somananda e-book

meant somehow to echo the thirteen Kālīs in the fourth, “nameless” phase.165 Regardless, it seems clear, in sum, that Somānanda’s is a work that, while heavily influenced by the Trika, also imports the aforementioned Krama tendencies into its system.

12. Somānanda and the Śaiva Siddhānta

It is well known that the scriptures of the dualist Śaiva Siddhāntins are clearly distinguished from the scriptures of the increasingly goddess-centered and non-dualistic Śākta Śaiva tantras. Yet both the post-scriptural tradition and the scriptures themselves acknowledge a hierarchy of scripture, one that conceives of the relationship between the Saiddhāntika and the Śākta Śaiva scriptural sources in a relative manner. Later scriptural works and traditions regularly consider their own teachings to be superior to what are regarded as the less powerful traditions that preceded them; and, at the same time, these traditions acknowledge a relative power and legitimacy for the scriptures and schools deemed to be of a lesser status. Indeed, scriptures of a lesser status are often quoted favorably and are treated as authoritative sources by subsequent traditions and texts. Somānanda, for his part, took this very approach to the dualist tantras of the Śaiva Siddhānta.

This may be said to be so because Somānanda favorably quotes Saiddhāntika sources in a number of places, implying that they are authoritative works whose doctrines he must account for in dealing with his opponents’ objections. At ŚD 3.10–12ab, Somānanda anticipates an objection from his Śākta opponent, who suggests that Śaiva scripture confirms his position that the divine feminine in the form of speech is supreme, for so much, the opponent argues, is said in the Saiddhāntika Kālottaratantra, a dualist scripture preserved in a number of versions of varying lengths.166 Instead of dismissing the scripture as a lower revelation, however, Somānanda goes to some length (ŚD 3.12cd–15ab) to explain that his opponent has misunderstood the Śaiva position. He follows this by anticipating a second objection based in the KT (ŚD 3.15cd–16ab), one that suggests that the goddess, as speech, may be understood to be supreme in Śaivism by an inference based on a mention in that Saiddhāntika scripture of the fact that speech is the means to acquiring the power of mantras. Somānanda again goes on to refute the claim by contextualizing the statements in the KT, rather than dismissing the scripture outright. In a word, Somānanda replies to these objections by engaging in an exegesis

165Of course, nothing but the presence of some flavor of the Krama in the ŚD and the numeric coincidence of thirteen stages in both the nameless and the ŚāVi invite speculation regarding this short work, whose attribution to Somānanda is highly doubtful. This is to say that this idea involves speculation on the basis of only the flimsiest of evidence.

166See Goodall 2004: xxv; Goodall 2007: 125–129.

of Saiddhāntika scripture, this to show their consistency with his position that Śiva alone, the possessor of the powers, is supreme. He even goes on to cite the Kālottaratantra favorably (at ŚD 3.63cd), suggesting, according to Utpaladeva, that Saiddhāntika scripture points to the existence of Śiva-nature in all beings.

In another place,167 Somānanda refers to a well-known Saiddhāntika dogma, namely, that the Lord uses the power of māyā to create the universe and uses the occasion of the equal strength of good and bad karma in the bound individual to grace him with liberating insight.168 Despite this, he suggests, the Lord may not be said not to be powerful, or sovereign. The same is true, he suggests, of Śiva in the Pratyabhijñā, for he subordinates certain powers to superior ones, as discussed above.

Clearly, then, while Somānanda would have real misgivings with their philosophical dualism, he understands all the scriptures of the Śaiva Siddhānta that he bothers to mention to be authoritative works against which his own philosophical theology may be measured. Here, then, is one instance where it may be said that Somānanda wished to accommodate his presentation of the Pratyabhijñā to an audience sympathetic to the tantric school most closely aligned with orthodox Brahminism and therefore with mainstream Hindu thought and practice.

