Jaina Narrative Refutations of Kumārila: Relative Chronology and the History of Jaina-Mīmām.sps h1.1sā Dialogues

Abstract

Assigning a date to Kumārila is notoriously difficult. Kumārila’s dates are usually assigned through a relative chronology of Brahmanical and Buddhist philosophers with whom Kumārila engages or is engaged. This is a precarious method because the dates of these interlocutors are equally unstable. But what if in considering systematic dialogues (śāstra) to be the primary medium for interreligious philosophical debate we have missed a source that does engage with Kumārila, and that can be reliably dated? In this article, I turn to a religious group whom, it has been previously thought, did not respond to Kumārila until the eighth century—Jainas—as well as to a genre that is not typically viewed as a site of systematic philosophical dialogue—narrative. I argue that the Padmacarita, a Jaina Rāmāyaṇa composed by a Digambara writer called Raviṣeṇa, contains a narrative refutation of Kumārila’s commentary to Mīmāṃsāsūtra 1.1.2. By bringing to light this refutation, and explaining how Raviṣeṇa’s Padmacarita can be reliably dated, I assign Kumārila’s terminus ante quem to the date of the Padmacarita’s composition, 676 CE. Finally, I suggest that Raviṣeṇa’s Padmacarita is the earliest extant Jaina text to discuss Kumārila’s claims, and that Jainas used narrative to reflect on Mīmāṃsā before they turned to śāstra as another medium for this dialogue.

Introduction

Kumārila’s dates are important for the history of Indian philosophy. His robust epistemological defense of the authorless Veda immediately became a touchstone of philosophical reflection for Brahmanical, Buddhist, and Jaina philosophers alike. To assign a historical date to Kumārila is not only to recover this little-known historical information but to identify a key moment in the history of premodern South Asian philosophy. But Kumārila never assigns a date, location, or order of composition to his works. And there exists no extant material evidence, such as inscriptions, that locates Kumārila in a particular time or place.

In the absence of such historical information, we rely on a relative chronology of philosophical authors. If we can determine who is responding to whom and when, we can estimate the order in which philosophical authors lived. Kumārila developed his discourses through dialogue with Brahmanical and Buddhist philosophers. Equally, Kumārila’s ideas immediately became a point of discussion among later Brahmanical and Buddhist authors.¹ The dialogical relations between Kumārila on the one hand and Brahmanical and Buddhist philosophers on the other thus constitute the primary means through which Kumārila’s dates are determined. By contrast, Jaina philosophers have proven a less valuable resource because Kumārila does not engage with any specific Jaina philosopher or text,² and Jaina philosophers do not begin to refute Kumārila until the middle of the eighth century.³ On the basis of Kumārila’s dialogical relations with Brahmanical and Buddhist authors, Kumārila is broadly assigned to the sixth or seventh century of the Common Era. That said, we lack a single Brahmanical and Buddhist interlocutor who cites a date for their composition with which we can firmly establish Kumārila’s terminus post quem or terminus ante quem.

But what if, while considering systematic dialogues (śāstra) to be the primary medium for interreligious philosophical debate, we have missed a source that does engage with Kumārila, and that can be reliably dated? In this article, I turn to a religious group whom, it has been previously thought, did not respond to Kumārila until the eighth century—Jainas—as well as to a genre that is not typically viewed as a site of systematic philosophical disputation—narrative. I argue that a Jaina epic called the Padmacarita (676 CE) targets Kumārila as one of its objects of refutation. The Padmacarita, composed by the Digambara Jaina, Raviṣeṇa, is a Sanskrit Jaina retelling of Vālmīki’s Rāmāyaṇa. It is the earliest extant Sanskrit Jaina text to classify itself under the genre category of “purāṇa.”⁴ If the Padmacarita can be shown to be explicitly reacting to Kumārila, and if we could date the text securely to 676 CE, it would provide external evidence for Kumārila’s terminus ante quem. Furthermore, it would suggest that narrative genres were a significant medium through which Jainas engaged in dialogue with Kumārila before they turned to śāstra as the primary medium for this dialogue from the eighth century onwards. In what follows, I survey previous attempts to assign a date to Kumārila before demonstrating that Raviṣeṇa’s Padmacarita contains a narrative in which Kumārila’s claims are debated. In bringing to light Raviṣeṇa’s engagement with Kumārila, I assign the terminus ante quem of Kumārila to the date of the Padmacarita’s composition, 676 CE, and reread Jaina narratives as an indispensable site for Jaina-Mīmāṃsā dialogues.

An Overview of Kumārila’s Dates

Unlike Buddhist philosophers in the first millennium, whose dates have undergone multiple revisions since Erich Frauwallner’s seminal essay, “Landmarks in the History of Indian Logic” (1961), discussions of Kumārila’s dates are typically appended to whichever Buddhist or other Brahmanical author is under examination. There has been little synthetic consideration of the numerous, distinct dialogical relations that have been used to assign Kumārila’s dates, much less the relative weight that each dialogical relation has in determining the upper and lower limits of Kumārila’s dates. This being so, a survey of previous estimations is necessary.

Let me begin with Kumārila’s relation to Mīmāṃsā authors. We know that Kumārila must postdate Śabara because he reacts to Śabara’s commentary on the Mīmāṃsāsūtras. But we do not have a precise date for Śabara beyond the estimate that he lived in the fifth century of the Common Era.⁵ Since the turn of the twentieth century, studies of Mīmāṃsā have identified Kumārila as a junior contemporary of Prabhākara, and on the basis of this relation it has been suggested that Kumārila lived in 600–650 CE.⁶ But the upper and lower limits of Prabhākara’s dates are just as contested as those of Kumārila. This is not to mention that the often-quoted fifty- or sixty-year lifespan allotted to Prabhākara, Kumārila, and indeed, most other South Asian philosophers in the first millennium is an arbitrary convention based on no hard evidence (Franco 2018, pp. 117–118).

