10 I already drew attention to the peculiar nature of EĀ 49.7 and EĀ 50.8. I first summarize the two Majjhima-nikāya discourses and their Madhyama-āgama parallels, then give a translation of the Ekottarika-āgama version, and in the final part of this chapter turn to a discussion of the transmission of the Ekottarika-āgama inasmuch as this seems relevant to the present study.

The Bhaddāli-sutta and the Laṭukikopama-sutta

The Bhaddāli-sutta and its Madhyama-āgama parallel report that the Buddha instructed his monks to take only a single meal per day. Bhaddāli refuses to comply,[n.2] [3] MN 65 at MN I 437,25 and its parallel MĀ 194 at T I 746b27; for a more detailed comparative study of MN 65 and MN 66 cf. Anālayo 2011c: 358– 367. a refusal he keeps up even when the Buddha offers him an alternative by way of compromise. For the three months of the rainy season Bhaddāli keeps up the same attitude. Just before the Buddha is about to set out wandering again, other monks prompt Bhaddāli to approach the Buddha and confess his transgression.

Before accepting this confession, the Buddha draws Bhaddāli's attention to the fact that his refusal to follow the rule promulgated by his teacher has become public knowledge. He then contrasts Bhaddāli's behaviour to the type of conduct that any out of a listing of seven types of noble disciple would have shown. Next he explains the importance of having a sound foundation in moral training in order to be able to withdraw into seclusion and practise successfully.

Bhaddāli then enquires why only some monks are repeatedly admonished. The Buddha explains that the degree of admonishment depends on how a particular monk reacts on being admonished. Bhaddāli has another query about why in earlier times there were fewer rules and more monks reached final knowledge. In reply, the Buddha points to the general growth of the monastic community in gains, renown, etc., which has led to the need for more rules and also to fewer monks becoming accomplished in the practice. The final parts of the Bhaddāli-sutta and its Madhyama-āgama parallel report the Buddha delivering a simile of a thoroughbred horse to Bhaddāli.

The Laṭukikopama-sutta and its Madhyama-āgama parallel begin with the monk Udāyin approaching the Buddha and expressing his appreciation for the rule on abstaining from food after noon.[n.3] [4] MN 66 at MN I 448,3 and its parallel MĀ 192 at T I 741a9. Udāyin relates an incident when a monk in the past had gone begging during a stormy night and thereby frightened a woman.

The Buddha then illustrates the predicament of those who do not submit to the training with the example of a weak animal's inability to break free from a feeble bond, whereas a strong elephant is able to break free even from a solid leash. Another set of two similes contrasts a poor man in wretched living conditions, unable to give them up, with a rich man who is able to let go of his many possessions and go forth.

Next the Buddha divides the arising of sensual thoughts into four distinct cases. Some tolerate them, others dispel them, either slowly or quickly, and still others are fully liberated from them. The topic of sensuality leads on to the contrast between ignoble pleasures and commendable pleasures. The latter pleasures are those experienced during absorption attainment, an indication which leads to a tour of the four absorptions and the four immaterial attainments from the viewpoint of what in each case needs to be overcome to progress from one to the next.

The discourse from the Ekottarika-āgama translated below combines the beginning part of the Bhaddāli-sutta, namely Bhaddāli's refusal to follow the rule on eating a single meal and his subsequent repentance, with part of the Laṭukikopama-sutta, namely begging at night and thereby frightening a woman.[n.4] [5] The translated discourse is EĀ 49.7 at T II 800b27 to 801c13. I have divided the discourse into sections and numbered these for ease of reference during my subsequent discussion; the numbering and the divisions are not found in the Chinese original and they also do not correspond to Ñāṇamoli 1995/2005.

---

Translation of EĀ 49.7

1. Thus have I heard. At one time the Buddha was staying at Sāvatthī in Jeta's Grove, Anāthapiṇḍika's Park.

2. At that time the Blessed One told the monks: "I always take my meal in a single sitting [per day], and my body is at ease, strong, and thriving. Monks, you should also take a single meal [per day], and your body will be at ease, strong, and thriving, enabling you to cultivate the holy life."

3. Then Bhaddāli said to the Blessed One:[n.5] [6] My identification of the proper name follows Akanuma 1930/1994: 86. "I cannot endure a single meal. The reason is that my strength will become feeble."[n.6] [7] According to MN 65 at MN I 437,27, he explained that this would worry him, which the commentary Ps III 148,12 glosses to mean that he would be worried if he could live like this for his whole life. According to MĀ 194 at T I 746b28, he was worried that with a single meal he would not be able to settle the matter (of nourishing himself). The Buddha said: "If you approach a house for meals, you can eat one part [there] and take one part back to your hut."[n.7] [8] In MN 65 at MN I 437,28 such a permission to keep food for later applies to occasions when Bhaddāli is invited for a meal; according to MĀ 194 at T I 746c1 he could take away food for later consumption after coming along when the Buddha had been invited.

Bhaddāli said to the Buddha: "I cannot endure undertaking this practice either." The Blessed One said: "I permit you to break your fast and partake of the food [you have taken back] throughout the day."[n.8] [9] Such an additional option is not recorded in MN 65 or MĀ 194. It does in fact not fit the context too well, since with such an allowance Bhaddāli's concerns about getting enough food would have been allayed, leaving little reason for him to refuse undertaking this mode of conduct.

