Postulating a fruit for a Vedic injunction

According to Mīmāṃsā hermeneutics, a prescription must include a fruit to be realised. This is not an optional element, but an unavoidable building block of its structure, just like the bhāvanā (the action prescribed). Thus, what to do in case no result is mentioned in a Vedic prescription? The Mīmāṃsā answer is the viśvajinnyāya 'the rule about the Viśvajit' sacrifice, so called because it has been elaborated in connection with the case of the Viśvajit sacrifice, which is in fact prescribed without any fruit to be realised through its performance. According to the viśvajinnyāya, whenever no result is mentioned, svarga, i.e., enduring happiness must be postulated as the result. Why exactly svarga? Because it is desired by everyone (unlike particular results, such as sons, cattle, rain, etc.).

A same problem applies in the case of the prescription to study the Veda (svādhyāyo 'dhyetavyaḥ), which also lacks the mention of a result. Vedānta Deśika discusses it in his commentary on Mīmāṃsāsūtra 1.1.1. There he says that one needs to postulate a result which should be desired  by everyone (sarvakāmya), analogically extended (atideśa-āgata) from other sacrificial prescriptions which mention it, obtained through commendatory statements, etc., which name it, and instrumental for the satisfaction of ancestors and gods. The relavant passage reads:

svādhyāyo 'dhyetavya iti vidhes svavākye phalādarśanāt, […] vidhyānarthakyaprasaṅgād viśvajinnyāyena svargaṃ vā bhāvyaṃ parikalpya […] adhyayanacodanāliṅgakātideśāgatajapādhyayanārthavādoktapitṛdevatṛptidvārakaṃ sārvakāmyaṃ phalam upādāya […]

Since one does not see the result (phala) of the prescription ''One has to study one's portion of the Veda'' in the sentence where it is itself [prescribed] (svavākya)[…] since there would be the undesired consequence that the prescription would be meaningless, we have alternatively (vā) postulated a heaven to be realised according to the Viśvajit-rule (and attributed it to the prescription to study). […]
Thus, we have accepted a fruit which is desirable to all, through which gods and ancestors (pitṛ) gain their satisfaction, expressed in commendatory statements (arthavāda), recitation and murmuring, and obtained (āgata) through an analogical extension whose mark is the injunction (codanā) to study.

Why does one's studying of the Veda satisfy gods and ancestors? Because one's study of the Veda is described in the Śatapatha Brāhmaṇa (9.5.6.3--8, but I am copying from Kane) as a Brahmayajña: "4--7 compare ṛk, yajus, sāman and Atharvāṅgiras texts respectively to offerings to gods of milk, ghee, soma and fat and it is stated that the gods being delighted and satisfied with these, bestow on the man who performs brahmayajña affluence and security, life-breath, seed, his whole self and all auspicious blessings and streams of ghee and honey flow for his departed pitṛs" (Kane, vol. 2.1 p. 701).

For further posts on Vedānta Deśika's Seśvaramīmāṃsā, see this one and this one (in Sanskrit).

What is the problem with the ambiguous nature of "action"?

Action (mostly referred to as karman) is a debated topic in Indian philosophy. Why? Because of its ambiguous nature. "Orthodox" schools believe that karman is a separate category, along with substance, quality, etc. However, karman inheres in substance just like quality and, consequently, some thinkers such as Bhāsarvajña or the anonymous objector in Maṇḍana Miśra's Bhāvanāviveka ("Discernment about action") claim that karman is nothing but a quality. The Vaiśeṣika way of distinguishing between qualities and actions is also unconvincing: both are non eternal and inhere in a substance, but only action produces conjunctions and disjunctions. This seems an ad hoc solution: why should one then not postulate a separate category for the quality "colour-and-form", which is the only one which produces visual data? Moreover, the Vaiśeṣika distinction cuts out all kind of action apart from movement. Other authors, such as the Mīmāṃsaka Kumārila, Maṇḍana, Someśvara and Pārthasārathi, reflected further on the nature of action and differentiated it (although with different gradations) from sheer movement. Contemporary Western physics and philosophy have also problems in explaining the phenomenon of "action" as distinguished from "movement" (both gross and subtle).

