Navigation – Plan du site
Résumés des conférences en Sciences religieuses
Religions of India: Tantric Studies

On the contributions of the Śivadṛṣṭi of Somānanda to the Intellectual History of the Pratyabhijñā

Conférence de John Nemec
p. 15-23


In a series of four lectures and in my research program of the last twelve months, I examined the history of Pratyabhijñā thought, primarily through an examination of the philosophical contributions of Somānanda, the founding author of the school. A study of Pratyabhijñā engagement with the realist Nyāya-Vaiśeṣika was followed by an examination of Somānanda’s particular contributions to the Pratyabhijñā arguments against the Buddhist Epistemologists and Dharmakīrti in particular. Finally, I examined the various arguments against Advaita Vedānta ontology in Somānanda’s Śivadṛṣṭi and suggested they exhibited a lesser concern with that school than was evinced in the writings of Abhinavagupta, Somānanda’s great-grand disciple.

Haut de page

Texte intégral


1Śaiva philosophy has increasingly come to be counted as an indispensable dimension of Indian thought, and that this is so may be credited to the evolution and maturation of the scholarly study of the relevant traditions. First examined with an eye toward their own internal logics—as inward-looking and self-contained sectarian theological schools—scholars have more recently mapped the ways in which Śaiva philosophers engage the writings of their Buddhist, Jain and other, orthodox “Hindu” interlocutors. So much is the case with the study of the famed Pratyabhijñā or “Recognition” School in particular, a Śaiva tantric philosophical tradition born in the Kashmir Valley around the turn of the tenth century. Following the printing of the major Pratyabhijñā works in the Kashmir Series of Texts and Studies in the first decades of the twentieth century, a number of translation projects treated them essentially as works of theological speculation, but scholars more recently have mapped the engagement of the authors of the Pratyabhijñā with their non-tantric, and often non-Śaiva, interlocutors, most notably the Buddhist Epistemologists.

2Guided by the presupposition that Śaiva philosophical traditions require the same degree of critical analysis that has led to the discovery of the particular contributions of individual authors in the history of, for example, Advaita Vedānta or Buddhist idealist traditions, I engaged this series of lectures precisely to contribute to such an intellectual history of the Pratyabhijñā. In doing so, I laid emphasis on the writings of the founding author of the school, Somānanda (fl. c. 900–950), who launched the Pratyabhijñā with the production of his Śivadṛṣṭi (ŚD). Attention was also paid to the writings of his disciple, Utpaladeva (fl. c. 925–975), who composed what is now counted as the preeminent work of Pratyabhijñā philosophy, the Īśvarapratyabhijñākārikās (ĪPK), along with two auto-commentaries, a Vtti and a Vivti, and whose commentary on the ŚD is the only one to survive to the present day (though it is incomplete). Finally, I took note of selections from the writings of the polymath Abhinavagupta (fl. c. 975–1025), the grand-disciple of Utpaladeva and author of, among a host of other items, two sub-commentaries on the ĪPK, the Īśvarapratyabhijñāvimarśinī (ĪPV) and Īśvarapratyabhijñāvivtivimarśinī (ĪPVV).

3Such an emphasis is justified by the fact that the study of the history of Pratyabhijñā thought has been impeded, to date, by the lack of any comprehensive accounting of Somānanda’s ŚD, a difficult work the comprehension of which allows one to see the progressive development of ideas across the three generations of these authors (and beyond). The fourth, fifth, and sixth chapters of the ŚD, in particular, have to date escaped almost all scholarly scrutiny, which is lamentable given Somānanda’s extensive engagement there with the Buddhist Epistemologists (Dharmakīrti in particular), the Nyāya-Vaiśeṣika and, to a lesser extent, the Vedānta, among other philosophical schools. Examining this evidence, I sought to clarify Somānanda’s positions vis-à-vis these opposing philosophical schools and did so in the course of four lectures, as follows.

