Skip to main content

Advertisement

SpringerLink
Log in

Mīmāṃsāsūtra 6.5.54 on bādha in Maṇḍanamiśra’s Brahmasiddhi

  • Published:
Journal of Indian Philosophy Aims and scope Submit manuscript

Abstract

This paper will show how the philosopher Maṇḍanamiśra discusses in his Brahmasiddhi the cancellation (bādha) of a former element by a latter, which is prescribed in Mīmāṃsāsūtra 6.5.54. We do not have yet a clear idea of what the value of this text holds for him. I would emphasize that probably more than we had expected, it forms an essential part of Maṇḍana’s philosophy. Its authority is sometimes stated explicitly and sometimes not; and we easily overlook the fact that his argument is highly dependent on the Mīmāṃsā scheme. This rule, 6.5.54, was originally purely concerned with the performance of the sacrifice, but Maṇḍana applies it to his epistemic analysis, i.e. his discussion on the relative strength of the valid means of knowledge, giving this rule the status of basic testimony. Furthermore, he interprets the rule in his own unique way, or at least differently from Kumārilabhaṭṭa, integrating it with the argument by his famous predecessor Bhartṛhari. The mutual relationship among Maṇḍana, Kumārila, and Bhartṛhari, will be illuminated by focusing on what Mīmāṃsāsūtra 6.5.54 really means.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1

Similar content being viewed by others

Notes

  1. This is supported by the earliest commentary by Vācaspatimiśra (10th c.). Kuroda (2011), suggesting that the prose portion of the Brahmasiddhi was composed some time after the composition of the verses, positively refers to Acharya’s claim. David (2013) footnote 32 (pp. 281–282) also approves of Acharya’s argument, and gives a detailed note about comparison with the four-fold division of the Brahmasūtra, as interpreted by Śaṅkara. Among the commentators, Ānandapūrṇa (Bhāvaśuddhi p. 149, ll. 7–9) clearly echoes this division when he introduces each chapter of the Brahmasiddhi.

  2. Vetter (1969), a German translation of the first chapter of the Brahmasiddhi, points out in its introduction (p. 16, ll. 20–29) that the first chapter is rather independent in its topic and not well connected with the following three chapters. Biardeau (1969), a French translation of the whole Brahmasiddhi, points out in the beginning note of the Tarkakāṇḍa (p. 192, note 1) that the content of the second chapter is comparable to the Tarkapāda (the first chapter of the first book) of the Śābarabhāṣya (the corresponding part related to our context could be Mīmāṃsāsūtra 1.1.4–5). According to that, while Śabara negates any possibility of conflict between direct perception and speech due to the difference of their scope, the former in visible objects and the latter in invisible objects, Maṇḍana would make a contrast between the two in terms of worldly truth and the ultimate reality. Her note implies that the name “Tarkakāṇḍa” is due to the contextual relation with the Tarkapāda.

  3. Auto-commentary on Brahmasiddhi III v.1 (p. 74, ll. 3–2): evaṃ pratyakṣādivirodhāt karmavidhivirodhāc ca śrutārthaparigrahe ye manyante vedāntānām upacaritārthatvaṃ tatpratibodhanāya vihitaḥ prayatnaḥ/ ‘Thus an effort has been made in order to awake those who think that the Upaniṣads have a figurative meaning, since if [we] accept (parigraha) a non-figurative meaning (śrutārtha), [they would] contradict [other means of knowledge] such as direct perception and also [they would] contradict the injunctions to [obligatorily perform ritual] activity (karmavidhi).’ The opponent claimed scripture would contradict the injunctions of ritual activity because they presuppose distinction of sādhya, sādhana, and itikartavyatā. Note that Maṇḍana enumerates the subject and purpose of each chapter in the very beginning of the Brahmasiddhi. See Vetter (1969, p. 36, footnotes 3–5); Bronkhorst (2007) pp. 53–57.

  4. In the rules of Mīmāṃsā cancellation is prescribed in various ways in accordance with different situations. In this paper I deal with cancellation in the sense “What follows cancels what precedes on the basis of prior-posteriority” (Mīmāṃsāsūtra 6.5.54) and also that in the sense “What precedes cancels what follows on the first-come basis” (Mīmāṃsāsūtra 3.3.2). The latter is similar to another cancellation prescribed in 3.3.14, which might be the most well-known rule applied to the relative strength of the six means of knowledge (pramāṇa) for the assignment of elements of a Vedic statement; but their territories are different. For the cancellation based on 3.3.14, see footnote 33.

  5. See Biardeau (1969, pp. 191–198) to grasp the whole argument and translation of this part.

  6. Though Maṇḍana keeps using the term pratyakṣādi ‘direct perception and the rest,’ the main target of the argument is direct perception. Therefore, from this part on, I clip ādi and use “direct perception” outside the translation, implying ordinary means of knowledge.

  7. The view “Scripture is stronger than direct perception etc.” is claimed also by Śaṅkarācārya, a possible contemporary of Maṇḍana, using the different reasonings that (1) it is free from mistakes because it has not been authored by any person; and (2) the metaphysical objects such as brahman cannot be grasped without the help of scripture. See Śaṅkara’s commentary on Brahmasūtra II 1.1.

  8. The sixth book of the Mīmāṃsāsūtra deals with “the qualification” (adhikāra), namely the relation between the result of a ritual activity and the one who enjoys it (bhoktṛ, adhikārin). The Adhikaraṇas 17–21 examine the expiation caused by the breaking of the chain of priests. For the general understanding of this argument, see Benson (2010, 6.5.17–6.5.19: p. 181, pp. 546–547).

  9. According to Ranade (2006, p. 242), this is called “a ritual procession” (prasarpaṇa) “taken out by the priests from the Sadas to the āstāva for chanting the Bahiṣpavamāna, participated by five or six priests and the sacrificer, led by the Adhvaryu, getting hold of one another’s garment; stealthily they proceed, bending their heads, licking lips behaving like a hunter in pursuit of a deer, since the sacrifice is a deer…” The hem of a lower garment is called kaccha, which is a crease made when the lower garment being folded is fixed on the part rolled up to the body. In our current context, this kaccha is the one on the back side of a priest.

  10. The sacrificer (yajamāna) is the one who enjoys the fruit of a ritual activity, and also he is the one onto whom the expiation is imposed. Priests are like cogwheels not in charge of any good or bad result. The content of expiations for the breaking of the chain by the Udgātṛ and Pratihartṛ priests are mentioned in the Śābarabhāṣya by the word adākṣiṇya for the former and sarvasva for the latter, which are interpreted by the commentators as ‘providing nothing as fee to the priests’ and ‘providing the whole of one’s property to them.’ Śābarabhāṣya on Mīmāṃsāsūtra 6.5.51 (p. 1475, ll. 25–26): yady udgātṛpratihartror yugapad apacchedo bhavati, tatra saṃdehaḥ — kim adākṣiṇyam, sarvasvaṃ veti bhavati vikalpaḥ, uta samuccaya iti/ ‘If the Udgātṛ and Pratihartṛ [priests] break [the chain of priests] at the same time, there is a doubt: does [one] optionally choose providing nothing or the whole of one’s property, or [does he] do the aggregation [of both expiations]?’ The description is based on the Brāhmaṇas, for example, Tāṇḍyamahābrāhmaṇa 6.7.13–15: pratihartā’pacchindyāt sarvave[d]asaṃ dadyāt/ yady udgātā’pacchindyād adakṣiṇaṃ taṃ yajñam iṣṭvā tena punar yajata/ tatra tad dadyād yat pūrvasmin dāsyan syāt/ “If the Pratihartṛ [priest] breaks [the chain], [the sacrificer] should donate all his property. If the Udgātṛ [priest] breaks [the chain], [the sacrificer] should perform the [same ritual] again after having performed that ritual without donation. In that case, [he] should donate what was supposed to be donated [after the second ritual].”

  11. Mīmāṃsāsūtra 6.5.51: tatra vipratiṣedhād vikalpaḥ syāt// ‘[Preliminary view (= Settled conclusion):] In that case (= when the Udgātṛ and Pratihartṛ priests let go the hems at the same time), there should be option [of one of the expiations] because of the contradiction [of two expiations].’ See section “Satisfaction of All Means Of Knowledge Is Impossible” of this paper for Mīmāṃsāsūtra 6.5.52–53 which follows the discussion above.

  12. See Jha (1934, pp. 1146—1149) to get the whole discussion and translation of the Śābarabhāṣya. Cf. Aṣṭādhyāyī 1.4.2: vipratiṣedhe paraṃ kāryaṃ// ‘When [two grammatical operations based on different rules] are in conflict, the grammatical operation [on the basis of] the latter [rule] is [applied].’ This rule is not about the actual time relationship like in the case of Mīmāṃsāsūtra 6.5.54, but similar principles (bādha, paratva, anavakāśatva) are in the scope. See Cardona (1970), Cardona (2013) (3.3.1) for the opinion which regards this rule as a meta-rule and for the criticism against it.