But why is the issue so important? Because "action" is not just an ontological category. For ontology, "movement" would be enough, and it could be abandoned once one adheres to the contemporary subatomic paradigm, where rest is no longer the "normal" state of affairs.

Action is much a more an ethical and juridical category. To define what an action is means to decide whether one is responsible for it and, lastly, liable to be punished. In this case, sheer movement is not enough. One is not ethically censured for having moved one's finger, but rather for having shot someone. Hence, one would say that ethics adheres to Someśvara's idea of action as effort/volition (see this post and the immediately preceding ones).

On the other hand, jurisdiction does not count only the intentional, conscious side of the action, but also its movement-aspect. One is not legally guilty because of having desired to kill someone if one has altogether failed to do it (for instance, because one was still looking for a gun to buy as one's target died out of a car accident). Similar difficult cases are examined in chapter 5 of Vasubandhu's Abhidharmakośa. Notwithstanding his rigid intentionalistic approach, Vasubandhu also compromises with the actual performance of a bodily movement.

Prescriptions embedding actions

According to the hermeneutic of Sacred Texts devised by the followers of Kumārila Bhaṭṭa, the linguistic bhāvanā constituting the prescription first requires (through expectation, ākāṅkṣā) an object to be brought about, that is, the purpose[-oriented] bhāvanā (arthabhāvanā). This is also conveyed by the prescriptive verbal suffix (just as it would be by any other finite verbal suffix), more precisely by its finite verbal component (ākhyātatva).

In order to fulfil the role of the result that must be accomplished, this bhāvanā must be somehow desirable (the action noun ākāṅkṣā retaining its desiderative shade of meaning). Since, however, the undertaking of an action is not desirable in itself, one must assume that it is indirectly desirable given that it is the means for achieving something desirable. In this way, although sacrificing is not in itself an agreeable activity, one undertakes it as if it were something desirable because one longs for the sacrifice's result, e.g., heaven.

In fact, the purpose[-oriented] bhāvanā also requires an object to be achieved thereby. The first candidate, because of proximity, is the meaning of the verbal root, since it is part of the same verbal form whose ending conveys the objective bhāvanā. However, the meaning of the verbal root is not by itself desirable and cannot hence fulfil the role of the bhāvya. This is, instead, revealed within the same prescription by an expression such as “one-who-is-desirous-of-heaven” (svargakāma), which indirectly indicates the sacrifice’s result, whereby “heaven” can be replaced by “cattle”, “a son”, “rain”, etc.

Immediately thereafter, the question arises “how can this result be achieved?” To answer this, the verbal root is connected to the bhāvanā as its instrument. Hence, by hearing “one who is desirous of heaven should sacrifice”, one simultaneously understands an urging (the linguistic bhāvanā), the contents of this urging (the undertaking of an action or purpose[-oriented] bhāvanā), its result (heaven) and the instrument for its achievement (the sacrifice). But the expectation is not completely appeased, insofar as the mere knowledge that heaven is to be achieved through a sacrifice is not enough. One also needs to know how this sacrifice will be instrumental for the arising of this result. Consequently, a procedure is also required, which is offered through the other sentences describing the sacrifice.

Summing up,

object to be brought about, e.g., heaven

↗

bhāvanā → instrument, e.g., sacrifice

↘

procedure, e.g., rites composing the sacrifice

And:

object to be brought about (heaven)

↗

object to be brought about, e.g.,bhāvanā →instrument (sacrifice)

↗ ↘

śabdabhāvanā → instrument procedure

↘

procedure

Lastly, since the sacrifice would not be a suitable instrument (as it does not last until the arousal of the result, as the Prābhākara point out), a new potency, arisen through the sacrifice and lasting until the result is also necessarily connected as an intermediate element.