13. The S´ivadr.s.t.i and the Philosophy of the Grammarians

We now turn to a consideration of Somānanda’s extended and vociferous arguments against Bhartṛhari’s non-dualism. The arguments put forth in the ŚD focus primarily on the first of three chapters (kāṇḍas) of the VP, and the explanations thereof furnished by Utpaladeva in the ŚDVṛ quote extensively from the VPVṛ, which, one must note, is attributed in the Pratyabhijñā literature to Bhartṛhari himself.169 In particular, Somānanda gives a great deal of attention to VP 1.159, where Bhartṛhari refers to the three levels of speech, paśyantī, madhyamā, and vaikharī, though he also offers detailed criticism of both VP 1.167ab, which describes paśyantī,170 and Bhartṛhari’s description of Brahman found in a verse Somānanda identifies as belonging to the Śabdadhātusamīkṣā

167See ŚD 4.4–5.

168For more on this doctrine, and Bhaṭṭa Rāmakaṇṭha’s detailed exposition thereof, see KT 1.20cd–22ab and Bhaṭṭa Rāmakaṇṭha’s Vṛtti on the same. Goodall explains the exegetical skill with which Rāmakaṇṭha interprets the present verse, in order to make it accord with Sadyojyotis’s account of what precedes a descent of power. See Goodall 1998: 215–216, fn. 171.

169Considerable differences of opinion remain over the authorship of the principal commentary on Bhartṛhari’s VP, the VPVṛ, though I agree with Iyer, as well as George Cardona (personal communication, 2002), that the commentary is likely to have been Bhartṛhari’s and not the work of another author. For

a review of the arguments, see Iyer [1969] 1992: 16–36. 170See ŚD 2.44cd–51.

(ŚDhāSam), one that happens to be identical to the first stanza of Bhartṛhari’s Nītiśataka (NŚ).171 Now, because Utpaladeva adopts much of the terminology of the VP and the VPVṛ, as well as many of the associated conceptual formulations regarding the relationship between the creator and the created universe, scholars today have been at a loss to explain why Somānanda’s arguments are so categorical and unforgiving of Bhartṛhari’s system. For the VP and VPVṛ present a non-dualism that has much in common with the philosophy of the Pratyabhijñā, particularly Utpaladeva’s formulation thereof.

Indeed, one may fairly ask why Somānanda would object so unyieldingly to Bhartṛhari’s views when he accepts, mutatis mutandis, many of the fundamental philosophical premises that, on his own reading, appear in the VP and VPVṛ. These include: the philosophical non-dualism of those works, which maintain that all of existence is nothing but the divine (i.e., Brahman),172 who, consisting of the power of cognition,173 is made up of a consciousness that is all-pervasive, real,174 and is identified, according to Somānanda, with the very individual who experiences quotidian life in the manifested universe.175 Somānanda and Utpaladeva further understand Bhartṛhari to envision the creation of the universe as a product of this divine consciousness, a creation that is accomplished through the application of a series of divine powers.176 The VP even refers to the individual agent as the enjoyer (bhoktṛ) of worldly delights, something Somānanda surely would appreciate.177 And we have already seen that Utpaladeva adopts Bhartṛhari’s notion that consciousness is a self-reflective power that reveals itself and its contents simultaneously, this idea being reflected in the formulation of the famous prakāśa-vimarśa pole on which the ĪPK and ĪPVṛ, and all subsequent Pratyabhijñā philosophy, for that matter, relies so fundamentally.178

In what is perhaps the leading theory to date, Torella has suggested (or at least he has implied) that Somānanda’s arguments stem from his inadequate familiarity with Bhartṛhari’s philosophy. The bases for this hypothesis are two. First, Somānanda is very possibly ignorant of the articulation in the VPVṛ of the existence of a supreme (parā) form of paśyantī, literally “seeing,” the highest level of speech in Bhartṛhari’s system, one that is identified with Brahman. Second, Somānanda seems to have misunderstood the ways in which the VP and VPVṛ

171See ŚD 2.72d–76.

172See ŚD 2.2.

173See ŚD 2.1, 2.2cd, and 2.3.