  13. Śabara uses in this context the word nimitta in the sense of what justifies the expiation, namely the breaking of the chain of priests, and naimittika in the sense of what is caused by the nimitta, namely the expiation. Expiations are included in “caused” ritual activity (naimittika) among the three categories of ritual activities, in addition to obligatory (nitya) and desiderative (kāmya) (Benson 2010, p. 34, ll. 20–33).

  14. When the two breakings take place one after the other, surely the cognition of the expiation caused by the first breaking arises and is present. Of this there is no doubt. We therefore have to think about whether another contradictory cognition can arise in such a situation.

  15. The expiation caused by the latter breaking can take place when the latter breaking alone happens. If there are two breakings in sequence, the result caused by the latter one is always weaker.

  16. Śābarabhāṣya on Mīmāṃsāsūtra 6.5.54 (p. 1476, ll. 20–24): pūrvasya balīyastvam, pūrvāpacchede yan naimittikaṃ prāptam, tasmin sati tadviruddhaṃ na śakyaṃ kartum/ na cāśakyam upadeśārhaṃ bhavati/ pūrvavijñānaṃ prāptam iti na saṃśayaḥ/ tasmāt tadavirodhenānyat kāryaṃ na viruddham/ kva tarhi tat syāt/ yatra kevalaṃ nimittam/ tasmāt pūrvavijñānaṃ balavat/

  17. I find neither a sūtra which explicitly mentions it nor a passage where it is discussed, but nonetheless this principle becomes manifest when we see arguments related to the Vedas as something we have to presuppose. I suspect that this is rather a meta-principle, which people agree on as far as they accept the validity of the Vedas.

  18. The principle “as far as you can” (yathā śakyate tathā), “according to your capacity” in other words, could be the same as the one which is applied to the performance of an obligatory rite (nitya) prescribed in Mīmāṃsāsūtra 6.3.1 onwards. But I did not find the exact passage which utilizes that rule for the performance of the enjoined activity in general. This principle is partly based on the presupposition of coherence of the Vedas. We need cancellation of the preceding cognition in order to overturn the impossibility of the expiation caused by the latter reason, because its arising is justified in scripture.

  19. In other words, the preceding cognition is already there, and now the subsequent cognition is coming into existence to contradict the first. In such a case, what brings about contradiction is not the preceding but the following cognition; therefore, the latter should be cancelled. This idea of putting predominance to the preceding one will get our attention later in this paper (section “Tantravārttika on Mīmāṃsāsūtra 3.3.2”).

  20. Here the text in the edition might be wrong. I interpret inserting na ‘no’ in the text, thinking the speaker changes here, but there might be a better solution.

  21. The latter cognition has not yet arisen and does not have a designable form at the time when the preceding cognition is present. One cannot judge such a cognition as contradictory or wrong. Kumārila explains this in concise and accurate words:

    Ṭupṭīkā (p. 1477, ll. 18–19): paraṃ tv anutpannatvenāsattvād aparyālocyatvād aviruddhatvād abādhyam/ ihaivaṃlakṣaṇo bādhaḥ/ ‘As for the following [cognition], since [it] has not yet arisen, [it] is not existent; therefore [it] is out of examination; and therefore [it is] not contradictory; and as a consequence [it] is not cancelled [by the preceding one]. In our current case, cancellation has such a characteristic.’

  22. Śābarabhāṣya on Mīmāṃsāsūtra 6.5.54 (p. 1476, l. 25–p. 1477, l. 4): paurvāparye pūrvadaurbalyaṃ syāt/ kutaḥ/ ākhyātena hi yo ’rthaḥ kartavya ity ucyate, tatraitad anubaddhaṃ yathā śakyate tatheti/ tat pūrvavijñānabādhena śakyate, nānyathā/ tena “pūrvavijñānaṃ bādhitvā idaṃ kartavyam” iti bhavati śabdārthaḥ/ nanu pūrvavijñāna idaṃ upapadyate, yad anyad virodhakaṃ vijñānaṃ bhaviṣyati tan mithyeti/ na (na: my emendation; omitted in the text), abhūtaṃ hi tan na śakyam āśrayitum “idaṃ nāma tat” iti/

  23. Śābarabhāṣya on Mīmāṃsāsūtra 6.5.54 (p. 1477, ll. 4–6): na cāpratiṣiddhe tasmin na pūrvavijñānaṃ saṃbhavati/ tasmād apratiṣiddhaṃ bhaviṣyat/ yadā tu tad bhavati, tadā pūrvavijñānaṃ bādhamānam evotpadyate/ tad idānīṃ bādhitaṃ na śaknoty uttaraṃ bādhitum iti/ ‘Furthermore, it is not the case that the preceding cognition cannot exist unless the [following cognition] is blocked [by the preceding one]. Therefore, [the preceding cognition] will arise without being blocked [by anything]. But when the [following cognition] arises, [it] arises only [under the condition of] cancelling the preceding cognition. The [preceding one] that is cancelled at this moment cannot cancel the following.’

  24. See Śābarabhāṣya on Mīmāṃsāsūtra 6.5.54 (p. 1477, ll. 6–9).

  25. Cf. Mīmāṃsākośa (vol. 2, p. 947): ānumānikānāṃ śrutyādīnāṃ pratyakṣaiḥ śrutyādibhir bādhaḥ, yathā śyenādau “śaramayaṃ barhiḥ” iti pratyakṣaśrutyā “barhiṣi havīṃṣy āsādayati” ity ānumānikaśruteḥ (kuśavidhāyikāyāḥ)/ ‘An explicit statement etc. obtained through inference is cancelled by the explicit statement etc. which is directly [stated]. For example, at [the context of] the Śyena sacrifice, the explicit statement [which prescribes kuśa grass] “[He] puts the oblation onto the sacrificial grass,” that has been obtained through inference, is cancelled by the explicit statement “The sacred grass is made of śara grass.”’

  26. See the Śābarabhāṣya on Mīmāṃsāsūtra 6.5.54 (p. 1477, ll. 10–13).

  27. This Opinion 3 is understood as Maṇḍana’s own idea by all the commentators.

  28. Maṇḍana intentionally uses the word tadapekṣa ‘which requires it, which is dependent on it.’ It is his unique idea when dealing with this rule. See “Dependence among Cognitions Justified by Experience”.

  29. Auto-commentary on Brahmasiddhi II v. 1 (p. 40, ll. 3–7): anyad darśanam — āmnāya eva balavāṃs tadvirodhe/ “paurvāparye pūrvadaurbalyaṃ prakṛtivat,” “pūrvābādhena notpattir uttarasya hi sidhyati” iti/ tathā hi — sarvasya nisargajaḥ pratyakṣādinibandhanaḥ kila vibhaktavastuparicchedaḥ/ tadapekṣas tu tatpūrvako ’nisargajaḥ kasya cid evāgantur advaitāvagamaḥ/ sa pūrvam anupamṛdyodetum aśaknuvaṃs tadapabādhātmodīyate/

  30. Note that for the word viparyaya, in my view Maṇḍana does not understand it in the sense of just ‘mistake’ but in the sense of ‘reversion’ in the form of “This is not X.” See the quotation of Sphoṭasiddhi v. 22 which comes later.

  31. sākṣādviparyayajñānāl laghvy eva tv apramāṇatā/ pūrvābādhena notpattir uttarasya hi sidhyati//

  32. The connection of this verse with Mīmāṃsāsūtra 6.5.54 is pointed out neither by Kumārila himself nor by the commentaries on the Ślokavārttika.

    Kataoka (2011, vol. 1, p. 266, footnote 229): “A previous cognition turns out to be false when another cognition arises which invalidates it, i.e. which proves that the object does not exist in the way that it was grasped by the previous cognition.” See Kataoka (2002) for a study on how Kumārila examined the validity of scripture and how the structure of cancellation is for him. Also according to Ishimura (2008), Kumārila has the same claim in the Ślokavārttika Vṛttikāragrantha vv. 5cd–6.

  33. Here Śaṅkhapāṇi (Vyākhyā p. 110, ll. 8–16) refers for the sake of comparison to the cancellation based on Mīmāṃsāsūtra 3.3.14: śrutiliṅgavākyaprakaraṇasthānasamākhyānāṃ samavāye pāradaurbalyam arthaviprakarṣāt// ‘When the explicit statement (śruti), indication (liṅga), sentence (vākya), context (prakaraṇa), place (sthāna), [and] name (samākhyā), are in conflict as being applicable, the latter is weaker, because it is more distant from the object.’ Here ‘latter’ means ‘latter in this list of pramāṇas,’ and the list is on the basis of how immediately (or directly) they function. For this rule, see Bronkhorst (1997) and Gajendragadkar and Karmakar (1934) (the translation and detailed notes of the Arthasaṃgraha); see also Yoshimizu (1987) for assignment by means of an indication. According to Śaṅkhapāṇi, in the case of these six pramāṇas beginning with the explicit statement, the former is stronger due to the difference of whether it is already present or not that is based on its immediacy (śīghrabhāvitva); and because of this, for example, the explicit statement, already present, cancels the indication which is not yet present. As we saw in Śabara’s response, it is difficult to justify cancellation of something just by the fact that it is not present (aprāpta). Therefore, it is plausible that immediacy is more emphasized by Śaṅkhapāṇi. And in such a case, the latter cannot be established so that prior-posterior relation does not take place.