Memory and perception

I recently read an interesting article about false memories. It seems that most people, if confronted with a made-up photo DO recall the event which would have been depicted in it. In the experiment, the event was a hot air balloon trip and I am surprised by how little we are sure about –if we can end up being persuaded we have actually lived such an extraordinary experience although we did not. This has important consequences on juridical matters, showing that it is useless (and may lead to distorsions), to urge witness to remember or to rember more clearly what they might have unconsciously seen. Vedānta Deśika makes a similar point in his Seśvaramīmāṃsā on 1.1.4:

[Obj.:] What about the fact that the very topmost level of visualisation (bhāvanā) makes [things] perceptible? [S:] This is not true. Out of visualisation it is not perception which arises, but rather only clearness of memory. In fact, the accumulation of mnestic traces (saṃskāra) supplies sharpness to memory. Even when, for instance, a love-sick meets his beloved one, nothing exceeding what has already been known appears. And the exceeding element appearing in “And on every tree I see a cloth (ambara) consisting in the skin of an antelope (kṛṣṇājina) and a garment (cīra), similar to Rāma with his arch, holding a noose in hand, the destroyer” and similar [verses], this is not directly perceivable, since it appears in a different way. What [is directly perceivable] is, instead, its external look.

Bhāvanā: Kumārila and Bhartṛhari

Bhartṛhari speaks of bhāvanā in his well-known passage about pratibhā (who taught to the cockoo bird to sing? The pratibhā). Bhāvanā should be a synonym of it. The vṛtti glosses it as śabdabhāvanā, since it is essentially linguistic. Abhinavagupta, who surely knew (and sometimes criticised) Kumārila, uses again śabdabhāvanā in the same sense as Bhartṛhari. Is it possible that two so different meanings of śabdabhāvanā co-existed without anyone feeling the need to specify "my śabdabhāvanā is not the same as Bhartṛhari's/Kumārila's one" or the like? I am not talking about bhāvanā, a common name throughout Sanskrit culture, but śabdabhāvanā does not seem as much broadly used (I do not know of other cases of its usage). Finally, could Kumārila have not been aware of Bhartṛhari when he called the linguistic bhāvanā "śabdabhāvanā"? Or did he have it somehow in mind? What are the implications of that?

Summary about the meaning of verbal roots and verbal endings

• According to Śabara, in every verb there is the general idea of "causing to be" something and the concept of a specific activity. They can be loosely connected to verbal ending and verbal root respectively.
• According to Kumārila, both the verbal root and the verbal ending express an activity, and the two are linked as universal and particular: the verbal ending expresses an action in general, the verbal root specifies it.
• According to Maṇḍana, the verbal root only expresses the action's result. The verbal ending alone expresses the action.
• According to Someśvara, the verbal root expresses the movement caused by effort, which is expressed by the verbal ending.
• According to Pārthasārathi, the verbal root and the verbal ending may have different specific meanings. The bhāvanā is however in every case characterised by the fact of leading to the production of something else (possibly, unlike the meaning of the verbal root).

Again on the argumentative structure of Pārthasārathi's Śāstradīpikā

So, points 1 to 3 and point 7 are definitely a pūrvapakṣa. Instead, points 4 to 6 are agreed upon by the siddhāntin as well. Points 8-9 are the transition to the siddhānta, opening at point 10.

Why did Pārthasārathi choose such a complex argumentative structure?

It is typical of him that he lets his own view merge into an uttarapakṣin's one (so, for instance, in the Nyāyaratnamālā, while dealing of śabdabhāvanā). He could thus have meant 1-3 as an objection becoming gradually more plausible. Then, 4-6 are definitely his view. 7 continues one of the elements of the 1-3 objection and 8-9 express it fully. 10 rejects it and the following part of the text expresses Pārthasārathis' view.

Is an action expressed by the verbal root?

Point 1. and 10. (listed in the above post) of Pārthasārathi's argument about bhāvanā –that is the assertion that the bhāvanā is the general form present in all verbal roots' meanings and the concept of its separateness from the meaning of verbal roots– contradict each other.