174See ŚD 2.3. See also ŚDVṛ ad ŚD 2.2 (which paraphrases VP 1.1), where Utpaladeva paraphrases VP 1.132 in suggesting that, according to Bhartṛhari, Brahman exists in the form of consciousness that

is supreme speech, called paśyantī.

175See ŚD 2.3ab.

176See, e.g., ŚD 2.9. Compare also VP 1.131ab (vivartate ’rthabhāvena prakriyā jagato yataḥ) to ŚD 1.21d: tataḥ sarvaṃ jagat sthitam.

177See Utpaladeva’s commentary on ŚD 2.3; cf. VP 1.4 and the related passage of the VPVṛ.

178Utpaladeva quotes the term prakāśa as it is used in the VPVṛ in ŚDVṛ ad ŚD 2.2. See also ŚDVṛ ad ŚD 2.8cd–11. Finally, see ŚDVṛ ad ŚD 2.56, where Utpaladeva may be seen to understand Bhartṛhari’s paśyantī to have prakāśa as its nature.

described both “nescience” (avidyā) and the manifestation of the divine in the phenomenal world (vivarta). For, Torella suggests, Somānanda was ignorant of the fact that Bhartṛhari conceived of nescience as a power of Brahman and not as an ontologically extraneous entity; and he did not conceive of the manifestation of the divine in the phenomenal world as involving any sort of de-realization of Brahman, but rather understood it to involve the extension of Brahman and none other into the form of the universe itself.179

It is of course possible that Somānanda was insufficiently versed in Bhartṛhari’s œuvre. Indeed, Somānanda does not show any awareness of the contents of the VP beyond the first kāṇḍa, as Torella already noted,180 just as he was right further to note that VP 3.7.39–41 (from the Sādhanasamuddeśa) articulates a view to which Somānanda himself easily could have subscribed—namely, that oneness is not the product of a separation from multiplicity, but rather is the nature of all reality.181 It is further possible, at least, that Somānanda did not know the philosophy of the grammarians as intimately as did Utpaladeva, for not only does Somānanda fail to display any familiarity with the contents of the second and third kāṇḍas of the VP, this contra Utpaladeva, who repeatedly invokes the later chapters in the ĪPK and ĪPVṛ,182 but there is also no proof that Somānanda had access to the VPVṛ: Somānanda nowhere quotes or paraphrases the VPVṛ, while Utpaladeva does so extensively, particularly, as already noted, in the ŚDVṛ.

On the other hand, just as real differences exist between the monism of the ŚD and that of the ĪPK and ĪPVṛ, so too is it possible that substantive differences exist between the monism of the former and that of Bhartṛhari, as Somānanda clearly believes. Now, it is certain that Somānanda was aware of the very terms and concepts of the VP and VPVṛ that Utpaladeva found so compelling, in particular the dual nature of light, the prakāśa-vimarśa pole, which is described in the first chapter of the VP.183 It therefore strikes one as likely that Somānanda’s critique of Bhartṛhari stems, in part, at least, from his interpretation of the

179See Torella 1994: xxvi, esp. fn. 37.

180This is so despite Kaul’s suggestion that Somānanda’s understanding of action depends on Bhartṛhari’s famous definition thereof, which may be found at VP 3.8.4 (see note 164 to the translation of the first chapter of the ŚD, below). As noted, already (see note 60, above), Patañjali’s definition of action sufficiently informs Somānanda’s understanding of the same as to make it unnecessary to assume he knew Bhartṛhari’s definition. Cf. ĪPVṛad ĪPK 1.2.9 and Torella 1994: 94, fn. 17.

181Ibid. Cf. VP 3.7.39–41: paramārthe tu naikatvaṃ pṛthaktvād bhinnalakṣaṇam / pṛthaktvaikatvarūpeṇa tattvam eva prakāśate. yat pṛthaktvam asaṃdigdhaṃ tad ekatvān na bhidyate / yad ekatvam asaṃdigdhaṃ tat pṛthaktvān na bhidyate. dyauḥ kṣamā vāyur ādityaḥ sāgarāḥ sarito diśaḥ / antaḥkaraṇatattvasya

bhāgā bahir avasthitāḥ.