  34. The same discussion is found in the Bhāmatī, Vācaspati’s sub-commentary on Śaṅkara’s commentary on the Brahmasūtra, which appears to be based on Maṇḍana’s discussion (See Shima (2012) pp. 63–65). In the Bhāmatī, Mīmāṃsāsūtra 6.5.54 and Kumārila’s Tantravārttika that will be explained later are quoted in continuity. Cf. Bhāmatī (p. 10, l. 5–p. 11, l. 3): jyeṣṭhatvaṃ cānapekṣitasya bādhyatve hetuḥ, na bādhakatve, rajatajñānasya jyāyasaḥ śuktijñānena kanīyasā bādhadarśanāt, tadanapabādhane tadapabādhātmanas tasyotpatter anutpatteḥ/ darśitaṃ ca tāttvikapramāṇabhāvasyānapekṣitatvam/ tathā ca pāramarṣaṃ sūtram — “paurvāparye pūrvadaurbalyaṃ prakṛtivat” iti/ tathā “pūrvāt parabalīyastvaṃ tatra nāma pratīyatām/ anyonyanirapekṣāṇāṃ yatra janma dhiyāṃ bhavet//” iti/ ‘Furthermore, priority of something which is not dependent [on others] is the ground for [its] being cancelled, but not the ground for [its] cancelling [the others]. This is because it is actually seen that the preceding cognition of silver is cancelled by the following cognition of a shell. For, the arising of the [cognition of a shell], since [it] consists in cancellation of the [cognition of silver], cannot arise without cancelling it. Also it has been already shown that [scripture] which has the ultimate validity is independent [from any other means of knowledge]. And in the same way, there is an aphorism composed by a great sage[, Jaimini] — “When there is a prior-posterior relation [between the two triggers], the preceding one is weaker, just as in the case of an archetype ritual” (Mīmāṃsāsūtra 6.5.54). Also “That the latter is stronger than the former is to be understood when the cognitions occur without depending on each other” (Tantravārttika 3.3.2.2).’

  35. This example means that we easily misunderstand the trees standing at a distant place as the elephant (or anything else). The false cognition of elephant comes first, and thereafter it is corrected by the right cognition of trees. To be more precise, at that moment we realize “What I see is actually not the elephant but the trees!” Therefore Maṇḍana claims that the subsequent cognition is stronger than the preceding one. Similar examples are seen between a rope and snake, and between a shell and silver.

  36. We cannot say that the right cognition of trees occurs just because of proximity. If that were the case, anyone who stands at a particular point should apprehend the trees, but in reality some understand them properly even from more distance, and some cannot understand them even from more proximity.

  37. Auto-commentary on Brahmasiddhi II v. 1 (p. 41, ll. 4–10): dṛṣṭaṃ ca hetutvenāpekṣitāyā api pūrvasyāḥ pratipatteḥ parasyāḥ pratipatter balīyastvam/ yathā dūrastheṣu vanaspatiṣu hastipratipattibhyo vanaspatipratipatteḥ/ apekṣitā hi hastipratipattayo vyaktavanaspatipratipattyā hetutvena/ na tasyā indriyārthasaṃnikarṣamātrāj janma, āpāte ’bhāvāt/ na ca deśaviśeṣāt, taddeśasthasyaivotpatteḥ/ tasmāt purovartiṣu vanaspatiṣu praṇihitamanasaḥ prācyaviparyāsānugatamatisaṃskārasacivendriyādisaṃyogakāritā seti mantavyam/

  38. See Thrasher (1993) Chapter 5: Error Leads to Truth. He refers exactly to this passage in sections 5.7 and 5.8 (p. 106, l. 18–p. 107, l. 21).

  39. Sphoṭasiddhi v. 19: ārūpālociteṣv asti hy anyathātvaprakāśanam/ tatsaṃskārakramāc cāpi vyaktaṃ tattvaṃ prakāśate// ‘Indeed, when [things] are perceived vaguely, they appear other [than they are]. Nevertheless, [their] true nature appears clearly through the gradual [intensification] of the latent impressions deposited by the [different apprehensions].’ In the auto-commentary on this verse (p. 139, l. 4–p. 142, l. 1) Maṇḍana sets the example of the trees and the elephant. Another example of number, which I do not quote in this paper, is also used in both the Brahmasiddhi and the Sphoṭasiddhi. See Thrasher (1993) sections 5.9 and 5.10 (p. 107, l. 23–p. 108, l. 34). See also Sastri (1937) Introduction (I; p. xxvi–xxvii).

  40. Sphoṭasiddhi v. 22: viparyāso nimittaṃ ca samyagbodhasya darśitaḥ/ kiṃcidbhedānukāre ’pi dṛṣṭa eva viparyayaḥ//

  41. Vākyapadīya I v. 86: bhedānukāro jñānasya vācaś copaplavo dhruvaḥ/ kramopasṛṣṭarūpā vāg jñānaṃ jñeyavyapāśrayam// ‘Constant fluctuation of cognition and speech exists [because of its] imitation of distinction (bhedānukāra) [which those cognized entities and sounds have]. [By the imitation of distinction,] speech is in form disturbed by the sequence, [and] cognition is dependent on cognized entities.’ Sphoṭasiddhi v. 22 is indeed a combination of Ślokavārttika Codanāsūtra v. 57 and Vākyapadīya I v. 86.

  42. The idea “This is not silver” (nedaṃ rajatam), which is the same as “This is a shell” (iyaṃ śuktikā), cannot exist without the cognition “This is silver” (idaṃ rajatam). It could be useful to note here that in the Niyogakāṇḍa of the Brahmasiddhi, Maṇḍana states that “cessation of the knowledge of silver” is the same as “arising of the knowledge of a shell.” Auto-commentary on Brahmasiddhi III vv. 105–106 (p. 121, l. 21–p. 122, l. 3): na cānyatvam, yato ’vidyāpagama evoktena prakāreṇa muktiḥ/ avidyā saṃsāraḥ; vidyaiva cāvidyānivṛttir yady agrahaṇam avidyā, yato bhāva evābhāvavyāvṛttiḥ; atha viparyāsaḥ, tathāpi virodhitattvajñānodaya eva tannivṛttiḥ; na hi śuktikājñānotpādād anyā rajatajñānanivṛttis tatsādhyā prayatnāntarasādhyā vā, tayor yaugapadyāt prayatnāntarānapekṣaṇāc ca/ ‘Furthermore, [liberation is] not something other [than true knowledge (vidyā)], because liberation is nothing but the cessation of nescience according to the manner that has been already told. Nescience is transmigration; and if nescience [means] “non-grasping,” its cessation is exactly true knowledge itself, because exclusion from non-existence is nothing but existence. If [nescience means] an erroneous cognition, nonetheless its cessation is exactly the arising of the knowledge of truth that is contradictory [to that]. For, the cessation of the knowledge of silver is nothing other than the arising of the knowledge of a shell: [it is] neither something established by such [knowledge of a shell] nor that established by any other effort. This is because they (the cessation of the knowledge of silver and the arising of the knowledge of shell) occur at the same time, and [they] do not require any other effort.’

  43. Iyer (1966, p. 59, ll. 13–19): “G. (Gopālikā) explains that when an error takes place, it is later sublated by the correct cognition. It can take a negative or a positive form. In the case of the mother-of-pearl being mistaken for silver, the later sublation takes the form “this is not silver” i.e. it is negative in form; if it takes the form “this is mother-of-pearl,” it would be positive in form. In the case of the word, the later sublation is of a positive character.” The Gopālikā is Ṛṣiputraparameśvara’s commentary on the Sphoṭasiddhi.

  44. Vākyapadīya I vv. 89–90: yathaiva darśanaiḥ pūrvair dūrāt santamase ’pi vā/ anyathākṛtya viṣayam anyathaiva adhyavasyati// vyajyamāne tathā vākye vākyābhivyaktihetubhiḥ/ bhāgāvagraharūpeṇa pūrvaṃ buddhiḥ pravartate// ‘Just as when [the object is] in a distant place or in a dark place, one identifies [it] completely otherwise after taking the object differently by means of previous experiences. In the same way, when the sentence is being manifested by the causes that manifest the sentence, initially the [unreal] cognition occurs in a form that is separated into parts.’

