The Rhetoric of Brahmanical Ritual in Early Buddhism: A Comparison of the Five-Sacrifice Formula in the Pāli Nikāyas and Chinese Āgamas

Abstract

The Pāli Nikāyas preserve a recurrent set of five sacrificial terms—assamedha, purisamedha, sammāpāsa, vājapeyya, and niraggaḷa—here termed the “five-sacrifice formula.” This article compares the Ujjaya-sutta, Yañña-sutta, and Brāhmaṇadhammika-sutta with their Chinese Āgama parallels to examine how this formula functions rhetorically. In the Pāli tradition, the list remains relatively stable despite differing narrative settings and audiences; the Chinese parallels, by contrast, abbreviate, replace, or reorganize its individual terms. Yet lexical variation does not eliminate the underlying argumentative structure: sacrifices involving killing are criticized, while sacrificial efficacy is reassessed rather than simply denied. These divergences are therefore better understood as recontextualizations shaped by differing transmission histories, translation choices, and editorial processes, rather than as mere translational inconsistencies. The analysis argues that early Buddhist criticism did not entail the outright rejection of Brahmanical ritual language or authority. Instead, inherited ritual categories were selectively adapted, and the criteria for legitimate and fruitful sacrifice were reoriented toward non-violence and generosity. Kosala is the primary regional setting for the texts examined; the broader eastern Vedic horizon, including Videha and Magadha, contextualizes the circulation of ritual language between Buddhist and Brahmanical communities. The five-sacrifice formula thus illustrates selective ritual reworking rather than wholesale repudiation.

1. Introduction

Early Buddhist texts engage Brahmanical sacrifice in several distinct registers, including criticism of violent sacrifice and the redefinition of sacrificial language in non-violent contexts. Within this broader ritual discourse, this article examines the relationship between mahāyañña—a designation for a large-scale sacrifice or offering—and a recurrent sequence of five sacrificial terms: assamedha, purisamedha, sammāpāsa, vājapeyya, and niraggaḷa. The list occurs in the Ujjaya-sutta (AN 4.39), Yañña-sutta (SN 3.9), Brāhmaṇadhammika-sutta (Sn 2.7), Mettā-sutta (AN 8.1), and Mettābhāvanā-sutta (It 27). For clarity, this article refers to this recurring sequence as the “five-sacrifice formula.”

The list discussed here belongs to a different ritual category from the later Brahmanical and Hindu pañcamahāyajña.1

Mahāyañña in the Pāli corpus does not necessarily entail the killing of animals or other living beings. The Ujjaya-sutta (AN 4.39) and Yañña-sutta (SN 3.9) condemn sārambha/mahārambha-yañña, sacrifices involving violence, yet non-violent practices are still termed yañña and are explicitly said to yield great fruit (Freiberger 1998, pp. 41–42). The Sela-sutta (Sn 3.7 = MN 92) offers a further example: the food offering made by Keṇiya to the Buddha and the Saṅgha is associated with a mahāyañña (Black 2009, p. 34, n. 15). It is therefore useful to retain “large-scale sacrifice/offering” as the basic sense of mahāyañña while distinguishing between its association with animal sacrifice and its redefinition in non-violent contexts.

The Vedic identity of the five rites raises a separate philological problem. Two analytical levels must be distinguished: first, which traditional Śrauta rites, if any, underlie the five Pāli names; and second, how Buddhist texts deploy those names in particular narrative and rhetorical settings. In his study of the Buddhist reinterpretation of sacrifice, Freiberger treats the five-sacrifice list in the Yañña-sutta as a representative example of Brahmanical sacrifice involving animal killing, while leaving the precise Vedic identification of the individual names to philological studies such as those of Falk and concentrating instead on its Buddhist reinterpretation (Freiberger 1998, p. 41).

Likewise, this study distinguishes these two levels. It first reviews the philological evidence relevant to the Vedic identification of the five ritual names before turning to their function in Buddhist discourse. Three terms—assamedha, purisamedha, and vājapeyya/vācapeyya—correspond relatively straightforwardly to the Sanskrit Aśvamedha, Puruṣamedha, and Vājapeya (Falk 1988, pp. 230–31). The Aśvamedha is a royal horse sacrifice; the Puruṣamedha is a rite in which human beings are constituted as sacrificial victims; and the Vājapeya is closely associated with kingship and ritual prestige and includes Soma drinking and a chariot race. The identities of sammāpāsa and niraggaḷa, however, remain disputed (Falk 1988, pp. 230–31). The fact that three items clearly correspond to Vedic ritual names, while the remaining two have long been debated in relation to traditional Śrauta practice, makes it difficult to treat the entire list as a Buddhist invention. At the same time, the extant Śrauta corpus cannot be assumed to preserve every regional form of ritual that once existed. Disagreement between Pāli ritual terminology and the surviving Śrautasūtras should therefore not be reduced too quickly to Buddhist ignorance or error. Although Falk (1988) and Thite (1971, 1972) provide the philological point of departure, the aim here is not to reconstruct each rite definitively, but to examine how Buddhist texts deploy the list.2

The article has three related aims. First, it distinguishes the relatively stable form of the five-sacrifice formula from the functions it performs in different Pāli contexts. The Ujjaya-sutta places it in an exchange with a Brahmin about the value of sacrifice; the Yañña-sutta uses it to evaluate King Pasenadi’s large-scale animal sacrifice; and the Brāhmaṇadhammika-sutta incorporates it into an account of the dhamma of the ancient Brahmins and its decline, addressed to elderly and wealthy Brahmins of Kosala. The recurrence of the formula must therefore be read in relation to speaker, audience, and narrative setting, rather than reduced to a single, fixed meaning.

Second, it compares these three Pāli texts with their Chinese Āgama parallels in order to trace the transmission and transformation of sacrificial discourse. Rather than treating the Pāli text as the original and the Chinese materials as derivative translations to be judged by their fidelity, it treats them as parallel witnesses. The analysis considers lexical correspondence and variation in the ritual names, the roles of speakers and audiences, the narrative presentation of the performance and results of sacrifice, and patterns of addition, omission, abbreviation, and expansion. This makes it possible to compare not only individual words but also narrative structures and evaluative patterns that persist across distinct transmissions.

Third, the comparison examines the rhetoric of early Buddhist criticism of Brahmanical sacrifice. It asks how ritual language is calibrated to different audiences, including Brahmins and bhikkhus; what regional and social conditions enabled ritual knowledge to circulate; and how inherited terms such as yañña, together with the authority attached to sacrifice, are repositioned when non-violence and giving become criteria of a legitimate and fruitful sacrifice.

Several methodological qualifications follow from this approach. A Nikāya–Āgama comparison should not assume the Pāli version to be the original against which the Chinese is measured. As Anālayo (2011) demonstrates in his comparative study of the Majjhima-Nikāya, elements shared by independently transmitted parallels can provide important evidence for a common core preceding the divergence of the extant traditions, but an element preserved in only one version cannot automatically be classified as a later addition. This article therefore attends not only to recurrent structures and evaluations but also to the ways in which particular witnesses expand, abbreviate, or reorganize them.

This study focuses on the comparison of Pāli and Chinese parallel witnesses at the level of the early textual traditions. Apart from the brief reference to Mp noted below, a systematic examination of the later Pāli commentarial tradition (Aṭṭhakathā) and its reception history lies beyond the scope of the present article and will require separate study.