Bhaddāli said to the Buddha: "I also can't endure being allowed to undertake this practice." Then the Buddha remained silent and did not reply.

4. At that time, towards nightfall,[n.9] [10] For what follows, the parallels are MN 66 and MĀ 192, both of which present the event as something that Udāyin told the Buddha. Kāḷudāyin put on his robes, took his bowl, and entered the town to beg alms. At that time it was extremely dark.[n.10] [11] In MN 66 at MN I 448,33 the monk who went begging was also Udāyin himself, whereas MĀ 192 at T I 741b9 just speaks of a monk in general. A similar incident is recorded in the Dharmaguptaka and Mahīśāsaka Vinayas, T 1428 at T XXII 662c8 and T 1421 at T XXII 54a19, as leading to the promulgation of a rule on not eating at the wrong time, and in the Mahāsāṅghika Vinaya, T 1425 at T XXII 359b25, for occasioning a rule on not going begging at the wrong time. Then Kāḷudāyin ⟨gradually⟩ approached the house of a householder.[n.11] [12] Adopting an emendation suggested in the CBETA edition of 慚 to 漸. Yet the wife of that householder was pregnant. When she heard that outside a recluse was begging alms,[n.12] [13] According to MĀ 192 at T I 741b10, she had been outdoors, washing a pot. MN 66 at MN I 448,34 also records that she had been washing a pot, which presumably would have happened outdoors. she took rice and went out to give him alms. However, Kāḷudāyin's complexion was very dark and at that time it was moreover about to rain, lightning was flashing all around.[n.13] [14] Adopting the variant 泄 instead of 抴. MĀ 192 at T I 741b11 also refers to lightning, which is not mentioned in MN 66.

Then the householder's wife, on coming out of the door, saw a recluse with a very dark complexion. She right away shouted in alarm: "This is a ghost!" Then she called out to herself: "Oh, I have seen a ghost!" Then she straight away had a miscarriage and the baby died.[n.14] [15] MĀ 192 at T I 741b12 also records that she had a miscarriage. MN 66 at MN I 449,1 does not mention a miscarriage, instead it reports that the frightened woman abused Udāyin once she found out that he was a monk (her abuse is also reported in MĀ 192). Another occurrence of the motif of a monk frightening a pregnant woman and thereby causing a miscarriage can be found in T 129 at T II 845a8, translated in Tokiwai 1898: 49; cf. also Lévi and Chavannes 1916: 264 or Strong 1979: 74. Then Kāḷudāyin returned to the monastery,[n.15] [16] What follows has no counterpart in MN 66 or MĀ 192. being worried and sad. He sat thinking to himself, regretting it in vain.

5. At that time in the city of Sāvatthī there was this bad rumour: "The recluses, sons of the Sakyan, have a charm whereby the children of others will be miscarried." The men and women said among each other: "Now these recluses behave improperly, they do not know the time for [begging] food. Compared to white-clad laymen, what is the difference?"

6. Then a group of many monks heard people discussing with each other about this matter: "The recluses, sons of the Sakyan, do not know what is proper, they come [begging] without scruples." Those among the precept-observing monks who were complete [in their practice of] the precepts also blamed themselves: "It is true that we are improperly eating without limits, acting without [regard for] the proper time. It is true that we are wrong." They together approached the Buddha, paid homage with their heads at his feet, and told the Blessed One all that had happened.

7. The Buddha then told one monk: "Go and summon Kāḷudāyin and bring him here." Having received the Buddha's instruction, that monk then approached Kāḷudāyin to call him. On hearing that the Buddha had summoned him, [wanting] to see him, Kāḷudāyin approached the Blessed One, paid homage with his head at [the Buddha's] feet, and sat down to one side.

8. Then the Blessed One asked Kāḷudāyin: "Did you indeed yesterday at nightfall enter the town to beg alms, go to the householder's house, and cause the householder's wife to have a miscarriage?" Kāḷudāyin said to the Buddha: "Indeed, Blessed One."

The Buddha said to Kāḷudāyin: "Why do you not distinguish the proper time, instead of entering the town to beg alms when it is about to rain? This is not proper for you. You are a clansman's son who has gone forth to train in the path, yet you are greedily attached to food." Then Kāḷudāyin rose from his seat and said to the Blessed One: "From now on I will not dare to transgress again. I only wish that the Blessed One accepts my repentance."

9. Then the Blessed One said to Ānanda: "Quickly strike the wood to gather all the monks in the community assembly hall." Having received the Buddha's instruction, Ānanda had all the monks gather. When they had gathered in the assembly hall, he went forward to inform the Buddha: "The monks have gathered, Blessed One, you know the right time [to join them]."

10. At that time the Blessed One went to the assembly hall, sat in its centre, and said to the monks: "Buddhas, Blessed Ones of the distant past, all ate in a single sitting and their disciples also ate in a single sitting. Even Buddhas and the community of their disciples in the future shall also all eat in a single sitting.