The oldest Brahmanical authors to cite Kumārila would, at first glance, seem relevant for dating Kumārila’s terminus ante quem. Kumārila must predate Maṇḍanamiśra because the latter cites the former.⁷ But previous scholarship has assigned Maṇḍana’s terminus post quem to the terminus ante quem of Kumārila (Thrasher 1993, pp. 111–128; Tola 1989).⁸ With this in mind, we cannot assign Kumārila’s dates on the basis of Maṇḍana’s dates lest we incur a circular argument. Even less useful are the dates of the oldest Mīmāṃsā commentator on Kumārila’s Ślokavārttika, Uṃveka. Uṃveka's terminus post quem relies on the dates of Maṇḍana which, we have now seen, depend on Kumārla’s dates.⁹ Uṃveka’s terminus ante quem is easier to identify than his terminus post quem because Uṃveka is cited by the Buddhist writer, Kamalaśīla, who, Dunhuang manuscripts record, was invited to a Tibetan monastery founded in 779 CE.¹⁰

So far, all we can estimate from Kumārila’s dialogical relations with Brahmanical philosophers is that Kumārila lived sometime between the sixth and eighth century CE though the upper and lower limits of this estimation remain fuzzy. Turning to Kumārila’s relation to Buddhist philosophers proves more useful yet equally frustrating. Kumārila’s refutation of the Buddhist thinker, Diṅnāga, provides a more specific terminus post quem for Kumārila. Kumārila must postdate Diṅnāga because the former cites the latter. That said, Diṅnāga’s dates have recently been revised from the previously accepted date, 480–540 CE, to the earlier date of 430–500 CE (Deleanu 2019, pp. 12–19), prompting us to revise in turn Kumārila’s terminus post quem.

Most attempts to date Kumārila rely on the date of Diṅnāga’s commentator, Dharmakīrti. Kumārila and Dharmakīrti are represented as direct contemporaries though there has been debate regarding their exact relation. It is not clear that Kumārila refers to Dharmakīrti, whereas it is clear that Dharmakīrti explicitly refers to Kumārila. Supposing that we could prove that Kumārila and Dharmakīrti were direct contemporaries does not make it any easier to assign a date to Kumārila because, as should by now be predictable, there is no scholarly consensus regarding the date of Dharmakīrti.¹¹ Accepting the most recent hypotheses that assign Dharmakīrti to 550–650 CE (Franco 2018; Deleanu 2019; Eltschinger 2019a) would restrict Kumārila’s dates to 550–650 CE.¹² Each time Dharmakīrti’s dates are revised, so too are Kumārila’s.

Examinations of Kumārila’s dialogical relation to Bhāviveka (Kataoka 2021; Krasser 2012) yields only more fragile evidence on which we can pin Kumārila’s dates. Bhāviveka’s dates are unresolved.¹³ In addition, although Bhāviveka refutes discourses that are found in Kumārila’s works, these refutations are interpreted by some scholars as interpolations of another author who postdates Kumārila. The possibility of interpolation complicates the relative chronology of Bhāviveka and Kumārila in ways that have not been concretely resolved.¹⁴

The date of Śāntarakṣita seems to be the most reliable extant evidence we have for determining Kumārila’s terminus ante quem. Kumārila is cited verbatim by Śāntarakṣita, and unlike all other interlocutors that I have discussed so far, Śāntarakṣita can be reliably dated. Early Tibetan sources record that Śāntarakṣita travelled to Tibet as a senior monk sometime between 763–775 CE.¹⁵ This external evidence for Śāntarakṣita’s date leads to the concrete claim that Kumārila lived before the early- to mid-eighth century.¹⁶

This brings us to the final source deemed relevant for assigning a date to Kumārila: Tibetan histories. In the seventeenth century, a Tibetan monk named Tāranātha composed A History of Buddhism in India.¹⁷ Some have claimed that Tāranātha connects Kumārila with the reign of a Tibetan ruler, Gyo-ba-brtan-pa, 627–650CE (Verpoorten 1987, pp. 22; Sharma 1980, pp. 13–14). Yet no such claim is found in Tāranātha’s text. Tāranātha cites several kings (including Gyo-ba-brtan-pa’s heir) before introducing Kumārila. This makes it unclear whether Tāranātha is, in fact, assigning Kumārila to the reign of Gyo-ba-brtan-pa, the reign of Gyo-ba-brtan-pa’s son, or indeed the reign of other kings cited in the same passage. What Tāranātha does explicitly state is that Kumārila defeated Diṅnāga.¹⁸ Taking seriously Tāranātha’s representation of Kumārila as contemporaneous not only with Dharmakīrti,¹⁹ but also with Diṅnāga would force us to reevaluate Diṅnāga’s date. Underlying all these readings of Tāranātha’s text is a broader methodological critique first highlighted by Frauwallner (1961, pp. 125–126): We cannot read a seventeenth century “history” (whatever “history” signifies for Tāranātha) as a mimetic representation of first millennium South Asia.²⁰

So, what are we left with? The reconstruction of the vast network that connects Kumārila with Brahmanical and Buddhist philosophers leads to an estimate that Kumārila lived sometime between the sixth to the early- to mid-eighth century. The date of Śāntarakṣita’s arrival in Tibet (763–775 CE) furnishes the only reliable external evidence we have for Kumārila’s terminus ante quem. If we try to specify that Kumārila lived in the sixth to seventh century because of his dialogical relations to those who lived before Uṃveka and Śāntarakṣita—such as Maṇḍaṇamiśra and Dharmakīrti—we must know that we are doing so in the absence of established dates for these interlocuters. Ultimately, to confine Kumārila’s dates to the seventh century requires us to play an elaborate game of who knows whom because none of these philosophers, who are chronologically more proximate to Kumārila than Śantarakṣita, cite a date for their compositions. If only, as scholars of Mīmāṃsā opine, we could establish Kumārila’s date independently of Brahmanical and Buddhist śāstra (Taber 2005, pp. 163, ft 2; Freschi 2014²¹).

