somananda e-book

Also, there is the (question of the) reality of the means of knowledge that establishes her. If it were real, there would be two real entities.334 If unreal, then there is nothing at all (that can be established).335

Also, if you maintain that the means of knowledge that establishes her is real, then that incurs the reality of two entities, paśyantī and the means of knowing her.

If the means of knowledge were unreal, nothing at all could be established.

2.69cd–71

sādhuśabdasamuccārāt kasya svargādiyogitā

2.70. paśyantyāś ced avidyātvaṃ tadbhogaunmukhyayogataḥ madhyamāder jaḍāyāḥ kiṃ bhogena śabalātmanaḥ

2.71. tasmād asādhuḥ sādhuḥ syāc chabdavidyāphalapradaḥ evaṃ vyākaraṇasyāpi samuccheda upaiti te

For whom does the uttering of correct speech lead to heaven, etc.?336 If for paśyantī, she would have nescience as her nature, because she would be associated with the eagerness for those enjoyments.337 If it is for the variegated madhyamā, etc., which are insentient, what is the use of the enjoyment?338 On that account, the proper [speech], which gives the reward for knowing speech, would be improper; in this way, even your grammar would dissolve.

The grammarians have said that the use of correct speech leads to heaven and liberation: “A single word used in accordance with the science (of grammar) becomes a wish-fulfilling cow in the heavenly world.”339 And:340

The supreme good nature of Prajāpati operates in those men in whom proper, good speech is established. The great energy, which has Prajāpati’s

334That is, both paśyantī and the means by which she is known would be real, thus creating an unwanted dualism.

335This passage takes up the question of the status of the means of knowledge, the first of two questions Somānanda raises in objection to the grammarians’ position, summarized in ŚD 2.67cd–68ab, viz., that everything other than paśyantī is unreal.

336In this passage, Somānanda considers the problems associated with an unreal agent of action (kartṛ), the second of a pair of questions raised in ŚD 2.67cd–68ab. Somānanda addressed the first, the question of the problems associated with the ontological status of the means of knowing paśyantī, in ŚD 2.68cd–69ab.

337That is, she would be associated with the desire to take up a place in heaven, etc. If paśyantī is granted liberation, then she must be bound in the world of transmigration (saṃsāra) prior to liberation, which by definition means she is ignorant (avidyātva) prior to her enlightenment.

338The point here made is that the various enjoyments would not have any use because madhyamā and the rest are insentient, and an insentient entity cannot enjoy heaven or other rewards. Note that while both śabalātman and madhyamādi are masculine, jaḍā (ŚD 2.70c) is a feminine adjective that reflects the gender of madhyamā.

339The present passage closely parallels Patañjali’s Mahābhāṣya, commentary on A 6.1.84.

340The following is a quotation of VP 1.126–127.

nature, is concealed by the subtle elements alone;341 (but,) when wise people die,342 it returns to its own source.

First of all, who is this so-called343 agent who uses proper speech, to whom the fruit of that [action], called “heaven,” accrues? If paśyantī, being intent on that reward, is the agent, she would be impure because, since she would be eager to enjoy heaven, she would (necessarily) be in contact with nescience.344

(Objection:) The conduct345 takes place in the madhyamā and vaikharī condition(s), at which346 one can perceive what is desired and not desired, because one is eager (there) to acquire and avoid (such entities).

Reply: Not so, for the supreme agent of cognition is separated from the objects;347 thus, this means that the very enjoyment in question could not arise, because, when a dualistic condition exists as a result of madhyamā and the rest348 having distinct parts, they349 are insentient insofar as they exist as the variegated form of nescience, and as a result there could be no enjoyment whatsoever that leads to the level of the supreme reality.350

Therefore, a proper word used, which yields enjoyment, etc.,351 i.e., the fruit (gained) from the knowledge of proper speech, would on the contrary be improper, that is, impure, because paśyantī would acquire an impurity.352 And

341Mention of the “subtle elements” (tanmātras) here refers to the five elements, speech, touch, form, taste, and smell. They correspond with five organs of sense (jñānendriyas) capable of cognizing each. Moreover, they are the subtle qualities associated with the five gross elements (mahābhūtas), ether, air, fire, water, and earth. The correlation with the mahābhūtas is cumulative: ether reflects the presence of the first tanmātra, air of the first and the second, fire of the first three, etc.