"The reason is that this way of practice is an essential teaching. One should eat in a single sitting. If one is able to eat in a single sitting, the body will be at ease and the mind will gain clarity of understanding. When the mind has gained understanding, one gains the roots of all that is wholesome. Having gained the roots of what is wholesome, one in turn gains concentration. Having gained concentration, one knows as it really is. What does one know as it really is?

"That is, one knows the truth of dukkha as it really is, one knows the truth of the arising of dukkha as it really is, one knows the truth of the cessation of dukkha as it really is, and one knows the truth of the way out [of dukkha] as it really is.[n.16] [17] For a discussion of the absence of the qualification "noble" in Ekottarika-āgama discourses cf. below p. 239ff.

"You are clansmen's sons who have gone forth to train in the path, leaving behind the eight deeds of the world,[n.17] [18] EĀ 49.7 at T II 801a18: 世八業. Perhaps this refers to the 世八法 (or 世間 八法), which are the eight worldly conditions of profit and loss, honour and disgrace, praise and blame, pleasure and misery. yet you do not know the proper time. You are like other people who have greedy desires, what is the difference? [In contrast], brahmins have their own distinct brahminical principles; heterodox practitioners have their own distinct heterodox practitioner's principles."[n.18] [19] EĀ 49.7 at T II 801a19: 外道, literally "outside path". For want of a better term, here and elsewhere I use the translation "heterodox", in the sense that these practitioners were, together with Buddhist monastics, members of the recluse (samaṇa) community in ancient India, but dissented from the Buddhists on points of doctrine and practice.

11. Then Upāli said to the Blessed One: "As Tathāgatas of the past and Buddhas of the future all eat in a single sitting, I only wish that the Blessed One restricts the time for the monks to eat."

The Blessed One said: "The Tathāgata as well has this understanding [that a restriction is required], it was just that there was no violation. There must be an offence before his eyes, only then shall he set a restriction."

12. Then the Blessed One told the monks: "I solely eat in a single sitting. You should also eat in a single sitting. [From] now on you should eat by noon, do not go beyond that time.

"You should also train in the practice of begging alms. How does a monk train in the practice of begging alms? Thus, monks, [train by begging alms just] for the purpose of supporting life, being neither pleased on getting them nor being distressed on not getting them. When you get food, then eat attentively. Be without greedy attachment in the mind, just wishing to make the body get its maintenance, to discard past feelings, without further creating new ones, and to make your strength be ample and full. Monks, like this one is reckoned to be begging alms [properly].

"Monks, you should eat in a single sitting. How does a monk eat in a single sitting? [As soon as] one gets up one has 'broken' the meal [session] and should not eat further. Monks, like this one is reckoned to be eating in a single sitting.

"Monks, you should also eat [just] the food you got. How does a monk eat [just] the food he got? Monks, upon having already got [food] and eaten [from] it, what about more [food] that might be prepared for you? If while eating one gets some further [food], then it is not proper to eat further. Like this monks, eat [just] the food you got.

13. "Monks you should also have [only] three robes to wear, you should sit under trees, sit in a quiet place, you should sit in the open as an ascetic practice, you should wear patchwork robes, you should stay in cemeteries, and you should wear rag robes.

"The reason is that I praise a person who has few desires. I shall teach you now, you should be like the monk Kassapa. The reason is that the monk Kassapa himself undertakes the eleven/twelve ascetic practices and also teaches others the undertaking of these important practices.[n.19] [20] The present passage in EĀ 49.7 at T II 801b12 refers to eleven ascetic practices, 頭陀十一法行, with a variant mentioning twelve (reading 二 instead of 一). The count of the ascetic practices varies between eleven and twelve in the Ekottarika-āgama. EĀ 5.1 at T II 558c24 and EĀ 49.2 at T II 795a26 refer to eleven, EĀ 49.3 at T II 795c10 refers to eleven with a variant reading twelve, EĀ 23.3 at T II 612a19 has twelve with a variant reading eleven, EĀ 4.2 at T II 557b8 and EĀ 48.3 at T II 788c27 refer to twelve (the listing of twelve in EĀ 4.2 has already been noticed by Boucher 2008b: 191 note 8). On variations in listings of the ascetic practices in general cf., e.g., Bapat 1937, Ganguly 1989: 21–23, Nanayakkara 1989: 584, Dantinne 1991: 24–30, Ray 1994: 293–323, and Wilson 2004: 33.

"I shall now admonish you, you should be like the monk Mogharāja.[n.20] [21] EĀ 49.7 at T II 801b13: 面王比丘; cf. Akanuma 1930/1994: 428. He also features in the list of outstanding monks in EĀ 4.6 at T II 558a14 and its counterpart in AN 1.14.4 at AN I 25,16. The two lists agree that Mogharāja was outstanding for the quality also highlighted in the present context, namely his wearing of coarse robes. The reason is that the monk Mogharāja wears coarse robes and does not wear them to adorn himself. Therefore I teach and admonish you, monks, so that you remember to practise in this way. Monks, you should undertake this training."