This comparative approach also engages with a broader debate concerning the status of recurring formulas in the early Buddhist literature. Shulman (2021, 2022) draws attention to formulas as important units of early Buddhist textuality and suggests that their recurrence in different narrative settings may reflect creative and audience-sensitive processes of composition, rather than merely the corruption or misremembering of a fixed original. Allon (1997, 2021, 2022), by contrast, emphasizes the role of repetition, waxing-syllable patterning, and standardized diction in supporting the memorization and communal recitation of texts. His work accordingly cautions against treating formulaic variation as evidence of the free recombination of autonomous units. The present analysis situates the evidence of the five-sacrifice formula in relation to these broader perspectives, without attempting to settle the debate between them. It examines how a relatively stable sequence of ritual names could be maintained while its deployment and, in some cases, its wording varied across narrative contexts and transmission lineages. Such variation may be shaped by rhetorical and interpretive considerations, without necessitating a definitive choice between compositional recombination, transmission-based change, or a combination of both.

The social relationship between Buddhists and Brahmins was more complicated than one of unqualified hostility. Brahmins appear as rivals for authority and patronage, but also as wealthy supporters who could become Buddhist lay followers without necessarily relinquishing their Brahmin social identity, the study of the Vedas, or ritual practice (Black 2009, p. 41; Black 2011, p. 153). The analysis below accordingly treats variation in the five-sacrifice formula as a rhetorical choice shaped by audience and context. The agnihotra–Sāvitrī hierarchy of the Sela-sutta and the ideal mahāyañña of the Kūṭadanta-sutta (DN 5) serve as supplementary cases of the same broader process.

The discussion proceeds in three stages. Section 2 examines the three principal Pāli contexts of the formula. Section 3 compares these with their Chinese parallels. Section 4 then considers the results in terms of audience, the circulation of ritual knowledge in the eastern Vedic milieu, and the Buddhist transformation of traditional ritual grammar.

2. The Five-Sacrifice Formula in Pāli Literature

The five-sacrifice formula occurs not only in the Ujjaya-sutta (AN 4.39), Yañña-sutta (SN 3.9), and Brāhmaṇadhammika-sutta (Sn 2.7), but also in the Mettā-sutta (AN 8.1) and Mettābhāvanā-sutta (It 27). In the latter two texts, the sacrifices of royal seers (rājisayo/rājīsayo) who conquered the world and performed the five rites are said not to equal even one sixteenth of the merit of a well-cultivated mind of loving-kindness (mettā). The list here functions less as an object of direct ritual controversy than as a benchmark against which the merit of mettā is magnified. The analysis that follows, however, concentrates on the three texts in which the list is embedded in distinct forms of sacrificial discourse: royal sacrifice, a direct question about the legitimacy of sacrifice, and a narrative about the ancient Brahmins.

2.1. Royal Sacrifice: SN 3.9

Because the five names recur in the same order across different Pāli texts, the list has been understood as a stereotyped prose formula that originally circulated independently of the verse contexts in which it now appears (Tsuchida 1991, p. 89). The formulaic character of the sequence also helps explain the metrical difficulties it creates in its present verse setting. Because the same stock formula recurs in several sources, removing an item solely to regularize the metre is methodologically problematic (Norman 2001, p. 226). At the same time, the metrical problem has led to the suggestion that one item may have been added in the course of transmission, with niraggaḷa proposed as a possible candidate (Ellis 2021, p. 102). The term “five-sacrifice formula” is retained here as a convenient designation for the sequence preserved in the surviving texts. The Yañña-sutta presents it as follows:

Assamedhaṃ purisamedhaṃ, sammāpāsaṃ vājapeyyaṃ niraggaḷaṃ3; Mahāyaññā mahārambhā, na te honti mahapphalā.

“Assamedha, purisamedha, sammāpāsa, vājapeyya, niraggaḷa—these great and violent sacrifices do not yield great fruit.”

—Yañña-sutta (SN 3.9; PTS SN I 75–76).4

In both discourses, the five rites are designated as mahāyaññā and mahārambhā. The latter expression occurs in contexts involving the slaughter of numerous animals (Tsuchida 1991, p. 89).

In the Yañña-sutta, the formula evaluates a royal sacrifice being prepared for King Pasenadi of Kosala. Five hundred bulls, five hundred young bullocks, five hundred heifers, five hundred goats, and five hundred rams have been brought as sacrificial victims (pasu) and tied to posts.5 The king’s slaves (dāsā), messengers or servants (pessā), and workers (kammakarā) are likewise compelled to labor:

daṇḍatajjitā bhayatajjitā assumukhā rudamānā parikammāni karonti

“Threatened with punishment and fear, with tearful faces, they perform their work while weeping.”

—Yañña-sutta (SN 3.9; PTS SN I 75–76).

When the bhikkhus report this situation, the Buddha responds with the five-sacrifice verse quoted above. The formula is thus embedded in the representation of a royal rite marked by both animal killing and coerced labor.

In traditional Śrauta ritual, sacrificial animals were selected and arranged according to prescribed rules. In the Aśvamedha, for example, the Yajurveda Saṃhitās prescribe in detail the kinds, numbers, and bodily characteristics of animals assigned to particular deities, and the lists themselves vary across recensions (Teshima 2016, pp. 1043–49). The schematic enumeration of five groups of five hundred animals in SN 3.9 is therefore unlikely to be a technical reproduction of an actual Śrauta victim list. It is better read as a literary representation that emphasizes the scale of a royal sacrifice and the violence it entails.

It is difficult to determine how much Buddhist transmitters knew about detailed Śrauta procedure. Buddhist transmitters were, however, clearly aware of the religious authority and social prestige attached to large-scale sacrifice (Tsuchida 1991, p. 89). The Yañña-sutta can thus be read as reflecting some knowledge of sacrificial practice while selectively reframing that practice within a Buddhist evaluative framework.

2.2. The Legitimacy of Sacrifice: AN 4.39

The Ujjaya-sutta, which shares the formula with the Yañña-sutta, places the same verse in a markedly different narrative context: a direct conversation between the Buddha and the Brahmin Ujjaya.

Bhavaṃ Gotamo yaññaṃ vaṇṇeti?

“Does Master Gotama praise sacrifice (yañña)?”

—Ujjaya-sutta (AN 4.39; PTS AN II 42–43).

The point at issue is not whether sacrifice should exist at all, but which kinds of sacrifice are fruitful. The Buddha distinguishes a sārambha-yañña, in which living beings are killed, from a nirārambha-yañña, which involves no killing. He does not praise the former, but praises the latter, citing as examples continual giving (niccadāna) and customary sacrifice (anukulayañña). The distinction is immediately followed by the five-sacrifice formula:

Assamedhaṃ purisamedhaṃ, sammāpāsaṃ vājapeyyaṃ niraggaḷaṃ; Mahāyaññā mahārambhā, na te honti mahapphalā.

“Assamedha, purisamedha, sammāpāsa, vājapeyya, niraggaḷa—these great and violent sacrifices do not yield great fruit.”

—Ujjaya-sutta (AN 4.39; PTS AN II 42–43).

