The Apūrva in Prābhākara Mīmāṃsā

Dear readers,

For my final blog post as guest host for March 2021, I’ll begin by putting the concept of the apūrva on the table, and then I’ll shift into my interpretation of Śālikanātha’s new model of the archetypal Vedic command. I should note that not everyone in the field agrees with my new interpretation of what apūrva means in Prābhākara Mīmāṃsā; that is cool, I’m happy to debate further.

First, to the concept of the apūrva. Mīmāṃsā begins with an investigation into the authority of the Veda via The Chapter on Reason, and only subsequently investigates the meaning of the Veda through hermeneutic principles. This means that with the Veda’s authority first established, our investigation into what the Veda tells us about extra-worldly matters must take everything the Veda communicates as truths that cannot subsequently be cancelled (but the onus is on us as interpreters to figure out exactly what the Veda communicates). For Śabara and Kumārila, the archetypal Vedic injunction: svargakāmo yajeta communicates sacrifice as the means to heaven, as for them this sentence has the semantic structure: yāgena svargam bhāvayet (‘one should bring heaven into being via sacrifice’). That is, with the Veda’s authority already established via The Chapter on Reason, this cause-n-effect relationship must be true. However, Śabara and Kumārila need a solution to the latency problem, i.e., to the issue that sacrifice and heaven defies our normative expectations of cause-n-effect relationships as immediate. To explain: if I shoot one billiard ball into another, they don’t stay connected motionless for several seconds and then the second billiard ball moves, rather the second billiard ball begins to move instantly upon contact with the first ball. However, heaven is not experienced instantly upon the performance of sacrifice: how can what the Veda tells us, that sacrifice produces heaven, be true? Their answer to this latency problem is the apūrva, literally: ‘the un-precedented (a-pūrva).’ The sacrifice produces an apūrva, a trans-temporally stable potency, which in turn produces heaven at a much later time. Thus, for Śabara and Kumārila the apūrva is necessarily postulated based on what the Veda tells us about sacrifice as the means to heaven, even though the language of the Veda is not explicit on the apūrva.

With the concept of the apūrva tabled, let us now turn to my interpretation of Śālikanātha’s new theory (i.e., unattested in Prabhākara’s Bṛhatī). In the Vākyārthamātṛā II, Śālikanātha presents his view on the apūrva as a direct engagement with Kumārila (in Subrahmanya Sastri’s 1961 edition, this is pp. 439-40). How can I be certain Śālikanātha is engaging Kumārila specifically? Let me explain:

Kumārila’s view is on a continuum with Śabara’s in that Kumārila accepts the apūrva as the solution to the latency problem. However, Kumārila faces a real post-Śabara Mīmāṃsā-internal interlocutor who denies the apūrva’s existence on the grounds that there is no epistemological instrument to cognize it (see Cummins 2021, pp. 314-5 and notes 114-6). Kumārila’s answer to his real interlocutor on the epistemological instrument behind how we understand the apūrva is necessary linguistic presumption (śrutārthāpatti): we postulate a term in the text which is not actually there to communicate this term-referent (i.e., the apūrva), based on the otherwise impossibility of sacrifice as a means to heaven (which we, as Mīmāṃsakas, accept as a truth based on the Veda’s infallible authority). Thus, yāgena svargam bhāvayet yāgena apūrveṇa svargam bhāvayet. Kumārila ad MS 2.1.5 is where we first find necessary linguistic presumption (śrutārthāpatti) as the epistemological instrument behind the apūrva.

In the Vākyārthamātṛkā II (Subrahmanya Sastri’s 1961 edition, pp. 439-40), Śālikanātha says: ‘no, the apūrva is not understood through necessary presumption (arthāpatti), it is understood through the language itself’ as per the Prābhākara theory of linguistic signification that terms directly signify their referents as co-construed (anvitābhidhāna). In Cummins 2021 pp. 315-6, I provide a translation and discussion of the relevant passage where Śālikanātha’s prima facie interlocutor says the apūrva is understood through necessary presumption (arthāpatti). Thus, it is beyond doubt here that Śālikanātha’s prima facie interlocutor is a version of Kumārila when the concept of the apūrva rears its head here in Prābhākara Mīmāṃsā, since Kumārila is the only person to hold this specific position that necessary postulation (arthāpatti) is the epistemological instrument by which we understand the apūrva. Moreover, Śālikanātha must have a shared understanding with Kumārila over what the object of debate is. To say: ‘it is not understood through arthāpatti, it is understood through anvitābhidhāna’ requires a shared understanding of what it is (even if specifics of its nature are under debate). Otherwise, their conversation would cease to be coherent. And no one would disagree that for Kumārila the apūrva is the sacrifice’s trans-temporal potency to produce its result.