14. At that time Bhaddāli did not approach the Blessed One until the three months [of the rainy season] had passed. Then, for the first time in these three months, Ānanda approached the monk Bhaddāli and said:[n.21] [22] MN 65 at MN I 438,9 and MĀ 194 at T I 746c26 report that the monks were making a robe for the Buddha. In both versions Bhaddāli then approached these monks, whereupon they told him that he should reconsider the situation, lest later on it will be more difficult for him. Bhaddāli then went to see the Buddha on his own, without being accompanied by Ānanda. "In the community all are now mending their robes. Thus the Tathāgata will [soon] be travelling among the people. [If] you don't approach him now, it will be of no use to regret it later."

15. Ānanda brought Bhaddāli to the Blessed One. [Bhaddāli] paid homage with his head at [the Buddha's] feet and further said to the Buddha: "May the Blessed One accept my repentance, from now on I shall not transgress further. The precept had been laid down by the Tathāgata, but I refused it. I wish to be forgiven." He spoke in this way three times.[n.22] [23] According to MN 65 at MN I 438,28 and MĀ 194 at T I 747a14, in reply to Bhaddāli's request to be forgiven, the Buddha drew Bhaddāli's attention to the fact that his obstinate behaviour had become public knowledge.

16. Then the Buddha said: "I accept your repentance. In future you should not again transgress.[n.23] [24] What follows is without a counterpart in MN 65 and MĀ 194. The only point of distant resemblance is that both versions at a later juncture refer to recollection of past lives as part of their description of the results to be expected by someone who fully submits to the training; cf. MN 65 at MN I 441,31 and MĀ 194 at T I 748a15. Note also the shift in the course of paragraphs 16 and 18 from the Buddha being in conversation with Bhaddāli to his addressing Upāli. The reason is that I recall my own innumerable births and deaths, where I have been a donkey, or a mule, a camel, an elephant, a horse, a pig, a sheep, and by relying on grass I nourished this body [made] of the four elements. Or else I have been in hell and swallowed hot iron balls. Or else I have been a hungry ghost, continuously eating pus and blood. Or I had a human body, eating the five cereals. Or I had the body of a deva, eating spontaneously [manifesting] ambrosia. During innumerable aeons I had a bodily appearance and lived in competition [for food], never getting sated.

"Upāli you should know, it is just as fire catches on firewood, never getting sated, it is just as the great ocean swallows rivers without getting sated. Now ordinary mankind is like this too, greedy for food without getting enough of it."

17. Then the Blessed One spoke this poem:
"Birth and death are without end,
All because of greedy desires;
Resentment increases the evil thereof,
Being cultivated by the fool."

18. "Therefore Bhaddāli, you should remember to have few desires and know contentment. Do not give rise to greedy perceptions and to disorderly thoughts. Like this, Upāli, you should undertake this training."

19. Having heard the Tathāgata's admonition, Bhaddāli then went to stay in a secluded place and reproached himself. Cultivating the supreme holy life for whose sake a clansman's son goes forth to train in the path, he came to know as it really is that birth and death have been extinguished, the holy life has been established, what had to be done has been done, there will be no further experiencing of becoming. At that time Bhaddāli became an arahant.

20. At that time the Blessed One said to the monks: "The monk Sirigutta is the foremost among my disciples for [receiving] much food and drink."[n.24] [25] EĀ 49.7 at T II 801c12: 吉護比丘; cf. Akanuma 1930/1994: 621 and T 2128 at T LIV 475c15. The Pāli commentarial tradition knows a layman and streamenterer by this name; cf. the tale in Dhp-a I 434,18 (translated Burlingame 1921: 92–99; cf. also Malalasekera 1937/1995: 753f s.v. Garahadinna); a reference to him occurs also in Th-a II 211,5. According to the tale in question, Sirigutta invited the Jains for a meal and humiliated them by causing them to fall into a ditch filled with filth, thereby proving that they were not omniscient. In revenge, Garahadinna tried a similar ruse with the Buddha, with the difference that the ditch was filled with glowing coals. The Buddha divined it all and through his magical power avoided that any harm occurred. The basic storyline recurs in a range of other texts, but with the difference that here Śrīgupta is the one who invited the Buddha to a meal intending to harm him with hidden fire and poisoned food; for a survey of various versions of this tale cf. Lamotte 1944/1981: 184f note 4 and Zin 2006: 124–133, to which the Śriguptāvadāna in Straube 2009: 94ff could now be added; cf. also below note 26.

Then the monks, having heard what the Buddha had said, were delighted and received it respectfully.

---

Study

The Narrative Flow of EĀ 49.7

Surveying the narrative flow of the above-translated discourse, a striking feature is the at times rather abrupt change between its various protagonists. The discourse begins with Bhaddāli who refuses to follow the ruling on taking a single meal (1–3). Then the scene shifts to Kāḷudāyin's begging and its repercussions, up to his repentance (4–8), followed by the Buddha giving a general instruction to the monks that have been summoned by Ānanda, among them also Upāli (9–13). Then the discourse reverts to Bhaddāli who repents and eventually becomes an arahant (14–19), with interim references to Upāli. The final part of the discourse has a conclusion that praises the monk Sirigutta (20), who has not been mentioned before.