The siddhāntin's view seems to be the latter, since Pārthasārathi himself asks (as any reader would have done): "But isn't it the case that the bhāvanā is exactly the general form common to verbal roots' meanings?" (nanv evaṃ dhātvarthasāmānyam eva bhāvanā syāt). The answer is no (neti brūmaḥ), "since the bhāvanā is different according to every single verbal root" (vilakṣaṇatvāt pratidhātvartham). In fact, a different bhāvanā is involved in cooking or in walking (anyad dhi pākasyaodanaṃ praty anukūlyam anyac ca calanādeḥ saṃyogādi prati, phalabhedāt).

But the answer does not address the more general point of the link between the bhāvanā and the verbal root's meaning. A bhāvanā specific to every verbal root may still be closely related to it. Luckily enough, Pārthasārathi adds a further argument. This claims that since the same thing (e.g. "decision") can be expressed by a verbal ending (as usually the case) or by a verbal root ("to decide", saṅkalp-), the bhāvanā cannot be the general form present in all verbal roots.

Even in this case, what one understands is that Pārthasārathi disagrees with the idea that a bhāvanā expresses an action in general (so to say, the general form of the action, the sāmānya) and that the verbal root connected to it expresses the specific action at stake in that case (the viśeṣa). So, Frauwallner is wrong in maintaining that Pārthasārathi totally agrees with Kumārila ("schon diese kurze Wiedergabe zeigt deutlich, daß sich Pārthasārathi bei seiner Behandlung der bhāvanā in den Grundzügen aufs stärkste an Kumārila anlehnt" Frauwallner 1938: S.245), since the latter explains the relation of bhāvanā and verbal root as one of universal/particular.

Finally, Pārthasārathi concludes "hence, it is established that the bhāvanā is different from the meaning of the verbal root" (tasmāt siddhaṃ dhātvarthātiriktaṃ bhāvanārūpam). Hence, the meaning of the verbal root is (just) the instrument of this bhāvanā (tatra ca dhātvarthaḥ karaṇam). Out of it is the apūrva also postulated (tato 'pūrvaṃ kalpyate).

Structure of Pārthasārathi's argument on action

I have already noted that Frauwallner's critique of Pārthasārathi may have been a little bit ungenerous insofar as Frauwallner seems to overlook the siddhāntin/pūrvapakṣin distinction in the Śāstradīpikā text about bhāvanā (ad 2.1.1, bhāvārthādhikaraṇa, pp.129-131 of the 1978 edition). Frauwallner may have overlooked it because Pārthasārathi is often (so also in his Nyāyaratnamālā) extremely cautious in marking the end of a pūrvapakṣa, so that it is often quite difficult to distinguish his own opinion.

In the following, the numeration follows Pārthasārathi's first enunciation of the themes. Point 1 clearly enunciates the pūrvapakṣa and point 10 the siddhānta. But where does the siddhānta start?

1. Pārthasārathi starts with a pūrvapakṣa stating that the bhāvanā is not distinct from the meaning of the verbal root (satyaṃ na dhātvārthātirekiṇī bhāvanā).

2. In fact, the meaning of the verbal root has two components: one which is not an action and has an established nature (dhātvārthānām eva tu dve rūpe […] ekam akriyātmakaṃ siddhasvabhāvam) and another one which is an action and has still to be realised (aparaṃ tu […] kriyātmakaṃ sādhyasvabhāvam).

3. The latter can be expressed by 'karoti' (he does), is present in all verbal roots and expressed by the verbal ending (sarvadhātvarthānuyāyi karotipratyayavedyaṃ […] ākhyātābhidheyam) (tac cākhyāteṣv avagamyamānatvāt tadabhāve pākādāv anavagamād ākhyātābhidheyam āśrīyate) (tac ca sarvadhātvarthānām astīti sarvavyāpy eva bhāvanārūpam).