  45. Vākyapadīya II v. 238: upāyāḥ śikṣamāṇānāṃ bālānām upalālanāḥ/ asatye vartmani sthitvā tataḥ satyaṃ samīhate// ‘Evasion is the means for children who are learning. [One] strives for the truth after having stood on the false path.’ Note that I chose the reading upalālanāḥ (supported by a manuscript according to Rau (1977) p. 80) instead of upalāpanāḥ in Iyer’s edition.

  46. Auto-commentary on Brahmasiddhi II v. 1 (p. 41, ll. 14–16): yata upāyam ākṣipati pratipattiḥ; nopāyasya paramārthatām, mithyājñānād api tattvapratipatteḥ/ vyāvahārikaṃ ca bhedasya satyatvam iṣṭam eveti/ ‘This is because the cognition allows you to understand (ākṣip) [its] means; [but] does not [allow you to presuppose] that the means [exists] from the viewpoint of the ultimate truth, because the cognition of the reality [occurs] even from the erroneous cognition. And [everyone] accepts that distinction has the empirical reality (vyāvahārikaṃ satyatvam).’ Cf. Sastri (1937) in his introduction mentions (p. xxxi, ll. 7–16): “Maṇḍana’s interpretation of the Īśāvāsya text “vidyāṃ cāvidyāṃ ca…” is entirely different from Śaṃkara’s interpretation of it […]. Maṇḍana explains this mantra in two ways; in his first explanation, the first half is taken to refer to the association of avidyā and vidyā as the means and the end (upāyopeyabhāvāt sahite) and the second half states the fact that, when a person removes his avidyā by avidyā, his self-realisation follows; […].”

  47. This is the issue related to reason 3 among the five reasons held by Opinion 1 (see the list at the beginning of section “Cancellation of Direct Perception, etc., by Scripture”), namely that scripture has room for different interpretations (avakāśa). Here the opponents in the Brahmasiddhi claim that we could interpret what is taught by the Upaniṣads, namely non-duality, not literally (mukhya) but figuratively (upacarita). This view and Maṇḍana’s answer follow the discussion in Mīmāṃsāsūtra 1.2.31–40, where the settled conclusion is that the meaning of expression is not different in ordinary speech and in the Veda. For this principle, see Harikai (1973, 1990, pp. 73–85, 262–269).

  48. Mīmāṃsāsūtra 6.5.52–53: prayogāntare vobhayānugrahaḥ syāt// na caikasaṃyogāt// ‘[Second view:] Rather (vā), both [expiations] are to be satisfied (or “assist mutually”?) in different performance[s]. [Settled conclusion:] But it is not the case (na ca), because [both expiations are] connected to one and the same [performance of the ritual].’

  49. See footnote 11.

  50. An arbitrary application is better to be avoided unless absolutely necessary, because due to choosing either one of prescriptions, the other would lose its validity. Śābarabhāṣya on Mīmāṃsāsūtra 6.5.52 (p. 1476, ll. 5–6): vikalpe hi pakṣe bādhaḥ, tasmāt samuccayaḥ/ ‘For, if [one option] is arbitrarily chosen, [there would be] cancellation of [the other] position. Therefore, the aggregation [of both options should be performed].’

  51. “If the Pratihartṛ [priest] breaks [the chain], [the sacrificer] should donate all his property. If the Udgātṛ [priest] breaks [the chain], [the sacrificer] should perform the [same] ritual again after having performed that ritual without donation.” See footnote 10.

  52. Śābarabhāṣya on Mīmāṃsāsūtra 6.5.52 (p. 1476, ll. 6–10): ucyate, prayogāntare vobhayānugrahaḥ syāt/ “tena punar yajeta” ity ucyate/ dvis tasya prayogaḥ/ tatraikasmin prayoge, ekaḥ kalpaḥ/ anyasminn aparo bhaviṣyati/ evam avirodhaḥ/ tasmād ubhayaṃ prāyaścittam iti/

  53. Śābarabhāṣya on Mīmāṃsāsūtra 6.5.53 (p. 1476, ll. 12–16): “ekasaṃyogāt”/ sa eva yāgaḥ punaḥ kriyeta/ yady anyatareṇa vinā, vaiguṇyam ity avadhāryeta, sarvasve kriyamāṇe ’dākṣiṇyābhāvād viguṇaḥ syāt/ nānyasmin prayoge kriyamāṇe ’nyaḥ prayogo “anugṛhyate”/ na cādākṣiṇyasya sarvasvadānasya ca prādhānyam/ guṇavatā prayogeṇa karma saṃbandhayitavyam/ tasmād vikalpaḥ/ ‘[Settled conclusion:] [No,] “because [both expiations are] connected to one and the same [performance of the ritual].” The same sacrifice should be performed again. If [you] think that “There is deficiency (vaiguṇya) if [the sacrifice is performed] without one [expiation],” at the time of performing “[the donation of] all the property,” there is no “non-donation”; [and accordingly] [the sacrifice] would be deficient. The performance of one [sacrifice] is not satisfied (anugrah) when another [sacrifice] is being performed. Furthermore, “non-donation” and “donation of all the property” are not predominant [in the sacrifice]. A ritual activity should be connected to the performance possessing the subsidiaries [including the expiation]. Therefore, [either one of the expiations] is arbitrarily applied.’

  54. None of the commentators (from this part on Vācaspati’s commentary Tattvasamīkṣā is available) does refer to 6.5.52–53. But since the word ubhayānugraha appears in the same context of the breaking of the chain of priests, in this paper I illustrate Maṇḍana’s discussion as being related to them.

  55. Remember that Mīmāṃsāsūtra 6.5.52–53 is under the condition where both priests break the chain at the same time.

  56. Auto-commentary on Brahmasiddhi II v. 1 (p. 42, ll. 9–10): atha matam — ubhayānugrahād evaṃ kalpyate/ evaṃ hi śabdaḥ pratyakṣādayaś cānugṛhītā bhavanti/ anyathā pratyakṣādayo ’tyantam apabādhitāḥ syuḥ/ The reading is ’tyantam apabādhitāḥ is ’tyantam eva bādhitaḥ in the edition but I follow the reading of Acharya (2006, p. 29, l. 12).

  57. The opposite case does not take place. We never see that direct perception makes understood an object figuratively by reducing its function. Note that this opinion is against the preceding argument of Maṇḍana that “Scripture’s validity is only from the viewpoint of the ultimate truth, and therefore it loses its validity if it is cancelled by the other means of knowledge; while direct perception and the rest can keep the worldly validity even though they are cancelled by scripture.” There Maṇḍana claimed that direct perception can survive at the worldly level even if its ultimate validity has to be given up.

  58. Auto-commentary on Brahmasiddhi II v. 1 (p. 42, ll. 11–14): naitat sāram/ pramāṇasāmarthyānusāreṇa prameye kā cit kalpanā syāt/ pramāṇe tu svarasasiddhaṃ parityajya kalpanāyāṃ na nibandhanam asti/ tasmāt pramāṇānusāreṇa vikalpasamuccayavyavasthāḥ kalpyante, nobhayānugrahāt/ nobhayam anugṛhītavyam iti pramāṇam asti/ ‘[Response:] It is not correct. For the object of the valid means of knowledge (prameya: cognition), there can be some postulation in accordance with the power of the means of knowledge. For the means of knowledge (pramāṇa) [itself], however, there is no ground for postulating [something else] having abandoned what has been established naturally. Therefore [either] arbitrary, aggregate, or respective (separate) application, are postulated [if needed] in accordance with [the power of] the means of knowledge; [but they] are not [postulated] on the basis of satisfying both [scripture and direct perception etc.]. There is no proof for the idea that both are to be satisfied.’

  59. Śaṅkhapāṇi (Vyākhyā p. 120, ll. 20–26) enumerates the situations where the object of cognition (= content of cognition) is arbitrarily, aggregately, or separately established. For example, both of the two prescriptions “[One] should sacrifice with rice grains” (vrīhibhir yajeta) and “[One] should sacrifice with barley grains” (yavair yajeta) are the prescriptions which restrict the oblation to rice or barley grains. They are equivalent and contradictory, and it is impossible to follow the two at the same time. If either one of them is refuted, the validity of scripture would be spoilt. Therefore, we understand that arbitrary application of them is the only way that would not damage the validity of two prescriptions. Also in the prescription “The one desirous of heaven should sacrifice with the new- and full-moon sacrifices,” we cannot think that the new-moon sacrifice is enough. The explicit statement of the Dvaṃdva-compound and the dual case-ending in darśapūrṇamāsābhyām makes us understand that aggregate application is necessary. In the case of the oblation after or before sunrise too they are incompatible; but it is understood that they are separately applied according to the schools. In this way, these three examples show different applications of the objects which are taught by scripture; but they are not for scripture itself. Namely, for example, we cannot interpret one of the restricting prescriptions figuratively in order to satisfy both of them.

  60. As I noted briefly in footnote 47, before the discussion on 6.5.52–53 Maṇḍana gives another claim based on Mīmāṃsāsūtra 1.2.40, which is also about the non-figurativeness of scripture.

