Reverberations of Dharmakirti’s Philosophy

by Birgit Kellner | 2020 | 264,305 words

This page relates ‘Another Look at avinabhava and niyama in Kumarila’s Exegetical Works’ of the study on the philosophy of Dharmakirti (6th century) and his predecessor Dignaga (5th century). This collection of articles reflects philosophical currents in India, China and Tibet during their time and investigates the Buddhist theories of Pramana (“instruments of trustworthy awareness”).

Another Look at avinābhāva and niyama in Kumārila’s Exegetical Works

(By Kiyotaka Yoshimizu)


In one of my recent papers (K. Yoshimizu 2007a) I argued that in a fragment quoted by Karṇakagomin from the Bṛhaṭṭīkā (BṬ), which I call the ‘avinābhāva fragment,’[1] Kumārila (ca. 560–620)[2] states that avinābhāva, the inseparability of the reason (hetu) from the thing to be inferred (sādhya), is not sufficient for a valid inference because it would even justify an inference by asādhāraṇahetu, which has no positive examples. In another fragment from the Bṛhaṭṭīkā, which I call the ‘niyama fragment,’[3] Kumārila designates niyama, natural restriction, as the foundation of inference, illustrating this by means of causality (kāryakāraṇabhāva).

In the avinābhāva fragment, however, Kumārila does not deny the necessity of avi-nābhāva for a valid inference. In the niyama fragment, avinābhāva is considered to have a sound footing in reality by means of niyama, which means that in the real world the existence of one thing is spatiotemporally determined by another thing, and therefore the existence of the latter is invariably inferred from the confirmation of the former’s exis-tence. Inasmuch as the inseparability of one thing from another is considered to require substantiation in reality, it follows that this inseparability is reflected in the consciousness of human beings. Therefore, the avinābhāva in the BṬ might rightly be called ‘epistemic inseparability’ in distinction from physical inseparability. In the present paper, I would like to elucidate how, before composing the BṬ, Kumārila attempted to substantiate avinābhāva through niyama4 by collecting from the Tantravārttika (TV) as many cases as possible in which he presents the invariable transition from one concept to another on the basis of the relation between two things in reality. In other words, I would like to reinforce the conclusion of another paper of mine (K. Yoshimizu 2011), that Kumārila was not indebted to Dharmakīrti for this idea, as assumed by Frauwallner (1962).

1. The conceptual transition modeled upon causality

1.1 Authorization of customs by assuming the existence of Vedic injunctions

In Mīmāṃsā, it is required that all provisions of man-made (pauruṣeya) scripture (smṛti) as well as traditional customs (sadācāra), if they pertain to religious duties (dharma), be based on the Veda. If the Vedic corpus one has inherited has no injunction that can be deemed as testifying these smṛti or customs, then it is, unlike in modern philology, allowed to assume the existence of corresponding Vedic injunctions in one way or another,[4] insofar as they are sanctioned by learned people (śiṣṭa), namely, by those who have received the cultural inheritance of the Vedic tradition.[5] The popular festivals celebrated by the common peoples of different regions belong to those Mīmāṃsā topics where authority was discussed. The opponent in Śabarasvāmin’s Bhāṣya (ŚBh) lists the names of three popular festivals held in east, south, and north India[6] and states that the authority of the Vedic injunctions assumed to prescribe these festivals is limited to the respective regions, in the same way as the authority of the Vedic injunctions assumed to prescribe the number of topknots arranged on one’s head is limited to one’s own kinship group.[7]

1.2 Difference in authorization between regional festivals and the customs of kinship group

Despite its conservative position, the Mīmāṃsā school never considers the Veda as forcing human beings to comply with its injunctions unconditionally. Vedic injunction urges only those who satisfy appropriate conditions (nimitta) of entitlement to perform duties. With regard to these conditions, however, one’s living place does not matter at all if one lives in the so-called “homeland of Aryans” (āryāvarta). No matter where someone was born among the four geographical quarters, they are not endowed with a generic property (jāti) distinguishing them from people who live in other quarters.[8] The generic properties recognized as real, at least by Mīmāṃsakas, such as humanness (manuṣyatva) and Brahmin-ness (brāhmaṇatva) never change regardless of where one lives, even if one moves to another country.[9] Therefore, even if Vedic injunctions existed that were deemed as testifying popular festivals, their authority could not be limited to a particular quarter because one’s living place has no effect on one’s social affiliation.

In contrast, regarding the traditional customs of kinship groups, Kumārila adduces some examples of ritual details prescribed in the Śrautasūtras specifically for people of particular kinship groups, adding that it is actually observed that such rituals are performed by these groups.[10] On the evidence of these explicit regulations given in ritual tradition, Kumārila concludes that with regard to traditional customs, if one observes a life-style restricted (niyata) to a caste or Brahmin lineage (pratijātigotra), then one can infer the existence of Vedic injunctions restricting each life-style to the corresponding kinship group.[11]

This argument reveals Mīmāṃsaka’s conservative view that one’s birth into a particular kinship group is just as deeply engraved in one’s identity as the generic properties of a natural species. This crystallization of one’s inborn social affiliation leads to the assumption that the life-style of Aryan kinship groups is regulated by the Veda, which has no regional difference in validity, just as the characteristics and the behaviors of natural species are regulated by their nature and not by where they live. Considering that this argument is modeled upon the laws of nature, Kumārila refers to a sort of rationale (nyāya)[12] found in causality, namely that effect is compliant to its cause (kāraṇānuvidhāyikāryanyāya), and uses this rationale to infer that when observing an effect it must have a conformable (anurūpa) cause (avaśyam upalabhyamānakāryānurūpakāraṇānumānair bhavitavyam).[13]

In this statement, the laws of nature regulating the behavior of individual entities through their generic properties are said to have universal validity because generic properties are not affected by the place where these entities are located or when they exist. Moreover, the effect (kārya) is said to occur passively as regulated by the cause (kāraṇa); the effect becomes the logical reason that actively leads to the existence of the cause.[14] This reversal of direction from cause to effect in reality and from effect to cause in cognition anticipates the formulation given by Kumārila in the niyama fragment from the BṬ[15] and reminds us that Dharmakīrti assigns only the effect, not the cause, as the logical reason based on the causality between a cause and its effect (PVSV 3,11–16 on PV 1.2abc).

2. The conceptual transition from species to genus

2.1 Various relations in reality between two epistemically inseparable things

In the TV we find various cases in which Kumārila assumes epistemic inseparability (avi-nābhāva) between two things. Illustrating the relation that makes one thing epistemically inseparable from another, Kumārila adduces the relation between cause and effect (kāryakā-raṇa), between owner and owned (svasvāmin), and the accompaniment (sahacarabhāva) of two particular things.[16] To illustrate the last, Kumārila says that if the constellation kṛttikā appears in the night sky, the rohiṇī will soon appear.[17] As regards the relation between owner and owned, this might be considered a typical example of the relation between a substrate and its properties, both essential ones and those that are accidental.[18]

Moreover, Kumārila considers avinābhāva to include any kind of necessary condition when he distinguishes avinābhāva from the relationship between the subordinate means and the primary purpose (śeṣa-śeṣi-bhāva), one of the Mīmāṃsā topics that is exegeti-cally determined from ritual texts.[19] For example, the study of the Veda (adhyayana), the installation (ādhāna) of sacred fires, the cultivation (kṛṣi) of crops for oblations, and the earning of money (dravyārjana) for rewards to be paid to priests are all necessary for the performance of sacrifices (kratu). But although a sacrifice cannot be performed without them (vinā-asaṃbhava), they are not exegetically subordinate (śeṣa) to a sacrifice because they are not prescribed in the text of the sacrifice.[20]

As a further instance of a necessary condition, Kumārila applies avinābhāva to his theory of human action and Vedic injunction. In the Mīmāṃsā position, the ending of a finite verb (tiṄ) denotes the general form of intentional action called “bringing into being” (bhāvanā).[21] Fully aware of the Pāṇinian rule that tiṄ in the active voice and the middle voice denotes the agent (kartṛ) of an action,[22] Kumārila maintains that the agent is invariably cognized from bhāvanā on account of avinābhāva because if there is no agent, no action can be performed.[23]

2.2 The inseparability of species from generic properties, and causality as its ontolog-ical background

Whereas an agent is indispensable for any action, action in many cases requires other kārakas in addition to agent. According to Kumārila, bhāvanā can bear relations with kārakas other than an agent without breaking the relation with the agent. To illustrate this, he refers to the following simile: When one sees a tree and notices that it is a siṃśapā, one can also optionally realize the same object in other ways–as a tree (vṛkṣa), an earthen thing (pārthiva), a substance (dravya), or a being (sat)–without denying that it is a siṃśapā.[24] This is because the specific property of a śiṃśapā, śiṃśapātva, exists only in an entity in which generic properties of its genus such as tree-ness and so on invariably exist. Kumārila expresses the inseparability (avinābhāva) of a species from each level of its genus by saying that siṃśapātva is not inconclusive (na … anaikāntikatvam) in the sense that it never loses its ability (pratyāyanaśakti) to make the existence of the generic properties known.[25]

Owing to this inseparability, śiṃśapātva can be compared to bhāvanā in the theory of action. Upon hearing a finite verb, one may comprehend many things, such as bhāvanā, action, agent, time and other ideas.[26] If the verbal root or the tense changes, the action or the time also changes; if the verbal ending remains the same, the comprehension with regard to the agent remains the same. According to Kumārila, however, we do not have to assume the agent to be the meaning of the verbal ending because, once bhāvanā is confirmed, it is invariably known that there is an agent. This can be explained by analogy with deriving the concepts of genus from the names of species; that is, we do not have to assume that the word śiṃśapā denotes any generic properties of a genus, such as tree-ness, because a tree is invariably known to be referred to when an object is called śiṃśapā, if it is accepted that the word śiṃśapā denotes only śiṃśapātva.[27] According to Kumārila, śiṃśapātva is inseparable from tree-ness because śiṃśapātva is restricted by tree-ness (vṛkṣatvaniyata) in its nature.[28]