182On Utpaladeva’s references to the later kāṇḍas of the VP, see Torella 1994: 94, fn. 17; 108, fn. 14; 121, fn. 29; 124, fn. 36; 150, fn. 12; 153, fn. 2; 154, fn. 5; and 164, fn. 8.

183See VP 1.132, which is echoed in ŚDVṛ ad ŚD 2.2 and 2.8cd0–11. Cf. ŚD 2.14cd–15ab, where Somānanda mentions (in ŚD 2.15a) the power of reflective awareness (parāmarśa) for which the VP argues.

merits of and problems with the grammarian’s philosophical system, rather than merely from his insufficient understanding of Bhartṛhari’s œuvre.184

On this alternative hypothesis, Somānanda’s various criticisms stem primarily from his desire to defend his particular formulation of the Pratyabhijñā, even if they reflect the sort of interpretation of the VP and VPVṛ favored by later Vedāntins, who were influenced by Śaṅkara’s non-dualism, as Torella argues.185 Indeed, while Somānanda focuses many of his various arguments against Bhartṛhari on the difficulties he perceives the grammarians to have in explaining the relationship between Brahman, conceived of as speech in the form of paśyantī, and the created universe—I will not here analyze these arguments in great detail, given that they have already been summarized by both Torella and Gnoli,186 and, anyway, they appear in full, along with the commentary, in the translation that follows—all of these arguments arise from Somānanda’s observation that the grammarians fail to conceive of the power of will in articulating Brahman’s creative powers.

S O M Ā N A N D A ’ S A R G U M E N T S A G A I N S T T H E

G R A M M A R I A N S’ P A Ś Y A N T Ī

It may be observed that the opening verse of the second chapter of the ŚD simply and clearly states that the grammarians’ paśyantī is an insufficiently elevated state to be considered supreme: it is the equivalent of the power of cognition (jñāna), which in Somānanda’s system is located at the level of the sadāśivatattva, the third of thirty-six tattvas. On Somānanda’s analysis, the problem here involves the incapacity of the grammarians’ system to account for the power of will (icchā) that the agent who sees the universe must exercise if any cognition is to take place. This he states simply in the closing verses of the second chapter of the ŚD, where he compares paśyantī’s act of cognition to the potter’s production of a pot: both require the agent in question to exercise his will prior to the action in question. Brahman, in the form of paśyantī, must choose to see the manifested universe, just as the potter must choose to make a pot before doing so.187 As Utpaladeva’s commentary makes explicit, this act of volition must accord with the formulation thereof found in the first chapter of the ŚD.188

184It must be noted, however, that these are not mutually exclusive explanations for the cause of Somānanda’s unyielding critique of Bhartṛhari and “the grammarians,” for it is possible that he both disagreed with and simultaneously was insufficiently versed in the philosophy of the VP and VPVṛ.

185See Torella 1994: xxvi, fn. 37.

186See Torella 1994: xix–xx; Gnoli 1959: 55–63.

187See ŚD 2.84–88.

188See ŚDVṛ ad ŚD 2.84–88: evaṃ cecchā darśanakriyāyāḥ pūrvā sthitā, tasyāś cecchāyā api cittattvasaṃbandhisūkṣmatarollāsam iṣṭajñeyakāryaunmukhyalakṣaṇaṃ vinā prāguktanyāyāt kathaṃ prasaraḥ, tasyā api cito nirvṛtyaunmukhyecchājñānakriyākramavyavasthāyā yat sāmarasyam ekībhāvaḥ samāveśaviṣayas tatra vyavasthāvān vyavasthāśrayo ’kramaḥ śivabhaṭṭārakaḥ sthita iti. “In this way, moreover, the desire is the first condition of the act of seeing, and how could that desire, for its part, come

This is not possible, however, because the grammarians conceive of Brahman as paśyantī; and by putting paśyantī at the top of the cosmic hierarchy they leave insufficient room above her, as it were, for the stages of will that must occur prior to cognition.