Śālikanātha’s position against Kumārila that the apūrva is directly signified by language (i.e., by an actual morpheme in svargakāmo yajeta) allows Śālikanātha to maintain the further position that the specific verbal suffix -eta directly signifies the apūrva (whereas on Kumārila’s position of ‘necessary linguistic presumption’, we postulate an extra term not there in the text to communicate the apūrva –> explained above: yāgena apūrveṇa svargam bhāvayet). In turn, saying the apūrva is directly signified by the verbal suffix –eta allows Śālikanātha to subordinately incorporate Kumārila and Maṇḍana’s theory of bhāvanā (the ‘bringing into being,’ best understood as intentional action to produce an outcome). Here is how:

All Mīmāṃsakas agree that in yajeta the verbal suffix –eta is the predominant element in relation to the verbal root yaj-, and Śabara, Kumārila and Maṇḍana all hold that the suffix –eta directly signifies the (in their view, predominant) bhāvanā. Śālikanātha accepts that the verbal suffix –eta signifies the ‘bringing into being’ (bhāvanā), but by advancing the position that the verbal suffix –eta also directly signifies the apūrva, opens the door for the bhāvanā to be positioned as the subordinate linguistic element on the basis of its ontological subordination (if both referents are communicated by the same suffix, we must look to the ontological structure to determine their relative hierarchy). Thus, when both the apūrva (the immediate effect of the bhāvanā) and the bhāvanā (the immediate cause of the apūrva) are established as directly signified by the verbal suffix –eta, this opens the door for the bhāvanā to be positioned as linguistically subordinate to the apūrva on the grounds of its ontological subordination in the causal chain of events.

The payout of this is a new model for Prābhākara Mīmāṃsā whereupon the bringing into being (bhāvanā) is subordinately incorporated to the apūrva – a model entirely unattested in Prabhākara’s own treatise(s). I will defer the interested reader to my recent article in JHS (esp. Cummins 2021 pp. 318-322), but put succinctly, the way this new model works is:

On Prabhākara’s theory of the command (niyoga), the commandee feels an obligation to the action expressed by the verbal root (e.g., yaj- ‘sacrifice’), and then Śabara’s injunctive theory kicks in at moment two the computational process. However, Prabhākara doesn’t really have a solution for where Śabara’s concept of bringing into being (bhāvanā) fits into the command’s computational process, he largely side-steps the issue.

In contrast, Śālikanātha has a solution to this very problem via the apūrva. On Śālikanātha’s new model, the commandee feels an obligation towards the apūrva, not the action (e.g., sacrifice). The payout is the exclusion of the bringing into being (bhāvanā) from being the item towards which one feels an obligation towards at moment one of the command’s computational process. Then, at moment two of the computational process – when bhāvanā theory explicitly kicks in (remember that for Śālikanātha, the bhāvanā is still directly signified by the verbal suffix –eta) – the bhāvanā is incorporated as subordinate in the language hierarchy to the apūrva. In short, this allows Śālikanātha to account for the bhāvanā and also systematically incorporate Kumārila’s new philosophical resources in a Prābhākara model (for an explanation of this in detail with citations, please see my article in JHS pp. 318-322).

4 Replies to “The Apūrva in Prābhākara Mīmāṃsā”

  1. Hi Patrick!

    Your position is now nicely set out and pretty clear, but there is a problem: you can’t overload a syntactic element in Sanskrit: the rules of kAraka exclude that. so the subordinate sense given to the apūrva falls away. I’m pretty sure that was why the manuscript trail ends with Śālikanātha. There’s a useful original commentary by Ganganath Jha, with his translation of Jaimini: he picks up Kumarila and Prabhakara, but not Śālikanātha, who was forgotten. And Jaimini (1.3.30) already allowed that the meaning of Veda depends on grammar.

    Jha speaks for Atharva-vāda, hermeneutic tradition, the Vedic hermeneutics, and it’s pretty clear that he takes that from the Atharvaveda in the way of Vedānta. But Śālikanātha’s ungrammatical computation seems to run rather with vijnana-vāda, from Vyāsa as Bhojaraja noted in his commentary. That’s why I asked about computation and theory of mind.

    A latency of ‘fruits’ is allowed as bija, ‘with seed’, and bija-corrections are calculated in the Siddhantas, giving elapsed time from the exact season of the year (as for planting or reaping). These were found difficult in early Orientalist work on Hindu calendars, and referred to Leonard Euler in St Petersburg: and his equations were then used by Frank P. Ramsey for his landmark 1928 paper in the Economic Journal, introducing an integral of infinite utility, which he called Bliss! Yes, that’s Ananda, as enjoyed by the gods!

    Assuming the integral exists, one can deduce a finite analogue, giving the optimal savings rate: the first of the classic Welfare Theorems. And through Sraffa, the policy behind the comfortable Russian pensions that swung sentiment in the Crimean conflict! So the computations have a very real history of their own.

    • Dear Orwin,

      Thanks for reading my post and for your response.

      I disagree.

      There is no rule in Vyākaraṇa that a single morpheme cannot communicate multiple referents. This should be obvious from the fact that a suffix such as –ena in rāmeṇa gamyate communicates agent, number and gender (three referents).

      In fact, to justify the apūrva as directly signified by the verbal suffix –eta, Śālikanātha points to the position in Bhāṭṭa Mīmāṃsā that the verbal suffix –eta is already accepted as communicating two referents: the bhāvanā and the agent’s number.

  2. I would love to see traditional mīmāmsā scholars opine on your argument here. I don’t see any references to them. Have you run your arguments by them? I think that’s the only way you truly get a good review of your position.

    • Thanks for the response.

      Hopefully this doesn’t come across the wrong way, but I disagree that the only way to get a certain answer is to ask a traditional paṇḍit.

      To me that rings of a bias of privileged access –> the idea that traditional Indian scholars can somehow access Sanskrit texts better than contemporary non-Indians scholars.

Leave a Reply

Your email address will not be published. Required fields are marked *