Here, Kumārila seems to refer to the hierarchical relationship between two generic properties, taking into account the Apoha chapter of the Pramāṇasamuccaya (PS)[29] as well as Patañjali’s Mahābhāṣya. Aṣṭādhyāyī (A) 2.2.30, upasarjanam pūrvam, lays down that the subordinate item (upasarjana) forms the first part of a compound. Anticipating this rule, A 2.1.57, viśeṣaṇaṃ viśeṣyena bahulam, lays down that whether a qualifier (viśeṣaṇa) and the item qualified (viśeṣya) do or do not form a tatpuruṣa compound (2.1.22) including karmadhāraya is determined according to the cases (bahulam).[30] In his commentary on this sūtra, Patañjali comments that the compound kṛṣṇatila (black sesame) has kṛṣṇa before tila because kṛṣṇa is subservient to tila; kṛṣṇa qualifies the object as black sesame by excluding white seasame and seasame of other colors.[31] In this context, Patañjali additionally remarks that the compound *vṛkṣaśiṃśapā is impossible because vṛkṣa does not qualify śiṃśapā as a particular kind, for there is no śiṃśapā that is not a tree (avṛkṣa).[32]

In this argument, Patañjali merely deals with the semantic connotation of tree by śiṃśapā without going into the ontological background, but we can say that Kumārila implies an ontological relation between śiṃśapātva and vṛkṣatva in the phrase “śiṃśapātva is restricted by tree-ness (vṛkṣatvaniyata),” for in his Ślokavārttika (ŚV) he has already discussed the idea that a species is ontologically restricted by a generic property. In the ŚV, Kumārila affirms that specific entities are endowed with generic properties through their own nature (svābhāvika); to possess them they do not need any cause (na hetumān).[33]

Kumārila regards generic properties as a kind of intrinsic ability (śakti) of an individual entity.[34] Generally speaking, entities of a certain kind are, by nature, invariably equipped with their own particular abilities, which can never be shared with entities of other kinds.[35] Moreover, because they are equipped with these abilities by nature, they start up their own activity of their own accord once they have been brought into existence.[36]

Taking his own remarks into account, Kumārila maintains that there must be some causes for an individual entity to be brought into existence, but in addition to these causes, nothing is required for possessing generic properties. Once an entity is brought into existence as an instance of a certain species (viśeṣeṣv eva labdheṣu keṣucit), no external condition is required (nānyavāñchana) for its acquisition of the generic property of its genus. What is required for this acquisition is only a cause to that extent (tāvanmātrapratīkṣaṇa), a cause that brings about the individual as an instance of some species or other that belongs to the genus.[37] In other words, an animal has already been equipped with the generic property of a cow when it is caused to be born as an instance of śābaleya, bāhuleya, or any other specific kind of a cow, by a pair of cows, male and female, of the respective species. This view is very close to Aristotle’s idea of “universals in the thing (in re)” (cf. K. Yoshimizu 2015b).

2.3 Kumārila’s reference to the proof of the impermanence by nature

In criticizing the Jaina theory that the soul (jīva), that is, the self (ātman) in transmigration, has the size of one’s body, Kumārila assumes the case that the soul consists of many parts (avayava), and from this assumption, he leads to the absurd consequence that the soul (jīva) would perish like a pot (ghaṭa) at some time because, he says, all forms of conjunction (saṃyoga) are inseparable from disjunction (viyogāvinābhūta).[38] Considering the production and the perishing of a pot adduced as an example, we can substitute produced-ness (kṛtakatva) for the conjunction of the parts that form a whole, and impermanence (anityatva) or perishing (vināśa) for the disjunction of conglomerated parts from one another.[39]

In another place (MmS 2.2.25), Kumārila briefly discusses the ontological relation that enables the inference of impermanence (anityatva) from produced-ness (kṛtakatva). Here Kumārila holds this type of inference to be different from those based on causality.[40] Presenting this inference, he remarks that even between two things that are not cause and effect (akāryakāraṇabhūta), it is well known that in some cases one thing indicates the other (gamyagamakabhāvaprasiddhi).

Moreover, in the ŚV chapter on the permanence of words (Śabdanityatādhikaraṇa), Kumārila enumerates three kinds of permanent elements (asaṃskṛta-dharma) in the Bud-dhist Abhidharma, namely, “cessation [of the occurrence of a dharma] by means of re-flection” (pratisaṃkhyānanirodha), “cessation [of the occurrence of a dharma] without reflection” (apratisaṃkhyānanirodha), and the ether (vyoman).[41] Thereafter, without distin-guishing between Sarvāstivāda and Sautrāntika, Kumārila refers to the Buddhist theory that things are destined to perish (vināśa) through their own nature (svabhāvasiddha) without any external cause (ahetuka).[42] Then adducing two examples, he compares generation with perishing in terms of whether or not an external cause is needed:[43] When something new is immediately generated after the perishing of an old thing, an extraneous cause is necessary. For example, a piece of fuel is changed into ashes when connected to fire, and a pot is changed into fragments when struck by a hammer. However, perishing is not produced (akriyamāna) within individual entities afterwards; rather it has been in progress from the beginning in a subtle form (sūkṣma) without being noticed; it manifests (abhivyajyate) in a gross form (sthūla) upon meeting a heterogeneous cause (vilakṣaṇahetu) such as fire or a hammer.[44] In short, perishing is the essential nature of an entity (svābhāvika) and is already established insofar as the entity is brought into existence (jātamātrapratiṣṭhita).

The phrase jātamātrapratiṣṭhita, which conforms to tāvanmātrapratīkṣaṇa (ŚV, Ākṛtivāda, v. 33), anticipates Dharmakīrti’s definition of the svabhāva to be proved by a svabhāvahetu, namely, bhāvamātrānurodhin (PV 1.2d & 39b).

3. Limits of logical formulation in exegetical works

From the abovementioned examples, we may say that in the TV Kumārila has already conceived the gist of the niyama fragment from the BṬ as stating that if one thing is restricted (niyamya) in nature, namely, brought into existence under the control of another thing (niyāmaka) in reality, then the confirmation of the former becomes the logical indicator (gamaka) of the existence of the latter thing (gamya) in cognition. He also had the idea of the invariable one-way transition in the concept from an effect to its cause, and from the existence of a thing to its generic properties on the basis of causality. He was even aware that the inference of impermanence from produced-ness is an example of the latter type of conceptual transition.[45] However, because every section of the TV deals with exegesis of scriptures, Kumārila refers only sporadically to the invariable transition from one concept to another on the basis of the natural restriction of various types.[46] Unlike Dharmakīrti, he never attempts to reduce various types of niyama to fundamental ones and relate them to each other. Moreover, in another exegetical work by Kumārila, the Ṭupṭīkā (ṬṬ), we find some cases in which the object to be known from an indicator that is inseparable (avinābhūta) from it is called “restricted” (niyata), instead of “the agent of restriction” (niyāmaka).

When one hears the Vedic injunction “One who wishes for heaven should perform a sacrifice” (svargakāmo yajeta), one may first, urged by the verb in the optative, feel obliged to perform a sacrifice. However, one usually performs an action in order to achieve a purpose one has set up in advance. The general form of action called “bringing into being” (bhāvanā) is, therefore, said to be inseparable from the motivation to achieve a purpose (prayojyādibhir avinābhūtā). Then, paying attention to another word in this injunction that describes the entitled person (adhikārin), “one who desires heaven” (svargakāmaḥ), one may come to be aware that the purpose to perform the sacrifice is a particular kind of something valuable for a human being (puruṣārtha), namely, heaven (svarga).[47] In this way, focused on the verb in the optative at first, one may simply feel one’s duty to perform a sacrifice (yāgasya kartavyatā), but this sense of unconditional duty is suspended and replaced by the motivation to achieve heaven by performing a sacrifice.[48]

However, accepting that the performance of a sacrifice is motivated by something valuable for human beings, Kumārila calls this goal of conceptual transition “restricted” (niyata-puruṣārtha-prayojyatva), instead of calling it the agent of restriction (niyāmaka) in view of the causa finalis. Nor does he consider the performance of a sacrifice to be the niyāmaka in view of the causa efficiens, for he assumes that the efficacy of a sacrifice is based on nothing other than the Vedic injunction (vidhi).[49]

In the case of Dharmakīrti’s logic, the arising of an effect (kāryotpāda) is inferred from the “complete cause” (kāraṇasāmagrī ), in other words, when all the necessary conditions are fulfilled in addition to the existence of the main cause (PVSV 6,22–7,12 on PV 1.7–8). Although Kumārila analogically compares the performance of a sacrifice to ordinary activities such as farming for a harvest,[50] it is impossible to consider the performance of a sacrifice a “complete cause” (kāraṇasāmagrī ) for attaining a result such as heaven because we cannot clarify why or how a sacrifice can bring about a result by empirically investigating its mechanism. This is because in Mīmāṃsā it is firmly presupposed that the Veda alone has authority over religious duties (dharma); there is no need to resort to empirical knowledge.


Despite assuming human rationality to be subordinate to Vedic authority as far as religious duties are concerned, Kumārila never advocates an extreme fundamentalism in which human beings are incapable of acquiring any knowledge without Vedic revelation. On the contrary, Kumārila clearly states that there are many fields to be investigated on the basis of empirical observation, giving many examples concerning the man-made scriptures (smṛtis) and the sciences ancillary to the Veda (vedāṅgas).[51] Accordingly, we may safely conjecture that after having completed the Ślokavārttika, Kumārila continued to study logical reasoning while compiling such exegetical works as the Tantravārttika and the Ṭupṭīkā. Finally, in the Bṛhaṭṭīkā, he declared natural restriction (niyama) to be the foundation of epistemic inseparability (avinābhāva).[52]

References and abbreviations

A Aṣṭādhyāyī.

AK Abhidharmakośa. See AKBh.

AKBh Abhidharmakośabhāṣya, ed. P. Pradhan. Patna[53] 1975. [1st ed. 1967].

ĀpŚS Āpastambaśrautasūtra, ed. R. Garbe. New Delhi[54] 1983. [1st ed. 1882–1902]. ĀśvŚS Āśvalāyanaśrautasūtra, ed. Ranbir Singh Bawa, et al., part 1. Hoshiarpur 1986. AV Atharvavedasaṃhitā, ed. R. Roth and W. D. Whitney. Berlin[55] 1924.

BhārŚS Bhāradvājaśrautasūtra, ed. C. G. Kashikar, part 1. Poona 1964.

Bodewitz 1990 H. W. Bodewitz, The Jyotiṣṭoma Ritual: Jaiminīya Brāhmaṇa I, 66–364. Leiden 1990.