342Literally, “when there is the separation [bhede] of wise men [viduṣām] from (their) bodies [śarīra].”

343The present phrase is a rendering of iti.

344The point here expressed is that if paśyantī herself is the agent who uses proper speech, then she would be expectant of and striving for the rewards associated with the use of proper speech, i.e., heaven and liberation. If this were true, then paśyantī would be impure: she would have to be in contact with the world, which requires a connection with nescience. What is more, she by definition would be ignorant, as an omniscient entity does not strive for liberation.

345That is, the conduct that is necessary for acquiring rewards and liberation takes place in a condition inferior to that of the (purportedly) non-dual paśyantī.

346Here, following Harunaga Isaacson’s suggestion, I read iṣṭāniṣṭopalambhabhāji as a compound in the locative, singular, literally meaning “that which takes part in the ascertaining of desired and undesired [entities].”

347That is to say that there is no contact with mundane things at the level of paśyantī.

348This refers to vaikharī, the third level of speech.

349This refers to madhyamā and vaikharī.

350The idea here expressed is that, insofar as madhyamā and vaikharī are insentient, they cannot experience the delight of attaining the rewards conferred on those who use correct speech. They cannot experience heaven, etc., because they are insentient, their insentience being the logical consequence of their being illusory, the very appearance of nescience. Finally, they are nescience itself insofar as they are multiple, variegated. In short, the problem in question is again one of the relationship of the divine agent to the (unreal, according to Somānanda’s read of the grammarians’ view) world he manifests. The divine cannot properly have contact with an unreal world, and he cannot experience the delight of escaping that world when he is not somehow directly involved with that world, as he is in Somānanda’s view.

351The term ādi here refers to heaven and liberation.

352That is to say that she would acquire the impurity of duality.

in this way, because she would make one impure,353 even your grammar, the fruits of which being as you described them, would dissolve, that is, would be something absolutely to be avoided.

2.72–73ab

2.72.vaiyākaraṇatāṃ tyaktvā vijñānānveṣaṇena kim bhavatām aprastutena na kevalam ihoditam

2.73.vijñānābhāsanaṃ yāvat samīkṣāyām udāhṛtam

Having given up being a grammarian, what is the use of your investigation of cognition, which is not (even) the matter at hand for you?354 It is not only here355 that you state your false knowledge, but you also declare it in the (Śabdadhātu-)samīkṣā.

Having given up being a grammarian, which is nothing but the activity of teaching correct words, which cause the cognition of meaning, there is no use whatsoever for your investigation—which is not (even) the matter at hand for you, i.e., which you are not supposed to do—into “correct cognition,” one that your śāstras, which aim at spiritual liberation, pursue.

Moreover, it is not only here356 that the learned Bhartṛhari speaks of his false “correct cognition” by referring to paśyantī, but he does so in the Śabdadhātusamīkṣā as well.357

2.73cd–74ab

dikkālādilakṣaṇena vyāpakatvaṃ vihanyate 2.74. avaśyaṃ vyāpako yo hi sarvadikṣu sa vartate

Omnipresence is destroyed by spatial, temporal, and other attributes; for the one who pervades inevitably exists everywhere.358

353That is to say that she would make the practitioner impure.

354As Utpaladeva explains, grammarians are supposed to concern themselves with grammar, not the processes of cognition and other such philosophical concerns.

355It appears that with iha, meaning literally “here,” Somānanda refers to the VP, and perhaps also to it’s primary commentary, the VPVṛ (if Somānanda knew it).

356As in the verse, “here” (atra) probably refers to the VP and VPVṛ, the authorship of both being attributed by the Pratyabhijñā authors to Bhartṛhari.