As per Somānanda’s system of overlapping pairs of powers, the manifestation of the power of cognition (jñāna) coincides not only with the premanifested form of the power of action (kriyā), the form of action in posse, as Utpaladeva has indicated,189 but it also must be preceded by the act of will, which on Somānanda’s view, necessarily must consist of two stages, that of will being fully manifested as icchā (at which point the power of cognition exists in its potential form) and, prior to this, a first movement of will, aunmukhya, which, in turn, is the manifested form of the power of delight (nirvṛti) that exists in its potential form within Śiva himself.

Left out of the grammarians’ system, then, is the possibility of locating either this first movement of will or the agent who exercises it. This is so, Utpaladeva explains, because even if a supreme, premanifested form of cognition, the form of cognition in posse, were imagined to exist in Bhartṛhari’s system, this at the level of the parā form of paśyantī identifiable in Somānanda’s system with the power of will fully manifested, there is no space, so to speak, for either the first movement of will (aunmukhya) or the agent who exercises this volition prior to this pre-manifested phase of cognition. In the nomenclature of the tattvas, the grammarians may account for a moment prior to cognition, which is found at the level of the second tattva, the śaktitattva, this being equivalent to the grammarians’ parā form of paśyantī, but they cannot account for the initial moment of divine will that initiates the very act of cognition, this occurring, Utpaladeva suggests, at an interstitial level that is associated with nirvṛti and aunmukhya, one that is found between the level of the śivatattva and that of the

śaktitattva.190

Thus, in identifying paśyantī with Brahman, Somānanda suggests, the grammarians incorporate a subject-object dichotomy between the seer and the object seen in the very nature of Brahman-as-speech, for such a dichotomy is inherent in the nature of paśyantī, the nature of “seeing,” this being the literal meaning of the term. Because the existence of such a dichotomy at the level of

forth, in the manner previously explained, in the absence of the extremely subtle joy, characterized by an eagerness for desired objects of cognition and action, that is connected to the nature of consciousness? As for that [consciousness], as you know, Śiva Bhaṭṭāraka, being without sequence, is established as the one possessed of, i.e., as the locus of, the equilibrium—the unity of penetration—of consciousness, abiding in the sequence or the absence thereof of delight, eagerness, will, cognition, and action.”

189See ŚDVṛ ad ŚD 2.1, where Utpaladeva quotes the following, heretofore untraced, maxim to suggest that the power of action (kriyā) exists (in potential form) at the level of the sadāśivatattva: jñānakriye sādākhyam. “Cognition and action exist at what is called the Sāda(-level).”

190See the chart provided in my notes to ŚDVṛ ad ŚD 2.1, note 10 in the translation of the second chapter of the ŚD.

Brahman itself signals a philosophical dualism that Bhartṛhari, of course, would not accept, Somānanda figures that the universe must either be explained away, somehow, in order to preserve its unity, or Brahman’s relationship to the universe must be finessed in some manner or another. This is so because any contact with the universe of diverse entities would threaten Brahman’s unity, as it would have to register such diversity in its very being, dividing itself in doing so. In short, Somānanda analyzes Bhartṛhari’s system in terms of the fundamental tenet of his own system, and in finding the VP unable to account for the supremacy of the divine agent, in the form of a willful consciousness, he develops the many arguments related to the Bhartṛhari’s understanding of avidyā and vivarta. Indeed, the very prospect of Brahman existing in the form of speech is off-putting to Somānanda, as speech itself involves the very sort of subject-object dichotomy that he finds impossible to locate in Brahman itself.191