The shift from Bhaddāli (3) to Kāḷudāyin (4) is unexpected, but in itself not problematic, since it could simply be that the Bhaddāli story has now come to an end and the narrative continues with another event. This assumption becomes somewhat problematic, however, when the Buddha promulgates the rule on eating in a single sitting (10), which he already did at the outset (2). The second promulgation (10) has a rationale, as it takes place in response to the incident of going begging at night and thereby frightening a woman. This follows a basic principle underlying Vinaya rules, namely that they are pronounced in response to a precedent. The discourse in fact explicitly refers to this principle, in that, for the Tathāgata to make a rule, "there must be an offence before his eyes, only then shall he set a restriction." From the viewpoint of narrative logic, the precedents leading to the rule described at the present juncture (10) should come before the rule to Bhaddāli (2), which is about the reactions the rule caused. The precedents should not come after the promulgation of the rule, as is currently the case.

Moreover, the ruling given at the present juncture (10) does not fully address the issue at hand, since eating at a single time does not necessarily exclude the possibility that someone may go begging late. To stop the possibility of future incidents of the type caused by Kāḷudāyin, a rule on abstaining from food after noon would be required. Since monks are not allowed to keep food overnight, this would automatically exclude the possibility that they go begging after noon has passed.

The choppy progression of the narration becomes particularly evident when Bhaddāli reappears (14), introduced by the indication that he "did not approach the Blessed One until the three months [of the rainy season] had passed". This would have its natural placement after the Buddha had remained silent, once Bhaddāli had refused to follow the rule (3). In contrast, it does not seem to have any connection to the intervening events related to Kāḷudāyin. This gives the impression that the part on Kāḷudāyin has been inserted in the midst of the Bhaddāli tale.

The textual confusion becomes worse when, while teaching Bhaddāli, the Buddha suddenly addresses the simile of the firewood to Upāli (16). Upāli had been present earlier, presumably as one of the monks Ānanda had summoned on behalf of the Buddha. But now, at the end of the rainy-season retreat, when Ānanda has brought Bhaddāli to the Buddha's presence, there is no reason why Upāli should be there and why the Buddha should suddenly turn to him. That this is indeed a textual corruption can be seen from the instruction given by the Buddha after he has spoken a poem on greedy desires (17). The instruction (18) begins "therefore Bhaddāli, you should remember to have few desires and know contentment", but then concludes: "like this, Upāli, you should undertake this training."[n.25] [26] EĀ 49.7 at T II 801c5: 是故, 跋提婆羅, 當念少欲知足, 無起貪想, 興諸 亂念. 如是, 優波離, 當作是學.

The perplexing shifting from one protagonist to another continues right up to the end of the discourse with the appearance of Sirigutta (Śrīgupta), who has not been part of any of the earlier events. He does appear in another discourse found earlier in the Ekottarika-āgama, although there he is a layman who offers beverages and food to the Buddha and his monks. [n.26] [27] EĀ 45.7 at T II 773c22, which uses the alternative 尸利掘 to render his name; for a summary of the tale cf. Lévi 1908: 158f. Here the layman Sirigutta/Śrīgupta has prepared a great variety of food and beverages to offer to the Buddha and his disciples, EĀ 45.7 at T II 774a18. This is part of an attempt to harm them, as both are poisoned. When the first part of his ruse to make the Buddha fall into a ditch filled with fire has not worked, Sirigutta/Śrīgupta confesses his evil intentions. The Buddha forgives him and then miraculously transforms the beverages and food by speaking a set of stanzas so that the poison disappears and they can be consumed by his disciples and himself. Yìjìng (義淨) reports that this transformation led to a custom to be observed by monastics on receiving food; cf. T 2125 at T LIV 209c17, translated in Takakusu 1966: 39. Fǎxiǎn (法顯) and Xuánzàng (玄奘) refer to the place where the ditch built by Sirigutta/Śrīgupta was found; cf. T 2085 at T LI 862c15, translated in Deeg 2005: 551 (107), and T 2087 at T LI 921a2, translated in Beal 1884/2001: 151; cf. also the discussion in Deeg 2005: 409f. In view of EĀ 45.7, I assume that the reference in EĀ 49.7 at T II 801c12 to being the foremost among disciples for "much beverages and food", 多飲食, refers back to the story about Sirigutta/Śrīgupta recorded in EĀ 45.7. Therefore I have supplemented in my translation that he "received" much of these, presumably as a karmic result from having offered food and drink to the Buddha and his community, even though originally done with evil intention. In fact just partaking of much food and drink would not be a quality that merits being highlighted, given that the according of the rank of being an outstanding disciple is to generate inspiration through qualities that are worth being emulated; cf. Anālayo 2014i. His appearance in the present context comes as an additional confirmation of a tendency to assemble different narrative bits and pieces that are in some way or another related to the topic of food.

The above inconsistencies make it safe to conclude that what we have here is not an originally single narrative preserved in the Ekottarika-āgama, which through subsequent expansion would have become two different discourses of the type now found in the Majjhima-nikāya and the Madhyama-āgama. Instead, it seems clear that the Ekottarika-āgama discourse translated above results from a merger of originally separate narrations related to the topic of monks and food, similar in kind to the instances of discourse merger surveyed in the previous chapter.