4. Further, it is favourable to the production of something else. This means that it leads to the production of something else beside the action (which is itself to be realised), although it is not tantamount to the production of something else –which would be an external act, not the inner disposition towards it– (anyotpādānukūlātmakam) (tac ca rūpam utpādyavastvantarakarmakam eva sarvadā pratīyate) (ato yadavastho yāgaḥ svargaṃ sampādayati tadavastho yāgo yena vyāpāreṇa sampadyate sa vyāpāraḥ svargakṛtir ity evaṃrūpeṇākhyāyābhidheyo bhavati).

5. It also entails an actor (tad eva ca bhavituḥ prayojakavyāpāro bhāvanety ucyate).

6. It is not tantamount to effort, nor to movement, not to both (ye 'pi prayatnaṃ bhāvanām āhuḥ, ye ca spandaṃ, ye cobhayaṃ, te 'pi na svarūpeṇa teṣāṃ bhāvanātvam āhuḥ, kintv anyotpādānukūlatvarūpam).

7. To that, the meaning of the verbal root is connected either as an object (e.g. "he does cooking") or as an instrument (e.g. "through cooking she does a rice-cake") (tasmiṃś ca kriyārūpe prakṛtyupātto yāgādiḥ kadācit karmatvena saṃbadhyate pākaṃ karoti yāgaṃ karotīti, kadācit karaṇatvena yāgena svargaṃ karoti pākenaudanam iti yathāvivakṣitam).

7.1 The meaning of the verbal ending also varies, depending on how the meaning of the verbal root is connected to it (tad yadā yāgādeḥ karmatvena saṅgatis tadā karmabhūtaḥ prayatnaḥ saṅkalpādir vā dhātvantarārtho 'pi yajetety atra pade yāgakṛtir ity evaṃrūpeṇākhyātābhidheyatāṃ bhajate. saṅkalpādikaṃ tu viśeṣarūpaṃ dhātvantarābhidheyaṃ pramāṇāntaravedyaṃ nākhyātena spṛśyate. yadā tu vidhivaśāt karaṇatvena yāgasya saṅgatiḥ, karmatvaṃ tu svargāder darśapūrḅamāsayāgena svargaṃ kuryād iti, tadā na yāgasya hetubhūtaiva kṛtir ākhyātenocyate, yāgasya karmatvābhāvāt, phalasyaiva karmatvāt tatkarmikā kṛtir ākhyātenocyate, yāgas tu karaṇam).

8. The verbal ending does not express the specific form of the action (tasyāpi vyāpārasya viśeṣarūpaṃ pramāṇāntaravedyaṃ nākhyātena spṛśyate).

9. The bhāvanā conveyed by a Vedic injunction stands in need of a further element, the procedure conveyed by nearby textual passages (vyāpāraviśeṣātmakakathaṃbhāvāpekṣāyām agnyanvādhānaprayājāvaghātādibhir upakārasampādanaṃ śāstrāntarair avagamyate, yāgasvarūpasya ca sampādanam arthapāttyā).

9.1 The procedure is instrumental to the action (which is instrumental to the arousal of the result), just like raising and sinking the axe is instrumental to chopping a wood-log into two (etad eva hi karaṇanām [so in the ed., I would emend into karaṇānāṃ] karaṇatvaṃ yaduta phaloddeśapravṛttakartṛvyāpāreṇa phalasādhanātmanā sampattiḥ, paraśor ivodyamananipatanābhyāṃ dvaidhībhāvasādhanatayā).

10. "the bhāvanā realised through these [procedure] is distinct from the meaning of the verbal root" (tat siddhā dhātvarthātiriktā bhāvanā).

Pārthasārathi on action

In his acute paper on Bhāvanā (Bhāvanā und Vidhiḥ in Maṇḍanamiśra. I. Bhāvanā, WZKM 1938), Erich Frauwallner seems irritated by Pārthasārathi's inconsistency in dealing with the issue of action (bhāvanā). Still, looking closer at Pārthasārathi's Śāstradīpikā, one sees that it encloses a long objection (pūrvapakṣa) and that the inconsistencies Frauwallner notices are between the objector's view and the one which is allegedly Pārthasārathi's own one. Is this enough to "rescue" Pārthasārathi's view? Or does the idea of the verbal root expressing an activity make it too similar to the meaning of the bhāvanā?