Jaina Tales of Rāma

One would be forgiven for having never considered the possibility that Jaina Rāmāyaṇas include dialogues with Kumārila’s Mīmāṃsā. Jaina tales about Rāma have, until now, always been examined in relation to Vālmīki’s Rāmāyaṇa or with other Jaina tales of Rāma. This methodological perspective makes sense in light of the history of Jaina Rāmāyaṇa retellings.²² From the fifth century of the common era, Jainas began to compose tales about Rāma, whom they call “Padma.” The earliest extant Jaina text to be dedicated to the tale of Rāma is Vimalasūri’s Paümacariya, composed at the beginning of the Common Era.²³ Vimalasūri presents his Paümacariya as a literary and theological refutation of Vālmīki’s Rāmāyaṇa by retelling the Rāma’s tale in such a way that rectifies what he views as hermeneutical or theological problems in Vālmīki’s text.²⁴

Vimalasūri’s Prakrit Paümacariya is retold for the first time in Sanskrit by Raviṣeṇa’s Padmacarita. Raviṣeṇa—as with all Jaina authors of Sanskrit purāṇas—cites a date for his composition. In the closing verses of the text, Raviṣeṇa states that he completed the Padmacarita 1203 years and six months after Mahāvīra attained the final liberation through death.²⁵ The date of the Padmacarita depends on the date of Mahāvīra’s death. This fact is less problematic than it might initially appear because Jaina sources from the first millennium of the common era are relatively consistent in assigning Mahāvīra’s death to 605 years and 5 months before Śaka’s reign, which would place Mahāvīra’s death in 527 BCE.²⁶ The frequency with which Digambara and Śvetāmbara sources before the seventh century of the common era assign Mahāvīra’s liberation to specifically 605 years and 5 months before Śaka’s reign gives us strong reason to believe that Raviṣeṇa is operating with this traditional date. Based on this traditional date for Mahāvīra’s death, 527 BCE, we can calculate that Raviṣeṇa completed his Padmacarita in 676 CE.²⁷

As a retelling composed several centuries after Vimalasūri’s Paümacariya, Raviṣeṇa recasts the tale of Rāma to engage in contemporaneous Brahmanical discourses. This is most clearly seen in Raviṣeṇa’s retelling of the tale of King Marutta and his priest, Saṃvarta, for it is through this subtale, I argue, that Raviṣeṇa discusses Kumārila’s Mīmāṃsā.

According to book 7 of Vālmīki’s Rāmāyaṇa, King Marutta sponsors a non-violent sacrifice that is officiated by the Brahmin, Saṃvarta.²⁸ The demon Rāvaṇa interrupts King Marutta’s sacrifice and demands that the king either fight him or acknowledge his defeat. Saṃvarta restrains his King and says, “If you want my advice, battle is not appropriate for you. A sacrifice to Maheśvara left incomplete will destroy your dynasty. How could one who is consecrated for sacrifice engage in battle? How can one engage in violent acts while they are consecrated?”²⁹ Saṃvarta defers to Vedic ideology to explain to King Marutta that the agent of sacrifice will no longer be fit to complete the sacrifice if he engages in battle. Implicit here is a Vedic belief that sacrifice guarantees the preservation of the cosmos and the continuation of the King’s lineage.³⁰ Marutta thus cannot fight Rāvaṇa without violating his ritual consecration and by extension, forfeiting the personal, social, and cosmic effects of his sacrifice. King Marutta presses ahead with the sacrifice to ensure its completion. Rāvaṇa tries to disrupt the rite. He devours the Brahmins who assembled for the sacrifice, but Marutta’s sacrifice is nevertheless completed, and the gods praise Marutta.

Valmīki’s Rāmāyaṇa paints King Marutta as an ideal patron of Vedic sacrifice because he abides by Vedic prescriptions. By contrast, Vimalasūri’s retelling of this tale undermines the validity of Vedic sacrifice. According to Vimalasūri’s Paümacariya, King Marutta and his priest, Samvarta, undertake a Vedic sacrifice that involves the slaughter of animals.³¹ The Jaina sage, Nārada, interrupts the sacrifice, rebukes the violence that Marutta and Saṃvarta inflict onto living beings, and reinterprets the Vedic sacrifice as a metaphor for Jaina asceticism. Saṃvarta and his Brahmin attendants reject Nārada’s arguments. But rather than dismissing Nārada through dialogue, the Brahmins forgo further discussion and beat Nārada within an inch of his life. News of Nārada’s plight reaches Rāvaṇa, who is represented by the Paümacariya as a demi-god (vidyādhara) and an adherent of the Jina’s teaching of non-violence. Rāvaṇa saves Nārada and frees the animals being prepared for sacrifice. We can read the Paümacariya’s retelling as an inversion of that of Vālmīki’s Rāmāyaṇa. In Vālmīki’s Rāmāyaṇa, Marutta maintains the stability of his lineage and the cosmos through the performance of Vedic sacrifice while Rāvaṇa’s rampage aims to disrupt social and cosmic order. But in Vimalasūri’s Paümacariya, Marutta’s Vedic sacrifice, and the violence it ensues, disrupts the Jaina order of the kingdom while Rāvaṇa saves the animals as a way of restoring order to the Jaina kingdom. In the broader context of Jaina texts, the Paümacariya’s tale of Marutta resembles earlier and contemporaneous Jaina representations of Vedic Brahmins.³²Just as Jaina suttas homogenize the diverse justifications underlying Vedic sacrifice into a single representation of Brahmins as violent humans who cannot be reasoned with, so too does Vimalasūri’s Paümacariya caricature King Marutta and Saṃvarta as barbaric Vedic practitioners without ever discussing the breadth and specificity of Vedic beliefs and practices that were being enjoined by contemporaneous Brahmanical texts.