The Translation of the Ekottarika-āgama

This kind of merger of what appear to be originally separate episodes, as in the Ekottarika-āgama discourse translated above, is an unusual occurrence for orally transmitted discourses. During oral transmission, a discourse can of course incorporate textual material from elsewhere, or else part of a discourse can be lost.[n.27] [28] An example of incorporation of material would be the Mahāsatipaṭṭhāna-sutta (DN 22); cf. Anālayo 2014d: 91–100 (on MN 10 cf. Anālayo 2011c: 73– 97 and 2013g). An example of loss of a substantial portion of text would be the Chabbisodhana-sutta (MN 112); cf. Anālayo 2008b and 2011c: 635–639. But this would normally not result in something similar to the cases of merger discussed in the previous and the present chapter.[n.28] [29] For a comparative study of the Majjhima-nikāya that shows how various types of transmission errors have had their impact on this collection as well as the parallel versions cf. Anālayo 2011c. That this is probably not the result of an error during oral transmission finds confirmation in the abrupt shift between the names of Bhaddāli and Upāli in the final part of the discourse translated above (18). Oral transmission tends to standardize rather than introduce variations of this type.

By way of providing a background for this characteristic of discourse merger in the Ekottarika-āgama collection, in what follows I survey other aspects relevant to an appreciation of the circumstances of the Ekottarika-āgama translation.

In addition to instances of merger, a polemical reference to the hīnayāna can be found in the Ekottarika-āgama. This reference gives the impression of having come into being in the written medium, more specifically in the Chinese written medium.[n.29] Cf. below p. 460ff. Moreover, differences in translation terminology make it safe to conclude that another discourse was added wholesale to the Ekottarika-āgama in China.[n.30] Cf. below p. 165ff.

Another indication pointing in the same direction emerges from a recent detailed study by Palumbo (2013) of the Ekottarika-āgama and a partial commentary on this work, the so-called Fēnbié gōngdé lùn (分別功德論), or according to Palumbo rather the Zēngyī āhán jīng shū (增一阿含經疏). Palumbo (2013: 127) convincingly shows that a Sarvāstivāda prātimokṣa preserved in Chinese translation from a Dūnhuáng (敦煌) manuscript exhibits such a close degree of similarity with part of a discourse in the Ekottarika-āgama as to make it clear "that the translator(s) or editor(s) of the received text of the Zengyi ahan jing [Ekottarika-āgama] made use of the prātimokṣa text … as a building block for the composite sūtra 48.2."[n.31] [33] Cf. also Palumbo 2013: 142–144.

Taken together, these indications give the impression that the Ekottarika-āgama now extant in the Taishō edition as number 125 went through some degree of development in China itself. This supports the impression that at least some of the cases of merging of discourses discussed above would also be the result of something that happened only when the Indic collection had already reached China.

In his detailed study of the translation of the Ekottarika-āgama and the so-called Fēnbié gōngdé lùn, Palumbo (2013) brings together a wealth of details that clearly confirm that some alteration did take place in China. This much can thus be taken as a firm basis for future studies, that is, the Ekottarika-āgama as we now have it is the result of a reworking of the material that reached China.

Regarding the actual circumstances of such a reworking, Palumbo's reconstruction of the situation could be summarized as follows: The reciter of the collection, Dharmanandin (or, according to Palumbo, Dharmananda),[n.32] [34] Palumbo 2013: 5 note 12; the introduction to the collection at T II 549a10 gives his name as 曇摩難提 (tánmónántí). had forgotten part of the Indic original. This motivated the translation team under the leadership of Dào'ān (道安) to supply other material to make up for the gaps.[n.33] [35] Palumbo 2013: 276 envisages that "the entire translation of the Ekottarika-āgama would transform into much more of a collective undertaking, and other members of the group – Zhu Fonian, Dao'an, the other foreign masters – could step in on occasion to supply the missing portions. Versions of individual sūtras that were known within the group might even have been chosen to replace those that Dharmananda had initially recited."

In the context of the present chapter it is of course not possible to do full justice to Palumbo's monograph, which would require a proper review. Hence in the remainder of this chapter I only critically examine some aspects of his reconstruction inasmuch as they seem to me to be of relevance for assessing the occurrences of discourse merger in the Ekottarika-āgama that forms the topic of the present and the previous chapter.

To begin with, it seems to me that this kind of scenario would not fully explain the type of merger in the Ekottarika-āgama. I find it highly improbable that a lapse of memory by a reciter renowned for having memorized two Āgamas could explain the shift of names from Bhaddāli to Udāyin within a single passage, resulting in an instruction being addressed to one of them in its first part and to the other in its second part. This is not the kind of variation to be expected from orally recited material, but rather something that can naturally happen during hasty or careless copying from a written passage.

The scenario reconstructed by Palumbo would also not explain the finding of a text with substantially different translation terminology as part of the collection. This is the case for discourse EĀ 50.4, which is one of two versions of the same story found in the Ekottarika-āgama, concerning King Mahādeva.