Maṇḍana's definition of action

In his Bhāvanāviveka, Maṇḍana reproduces the view of an objector who maintains that karman is not a separate category but just an unnecessary postulation. All that we can see is conjunctions and disjunction (which are qualities) and there is no necessity to postulate anything else beyond that. Maṇḍana criticises this view and then presents his theory of bhāvanā. Due to this sequence, it is easy to argue that he considers bhāvanā as an alternative explanation of movement.

He repeats Kumārila's definition of bhāvanā as the “interruption of a previous state of rest (audāsīnya)”, and interpreted this “rest” rather loosely, in both physical and psychical sense. Accordingly, he further defins the bhāvanā as consisting in effort (prayatna) and movement (parispanda). In this way, mental activities can also be labelled by him bhāvanā). Finally, Maṇḍana explicitly acknowledges the differences between his position and the Vaiśeṣika one:

For us, it is not so that movement alone is the only action (kriyā), as it is [instead] for Kaṇāda (the author of the Vaiśeṣikasūtra, the foundational text of the Vaiśeṣika school).

Due to this double criterion, Maṇḍana's definition of action (bhāvanā) can apply to both movement and effort without recurring to metaphorical usages. It can hence easily explain cases such as "the chariot goes", where no effort is seen.

As for the objector's claim, one could integrate Maṇḍana's exposure by making explicit that conjunctions and disjunctions inhering in a thing as their qualities are instead just the movement's result.

Did Mīmāṃsā authors develop a general theory of action?

The Mīmāṃsā theories on bhāvanā are originally exegetical/linguistic devices. But do they also constitute or presuppose a general theory of action? My provisional answer is yes, insofar as 1. they are used not just in an exegetical context (Kumārila mentions examples such as "(s)he cooks", Someśvara speaks of chariots), 2. they are used to address topics such as intentionality, effort, atoms' movements, which have nothing to do with both exegesis and linguistic analysis 3. they are used in debates against opponents of other schools who speak about movement and do not share the Mīmāṃsā theory of bhāvanā.

Movement is not just conjunction and disjunction

Maṇḍana Miśra, who lived short after Kumārila, wrote a separate study on bhāvanā called Bhāvanāviveka (“Discernment about bhāvanā”). Apart from its intrinsic philosophical value, this text is also precious because it reproduces many objections raised against Kumārila's theory but not otherwise recorded. The main objector maintains, à la Zeno of Elea, that movement just does not exist. It is nothing but the conjunctions and disjunctions (saṃyoga-vibhāga) inhering in a thing (BhV, p.25). For instance, walking is nothing but conjunctions and disjunctions inhering in feet and ground. In fact, only conjunctions and disjunctions are, indeed, seen (pp. 29-30¹) and one then infers movement out of them (pp.33-34) at a time when the alleged movement has actually already vanished and one only sees a new disjunction or conjunction (Uṃveka, p.35, BhV, p.35). But, since the latter are enough to explain what one sees, why should one postulate a further entity?² One could complete the argument by specifying that the illusion of a separate category, movement, is just created by the succession of disjonctions and conjunctions.

This interesting position is criticised by Maṇḍana, who upheld the specificity of activities. Since conjunctions and disjunctions do not cease to be there, argues Maṇḍana, one would keep on inferring an activity even when one pauses after having walked³.

1api ca pratyakṣāpratyakṣavṛttyor api saṃyogavibhāgayoḥ siddhayoḥ pratyakṣatvakalpanā yuktā na tv asiddhasya karmaṇaḥ. tābhyām eva tarhi saṃyogavibhāgābhyāṃ kriyām anumimīmahe (BhV p.30, ll.1-3).