The five names occur in the same sequence as in the Yañña-sutta, but they perform different rhetorical functions. In SN 3.9, the formula evaluates a large-scale sacrifice actually being prepared for Pasenadi. In AN 4.39, it answers a Brahmin’s question about what kinds of sacrifice deserve praise, and thus serves as a negative example within an argument about ritual efficacy. Non-violent sacrifice is evaluated in the opposite terms:

Etaṃ yajetha medhāvī, eso yañño mahapphalo.

“The wise should perform this sacrifice; this sacrifice bears great fruit.”

—Ujjaya-sutta (AN 4.39; PTS AN II 42–43).

The remainder of the verse links the sacrificer (yajamāna), the sacrifice (yañña), and the deities (devatā) who rejoice in it:

Etañhi yajamānassa, seyyo hoti na pāpiyo; Yañño ca vipulo hoti, pasīdanti ca devatāti.

“For the sacrificer this is better, not worse; the sacrifice flourishes, and the deities are pleased.”

—Ujjaya-sutta (AN 4.39; PTS AN II 42–43) and Yañña-sutta (SN 3.9; PTS SN I 75–76).

The relation among sacrificer, sacrifice, and deity is central to the reciprocal logic of Vedic ritual. Vedic texts express this reciprocity in formulations such as dehi me dadāmi te, “give to me, and I will give to you” (TS 1.8.4.1; VS 3.50). As Jamison and Witzel (2003, p. 94) observe, offerings and hymns presented in ritual both acknowledge benefits received from the gods and seek their continued support, thereby mediating an ongoing exchange between humans and gods. The Ujjaya-sutta does not reject this reciprocal ritual relationship outright. Instead, it relocates efficacy within it. A non-violent sacrifice benefits the sacrificer and pleases the gods; the issue is therefore not whether ritual can be effective, but what kind of ritual deserves to be called fruitful.

2.3. The Dhamma of the Ancient Brahmins: Sn 2.7

The Brāhmaṇadhammika-sutta places the five-sacrifice formula in a third narrative context. The Buddha is approached by elderly and wealthy Brahmins of Kosala (Kosalakā brāhmaṇamahāsālā). The text repeatedly characterizes them as old (jiṇṇā), aged (vuḍḍhā), elderly (mahallakā), advanced in years (addhagatā), and at the end of life (vayoanuppattā). They ask whether the dhamma of the ancient Brahmins is still found among Brahmins of the present day. When the Buddha answers that it is not, they ask him to explain the dhamma of the ancient Brahmins.

In the Buddha’s account, the ancient Brahmins are characterized by celibacy (brahmacariya), recitation (sajjhāya), and non-harming (avihiṃsā). They acquire the materials of sacrifice in accordance with dhamma and refrain from killing cows:

Dhammena samodhānetvā, tato yaññam akappayuṃ. Upaṭṭhitasmiṃ yaññasmiṃ, nāssu gāvo haniṃsu te.

“Having gathered [the materials] in accordance with dhamma, they prepared the sacrifice; and when the sacrifice was ready, they did not kill a cow.”

—Brāhmaṇadhammika-sutta (Sn 2.7, vv. 295–96).

These verses present the sacrifice of the ancient Brahmins as governed by dhamma and depict the Brahmins as refraining from killing cows, which are later portrayed as beneficent friends who provide food, drink, and sustenance. This ideal provides the baseline for the subsequent narrative of ritual decline.

The narrative shifts when the Brahmins encounter royal luxury and wealth. The Brāhmaṇadhammika-sutta links this change to the royal sacrifices of King Okkāka and presents Brahmanical ritual as transformed through its association with kingship and wealth. The Pāli name Okkāka corresponds to Sanskrit Ikṣvāku.6 Captivated by courtly luxury and wealth, the Brahmins experience a reversal (vipallāsa) and urge King Okkāka to sacrifice:

Pahūtadhanadhaññosi, yajassu bahu te vittaṃ; yajassu bahu te dhanaṃ.

“You have abundant wealth and grain; sacrifice, for you have great riches; sacrifice, for you have great wealth.”

—Brāhmaṇadhammika-sutta (Sn 2.7, v. 302).

The five-sacrifice formula appears in the account of how, at the Brahmins’ request, King Okkāka performed the five rites:

Tato ca rājā saññatto brāhmaṇehi rathesabho assamedhaṃ purisamedhaṃ sammāpāsaṃ vācapeyyaṃ niraggaḷaṃ, ete yāge yajitvāna brāhmaṇānaṃ adā dhanaṃ.

“Then the king, the foremost of chariot-warriors, persuaded by the Brahmins, performed the assamedha, purisamedha, sammāpāsa, vācapeyya, and niraggaḷa; having performed these sacrifices, he gave wealth to the Brahmins.”

—Brāhmaṇadhammika-sutta (Sn 2.7, v. 303).

Here the formula does not operate as an external evaluation of sacrifice. It is inserted into the narrated past as a list of royal rites performed at the Brahmins’ instigation. Its position in the story is crucial. The ancient Brahmins had prepared sacrifice according to dhamma without killing cattle (especially cows); after becoming captivated by royal wealth, their successors persuaded the king to perform the five sacrifices and received wealth from him. The formula thus marks the transition from the non-violent sacrifice of an idealized past to the later slaughter of cows, linking sacrificial practice to royal power and material acquisition.

Yet the wealth acquired through these sacrifices does not satisfy the Brahmins. Their craving (taṇhā) increases, and they eventually induce King Okkāka to have cows slaughtered in sacrifice. As the knife falls upon a cow, the following gāthā declares:

Adhammo iti pakkanduṃ, yaṃ satthaṃ nipatī gave.

“‘This is adhamma,’ they cried, when the knife fell upon the cow.”

—Brāhmaṇadhammika-sutta (Sn 2.7, v. 310).

3. The Five-Sacrifice Discourse in the Chinese Āgama Parallels

The comparisons in this section treat the Pāli and Chinese materials as parallel textual witnesses rather than as a simple original and its translation. The analysis focuses on three dimensions: the narrative setting of each discourse, the selection and arrangement of ritual terms, and the evaluative function performed by the five-sacrifice formula. The aim is therefore not to reconstruct a single Indic source text, but to identify how related traditions preserve, abbreviate, expand, or reframe a shared ritual vocabulary.

The differences examined below cannot, however, be assigned with confidence to a single stage of transmission. Some may reflect divergent Indic recensions that predate Chinese translation, whereas others may have arisen through translation choices, editorial reorganization, or the adaptation of ritual vocabulary within Chinese Buddhist literary settings. In most cases, the available evidence does not permit these possibilities to be separated conclusively. The comparison therefore identifies the textual and rhetorical effects of variation without presupposing a single mechanism for its production. At the same time, recurrent expansions of evaluative or affective language in the Chinese witnesses—for example, the emphasis on unwillingness and suffering in ASA 61 and SA 89—make translation and redaction plausible factors to consider alongside differences in transmission lineage.

Identifications of particular parallels follow previous studies of parallel discourses. In particular, the identification of SA 89 and ASA 89 as parallels to AN 4.39, the Ujjaya-sutta, follows Choong (2009, p. 373).7 The following sections compare the transmission and transformation of the five-sacrifice formula and the broader discourse on sacrifice across these parallel traditions. The comparative corpus is summarized in Table 1.

Table 1. Comparative corpus.