BṬ Bṛhaṭṭīkā.

Caland 1982 W. Caland, Pañcaviṃśa-brāhmaṇa: the Brāhmaṇa of Twenty Five Chapters.

Calcutta 1982. [1st ed. 1931].

Ch. Yoshimizu 2011 Chizuko Yoshimizu, What makes all that is produced impermanent? The proof of impermanence and the theory of causality. In: Religion and Logic in Buddhist Philosophical Analysis. Proceedings of the Fourth International Dharma-kīrti Conference. Vienna, August 23–27, 2005, ed. H. Krasser, H. Lasic, E. Franco, and B. Kellner. Vienna 2011, 491–506.

Franco 2019 E. Franco, Xuanzang’s Silence and Dharmakīrti’s Dates. Wiener Zeitschrift für die Kunde Südasiens 56–57 (2019) 117–141.

Frauwallner 1962 E. Frauwallner, Kumārila’s Bṛhaṭṭīkā. Wiener Zeitschrift für die Kunde Süd-und Ostasiens 6 (1962) 78–91.

Harikai 2006 K. Harikai, The Mīmāṃsā Interpretation of the Particle Hi in Brāhmaṇa Literature: Is Anumāna Possible in the Veda? In: Expanding and Merging Horizons. Contributions to South Asian and Cross-Cultural Studies in Commemoration of Wilhelm Halbfass, ed. K. Preisendanz. Vienna 2006, 303–313.

HirŚS Hiraṇyakeśiśrautasūtra (= Satyāṣāḍhaśrautasūtra). Poona 1907.

IO Manuscript of the Tantravārttika, preserved in the British Library, London. Eggeling No. 2157 & 2158.

JB Jaiminīyabrāhmaṇa, ed. Raghu Vira and Lokesh Chandra. Delhi 1986. [1st ed. 1954].

Joshi/Roodbergen 1971 S.D. Joshi and J.A.F. Roodbergen, Patañjalis Vyākaraṇa-Mahā-bhāṣya: Karmadhārayāhnika (P. 2.1.51–2.1.72). Poona 1971.

K. Yoshimizu 2007a Kiyotaka Yoshimizu, Reconsidering the fragment of the Bṛhaṭṭīkā on inseparable connection (avinābhāva). In: Pramāṇakīrtiḥ, Papers Dedicated to Ernst Steinkellner on the Occasion of his 70th Birthday, ed. B. Kellner, H. Krasser,

H. Lasic, M.T. Much, and H. Tauscher. Vienna 2007, 1079–1103.

K. Yoshimizu 2007b Kiyotaka Yoshimizu, Kumārila’s Reevaluation of the Sacrifice and the Veda from a Vedānta Perspective. In: Mīmāṃsā and Vedānta, Interaction and Continuity, ed. J. Bronkhorst. Delhi 2007, 201–253.

K. Yoshimizu 2011 Kiyotaka Yoshimizu, Reconsidering the fragment of the Bṛhaṭṭīkā on restriction (niyama). In: Religion and Logic in Buddhist Philosophical Analysis. Proceedings of the Fourth International Dharmakīrti Conference. Vienna, August 23–27, 2005, ed. H. Krasser, H. Lasic, E. Franco, and B. Kellner. Vienna 2011, 507–521.

K. Yoshimizu 2012a Kiyotaka Yoshimizu, Kumārila and Medhātithi on the Authority of Codified Sources of dharma. In: Devadattīyam. Johannes Bronkhorst Felicitation Volume, ed. F. Voegeli, V. Eltschinger, and D. Feller. Bern 2012, 643–681.

K. Yoshimizu 2012b Kiyotaka Yoshimizu, Tradition and Reflection in Kumārila’s Last Stand against the Grammarians’ Theories of Verbal Denotation. In: Saṁskṛta-sādhutā: Goodness of Sanskrit. Studies in Honour of Professor Ashok N. Aklujkar, ed. Ch. Watanabe, M. Desmarais, and Y. Honda. New Delhi 2012, 552–586.

K. Yoshimizu 2015a Kiyotaka Yoshimizu, Kumārila’s Criticism of Buddhism as a Reli-gious Movement in his Views on the Sources of dharma. Acta Asiatica 108 (2015), 43–62.

K. Yoshimizu 2015b Kiyotaka Yoshimizu, Kumārila ni okeru kotai to huhen no hi-betsui: Aristotle tono hikaku. (Kumārila on the Non-difference between an Individual and Its Generic Property: A Comparison with Aristotle). Indo Ronrigaku Kenkyū (Indian Logic) 8 (2015) 5–30.

Kanazawa 1983 A. Kanazawa, Kumārila to , Journal of Indian and Buddhist Studies 31(2) (1983) 928–933.

Kataoka 2011 K. Kataoka, Kumārila on Truth, Omniscience, and Killing. Part 2. Vienna 2011.

Katre 1989 S. M. Katre, Aṣṭādhyāyī of Pāṇini. Delhi 1989. [1st ed. 1987].

MānŚS Mānavaśrautasūtra, ed. J. M. van Gelder. Delhi 1985. [1st ed. 1961]. MBh Mahābhārata, critical ed. V. S. Sukthankar, et. al. Poona 1933–1966. MmS Mīmāṃsāsūtra. See TV.

Oetke 1992 C. Oetke, Pragmatic Implicatures and Text-Interpretations. Studien zur In-dologie und Iranistik 16 (1992) 185–233.

Ogawa 1993 H. Ogawa, Kumārila no śabdārtha ron. (Kumārila’s Theory of śabdārtha: abhidheya and gamya). Journal of Indian and Buddhist Studies 41(2) (1993) 1051–1055.

Pandey 2006 R. Pandey, Hindu Saṁskāras: Socio-Religious Study of the Hindu Sacra-ments. Delhi 2006. [1st ed. 1969].

PB Pañcaviṃśabrāhmaṇa, ed. A. Ch. Vedantavagisha, 2 vols. Varanasi 1989. [1st ed. 1870–1874].

Pind 2015 O. H. Pind, Dignāgas Philosophy of Language. Dignāga. Pramāṇasamucca-yavṛtti V on anyāpoha. Part I: Text. Part II: Translation and Annotation, ed. E. Steinkellner. Vienna 2015.

PS 5 Pramāṇasamuccaya, the fifth chapter. See Pind 2015: Appendix 1.

PV 1 Pramāṇavārttika, the first chapter. See PVSV.

PVSV Pramāṇavārttikasvavṛtti, ed. R. Gnoli. Rome 1960.

PVSVṬ Pramāṇavārttikasvavṛttiṭīkā, ed. R. Samkrityayana. Kyoto 1982.

RV Ṛgvedasaṁhitā, ed. T. Aufrecht, Wiesbaden 1968. [1st ed. Bonn 1877].

ŚāṅkhŚS Śāṅkhāyanaśrautasūtra, ed. A. Hillebrandt, vol. 1. New Delhi 2002. [1st ed. 1888–1899].

ŚBh Śābarabhāṣya. See TV.

Schwab 1886 J. Schwab, Das altindische Thieropfer: mit Benützung handschriftlicher Quellen bearbeitet. Erlangen 1886.

Steinkellner 1971 E. Steinkellner, Wirklichkeit und Begriff bei Dharmakīrti. Wiener Zeit-schrift für die Kunde Südasiens 15 (1971) 179–211.

Steinkellner 1991 E. Steinkellner, Dharmakīrti on the Inference of Effect (kārya). In: Papers in Honour of Prof. Dr. Ji Xianlin on the Occasion of His 80th Birthday (II), ed. L. Zheng, J. Zhongxin, D. Qing, and Q. Wenzhong. Beijing 1991, 711–736.

Steinkellner 1997 E. Steinkellner, Kumārila, Īśvarasena, and Dharmakīrti in Dialogue, A New Interpretation of Pramāṇavārttika I.33. In: Bauddhavidyāsudhākaraḥ, Studies in Honour of Heinz Bechert. ed. P. Kieffer-Pülz and J.-U. Hartmann. Swisttal Odendorf 1997, 625–646.

Steinkellner 2007–2008 E. Steinkellner, Further Remarks on the Compound avinābhā-vaniyama in the Early Dharmakīrti. Wiener Zeitschrift für die Kunde Südasiens 51 (2007–2008) 193–205.

Steinkellner 2013 E. Steinkellner, Dharmakīrtis frühe Logik: Annotierte Übersetzung der logischen Teile von Pramāṇavārttika 1 mit der Vṛtti. Tokyo 2013.

ŚV Ślokavārttika, ed. S. Dvarikadāsa Śāstrī. Varanasi 1978.

Taber 2005 J. Taber, A Hindu Critique of Buddhist Epistemology: Kumārila on Perception. London/New York 2005.

TR Tantraratna, ed. Ganganath Jha. Varanasi 1963.

ṬṬ Ṭupṭīkā. See TV.

TV Tantravārttika. In: Mīmāṃsādarśanam, ed. Subbāśāstrī, 6 parts. Poona 1929–53.

VMBh Vyākaraṇamahābhāṣya, ed. F. Kielhorn, 3 parts. Revised by K.V. Abhyankar. Poona 41985. [1st ed. 1880].

von Rospatt 1995 A. von Rospatt, The Buddhist Doctrine of Momentariness. A Survey of the Origins and Early Phase of this Doctrine up to Vasubandhu. Stuttgart 1995.

Appendix: A reply to E. Steinkellner (2007/2008)

In his PVSV on PV 1.1, after having concisely explained the three kinds of logical reason (PV 1.1b: hetus tridhaiva saḥ), namely, effect (kārya), essential property (svabhāva), and the absence of the perception (anupalabdhi) of a perceptible thing, Dharmakīrti summarizes that the first two are reasons for affirmation (vastusādhanau) and the last one for negation (pratiṣedhahetuḥ) (PVSV 2,19). Then, in order to briefly answer why logical reasons must be one of these three kinds, he introduces svabhāvapratibandha, which takes either one of the two forms: tādātmya (PVSV 2,21: tadātmatva), that is, the relation between an essential property (bhāva) and another essential property (svabhāva) of one and the same thing; or tadutpatti (PVSV 3,4), that is, causality between two different (anya) things.