357Note that ŚD 2.73cd–74ab is not a quotation of the ŚDhāSam but a criticism of it. Utpaladeva provides the quotation in full in his commentary.

358Literally, the text says it exists “in all the directions” (sarvadikṣu). Here, Somānanda criticizes Bhartṛhari’s description of Brahman in a passage of the ŚDhāSam that is identical to NŚ 1.1: dikkālādyanavacchinnānantacinmātramūrtaye / svānubhūtyekamānāya namaḥ śāntāya tejase. (See Iyer [1969] 1992: 10.) In referring to this passage, Somānanda wishes to criticize the grammarians for their inability to explain how their monistic, transcendent Brahman can be manifested in such a manner as to provide a soteriological experience for the individual practitioner. His approach here is rhetorical: he criticizes the terminology that Bhartṛhari uses in the above-quoted propitiatory verse (maṅgala). In ŚD 2.74cd–75ab Somānanda questions the possibility of the unity of the practitioner’s experience. He criticizes the possibility that an endless deity can have any experience in ŚD 2.75cd–76, where he also questions the apparent

dikkālādyanavacchinnānantacinmātramūrtaye svānubhūtyekamānāya namaḥ śāntāya tejase

Homage to the image of pure, endless consciousness, which is not limited by space, time, etc., to the one of whom the only means of knowledge is one’s own experience, to (the one in the form of) energy, the one who is peaceful.359

As a result of the aforementioned attributes,360 delimitation by space, that is, place,361 and time exists, i.e., the fact of being distinguished exists, which is prohibited. This, moreover, is not logically coherent: since the meaning is that there is one who is spatially and temporally limited,362 it must be delimited (by them). Otherwise, it would not be omnipresent, etc.,363 when all of the spatial, temporal, and other attributes exist; for, the one who pervades inevitably is present everywhere364 and should be eternally delimited by all (moments of) time (past, present, and future).

With (the word) “omnipresence” [Somānanda] has pointed to something that he has not explicitly taught.365

2.74cd–75ab

If they maintain that spatial and temporal limitations do not exist, then they should state as much explicitly and not vaguely, and even that is not viable. Thus, [Somānanda] says:

svānubhūtir vartamānakālenāsya vibhāvyate

2.75. evaṃ kālānavacchedaḥ katham asyodito hi taiḥ

contradiction in describing the deity as both peaceful and active. Finally, in the same passage he takes issue with Bhartṛhari’s reference to a material icon (mūrti) as representative of the divine. Somānanda’s underlying argument is that the grammarians cannot account for the relationship between an eternal deity and a world of temporally and spatially diverse entities.

359This is NŚ 1.1, quoted by Utpaladeva to open his commentary on ŚD 2.73cd–74ab.

360Kaul defines a characteristic (lakṣaṇa) as the narration of an entity’s peculiar nature: asādhāraṇas-

varūpakathanaṃ lakṣaṇam.

361Cf. ŚD 2.48a–b and the relevant commentary for the use of the terms here in question, viz., dik and deśa. Here, Utpaladeva is merely making clear that dikkāla (ŚD 2.73c; NŚ 1.1a) means what Somānanda intended it to mean when he twice used the term dikdeśa in ŚD 2.48a–b.

362Parimitadeśakāla is a bahuvrīhi compound referring to Brahman.

363“Etcetera” (-ādi) here refers to the eternality of Brahman.

364As mentioned above in the notes to Somānanda’s text, the term here used, sarvadikṣu, literally means “in all the directions.”

365This is to say that, just as Brahman’s omnipresence is impossible in the grammarians’ system, so too is its eternality. Thus, Utpaladeva suggests that by using the word “omnipresence” (vyāpakatva) Somānanda secondarily indicates (upalakṣaṇa) as much. Therefore, when Somānanda says dikkālādilakṣaṇena vyāpakatvaṃ vihanyate in ŚD 2.73cd, he wishes to refer to the destruction of Brahman’s omnipresence and eternality both.