Comparative Group	Chinese Parallel	Pāli Text	Section
Royal sacrifice	ASA 61 (T02, no. 100)	SN 3.9 Yañña-sutta	Section 3.1
Legitimacy of sacrifice	SA 89 (T02, no. 99); ASA 89 (T02, no. 100)	AN 4.39 Ujjaya-sutta	Section 3.2
Dhamma of the ancient Brahmins	MA 156 (T01, no. 26)	Sn 2.7 Brāhmaṇadhammika-sutta	Section 3.3

3.1. Royal Sacrifice: SN 3.9 and ASA 61

The Chinese parallel to the Yañña-sutta, ASA 61 (T02, no. 100), does not enumerate the five Pāli ritual names. Instead, it describes the sacrifice being prepared by King Pasenadi as a ⼤祀 (“great sacrifice”) and expands the description of the animals tied at the sacrificial ground:

波斯匿王欲設⼤祀，飬千⽜王，皆繫於柱，幷及犛⽜、⽔⽜、乳⽜犢，及⼩⽜，盡各數千；䍧羖⽺等，亦復數千；種種畜⽣，皆繫祀場。

“King Pasenadi wished to institute a great sacrifice (⼤祀). He raised one thousand leading bulls, all of them tied to posts; there were also black cattle, water buffaloes, milk cows and calves, and young cattle, each numbering in the thousands. Ewes, goats, and other animals likewise numbered in the thousands. All kinds of animals were bound at the sacrificial ground.”

—《別譯雜阿含經》卷3, T02, no. 100, pp. 394c27–395a01.

The two traditions also correspond closely in their descriptions of the laborers compelled to work at the sacrifice. The Pāli text speaks of the king’s slaves (dāsā), servants (pessā), and workers (kammakarā) as “threatened with punishment, threatened with fear, with tearful faces, weeping as they work” (daṇḍatajjitā bhayatajjitā assumukhā rudamānā parikammāni karonti). The Chinese parallel reads:

僕使作⼈，鞭笞恐怛，悲泣號呼，不喜不樂，衆苦作役。

“Slaves, servants, and laborers were whipped and beaten, terrified, crying and wailing, neither glad nor joyful, and made to labor amid manifold suffering.”

—T02, no. 100, p. 395a02–05.

The two accounts align closely in their description of coerced labor. Pāli dāsā, pessā, and kammakarā correspond to the Chinese 僕, 使, and 作人, while daṇḍatajjitā, bhayatajjitā, and assumukhā rudamānā closely match 鞭笞, 恐怛, and 悲泣號呼. Yet the Chinese version adds 不喜不樂, “neither glad nor joyful,” for which the Pāli has no direct equivalent, and 衆苦, “manifold suffering,” makes the affective dimension of the labor described by parikammāni karonti more explicit. The Chinese parallel preserves the basic structure and much of the vocabulary of the Pāli account, but it foregrounds the coercive and painful nature of the labor more explicitly.

The scale of the animal sacrifice is likewise enlarged. The Pāli version lists five groups of animals, each numbering five hundred; the Chinese version subdivides the animals into more categories and raises the quantities to one thousand or several thousand. The larger narrative frame remains the same: bhikkhus learn of the king’s sacrifice, report it to the Buddha, and receive an evaluative response.

What changes most markedly is the verse that follows. The five-sacrifice formula does not appear in its Pāli form. Instead, the Chinese text contrasts the merit of repeated sacrifice directly with Buddhist faith and practice:

⽉⽉百千祀 修此以求福 不如⼀信佛 ⼗六分中⼀

“Even if one performed hundreds of thousands of sacrifices month after month in search of merit, it would not amount to one sixteenth of a single act of faith in the Buddha.”

—T02, no. 100, p. 395a12–13.

In the following verses, the same pattern recurs, linking “faith” (信) in turn to the Buddha (佛), the Dharma (法), and the Saṅgha (僧), before turning to a mind of loving-kindness (慈⼼). Each is declared superior in merit to innumerable sacrifices. In the Chinese parallel, the evaluation expressed by the Pāli formula is recast as an explicit comparison between the merit of repeated sacrifice and that of Buddhist faith and practice.

3.2. The Legitimacy of Sacrifice: AN 4.39, SA 89, and ASA 89

Two Chinese texts parallel the sacrificial dialogue of AN 4.39, the Ujjaya-sutta: SA 89 and ASA 89. Both retain the basic dialogic structure in which a Brahmin asks the Buddha about sacrifice and the Buddha does not reject every sacrifice indiscriminately, distinguishing instead between rites that involve killing and those that do not. The parallels diverge, however, in the ritual terminology used to represent the Pāli five-sacrifice formula and in the vocabulary through which a legitimate sacrifice is described.

In SA 89 (T02, no. 99), the interlocutor corresponding to Ujjaya in AN 4.39 is transcribed as 優波迦 and explicitly identified as a “young Brahmin” (年少婆羅⾨). He asks whether the Buddha likewise praises what the text calls an “improper grand sacrificial assembly” (邪盛⼤會) performed by Brahmins. The Buddha responds by distinguishing gatherings that can be praised from those that cannot (《雜阿含經》卷4, T02, no. 99, pp. 22c04–23a04). As in ASA 61, the Chinese description of sacrificial labor incorporates the phrase 不喜不樂 (“neither glad nor joyful”) alongside 鞭笞恐怛 and 悲泣號呼, thereby rendering the elements of coercion and unwillingness more explicit than in the Pāli parallel.

The five-sacrifice formula is abbreviated rather than reproduced. Only ⾺祀, “horse sacrifice,” corresponding to the first Pāli item, is named; the remaining rites are covered by 等, “and the like”:

⾺祀等⼤會 造諸⼤難事 如是等邪盛 ⼤仙不稱歎

“Great assemblies such as the horse sacrifice create great affliction; such improper sacrificial gatherings are not praised by the great sage.”

—《雜阿含經》卷4, T02, no. 99, p. 22c20–21.

The text contrasts sacrifices involving killing—termed 邪盛大會—with non-violent rites, called 正會 and praised by the Great Sage (大仙). Such non-violent rites yield 大果, “great fruit,” and 諸天皆歡喜, “all the gods rejoice.” In SA 90, this legitimate sacrificial mode is further associated with 惠施 (giving), 供養 (offering), 福田 (field of merit), and 羅漢田 (field of arhats). Pāli nirārambha-yañña and niccadāna are therefore represented in the Chinese tradition through a more explicit and practical vocabulary of giving, offering, and fields of merit.

ASA 89 (T02, no. 100) likewise presents the interlocutor as 優北伽, an alternative transcription of Ujjaya, and identifies him as a māṇava (摩納), again indicating a young Brahmin. The text frames Ujjaya’s question around Brahmins’ acquisition of wealth and the legitimacy of a great sacrifice (大祀) (《別譯雜阿含經》卷5, T02, no. 100, pp. 404a27–404b02). In place of the five Pāli ritual names, however, it gives a distinct list of six items:

⾺脂及⼈脂 ⽜脂幷美⾷ 吸⾵開祀⾨ 此六名⼤祀 作業雖廣⼤ 仙聖所毀呰

“Horse fat, human fat, cattle fat, fine food, inhaling the wind, and opening the sacrificial gate—these six are called great sacrifices. Though the undertaking is vast, sages and holy men condemn them.”

—《別譯雜阿含經》卷5, T02, no. 100, p. 404b03–05.