Raviṣeṇa’s retelling in the seventh century Padmacarita follows the sequence of events as told earlier by Vimalasūri’s Paümacariya. King Marutta and Saṃvarta perform an animal sacrifice that is interrupted by Nārada who proclaims that animal sacrifice will lead to hell. And once again, Nārada is beaten to pulp by the Brahmins. The difference between Raviṣeṇa’s Padmacarita and Vimalasūri’s Paümacariya lies less in the plotline of this subtale than in the content of the characters’ dialogue. Raviṣeṇa inserts a lengthy conversation between Marutta’s Vedic officiant, Saṃvarta, and the Jaina sage, Nārada, that is absent in earlier versions of this tale. Through this dialogue, which spans over a hundred verses, Nārada and Saṃvarta debates claims that, I argue, are voiced by Kumārila in his Ślokavārttika. Let me be transparent here. Nowhere in Raviṣeṇa’s subtale or in the Padmacarita entire does Raviṣeṇa label Saṃvarta a “Mīmāṃsaka” or a follower of Kumārila. In fact, Raviṣeṇa never identifies any of the tradition(s) or intertext(s) that are discussed by Nārada and Saṃvarta. This textual practice contrasts with later Jaina narratives from the eleventh century onwards, such as Amitagati’s Dharmaparīkṣā (1014 CE), which explicitly identifies one of the antithetical positions as “Mīmāṃsaka”.³³ The absence of such explicit identifications means that we need to prove through a close reading of the dialogue that Saṃvarta’s position voices Kumārila’s Mīmāṃsā.

Nārada’s refutation of Saṃvarta in Raviṣeṇa’s Padmacarita

In the Padmacarita, the debate between Saṃvarta and Nārada arises because Nārada rejects the validity of Vedic animal sacrifice. Nārada, witnessing the animals being lined up for slaughter, interrupts the sacrifice in a bid to save the animals. He declares that the omniscient Jinas have previously taught that the violence inflicted on animals during Vedic sacrifices leads to undesirable results on the part of the sacrificer. Such results include the attainment of hell.³⁴ Saṃvarta, the Brahmin officiant of the sacrifice, is outraged. He explains his position on the validity of the Veda and its sacrificial injunctions through the following verses.

Saṃvarta, outraged, said, “Oh, because of your complete stupidity, you are making a claim that is entirely incoherent; it has no logical grounds. (164) You believe that someone who is omniscient must be devoid of desire. [But,] He cannot [be omniscient] if he possesses properties such as speaking. The opposite would also apply: [He cannot be a speaker if he is omniscient]. (165) Words spoken by imperfect authors are full of imperfections, and there exists no-one dissimilar to this because there is no proof (of a speaker who is devoid of imperfections). (166) Therefore, the Veda, being authorless, must be the [only] valid means of knowing (pramāṇa) with respect [to objects] that lie beyond the sense faculties. Furthermore, it enjoins the three classes to perform ritual acts. (167) The fixed dharma known as “apūrva,” which manifests through sacrifice, produces in heaven a result that arises from desirable sense objects. (168) Moreover, killing animals in sacrificial contexts does not lead to a negative effect. One should perform rituals such as sacrifice because it is enjoined by the Veda. (169) Indeed, Svayambhū created animals for the sake of sacrifice. So, what fault is there in killing those [animals] who are created for this reason? (170)³⁵

In the above verses, Saṃvarta justifies the validity of the Vedic injunctions to perform animal sacrifice. First, he rejects the possibility that an omniscient speaker exists by calling out what he sees as a logical contradiction in the definition of the Jina as an omniscient speaker. An omniscient being cannot engage in the act of speaking because all activity, including speaking, is predicated on a desire. The Jina must either be an omniscient being who does not teach on account that he has no desire to do so, or else the Jina teaches but the presence of a desire to act renders him susceptible to error. For Saṃvarta, this means that the Jina’s words, and indeed all authored testimony, is fallible. By contrast, Saṃvarta understands the Veda to be infallible because it is authorless. Without an author, there is no locus to which one could attribute any fault. Thus, on Saṃvarta’s reading, the authorless Veda is infallible. More precisely, the Veda is the valid means of knowing (pramāṇa) with respect to “that which lies beyond the sense faculties”—a phrasing that, as we will later see, Mīmāṃsakas use to define “dharma.”

Establishing the validity of the Veda is paramount for justifying the validity of animal sacrifice. If Saṃvarta can prove that the Veda is authoritative, then he can justify the belief that animal sacrifices enjoined by the Veda lead to the attainment of heaven. Therefore, having demonstrated the validity of the authorless Veda over the Jina’s words, Saṃvarta explains that there is no reason to believe that animal sacrifices will lead to undesirable results. The Veda enjoins animal sacrifice for the attainment of heaven. Since the Veda is authoritative and there is no forthcoming cognition that reveals the effects of vedic sacrifice to be otherwise, we are justified in believing that Vedic animal sacrifice leads to the attainment of heaven.