2tasmād guṇaviśeṣa eva dhātūpādānaḥ kriyā na tu tadatiricyamānātmā kriyāpadārthaḥ, yaḥ pratyayasya dhātor vābhidheyaḥ syāt (BhV, p.35). And siddher guṇaviśeṣeṇa pacatīty api saṃvidaḥ | kriyāpadārthasyānyasya nānumānaṃ prakalpyate || 24 || (BhV, p. 36).

3na caiṣa tajjanyābhimatasaṃyogavibhāgālambana eva pratyayaḥ. […] na hi calitvā sthitasya kriyāprabhavapūrvottaradeśasaṃyogavibhāgābhāvaḥ. atas tadālambano vyāpārapratyayo na jātu viramet (BhV, p. 83, ll.6-12).

Action as Effort

I am changing my mind about what Kumārila meant with arthabhāvanā. In fact, he defines it as parispandarūpam ("having the nature of a movement") and never uses the term prayatna (effort). Moreover, Maṇḍana, who followed Kumārila in admitting two different forces, described the bhāvanā as meaning both effort and movement. It is only with Someśvara that the definition of bhāvanā as exclusively effort becomes current. This does not clash with Kumārila, who defined the bhāvanā as "activity of the promoting subject", thus preparing the equation with effort. Still, it is not Kumārila's own elaboration. But to detect an innovation in Someśvara makes his claim that arthabhāvanā means "the force caused by the desiring [subject]" (see previous posts) less plausible. It is easier to figure out that Someśvara (consciously or not) innovated in regard to the equation bhāvanā/effort and, consequently, interpreted arthabhāvanā as synonym of prayatnavantaḥ bhāvanā (the activity of one who makes an effort). On the other hand, the definitions of bhāvanā by Kumārila seem still less precise.

Bodily movements non caused by efforts

Someśvara is aware of the fact that movements are not enough to detect an intentional action (which he links with effort, prayatna). On the one hand, one can meditate and hardly move, on the other, there can be unintentional movements.

In case a bodily movement is not intentional (in case of a illness, for instance), then it cannot be said to be caused by an effort, maintains Someśvara. One can say that "there is a movement" (spando bhavati), not that "(s)he makes a movement" (spandaṃ karoti).

What about the well-known example "the chariot goes"? Someśvara replies by saying that linguistic use does not always reflect reality. In fact, in khaṭvā (couch) there is a feminine suffix (-ā, technically called ṭāp), although no feminine entity is there, "in the same way, in regard to an unconscious [entity] a verbal suffix is employed although there is no effort" (Nyāyasudhā 1909, p. 579).

Someśvara on effort and the meaning of arthabhāvanā

Someśvara aptly connects effort (prayatna) and its result, bodily movement (śarīraparispanda). The former implies even subtler forms of effort, since among the examples mentioned there are not just going or cooking or sacrificing, but also dhyāna (meditation): dhyāne 'ham prayate (Nyāyasudhā 1909, p. 577). Hence, Someśvara can consequently explain artha- in arthabhāvanā as referring to the subject who strives for a certain result (arthayate). He might be historically wrong, insofar as his interpretation of the bhāvanā as mainly located in the person who strives for something is not necessarily the same as Kumārila's one. The latter seems to be still more inclined to a concept of bhāvanā as movement (or activity, vyāpāra). Still, Someśvara's proposal is consistent and interesting.

Effort and other minds

Someśvara explains that every bodily movement (śarīraparispanda) is caused by an effort (prayatna). In Nyāya-Vaiśeṣika texts, this link was used in order to prove the existence of a Self, insofar as prayatna was believed to be a quality of the Self (and, hence, inseparable from it). Hence, by seeing other people moving one could infer the existence of their Selves (in contemporary terms: the existence of other minds).