Then, after a short interlude, he briefly explains how an inferential cognition can be a pramāṇa on account of its non-deviation (avyabhicāra) from its object, saying “Because, between the [two] things one of which is not dependent on the other, there is no restriction [that would account] for an invariable occurrence [of the latter] with [the former]” (PVSV 3,9: anāyattarūpāṇām sahabhāvaniyamābhāvāt). This phrase can be considered the logical inverse (¬P→¬Q) of the general law (P→Q) given in PVSV 2,19–20: “Because, if [one thing] is bound [by the other] by nature, the former does not deviate from the latter” (svabhāvapratibandhe hi saty artho arthaṃ na vyabhicarati). If P and Q are logically equivalent, the conditional sentence “P→Q” as well as its inverse “¬P→¬Q” holds good (cf. Oetke 1992: 198–199). What is called āyattarūpa is something A whose essence (rūpa, i.e., svabhāva) is dependent on (āyatta), in other words, bound (pratibaddha) by, the other thing B.[56] Sahabhāva, that is, the invariable occurrence (bhāva) of B with (saha) A, is logically equivalent to its contraposition, namely, avinābhāva, that is, the never-occurrence (a-bhāva) of A without (vinā) B, which can be paraphrased as artho (A) arthaṃ (B) na vyabhicarati. Thus, sahabhāvaniyama is said to be the restriction of A’s nature by B for making the conceptual occurrence of B with A invariable; but it is unlikely that in this context sahabhāva is said to be restricted (*niyata) to the three modes in accord with the three kinds of logical reason including non-perception (anupalabdhi) that leads to non-existence, because what is argued about is the cognition of the two kinds of object to be affirmatively inferred (PVSV 3,5: etāv dvāv anumeyapratyayau), namely, svabhāva and cause (kāraṇa).

In the PVSV on PV 1.2a–c, in which kāryahetu is defined, niyama is the name of the state of real affairs in which the production of effect is regulated by its cause (3,11–12: tadutpattiniyama; 3,16: tatkāryatvaniyama). Later, concerning an inference of one thing from another different (anya) thing that is neither its effect nor its cause (PVSV 7,12: akāryakāraṇabhūta, cf. Steinkellner 2013, vol. I: 19), Dharmakīrti repeats the content of the phrase anāyattarūpāṇām sahabhāvaniyamābhāvāt in PVSV 8,12–13: apratibaddha-svabhāvasya^avinābhāvaniyamābhāvāt, in which sahabhāvaniyama is explicitly replaced with avinābhāvaniyama. As regards the immediately preceding sentence, “Hence there is no other logical reason that makes [it] known” (PVSV 8,12: tena nānyo hetur gamako’sti), Karṇakagomin glosses “no other” (nānyo) with “nothing other than a hetu of [either one of] the three kinds” (PVSVṬ 49,14: trividhād dhetor nānyo; 50,6: hetutrayavyatirekeṇa nānyo). But this gloss is off the mark. What is intended here with “no other” is “nothing other than a kāryahetu,” because this sentence is Dharmakīrti’s own comment on the last part of PV 1.10, tat kāryaliṅgajā, which concludes that the inferential cognition from a particular taste (rasa) to a particular color (rūpa) arises from the epistemic application of a special type of causality that brings about two effects simultaneously. In the rest of his commentary on PV 1.10 and from PV 1.11 onwards, Dharmakīrti goes on arguing about how to apply causality in inference.

Moreover, in the PVSV on PV 1.2cd, which defines svabhāvahetu, it is said that what is not spatiotemporally present (B) in the presence of something (A) is not restricted to its later presence (4,3: tadbhāve’bhūtasya paścād bhāvaniyamābhāvāt), in other words, will not necessarily appear later on. With this lack of paścādbhāva-niyama in case of the absence of something B when another thing A is present (tadbhāve’bhūtasya), Dharmakīrti formulates the contraposition (¬Q→¬P) of the general law (P→Q), svabhāvapratibandhe hi saty artho (A) arthaṃ (B) na vyabhicarati, in the case of pseudo tādātmya. On the other hand, as mentioned above, he formulates the inverse (¬P→¬Q) arguing the lack of avinābhāvaniyama in the case of pseudo tadutpatti in PVSV 8,12–13: apratibaddhasva-bhāvasya^avinābhāvaniyamābhāvāt.

In all these cases of PVSV 3,9, 4,3 and 8,12–13, the “restriction” (niyama) that enables sahabhāva, tadbhāve-bhāva, and even avinābhāva between two concepts is completed within either one of the two binominal forms of svabhāvapratibandha, tādātmya or tad-utpatti, respectively. For the spatiotemporal occurrence in reality, the nature of one thing is determined by another thing that is either its own essential property or its cause. Later, Dharmakīrti refers to this ontological restriction with the term svabhāvaniyama (cf. Stein-kellner 1971: 188f.). In all of these three cases, however, niyama does not restrict the total number of the types of epistemic inseparability: It is not discussed how many modes there are of the invariable transition from one concept to another.

Seeking Dharmakīrti’s own paraphrase of avinābhāvaniyamāt (PV 1.1c) in his PVSV on PV 1.1, we find only anāyattarūpāṇām sahabhāvaniyamābhāvāt (PVSV 3,9) as an approximate phrase in the inverse form. If the avinābhāvaniyama in PV 1.1c is paraphrased as sahabhāvaniyama in PVSV 3,9, and this sahabhāvaniyama is replaced with avinābhā-vaniyama in PVSV 8,12–13, then the avinābhāvaniyama in PV 1.1c, too, cannot be the restriction “of” avinābhāva in three modes in accord with the three kinds of logical reason, because anupalabdhihetu does not require its own ontological restriction for conceptual transition. Dharmakīrti does not separately assume a third form of svabhāvapratibandha for anupalabdhihetu because this kind of logical reason is based on the tādātmya of the complete cause (kāraṇasāmagrī ) of the perception of a visible thing (cf. PVSV 105, 1–3; Steinkellner 1991: 712). The state of real affairs presupposed in applying an anupalabdhi-hetu is that the complete cause of perception including the existence of a visible thing is restricted by, or, naturally equipped with, their essence (svabhāva), which is the fitness (yogyatā) to bring about its perception as their effect. It is only on an epistemic level that one can, seeing an existing empty purse, infer the non-existence of money from its non-perception on the grounds of the non-existence of the impediments of its perception, because the existence of money in a purse without impediments naturally possesses the fitness to be perceived.

That being the case, we may construe even the avinābhāvaniyama in the c-pāda of PV 1.31, which Dharmakīrti inserts as one of the two “intermediate (or, transitive) verses” (antaraślokau) after finishing a discussion about anupalabdhihetu (PVSV 19,23–20,13 on PV 1.29–30). He does this in order to introduce a new section, referring to the ontological restriction in either one of the two positive forms, but not the epistemic restriction “of” avinābhāva to the three modes of conceptual transition. This is because in the ab-pādas of the same verse Dharmakīrti declares avinābhāvaniyama to be brought into existence in “either” () one of the two positive forms: kāryakāraṇabhāvād vā svabhāvād vā niyāmakāt, as I argued in K. Yoshimizu 2007a: n. 62.[57] Accordingly, as far as PV 1 and the PVSV are concerned, we may conjecture that Dharmakīrti uses niyama in the compound avinā-bhāvaniyama “[natural] restriction [that accounts] for [epistemic] inseparability” as an ontological term allied with svabhāvaniyama and svabhāvapratibandha when he concisely asserts that the restriction of the nature of things in either one of the two positive forms brings about epistemic inseparability.

Finally, let me quote the final phrase of K. Yoshimizu 2007a: 1100:

Dharmakīrti revises Kumārila’s concept of niyama as advocated in the BṬ by ontologically confining it to two kinds, tādātmya and tadutpatti, both of which he later asserts not established between a word and its meaning (PV I v. 336) in order to disprove the Mīmāṃsā doctrine that a word is intrinsically (autpattika) related to its meaning.

Footnotes and references:


Pramāṇavārttikasvavṛttiṭīkā (PVSVṬ) 87,12–17: avinābhāvaśabdo’py a[*darśanāt] sakalārthabhāk / nānumāyogyasambandhapratipattiṃ karoti naḥ // yadi tāvad vinābhāvo na sa paścād viśiṣyate / tato’sādhāraṇe’py asti sa iti syād akāraṇaṃ // yo hy asādhāraṇo dharmaḥ sa tenaivātmasātkṛtaḥ / vinā na bhavatīty eva jñāto hetuḥ prasajyate // For an English translation, see K. Yoshimizu 2007a: 1080.


Franco (2019) convincingly demonstrates that Dharmakīrti’s dates, which were assumed by Frauwallner as ca. 600–660, cannot be moved back to the sixth century. I surmise that Kumārila passed away about 620 CE (K. Yoshimizu 2015a: n. 1), before coming to know about young Dharmakīrti, whereas Dharmakīrti wrote his first work, the first chapter of the PV refuting the Mīmāṃsā view of the intrinsic relation between a word and its meaning, by thoroughly remodeling the idea of Kumārila advocated in the BṬ that niyama is the foundation of avinābhāva. Because Dharmakīrti made use of this idea in order to attack the Mīmāṃsā theory of word and meaning, we may rightly assume as follows: If Kumārila, a hard-core Mīmāṃsaka, had first come to be informed of this idea through Dharmakīrti, he would have held it too provocative to be adopted into his own philosophical system, contrary to the assumption of Frauwallner (1962). See the appendix of the present article. Moreover, showing that PV 1.335cd is a quotation from the BṬ, Franco (2019: 132–137) corroborates young Dharmakīrti’s acquaintance with the BṬ. Following Frauwallner’s unacceptable assumption of the relationship between Kumārila and Dharmakīrti, Kataoka (2011: 47–60) errs in assuming Kumārila’s date as ca. 600–650.


PVSVṬ 87,21–30: evam anyoktasaṃbandhapratyākhyāne kṛte sati / niyamo nāma saṃbandhaḥ svamatenocyate’dhunā // kāryakāraṇabhāvādisaṃbandhānāṃ dvayī gatiḥ / niyamāniyamābhyāṃ syān niya-masyānumāṅgatā // sarve’py aniyamā hy ete nānumotpattikāraṇam / niyamāt kevalād evaṃ na kiṃcin nānumīyate // tasmān niyama evaikaḥ saṃbandho’trāvadhāryate / gamakasyaiva gamyena sa ceṣṭaḥ prāṅ nirūpitaḥ // niyamasmarataḥ samyag niyamyaikāṅgadarśanāt / niyāmakāṅgavijñānam anumānaṃ tadaṅgiṣu // For an English translation, see Steinkellner 1997: n. 28.