  61. The “sentence” here includes the collection of smaller sentences which have the same purpose (ekārtha = ekaprayojana) on the basis of Mīmāṃsāsūtra 2.1.46: arthaikatvād ekaṃ vākyaṃ sākāṅkṣaṃ ced vibhāge syāt// ‘[Condition 1:] If the purpose is the same, [and] [Condition 2:] if [the parts] expect [each other] once it is divided, it is a single sentence.’ On 2.1.46 see Fujii (2014), Kawakami (1994), and Jha (1942, p. 189, ll. 1–18). Harikai (1990, pp. 37–41) discusses “unity of a sentence” (ekavākyatā) between the explanatory and prescriptive statements.

  62. Śatapathabrāhmaṇa 11.5.8.3: prajāpatir akāmayata prajāḥ sṛjeyeti/ sa tapo ’tapyata/ tasmāt tepānāt trayo devā asṛjyanta, agnir vāyur ādityaḥ/ te tapo ’tapyanta/ tebhyas tepānebhyas trayo vedā ’sṛjyanta/ agner ṛgvedo vāyunā yajurveda ādityāt sāmavedaḥ/ ‘Prajāpati desired: “I want to create offspring.” He did a penance. From him who finished a penance, three deities were created: Agni, Vāyu, and Āditya. They did a penance. From them who finished a penance, three Vedas were created: the Ṛgveda from Agni, the Yajurveda from Vāyu, and the Sāmaveda from Āditya.’

  63. Maitrāyaṇī Saṃhitā 3.6.5: uccair ṛcā kriyate, uccais sāmnā, upāṃśu yajuṣā/ ‘One performs with ṛc, [recited] loudly; with sāman, [recited] loudly; with yajus, [recited] in a whisper.’

  64. Yoshimizu (2013) investigates this Adhikaraṇa questioning why two different texts, the Śatapathabrāhmaṇa and Maitrāyaṇī Saṃhitā, are regarded as a single sentence. The Maitrāyaṇī Saṃhitā refers to a verse belonging to the Ṛgveda as a spell to be recited during the ritual called Audgarbhana, which is performed by the Yajurveda priests. Since the Maitrāyaṇī Saṃhitā itself regards the verse as ṛc, namely as belonging to the Ṛgveda, if we take it naturally, the Yajurveda priests have to recite it in the manner of the Ṛgveda. To avoid this the Mīmāṃsakas are forced to think that the text is connected to the Śatapathabrāhmaṇa and therefore the word ṛc should not be taken literally. Mīmāṃsāsūtra 3.3.2 exists in this context. Yoshimizu points out that there is a priority of the Yajurveda in recitation and it is found in other Adhikaraṇas.

  65. Śabara rephrases jāta as jāti, but in the Tantravārttika this identity is criticized from the viewpoint of the opponent. Although most probably Śabara did not intend a strict sense in the word jāti, Kumārila claimed that jāta cannot be taken as jāti ‘generality.’ For example, a forest (vana) which is an aggregate of trees does not have “the property of being a forest” (vanatva); also there is no “property of being a word gauḥ” (gośabdatva) in a word gauḥ that is an aggregate of phonemes. In the same way, the verse (ṛc), prose formula (yajus), and song (sāman), are the aggregates of pāda, pada, and parvan, and they do not have the generality such as “the property of being a verse.” What the preliminary view claims here according to Kumārila is that the singular-form word “ṛcā” expresses ‘the verse’ as the aggregate. As Yoshimizu (2013) explains, the point is that in this case, such a verse still belongs to the Ṛgveda even though it is quoted in the other Vedas, the Yajurveda for instance. Therefore, the Yajurveda priests have to recite it loudly in accordance of the manner of the Ṛgveda priests.

  66. If we rephrase this question according to the issue of assignment by means of sentence: is a property like “loudness” assigned to “the Ṛgveda” or “the verse” on the basis of the power of a sentence? Śābarabhāṣya on Mīmāṃsāsūtra 3.3.1 (p. 803, ll. 3–5): jyotiṣṭome śrūyate “uccair ṛcā kriyate, uccaiḥ sāmnā, upāṃśu yajuṣā” iti/ tatra saṃdehaḥ/ kim ṛgādijātim adhikṛtya ete śabdāḥ pravṛttā uta vedam adhikṛyeti/ ‘In the [context of] the Jyotiṣṭoma sacrifice it is taught that “One performs with ṛc, [recited] loudly; with sāman, [recited] loudly; with yajus, [recited] in a whisper.” On that [topic] (tatra), there is a doubt: Are these words (ṛc etc.) uttered having in scope (adhikṛ) a group (jāti) like ‘verse’; or having in view a [corresponding] ‘Veda’ (an individual Veda)?’ According to Yoshimizu, in the settled conclusion in which ṛc expresses ‘the Ṛgveda,’ a verse quoted in the Yajurveda no more belongs to the Ṛgveda. Therefore, the Yajurveda priests should recite it in a whisper, as in the case of prose formula (yajus).

  67. The same question is raised also for the words yajus and yajurveda, and sāman and sāmaveda; but in my discussion I focus only on ṛc and ṛgveda.

  68. See Tantravārttika on Mīmāṃsāsūtra 3.3.1 (p. 803, ll. 15–17). It is an important issue regarding the unity of a sentence why such doubt of the word-meaning arises between the prescriptive and explanatory statements. On this account, Mīmāṃsāsūtra 1.4.29 deals with the relative strength in the case where the prescriptive statement that precedes refers to the generality (sāmānya) and the explanatory statement that follows refers to a particular aspect (viśeṣa).

  69. Mīmāṃsāsūtra 3.3.1–2: śruter jātādhikāraḥ syāt// vede vā prāyadarśanāt// ‘[Preliminary view:] [The words such as ṛc] have a group as [their] scope, because [they are] explicitly stated [in the prescription]. [Settled conclusion:] Rather, [they are stated] in the sense of the Vedas, because [they] are seen in the beginning.’ This Adhikaraṇa continues up to 3.3.8; and from 3.3.3 on the other reasons are listed which justify the settled conclusion. In this paper, however, we do not touch them because our purpose is the Tantravārttika on 3.3.2. See Jha (1933, pp. 437–440) to get the whole argument of this Adhikaraṇa. See Benson (2010) 3.3.1 too.

  70. See Śābarabhāṣya on Mīmāṃsāsūtra 3.3.1 (p. 803, l. 5–p. 804, l. 3). This preliminary view is at a glance similar to the settled conclusion in 6.5.54; but the difference is the ground for the argument, which is “the explicit statement in the prescriptive statement.”

  71. See Śābarabhāṣya on Mīmāṃsāsūtra 3.3.2 (p. 805, l. 4–p. 807, l. 3).

  72. Tantravārttika on Mīmāṃsāsūtra 3.3.2 (p. 806, ll. 12–14): ye hi vidhyuddeśāt parastād arthavādāḥ śrūyante teṣām asti daurbalyam/ ye purastāc chrūyante te mukhyatvād balīyāṃso bhavanti/ sarvasyaiva hi śabdasya svārthātilaṅghane virodhaḥ kāraṇam/ sa ca pratiyogini dṛṣṭe virodho vijñāyate/ Note that I do not translate hi in the first sentence to avoid further explanation.

  73. See Tantravārttika on Mīmāṃsāsūtra 3.3.2 (p. 806, ll. 14–17).

  74. This reasoning “what comes first is stronger” which has appeared several times so far, according to Kumārila, is based on Mīmāṃsāsūtra 12.2.23: mukhyaṃ vai pūrvacodanāl lokavat// ‘Rather, the prior/first one [is satisfied], because [it is] prescribed/mentioned first, as is the case in the world.’ Note that the rules of the Mīmāṃsāsūtra are not irreversible. Note also that when the prescriptive statement comes first, the situation changes slightly: in that case, another rule 1.4.29 plays a role; and furthermore, the definite content of the prescriptive statement too becomes subject to debate.

  75. Tantravārttika 3.3.2.2 (Tantravārttika v. 2 on Mīmāṃsāsūtra 3.3.2): paurvāparyabalīyastvaṃ tatra nāma pratīyate/ anyonyanirapekṣāṇāṃ yatra janma dhiyāṃ bhavet//

  76. Here Kumārila states explicitly the two conditions for the cancellation to be applied, which we saw in the discussion on Mīmāṃsāsūtra 6.5.54. Tantravārttika on Mīmāṃsāsūtra 3.3.2 (p. 806, ll. 23–25): ye hi bhinnavākyagatāḥ parasparanirapekṣā virodhino ’rthāḥ śrūyante teṣāṃ pūrvasyottarānupamardenaiva labdhātmakatvād uttarasya pūrvopamardena vinātmalābhānupapatteḥ paurvāparyabalīyastvanyāyo bhavati/ ‘For, if [we] hear the objects which are existing in the different sentences [and] which are not dependent on each other [and] which are contradictory [to each other], for them there is the principle of relative strength that exists in the successive [elements] on the basis of that [Condition 1:] The preceding one comes into existence just without cancelling the following one; [and] [Condition 2:] The following one cannot arise unless [it] cancels the preceding.’