The Indic source terms underlying these individual expressions cannot be identified with confidence. Nevertheless, some elements clearly invite comparison with the Pāli list. ⾺脂 and ⼈脂8, involving horse and human victims, recall assamedha and purisamedha, whereas ⽜脂 is better understood as an additional sacrificial element involving a cow than as a direct equivalent of a single Pāli item. The final three expressions—美⾷, 吸⾵, and 開祀⾨—cannot simply be mapped in order onto sammāpāsa, vājapeyya, and niraggaḷa. The last, however, has a more specific connection with niraggaḷa. Sanskrit nirargala literally means “without a bar,” hence “unbarred” or “unrestricted,” and other Chinese Buddhist texts render comparable expressions through the imagery of 無遮會, an “unrestricted assembly,” and 開⾨, “opening the gate.” The Foben xingji jing (《佛本⾏集經》) contains 汝昔施⼀無遮會, “you once gave a single unrestricted assembly,” and 我昔祭祀無遮會, “I once performed sacrifice at an unrestricted assembly” (T03, no. 190, p. 791a), while the Foben xing jing (《佛本⾏經》) has 吾祀⼤開⾨, “I performed a great sacrifice with the gate opened,” and 於此地開⾨⼤祀, “at this place I opened the gate for a great sacrifice” (T04, no. 193, p. 78a; Teshima 2022, p. 74). 開祀⾨ is therefore more plausibly read as a ritual designation that renders the imagery of an “unbarred” rite associated with niraggaḷa/nirargala than as a merely descriptive statement that a sacrificial gate was opened.

The evaluative structure, however, remains remarkably stable. ASA 89 distinguishes 邪祀, “improper sacrifice” involving killing, from 正祀, “proper sacrifice” that harms no living beings. Thus, even where the ritual names and the composition of the list vary considerably, the dialogic structure is preserved: a Brahmin asks about sacrifice, and its legitimacy and efficacy are judged according to whether it involves killing. The parallel traditions thus combine considerable lexical variation with a stable underlying argumentative structure.

3.3. The Dhamma of the Ancient Brahmins: Sn 2.7 and MA 156

The Chinese parallel to the Brāhmaṇadhammika-sutta, MA 156, begins with Brahmins asking whether any contemporary Brahmins still study the dhamma of the ancient Brahmins. Whereas the Pāli version specifies its audience as elderly and wealthy Brahmins of Kosala (Kosalakā brāhmaṇamahāsālā), the Chinese version places less emphasis on their age and social status and more directly foregrounds the relation between contemporary Brahmins and the ancient dhamma. The Buddha replies that no Brahmins now study that dhamma and that they departed from it long ago (《中阿含經》卷39, T01, no. 26, pp. 678a29–678b02).

The ancient Brahmins are then said to acquire what is needed “in accordance with dhamma” (如法), to establish ritual observances and engage in giving (⽴齋⾏布施), while not killing a cow (不有殺⽜):

乞求皆如法 ⽴齋⾏布施 … ⽴齋⾏施時 彼不有殺⽜

“They sought what they needed in accordance with dhamma, established the observance, and practiced giving; […] even so, they did not kill a cow.”

—《中阿含經》卷39, T01, no. 26, pp. 678b29–c02.

This closely parallels the Pāli depiction of ancient Brahmins preparing sacrifice according to dhamma without killing cows. The Chinese terminology is nevertheless distinctive: it renders sacrifice as 齋, an observance, and explicitly associates it with 布施 (giving).

The subsequent transformation is narrated in psychological terms. Royal splendor and wealth give rise to a repeatedly increasing attachment (愛). MA 156 describes the process as beginning with the “subtle human desires” (⼈間微妙欲) and progressing toward “repeatedly increasing attachment” (數數增⾧愛) (《中阿含經》卷39, T01, no. 26, p. 678c14, c24). The Pāli compresses the corresponding moment into vipallāsa, a reversal or distortion that overtakes the Brahmins when they are captivated by royal wealth. The Chinese version instead narrates a gradual intensification from desire to attachment.

At the point of royal sacrifice, the Chinese text again departs from the Pāli five-item list:

象齋及⾺齋 ⾺齋不障⾨

“Elephant sacrifice and horse sacrifice; a horse sacrifice with no barred gate.”

—T01, no. 26, p. 678c21.

象齋, an elephant sacrifice, has no counterpart in the Pāli five, while ⾺齋, a horse sacrifice, corresponds to assamedha. The phrase ⾺齋不障⾨ also raises the possibility that ⾺齋 and 不障⾨ should not be segmented as two independent ritual names. Kosambi suggested that whereas an early horse sacrifice may have involved the direct offering of the horse, a later form allowed the horse to wander freely for a year before the rite. On this interpretation, niraggaḷa could function not as an independent ritual name but as an adjective modifying assamedha (Kosambi 1951, p. 54, cited in Thite 1972, p. 195). ⾺齋不障⾨ could accordingly mean an “unobstructed horse sacrifice,” a rite involving a horse allowed to roam without hindrance. MA 156 may thus preserve assamedha and niraggaḷa not as two separate entries but as elements combined within a single description of horse sacrifice.

This differs from ASA 89, where 開祀⾨ may preserve niraggaḷa in a form closer to an independent ritual designation. The contrast suggests that the Pāli item could be interpreted differently in Chinese transmission: as an independent sacrificial term in one context, and as a qualification of horse sacrifice in another. The divergence concerns not merely translation equivalents but the very segmentation of the list and the perceived relationship among its ritual names.

Read in isolation, such divergence might be mistaken for a translation error or an imprecise rendering of an Indic term. Shulman’s (2021) literary approach to parallel witnesses provides a firmer basis for the alternative reading adopted here. Rather than measuring each parallel against a hypothetical original from which it has “erred,” this approach allows each version to be read as a legitimate articulation shaped by the narrative logic and formulaic resources available to its respective transmitting community. On this reading, ASA 89’s 開祀門 and MA 156’s 馬齋不障門 are not competing translations of a single Indic term to be ranked by fidelity. Rather, they offer two distinct ways of segmenting and interpreting a shared ritual designation, each internally consistent with the surrounding formula in its respective recension.

MA 156 places particular emphasis on the relationship between human beings and cows before describing how this relationship is transformed as sacrificial practices expand. Cows, initially portrayed as beneficent beings and friends of human beings, become sacrificial victims slaughtered in countless numbers as wealth and attachment increase. The Chinese account depicts the physical act of killing with unusual vividness: the sacrificer seizes a cow by the horns (捉⽜⾓) and kills it with a sharp knife (持利⼑殺⽜).

往⾄捉⽜⾓ 持利⼑殺⽜

“They went forward, seized the horns of the cow, and killed it with a sharp knife.”

—T01, no. 26, p. 679a05.

At the moment of slaughter, the act is explicitly condemned as 非法 (adharma):

彼喚呼⾮法 以⼑刺⽜時

“They cried out, ‘This is not in accordance with dhamma,’ when the knife pierced the cow.”

—T01, no. 26, p. 679a07.