In early Mīmāṃsā, this problem was solved by assuming that the Vedic text containing the needed injunctions had disappeared. Kumārila, however, rejected this view in order to prevent heretics from making use of this excuse to claim orthodoxy. Instead, he proposed assuming that a corroborating injunction was preserved somewhere in other branches of the Veda. Cf. K. Yoshimizu 2012a: 650–654.


Cf. K. Yoshimizu 2015a, section III.


ŚBh 243,5–244,2: anumānāt smṛter ācārāṇāṃ ca prāmāṇyam iṣyate. yenaiva hetunā te pramāṇam, tenaiva vyavasthitāḥ prāmāṇyam arhanti. tasmād holākāda[ya]ḥ prācyair eva kartavyāḥ, āhnīnai-bukādayo dākṣiṇātyair eva, udvṛṣabhayajñādaya udīcyair eva. “Man-made scripture and traditional customs are regarded as authoritative on the basis of the assumption [of corroborating injunctions of the Veda]. On account of this very reason to make them authoritative, they (i.e., traditional customs) have authority only in respective (regions). Therefore, Holākā, etc., are to be performed only by easterners, Āhnīnaibuka, etc., only by southerners, and Udvṛṣabhayajña, etc., only by northerners.”


ŚBh 244,2–3: yathā śikhākalpo vyavatiṣṭhate, kecit triśikhāḥ, kecit pañcaśikhā iti. “Just as [the number of] topknots one arranges [on one’s head] is regulated in respective (families), for example, some (families) keep three, some five.” For the śikhā to be arranged at the cūḍākaraṇa for the children of a twice-born (dvija) family, see Pandey 2006: 94–101.


TV 246,10–11: na tāvat prācyatvadākṣiṇātyatvādijātiḥ pratīcyodīcyādivyaktivyāvṛttā sarvaprācyādivyaktiṣv anugatā kācid upapadyate, yadvacanam upapadaṃ holākādyadhikāraṃ viśiṃṣyāt. “First, easterners or southerners do not have any generic properties which would be separate from every westerner or every northerner and inhere in every easterner [or in every southerner], and which would be denoted by an accompanying word, if any, that qualifies the entitlement to Holākā, etc.”


TV 246,12–13: yās tu manuṣyatvabrāhmaṇatvādijātayaḥ teṣu vidyante tāḥ sarvadeśavāsivyaktiṣvaviśiṣṭā iti nācārānurūpaviśeṣaṇasamarthatvena jñāyante. “On the other hand, the generic properties found in them, such as humanness, Brahmin-ness, and so on, are equally inherent in all individuals who live in any region. Therefore, no [generic properties] are recognized as able to qualify [the assumed Vedic injunction] to correspond to the [localized] customs.”


TV 248,23–25: yathaiva pañcāvattaṃ tu bhṛgūṇāṃ, vasiṣṭha-śunaka-atri-vadhryaśva-kāṇva-saṃkṛtirājanyānāmnārāśaṃso dvitīyaḥ prayājas tanūnapād anyeṣāmity anvayato vyatirekataś copalakṣa-ṇasaṃbhavād vyavasthitavidhyavasānaṃ. “For example, the pañcāvatta is prescribed for the members of the Bhṛgu family.* For the members of the families of Vasiṣṭha, Śunaka, Atri, Vadhryaśva, Kāṇva, Saṃkṛti, and royal families, the second prayāja [verse that begins with] narāśaṃsa is prescribed whereas that [which begins with] tanūnapāt is prescribed for the others.** As we observe [these details performed] in this way through association (anvaya) [of practice with prescription] and dissociation (vyatireka) [of practice with the absence of prescription], we conclude that there are [Vedic] injunctions differentiated [by these kinship groups].” *Among the offerings in the new-and full-moon sacrifices, the ājyabhāga (ghee-portion) scooped four times should be offered for Agni and Soma, respectively. But in the case of the sacrifices of the Jamadagni family, the portion scooped five times (pañcāvatta) should be offered. Cf. ĀpŚS 2.18.1; BhārŚS 2.17.7; HirŚS 2.2: 190; MānŚS Jamadagni is compared to Bhṛgu in Jaiminīyabrāhmaṇa (JB) 1.152. Cf. Bodewitz 1990: 85 and 256, n. 22; Atharvavedasaṃhitā (AV) 5.19.1. In the Mahābhārata (MBh 3.115.28), Jamadagni is said to have been born in the clan of Bhṛgu (bhārgava). ** The hymn to be recited at the eleven fore-offerings (prayāja) of the animal sacrifice is called Āprī. The most generally used Āprī hymn is RV 10.110, whose second verse begins with tánūnapāt. Instead of RV 10.110.2, however, the Vasiṣṭa and related families used RV 7.2.2, which begins with nárāśáṃsa, as the second verse of the Āprī hymn. Cf. ŚāṅkhŚS 1.7.2: tanūnapād agna ājyasya vetv iti dvitīyaḥ; 3: narāśaṃso agna ājyasya vetv iti dvitīyo vasiṣṭhaśunakānām atrivadhryaśvā-nāṃ kaṇvasaṃkṛtīnāṃ rājanyānāṃ prajākāmānāṃ ca; ĀśvŚS 1.5.21–22: tanūnapād agna ājyasya vetv iti dvitīyo’nyatra vasiṣṭha-śunaka-atri-vadhryaśva-rājanyebhyaḥ; 22: narāśaṃso agna ājyasya vetv iti teṣām; Schwab 1886: 90–91. Kumārila’s remark is much closer to ŚāṅkhŚS than to ĀśvŚS, since ŚāṅkhŚS mentions the Kāṇva and the Saṃkṛti families. RV 10.110 is a hymn of Jamadagni, and RV 7 is the collection of Vasiṣṭha’s hymns. See Th. Aufrecht’s “Verzeichnis der angeblichen Hymnendichter gemäß der Anukramaṇikā” appended to his edition of the RV.


TV 248,25–26: tathaiva pratijātigotraniyatatriśikhaikaśikhādikalpavyavasthitavidhiviśeṣānumānopapattir astīti. “In the same way, from the fact that whether one arranges three topknots or one topknot is restricted (niyata) to each caste and each Brahmin lineage (pratijātigotra), it is appropriate to infer the existence of a particular injunction differentiated [by the respective kinship group].”


The first of the proponent sūtras (Mīmāṃsāsūtra [MmS] 1.3.16: api vā sarvadharmaḥ syāt, tannyāyatvād vidhānasya) asserts that a Vedic injunction, if it exists, must be a norm issued to all people (sarvadharma) irrespective of where they live. To assert this, however, this sūtra merely adds the tautological reason that it is reasonable (tannyāyatva) to consider that Vedic injunction works (vidhāna) in this way. Kumārila uses the reason of causality to avoid this tautology.


TV 245,14–15: iha smṛtīr ācārāṃś copalabhya mūlaśrutiṣv anumīyamānāsu kāraṇānuvidhāyikāryanyāyenāvaśyam upalabhyamānakāryānurūpakāraṇānumānair bhavitavyam. “When one, having observed some man-made scriptures and customs in this world, attempts to assume the corroborating injunctions [of the Veda], one should invariably infer the cause that is conformable to the effect one has observed, following the rationale that the effect is compliant to its cause.”


On another occasion, Kumārila remarks that unlike effect, cause does not work as an indicator (gamaka) because in many cases it is inconclusive (vyabhicārabahutva), namely, when auxiliary conditions are not fulfilled, it fails to indicate the occurrence of its effect. TV 544,20–21: kāryakāraṇayor api ca na tathā kāraṇaṃ gamakaṃ vyabhicārabahutvād yathā kāryam. This remark is made immediately after the argument that the inference from kṛtakatva to anitytva is not based upon causality. Cf. n. 41.


PVSVṬ 87,29–30: niyamasmarataḥ samyag niyamyaikāṅgadarśanāt / niyāmakāṅgavijñānam anu-mānaṃ tadaṅgiṣu // “If one who correctly remembers the restriction sees one member [of the restriction-relation], (namely) the restricted (member) (niyamya), the resultant cognition of the (other) member which is the restricting (niyāmaka) is an inference in regard to such (entities) that possess these (two) members [i.e. pakṣas].” (Tr. Steinkellner 1997: 634, n. 28.)


TV 139,13–14: atha vodāhṛtaviṣayahetulakṣaṇam etat*. avinābhāvo hy aneka-kāryakāraṇa-svasvāmisahacarabhāvādi-prabhedabhinnaḥ. “Or, this [merely] characterizes the logical reason given in the quoted ‘example sentence of exegesis’ (viṣaya). For, the epistemic inseparability differentiates according to the difference of the relation between cause and effect, that between owner and owned, the accompa-niment, and so on.” Cf. Kanazawa 1983: 933; Harikai 2006: 307, n. 10. In view of the context of the section in question (MmS 1.2.26–30), this portion of the TV is included in the part of the pūrvapakṣa that construes śūrpeṇa juhoti, tena hy annaṃ kriyate as a reason justifying the use of a winnowing basket (śūrpa) for an offering. For the contents of this section, see Harikai 2006. * This etat refers to the kāryakāraṇabhāva in ŚBh 139,1: nanv aprasiddhe kāryakāraṇabhāve na hetūpadeśaḥ. “[Some people] may contend that the presentation of a logical reason is made impossible unless a causal relationship is established [between the reason and the thing to be proved].”


TV 139,8–9: tat* tv ayuktam. akāryakāraṇabhūtānām api kṛttikādīnām acirodgatarohiṇyādipratipattihetutvadarśanāt. Cf. ŚV, Anumānapariccheda, vv. 12d–13ab: kva cic cāpi viśeṣayoḥ // kṛttikodayam ālakṣya rohiṇyāsattikḷptivat / *This tat refers to TV 139,7–8: kecid āhuḥ. kāryakāraṇayor evānumānam.