B H A R T Ṛ H A R I’ S AVIDYĀ A N D U T P A L A D E VA ’ S

ABHEDĀKHYĀTI

Turning now to a particular contribution of the ŚDVṛ, one not found in the ŚD itself, it may be noted that Utpaladeva puts forward a unique argument to explain how one could be ignorant of one’s own identity as Śiva. And one must further note that the formulation of this argument stands in direct contrast to the one Somānanda anticipates Bhartṛhari making for nescience (avidyā). In ŚD 2.34–35, Somānanda considers the possible nature of nescience by analyzing the term by which it is named. Insofar as nescience impedes one’s awareness of Brahman’s non-dual nature, he suggests, it must somehow negate, block, or alter the appearance of Brahman’s true nature (see ŚDVṛ ad ŚD 2.21cd–22ab; cf. ŚDVṛ ad ŚD 2.8cd–11). Nescience involves either the cognition of that which is not real or the cognition of some entity as something other than it truly is. In order to perform its function, nescience therefore must somehow exist as something other than Brahman. A problem arises, however, with the very fact that nescience is thereby conceived of as other than Brahman, thus producing a dualism of Brahman, on the one hand, and nescience, on the other.

Somānanda anticipates that the grammarians will have considered this problem, and ponders the possibility that they understand nescience in a different

191This is to say that I disagree with Torella’s suggestion (1994: xxvi, fn. 38) that Somānanda “essentially pass[es] over” the notion that speech is inherent in all entities, having considered this fact in ŚD 2.19–20. I rather understand Somānanda simply to disagree with Bhartṛhari’s conception of the hierarchy of speech, regardless of the levels of subtlety applied to it, as it necessarily involves a level of duality that is best located at the parāparā level and best associated with Sadāśiva, not Śiva himself. Much of the balance of the chapter, then, is dedicated to a critique of Bhartṛhari’s very conception of a self-reflexive speech as supreme, as its very self-reflexivity involves an inherent dualism, however subtle, that can only be transcended at the level prior to the intention of the agent who speaks.

manner, as simply the absence of knowledge, as “non-knowledge” (a-vidyā).192 However, this also fails, both Somānanda and Utpaladeva argue, because such a view of nescience would require something that does not exist, “nonknowledge” (a-vidyā), to impede something that does exist, Brahman. This is impossible, because logically speaking something that does not exist, something that has no form (akiñcidrūpa), cannot do anything at all:193 it would be astonishing if something that has no real form were able to block something that does.194

How, then, is it possible for one not to know one’s own identity as Śiva, according to the Pratyabhijñā, and how can one subsequently “recognize” it? Utpaladeva’s explanation in the ŚDVṛ, which does not appear, incidentally, anywhere in the ĪPK and ĪPVṛ, is as follows. That which keeps one from seeing the multiple universe as what it really is is simply the noncognition (a-khyāti) of Śiva’s non-duality (a-bheda).195 It is the nonrecognition of the non-duality that one inevitably sees. Now, while the nomenclature here used echoes that of the Mīmāṃsaka Prabhākara, who spoke of the nonappearance of difference (bhedākhyāti), Utpaladeva does not seem to wish to respond to that position, as I have suggested elsewhere, but rather focuses on developing a theory of error that is peculiar to the Pratyabhijñā.196 He suggests that the type of error in question is similar to that of a man who, standing before the woman who loves him after an absence of many years, gives her no pleasure until the moment she recognizes the man in front of her as her very own: her cognition of him is the same both prior to and after recognizing him, but the recognition makes all the difference.197 In the same way, one always and only sees Śiva performing the activity

192That is, Somānanda (ŚD 2.34–35) analyzes the term avidyā etymologically, it being a compound of two terms, vidyā, roughly meaning “knowledge,” prefixed by a negative particle, the so–called “alpha-privative” (a-). Here, he suggests that the term should be understood as a pure negation (prasajyapratiṣedha), referring to the absence of knowledge. As outlined, above, he also considers the possibility that the term is a negative compound that merely indicates what something is not (paryudāsapratiṣedha), one in other words that suggests that avidyā is something, but what it is is “not-knowledge.” This he rejects, as already noted, on the grounds that it would require two entities to exist: a knowledge associated with Brahman-as-paśyantī and something other than this, namely avidyā. See also ŚDVṛ ad

ad ŚD 2.30cd–31, where Utpaladeva suggests that the grammarians argue that nescience is not a thing and, as such, has no nature of its own: avastu punar avidyā niḥsvabhāvā.

194As Utpaladeva put it, “something that has no form [akiñcidrūpa] does not have the power, which means that it is not possible (for it), to block (anything).” See ŚDVṛ ad ŚD 2.34–35.

195See ŚDVṛ ad ŚD 1.1, 1.7cd–8, and 1.11cd–13ab.

196See my forthcoming “Two Pratyabhijñā Theories of Error.” I also examine the development of this theory in the writings of Abhinavagupta in the same article. See also Rastogi 1986, esp. p. 4, where he notes that Abhinavagupta is uninterested in Prabhākara’s theory of error, despite the existence of any terminological affinities.

197This is the example given on ĪPK 4.17.

that is inherent to his nature as consciousness, and yet one does not always recognize what one sees.

More important, the noncognition of Śiva’s non-duality has no form (it is akiñcidrūpa), because it is in reality nothing at all: it is merely the absence of a cognition.198 There is in reality nothing there, as it were, to negate, block, or alter one’s awareness of Śiva. Thus, while (according to Somānanda) the grammarians understand some entity other than Brahman to be a negative entity, be it a negation or an absence of the real Brahman, Utpaladeva understands the very cognition of duality to itself be nothing but an absence of the cognition of non-duality, or in other words to itself be a nonentity.199 In fact, Utpaladeva even suggests that the noncognition of Śiva’s non-duality is the very power of māyā, itself.200 As a consequence, Utpaladeva suggests, the Pratyabhijñā does not have to explain the ontological status of that which impedes one’s awareness of non-duality. It is simply the absence of the awareness of unity. In either cognition, of unity or of multiplicity, the object of contemplation is the same, and nothing changes in the moments prior to and following the recognition of non-duality. Moreover, the very fact of the noncognition of non-duality is a function of the very nature of consciousness, which by nature sees diverse forms in itself. Thus, it is only the one unchanging but dynamic nature of Śiva’s consciousness that one sees. In this way, Utpaladeva offers a detailed and in many ways compelling explanation for the sort of cognitive error that causes one to perceive a multiplicity of entities in the universe, where only one entity exists.

O N W H AT D I F F E R E N T I AT E S T H E T W O S C H O O L S

Somānanda suggests that the grammarians could have avoided the various problems of dualism arising from their problematic formulation of the relationship between Brahman and the universe it creates, if they only had chosen to describe paśyantī as that which sees a subtle entity that is not autonomous from her own self (ŚD 2.57). But the language of the grammarians points toward a rather different conception of Brahman-as-speech: not only does the idea of “seeing” (paśyantī) literally refer to an entity that is fully distinguishable from the object of sight (something Somānanda takes very seriously);201 but the grammarians also understand paśyantī as the one in whom or by whom sequence is concluded.

198See, e.g., ŚDVṛ ad ŚD 1.11cd–13.

199Insofar as it is an error, it is impure, but it is nothing at all because it is merely an absence. See

ŚDVṛ ad 1.11cd–13ab: abhedāparāmarśanam eva bhrāntirūpaṃ kutsitam, tac ca na kiñcid akhyātirūpamātratvāt.

200See ŚDVṛ ad ŚD 2.89–91: māyāśaktyā śivābhedākhyātyā.

201See ŚD 2.81 (also discussed below), where Somānanda suggests that to abandon the meaning of the term paśyantī is to abandon the notion that she embodies the power of cognition. See also ŚD 2.45cd, where Somānanda suggests that paśyantī cannot be described as “nondistinct” if she truly is one who “sees:” to see requires that the agent who sees registers the differences apparent in the diversity of the object(s) of sight.