This closely corresponds to the Pāli declaration of adhamma when the knife falls upon the cow. The Chinese version, however, first depicts the act itself—grasping the cow by the horns and cutting it—before condemning it as ⾮法.9

The two traditions do not end with the condemnation of slaughter. After hearing the discourse, the Brahmin audience accepts the Buddha’s teaching. In the Pāli version, the Brahmins take refuge in the Buddha, Dhamma, and Saṅgha and ask to be accepted as lay followers (upāsaka). MA 156 concludes with a corresponding declaration:

我今⾃歸於佛、法及⽐丘眾，唯願世尊受我為優婆塞，從今⽇始，終⾝⾃歸，乃⾄命盡。

“We now take refuge in the Buddha, the Dharma, and the community of bhikkhus. We ask the Blessed One to accept us as lay followers. From this day onward, for the rest of our lives and until life ends, we take refuge.”

—T01, no. 26, p. 679a19–23.

The narrative of the dhamma of the ancient Brahmins and its transformation therefore does not end as a critique of the past. It culminates in a present Brahmin audience asking to be accepted as Buddhist lay followers. The social implications of this concluding movement, particularly the relation between Brahmin identity and Buddhist affiliation, will be reconsidered in Section 4.3.

4. Recontextualizing Brahmanical Ritual

4.1. Audience and the Selective Deployment of Ritual Discourse

Dialogues and debates involving two or more speakers are prominent literary features in the Nikāyas. These texts are indispensable evidence for the social organization, beliefs, and interactions of early Indian Buddhist communities, but they are not verbatim records of historical conversations. They passed through extended processes of oral transmission, editing, and formalization (Gombrich [1996] 2010, p. 29; Nichols 2015, p. 173). At the same time, the choice of interlocutors, questions, concepts, and points of controversy can still illuminate how Buddhist communities perceived and represented their relationships with rival groups (Nichols 2015, p. 173). In this sense, the suttas are best approached not as homogeneous declarations from a single timeless source, but as products of debate and negotiation among communities over an extended period (Walser 2018, p. 107).

From this premise, the central question is how the same five-sacrifice formula functions in texts addressed to different audiences and embedded in different narrative settings. As the preceding sections have shown, its function is not fixed by the list itself. In the Yañña-sutta, it evaluates the violence and coercion of a royal sacrifice. In the Ujjaya-sutta, it appears in response to a Brahmin’s question and helps redefine ritual legitimacy and efficacy. In the Brāhmaṇadhammika-sutta, it is embedded in a reconstruction of the authoritative Brahmanical past. The recurrence of the formula should therefore not be treated as the mechanical repetition of a single anti-sacrificial message. Its ritual vocabulary and evaluative force are selectively deployed according to audience and discursive situation.

This pattern resonates with Shulman’s (2022, pp. 558–59) theory of the “play of formulas,” which treats recurring sequences less as inert mnemonic devices than as expressive units whose combination with distinct narrative frames produces legitimate, audience-sensitive articulations of Buddhist teaching. From this perspective, the recurrence of the five-sacrifice formula across the Ujjaya-sutta, Yañña-sutta, and Brāhmaṇadhammika-sutta need not be traced to a single originating occasion; rather, each occurrence may reflect the formula’s adaptation to a particular narrative design. Allon (2021, pp. 105–8; 2022), however, offers an important caution against treating the formula as an autonomous compositional unit. He argues that a sequence is recognized as a formula through repeated use within otherwise fixed texts, and that its deployment on a given occasion is shaped by plot, audience, and argumentative purpose rather than by the formula in isolation. The pattern documented here is broadly consistent with Allon’s emphasis on the stability of formulaic sequences. Across all three Pāli settings, the five-sacrifice formula retains a stable core—the same five ritual names, in the same order, evaluated as mahārambha—even as its surrounding narrative and argumentative frame changes. This stability across settings, together with its sensitivity to audience and context, is further illuminated by the Pāli–Chinese comparison in Section 3.

This variation in rhetorical deployment can be obscured if the Buddha’s role as the principal speaker in all three texts is taken to imply a single, consistent position. Walser criticizes a common reconstruction of the historical Buddha as a coherent teacher who engages Brahmin opponents by gradually assigning specifically Buddhist meanings to shared concepts (Walser 2018, p. 105).

The Brahmin audiences represented in early Buddhist texts should likewise not be reduced to a single social type, particularly that of specialist officiants at large-scale sacrifices. The Nikāya suttas portray Brahmins as occupying different social positions and engaging in distinct forms of religious activity. Among them, mahāsāla and jaṭila designate distinct social configurations. Mahāsāla is not an occupational term for a priest who specializes in large-scale ritual; it primarily marks wealth and high social standing, and brāhmaṇa-mahāsāla overlaps substantially with a class of prosperous householders who also possessed cultural authority through the study of the Vedas and adherence to Vedic tradition (Tsuchida 1991, pp. 53, 60–61, 65). By contrast, the performance of elaborate Śrauta rites and possession of specialized ritual knowledge are more closely connected with forest-dwelling jaṭila Brahmins such as Bāvari and the Kassapa brothers, some of whom combined ritual expertise with the distinctive religious prestige of asceticism (Tsuchida 1991, pp. 67, 84, 88–90). It is therefore useful to distinguish specialist ritualists, Brahmin elites who shared the cultural authority and vocabulary of ritual, and royal patrons who financed large sacrifices.

Comparative work on the Droṇa traditions likewise shows that the representation of a Brahmin interlocutor can vary across parallel witnesses without reducing comparison to a search for a single recoverable wording. Anālayo (2017, pp. 15–41) traces such variation across Pāli, Chinese, and other versions, while Feinberg (2026, p. 273) emphasizes the philosophical and literary productivity of reading these divergences together; the present analysis of the five-sacrifice formula adopts a similar orientation toward its own Pāli–Chinese parallels.

The special position of jaṭilas is also visible in their relation to the Buddhist Saṅgha. Vinaya regulations allowing fire-worshipping jaṭilas to receive upasampadā ordination without the usual probationary period (parivāsa) suggest that a degree of affinity was recognized between practitioners of agnihotra and Buddhist renouncers (Rhys Davids and Oldenberg 1899, pp. 190–91, cited in Walser 2018, p. 112).

This social differentiation helps clarify the ritual language addressed to Brahmins in the Ujjaya-sutta and Brāhmaṇadhammika-sutta. Neither discourse is best understood as a technical explanation of Śrauta procedure delivered to specialist officiants. Rather, each draws on ritual values and forms of authority already meaningful to Brahmin interlocutors. In the Ujjaya-sutta, yañña is retained, but its legitimacy and efficacy are redefined in terms of non-killing and ritual fruit. In the Brāhmaṇadhammika-sutta, the dhamma of the ancient Brahmins and their non-violent rites become a normative standard for judging present conduct. The texts do not presuppose that their audiences possess the same level of specialist ritual knowledge. Rather, they require only that yañña, ritual fruit, and the authority of the dhamma of the ancient Brahmins serve as shared points of reference within the exchange.

The three Pāli texts, however, involve different audiences. The Ujjaya-sutta and Brāhmaṇadhammika-sutta present Brahmins as the Buddha’s direct interlocutors; in the Yañña-sutta, by contrast, he addresses bhikkhus who report a royal sacrifice and then evaluates it. The same formula could thus serve both as a vehicle for polemical engagement with Brahmins and as a means of internal instruction and self-definition within the Saṅgha.