All four proper names that occur in the two tales are rendered differently, and otherwise similar descriptions are translated in different ways. Standard pericope descriptions and key terminology are translated differently. Translations used in discourse EĀ 50.4 never occur anywhere else in the Ekottarika-āgama. Even the introductory and the concluding phrase used in this discourse are never found anywhere else in the Ekottarika-āgama.[n.34] Cf. below p. 183f. Given the staggering amount of differences in what is a relatively short narrative piece held in common between the two versions of the Mahādeva tale in the Ekottarika-āgama, it is simply impossible to assume that this discourse could have been produced by the same translator(s) that is/are responsible for the rest of the collection.[n.35] [37] Palumbo 2013: 275 note 13 sees even such major differences as "the expression of a different and arguably earlier stage in the process of translation of the Zengyi ahan jing rather than as the product of one or more different translators" (note that according to his reconstruction the earlier and later stages of translation would have taken place shortly after each other). Palumbo 2013: 280 note 21 then also objects to the quantitative analysis by Hung 2013: 129f, stating that "there does not seem to be any cogent reasoning behind the … conclusion" that individually translated Madhyama-āgama discourses "cannot be ascribed to Dharmananda and Zhu Fonian", a criticism raised again by Palumbo 2013: 132 note 76 as follows: "Hung 2013 rejects the attribution of these parallels [i.e. the individually translated Madhyama-āgama discourses] to the initial translation by Zhu Fonian and Dharmananda." Yet Hung 2013: 130 only states that "the individually translated Madhyama-āgama discourses also differ from the translation terminology used in the rest of T 125. This makes it improbable that these discourses and T 125 stem from the same translator, although firm conclusions in this respect require further research." This to my mind does not convey a wholesale rejection, nor does it state that these "cannot be ascribed" to certain translators. Instead I see Hung as correctly stating that his research points in that direction, which it indeed does, but leaving the situation open for future research to confirm his findings or otherwise bring up evidence that disproves them. With all due allowance for changes in translation terminology, etc., the sheer amount of differences found does not leave room for any other conclusion.

Not only does Palumbo's reconstruction not satisfactorily explain these findings, it is also based on assumptions which I find difficult to follow.[n.36] [38] I am also not convinced by Palumbo's assumption that there have been four redactions of the Ekottarika-āgama translation, a topic which in the context of the present chapter I cannot discuss in detail. One of these is the rationale for a revision which, according to his reconstruction, was that the reciter of the collection had forgotten parts of the Ekottarika-āgama. Another is his dating of the so-called Fēnbié gōngdé lùn, based on which it would seem that the revision of the Ekottarika-āgama must have taken place before the time of Kumārajīva and thus was probably undertaken by the translation team itself. In what follows I will first discuss the dating issue, and then turn to the presumed rationale for the revision.

Palumbo convincingly shows that the commentary on the first part of the Ekottarika-āgama extant in the Fēnbié gōngdé lùn must have come into being as part of the original translation efforts. Given that the Fēnbié gōngdé lùn reflects the part of the Ekottarika-āgama now extant in the Taishō edition on which it comments, this version would in essence correspond to the text finalized by the translation team under the guidance of Dào'ān.[n.37] [39] According to Palumbo 2013: 281, "what has been handed down to us is in essence, if certainly not in shape, the very improbable Ekottarika-āgama that Dharmananda, Zhu Fonian, Dao'an and the others laboriously produced."

Without intending to dispute in any way that elements in the Fēnbié gōngdé lùn must indeed go back to the original translation efforts, it seems to me that we cannot exclude that at that time mere notes were taken of discussions that took place in relation to the actual translation. Such notes might originally not have been intended for publication. If someone later should have undertaken a revision of the Ekottarika-āgama, it would have been only natural for the same person(s) to produce a version of these notes that is in line with the revised Ekottarika-āgama.

The arguments offered by Palumbo for dating the Fēnbié gōngdé lùn that I will take up in what follows are related to the translation activities of Kumārajīva. One of these arguments is that the Fēnbié gōngdé lùn espouses the view that Vinaya material is not meant for circulation outside the circle of those who have received full ordination.[n.38] [40] T 1507 at TXXV 46c21: 不可示沙彌及以白衣; a view also expressed in the introduction to the Ekottarika-āgama, T II 549a27: 外國不通與沙彌, 白 衣共視也; cf. also T 2145 at T LV 64b23. According to Palumbo (2013: 204), this was "blatantly disavowed in 405" when Kumārajīva received a manuscript enabling him to complete his translation of the Sarvāstivāda Vinaya in 406, whereupon "a new era had started, in which the precepts could not only be circulated, but also commented upon in written form, and even made the object of public lectures."

Another argument for dating the Fēnbié gōngdé lùn takes up a reference to the Buddha's disciple Kātyāyana as the author of the Jñānaprasthāna of the Sarvāstivāda Abhidharma, which Palumbo (2013: 210) believes to be a view that must have "already been discarded by the time Kumārajīva translated the Da zhidu lun 大 智度論 (A.D. 402–406)".[n.39] [41] T 1507 at T XXV 42c21. Moreover, two quotations in the Fēnbié gōngdé lùn from the so-called Larger Prajñāpāramitā and from the Vimalakīrtinirdeśa-sūtra rely on translations of these works done before Kumārajīva, instead of quoting the translations done by him.[n.40] [42] Palumbo 2013: 229–234.

These quotations do indeed give the impression that at the time the respective passages came into being, Kumārajīva's translations were not yet in circulation. Once the quotations were in place, however, there would have been little need for someone publishing these notes to replace them with the translations by Kumārajīva that in the meantime would have become available.

Regarding the reference to Kātyāyana, other works of the Sarvāstivāda Abhidharma, such as the Dharmaskandha and the Saṅgītiparyāya, are attributed by tradition to chief disciples of the Buddha, such as Śāriputra, Mahāmaudgalyāyana, and Mahākauṣṭhila.[n.41] [43] Anālayo 2014d: 153 note 76. This makes it less surprising for the notion that Kātyāyana was the author of the Jñānaprasthāna not be immediately rectified as soon as Kumārajīva had translated the Dà zhìdù lùn.

As for the suggestion by Palumbo (2013: 256) that "the esoteric view of the vinaya, which is repeatedly expressed in the commentary [i.e. the Fēnbié gōngdé lùn] … cannot be reconciled with the state of things after the translation of the Sarvāstivāda vinaya in A.D. 405–406", I am under the impression that this conclusion does not fully take into account the nature of such monastic attitudes. The stricture against allowing access to Vinaya regulations for those who have not received full ordination is an aspect of monastic conduct. The issue at stake is that a fully ordained monastic should not teach such matters to laity or even novices. In fact the Dà zhìdù lùn, translated by Kumārajīva, takes the same position.[n.42] [44] T 1509 at T XXV 66a12: 此毘尼中說, 白衣不得聞. Such an attitude also finds explicit expression in the translation of the Mūlasarvāstivāda Vinaya, undertaken three centuries later.[n.43] [45] T 1442 at T XXIII 672c4: 毘奈耶教是出家軌式, 俗不合聞. It is thus not possible to assume that any such indication could only have come into being before the time of Kumārajīva.

In my view the above arguments raised by Palumbo for dating the Fēnbié gōngdé lùn are not conclusive and the possibility remains open that this work could have been finalized during or after the time of Kumārajīva's translation activities.

The other assumption by Palumbo that I would like to discuss here is that part of the Indic original of the Ekottarika-āgama had been forgotten, which motivated the translation team to supply their own discourses. This assumption is based on the indication by Dào'ān that the reciter had forgotten some of the summary verses (uddāna).[n.44] [46] T II 549a17: 失其錄偈. Palumbo (2013: 276) comments:

how could Dharmananda forget the brief uddānas, and not the much longer sūtras that those mnemonic verses were supposed to summarize? The impression is that the Chinese monk [i.e., Dào'ān] is glossing over a far more embarrassing situation … Dharmananda may in fact have been unable to recite at least part of the sūtras of the relevant vargas.

Here I think it needs to be kept in mind that uddānas are not just summaries, but rather often meaningless strings of words, taken from different discourses in a particular chapter. Such meaningless strings of words are considerably more difficult to memorize and keep in memory than the discourses to which they refer, because the latter are built out of meaningful phrases that together form the theme of a particular discourse.

Moreover, these uddānas are not required in an actual teaching situation, but only come into their own when the whole collection needs to be recited chapter-wise, such as for the purpose of translation. In such a situation they serve to ascertain that the discourses are recited in their proper order.

Judging from variations between renderings of the same term found in an uddāna and in the respective discourse elsewhere in the Chinese Āgamas, it seems that the uddānas were recited apart from the collection and thus also translated separately.[n.45] [47] Cf. the study of this feature of the Ekottarika-āgama's uddānas by Su 2013: 205–207. In the case of a Dharma teacher like Dharmanandin/Dharmananda, who apparently had been travelling for quite some time, it seems quite conceivable that he still remembered the material he had been using regularly for preaching purposes, but no longer remembered all of the uddānas.

In fact, once Dào'ān was willing to record explicitly that part of the uddānas had been forgotten, why would he not similarly have been willing to record that part of the discourses had been forgotten, if that had indeed been the case? Numbers of Āgama discourses have been translated individually into Chinese, as well as partial Āgama collections or extracts (such as T 100, T 101, and T 150A). There seems to be no compelling reason why Dào'ān and his collaborators could not have produced a partial Ekottarika-āgama collection, had the reciter of the original indeed forgotten major parts of it. Thus it seems to me that the assumption that there was a need for the translation team to supplement discourses that had been forgotten is not convincing.

In sum, the hypothesis that an integration of new material into the Ekottarika-āgama took place during the actual translation and with the sanction of the entire team, including the reciter of the collection and Dào'ān, is to my mind not persuasive. Instead, as far as I can see, a more probable scenario would be that something took place subsequently, after the translation.

Be that as it may, the detailed research by Palumbo confirms the basic impression derived from other studies of the Ekottarika-āgama in so far as it can safely be concluded that this collection did undergo substantial change in China. It thus needs to be reckoned as a case apart from the other Āgamas and their Pāli Nikāya counterparts, which show no signs of having remained open to comparable changes at so late a time.[n.46] [48] On the main time frame reflected by the discourses in the four Pāli Nikāyas cf. Anālayo 2012e.