Engagement with the Realist Nyāya-Vaiśeṣika

4First, I prosecuted a diachronic reading of Pratyabhijñā philosophy to unearth a rather more significant debt to the Nyāya-Vaiśeṣika in that tradition than has heretofore been recognized. In doing so, I offered an analysis of the ways in which Somānanda apparently modifies—or at least has in mind—Nyāya-Vaiśeṣika ontology in the formulation of his theory of Śiva-as-consciousness (most notably as found in Praśastapāda’s Padārthadharmasagraha).

5Somānanda’s concern with the Vaiśeṣika is explicit, and is expressed in the opening lines of the sixth chapter of the ŚD (for which no commentary remains extant):

atha śakteḥ śaktimato na bhedo dravyakarmavat |
sthāpito dravyato bhinnā kriyā no na ca nāsti sā | | 6.1. | |
evaṃ tathā śaktimataḥ śaktasya samavasthitā |
jagadvicitratā śaive na punar darśanāntare | | 6.2. | |

Now, no [ontological] distinction is established [in our view] between the power and the possessor of the power just as [we make no such distinction] between substance (dravya) and action (karman). Action (kriyā) is simply not different from substance (dravyatas), nor is it the case that [action] does not exist. And, in this way, the variegation of the world is fully established for the empowered possessor of power in Śaivism [alone], but not in any other philosophical system.

6This is to say that Somānanda explicitly criticizes the ontological distinctions found in the Nyāya-Vaiśeṣika, which distinguish between substance (dravya) and action (karman), each being classed as categories of real being or padārthas. Yet, he, and after him Utpaladeva and Abhinavagupta, accepted a notion of the ātman or self—identical as it is with Śiva—that was similar to its Nyāya-Vaiśeṣika counterpart in being conceived of as eternal (nitya), all pervasive (vibhu, sarvagata), formless (amūrta), and—what is key—as a volitional agent of action (this in contrast to the Sāṅkhya formulation, which posits the existence of a self as the passive, though fully conscious, purua). Indeed, to fuse the natures of substance (dravya) and action (karman) as they were understood by the Nyāya-Vaiśeika precisely accounts for Somānanda’s conception of Śiva as a volitional agent who is at the same time active and ubiquitous.

7Following this, I illustrated the ways in which Utpaladeva modified this view. Somānanda understood Śiva to have no form, to be possessed of an amūrtatva, and for this formlessness to be a universal quality of all reality; but Utpaladeva, following Bhartṛhari, suggested that spatial form (mūrti) and time (kāla) could be manifested in the phenomenal universe, this at ĪPK 2.1.5:

mūrtivaicitryato deśakramam ābhāsayaty asau |
kriyāvaicitryanirbhāsāt kālakramam apīśvaraḥ | | 2.1.5 | |

Through the variety of physical forms he causes spatial succession to appear; through the manifestation of the variety of actions the Lord also causes temporal succession to appear. (Transl. Torella.)

8So much exemplifies Utpaladeva’s willingness to adopt positions maintained in the commentaries on Bhartṛhari’s Vākyapadīya, where Harivṛṣabha and Helārāja allow for such a real existence of form in a manner that is not found in evidence in the ŚD, and explain it by appealing to a deśakrama, just as they explain action by appealing to a kālakrama. Thus, Harivṛṣabha’s Vākyapadīyavtti (see VPVṛ ad VP 1.1, vol I, p. 9, ll. 1–3) describes two types of vivarta, one related to mūrti, the other to kriyā:

mūrtikriyāvivartāv avidyāśaktipravṛttimātram. tau vidyātmani tattvānyatvābhyām anākhyeyau. etad dhy avidyāyā avidyātvam.

Spatial (mūrti) and temporal (kriyā) manifestations are nothing but the functioning of the powers of nescience (avidyā). The two cannot be expressed as either identical with nor different from the one whose nature is vidyā. For, the fact of being nescience consists in just that. (Transl. an adaptation from Iyer’s.)

9And in the commentary of Helārāja on verse 14 of the dravyasamuddeśa (what is VP 3.2.14), kriyā is associated with the power of time (kālaśakti), while mūrti is associated with the power of space (dikśakti):

kālaśaktyavacchinno hi kriyāvivartaḥ dikśaktyavacchinnaś ca mūrtivivarta
iti mūrtikriyāvivartarūpaṃ viśvaṃ pratipāditam.

For, the transformation of action (kriyāvivarta) is circumscribed by the power of time (kālaśakti), while the transformation of form (mūrtivivarta) is circumscribed by the power of space (dikśakti). Therefore, the universe (viśva) is taught to have the transformation of form and of action as its nature.

10Of course, this offers a formulation that is very close to the one propounded by Utpaladeva and stands, I proposed, as the model for what is stated in the ĪPK.

11Finally, I suggested that Abhinava, in turn, echoed elements of both his predecessors’ positions on the nature of (physical) form and action in the manifested universe. The lecture illustrated, then, the significance of the “realist” schools on Somānanda’s thought, and the degree to which the authors who followed him modified his position on the nature of things in the world.

New Evidence of Śaiva Arguments Against the Buddhist Pramāṇa Theorists

12The second lecture, along with a part of the fourth, was devoted to a study of the influence of the Buddhist Epistemologists and Dharmakīrti in particular on Somānanda’s ŚD. Now, it is well known that Utpaladeva’s ĪPK exhibits extensive engagement with this tradition of Buddhist thought, and that Utpaladeva adopted much, conceptually and terminologically, from it. Seeking to illustrate the degree to which Somānanda, too, engaged this tradition, I argued that he knew it intimately and had more to say about it than had been known previously, and, moreover, that there is a great degree of philosophical continuity in the tradition as regards Pratyabhijñā arguments against the Buddhist Epistemologists.

13Somānanda’s arguments presage similar approaches to the Buddhists that are found in the ĪPK. Thus, for example, Somānanda attacks the Buddhist argument for the momentariness of cognitions, this by suggesting that it is impossible for their system to countenance any invalidation of any cognition in the absence of the existence of a single agent—Śiva—who is the basis for and is possessed of such moments of awareness. This is so, Somānanda argues (at ŚD 4.23), because the very momentariness of cognitions renders it impossible for the Buddhist to account for any relationship between one cognition and another, the one being meant to invalidate the other (this, it is implied, unless that Buddhist opponent will accept the existence of a unitary agent of consciousness, of Śiva-as-consciousness, who can unite these moments of cognition):

jñānāntareṇa jñānaṃ tadvirodhād atha bādhyate |
na bādho bhinnakālatvāt prāktanasyāpy abhāvataḥ | | 4.23 | |

Now, if you argue that the cognition [of, e.g., silver] is invalidated by another cognition [i.e., by that of mother of pearl], since it is incompatible with it, [we reply:] there is no invalidation since the preceding [cognition] itself no longer exists [by the time the subsequent cognition arises], because it occurs in a different [i.e., earlier] moment of time [and is, by nature, instantaneous].

14In doing so, Somānanda explicitly rejected the possibility that either of the two forms of contradiction conceived of by Dharmakīrti, in his Nyāyabindu (NB), could obtain in such circumstances. Dharmakīrti’s two-fold schema of contradiction (virodha) is offered as follows:

NB 3.74 : dvividho hi padārthānāṃ virodhaḥ.
NB 3.75 : avikalakāraṇasya bhavato ’nyabhāve ’bhāvād virodhagatiḥ.
NB 3.76 : śītoṣṇasparśavat.
NB 3.77 : parasparaparihārasthitalakṣaṇatayā vā bhāvābhāvavat.

[Mutual] contradiction between objects is of two kinds. A contradiction arises when an entity whose cause is unimpaired ceases to exist when another entity comes into being… just as [is the case] with [the sensations of] heat and cold. [A contradiction] also [arises] as a result of the nature [of two entities] being mutually exclusive, as [is the case] with the affirmation and negation [of the same entity].

15Somānanda dismisses the first of the two forms of contradiction, that of sahānavasthiti, at ŚD 4.24ab:

sahānavasthitir nāsti virodhaḥ prāgvināśataḥ |

There is no contradiction caused by the co-presence of mutually distinct entities, due to the prior destruction [of the earlier cognition, before the subsequent cognition arises].

16In explaining the passage, moreover, Utpaladeva’s ŚDVṛ echoes Nyāyabindu 3.75–76:

sahānavasthitir api virodho nāsti. na hi jñānaṃ kṣaṇikatvena prāgvinaṣṭam anyena nivartanīyam uṣṇena śītam iva viruddhaṃ bhavati rajataśuktijñānayoḥ.

There is no contradiction caused by the co-presence of mutually distinct entities, either. For, when it comes to the cognitions of silver and mother of pearl, the cognition that on account of its being momentary is destroyed in an earlier moment is not suppressed by another [cognition], as cold is invalidated by heat.

17Next, Somānanda rejects the possibility that the second form of contradiction, that of parasparaparihāra, could obtain, this at ŚD 4.24cd–25:

anyonyaparihāro vā jñānājñānātmakaḥ sthitaḥ | | 4.24 | |
ajñānatve parijñāte tadā syāt svavirodhitā |
ajñānatve svabhāvena virodhaḥ kena vāryate | | 4.25 | |

Alternatively, [you might argue that] there exists a contradiction by the mutual exclusion of awareness and the absence of awareness. [Reply:] When an absence of awareness is perceived, it is necessarily self-contradictory at that time. When the nature [of the cognition] is that of non-awareness, there is a contradiction [of this nature] with its own nature. By what could this [contradiction] be averted?

  • 1 Some of these materials were presented in J. Nemec, « Two Pratyabhijñā Theories of Error », Journal (...)

18Here, then, is a moment of consistency among Pratyabhijñā authors as regards their criticism of the momentariness doctrine in Buddhist Epistemology; yet, Somānanda’s contributions to the matter were not well known until recently.1

19I also argued that Somānanda engages the Buddhist Epistemologists by way of appealing (approvingly, for the most part) to the philosophy of the two Mīmāṃsaka authors, Kumārila and Śabara. He does so by arguing, in sympathy with these Mīmāṃsakas, who made similar claims though much before him, that a sound philosophical system must be able to account for the existence of a connection between speech and objects of speech, between śabda and artha, and that if one were to posit the existence of a moment when such a relationship did not exist, it would be impossible to account for any possible creation of such a relationship. Of course, Somānanda parted ways with his Mīmāṃsaka counterparts in accounting for this relationship by virtue of his appeal to a non-dualist ontology—the Mīmāṃsakas’ of course were dualists—but he apparently accepted the Mīmāṃsakas position regarding the impossibility of any fabricated connection of speech and object-of-speech, agreeing with them that this presented insurmountable challenges to their Buddhist opponents.

20The matter is first raised at ŚD 4.71–73ab, as follows:

ekatattvaṃ vinaitac ca vyavahāro na jāyate |
śabdārthayor na saṃbandho bhinnayor bhinnadeśayoḥ | | 4.71 | |
viruddharūpayor bhinnakaraṇagrāhyayor api |
mukhe hi śabdo bhūmau ca vidyate ’rthaḥ kva saṃgamaḥ | | 4.72 | |
amūrta eko mūrtaś ca dvitīyo yogitā katham |

  • 2 Presumably Somānanda here refers with etat to tattvasyaikyam, which is found at ŚD 4.70a.

Moreover, [human] discourse (vyavahāra) could not occur in the absence of this2 single reality: There could be no relation between speech and object, the two being distinct and situated in different places. Indeed (hi), where can there be a meeting (sagama) of two [phenomena] of [mutually] contradictory forms (viruddharūpayo) that also (api) are apprehended by different instruments (bhinnakaraagrāhyayo). For, speech (śabda) is found (vidyate) in the mouth (mukhe) and the object (artha) [denoted by it] on the ground (bhūmau); the one is without form (amūrta eka), and the other has a form (mūrtaś ca dvitīya); how (katham) can they be [mutually] associated [one with the other, this in the absence of the existence of a single reality]?

21In referring here to the mutual differences between speech and object of speech, Somānanda makes clear reference to Śabara’s commentary on Mīmāsāsūtra 1.1.5, specifically to the notion that the two are perceived as mutually separate entities. The passage in question, which presents the views of a pūrvapakin who wishes to challenge the Mīmāṃsā position, reads in part as follows:

yadi prathamaśruto na pratyāyayati, ktakas tarhi śabdasya arthena sabandha | kuta? svabhāvato hy asabandhāv etau śabdārthau | mukhe hi śabdam upalabhāmahe, bhūmāv artham | śabdo ya na tv artha, artho ya na śabda iti ca vyapadiśanti |

[Objection:] If [a word] does not denote meaning on first being heard, then it follows that the relation of the word with the object [to which it refers] is created/artificial (ktaka) [and not eternal]. Why? Because (hi) these two, word and object, are by nature not related. For, we perceive the word in the mouth, the object on the ground. Moreover, [people] make the distinction: « this is the word and not the object ; this is the object, not the word. »

22The wider argument concerning the denotative power of speech—the very capacity of a word to refer to a thing—is articulated only a few verses later, at ŚD 4.78–81c:

kathaṃ na devadattasya yajñadattavad ākhyayā |
atra saṃketitatvāc cet saṃketenātra kiṃ kṛtam | | 4.78 | |
saṃyogaś cen na dūreṣu mūrtāmūrteṣu yujyate |
śabdasyoccāritadhvaṃsān naṣṭānaṣṭadvaye na ca | | 4.79 | |
vācyavācakarūpaś cet sa eṣa niyamaḥ kutaḥ |
yat tasya vācakatvaṃ hi vācyatvam aparasya tu | | 4.80 | |
vācyavācakarūpatve vācyavācakatānvayaḥ |
itaś cāsti jagaty aikyaṃ

How is one like Yajñadatta not [designated] by Devadatta’s name? [Opponent’s] Objection: because of the fact that there is agreement (saketitatva) here [regarding the referent of the name]. [Reply:] What service is done here by the agreement? [Opponent’s] Objection: a connection (sayogas) [of the name with the person named]. [Reply:] That [sayoga] is not possible (yujyate) for those [mutually] distant (dūreu) [entities], which are [respectively] material and immaterial (mūrtāmūrteu); nor (na ca), given that speech expires after it is uttered, is it [possible] for the two that have [respectively] perished [i.e., the name ‘Devadatta’, after it is uttered] and not-perished [i.e., Devadatta himself]. [Opponent’s] Objection: it [i.e., the sayoga or connection] is in the form of designatum and designator (vācyavācakarūpa). [Reply:] Wherefrom this necessity [of the connection] (niyama) ; for, it [i.e., speech; the name] is that which designates (vācakatvam), while (tu) that which is designated (vācyatvam) is something different (aparasya). If [you argue:] it is [found in] the [very] nature of the designatum and designator (vācyavācakarūpatve), [we reply:] there is [, in that case,] a definite link of the nature of the designatum to that of the designator. And, hence, a unity exists in the world.

23Now, the argument Somānanda offers is strikingly similar to those maintained by Kumārila in this sense: like Kumārila, he suggests that śabda and artha must be perpetually connected—the sabandha or connection between them must be eternal (nitya)—; for if there were a time, prior to their mutual association, when they were mutually unrelated, there could be no way to begin to connect them.

24At the core of this is a more fundamental claim, however, namely, the notion that concept and percept, speech and object known or denoted by speech, must be mutually implicated. Kumārila articulates this position in a manner that is careful to distinguish it from Bhartṛhari’s pure identification of the two: he speaks of a “double nature” of perception. There is, that is, no perception that is entirely divorced from conceptualization, Kumārila maintains, (this in his Ślokavārttika, at pratyakapariccheda 112–113), not in the monistic manner presented in Bhartṛhari’s Vākyapadīya, but in the more limited and realist sense that speech is implicitly present even in non-conceptual perceptions, just as it is implicit in the awareness of children, those who are unintelligent, or the like—not as an ontological unity, but as an implicit but unexpressed dimension of the perception in question. Kumārila says:

asti hy ālocanājñānaṃ prathamaṃ nirvikalpakam |
bālamūkādivijñānasadṛśaṃ śuddhavastujam | | 112 | |
na viśeṣo na sāmānyaṃ tadānīm anubhūyate |
tayor ādhāṛabhūtā tu vyaktir evāvasīyate | | 113 | |

First there is an initial perception (ālocanā) —a non-conceptual awareness, similar to the awareness of children, mutes, and the like, and arising purely from the object. At that time, neither particular nor universal is experienced, but, rather, the individual (vyakti), which is the basis (ādhāra) of both, is apprehended. (Transl. McCrea’s).

25In such a simple initial perception, merely the individual object of sight is seen, but it embodies implicitly in itself both a universal, abstract entity that is closely associated with language and conceptualization—the fact of being a pot, for example—and the unique particularity of the object in question—that particular pot with all its unique qualities.

26Somānanda makes clear and explicit reference to this idea at ŚD 4.81d–83ab, as he there quotes Kumārila’s verse 112c–d, as follows:

  • 3 The reading here recorded, pratyakād grahaād is that of the manuscript of the Calcutta Sanskrit C (...)
  • 4 The present reading, that of vā vyavahāragā is a conjectural emendation. Both the editio princeps a (...)

pratyakṣād grahaṇād3 api | | 4.81 | |
svalakṣaṇena yogitvād vyavahārasya sarvataḥ |
loke cānupapatteś ca pramā vā vyavahāragā4 | | 4.82 | |
bālamūkādivijñānasadṛśī kidṛśī kriyā |

27This passage of the ŚD is a difficult one, and any proper interpretation of it depends on accepting a variant reading recorded in one manuscript, that of pratyakād grahaād where the editio princeps reads pratyakāgrahaād. It also requires one to emend, as I tentatively have done, in the last quarter of verse 4.82. Taken together, one can tentatively suggest that this passage may be rendered with the following:

Indeed (), valid knowledge (pramā) is discursive (vyavahāragā), because discourse (vyavahārasya) is invariably connected with the unique particular (svalakaa) in the world, even as the result [merely] of the apprehension that is direct cognition, and because of the impossibility [of such discourse, if matters were otherwise]. [After all:] Of what kind is the action that is similar to the awareness of children, mute people, and the like?

28To put the matter plainly, cognition, even what could be conceived of as purely perceptual cognition, must be connected with speech. Somānanda here agrees, that is, with the Mīmāṃsakas, and he quotes Kumārila to support his point of view: the Buddhists cannot posit a momentary and instantaneous real existent, a svalakaa, that is entirely non-conceptual in nature, for it would be impossible to connect it to concepts if it weren’t already so connected, and because we see the two connected in the world at all times, anyhow.

29I know of no passages in the ĪPK or its Vtti, nor in the writings of Abhinavagupta, which similarly engage Kumārila and Śabara in treating the theories of the Buddhist Epistemologists. Perhaps the reason for this is simply that Kumārila offers his point of view in explicit contrast to Bhartṛhari’s, and unlike Somānanda, Utpaladeva and Abhinavagupta frequently accommodate Bhartṛhari’s views in their writings. Indeed, both claim that language and reality are interwoven not in the provisional manner allowed by Kumārila, but decidedly so, in the form of an ontological identity that accords with Bhartṛhari’s view of the matter. It does not appear, in this context at least, that the same could be said of Somānanda.

Pratyabhijñā Arguments Against the Vedāntins

30The third lecture explored the Pratyabhijñā’s somewhat uneven but fascinating history of engagement with the Vedānta. While virtually absent from the ĪPK and their Vtti, the Vedānta was known in myriad forms to Somānanda, who refers (in ŚD Ch. 6) to no fewer than twelve distinguishable Vedāntic perspectives on how it is that a multiple world can be manifested by a unitary reality. That is, Somānanda queried and rejected a dozen Vedāntic ways of justifying their non-dualism. In doing so, he considered both illusionist and transformationalist arguments, those theories that, respectively, claimed the multiple world to be some sort of illusion (the māyāvāda) or claimed the world to be a natural extension of the unitary divine (the pariāmavāda). Also of note is the fact that Somānanda joined his critique of the Vedānta with one of the Pañcarātra in a manner that mirrors the writings of his dualist, Śaiva Siddhānta counterparts, specifically the Paramokanirāsakārikā of Sadyojyotis and the Vtti thereon, which was composed by Rāmakaṇṭha, who was more or less a contemporary of Somānanda and himself a Kashmiri.

31I conjectured that this evidence, somewhat scarce as it is, suggested that Somānanda may well have had the writings of his Saiddhāntika counterparts in mind when composing his brief and doxographic account of the Vedānta (and the Pañcarātra); and, apart from this, that his dismissive and apparently incidental interest in the Vedānta—he devoted only 23 verses to their thought, and the passage is adventitious to the larger argument of the ŚD—stood in contrast to the body of work composed by Abhinavagupta, who not only makes more frequent reference to the Vedānta in his ĪPV and ĪPVV but also penned the Paramārthasāra, a text that incorporates, somewhat modifies, and adds to a Vedāntic work of the same name in what appears—to me, anyhow—to be an effort to appropriate any influence such a Vedāntic work might have had in Abhinavagupta’s day.

Philosophy in Context: The Evidence of the Pratyabhijñā

32The final lecture served to summarize the findings of the previous three, and began to address questions relating to the degree to which each of the respective authors of the Pratyabhijñā innovated novel philosophical arguments, and how well those arguments could be said to conform to a unified tradition—that could be said, that is, to maintain fidelity to a shared notion of their common “theological truths.” The answers I offered, tentative though they were, suggested that these authors did, indeed, write new philosophy and, more importantly, did so in a shared spirit of tradition. Somānanda’s contributions, moreover, insufficiently understood as they might be, were substantial, and often novel; and while sometimes they were left without a legacy in the writings of his successors, engaging them promises deeply to enrich our understanding of Kashmiri—and Indian—intellectual history.

Haut de page


1 Some of these materials were presented in J. Nemec, « Two Pratyabhijñā Theories of Error », Journal of Indian Philosophy 40/2 (2012), p. 225–257.

2 Presumably Somānanda here refers with etat to tattvasyaikyam, which is found at ŚD 4.70a.

3 The reading here recorded, pratyakād grahaād is that of the manuscript of the Calcutta Sanskrit College. Both the editio princeps and the manuscript of Trivandrum University read pratyakāgrahanād.

4 The present reading, that of vā vyavahāragā is a conjectural emendation. Both the editio princeps and the manuscript of Trivandrum University read ca vyavahāragā; the manuscript of Calcutta Sanskrit College reads vācyavahārata. Another possible, though to my mind less likely, reading is vāgvyavahāragā.

Haut de page

Pour citer cet article

Référence papier

John Nemec, « On the contributions of the Śivadṛṣṭi of Somānanda to the Intellectual History of the Pratyabhijñā », Annuaire de l'École pratique des hautes études (EPHE), Section des sciences religieuses, 124 | 2017, 15-23.

Référence électronique

John Nemec, « On the contributions of the Śivadṛṣṭi of Somānanda to the Intellectual History of the Pratyabhijñā », Annuaire de l'École pratique des hautes études (EPHE), Section des sciences religieuses [En ligne], 124 | 2017, mis en ligne le 04 septembre 2017, consulté le 20 septembre 2017. URL :

Haut de page


John Nemec

Directeur d’études invité (2015-2016), M., École pratique des hautes études – Section des sciences religieuses, University of Virginia

Haut de page

Droits d’auteur

Tous droits réservés : EPHE

Haut de page
  • Les cahiers de