Someśvara develops an inference to the same purpose in the following way. In fact, one knows through mental awareness (mānasapratyakṣa, the kind of direct perception allowing one to seize one's own inner states) that one's bodily movements while going, sacrificing, cooking, etc., are caused by the self's effort. So:

1. all movements of one's body (svaśarīraparispanda) are caused by the self's effort (ātmaprayatna), as one knows through mental awareness (mānasapratyakṣa),

2. other bodies also move,

3 hence, their movements (paraśarīraspanda) are also caused by a self's efforts, just like in one's own case (Nyāyasudhā 1909, p. 577).

The rationale behind it is that one is aware of something else beyond the movement one performs in sacrificing, etc. (which are expressed, according to Mīmāṃsā authors, by the verbal root).

Is effort a mental action or a quality of the self?

Nyāya and Vaiśeṣika authors maintain that effort (prayatna) is a quality of the self (ātman). This implies that actions are just bodily ones, involving atoms' movements. Since, in fact, the self is composed of just one atom, its changes cannot be explained as atoms' movements.

The theory is plausible, although it reflects a rather static conception of the Self (as if changes were only accidental to it).

On the other hand, Mīmāṃsā authors developed a theory of bhāvanā which seems not to differentiate between bodily and mental actions. In this way, they can make sense of Vedic prescriptions as exhortations to action even in case one does not immediately perform the action enjoined. Someśvara (12 c.?) explicitly says that prayatna is a kriyā. On the other hand, Rāmānujācārya lists it together with the guṇas. But, he also says that bhāvanā is prayatna (!). This might mean that a certain point of the history of Mīmāṃsā confusion reigned in this regard, or that Mīmāṃsā authors just repeated the list of guṇas they found in Nyāya texts, without thoroughly reflecting about its implications.

Action in Indian philosophy

Broadly speaking, the theme “action” is recurrent in Indian philosophy in at least three context: 1. Action as an evidence for the existence of a Self, 2. Action in the context of language use, 3. Moral Action (I will not distinguish between moral and ethics).

1. Action as an evidence for the existence of a Self

Indian darśanas share commonly agreed notions and terms. One of those is the idea of categories as actually existing classes into which the reality can be organized. The most ancient ones are probably substance, quality (guṇa) and movement (karman). The latter is of direct relevance for the present study. The Vaiśeṣika system, which fixed as such at an early stage (so Frauwallner), does not sharply distinguish between movements and intentional actions. Both are described as movements of atoms (which are conceived as the minutest particles of matter, unchangeable and eternal). Their difference lies only in the fact that the first ones are produced by material causes, the latter by effort (prayatna). Their development, though, follows the same causal scheme. Prayatna is again determined by volition and that by desire or aversion. A similar causal scheme is adopted by Nyāya and later Vaiśeṣika as well. As remote causes of desire and aversion either non-knowledge (erroneous knowledge, leading to an erroneous attachment to worldly things) or dharma and adharma (merit and demerit) are mentioned. Nyāya authors make the string longer by adding before the remote cause a “connection with a recollection” (smṛtyanubandha) as a determinant for the arousal of volition and then of efforts.

What is an action?

Śabara introduces the label bhāvanā to define an action as something which "causes to be" a result, such as rain in the case of a sacrifice performed in order to get rain.

Kumārila further distinguishes between: 1. activities inhering in a conscious agent (arthabhāvanā) and producing a result; 2. activities non producing a result (and possibly inhering in a non-conscious agent as well), such as "to dwell"; 3. activities inhering in words (śabdabhāvanā) and causing the first ones to be.

His commentators debated about the understanding of those categories. Maṇḍana abolished the distinction between 1. and 2. and defined bhāvanā in general as "prayatna", effort. Someśvara repeated this definition, whereas Pārthasārathi Miśra kept to the older definition and preferred to describe a bhāvanā as anyotpādakavyāpāra, an activity producing something else. The Bhāṭṭa authors successive to them all sided with Someśvara or Pārthasārathi. The first ones explained that, though a bhāvanā typically inheres in a conscious agent, it can only be metaphorically said to inhere in an unconscious one.

Prābhākara authors reply by describing alleged actions as just successions of conjunctions and disjunctions with various locations in the space.