The Yañña-sutta in particular should first be situated within this internal pedagogical context. Stories of the Buddha responding to Brahmanical practices may have offered monks exemplary ways of understanding and responding to issues encountered in a Brahmin-dominated social environment (Manné 1990, p. 71). Yet their effect need not have been confined to monastic audiences. For lay Buddhists living amid Brahmanical authority, narratives in which prominent Brahmins and Buddhist teachings are brought into relation could also legitimize Buddhist affiliation and provide reassurance about its place in the surrounding social order (Black 2009, pp. 41–42).

The dialogues with Brahmin interlocutors likewise cannot be treated as direct records of historical conversations. Nevertheless, their literary construction can reveal which questions and concepts Buddhist transmitters considered appropriate to a Brahmin audience and which values they chose to emphasize (Nichols 2015, pp. 173–74). Even where it is impossible to establish whether a particular Brahmanical practice has been represented accurately, misunderstood, or deliberately reshaped, the polemical techniques used in the representation remain open to analysis (Freiberger 2009, pp. 62–63).10

These differences in audience help account for the different functions of the same formula across the suttas. They also raise the historical question addressed in the next section: under what regional conditions could such ritual language have circulated between Brahmanical and Buddhist communities?

4.2. The Eastern Vedic Milieu and the Circulation of Ritual Knowledge

The regional discussion that follows does not seek to establish direct textual dependence or to assign individual Buddhist ritual terms to a particular Vedic school. Rather, it delineates the eastern Vedic milieu in which Buddhist communities could encounter ritual terminology, sacrificial prestige, and ideas of kingship in historically variable forms.

The prominence of Kosala in the principal texts containing the five-sacrifice formula calls for closer attention to their regional context. Kosala appears in early Buddhist literature as one of the sixteen mahājanapadas, a major political power and an important center of Buddhist activity. The Pabbajjā-sutta likewise presents the renunciant Gotama, in his exchange with King Bimbisāra, as locating the Śākyas in a prosperous country “dwelling among the Kosalans” (Kosalesu niketino) and identifying himself as a Śākya of the Ādicca lineage (Sn 3.1). Kosala lay east of the Kuru–Pañcāla Vedic heartland and was associated in particular with the Kāṇva branch (śākhā) of the Vājasaneyin tradition. Witzel describes the eastward expansion of Vedic culture not simply as a movement of populations but as a “ritual occupation” of Kosala(-Videha) by representatives of orthodox Kuru culture (Witzel 1995, pp. 11, 17, 22–23). The Videgha Māthava narrative in the Śatapatha Brāhmaṇa, together with accounts of eastern kings receiving Brahmins, similarly points to the eastward spread of western Vedic ritual knowledge and authority.11

A more concrete indication of contact with regional Vedic practice appears in the Sundarikabhāradvāja-sutta, where a Brahmin, after performing the agnihotra, seeks another Brahmin to consume the sacrificial remainder. Bausch compares this detail with agnihotra prescriptions of the Vājasaneyin tradition (Bausch 2015, pp. 57–58). This correspondence does not establish direct dependence, but it shows why early Buddhist ritual language should be considered in relation to regional Vedic traditions rather than against an abstract pan-Vedic background alone.

This eastward spread was supported by political and social institutions. Coronation rites such as the royal abhiṣeka show how Brahmanical ritual expertise could function as a resource for legitimating kingship, while royal patronage, including grants of land to Brahmins (brahmadeya), provided material conditions for Brahmin communities to settle and transmit their traditions. The distribution of named Brahmin villages in the Theravāda Tipiṭaka offers indirect support for this picture: nine of fourteen identifiable Brahmin villages are located in Kosala, compared with four in Magadha and one in Malla (von Hinüber 2006, pp. 199–201; Tsuchida 1991, p. 57). Kosala therefore offered favorable conditions for sustained contact between Buddhist and Brahmanical communities and for the circulation of culturally significant ritual knowledge.

Within this setting, early Buddhist ritual vocabulary should therefore be considered not only against an abstract pan-Vedic background but also against Vedic traditions that could have been encountered in Kosala and regions further east (Bausch 2018, pp. 28–31). This regional connection does not, however, establish that the ritual vocabulary of any particular sutta derives directly from the Kāṇva tradition, nor does it imply that early Buddhist doctrine derived from an eastern Vedic tradition. It does suggest, however, that early Buddhist texts may preserve traces of engagement with ritual and conceptual materials that were locally available in Kosala and the wider eastern region.

The case of sammāpāsa illustrates the point. Falk observes that the terminology and form of the śamyā-throwing rite vary by Vedic school and region, and he connects Pāli sammāpāsa with eastern ritual usage in Magadha and Videha (Falk 1988, pp. 239–40). More importantly, Falk cautions against explaining such discrepancies simply as deficient Buddhist knowledge. The surviving Śrauta texts do not constitute a complete record of every ritual form that once existed, while Pāli and Sanskrit epic sources may preserve evidence of regional and historical variability not otherwise visible in the extant ritual corpus (Falk 1988, pp. 240, 246).12

The uneven preservation of the eastern Vājasaneyin tradition makes this caution especially important. The Śatapatha Brāhmaṇa survives in Mādhyandina and Kāṇva recensions: the former is closely associated with Videha, whereas the compilation and redaction of the latter have been localized in Kosala (Brereton 2006, pp. 323–24; Bausch 2018, p. 31). Although the Mādhyandina tradition is accompanied by the Kātyāyanaśrautasūtra (KŚS), no independent Śrautasūtra survives for the Kosalan Kāṇva tradition (Bausch 2015, pp. 49–50). The absence of exact Śrauta counterparts to sammāpāsa or niraggaḷa therefore cannot by itself show that Buddhist authors invented or misunderstood these terms.

Repeated contact with Brahmanical groups need not have given Buddhist communities specialist knowledge of Śrauta ritual, but it could have familiarized them with ritual names, the prestige and authority of sacrifice, and its association with kingship. The five-sacrifice formula should therefore be considered in light of this regional circulation of ritual knowledge and the uneven survival of Vedic ritual texts, rather than judged solely by whether each of its terms has an exact counterpart in an extant Śrautasūtra. The question, then, is how Buddhist texts drew on and reshaped the ritual language and authority available to them.

4.3. The Buddhist Transformation of Traditional Ritual Grammar

Early Buddhist sources are widely understood to criticize Brahmanical sacrifices involving violence and socially harmful practices (Freiberger 1998, pp. 39–40). Such criticism, however, does not amount to an unqualified repudiation of Brahmanical ritual. Rather, early Buddhist texts retain ritual terminology and evaluative hierarchies while redefining what counts as legitimate and efficacious sacrifice. This process can be situated within a broader pattern in which Buddhism defined itself in relation to a preexisting Brahmanical hierarchy (Bailey and Mabbett 2003, p. 121). The texts examined here show that this relation involved not only opposition but also the selective retention and reassignment of ritual categories.

This retention and reassignment are clearest in the five-sacrifice discourse itself, where familiar elements of Brahmanical sacrifice are kept in place but evaluated according to different criteria. In the Ujjaya-sutta and Yañña-sutta, the category yañña is retained, as are the sacrificer (yajamāna), ritual fruit (phala), and the pleasure of the deities (devatā). Large sacrifices involving animal killing are said not to yield great fruit, whereas non-violent sacrifice and continual giving (niccadāna) are said to do so. Freiberger describes this as a reinterpretation that preserves the “conventional frame” of traditional sacrifice while eliminating or replacing problematic elements within it (Freiberger 1998, pp. 41–42). The Chinese terms 正會 (“proper assembly”), 正祀 (“proper sacrifice”), 惠施 (“giving”), 供養 (“offering”), 福田 (“field of merit”), 羅漢田 (“field of arhats”), and 羅漢會 (“assembly of arhats”) make this recalibration still more explicit, recasting ritual efficacy in an explicitly Buddhist vocabulary of non-violent sacrifice, giving, offering, and fields of merit.