  77. On the basis of Mīmāṃsāsūtra 2.1.46 (See footnote 61), the explanatory and prescriptive statements become a big single sentence beyond each smaller sentence due to “the expectation” arising between the word-objects of those statements. The same thing happens when the prescriptive statement is heard first. See Raja (1963, pp. 151–164) for this ākāṅkṣā.

  78. These two anuguṇa and adhika could be related to what we find in Vākyapadīya II v. 42, and v. 44, where Bhartṛhari introduces several views on the sentence object. A scholar who holds the view that the sentence object is “the supplement” (ādhikya) of the word objects is Patañjali on the basis of his statement in the Mahābhāṣya on Aṣṭādhyāyī 2.3.46; while for the view that the sentence object is “the conformity” (ānuguṇya) to the particularities, Bhartṛhari gives no clue as to who claimed this. Puṇyarāja, a commentator of the second book of the Vākyapadīya, attributes these two views respectively to the abhihitānvaya and anvitābhidhāna views, but it might be an impetuous speculation. Most probably these two had been widely known as the elements to explain the sentence object, and Kumārila took them over as the factors of ākāṅkṣā. Cf. Vākyapadīya II v. 42: sambandhe sati yat tv anyad ādhikyam upajāyate/ vākyārtham eva taṃ prāhur anekapadasaṃśrayam// ‘[They] say: As for [the object] other [than the word object], that is a supplement (ādhikya) which arises when there is a relation [among the words in a sentence]; such [a supplement] is the sentence object that resides in more than one word.’ II v. 44: sarvabhedānuguṇyaṃ tu sāmānyam apare viduḥ/ tad arthāntarasaṃsargād bhajate bhedarūpatām// ‘On the other hand, other [scholars] consider [as follows]: [Some] common property is the conformity (ānuguṇya) to all the particularities. Such [a common property] gets the nature of a particular (= specific sentence object) on the basis of the association with the other object.’ See David (2017) for the definition of the sentence discussed by the Grammarians and Bhartṛhari (and the Mīmāṃsakas).

  79. Kumārila has the same idea as Śabara that cancellation of X is possible only when X is existent. We saw a similar discussion in Mīmāṃsāsūtra 6.5.54.

  80. Tantravārttika on Mīmāṃsāsūtra 3.3.2 (p. 806, l. 25–p. 807, l. 11): yatra tv ekavākyatayā parasparāpekṣāṇāṃ śravaṇaṃ na tatraitad bhavati/ tatra hy ākāṅkṣāvaśenaikavākyatā bhavati/ pūrvapratīte cottaram ātmānuguṇam adhikaṃ vākāṅkṣati/ ataś ca yad anadhikam ananuguṇaṃ vā tad anapekṣitatvān naiva tasmin vākye ’sti/ na cāvidyamānena pūrvabādhaḥ śakyata iti pūrvabalīyastvaṃ bhavati/

  81. According to Bhartṛhari, ānuguṇya is suitability (yogyatā). There is conformity in general which has the potential to assimilate itself into all the particular word objects. And once it is particularized, it becomes a particular sentence object. See the auto-commentary on Vākyapadīya II v. 44.

  82. For ādhikya, according to Bhartṛhari there are vākyādhikya and padādhikya. Vākyādhikya is explained as the additional element which is beyond each word object. In “vīraḥ puruṣaḥ,” the coreference (sāmānādhikaraṇya) based on their relation of qualifier and qualified is the supplement. Padādhikya is explained as the specificity (vyatireka) given to each single word, a kāraka relationship in particular. See the auto-commentary on Vākyapadīya II v. 42. I do not find so far any further explanation by Kumārila himself on this issue. It should be examined more in the future.

  83. See Tantravārttika on Mīmāṃsāsūtra 3.3.2 (p. 807, ll. 11–17). What is meant is that a thing which is not expected (anapekṣita) is not an element of a sentence. What is not an element of a sentence is not intended (avivakṣita); and therefore it does not exist in that sentence. Here I took “intention” (vivakṣā) as the intention of the Veda, fearing that taking a person into account might come into conflict with the non-humanmadeness (apauruṣeyatva) of the Vedas, but it is not precise enough. See Yoshimizu (2008) for Kumārila’s three kinds of interpretations of vivakṣā.

  84. The Vedas have no mistake and the relation between the words and their objects is eternally established. See Mīmāṃsāsūtra 1.1.5, Jha (1942, pp. 174–178).

  85. Auto-commentary on Brahmasiddhi II v. 1 (p. 43, ll. 6–11): na ca tadanusāreṇārthakalpanām arhati, anapekṣatvāt/ na tadadhīnam asya pramāṇatvam/ na taiḥ saṃbhūyakāritvam, nāmnāyasya pratyakṣādīnāṃ caikavākyabhāvaḥ/ parasparānapekṣo hy ubhayatra buddhyutpādaḥ/ tatra parabalīyastvaṃ nyāyyam/ yathoktam — “pūrvāt parabalīyastvaṃ tatra nāma pratīyatām/ anyonyanirapekṣāṇāṃ yatra janma dhiyāṃ bhavet//” There is a variant reading: the first line of Tantravārttika 3.3.2.2 is paurvāparyabalīyastvaṃ tatra nāma pratīyate. The reading in the Brahmasiddhi is shared by the Bhāmatī (see footnote 34 above). The idea is the same, and I keep the reading of the edition in this paper.

  86. It is not always appropriate to translate the word ekavākyatā as ‘the unity of a sentence,’ because as in our current case, this concept is used also to refer to the mutual agreement in general. In the relation between direct perception and scripture there is no issue of “sentence” or “language” but Maṇḍana calls it ekavākyatā focusing on their cooperative interaction. For the usage of ekavākyatā, see David (2020) n. 359 (p. 387) and n. 360 (pp. 387–388).

  87. Kataoka points out in footnote 228 that the passage of the Bṛhaṭṭīkā is found in Śāntarakṣita’s Tattvasaṃgraha v. 2865, where Kumārila adds the condition “anapekṣapramāṇatvāt.”

  88. Kataoka (2011, p. 266, footnote 228): “In the Ślokavārttika, Kumārila describes the simple fact that an invalidating cognition never arises without it setting aside a previous invalid cognition. In the Bṛhaṭṭīkā, however, he explains the reason this is possible by pointing out the independence of an invalidating cognition.”

  89. This is the first reason among the five reasons held by Opinion 1 which claims direct perception is stronger. See the list at the beginning of section “Cancellation of Direct Perception, etc., by Scripture”.

  90. Though Maṇḍana does not explain it in detail, in the example of trees and an elephant, he said that direct perception is caused by the contact (saṃnikarṣa) between the sense-organs and the object. According to Citsukha, conceptual perception is dependent on non-conceptual perception. Abhiprāyaprakāśikā (p. 156, ll. 22–24): anumānādeḥ pratyakṣāpekṣatvāt, pratyakṣasya ca savikalpakasya nirvikalpāpekṣatvād ity āha — pratyakṣādayo ’pīti/ ‘This is because inference and the rest are dependent on direct perception; and for direct perception too, conceptual [perception] is dependent on [unconceptual]. Thus [Maṇḍana] says — “[It is] also [not the case that] direct perception etc…”’

  91. Auto-commentary on Brahmasiddhi II v. 1 (p. 40, l. 24–p. 41, l. 1): na pramitāv apekṣāvattā śabdasya pratyakṣādiṣu, kiṃ tu svarūpasiddhau/ anyathā pramāṇam eva na syāt/ tathā ca svakārye ’napekṣatvān na pratyakṣādibhyo hīyate/ pratyakṣādayo ’pi svarūpasiddhau nānyānapekṣāḥ/ kārye hi sāpekṣatvaṃ sāmarthyam apakarṣatīti/

  92. In Fig. 1, the solid line shows dependence on the arrowed object, while the dotted line shows independence from each side. Though the expressions differ, on “kāraṇāpekṣatā,” svarūpasiddhau (= ātmalābhe) is equal to hetutvena/upāyatvena; and on “svakāryeṣu nirpekṣatā,” buddhyutpāde is equal to pramitau (= svakārye). The relation between direct perception and scripture and that between the cognition of an elephant and of trees do not perfectly match since the former is the means of knowledge while the latter is cognition. But we can assume that for Maṇḍana, in both cases it is the cognition that has two aspects, dependence and independence. In other words, between the cognitions of bheda and abheda too, there exists the relation of dependence of the latter on the former through their means of knowledge (I do not show this in the figure though). This is implied by Maṇḍana in his statement “tadapekṣas tu…” that I quoted in the beginning of section “Dependence among Cognitions Justified by Experience”. Note that “the cognition of phonemes” and “that of sphoṭa” in Maṇḍana’s sphoṭa theory are comparable to the cognitions of an elephant and of trees. Remember that for Maṇḍana the sphoṭa theory is the theory of how to perceive speech, but not the theory of how to understand its meaning.

  93. atmalābhe hi bhāvānāṃ kāraṇāpekṣatā bhavet/ labdhātmanāṃ svakāryeṣu pravṛttiḥ svayam eva tu// See Kataoka (2011, pp. 250–254) to get the whole context of the discussion around this.

  94. Kataoka points out in his footnote 207 “in which Kumārila emphasizes that cognition operates innately (pramāṇaṃ svataḥ pravartate), though it is dependent on its causes (pramāṇaṃ parata utpadyate).” See Taber (1992), an investigation on the notion of svataḥ prāmāṇya ‘intrinsic validity’ for Kumārila, pp. 207–212 in particular. As he points out (p. 209, right column l. 15–17), the word bhāva ‘thing’ in the verse includes the valid cognitions. On the basis of Pārthasārathimiśra’s interpretation, Taber says (p. 210, right column ll. 25–30): “All cognitions, not just true cognitions, have intrinsic validity […]. The difference between true and false cognitions is that the latter are always eventually overturned by other cognitions, whereas the former retain their intrinsic validity indefinitely.” For this issue, see Tanizawa (2000); Arnold (2001) too.

  95. As far as I know, no one except Maṇḍana claims the dependence between the erroneous cognition and right knowledge. Kumārila as well as Vācaspati simply say that they are not dependent. In this Codanāsūtra v. 48, the existence of a cause for its arising does not affect the independence of the operation of cognition. This idea is the same for both Kumārila and Maṇḍana, but in Maṇḍana’s case, he extends his argument, stating that “there must be the relation of dependence at the fundamental level,” which makes his philosophy unique.

  96. None of the commentators refer to the point that Maṇḍana’s answer to the objection that “Scripture is dependent on direct perception etc.” is deeply connected to the interpretation of Mīmāṃsāsūtra 6.5.54 as well as 3.3.2. They overlook the fact that 6.5.54 is applicable and 3.3.2 becomes inapplicable only through this argument. Note that the Tattvasamīkṣā for this part is lost so we cannot judge Vācaspati’s understanding. As for his Bhāmatī quoted in footnote 34, it basically follows the statement in the Brahmasiddhi, but it is not explicit about the dependence of the erroneous cognition (= the one to be cancelled) and the right knowledge (= the one to cancel). The fact that Vācaspati does not mention it even though he quotes both the Mīmāṃsāsūtra and Tantravārttika may suggest his suspicion against such dependence proposed by Maṇḍana, and his intention to go back to Kumārila’s idea.

References

Primary Sources

  • - Aṣṭādhyāyī: Pāṇini’s Grammatik, edited and translated by Otto Böhtlingk, 2 vols., Rinsen Book Company, Kyoto, 1977.

  • - Abhiprāyaprakāśikā: Bhāvaśuddhi of Ānandapūrṇamuni and Abhiprāyaprakāśikā of Citsukhamuni: Two Commentaries on Brahmasiddhi, edited with introduction and appendices by N. S. Anantakrishna Sastri, Government Oriental Manuscript Library, Madras, 1963 (Madras Government Oriental Series No.161).

  • - Ṭupṭīkā: Kumārilabhaṭṭa’s commentary on the Śābarabhāṣya. See the Mīmāṃsāsūtra.

  • - Tattvasamīkṣā: Vācaspatimiśra’s commentary on the Brahmasiddhi. See Acharya [2006].

  • - Tattvasaṃgraha: Tattvasaṅgraha of Ācārya Shāntarakṣita with the Commentary ‘Pañjikā’ of Shri Kamalashīla, edited by S. D. Shastri, Bauddha Bharati, Varanasi, 1968.

  • - Tantravārttika: Kumārilabhaṭṭa’s commentary on the Śābarabhāṣya. See the Mīmāṃsāsūtra.

  • - Tāṇḍyamahābrāhmaṇa: Tāṇḍyamahābrāhmaṇa, Belonging to the Sāma Veda, With the Commentary of Sāyaṇācārya, edited by Chinnaswami Śastri, Kāshī Sanskrit Series 105, Benares, 1935, 1936.

  • - Brahmasiddhi: See Sastri [1937].

  • - Bhāmatī: Brahmasūtra Śānkara Bhāṣya with the Commentaries Bhāmatī, Kalpataru and Parimala and with Index etc., edited with notes etc. by Pandita Anantakrishna Shastri and Vasudev Laxman Shastri Pansikar, Krishnadas Academy (Krishnadas Sanskrit Series 25), Varanasi, 2000 (Reprinted).

  • - Bhāvaśuddhi: Ānandapūrṇa’s commentary on the Brahmasiddhi. See the Abhiprāyaprakāśikā.

  • - Mīmāṃsākośa: Mīmāṃsākośaḥ, edited by Kevalānandasaraswatī, 7 vols., Prajna Pathasthala Mandal, 1952–1966, (Prajñā Pāṭhasthālā Maṇḍala Grantha Mālā).

  • - Mīmāṃsāsūtra: Śrīmajjaiminipraṇītaṃ Mīmāṃsādarśanaṃ Śrīkumārilabhaṭṭaviracitatantravārttikākhyavyākhyāsahitaśābarabhāṣyasametam (vol. 2–4), Śrīkumārilabhaṭṭaviracitaṭupṭīkākhyavyākhyāsahitaśābarabhāṣyasametam (vol. 5–7), edited by Vināyaka Gaṇeśa and Mahādeva Cimaṇājī, Ānandāśramamudraṇālaya, 7 vols., 1929–1953, (Ānandāśramasaṃskṛtagranthāvali No. 97).

  • - Maitrāyaṇī Saṃhitā: Maitrāyaṇi Saṃhitā: Die Saṃhitā der Maitrāyaṇīya-Śākhā, edited by Leopold von Schroeder, F.A. Brockhaus, Leipzig, 1972.

  • - Vākyapadīya I: Vākyapadīya of Bhartṛhari with the Vṛtti and the Paddhati of Vṛṣabhadeva (Kāṇḍa I), edited by K. A. Subramania Iyer, Decan College Postgraduate and Research Institute, Poona, 1966 (Deccan College Monograph Series).

  • - Vākyapadīya II: Vākyapadīya of Bhartṛhari Kāṇḍa II with the Commentary of Puṇyarāja and the ancient Vṛtti, edited by K. A. Subramania Iyer, Motilal Banarsidass, Delhi/Varanasi/Patna, 1983.

  • - Vyākhyā: Śaṅkhapāṇi’s commentary on the Brahmasiddhi. See the Brahmasiddhi.

  • - Śatapathabrāhmaṇa: The Śatapatha-Brāhmaṇa in the Mādhyandina-Śākhā with Extracts from the Commentaries of Sāyana, Harisvāmin and Dvivedaganga, edited by A. Weber, Berlin and London, 1849.

  • - Śābarabhāṣya: Śabarasvāmin’s commentary on the Mīmāṃsāsūtra. See the Mīmāṃsāsūtra.

  • - Ślokavārttika: Ślokavārttika of Śrī Kumārila Bhaṭṭa with The Commentary Nyāyaratnākara of Śrī Pārthasārathi Miśra, edited and revised by Ganga Sagar Rai, Ratna Publications, Varanasi, 1993.

  • - Sphoṭasiddhi: The Sphoṭasiddhi of Ācārya Maṇḍanamiśra with the Gopālikā of Ṛṣiputra Parameśvara, edited by S. K. Rāmanātha Śastrī, University of Madras, 1931 (Madras University Sanskrit Series, No. 6).

Secondary Sources

  • Acharya, D. (2006). Vācaspatimiśra’s Tattvasamīkṣā, the Earliest Commentary on Maṇḍanamiśra’s Brahmasiddhi. Stuttgart: Franz Steiner Verlag.

    Google Scholar 

  • Arnold, D. (2001). Of intrinsic validity: A study on the relevance of Pūrva Mīmāṃsā. Philosophy East and West, 51(1), 26–53.

    Article  Google Scholar 

  • Benson, J. (2010). Mahādeva Vedāntin Mīmāṃsānyāyasaṃgraha: A compendium of the principles of Mīmāṃsā. Wiesbaden: Harrassowitz Verlag.

    Google Scholar 

  • Biardeau, M. (1958). Sphoṭa Siddhi (La démonstration du Sphoṭa) par Maṇḍana Miśra. Pondichéry: Institut français d’indologie.

    Google Scholar 

  • Biardeau, M. (1969). Le Philosophie de Maṇḍana Miśra. Vue à partir de la Brahmasiddhi. Paris: École français d’Extrême-orient.

  • Bronkhorst, J. (1997). Philosophy and Vedic exegesis in the Mīmāṃsā. In E. Franco & K. Preisendanz (Eds.), Beyond orientalism: The work of Wilhelm Halbfass and its impact on Indian and cross-cultural studies. Amsterdam: Rodopi (Poznan Studies in the Philosophy of the Sciences and the Humanities 59).

  • Bronkhorst, J. (2007). “Vedānta as Mīmāṃsā,” Mīmāṃsā and Vedānta: Interaction and continuity (Papers of the 12th World Sanskrit Conference 10.3), Ed. J. Bronkhorst, Delhi: Motilal Banarsidass Publishers Private Limited.

  • Cardona, G. (1970). Some principles of Pāṇini’s grammar. Journal of Indian Philosophy, 1(1), 40–74.

    Article  Google Scholar 

  • Cardona, G. (2013). Mīmāṃsā and the Pāṇinian System. Annals of the Bhandarkar Oriental Research Institute, 64, 7–110.

    Google Scholar 

  • Fujii, T. (藤井 隆道). (2014). On the definition of a sentence in classical Mīmāṃsā (古典ミーマーンサー学における文の定義), Indian and Buddhist Studies in Honor of President Dr. Shouou (Kiyoaki) Okuda in Recognition of his Lifelong Scholarship (pp. 299–312).

  • David, H. (2013). Action theory and scriptural exegesis in early Advaita-Vedānta (1): Maṇḍana Miśra on upadeśa and iṣṭasādhanatā. In V. Eltschinger & H. Krasser (Eds.). Scriptural authority, reason and action. Proceedings of a panel at the 14th World Sanskrit Conference, Kyoto, September 1st–5th 2009 (pp. 271–318). Wien: Verlag der Österreichische Akademie der Wissenschaften (Österreichische Akademie der Wissenschaften Philosophisch-Historische Klasse Sitzungsberichte 847).

  • David, H. (2017). Les définitions de l’énoncé (vākya) dans la tradition sanskrite : entre grammaire et exégèse. Language, 205, 27–42.

    Google Scholar 

  • David, H. (2020). Une philosophie de la parole. L’Enquête sur la connaissance verbale (Śābdanirṇaya) de Prakāśātman, maître advaitin du xe siècle (édition critique, traduction, commentaire, avec une nouvelle edition du commentaire d’Ānandabodha). 2 vols. Paris: École française d’Extrême-Orient / Verlag der Österreichischen Akademie der Wissenschaften (Monographies 198).

  • Gajendragadkar, A. B., & Karmarkar, R. D. (1934). The Arthasaṃgraha of Laugākṣi Bhāskara. Edited with an Introduction, Translation into English and Notes (Explanatory and Critical). Delhi: Motilal Banarsidass Publishers Private Limited.

  • Harikai, K. (針貝 邦生). (1973). Mantora no kinou ni tsuite (Mantraの機能について). Journal of Indian and Buddhist Studies, 21(2), 60–65.

  • Harikai, K. (針貝 邦生). (1990). Koten indo seiten kaishakugaku kenkyuu (古典インド聖典解釈学研究). Fukuoka: Kyushu daigaku shuppankai.

  • Ishimura, S. (石村 克). (2008). Kumārila’s theory of bādha ‘Sublation’ (ミーマーンサー学派の情報選別理論). Journal of Indian and Buddhist Studies, 56(2), 798–795 (239–242).

  • Ihara, S. (伊原 照蓮). (1963). Man-dana mishura no chosaku (マンダナ・ミシュラの著作). Iwai hakase koki kinen ronbunshū (pp. 90–98) Tokyo.

  • Iyer, K. A. S. (1966). Sphoṭasiddhi of Maṇḍana Miśra (English Translation). Poona: Deccan College Post-graduate and Research Institute.

    Google Scholar 

  • Jha, G. N. (1933). Shabara-Bhāsya: Translated into English I (Adhyāyas I-III). Baroda: Oriental Institute Baroda.

    Google Scholar 

  • Jha, G. N. (1934). Shabara-Bhāsya: Translated into English II (Adhyāyas IV–VIII). Baroda: Oriental Institute Baroda.

    Google Scholar 

  • Jha, G. N. (1942). Pūrvamīmāṃsā and its sources. Banaras: The Banaras Hindu University.

    Google Scholar 

  • Kataoka, K. (片岡 啓). (2002). Validity of cognition and authority of scripture. Journal of Indian and Buddhist Studies, 50(2), 1026–1022 (11–15).

  • Kataoka, K. (片岡 啓). (2011). Kumārila on Truth, Omniscience, and Killing Part 2, An Annotated Translation of Mīmāṃsā Ślokavārttika ad 1.1.2 (Codanāsūtra). Wien: Verlag der Österreichischen Akademie der Wissenschaften (Österreichische Akademie der Wissenschaften Philosophisch-Historische Klasse Sitzungsberichte 814).

  • Kawakami, S. (川上 真一). (1994). Kumaarira no bun-iron (クマーリラの文意論). Journal of the Nanto Society for Buddhist Studies (南都仏教), 69, 57–82.

  • Kuroda, H. (黒田 泰司). (2011). On the verse part of the Brahmasiddhi” (Brahmasiddhi韻文部に関する一考察). Journal of Religious Studies (宗教研究), 84(4), 254–255.

  • Raja, K. K. (1963). Indian theories of meaning. Madras: The Adyar Library and Research Centre.

    Google Scholar 

  • Ranade, H. G. (2006). Illustrated dictionary of Vedic rituals. New Delhi: Indira Gandhi National Centre for the Arts, Aryan Books International.

    Google Scholar 

  • Rau, W. (1977). Bhartṛharis Vākyapadīya, die Mūlakārikās nach den Handschriften herausgegeben und mit einem Pāda-index versehen. Wiesbaden: Deutsche Morgenländische Gesellschaft, Kommissionsverlag Franz Steiner Gmbh.

    Google Scholar 

  • Sastri, K. (1937). Brahmasiddhi by Ācārya Maṇḍanamiśra with commentary by Śaṅkhapāṇi, Ed. Introduction, appendices and indexes by S. Kuppuswami Sastri, The Superintendent. Madras: Government Press (Madras Government Oriental Manuscript Series No. 4).

  • Shima, I. (島 岩). (2012). A philological study of Bhāmatī (『バーマティー』の文献学的研究), Ed. J. Takashima & M. Mori. Study of language and cultures of Asia and Africa monograph series (Vol. 51). Tokyo: Research Institute for Language and Cultures of Asia and Africa (ILCAA), Tokyo University of Foreign Studies.

  • Taber, J. (1992). What did Kumārila Bhaṭṭa mean by svataḥ prāmāṇya? Journal of the American Oriental Society, 112, 204–221.

    Article  Google Scholar 

  • Tanizawa, J. (谷沢 淳三). (2000). Indian Svataḥ theory and Parataḥ theory of Prāmāṇya from the point of view of modern philosophy (インド哲学におけるprāmāṇyaの〈自ら〉説と〈他から〉説 ―現代哲学の観点から―). Shinshu Studies in Humanities, 34, 1–14.

  • Thrasher, A. W. (1993). The Advaita Vedānta of Brahma-Siddhi. Delhi: Motilal Banarsidass Publishers Private Limited.

    Google Scholar 

  • Vetter, T. (1969). Maṇḍanamiśra’s Brahmasiddhiḥ. Brahmakāṇḍaḥ. Übersetzung, Einleitung und Anmerkungen, Wien: Hermann Böhlaus Nachf. / Österreichische Akademie der Wissenschaften (Veröffentlichungen der Kommission für Sprachen und Kulturen Süd- und Ostasiens 7).

  • Yoshimizu, K. (吉水 清孝). (1987). On the function of Liṅga in the Exegetics of the Mīmāṃsā School (ミーマーンサー学派の聖典解釈学における liṅga の機能について). The Journal of Oriental Studies, 26(2), 150–169.

  • Yoshimizu, K. (吉水 清孝). (2008). The intention of expression (vivakṣā), the expounding (vyākhyā) of a text, and the authorlessness of the Veda. Zeitschrift der Deutschen Morgenländischen Gesellschaft, 158(1), 51–71.

  • Yoshimizu, K. (吉水 清孝). (2013). On the centralism of the Yajurveda in Mīmāṃsā (ミーマーンサーにおける Yajurveda 中心主義について). Studies in Religions East and West Ronshu, 40, 178–167.

Download references

Acknowledgements

This research has been done under the support of Japan Society for the Promotion of Science (JSPS). I would like to thank Dr. S. L. P. Anjaneya Sharma and Dr. Hugo David for their kindly having joined the reading sessions and given me the valuable suggestions during my stay in Pondicherry in 2018–2019.

Author information

Authors and Affiliations

  1. EFEO, Pondicherry, India

    Akane Saito

Authors
  1. Akane Saito
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Akane Saito.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Saito, A. Mīmāṃsāsūtra 6.5.54 on bādha in Maṇḍanamiśra’s Brahmasiddhi. J Indian Philos 48, 915–944 (2020). https://doi.org/10.1007/s10781-020-09447-w

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10781-020-09447-w

Keywords

Search

Navigation