“For example, [hearing] that he has a vajra makes one infer that [the deity concerned is] Indra” (ṬṬ 1614,25–1615,15: yathā vajrītīndram anumāpayati) is used by Kumārila as a simple example of the exegetical inference from a qualifier (viśeṣaṇa) mentioned in a text to the sacrifice qualified (viśeṣya) by it. ṬṬ 1614,24–25: ahaḥśabdo yāgaviśeṣavacano’śrutam api viśeṣyam anumāpayati. nedaṃ viśeṣaṇaṃ yāgam antareṇa saṃbhavati. “The word ‘day,’ which expresses a particular sacrifice [in the text concerned, that is, Pañcaviṃśabrāhmaṇa (PB) 22.9.1*], makes one infer the [sacrifice] qualified [by it] even if [the sacrifice] is not directly expressed. This qualifier cannot occur [in the text] without the sacrifice.” * Cf. PB 22.9.1: catvāri trivṛnty ahāny agniṣṭomamukhāni viśvajin mahāvrataṃ jyotiṣṭomo’tirātraḥ “Four nine-versed (trivṛt) days, the first of which is an agniṣṭoma (the three others being ukthyas); a Viśvajit-day; a mahāvrata-day, (and) a jyotiṣṭoma as over-night-rite.” (Tr. Caland 1982: 578.) This is the breakdown of a seven-day-sacrifice called janakasaptarātra (cf. Caland 1982: 578, n. 1). Moreover, Kumārila considers the restriction of parts by the whole (samudāya) as the basis for the conceptual transition from a piece of text to the entire text by which it is restricted (niyamya), in other words, to which it belongs. See n. 47. In addition, we may also classify the inference from an accidental property (guṇa) to its substrate into this type. Kumārila offers the example that one can infer the existence of a substance (dravya) from the cognition of color. Cf. ṬṬ 1368,15–18: yathā śuklādiguṇo dravyeṇa vinā na bhavatīty avinābhāvād dravyaṃ lakṣayati, naivam iha; Kanazawa 1983: 932.


TV 654,10–15: ayuktam evaṃ śeṣatvam avinābhāvalakṣaṇam / vyabhicārāt tathāhīdam aśeṣeṣv api dṛśyate // sarvadā hy avinābhūtā rūpasparśādayaḥ kṣitau / na caiṣāṃ tulyakalpatvād bhavaty anyo-nyaśeṣatā // svāminā ca vinā dāsā na bhavanti kadācana / tathā gṛhādayas teṣāṃ śeṣaḥ svāmī ca neṣyate // “It is not right to say that subordination is defined as epistemic inseparability on account of inconclusiveness, for we find this also in those cases that are not subordinate. As a matter of fact, color, touch and other [qualities of elements] are always inseparable from one another in earth, but among them, there is no subordination because they are equally arranged [in earth]. Without a master, moreover, there can never be a slave, a house, and so on, but the master is not deemed to be subordinate to them.” Cf. Kanazawa 1983: 932–933.


TV 654,20–21: etenādhyayanādhānakṛṣidravyārjanādiṣu // vācyaṃ prasaṅgiśeṣa(IO 2157, 6b6: śeṣa; Ānandāśrama ed.: śeṣi)tvaṃ tair vinā kratvasaṃbhavāt / “By this (i.e., assuming avinābhāva to be the definition of subordination), one would have to say that Vedic study, fire installation, cultivation, earning money, and so on, come to be subordinate because without them the performance of a sacrifice is impossible.” Moreover, Kumārila finds inseparability in the hierarchy of social order, saying that when a king is invited somewhere, all the subordinates who are inseparable from him must follow him. TV 486,23–26: loke’pi ca yat pradhānaṃ rājaprabhṛti kvacin nīyate tadātmanāvinābhūtāny amātyā-dīny aṅgāny apy aśeṣāṇy ākṣipati. yadā tu kaścid bhṛtyānāṃ kvacid gacchati tadā tadasaṃbandhīni bhṛtyāntarāṇi padam api na calanti. “Even in this world, a chief person, for example, a king, when invited to some place, mobilizes all the subordinate people who are inseparable from him, for example, ministers, etc. On the other hand, when one of the servants goes somewhere, the others do not take even a single step because they are not connected to that (servant).”


For the reason why Mīmāṃsakas insist on denoting bhāvanā through verbal endings, in defiance of the Pāṇinian rule, see K. Yoshimizu 2012b: 555–557.


A 3.4.67: kartari kṛt; 69: laḥ karmaṇi ca bhāve cākarmakebhyaḥ.


TV 914,1–2: yādṛśaś ca guṇabhūtaḥ kartātrāvagamyate na tādṛśena vinā bhāvanopapadyata ityarthā-pattyānumānena vā śaktā gamayitum. “No matter in whichever manner an agent that is subordinate [to the action] is cognized in this (finite verb), bhāvanā is impossible without an agent. In this way, [bhāvanā] is able to make [an agent] known through logical derivation or through inference.”


TV 914,7–8: yas tv anyatrāpi vartamānaḥ pūrvāvagatasaṃbandhyaparityāgenaiva vartate yathā śiṃśa-pātvaṃ vṛkṣatvapārthivatvadravyatvasattveṣu. “On the other hand, there is a case in which a thing does not renounce the formerly cognized relation with another thing, even if it has come into the relation with something else. For example, [when an object is cognized as śiṃśapā] the śiṃśapā-ness remains [in one’s cognition] even if [the object is cognized as having] the generic properties of tree, earthen thing, substance, and being.”


TV 914,8: na tasyānaikāntikatvaṃ, sarveṣu pratyāyanaśaktyavighātāt. “It (i.e., siṃśapātva) is not inconclusive in the sense that it never loses its ability to make [the existence of] all of them known.”


TV 913,16–18: tad iha pacatyādiśabdoccāraṇād iyanto’rthā gamyante, bhāvanādhātvarthakartṛta-tsaṃkhyāpuruṣopagrahakālaviśeṣāḥ. teṣāṃ tu kaḥ śābdaḥ ko’rthād iti vibhāgo na jñāyate. lakṣaṇaṃ caitāvad yo’rthān na gamyte sa śabdārtha iti. “Now, when hearing the word pacati, etc., pronounced, one cognizes so many things, such as bhāvanā, the meaning of the verbal root (i.e., an action), agent, its number, the grammatical person, grammatical voice, and a certain time. Among them, however, it is not yet ascertained which one is known directly from the word and which one is from the meaning [of the word], and the meaning of a word is merely defined as that which cannot be cognized through anything else that is assumed to be the meaning [of that word].” Cf. Ogawa 1993: 1052–1053.


TV 940,12–14: śiṃśapāśabdo hi na tāvad vṛkṣatvapārthivatvadravyatvasattāprameyatvajñeyatvānām abhidhāyako’tha ca tadviśeṣam eva śiṃśapātvam(*-pātvam; IO 2158, 39(152) b2 & Ānandāśrama ed.: -pām) abhidhatte. “For the word śiṃśapā does not denote tree-ness, nor earthen-ness, nor substance-ness, nor being-ness, nor the property of being correctly known, nor the property of being known, but it denotes only their particular form, namely, śiṃśapā[-ness].” Cf. TV 982,10–13; TV 1048,15–16.


TV 932,16–18: nābhidhātā kartṛkarmaṇoḥ pacatyādiśabdas tadatyantāvinābhūtārthābhidhāyitvāt. yo yadatyantāvinābhūtārthābhidhāyī sa tasya na vācakaḥ, yathā śiṃśapāśabdo vṛkṣatvaniyataśiṃ-śapātvasya. “The word ‘pacati,’ etc., does not denote agent or object because it denotes a meaning that is completely inseparable from them. If a word denotes something that is completely inseparable from something else, then the word does not denote the latter, just as the word śiṃśapā [denotes only] śiṃśapā-ness, which is restricted by tree-ness.” Cf. Kanazawa 1983: 929.


Kumārila may have collected the essential predicates applicable to a śiṃśapā from PS 5.35: vṛkṣatvapārthivadravyasajjñeyāḥ pratilomyataḥ / catustridvyeka sandehe nimittam niścaye’nyathā // “‘treeness,’ ‘earthen,’ ‘substance,’ ‘existent,’ and ‘knowable’ are [each] a cause of doubt, in reverse order, about four, three, two, and one [properties]. In opposite order they serve the purpose of ascertainment.” (Tr. Pind 2015: I 125.)


The case in which a compound is obligatory, such as kṛṣṇasarpa (cobra = “black snake”), is an exception to the governing rule (adhikārasūtra), A 2.1.11: “optional” (vibhāṣā). Cf. Katre 1989: 121.


VMBh I, 399: 3–15: anyatarasya pradhānabhāvāt tadviśeṣakatvāc cāparasyopasarjanasaṃjñā bhavi-ṣyati. yadāsya tilāḥ prādhānyena vivakṣitā bhavanti kṛṣṇo viśeṣaṇatvena tadā tilāḥ pradhānaṃ kṛṣṇo viśeṣaṇam. “(Therefore,) since one of the two is the main (word) and since the other functions as the qualifier of that, the designation upasarjana will apply. When the sesame seeds are intended as the main thing by somebody, and black as the qualifier, then (the word) tilāḥ becomes the main (word) and (the word) kṛṣṇāḥ (becomes) the qualifying word.” (Tr. Joshi/Roodbergen 1971: 140.)


VMBh I, 399: 25–26: kathaṃ tarhīmau dvau pradhānaśabdāv ekasminn arthe yugapad avarudhyete vṛkṣaḥ śiṃśapeti. naitayor āvaśyakaḥ samāveśaḥ. na hy avṛkṣaḥ śiṃśapāsti. “Then in what way can these two main words vṛkṣaḥ śiṃśapā ‘the tree śiṃśapā’ be locked up together to refer to one and the same object? The co-usage of these two words in the order vṛkṣaḥ śiṃśapā is not necessary, because there is no śiṃśapā which is not a tree.” (Tr. Joshi/Roodbergen 1971: 147–149.)


ŚV, Ākṛtivāda, vv. 31cd–32: svābhāvikaś ca sambandho jātivyaktyor na hetumān // tenaitasya prasiddhyarthaṃ nānyat sāmānyam iṣyate / śaktisiddhivad etasya svabhāvo’tra na vāryate // “The relation between an individual and its generic property is based on its own natural disposition and has no [extra-neous] cause. Therefore, for the purpose of the accomplishment of this (relation), no more universal is needed. It is not denied that an individual is innately related with this (particular generic property) in the same way that [it is innately] equipped with some abilities.”