The above arguments recall those that Kumārila advances in his commentary on Mīmāṃsāsūtra 1.1.2 in the Ślokavārttika. For instance, Saṃvarta’s definition and rejection of omniscient speakers run parallel to those of Kumārila. According to Kumārila, an omniscient speaker is one who simultaneously has no desires (rāgādirahita) and who does not engage in any activity (nirvyāpāra).³⁶ Kumārila applies the premise that all action is preceded by a desire to his definition of omniscience speakers to bring into relief the logical contradiction in the definition of the Buddha and the Jina as a speaker who engages in the act of teaching while remaining desireless. The Buddha and the Jina cannot teach (deśanā) without a desire because all actions are predicated on a desire. Yet the Buddha and the Jina cannot be devoid of desire because they engage in the act of teaching. In both cases, the existence of an infallible omniscient speaker is, on Kumārila’s account, impossible. As we can see, Kumārila’s definition of omniscient speakers is identical with that of Saṃvarta.

That Kumārila’s discussion of omniscient speakers departed from earlier extant Mīmāṃsakas further substantiates the claim that Saṃvarta is voicing Kumārila rather than any other author. Kumārila’s discussion of omniscient beings arises precisely because his predecessor, Śabara, never addresses the topic. Śabara claims that the testimony of reliable men is valid when it pertains to perceivable objects. In Kumārila’s reading, Śabara’s claim leads to an undesirable consequence, namely that Buddhist and Jaina scriptures can be taken as valid since their authors, the Buddha and the Jina, are believed by Buddhists and Jainas respectively to be omniscient and, by extension, reliable speakers. Kumārila includes a rejection of omniscient speakers to curtail the possibility that Jaina and Buddhist scriptures constitute a valid means of knowing over and above the authorless Veda. The conceptual parallels between Kumārila’s and Saṃvarta’s discussion of omniscient speakers, combined with the claim that Kumārila is the earliest extant Mīmāṃsaka to reject omniscient speakers further suggests that Saṃvarta is a mouthpiece for Kumārila’s ideas.

Second, Saṃvarta’s contrast between the invalidity of authored texts and the validity of the authorless Veda (PC 11.166–167) is suggestive of the same contrast that is introduced by Kumārila. Kumārila rejects the possibility that omniscience and authorlessness constitute equally valid reasons for establishing a scripture as valid.³⁷ To claim that an omniscient speaker exists is to postulate something unseen. But to claim that the Veda is authorless does not require one to postulate something unseen. We observe the Veda being transmitted from teacher to student across multiple generations. Based on this perception, we can justifiably believe that the Veda has been transmitted from teacher to student ad infinitum. There is no forthcoming cognition of the Veda’s creation to suggest otherwise. In this way, the Veda’s authorlessness is established through perception, unlike the existence of an omniscient speaker which goes unseen. Kumārila’s contrast between the postulation of omniscient speakers and the postulation of the authorless Veda lies in the background of verses 166–167 of Saṃvarta’s discussion, which contrast the authored, fallible teachings of the Jina with the authorless, infallible Veda, though it is left undeveloped. Nevertheless, Saṃvarta follows the order of arguments given in Kumārila’s commentary. Just as Kumārila shifts from a rejection of omniscient speakers to a justification of the Veda’s validity, so too does Saṃvarta. Saṃvarta synthesizes Kumārila’s claim that the absence of an author means that there is no locus (i.e a speaker) to whom we could attribute any faults (PC 11.167; ŚV 2.169). They both conclude that the Veda is the only valid means of knowing dharma because dharma, defined as that which lies beyond sense faculties, can only be grasped by the authorless injunctions that constitute the Veda.

Saṃvarta’s final comments regarding the justification to slaughter animals in Vedic sacrifice is one of the few arguments that does not align exactly with Kumārila’s Mīmāṃsā. Saṃvarta’s summary of “apūrva dharma” resonates with older Mīmāṃsā ideas just as much as it does with contemporaneous Naiyāyikas. Saṃvarta’s final claim that God created animals for the sake of sacrifice contradicts Kumārila who vehemently rejects the existence of a creator deity.³⁸ Although these broader arguments are not found in Mīmāṃsā or Kumārila’s Ślokavārttika, they nevertheless lead to Kumārila’s idiosyncratic commitment to animal sacrifice and his belief that Vedic sacrifice constitutes a distinct context in which violence is justified.

Śabara briefly contends that sacrifices such as the Śyena sacrifice, which is performed to bring harm to the agent’s enemies, leads to negative effects. But Śabara never addresses the validity of animal sacrifice at large. Kumārila not only engages with the question of violence in animal sacrifices, but he defends it. Immediately after justifying the validity of the Veda, Kumārila claims that animal sacrifice in the specific context of Vedic ritual leads to beneficial results. This unique defense of animal sacrifice and the position in which this argument appears in Kumārila’s commentary once again mirrors that of Saṃvarta from verse 169. While Saṃvarta’s claims about “apūrvadharma” and a creator are not found in Kumārila’s commentary, they nevertheless support Kumārila’s overarching commitment to animal sacrifice. Indeed, as I later explain, Saṃvarta’s discussion of “apūrva dharma”, though different from Kumārila’s rendition, is located in the same position that Kumārila discusses “apūrva” rather than in the position that Śabara discusses “apūrva”.

Despite the brevity of Saṃvarta’s arguments, there are conceptual and sequential parallels between Saṃvarta’s claims in the Padmacarita and Kumārila’s commentary on MS 1.1.2. Saṃvarta articulates innovations that Kumārila uniquely made to Śabara’s commentary—the rejection of the existence of omniscient beings, the contrast between omniscient beings and the authorless Veda, and the justification to perform animal sacrifice.

We find further evidence for this intertextual relation when we turn to Nārada’s refutation.³⁹ Nārada responds to specific subordinate claims that can be found in Kumārila’s discussion of omniscient speakers, the Veda, and animal sacrifice, following the same order in which those arguments are presented by Kumārila in his commentary on MS 1.1.2.

Nārada introduces his response by criticizing Saṃvarta’s first claim that there exists no omniscient speaker. He begins with a linguistic argument. An omniscient being (sarvajña) must exist because the word “sarvajña” would not yield a cognition unless there existed a referent—an actually existing omniscient speaker—to whom the term refers. For Nārada, cognitions correspond to the actual state of affairs in the world. Speech would be impossible if linguistic expressions (śabda) and cognitions (buddhi) did not depend on the existence of a referent (artha), external to cognition, to which words and cognitions refer.⁴⁰ For instance, the linguistic expression “go” results from having a cognition of a referent, a cow, that exists in the world. Similarly, Nārada explains, the linguistic expression “sarvajña” depends on a perceptual cognition of an omniscient being, the Jina, who exists in the world.⁴¹ In the context of the Padmacarita’s narrative, Nārada’s argument ought to be taken literally because Nārada, unlike the readers of the Padmacarita, inhabits an era in which the Jinas exist.

Read in the context of the Ślokavārttika, Nārada’s argument seems to target Kumārila’s claim that no pramāṇa establishes the existence of an omniscient speaker (ŚV 2.117–155). We do not have a perception of an omniscient speaker in our era. We cannot infer the existence of an omniscient being since there are no inferential marks. And, there is no scriptural testimony that attests to the existence of omniscient beings.⁴² Kumārila’s claim that no pramāṇa can establish the existence of omniscient beings seems to be the target of Nārada’s rejoinder in PC 11.172–178 because Nārada brings to light the ways in which the pramāṇa of Vedic testimony does establish the existence of an omniscient being.

Nārada cites a verse from Muṇḍaka Upaniṣad that affirms the existence of an omniscient being (sarvajña).⁴³ This citation is provocative when read in the context of ŚV 2.118–119 where Kumārila claims that there exists no Vedic passage (āgamābhāva) that refers to the existence of omniscient beings.⁴⁴ Kumārila explains that we cannot defer to scriptural passages that are written by omniscient beings, including the Jina, because this would incur the fault of mutual reliance (ŚV 141–142ab). The sole scripture that is authoritative is the Veda on account of its authorlessness. But according to Kumārila, no Vedic passage expresses the existence of an omniscient being (ŚV 2.119ab).⁴⁵ Read in the context of this argument, Nārada’s citation of the Muṇḍaka Upaniṣad is not fortuitous. He exposes a contradiction in his opponent’s reasoning (PC 11.178ab). The opponent’s claims about the Veda do not accurately capture what is expressed by those very scriptures.⁴⁶ And in citing a Vedic text over a Jaina text, Nārada circumvents the fault of mutual reliance and proves his argument through a scripture that his opponent deems valid.

Nārada moves on to the question of whether an omniscient speaker can engage in action if that speaker is, by definition, devoid of desire. As seen earlier, the opponent’s definition of omniscient speakers (PC 11.165–166) parallels that of Kumārila in the Ślokavārttika (ŚV 2.137–140). Nārada responds to this definition in PC 11.179–186 with the rejoinder that there exists no contradiction between being omniscient and being a speaker. In Nārada’s definition, an omniscient individual must be a speaker. He contends that we cannot equate omniscient speakers such as the Jina with non-omniscient speakers, such as Devadatta, because the former, being omniscient, does not possess the faults (doṣa) that arise from not possessing omniscience.⁴⁷

As we can see so far, Nārada’s arguments regarding the existence of an omniscient speaker follow the order and content of claims that are forwarded by Kumārila in ŚV 2.117–140. Nārada continues to address the opponent’s arguments on his terms. He argues that the Vedas’ authorlessness cannot be established through an absence of proof.

The claim that the Veda has no author is not established through the absence of proofs. On the contrary, the claim that the Veda has an author can be proven in the same way we prove perceptible objects. [189] Moreover, the fact that the Veda has a particular arrangement of words and expressions, which make it possible [for it to express] the meaning of injunctions and prohibitions, is proof that the Veda has an author; just as in the case of the poetry of Maitra. [190] Furthermore, some say that the Veda was created by men such as Brahmā-Prajāpati. It is not possible to shrug off this belief. [191] If you think that they are not authors but reciters of scripture, even then, reciters possess faults such as attachment and aversion. [192]. If they are truly omniscient, then why would they author a teaching of the text in one way and an explanation of its meaning in another way given that their teaching is considered a pramāṇa? [193]⁴⁸

As I explained earlier, Kumārila concludes that we can take seriously our initial perception that the Veda has been transmitted from teacher to student ad infinitum unless or until there is a forthcoming perception that proves the Veda to have been created at a particular point in time.⁴⁹ That said, no such perception is forthcoming because we cannot ever perceive the creation of the Veda. Narada rejects these arguments. Contrary to Kumārila’s claim that the Veda’s authorlessness is established through an absence of proofs, Nārada highlights two proofs that establish the existence of the Veda’s authorship. In the first case, the arrangement of words into expressions that convey meaning would not be possible without an author to arrange the words in such a way.⁵⁰ Even if this argument is dismissed, as incidentally it is by Kumārila in ŚV 2.170,⁵¹ Nārada notes that not everyone who subscribes to the authority of the Veda accepts that the Veda is authorless. Some believe that the deity, Brahmā-Prajāpati, authored the Veda (PC 11.191).⁵² Nārada is not wrong to point out the existence of these alternative beliefs. The Veda, as well as in Brahmanical epics and purāṇas, narrate stories about Brahmā’s creation of the Veda.⁵³ Nārada could be citing Vedic scripture as evidence for the claim that the Veda was authored. But the literary context—that is, the narrative world that Nārada and Saṃvarta inhabit—suggests that verse 191 is referring to perceptual evidence. Prior to meeting Saṃvarta, Nārada chastises a character who goes by the name of “Brahmā” and who authors the Veda.⁵⁴ Nārada’s argument in verse 191 relies on perception because he, along with the rest of the kingdom, has witnessed the Vedas’ authorship. According to Kumārila’s epistemology, such a perception invalidates the belief that the Veda is unauthored. Raviṣeṇa’s narrative framing therefore allows him to jettison the Mīmāṃsā claim that we can never perceive the Veda’s creation. The characters can employ perceptual evidence for the Veda’s authorship that is impossible on the part of Raviṣeṇa himself. Nārada’s clarification in PC 11.192 that “Brahmā” is subject to attachments and aversions is again based on perceptual evidence. Nārada and the kingdom have seen that “Brahmā” is none other than a demon who desires to delude the Jaina kingdom.⁵⁵

The final section of Nārada’s refutation responds to the justification to perform animal sacrifice. Broadly speaking, Nārada questions the Mīmāṃsā definition of “dharma” as a ritual act enjoined by the Veda for the sake of producing beneficial results (artha), including the attainment of heaven.⁵⁶ Nārada begins by challenging the claim that the result of Vedic sacrifice is not immediately perceptible (apūrva).

The dharma known as “apūrva” cannot manifest as an effect of ritual action if it is eternal, like the sky. If it manifests then it is impermanent, just as in the case of a pot.⁵⁷ [206] Just as the discrimination of forms is an effect that is perceived after light has manifested, in the same way, the result (of performing sacrifices) should be perceived here in the world after apūrva dharma has manifested. [But it is not.] [207]⁵⁸

The opponent’s understanding of “apūrva” recalls ideologies that circulated as the mainstream among Mīmāṃsakas prior to Kumārila.⁵⁹ For instance, Śabara defines “apūrva” as a “dharma” that manifests as something new.⁶⁰ This definition armed Mīmāṃsakas with a unique defense of the efficacy of Vedic sacrifice. To any opponent who claimed that we do not perceive the beneficial effects of Vedic sacrifice, Mīmāmṣakas could simply respond with the rejoinder that such effects had not yet come into fruition.

Kumārila departs from his predecessors on the definition of “apūrva”.⁶¹ In ŚV 2.196–200, which commences Kumārila’s commentary on “artha,” Kumārila agrees that the effects of Vedic sacrifice are not immediately perceived but redefines “apūrva” as “a mere capacity of a sacrifice that operates towards a fruit.”⁶² This new definition circumvents the ontological criticisms that arise from understanding “apūrva” as a substance which Nārada articulates.

For Nārada, apūrva dharma must be an impermanent substance that manifests at a particular place and time. This definition allows Nārada to reject the purportedly beneficial results of animal sacrifice. If apūrva dharma is an effect of Vedic sacrifice that arises at some place and time, then the fact that we never perceive these effects, such as the attainment of heaven, constitutes proof that Vedic injunctions are inefficacious. Nārada’s argument targets the mainstream definition of “apūrva” voiced by pre-seventh century Mīmāṃsakas rather than Kumārila’s innovation.⁶³ Nevertheless, the location in which Nārada addresses “apūrva” is the same as in Kumārila’s discussion.⁶⁴ Śabara defines “apūrva dharma” outside his commentary to MS 1.1.2. whereas Kumārila sandwiches his discussion of apūrva between his discussion of the authorless Veda and his discussion animal sacrifices in MS 1.1.2. Though Nārada does not deal with Kumārila’s unique definition, the positionality of his argument still strongly suggests that Raviṣeṇa is following the sequence of Kumārila’s arguments as they develop in the commentary to MS 1.1.2.

The remainder of Nārada’s arguments do counter Kumārila’s unique defense of animal sacrifice. According to the Mīmāṃsāsūtra 1.1.2, dharma is defined as a ritual act enjoined by Vedic injunctions for the sake of beneficial results (artha). At stake is understanding the extent to which violence (hiṃsā) is justifiable in the context of Vedic ritual.⁶⁵ Kumārila defends animal sacrifice against Sāṃkhya writers, Buddhists, and Jainas who prohibit all violent acts on the grounds that they incur negative results on the part of the agent.⁶⁶ He achieves this through reference to his justification that the Veda is the exclusive means of knowing dharma. The Veda, constituted by authorless injunctions, uniquely conveys the future effects of Vedic sacrifice in a way that all other forms of expression cannot. Kumārila takes this a step further by distinguishing between different Vedic commands. Injunctions express Vedic rituals that lead to beneficial results whereas prohibitions express actions that lead to negative results (ŚV 2.214-216). Therefore, if animal sacrifices were to incur negative results, the Veda would convey this effect through a prohibition. The fact that the Veda enjoins animal sacrifice demonstrates, on Kumārila’s reading, that animal sacrifice produces beneficial results. Moreover, Kumārila contends, we never perceive a negative result after a Vedic sacrifice is completed (ŚV 2.232-6). In the absence of a forthcoming perception of a negative effect of Vedic sacrifice, we are entitled to believe that animal sacrifice leads to beneficial results. Note that Kumārila’s argument does not, justify violence in all contexts. He distinguishes violence performed in Vedic sacrifice from violent rituals undertaken outside of Vedic contexts. Violent acts undertaken outside the context of Vedic ritual leads to demerit precisely because it is not enjoined by Vedic injunctions (ŚV 2.260-268).⁶⁷

Kumārila’s defense of animal sacrifice, synthesized by Saṃvarta in verse 169, is unpacked by Nārada. For instance, Kumārila’s distinction between injunctions and prohibitions, while not explicitly stated by Saṃvarta, is presupposed in Nārada’s response. Nārada cites a Vedic text, the Chāndogya Brāhmaṇa, that enjoins expiation rites (prāyaścitta) to remove the sin incurred through animal sacrifice.⁶⁸ Such expiation rites furnish evidence that the Veda does enjoin rites that incur negative results for why else would an expiation rite be enjoined if not to expiate the negative result incurred through animal sacrifice?⁶⁹ By referring to Vedic expiation rites, Nārada overturns Kumārila’s claim that Vedic injunctions do not enjoin rituals that incur negative results.

Next, Nārada addresses the distinction between killing animals in the context of Vedic sacrifice and killing animals in non-Vedic contexts. Once again, this argument presupposes Kumārila’s attempt to distinguish the two contexts. Nārada contends that the context does not impact the result of an action. Killing animals, whether in Vedic sacrifices or in hunting, leads to the same negative results, the accumulation of sin, because in both cases violence is inflicted onto an animal (PC 11.216). Here, Nārada implicitly leans on the Jaina theory of karma to collapse the distinction between Vedic and non-Vedic actions. For Nārada, the laws of karmic causality are universal. This universality means that violence inflicted on animals will always result in the accumulation of sin on the part of the agent irrespective of the context in which the act is performed.

Finally, Nārada tackles an argument that is accepted by Śabara and Kumārila—namely, that there is no forthcoming perception of any negative effects of Vedic sacrifice.⁷⁰ Nārada asks, “If we accept that humans go to heaven as a result of performing animal sacrifice, then why did Vasu fall to hell?”⁷¹ Nārada recounts how King Vasu entered hell because he enjoined the sacrifice of animals rather than the offering of rice seeds to the Jina.⁷² The tale of King Vasu is told and retold by Brahmanical texts such as Vyāsa’s Mahābhārata.⁷³ So, we might understand Nārada to be citing scriptural testimony. But once again, Nārada is, in fact, employing perceptual evidence. At the beginning of Chapter 11, just prior to the narration of Nārada’s debate with Saṃvarta, the Padmacarita describes the lives of Nārada and his childhood friend, King Vasu.⁷⁴ The two characters grew up together under the same Jaina teacher. As adults, Nārada and Vasu interpret the Jina’s teaching differently. Vasu proclaims that the Jina enjoined the sacrifice of animals while Nārada argues that the Jina enjoins offerings of rice seeds. Vasu’s interpretation of the Jina’s words, and his subsequent performance of animal sacrifice, cause him to fall to hell. This is witnessed by the entire kingdom including Nārada. Therefore, Nārada’s reference to King Vasu constitutes perceptual evidence. Nārada and the kingdom have a perception that invalidates the claim that Vedic animal sacrifice leads to heaven. The storyworld in which Nārada’s dialogue is embedded allows Raviṣeṇa to challenge the Mīmāṃsā claim that we never perceive negative effects of Vedic sacrifice on epistemological grounds that Mīmāṃsakas accept.

Conclusion: Relative Chronology and the History of Jaina-Mīmāṃsā dialogues

In sum, there is good reason to believe that Saṃvarta’s pūrvapakṣa position, which is summarized by Saṃvarta and unpacked by Nārada, voices Kumārila’s commentary to MS 1.1.2 in the Ślokavārttika. In the Padmacarita, the opponent’s arguments align with those that Kumārila introduced into Mīmāṃsā. Most notably, Kumārila’s rejection of omniscient speakers and his defense of animal sacrifice are summarized by Saṃvarta. Furthermore, the order in which the opponent’s arguments proceed in the Padmacarita follows the same order of arguments found in Kumārila’s commentary. We can see this at the macro-level of the debate overall. Saṃvarta tackles the existence of omniscient speakers, the validity of the authorless Veda, the concept of apūrva, and the validity of animal sacrifice in that order. This is same sequential order in which these arguments are laid out in Kumārila’s commentary to MS 1.1.2. To take a counter example, Śabara does not discuss the validity of omniscient speakers or the violence in sacrifices, and he locates his discussion of “apūrva” outside the commentary to MS 1.1.2. At the micro level of Nārada’s position, most clearly seen in Nārada’s discussion of omniscient speakers, Nārada’s responses tend to follow the content and order of subordinate arguments that unfold in Kumārila’s commentary to MS 1.1.2. Of course, there are points where the opponent’s arguments diverge from Kumārila’s commentary. My point is not that Kumārila’s Ślokavārttika constitutes the sole intertext for Saṃvarta’s position. Rather, my point is to highlight that Kumārila’s commentary to MS 1.1.2 is a significant and consistent primary source that Raviṣeṇa draws upon to frame Saṃvarta’s Vedic position.

Identifying Kumārila as one of the primary targets of the Padmacarita impacts the terminus ante quem that we assign to Kumārila. Raviṣeṇa—as with all Jaina authors of Sanskrit purāṇas—cites a date for his composition of the Padmacarita. Raviṣeṇa completed the Padmacarita in 676 CE. If we accept that Raviṣeṇa’s Padmacarita engages in a dialogue with Kumārila’s thesis, then the year of the Padmacarita’s completion marks the terminus ante quem for Kumārila.

Equally important is that the Padmacarita revises our understanding of the history of Jaina-Mīmāṃsā dialogues in the first millennium of the Common Era. Reconstructing Jaina-Mīmāṃsā dialogues solely through the study of Jaina śāstra leads to the presumption that Jainas did not respond to Kumārila until sometime in middle of the eighth century, when Akalaṅka and Haribhadra composed śāstric refutations of Kumārila.⁷⁵ But when we turn to Jaina narratives, which lie outside the strict confines of śāstra, we can see that Jainas were far quicker to react in writing to Kumārila’s discourses than has been initially thought. The Padmacarita, a self-proclaimed “purāṇa” and “carita,” is the earliest extant Jaina text across any genre to discuss Kumārila’s discourses. This claim is liable to change as research into understudied Jaina texts continues. Nevertheless, the Padmacarita impels us to recast narratives as a site in which Jainas reflected on Mīmāṃsā thought.