The experience one has of it366 appears in the present. Thus, how can they say that it is not limited by time?367

Indeed, the one who dwells in the world of transmigration, the agent of cognition, is located in a moment of time, and it is only in that present moment of time, since he is not situated at that time in the future, etc.,368 that the experience one has of it is observed. Thus, how is it not limited even by temporal distinctions?369

2.75cd–76

anantasyānubhūtiḥ kā paricchedaṃ vinātmanaḥ 2.76. anante ’vagamaḥ kutra tejastve śāntatā katham asarvagapramāṇaṃ hi mūrtir no lakṣyate citaḥ

What experience can there be of the endless [Brahman] without its being divided? Where is there understanding in an endless entity? If it is energy, then how can it be peaceful? For, consciousness does not appear as having a limited measure or as an image.370

Since six ontological categories are accepted in the doctrine of the dualists,371 space and time must exist.372 Therefore, endlessness must be said to be the cessation of space and time,373 and thus what experience is there of one of indeterminate physical and temporal location,374 which means that it is not appropriate for the two to be connected to it.375

366That is, of Brahman as described in the ŚDhāSam (NŚ 1.1), where Brahman is said to be “[consciousness] that is not limited by time, space, etc.” (dikkālādyanavacchinna ).

367Somānanda again criticizes Bhartṛhari’s description of Brahman in the ŚDhāSam. See ŚD 2.73cd–74ab and Utpaladeva’s commentary on the same.

368The present passage remains a bit unclear, and my translation reads as if the word order of the text were tadā bhaviṣyadādinā rather than bhaviṣyadādinā tadā.

369It is impossible for Brahman not to be temporally distinguished if one has an experience of it in a particular moment of time. Otherwise, Brahman would cease to exist at certain moments of time, i.e., when one has an experience of it. See also ŚDVṛ ad ŚD 2.75cd–76.

370Somānanda continues his critique of NŚ 1.1, here questioning Bhartṛhari’s description of Brahman as endless (ananta), the possibility of an energetic (tejastva) deity being peaceful (śānta), and the possibility of the existence of a stone image that is “pure consciousness” (cinmātramūrti).

371That is, in the system of the Naiyāyikas, etc. Kaul glosses “in the doctrine of the dualists” (bhedavāde) with “in the thought of the Naiyāyikas, etc.” (naiyāyikādimate). Here, Utpaladeva is appealing to the authority of the Nyāya in explaining the nature of the mundane world. The term dhātu normally refers to the five elements, the mahābhūtas, but later on (in Yājñavalkyasmṛti) a sixth element, Brahman, is added to the head of the list. The six categories, then, are: (1) Brahman, (2) “ether” (ākāśa, kha), (3) “wind” (vāyu, anila), (4) “fire” (agni), (5) “water” (jala), and (6) “earth” (pṛthivī, bhū). See YājSmṛ 3.145ab.

372Literally, the text says “there must be a connection with space and time.”

373This present compound, deśakālaparyavasāna, could be translated “the limit of space and time.”

374Aparyavasitadeśakāla is an exocentric (bahuvrīhi) compound that refers to Brahman.

375In other words, Brahman is not connected with either space or time. Alternatively, the passage in question could mean that the individual’s experience of Brahman and Brahman itself are not connected.

For instance, where in an endless thing, i.e., in what portion, can there be understanding in the absence of such376 division of it, i.e., its nature,377 given that its parts are indeterminate?378 Yet, the cognition of endlessness appears as a grasping determinately only of a thing whose beginning and end and length and width cannot be seen, having first excluded things that can be seen completely, but it is not possible for an endless, undivided form to appear. Indeed, cognition is certainty regarding finite379 forms.

If it has an energetic form,380 (then,) given that it is not the case that it does not appear,381 how can it be peaceful? So, what is (the meaning of) this statement: “(homage) to (the one in the form of) energy, the one who is peaceful?”382

Also, what is the use of figurative speech? Consciousness is not a substance of a limited measure, nor is it an image, i.e., a solid entity. So, why do you say “(homage) to the image of (pure, endless) consciousness?”383

2.77

2.77. atraiva śabdanityatvavādino rūḍhatāṃ gatāḥ anādinātha tenaiva śabdatattvena tulyatā

Those who profess the doctrine of eternal speech have become famous for this very proposition.384 Now, there is a similarity (of this view) with the very one who is beginningless, who has speech as its true nature.385

Others, themselves grammarians, hold that the eternal [Brahman], which has speech as its nature, is absolutely different from sphoṭa, and they are famous for that very proposition.386

376Kaul glosses iyattā, literally “the fact of being so much,” with “by space and time” (deśena kālena ca).

377Utpaladeva here glosses ātman (ŚD 2.75d) with svarūpa.

378In other words, it is impossible to experience an endless entity, because there would be no cognizable part of it to cognize. A single, endless entity, because it lacks in every distinction, cannot be registered in experience.

379Literally, the text says “fully concluded.”

380Here, Utpaladeva suggests that the grammarian might argue, in a manner similar to the Śaivas, that the divine appears in many forms because it is active.

381Nirābhāsatvābhāva literally means “the nonexistence of the state of being without appearances.”

382See NŚ 1.1d, quoted in ŚDVṛ ad ŚD 2.73cd–74ab. The point is that Brahman cannot simultaneously be energetic and peaceful.

383See NŚ 1.1b, quoted in ŚDVṛ ad ŚD 2.73cd–74ab.

384Literally meaning “in this alone,” I here render atraiva idiomatically with “for this very proposi-

tion.”

385Somānanda here suggests that the Śabdanityatvavādins, who hold that Brahman is speech but is distinct from sphoṭa, can be criticized in the same manner as the Sphoṭanityatvavādins, those who hold that Brahman is speech in the form of sphoṭa. See ŚD 2.58–71.

386Literally meaning “in that alone,” I here render tatraiva idiomatically with “for that very proposition.”

Now, because of the absence of a means of knowledge, etc., they err in that [view], as well, in a way similar with (the view of) it being in the form of paśyantī, being beginningless and endless, (and) being one who has speech as its true nature, or (the view of) it having sphoṭa as its nature.

2.78–79ab

Therefore, [Somānanda] says:

2.78.āptānāptabhāṣitatve viśeṣo nāsti śabdagaḥ nityatve śabdatattvasya vyaṅgyatvaṃ dhvanibhir na ca

2.79.vyomavac cen na tulyatvaṃ sadā vyomny anumeyatā

Whether uttered by a reliable person or an unreliable person, speech is not differentiated if that which has speech as its true nature is eternal. Nor can it be made perceptible by sounds. If you argue that it is like the ether, (we reply:) there is no similarity. The ether is always inferred.387

Even what is uttered by an unreliable person would similarly be a valid means of knowledge,388 because one cannot differentiate (parts of) eternality.389

387This criticism is meant to parallel the criticism of the Sphoṭanityatvavādins in ŚD 2.58–62. In ŚD 2.58–59, we are told that if sphoṭa is different from paśyantī, then dualism persists. On the other hand, if it is identical to the latter, then all sphoṭas are equally real, the consequence of which is the fact that any statement, however ridiculous, would be true. Thus, the words of a reliable person would be no better than those of an unreliable person. In ŚD 2.60–61ab, Somānanda argues that, even if the grammarians argue that paśyantī, which is real, is identical to sphoṭa, they cannot claim that the latter is made perceptible by unreal words and sounds. Finally, Somānanda suggests that speech cannot be inferred in ŚD 2.61cd–62.

The first two arguments—the ones concerning the utterances of reliable and unreliable people and the manifestation of a real entity by unreal ones—are in this passage applied to the arguments of the Śabdanityatvavādins, those who think that Brahman is speech, but is distinct from the sphoṭas. The third argument is also invoked, but Somānanda does so indirectly. Here, Somānanda considers the possible objection that, when Brahman has speech as its nature, it can be known in the same manner as the existence of the ether is known from the sounds that are manifested in it. This, Somānanda argues, is not a good analogy, because the ether is not directly perceived but is inferred. We have already seen that, on Somānanda’s view, the grammarians do not accept inference as a valid means of knowledge (ŚD 2.33, 2.62b, and 2.64cd–65ab), except in mundane matters. In addition, it would seem that Somānanda, in arguing that ether is not an entity that appears in one’s direct experience, also wishes to suggest that the ultimate goal cannot be the inference of Brahman but rather it must be constituted by a direct experience of it. This last point should be taken in the context of the arguments Somānanda makes concerning the nature of religious experience in ŚD 2.72–76.

388That is, the statement of an unreliable person would be valid in a manner similar to the way in which a statements of a reliable person is valid.

389Nityatvāviśeṣāt may be taken literally to mean “because of the nondifferentiation of eternality.” Thus, the present passage suggests that if speech (śabda) is eternal and exists in the form of Brahman, as the Śabdanityatvavādins suggest, then it is impossible to distinguish the words of reliable people from those of unreliable people.

2.81

Thus, having mentioned this undesirable implication,396 [Somānanda], considering paśyantī herself, says:397

2.81. atha nāmnaiva paśyantī sphuṭam eva jaḍā tataḥ jñānaśaktiḥ smṛtā bhaṅgyā strīliṅgavyapadeśataḥ

Now, you might argue that she is “seeing” only by name, that you consider her to be the power of cognition merely because she is designated by the feminine gender. (We reply:) As a result, she is clearly insentient.398

Now, you might argue that, since there is error in associating (the nature of paśyantī) with the meaning (of the word) “seeing,” you hold that paśyantī is merely a name, i.e., is arbitrarily named, and, by the mere appearance of the feminine gender,399 you consider her to exist in the form of the power of cognition, which is associated with Brahman.

396This passage refers to the proposition, put in the mouths of Somānanda’s grammarian opponents (ŚD 2.79cd–80), that his own view suffers the same faults as that of the grammarians.

397Here, Somānanda considers the possibility that the grammarians will argue that the term paśyantī should not be taken literally, as Somānanda has been so apt to do in the preceding arguments. Kaul comments: “There is a possible objection: Be done with raising your objections (to our view, saying), ‘you must explain, what is paśyantī?’ For, it is not the case that the meaning (of a word) is always etymological, since there are also words like (the meaningless name) Ḍittha. And therefore, the reference here to paśyantī is similarly just by the force of common usage, like kuśala and so on (for Śiva), and you shouldn’t say: ‘what use is this?’ Nevertheless, she is designated by the word paśyantī, because she has the power of cognition as her form, simply because she is referred to in the feminine gender. Refuting this as well, [Somānanda] says ‘Now, you might argue,’ etc.” (nanu kṛtaṃ bhavaddoṣāropaṇena kiṃ paśyantīti kathyatām iti. na hi sarvatra yaugiko ’rthaḥ saṃbhavati. santi hi ḍitthādayo ’pi śabdāḥ. tataś ca rūḍhimahimnaiva kuśalādivad atrāpi paśyantīnirdeśaḥ. na ca vaktavyaṃ kim anena tathāpi paśyantīśabdena vyapadeśyaṃ, strīliṅganirdeśād eva jñānaśaktirūpatvād asyā ity api nirasayann āhāthetyādi). Note that, to reiterate, reference here to “Ḍittha” is meant to invoke a term—a proper name, in this instance—that has no etymological meaning, but nevertheless possesses denotative power.

398The word order in this verse is unusual. I have translated in the following order: 2.81a, c and d, and, finally, b. There is no reason that Somānanda could not have presented 2.81b for 2.81d, and vice versa, but this is not what is given in the KSTS edition. It is possible that the text is corrupt. All of the manuscripts that I have consulted, however, agree with the reading of the KSTS.

Somānanda here anticipates that the grammarians might argue that paśyantī is merely a word, one that does not describe the entity in question, but rather simply refers to the power of cognition. Somānanda responds by pointing out that, if this were the case, then paśyantī would have to be insentient. The argument is a rhetorical one (about which, see the Introduction, section 14). If the term in question does not describe the entity in question, then the entity in question cannot be considered to have the qualities that the term denotes, in this case sentience. Paśyantī, the present participle of a transitive verb, refers to something that is “seeing,” and by implication, knows or cognizes an object. If, on the other hand, the entity in question possesses the qualities denoted by the term, then the term accurately describes the entity in question.

399In other words, it is not the meaning of the term in question that suggests she is Brahman’s power of cognition, but merely the gender of the term in question that suggests as much.

As a result, this fails because, being unassociated with seeing, she is consequently insentient, and, as you know,400 something insentient cannot be the power of cognition.401

2.82–83

2.82.śabdasya viṣayākhyasya miśratvenendriyasya tu sarvadarśanavijñānaśūnyatā padavedinām

2.83.yasmād anādinidhanaṃ śabdatattvaṃ parā hi vāk paśyantyā varṇyamānatve haste grāhyaikatāpatet

By conflating speech, which is cognized as the object, and the organ, the knowers of words402 are ignorant of every (other) philosophical view,403 since, in describing paśyantī—“indeed, (Brahman,) beginningless and endless, it has speech as its true nature, is (paśyantī,) supreme speech”—there arises404 the consequence that the hand would be one with its object.405

By conflating, i.e., by identifying, speech, the cognition of which406 is in the form of the object, that is, the object of the action (of seeing), and the organ that knows it,407 which is well-known as the instrument of the action in the form of

400“As you know” renders the flavor of iti.

401In other words, the entity called paśyantī, because it is not associated with the qualities suggested by the meaning of the term, is not “seeing,” but rather is insentient. For this reason, paśyantī cannot be the power of cognition.

402This is a reference to the grammarians. It suggests that the grammarians should focus on explaining language, rather than philosophical matters. See ŚD 2.72–73ab.

403Utpaladeva glosses vijñānaśūnyatā with ajñānatā, which Kaul glosses with anabhijñatvam, and I follow this interpretation in my (slightly free) translation. Sarvadarśanavijñānaśūnyatā padavedinām could more literally be translated, “the knowers of words are devoid of an understanding of every (other) philosophical view.” In identifying the organ and the object of cognition, the grammarians contradict every other philosophical understanding of the matter.

404I interpret 2.83d to read haste grāhyaikatāpatet, not haste grāhyaikatā patet, meaning that I understand the verb to be āpatet, not patet.

405Somānanda here suggests that the grammarians identify Brahman-as-speech, the object of cognition, with paśyantī, which is that which cognizes speech. The identification of the two is justified on the basis of Somānanda’s interpretation of the grammarians’ view of Brahman: it is both “one who has speech as its true nature” (śabdatattva), as it is described in the first verse of the Vākyapadīya, and it is identical with paśyantī, who is “supreme speech” or parā vāk. The former is that which is cognized (grāhya), the latter that which cognizes (grāhaka), and thus, the two are equated. Put differently, Somānanda here suggests that the grammarians identify paśyantī with śabda, the instrument of action, as opposed to vāc, the organ of action. In doing so, they identify the instrument and the object of action, which is unheard of in Indian philosophy.

406Here, the relative pronoun, yasya is an objective genitive, not a subjective genitive. Note also that the commentary here glosses ākhya with pratīti. As the terms are virtually synonymous, and given that I have uniformly translated both with “cognition” throughout, I have not rendered this gloss in the translation, though the reader should be aware of its existence.

407Literally, tasyendriyasya means “the organ (of knowing) of it.” In other words, I take sa to be a possessive genitive referring to śabda, and my idiomatic translation is meant to reflect as much.