ŚV, Ākṛtivāda, vv. 17–18: bhinnā viśeṣaśaktibhyaḥ sarvatrānugatāpi ca / pratyekaṃ samavetā ca tasmājjātir apīṣyatām // tenātmadharmo bhedānām ekadhīviṣayo’sti naḥ / sāmānyam ākṛtir jātiḥ śaktir vā so’bhidhīyatām // “[This generic ability (śakti) of an individual] is different from specific abilities; [it] recurs everywhere [in all individuals] and completely inheres in each individual, and thus should be accepted as a generic property [called cow-ness]. Therefore, in our position, entities are endowed with an essential property that is the object of the cognition of the same kind, and one may call it universal, form, generic property, or ability.” For pratyekaparisamāpti as a characteristic of apoha in addition to its ekatva and nityatva, see PS 5 44,11.


ŚV, Ākṛtivāda, vv. 28–29ab: bhinnatve’pi hi kāsāñ cic chaktiḥ kāś cid aśaktikāḥ / na ca paryanuyogo’sti vastuśakteḥ kadā cana // vahnir dahati nākāśaṃ ko’tra paryanuyujyatām / “Although [all individuals are] different from one another, some of them have a particular ability and others do not, and one should never bring the ability of a real thing into question. Who would be questioned about the fact that fire burns but the ether does not?”


ŚV, Codanādhikaraṇa, vv. 47cd–48: na hi svato’satī śaktiḥ kartum anyena śakyate // ātmalābhe ca bhāvānāṃ kāraṇāpekṣatā bhavet / labdhātmanāṃ svakāryeṣu pravṛttiḥ svayam eva tu // “For any ability that would not be intrinsic [in an entity] cannot be produced by something else. When coming into existence, entities may call for [their own] cause. Once having been brought into existence, however, they start up their own activity of their own accord.”


ŚV, Ākṛtivāda, v. 33: yad vā naimittikatve’pi tāvanmātrapratīkṣaṇāt / viśeṣeṣv eva labdheṣu keṣucin nānyavāñchanam // “Even if [assumed to be] dependent on a certain cause, [the relation between an individual and its generic property] merely requires [a cause] to the following extent: once [an instance of] any species has been brought into existence, nothing more is required.” With “a cause to the following extent” (tāvanmātra), Kumārila may exclude the Vaiśeṣika’s idea that inherence (samavāya) must exist between two things that are inseparable (ayutasiddha), such as substance and quality, individual and universal. In ŚV, Pratyakṣasūtra, vv. 146–148, Kumārila criticizes this idea as follows: the assumption that two things have already been established (niṣpanna) as inseparable would make inherence useless, whereas the assumption that two things require inherence for their inseparability leads to a regress (avyavasthā) of further intermediary (see Taber 2005: 108–109).


TV 403,1–5: sarve hy avayavasadbhāvādayaḥ pratyakṣādyaviṣayāḥ santa utprekṣāmātreṇa kalpyante. te śabdavac cehāpi pratyākhyeyāḥ. śarīramātrāvadhikānāṃ caiṣām utprekṣitum apy ānantyād aśakyaṃ, na caiṣāṃ dravatvādinā vinā kaścit saṃśleṣahetur vidyate. na cāsaṃśliṣṭānām ekajīvārambhasāmarthyam. sarvasaṃyogānāṃ ca viyogāvinābhūtatvād ghaṭāditulyaḥ kadācij jīvasya pradhvaṃso bhavet. “Since all these (assumed things, cf. TV 402,27–30), namely, the existence of the parts [of the soul], etc., are not the object of perception, they are assumed only by imagination and are to be rejected here just like [the parts of] a word [in the sense of phonemes (varṇa) as rejected in Sphoṭavāda vv. 11–13 of the ŚV]. Because they are boundless, one can never imagine that they are limited to the extent of the body. Without some fluid substances, moreover, there cannot be any ground for their conglomeration, and they are unable to bring about a life unless conglomerated. Besides, because all sorts of conjunction are epistemically inseparable from disjunction, the soul would suffer perishing at some time in the same way as a pot.”


The logical inseparability of kṛtakatva and sāvayavatva from vināśatva has already been mentioned in ŚV, Anumānapariccheda vv. 20cd–21ab: kṛtasāvayavatvādiprayuktā ca vināśitā // prayatnānantarajñā-nasadṛśair na prayujyate / “Perishability is caused [to be known] by produced-ness, composite-ness, etc., but it is not caused [to be known] by the fact of being cognized immediately after exertion, etc.”


TV 544,18–20: ato vacanāt* kāryakāraṇayor eva gamyagamakabhāvo bhāṣyakārābhipreta iti kecin manyante. tat tv ayuktam. akāryakāraṇabhūtānām apy anityatvakṛtakatvādīnāṃ gamyagamakabhāva-prasiddheḥ. “Some people assume from this statement that the author of the Bhāṣya has the idea in mind that only a cause and its effect have the relation between them of logical indicator and the thing to be indicated. This is, however, not correct. Even between two things that are not a cause and its effect, it is well known that in some cases one thing indicates the other, for example, between impermanence and produced-ness.” * This vacana refers to ŚBh 544,2–4: yad dhi yasya kāraṇabhūtaṃ dṛṣṭaṃ siddhe, tac cet sādhye’pi kāraṇabhūtam ity avagamyate. bhavati tat tasya sādhakam. “As a matter of fact, when A has already been found to be the cause of B in a well-known example, and if in the subject of inference, too, A is found to be the cause, then A is the logical reason for inferring [the existence of] B [in the subject of inference].” As regards causality, Kumārila, unlike Śabara and like Dharmakīrti, remarks that an inference from cause to effect is inconclusive. Cf. n. 15.


ŚV, Śabdanityatādhikaraṇa, vv. 22–23ab.


ŚV, Śabdanityatādhikaraṇa, vv. 23cd–24: tau ca dvāv apy anāśitvād iṣṭāv akṛtakāv api // āhuḥ svabhā-vasiddhaṃ hi te vināśam ahetukam / hetur yasya vināśo’pi tasya dṛṣṭo’ṅkurādivat // “And both of them (i.e., pratisaṃkhyānanirodha and apratisaṃkhyānanirodha) are considered not to be produced because of their imperishability. For they (i.e., Buddhists) say the perishing [of an entity] is established by nature without [extraneous] cause because anything that has a cause is also observed to be perishable, just like sprouts, etc.” I am indebted to Dr. Taisei Shida for telling me about this reference by Kumārila to the Buddhist proof of impermanence without extraneous cause in the Śabdanityatādhikaraṇa.


ŚV, Śabdanityatādhikaraṇa, vv. 25cd–29: bhavati hy agnisambandhāt kāṣṭād aṅgārasantatiḥ // mudgarādihatāc cāpi kapālaṃ jāyate ghaṭāt / svābhāviko vināśas tu jātamātrapratiṣṭhitaḥ // sūkṣmaḥ sadṛśasantānavṛtter anupalakṣitaḥ / yadā vilakṣaṇo hetuḥ patet sadṛśasantatau // vilakṣaṇena kā-ryeṇa sthūlo’bhivyajyate tadā / tenāsadṛśasantāno hetoḥ sañjāyate yataḥ // tenaivākriyamāṇo’pi nāśo’bhivyajyate sphuṭaḥ / sa mudgaraprahārādiprayatnānantarīyakaḥ // “The temporal succession of embers proceeds from the fuel if it is connected to fire; and when a pot is struck by a hammer, a fragment is born from that (pot). The perishing [of the fuel and the pot], however, is based on their own natural disposition and established insofar as they are produced. [The inborn perishing] in subtle form is imperceptible on account of their existence in homogenous succession. When a heterogeneous cause happens to the homogenous succession, then [the inborn perishing] is manifest in gross form by means of heterogeneous efficacy. In this manner, there occurs a heterogeneous succession on account of the (heterogeneous) cause. By means of this (cause), the perishing, despite not being produced, is manifest in explicit form. This [explicit perishing] appears as soon as one makes an effort, for exam-ple, to hit a hammer [on a pot].” Kumārila’s explanation of spontaneous perishing seems to be based on Abhidharmakoṣabhāṣya (AKBh) 193,5–7: saṃskṛtasyāvaśyaṃvyayāt” (AK 4.2d). ākasmiko hi bhāvānāṃ vināśaḥ “[The momentariness of everything is established] because the conditioned entity necessarily ‘perishes.’ For the destruction of things is spontaneous” (tr. von Rospatt 1995: n. 397), because the simile of the contact of the fuel with fire is applied in AKBh 194,1–6: yadi ca kāṣṭhādīnām agnyādisaṃyogahetuko vināśaḥ syād evaṃ sati pākajānāṃ guṇānāṃ pakvataratamotpattauhetuḥ syāc ca vināśakaḥ” (AK 4.3b). hetur eva ca vināśakaḥ syāt. “And if the destruction of firewood etc. was caused by the contact with fire etc., this being so, in the case of the more and more progressed origination of the qualities borne from burning (pākaja) ‘the cause [for these qualities] would be [the factor] destroying [them],’ [that is,] precisely [their generative] cause would be the [factor] destroying [them].” (Tr. von Rospatt 1995: n. 402.)


Kumārila insists, however, that the proof of impermanence cannot be applied to words (śabda) because words are not produced but only made manifest by means of an utterance. Cf. ŚV, Śabdanityatādhi-karaṇa, v. 39ab: vyañjakābhāvataḥ śabde’py abodho badhirādivat / “As regards words, too, if there is no cause of manifestation, one cannot perceive them just like deaf people;” vv. 40cd–41: sann eva sādhanābhāvāc chabdo naivopalabhyate // kṣaṇikaṃ sādhanaṃ cāsya buddhir apy anuvartate / me-ghāndhakāraśarvaryāṃ vidyujjanitadṛṣṭivat // “In spite of existing, a word is never perceived insofar as the means [of its perception, i.e., an utterance] is absent. And the cognition of that (word) conforms to the means that remains only for a moment, just like the sight caused by a flash of lightning during a dark night covered with clouds.” Cf. ŚV, Śabdanityatādhikaraṇa, vv. 424–425; von Rospatt 1995: n. 369. Kumārila also denies the momentariness of things for everyday use. ŚV, Śabdanityatādhikaraṇa, v. 426: kṣaṇabhaṅgo ghaṭādīnāṃ vāryas tair eva sādhanaiḥ / tathaiva pratyabhijñānād, yāvad dṛṣṭaṃ na bādhate // “The momentary perishing of a pot, etc., is to be rejected for the same reasons [as those given in vv. 416–421 in order to disprove the momentary perishing of words]. Because [a pot] is recognized to be the same [as a few moments before], [the continuation of its identity] is not denied insofar as it remains to be perceived.”