The same strategy extends to other Buddhist reworkings of Brahmanical ritual. The Sela-sutta and related verses acknowledge the agnihotra as the foremost sacrifice and the Sāvitrī as the foremost metre, then extend this comparative structure to the Buddha as supreme among human and divine beings. The description of the Sāvitrī as consisting of three lines and twenty-four syllables reflects familiarity with the technical characterization of the Gāyatrī metre (Shults 2014, pp. 114–25). The choice of agnihotra is especially significant: as a daily fire offering performed by the sacrificer himself, it represents a familiar Brahmanical norm that these verses retain and adapt to affirm the Buddha’s supremacy. The Kūṭadanta-sutta follows the same logic with its ideal mahāyañña in which no animals are killed, no trees are felled, no grass is cut, workers are neither coerced nor abused, and wealth is redistributed rather than accumulated (Black 2009, p. 30). The traditional category and prestige of mahāyañña remain in place, but the practices and values that constitute a legitimate great sacrifice are changed. Violence, coercion, destruction, and accumulation are removed; non-killing, voluntary participation, and redistribution take their place.

The account of the ancient Brahmins offers another form of this reinterpretation, this time by appealing to an idealized Brahmanical past. In the Pāli version, Brahmins captivated by royal wealth undergo vipallāsa, receive riches through the five sacrifices, and experience increasing taṇhā until the process culminates in cattle slaughter and the declaration of adhamma. The Chinese version narrates the same movement more gradually, from subtle human desires to increasing attachment and, finally, slaughter. Violent sacrifice is thus presented as a departure from “the dhamma of the ancient Brahmins.” This past functions less as a historical account of Brahmanical ritual than as an appeal to values the Brahmin interlocutors themselves are expected to recognize (Freiberger 2009, pp. 66, 72–73).

Buddhist accounts of Brahmanical ritual, however, cannot be taken as straightforward descriptions of actual practice. Pāli texts participate in competition among religious groups and may make Buddhist teaching more attractive by selecting, simplifying, or exaggerating negative features of rival practices (Freiberger 2009, pp. 61–63). The large scale of royal sacrifice, the coercion of laborers, and Brahmanical greed may thus reflect familiarity with such practices while also being heightened for polemical purposes.

The representation of Brahmin social identity raises a related complexity. As seen in Section 3.3, the Brahmin audiences of the Brāhmaṇadhammika-sutta and MA 156 end by asking to be accepted as lay followers. In the Nikāyas, Brahmins who become upāsakas are not necessarily presented as relinquishing their social identity, abandoning the study of the Vedas, or ceasing ritual practice. Buddhist lay affiliation could thus coexist with a continuing Brahmin identity (Black 2009, p. 41; Black 2011, p. 153). The concluding lay-follower formula can thus be read not as the total erasure of Brahmin identity but as another form of recalibration: existing social and cultural frameworks remain, while the standards governing legitimate practice and religious allegiance are reorganized.

The central rhetorical move of the five-sacrifice discourse, then, lies in the selective adaptation of Brahmanical ritual categories within a Buddhist evaluative framework. Traditional categories—yañña and its fruit, the relation between sacrificer and gods, and the dhamma of the ancient Brahmins—remain within the argument. Violence and acquisitiveness are recast as marks of fruitless sacrifice, whereas non-harming and giving become the conditions of legitimate ritual. Together, these patterns show how the five-sacrifice formula was adapted to different audiences and used to redefine inherited ritual values within the shared ritual environment of eastern India.

5. Conclusions

This study has examined the five-sacrifice formula across three Pāli contexts and their Chinese Āgama parallels in order to clarify the rhetoric of early Buddhist criticism of Brahmanical sacrifice. The sequence assamedha, purisamedha, sammāpāsa, vājapeyya, and niraggaḷa is relatively stable in the extant Pāli evidence, whereas the Chinese materials abbreviate, replace, or reorganize its individual terms. These divergences should not be reduced to translation errors or inconsistencies. Rather, they show how sacrificial discourse was reshaped across distinct textual witnesses and transmission histories while preserving a broader argumentative concern with the legitimacy and efficacy of sacrifice.

First, the five-sacrifice formula should be understood not as a transparent catalogue of five fixed Śrauta rites, but as a recognizable ritual sequence with a stable evaluative structure. Its recurrence in different Pāli settings preserves a shared vocabulary of sacrifice, even as its narrative function changes. The formula may evaluate the violence and coercion of a royal sacrifice, respond to a question posed by a Brahmin about the value of sacrifice, or appear in an account of the dhamma of the ancient Brahmins and its decline. Its repetition therefore does not entail a single, uniform Buddhist message about sacrifice; rather, it enables distinct rhetorical uses in different discursive situations. In this respect, the findings bear on the debate between Shulman (2022) and Allon (2022) concerning the compositional status of formulas in the early Buddhist literature. They suggest that the formula was neither wholly free-floating nor merely static: it remained a recognizable ritual sequence while being adapted to distinct narrative settings, audiences, and transmission lineages. In some Chinese parallels, the sequence is condensed around emblematic designations such as horse sacrifice or supplemented by terms such as elephant sacrifice. These variations shift attention away from identifying each item with a specific Śrauta rite and toward a broader image of large-scale, violent sacrifice.

Second, the regional setting helps explain how such ritual language could circulate between Brahmanical and Buddhist communities. Kosala is the primary regional setting of the principal texts, while the broader eastern Vedic horizon, including Videha and Magadha, provides the relevant context for considering the movement of ritual terminology, sacrificial prestige, and ideas concerning kingship and ritual. This is not evidence that Buddhist communities possessed systematic knowledge of Brahmanical sacrificial traditions. Nor can Buddhist ritual terminology be judged solely against the surviving Śrauta texts, since regional ritual traditions may have existed in forms no longer preserved. The relevant historical question is therefore not merely how accurately Buddhist authors understood Vedic sacrifice, but how they selected and repositioned ritual knowledge acquired through contact for particular audiences and argumentative purposes.

Finally, the five-sacrifice discourse can be situated alongside the Sela-sutta and the Kūṭadanta-sutta as complementary instances of Buddhist ritual recontextualization. The Sela-sutta retains the hierarchy of agnihotra and Sāvitrī but redirects it toward the Buddha’s supremacy; the Kūṭadanta-sutta retains the prestige of mahāyañña while redefining it through non-killing, voluntary labor, and redistribution. Likewise, the five-sacrifice discourse retains yañña, the sacrificer, ritual fruit, and the pleasure of the gods, but makes non-harming and giving the criteria of legitimate sacrifice.

Taken together, these findings show that early Buddhist engagement with Brahmanical sacrifice was not confined to denunciation or critique. It involved the selective adaptation of inherited ritual forms and terminology. The five-sacrifice formula thus makes visible a contested field of ritual authority in which inherited Brahmanical categories were retained but their meaning and efficacy were redefined in Buddhist terms.