Cf. n. 41.


Based on a holistic view that the whole pervades (vyāpaka) its parts and parts are restricted (niyamya) by the whole they form, Kumārila maintains an exegetical principle that a word that directly refers to a Ṛgveda verse (ṛc) indirectly indicates the whole text of the Ṛgvedasaṃhitā. Cf. TV 807,17–18: vedaś ca vyāpakatvād ahetulakṣaṇayuktaḥ san na śaknoty evāvayavaṃ lakṣayitum. ṛgādayas tu niyamyatvāt samudāyaṃ lakṣayanti. “Because [an entire text of] the Veda pervades [all of its parts], it is not fit for the definition of a logical reason. Therefore, it cannot indirectly denote its part. The verses of ṛc, etc., on the other hand, indirectly denote the whole [text of the Ṛgvedasaṃhitā] on account of being restricted [by the whole].”


ṬṬ 1352,16–18: ānantaryaśrutyā yāgasya kartavyatā. pratyayaḥ punaḥ svārthe bhāvanāyāṃ puruṣaṃ pravartayati. sā ca prayojyādibhir avinābhūtā, tena yan niyamitaṃ tad eva puruṣārthaṃ prayojyatve-nākāṅkṣati, na yāgam. “No sooner than one has heard the injunction, one recognizes that the sacrifice is to be performed. The ending [of the verb in the injunction], however, prompts one to exert oneself for bhāvanā, its own meaning. And this (bhāvanā [of one’s activity]) is inseparable from its purpose, etc. [that is, purpose (sādhya), means (sādhana) and the mode of performance (itikartavyatā)]. For its purpose (prayojya), [the bhāvanā] requires only the things valuable for a human being insofar as they are restricted (niyamita) by that (injunction), not the sacrifice itself.” Kumārila distinguishes two kinds of bhāvanā (cf. TV 378,20–21). The bhāvanā discussed in this context is the “intentional force for activity” (arthātmabhāvanā/arthībhāvanā), the general form of one’s action that requires a purpose to be achieved by the one who performed the action, but not the “verbal force for activity” (abhidhābhāvanā/śabdībhāvanā), which is the function of the verbal ending of an injunction to urge one to exert one’s bhāvanā.


ṬṬ 1352,23–25: tasmān naiva kadācid yāge pratyaya upanipatati saty evānantarye. ato yatra kartavyatāvacanas tatra puruṣārthaḥ sākṣāt pāramparyeṇa vāvinābhūtaḥ. tasmād yāgasya kartavyatā prāptā, sā bhāvanāśabdasya niyatapuruṣārthaprayojyatvād bādhyate. “Therefore, the verbal ending never happens to be directed to the sacrifice even though it is immediately cognized. Thus, no matter in what statement a duty is enjoined, something valuable for a human being is inseparably connected [to that duty], whether directly or indirectly. Therefore, even if it is assumed that the sacrifice is to be performed [for its own sake], this assumption is given up because the word that [urges the hearer to exert his] bhāvanā has [the hearer’s] purpose [in performing the sacrifice] restricted [by the injunction] to something valuable for a human being.”


In his commentary on the ṬṬ, Pārthasārathimiśra remarks that what Kumārila refers to by the pronoun tena, by which the purpose in performing the sacrifice is restricted to a particular puruṣārtha (ṬṬ 1352,17: tena yan niyamitaṃ tad eva puruṣārthaṃ prayojyatvenākāṅkṣati), is not the action of sacrifice but the injunction (vidhi) that prompts human beings by nature (puruṣapravartanātmaka). Cf. TR 390,10–13.


TV 395,3–6: laukikaṃ cāpi yat karma phale kālāntarodgatau / tatrāpi śaktir evāste na tv apūrvam iheṣyate // yāny api ca laukikāni kṛṣighṛtapānādhyayanaprabhṛtīti karmāṇi kālāntaraphalatveneṣyante teṣāṃ api svarūpāvasthānāsaṃbhavāt saṃskārair eva tiṣṭhadbhir vyavahārasiddhiḥ. “Even in worldly activities, there must remain an ability that brings about its result later on, but [such ability] is not recognized as apūrva in this (science of exegesis, mīmāṃsā). Such worldly activities as cultivation, drinking ghee, and study are accepted to have their result later on. Because they do not continue to exist in their own form [until the time of the appearing of the result], these ordinary activities are accomplished through certain dispositions that continue [until then].”


Cf. TV 166,27–28: tatra yāvad dharmamokṣasaṃbandhi tad vedaprabhavam. yat tv arthasukhaviṣayaṃ tal lokavyavahārapūrvakam iti vivektavyam “Among those (smṛti texts), the portions have their origin in the Veda insofar as they pertain to duties and emancipation, but those concerned with profit and pleasure are to be distinguished because they are based on worldly activities;” 166,29–167,1: yat tu pṛthivīvi-bhāgakathanam tad dharmādharmasādhanaphalopabhogapradeśavivekāya kiṃcid darśanapūrvakaṃ kiṃcid vedamūlam, vaṃśānukramaṇam api brāhmaṇakṣatriyajātigotrajñānārthaṃ darśanasmaraṇamū-lam “The reports of the various parts of the earth are partly based on observance and partly on the Veda, for the purpose of distinguishing the regions appropriate for receiving the reward for having performed dharma and adharma. In addition, the chronicles of genealogies, which serve to know the caste and lineage of Brahmins and Kṣatriyas, are based on observance and legends;” 167,4: aṅgavidyānām api kratvarthapuruṣārthapratipādanaṃ lokavedapūrvakatvena vivektavyam “Also in the ancillary sciences [of the Veda], one should distinguish between the portions based on the Veda for informing what is valuable for the sacrifice, and the portions based on the ordinary world for informing what is valuable for human beings.”


See appendix for a reply to Ernst Steinkellner.


Franco (2019) convincingly demonstrates that Dharmakīrti’s dates, which were assumed by Frauwallner as ca. 600–660, cannot be moved back to the sixth century. I surmise that Kumārila passed away about 620 CE (K. Yoshimizu 2015a: n. 1), before coming to know about young Dharmakīrti, whereas Dharmakīrti wrote his first work, the first chapter of the PV refuting the Mīmāṃsā view of the intrinsic relation between a word and its meaning, by thoroughly remodeling the idea of Kumārila advocated in the BṬ that niyama is the foundation of avinābhāva. Because Dharmakīrti made use of this idea in order to attack the Mīmāṃsā theory of word and meaning, we may rightly assume as follows: If Kumārila, a hard-core Mīmāṃsaka, had first come to be informed of this idea through Dharmakīrti, he would have held it too provocative to be adopted into his own philosophical system, contrary to the assumption of Frauwallner (1962). See the appendix of the present article. Moreover, showing that PV 1.335cd is a quotation from the BṬ, Franco (2019: 132–137) corroborates young Dharmakīrti’s acquaintance with the BṬ. Following Frauwallner’s unacceptable assumption of the relationship between Kumārila and Dharmakīrti, Kataoka (2011: 47–60) errs in assuming Kumārila’s date as ca. 600–650.


Franco (2019) convincingly demonstrates that Dharmakīrti’s dates, which were assumed by Frauwallner as ca. 600–660, cannot be moved back to the sixth century. I surmise that Kumārila passed away about 620 CE (K. Yoshimizu 2015a: n. 1), before coming to know about young Dharmakīrti, whereas Dharmakīrti wrote his first work, the first chapter of the PV refuting the Mīmāṃsā view of the intrinsic relation between a word and its meaning, by thoroughly remodeling the idea of Kumārila advocated in the BṬ that niyama is the foundation of avinābhāva. Because Dharmakīrti made use of this idea in order to attack the Mīmāṃsā theory of word and meaning, we may rightly assume as follows: If Kumārila, a hard-core Mīmāṃsaka, had first come to be informed of this idea through Dharmakīrti, he would have held it too provocative to be adopted into his own philosophical system, contrary to the assumption of Frauwallner (1962). See the appendix of the present article. Moreover, showing that PV 1.335cd is a quotation from the BṬ, Franco (2019: 132–137) corroborates young Dharmakīrti’s acquaintance with the BṬ. Following Frauwallner’s unacceptable assumption of the relationship between Kumārila and Dharmakīrti, Kataoka (2011: 47–60) errs in assuming Kumārila’s date as ca. 600–650.


Franco (2019) convincingly demonstrates that Dharmakīrti’s dates, which were assumed by Frauwallner as ca. 600–660, cannot be moved back to the sixth century. I surmise that Kumārila passed away about 620 CE (K. Yoshimizu 2015a: n. 1), before coming to know about young Dharmakīrti, whereas Dharmakīrti wrote his first work, the first chapter of the PV refuting the Mīmāṃsā view of the intrinsic relation between a word and its meaning, by thoroughly remodeling the idea of Kumārila advocated in the BṬ that niyama is the foundation of avinābhāva. Because Dharmakīrti made use of this idea in order to attack the Mīmāṃsā theory of word and meaning, we may rightly assume as follows: If Kumārila, a hard-core Mīmāṃsaka, had first come to be informed of this idea through Dharmakīrti, he would have held it too provocative to be adopted into his own philosophical system, contrary to the assumption of Frauwallner (1962). See the appendix of the present article. Moreover, showing that PV 1.335cd is a quotation from the BṬ, Franco (2019: 132–137) corroborates young Dharmakīrti’s acquaintance with the BṬ. Following Frauwallner’s unacceptable assumption of the relationship between Kumārila and Dharmakīrti, Kataoka (2011: 47–60) errs in assuming Kumārila’s date as ca. 600–650.


For the synonymy between āyatta and pratibaddha, see Y. Fukuda’s article in the present volume.


See also Ch. Yoshimizu 2011, Appendix 1.

Help me keep this site Ad-Free

For over a decade, this site has never bothered you with ads. I want to keep it that way. But I humbly request your help to keep doing what I do best: provide the world with unbiased truth, wisdom and knowledge.

Let's make the world a better place together!

Like what you read? Consider supporting this website: