Studies on Utpaladeva's Ishvara-pratyabhijna-vivritti Part II-Torella

!!

WIENER!STUDIEN!ZUR!TIBETOLOGIE!UND!BUDDHISMUSKUNDE!HEFT!66!

!!

!!!

INDICA!!ET!!TIBETICA!

Festschrift!für!Michael!Hahn!Zum!65.!Geburtstag!

von!Freunden!und!Schülern!überreicht!

!!!

HERAUSGEGEBEN!VON!KONRAD!KLAUS!UND!JENS-UWE!!HARTMANN!

!!!!!!!

ARBEITSKREIS!FÜR!TIBETISCHE!UND!BUDDHISTISCHE!STUDIEN!UNIVERSITÄT!WIEN!

WIEN!2007!

WIENER STUDIEN

ZUR TIBETOLOGIE UND BUDDHISMUSKUNDE!!!

GEGRÜNDET!!VON!ERNST!!STEINKELLNER!

!!!

HERAUSGEGEBEN!VON!BIRGIT!KELLNER,!HELMUT!KRASSER,!

HELMUT!TAUSCHER!!!!

HEFT!66!!!!!!!!!!

WIEN!2007!!!

ARBEITSKREIS!FÜR!TIBETISCHE!UND!BUDDHISTISCHE!STUDIEN!UNIVERSITÄT!WIEN

!!!

!

INDICA!!ET!!TIBETICA!

Festschrift!für!Michael!Hahn!Zum!65.!Geburtstag!

von!Freunden!und!Schülern!überreicht!

!!!

HERAUSGEGEBEN!VON!KONRAD!KLAUS!UND!JENS-UWE!!HARTMANN!

!

!

!

!

!

!

!

!

!

!

!WIEN!2007!

!!

ARBEITSKREIS!FÜR!TIBETISCHE!UND!BUDDHISTISCHE!STUDIEN!UNIVERSITÄT!WIEN

Inhaltsverzeichnis

Verzeichnis der Schriften von Michael Hahn . . . . . . . . . . . . . . . . . . . 13

AN¨LAYOWho said it? Authorship Disagreements between P¢li andChinese Discourses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

NALINI BALBIRÀ propos des hymnes jaina multilingues (sanskrit, prakrit, per-san) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

CHRISTINE CHOJNACKILes dialogues dans le Kuvalayam¢l¢. L'inventivité d'Uddyo-tana au service de la foi jaina . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

RAHUL PETER DASOn the P¢da-End Break in ¹lokas of ÿurvedic Sa¾hit¢s . . . . 85

SIGLINDE DIETZThe Saptas‹ryodayas‹tra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

DRAGOMIR DIMITROVRatn¢kara¡¢nti's Chandoratn¢kara and Tath¢gatad¢sa'sChandom¢½ikya . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

AKIMICHI EDAFreigebigkeit in Bezug auf Frauen?! Eine Untersuchung zuden Quellen der Ratn¢val¤ des N¢g¢rjuna . . . . . . . . . . . . . . . . 139

GERHARD EHLERSSieben Seher, sieben Wasser . . . . . . . . . . . . . . . . . . . . . . . . . . 145

FRANZ-KARL EHRHARDA Short History of the g.Yu thog snying thig . . . . . . . . . . . . . . 151

HELMUT EIMERDie Liste der Mah¢y¢na-Texte im tibetischen Nandimitra-Avad¢na . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171

KARL-HEINZ GOLZIO¹iva nur noch auf Platz Zwei!? Der Buddhismus des Angkor-Herrschers Jayavarman VII. und die Integration des Hinduis-mus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

10 Inhaltsverzeichnis

ALBRECHT HANISCHNew Evidence of A¡vagho¼a´s S‹tr¢la¾k¢ra: Quotationsfrom the mDo sde rgyan of g an la phan pa'i dbyaËs in theTibetan Version of Dharmak¤rti's J¢takam¢l¢¿¤k¢ . . . . . . . . . 193

JÜRGEN HANNEDERVasubandhus Vi¾¡atik¢ 1 2 anhand der Sanskrit- und tibeti-schen Fassungen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207

PAUL HARRISONThe Case of the Vanishing Poet. New Light on ¹¢ntideva andthe ¹ik¼¢-samuccaya . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215

JENS-UWE HARTMANNEin Schauspielfragment aus Afghanistan . . . . . . . . . . . . . . . . 249

MONIKA HORSTMANNCaukasr¢ms Vermächtnis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259

TAKASHI IWATADharmak¤rti's Interpretation of the Word i¼¿a in the Definitionof the Thesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275

PETER KHOROCHEOn the Vocabulary of ¨rya-¹‹ra´s J¢takam¢l¢ . . . . . . . . . . . . 289

SIGNE KIRDEWieviele Wirkungen hat Schlangengift? Bemerkungen zurToxikologie im Schauspiel Bhagavadajjuka . . . . . . . . . . . . . . 295

KONRAD KLAUSZu der formelhaften Einleitung der buddhistischen S‹tras . . . 309

KLAUS-DIETER MATHESThe Ontological Status of the Dependent ( paratantra) in theSa¾dhinirmocanas‹tra and the Vy¢khy¢yukti . . . . . . . . . . . . . 323

MAREK MEJORA Tibetan Prose Version of K¼emendra's Bodhisattv¢vad¢na-kalpalat¢ X: Garbh¢vakr¢ntyavad¢na . . . . . . . . . . . . . . . . . . 341

ADELHEID METTEBuddhistische Sanskritstrophen aus dem Rotkuppelraum derMing-öi von Qizil: Proben aus der Fragmentsammlung SHT25 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351

ULRICH PAGELSt‹pa Festivals in Buddhist Narrative Literature . . . . . . . . . . . 369

Inhaltsverzeichnis 11

BHIKKHU P¨S¨DIKAThe Ekottar¢gama Parallel to J¢taka 77 . . . . . . . . . . . . . . . . . 395

ULRIKE ROESLERMaterialien zur Redaktionsgeschichte des mDzaËs blun: DieSelbstaufopferung des Prinzen Suj¢ta . . . . . . . . . . . . . . . . . . . 405

LAMBERT SCHMITHAUSENZur Frage, ob ein Bodhisattva unter bestimmten Vorausset-zungen in einer neutralen Geisteshaltung (avy¢k’ta-citta) tötendarf . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423

JOHANNES SCHNEIDERCandragomins ÿurvardhanavidhi. Ein Ritual zur Erzielungeines langen Lebens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441

PETER SCHWIEGERA Glance at the Problematic Relations Between DifferentBuddhist Traditions in Tibet . . . . . . . . . . . . . . . . . . . . . . . . . . 453

FRANCESCO SFERRAFragments of Pu½Îar¤ka's Param¢rthasev¢ . . . . . . . . . . . . . . . 459

JAYANDRA SONIAnek¢ntav¢da Revisited ) for do¼as . . . . . . . . . . . . . . . . . . . . 477

ULRIKE STARKMakkhanl¢l's Sukhsagar (1846/47): The First Complete Ver-sion of the Bh¢gavata Pur¢½a in Modern Hindi Prose? . . . . . 491

ROLAND STEINERDas Œdreifache Leiden– in S¢¾khyak¢rik¢ 1 . . . . . . . . . . . . . . 507

MARTIN STRAUBEDie Adaptation von K¼emendras Sudhanakinnaryavad¢na imBhadrakalp¢vad¢na . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 521

RAFFAELE TORELLAStudies on Utpaladeva's ª¡varapratyabhijñ¢-viv’ti. Part II:What is Memory? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 539

HELGA UEBACHGeheime Wörter und Zeichen aus dem Kreis der Heruka-Tantras . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 565

CLAUS VOGELThe Propitiation of the Planets in Indian Ritual and AlliedLiterature with Special Reference to Colours and Flowers . . . 587

12 Inhaltsverzeichnis

NOBUYUKI YAMAGIWAVinaya Manuscripts: State of the Field . . . . . . . . . . . . . . . . . . 607

This paper that I have dedicated to PROF. MICHAEL HAHN as a token of appreciation to1

the scholar and of affection to the old friend is perhaps a bit far away from his preferredfield of research, but I remember what once he told me during my stay at his house at Oden-dorf: that the fact that we both liked to work on unpublished sources created a stronger linkbetween us than the mere sharing the same subject ...

TORELLA 2002: XL!XLV.2

Studies on Utpaladeva's ª¡varapratyabhijñ¢-viv’tiPart II: What is Memory? 1

RAFFAELE TORELLA, Rome

The present paper is the second of a series of papers in which I have beenpresenting the critical edition and English translation of the fragmentarycodex unicus of the ª¡varapratyabhijñ¢-viv’ti, the long commentary thatUtpaladeva composed on his own ª¡varapratyabhijñ¢-k¢rik¢ (ªPK) andsvav’tti, a work of outstanding importance for the philosophy of Ka¡mirian¹aiv¢dvaita, and for Indian philosophy as a whole. I need not repeat herewhat I have already said elsewhere on the nature of the Viv’ti and its rela-tion to the other commentaries. Suffice it to recall that Utpaladeva is said2

to have composed the ªPK and the concise V’tti at the same time, and lateron to have devoted an analytic commentary to the complex K¢rik¢-V’tti,i.e. our Viv’ti (or Â¤k¢), in which he discussed possible alternative viewsand rejected them, also making occasionally quite long digressions on par-ticular subjects. Of this lengthy work ) corresponding to the extent of 8000¡lokas (hence the traditional denomination of A¼¿as¢hasr¤) ) only a com-paratively small fragment has come down to us, covering the section ªPKI.3.6 through I.5.3. A detailed exposition of my discovery of the originalmanuscript ) after a cursory consultation, some 15 years ago, of a transcriptof the same made by PROF. K.CH. PANDEY ) can be found in the first studythat I have devoted to this text (TORELLA forthcoming), which also containsa description of the manuscript (National Archives, Delhi, Skt. Mss. No.30).

The present paper deals with the Viv’ti on ªPK I.4.1!2. In the previouschapter Utpaladeva, referring to an enigmatic statement in the Bhagavad-

540 RAFFAELE TORELLA

Bhagavadg¤t¢ XV.15b mattaÀ sm’tir jñ¢nam apohana¾ ca “From Me derive memory,3

knowledge and exclusion.” So we read at the very beginning of the Viv’ti on I.4.1; see below, p. 535.4

See below, fn. 83.5

g¤t¢, had identified three powers (¡akti) in the Lord: Cognition, Memory3

and Exclusion. After making some preliminary remarks concerning them asa whole, he starts now a detailed inquiry into each of them. His aim is toshow that cognition, memory and exclusion, which constitute the very basisof the knowledge process in human mind, are indirectly also a proof of thecoinciding of the individual subject with universal Consciousness. None ofthese phenomena can be really explained and their complex functioningaccounted for satisfactorily in merely `mechanic´ terms, as first of all theBuddhists do. The individual subject can cognize, remember and excludeonly if it is conceived of as inscribed within an eternal and, at the sametime, dynamic universal I-ness, i.e. ¹iva.

If Utpaladeva's investigation starts with memory, by infringing the a-bove stated order, it is “because in a very clear manner (su¼pa¼¿am) mem-ory can serve as a logical reason for the establishment of the identity of theself with the Lord”. The starting point is the classical definition of memory4

given in Yogas‹tra I.11: “Memory is the non-extinction of the objectformerly perceived” (anubh‹tavi¼ay¢sampramo¼aÀ sm’tiÀ). The sustainedanalysis of Utpaladeva singles out a few crucial points contained in such anapparently simple process: How is it possible to attribute temporal differen-tiation to a cognizer that is permanent in his essential nature? What is therelationship between the cognitive act of the original perception and thecognitive act of the subsequent memory? How can the latter bring theformer to light again without objectifying it? On this point, in fact, the¹aiva and his principal opponent, the Buddhist epistemologist, are in fullagreement: a cognition is self-luminous and cannot be the object of anothercognition. The standard Buddhist explanation is far from being satisfactory:saying that the perception produces a sa¾sk¢ra, which in turn will producethe phenomenon of memory, only accounts for the fact that memory has acertain objective content but leaves out the `subjective´ component repre-sented by the fact that the object has been `coloured´ by the previous per-ception, or, to be more precise, by its having been àlready´ perceived in acertain past moment. Memory, in fact, is indeed the memory of the past ob-ject but also of the past perception of it. Instead, as Abhinavagupta says,5

what the sa¾sk¢ra is able to convey (or resurrect) is neither the original

Studies on Utpaladeva's ª¡varapratyabhijñ¢-viv’ti 541

The point has been explicitly touched, in a different context, by Utpaladeva in ªPK6

I.3.2cd [...] sa¾sk¢rajatva¾ tu tattulyatva¾ na tadgatiÀ “The fact that [memory] arisesfrom latent impressions implies its similarity to the former perception, but not its cognitionof that.” The sa¾sk¢ra of the former perception is awakened by a present perception ) sim-ilar to the other ) which gives rise to the memory. The sa¾sk¢ra, therefore, ensures this`similarity´ in the memory, but the memory itself has no direct access to (cannot `know´) theformer perception, nor can it, strictly speaking, establish the similarity between the latterand the present perception which has been reawakened by the sa¾sk¢ra (TORELLA 2002:99f., fn. 4). V’tti thereon: “Since memory arises from the latent impression left by the for-mer perception, it only bears a similarity to that perception but does not have direct cogni-tion of the latter; and, moreover, as there is no cognition of the former perception not eventhe similarity to it can be maintained.”

A lyuÎanta word like asampramo¼a½am is more inclined to express a process than the7

ghañanta word asampramo¼aÀ. See below, fn. 78. See below, p. 539.8

See below, p. 542.9

perception nor the object insofar as it was cognized by such past percep-tion. This presupposes a living organism at work, a dynamic and unitaryconsciousness able to freely move between different moments of time.6

Having this in mind, Utpaladeva deliberately introduces an apparently littlebut in fact quite significant change in the Yogas‹tra definition, by readingasampramo¼a½am in the place of asampramo¼aÀ. Due to the very nature7

of the phenomenon of memory, consciousness is expected to work at thelevel of individual subject, fully within the world of m¢y¢. As Utpaladevaputs it: “For this function belongs to the Lord alone, identical with con-8

sciousness, and takes place due to His freedom, in these terms: it is theLord that, having assumed the form of the [limited] knower, identified withthe purya¼¿aka and other planes onto which freedom is superimposed, cog-nizes, remembers or ascertains.”

But how, analytically, does the process of memory work? Both the act ofascertainment (ni¡caya, adhyavas¢ya) and memory belong to the categoryof vikalpa, being the m¢yic form of vimar¡a. The main difference betweenadhyavas¢ya, immediately following the original manifestation (or `shin-ing´) of the object ) that is, its perception ), and sm’ti, which is insteadmore or less distant from it, is that in the former case we have the reflectiveawareness (par¢mar¡a) `this´, while in the latter we have the reflectiveawareness `that´. However, according to the Pratyabhijñ¢ philosophy, onlya par¢mar¡a of a presently `shining´ object is possible. So memory cannothave as its object something which only `shone´ in the past (V’tti: prak¢¡i-tasya par¢mar¡o na k’taÀ sy¢t). Once the matter has been put in these9


Cf. ªPVV II, p. 17, ll. 22f. anubhavasm’tyor eka¾ svasa¾vedanar‹pam ekavi¼ayato-10

palambh¢t. Utpaladeva's concept of svalak¼a½a seems (and, in a sense, indeed is) just the opposite11

of the Buddhist's (for a detailed discussion see TORELLA 1992: 332!336). For the Buddhistepistemologist the starting point is the particular ; perception grasps it in its entirety but isalso inexpressible and uncommunicable; many different ni¡cayas may stem from this singleperceptual content, each of them captures a part of it and connects it with a word, whichtherefore denotes a certain s¢m¢nya (or rather the negation of what is other than that feat-ure). For Utpaladeva, each pram¢½a grasps an individual ¢bh¢sa (which is a s¢m¢nya), ex-pressed by a determinate word, depending on a determinate reflective awareness, or grasps

terms (the possibility that memory might have as its òbject' the former per-ception had been discarded at the outset), Utpaladeva is able to point to thecentrality of a dynamic I as the only way to get out of the impasse. It is theI that ensures the possibility of unifying the various cognitions occurring atdifferent times, thus resolving the apparent inconsistency between a (pres-ent) vimar¡a and a (past) anubhava. The one and same svasa¾vedana ofboth cognitions creates that necessary bridge between them which the Bud-dhist epistemologist fails to account for. A further clarification is provided10

by Abhinavagupta in the ªPVV (II, p. 32, ll. 10!13): the prak¢¡a concern-ing the part-object (arth¢¾¡e) belongs to the past; but the prak¢¡a as grasp-ed by the vimar¡a, concerning the part-self (sv¢tm¢¾¡e), is not limited bytime. Thus the vimar¡a in the memory can connect itself with the vimar¡ain the perception and, through it, with the former light of the object ) in thisway meeting both requirements: taking place in the present and not beingdivorced from prak¢¡a (TORELLA 2002: 106f., fn. 12).

In the course of the exploration of the mechanism of memory, Utpala-deva has to explain his position with regard to the object of memory. Theobject of memory is, in principle, the same as the object of perception: theunique particular, the svalak¼a½a of the Buddhists, belonging to a specifictime and space and having a specific form. But just as the original percep-tion had catched with definiteness only some aspects of the svalak¼a½a, sothe memory usually resurrects only some aspects of it, and not necessarilythe same. The starting point of this long digression, which will turn into aninvestigation into the nature of the universal, the particular and their rela-tionship, is the degree of vividness of the remembered object. Vividness(sphu¿atva) depends on how many aspects or components of the thing aremanifested in memory. In fact, the apparently unitary svalak¼a½a is com-posed of a number of `manifestations´ (¢bh¢sa), each of them having thenature of universal. But if it is true that the more numerous s¢m¢nyas11


) in the perception itself and not in a later cognitive act ) a group of ¢bh¢sas coordinated bythe Lord's power of necessity around a dominant ¢bh¢sa, which allows the perception toremain unitary. The group of ¢bh¢sas taken in its totality ultimately corresponds to the Bud-dhist svalak¼a½a. The two conceptions are after all not so opposed to each other: Utpala-deva's svalak¼a½a clearly derives from the svalak¼a½a of the Buddhists, only with a signifi-cantly inverted perspective.

See in particular I.2.3 (Candr¢nanda) s¢m¢nya¾ vi¡e¼a iti buddhyapek¼am, I.2.4 bh¢-12

vaÀ s¢m¢nyam eva, I.2.5 dravyatva¾ gu½atva¾ karmatva¾ ca s¢m¢nyani vi¡e¼¢¡ ca, etc. Cf. p. 741 s¢m¢nya¾ dvividha¾ ) param apara¾ ca; p. 743 tatra satt¢s¢m¢nya¾13

param anuv’ttik¢ra½am eva; p. 746 apara¾ [...] anuv’ttivy¢v’ttihetutv¢t s¢m¢nya¾ vi¡e¼a¡ca bhavati. See also Bhart’hari, J¢tisamudde¡a, v.14 anuprav’ttidharmo v¢ j¢tis sy¢t sarva-j¢ti¼u › vy¢v’ttidharmas¢m¢nya¾ vi¡e¼e j¢tir i¼yate.

unite, the more individualized is the object, and the more vivid is its mani-festation in memory, it is also true that s¢m¢nyas are not on the same plane.Even a single s¢m¢nya can as well be manifested vividly, when it compris-es a multiplicity of broader universals (this is the case, Utpaladeva says, ofthe universal `dhava´ [Grislea tomentosa] with respect, e.g., to the uni-versal `treeness´). In referring to a hierarchy of s¢m¢nyas, Utpaladeva islikely to have in mind Bhart’hari's conception, as expressed in J¢tisamud-de¡a, v. 33 sambandhibhed¢t sattaiva bhidyam¢n¢ gav¢di¼u › j¢tir ity ucya-te [...] (cf. Helar¢ja's commentary, p. 41 bhidyam¢n¢ upacaritabhed¢ gav¢-¡v¢di¼u sattaiva mah¢s¢m¢nyam eva j¢tiÀ gotv¢¡vatv¢dik¢, aparas¢m¢-nyam). Bhart’hari, in his turn, seems to derive this conception from the Vai-¼e¡ikas‹tra (I.21ff.), where the term s¢m¢nyavi¡e¼a appears for the first12

time. s¢m¢nyavi¡e¼a can be taken in a double meaning: “specific s¢m¢nya”(versus the highest s¢m¢nya: bh¢va) and “being at the same time s¢m¢nyaand vi¡e¼a”. Pra¡astap¢da calls them para and apara s¢m¢nya: the formercauses the notion of continuity, the latter that of distinction.13


The words underlined in the text are those literally cited in the ªPVV. Punctuation is14

mine (that found in the MS is often misleading). In the edition, <...> means àddition withrespect to the transmitted text´; [...], means èlimination of portions of the transmitted text´(also the indications `folio´ and `line´ are between square brackets). The sandhi has been`normalized´. The establishing of the text has resulted from the delicate balance of some-times divergent factors: the text as transmitted by the codex unicus, internal coherence,literal citations in the ªPVV, paraphrases in the ªPVV, parallel texts. A few literal citationsfrom the Viv’ti can be found in the footnotes of the KSTS ed. of the ªPV, which have beenderived from the marginal notes of an important ms. of ªPV, that sigled Gh in the edition ofthe ªPV (the ms. is now at the National Archives, Delhi, where I have consulted it). Whenthe transmitted text has been modified, this has been pointed out by using three differentexpressions: correction, conjectural emendation and tentative restoration. The first is felt asvirtually certain, the second as highly probable, the third as a mere attempt.

The MS regularly reads viv’tti- instead of viv’ti-.15

The avatara½ik¢ of the Viv’ti is cited in ªPV I, pp. 115f., fn. 17.16

The K¢rik¢ is followed by the text of the ªPV thereon.17

Abhinavagupta seems to have a partly different text: ªPVV II, p. 5, ll. 10f. `tena´ iti18

bodhena › `te´ iti anubh‹yam¢n¢À [...]. But neither tena nor te fit the context of the trans-mitted text, which on the other hand is confirmed by the paraphrase found in ªPVV imme-diately after the above passage: p. 5, ll. 11!14 `tath¢ ´ iti svar‹p¢d avibhakt¢ api bodhasv¢-tantryaprak¢re½a nirmitavibhaktat¢m iva ¢p¢dit¢À, ata eva vicitra¾ k’tv¢ bhinn¢ bodh¢cca anyonya¾ ca, tata eva de¡ak¢lavi¡e¼a½¢ ye arth¢À, taiÀ prak¢¡am¢½air yaÀ upar¢gaÀ,tena hetun¢ [...]. The reading ºvi¡e¼a½aº would seem preferable to ºupasarjanaº MS, but cf.p. 8, ll. 24f. kevalam apradh¢natay¢ tasya prathanam iti vi¡e¼a½atvam upasarjanatv¢t.

ºup¢d¢naÀ, my conjectural emendation, ºup¢d¢n¢t MS (cf. ibid. p. 5, ll. 21f. àta eva´19

¡aËkita¾ p‹rvapak¼am artha¡abdop¢d¢ne hetutvena upaj¤vati).

Text14

[9v13] viv’tau sm’ty¢d¤n¢m id¢n¤¾ jñ¢naprakara½en¢nena svar‹pa¾ nir‹pa-15 16

yi¼yann a¡e¼ak¢l¢nugatasvatantr¢vabodhalak¼a½¢tmatattvasvabh¢ve¡vara¡aktir‹pa-t¢¾ pratipip¢dayi¼uÀ sm’ter eva t¢vat suspa¼¿am ¤¡var¢tmasiddhihetuta[10r]y¢prathama¾ sambhavam ¢ha ››

sa hi p‹rv¢nubh‹t¢rthopalabdh¢ parato 'pi san › vim’¡an sa iti svair¤ smarat¤ty apadi¡yate ›› <1.4.1>

[...] ›› [11r10] v’ttiÀ ›› pa¡c¢d api p‹rv¢nubh‹t¢rth¢nubhavit’tv¢t p‹rv¢nubh‹-17

t¢rthaprak¢¡¢sampramo¼a½a¾ tasyaikasya vibhoÀ kartuÀ sa ity atra p‹rv¢nubh‹ta-tvena pratyavamar¡aÀ sm’tir n¢ma vy¢p¢raÀ ›› viv’tiÀ ›› bodh¢tmano nityatvenaitadavibhakt¢y¢À boddh’t¢y¢À nityatve 'pi tat-18

tattath¢nirmitavibhinnade¡ak¢lopasarjan¢rthopar¢ge½a bodhaikar‹pasy¢[11v]rth¢-nubhavasy¢pi ca bhinnak¢latvena p‹rvapa¡c¢dvyavah¢raÀ › tad ¢ha “pa¡c¢d api”iti › ata eva prak¢¡am¢n¢rth¢dh¤natv¢d asya k¢lavyavah¢rasya s‹tre v’ttau c¢rtha-¡abdop¢d¢naÀ › vi¼ayopalabdh’t¢py arthasya p’thagbh¢ve nopapannety anta-19

rarth¢nuprave¡adar¡an¢rtham arth¢k¢ro 'nubhavitety evamarthat¢tparye½a s‹tra-


k¢laº, my correction for ak¢laº MS (cf. p. 8, l. 9 kart’¡akteÀ k¢lakram¢vabh¢sana¾20

n¢ma vy¢p¢raÀ). Apparently in the text that Abhinavagupta had before him there was tatra in the place21

of tad¢n¤¾ (p. 8, l. 8 `tatra´ iti m¢y¢vi¼aye). e¼aiva, my correction for e¼ava MS.22

parame¡varasy¢º, my correction for parame¡var¢º MS.23

y¢ added above the line.24

asyaivai¼aiv¢ntaÀº, my conjectural emendation for asyaivai¼av¢ntaÀº MS.25

ºsv¢tantryaº cit. p. 11, l. 14, incorrectly, as ºsv¢tantrye.26

v¢ na tu, my correction for v¢ 'nantaº MS (na tu, strictly required by meaning, is cit.27

p. 11, l. 25). anubhavo 'rthasya prav’ttaº, my conjectural emendation for anubhavam arthasya pra-28

v’ttam MS. In emending the transmitted text, I have taken into account the long and com-plex explanation of the passage given by Abhinavagupta ibid. p. 14, l. 23 ! p. 15, l. 17.

atatº, my correction for ataÀ MS.29

MS has: ›› 25 ›› 1 ››30

v’ttyor artha¡abda¼a¼¿hy¢s t’c¢pi k’taÀ sam¢saÀ, sa ca jñ¢pak¢t kvacid i¼¿aÀ,t’nnantena v¢ s¢dhana¾ k’teti dvit¤y¢sam¢saÀ › nanv arthasy¢nubhavamukhenaivak¢lasa¾spar¡o vak¼yate vi¡e¼ato 'nubh‹tat¢nubh‹yam¢nat¢dir‹pe½a › satyam etat,kintv an¢dinidhanasyaiv¢sya sarva¡akte¡ cittattvasya m¢y¢ yat k¢lakram¢vabh¢-20

sanavy¢p¢r¢ kart’¡aktiÀ saiva ca tad¢n¤¾ k¢la¡aktisa¾jñ¢ tattadvividh¢vabh¢sa-21

n¤y¢dikakarmavastusa¾lagnatay¢ k¢lakramopasarjanavicitr¢vabh¢san¢dikriy¢r‹-patay¢ prathate › e¼aiva hi s¢ parame¡varasy¢vabh¢san¢khy¢ kriy¢ntastattat-22 23

pad¢rth¢vavibh¢yi¼¢tmakecch¢[12r]pr¢rambh¢ bahistadavabh¢sanaparyant¢ n¢nyatkiñcid iti vak¼yate › sm’tikriy¢ py asyaivai¼aiv¢ntaÀsthit¢nubh‹tap‹rv¢rthavi-24 25

mar¡ecchopakram¢ bahiÀ sa iti tatp‹rvak¢loparakt¢nubh‹tabh¢v¢vamar¡an¢va-s¢n¢ › tad ¢ha “anubh‹t¢rthaprak¢¡a” iti › sampramo¼a½am iti lyuÎantena kart’vy¢-p¢rat¢sphu¿¤kara½¢ya nirde¡aÀ gha¿¢divailak¼a½yena s¢dhyam¢nat¢pr¢dh¢nyasyar‹ÎheÀ › cinmayasye¡varasyaiva hi sv¢tantry¢d vy¢p¢ro 'ya¾ yad e¼a sam¢ropita-sv¢tantrya purya¼¿ak¢dipram¢t’r‹po j¢n¢ti smaraty avasyati v¢ na tu svas¢mar-26 27

thyena p’thagbh‹t¢n¢¾ jñ¢nasm’ty¢divast‹n¢¾ k¢ryak¢ra½abh¢vaÀ svatantr¢½¢¾yena vi¼ayabheda¡ codyate › ata eva “svair¤ ” ity up¢tta¾ viv’ta¾ ca “kartur” iti,anubhavasmara½ak¢lavy¢pitva¾ “sa parato 'pi sann” iti s‹tre nirdi¼¿a¾ viv’ta¾ ca“pa¡c¢d api p‹rv¢nubh‹t¢rth¢nubhavit’tv¢d” iti › tath¢ “ekasya” iti kart’taiva caai¡varyam uktam “vibhor” iti › kevalam anubhavo 'rthasya prav’ttatadarth¢va-28

bh¢[12v]sapram¢trantaraik¤bh¢vanirm¢½amukhena tad¢n¤¾ tadanubhavit’pratya-g¢tmani vak¼yam¢½anayen¢nuprave¡o nirm¢½am iv¢sy¢bh¢san¢d › avabh¢ty apitasminn atatsam¢rop¢pohan¢tman¢ ni¡cayena vin¢ vyavah¢ro na pravartate, sa ca29

ni¡cayo 'nubhavak¢la eva v¢ bh¢vyate 'yam iti, anyad¢ v¢ sa iti, ubhayath¢pi v¢ ›k¢l¢ntare tu sa ni¡cayaÀ pratyavamar¡¢tm¢ sm’tir ity ucyate › tad uktam “prak¢¡¢-sampramo¼a½a¾ pratyavamar¡a” iti › anugatade¡ak¢l¢navacchinnatattad¢bh¢sa-m¢trapratyavamar¡e tu vikalpam¢tram etac cottaratra vak¼yate <›› I.4.1 ››> 30


ºnirde¡yasya, my correction for ºnirde¡asya MS.31

Here MS adds sy¢d eva, then deleted.32

The avatara½ik¢ of the ªPV follows.33

The text of ªPV follows.34

All the mss. of the V’tti (and, accordingly, my edition) have ¢bh¢sayati.35

I add ca, omitted in MS (and also in one of the mss. I used for my edition of the V’tti)36

but explicitly cited by Abhinavagupta (p. 12, ll. 19!21 [...] “svasatt¢k¢la eva ca” ity anenav’ttau sphu¿¤k’taÀ › tena sm’tik¢le eva ca ¢bh¢sayat¤ti v’ttiyojan¢ ).

The evidence of the Viv’ti induces me to accept the reading asphu¿¢vabh¢saÀ instead37

of sphu¿¢vabh¢saÀ, as found in my edition of the V’tti (both readings had been transmittedby the mss.). I read ºavabh¢saÀ instead of º¢bh¢saÀ MS (all the mss. of the V’tti read º¢va-bh¢saÀ, which is also confirmed p. 25, l. 19 y¢ v’ttiÀ “avabh¢sa” ity ant¢.

MS reads tathaiva sphu¿a eva; the same reading can be found in three mss. of the V’tti38

(J, T; Ch; in Ch sphu¿a eva is added in the margin), while the rest of the mss. read tathaivaonly. In my edition of the V’tti I had accepted tathaiva and considered sphu¿a eva an oldgloss interpolated into the text. Now, if, as I believe after taking into consideration the argu-ments of the Viv’ti, in the previous line we must read asphu¿¢vabh¢saÀ, it is instead tathaivato be deleted and only sphu¿a eva to be left.

See below fn. 42.39

MS has ºdyate then corrected to ºdyeta.40

º taº added above the line.41

What may appear p. 22, ll. 12!14 as a quotation is in fact a paraphrase: ànubh‹tasy¢-42

va¡ya¾ prak¢¡aÀ´ iti `p‹rv¢bh¢sitatvenaiva prak¢¡aÀ´ iti ca vy¢khy¢tam [...].

<Viv’ti> sm’tivikalpar‹papar¢mar¡asamaye de¡ak¢l¢dyavacched¢t tath¢ntaÀsthi-tasya bahir api tad idam ity¢disarvan¢ma nirde¡yasy¢rthasya s¢marthy¢d eva31 32

prak¢¡o bhavat¤ty ¢ha ›› [...]33

[13r6] bh¢sayec ca svak¢le 'rth¢t p‹rv¢bh¢sitam ¢m’¡an › svalak¼a½a¾ gha¿¢bh¢sam¢tre½¢th¢khil¢tman¢ ›› <I.4.2>

[...] [14r8] V’ttiÀ ›› sm’ti¡akty¢ sa iti p‹rv¢nubh‹ta¾ svalak¼a½a¾ par¢m’¡an34

bh¢sayaty ev¢nyath¢ prak¢¡itasya par¢mar¡o na k’taÀ sy¢t svasatt¢k¢la eva ca ›35 36

tena smara½ak¢le na¼¿asy¢py ¢bh¢so na du¼yati › kad¢cit tv arthit¢va¡¢d gha¿a-k¢ñcanadravyasatt¢dyanyatamaik¢bh¢sar‹pe½aiv¢sy¢sphu¿¢ vabh¢saÀ, anyad¢ tu37

sarv¢tman¢rthitvena sphu¿a eva, ati¡ayanirantar¢vahitacetasas tu d’¼¿¢rthapratya-38

k¼¤k¢ra eva ››Viv’tiÀ ›› sm’tau de¡ak¢l¢dyavacchedena p‹rvatay¢nubh‹ta¾ par¢m’¡an niyata-r‹patay¢ svalak¼a½¢tm¢nam artha¾ par¢m’¡ati › tad ¢ha “sm’ti¡akty¢ ” iti “sva-lak¼a½am” iti ca › prak¢¡asvabh¢[14v]va¡ ca par¢mar¡o yath¢sau tasy¢va¡ya¾ pra-39

k¢¡anena tath¢ bh¢vyam, anyath¢ par¢mar¡anam eva nopapadyeta , ato 'tra n¢sti40

viv¢daÀ › ata eva “bh¢sayaty eva” ity avadh¢ra½a¾ tena ca s‹tre liËartho niyoga-kara½alak¼a½e viv’taÀ › kevala¾ yad yath¢ par¢m’¡yate tat tathaiv¢vabh¢syatesm’tau ca p‹rvaprak¢¡itasya par¢mar¡¢t p‹rvaprak¢¡ita tvenaiv¢bh¢saÀ › tad ¢ha41 42

“anyath¢ prak¢¡itasya par¢mar¡o na k’taÀ sy¢d” iti › e¼a eva ca anubh‹tavi¼ay¢-


I would be tempted to add here something like id¢n¤ntaÀ making contrast with the pre-43

vious tad¢n¤ntanaÀ; id¢n¤ntaÀ is in any case to be implicitly understood, if we want to ob-tain the expected meaning.

Fn. 62 of ªPV I, p. 124 reads as follows: yathokta¾ madhyapratyabhijñ¢y¢¾ ) arthit¢-44

va¡¢t sm’t¢v upaj¢yam¢n¢y¢¾ gha¿¢dy¢bh¢sasyaiva k¢ñcan¢dy¢bh¢sasa¾bhinnasy¢piyathopayoga¾ p‹rvade¡ak¢l¢bh¢s¢dyavacchedopar¢g¢t svalak¼a½ar‹pasya smara½¢d as¢vavabh¢so bhavati iti. The quotation shows some significant divergence from the text I haveestablished, also confirmed by Abhinavagupta's comments. Some substantial doubt remainsonly with regard to upaj¢t¢y¢¾/upaj¢yam¢n¢y¢¾, the latter fitting the context better.

¢bhogaº, my correction for bhogaº MS. The detailed comments of Abhinavagupta refer45

to the reading ¢bhogaº (ªPVV II, p. 25, ll. 20!23 ¢ samant¢t bhogaÀ sa¾vedan¢ntarbruÎi-t¢nantabh¢vasa¾sk¢ramadhyanimagnasya cidbh¢vasya unme¼a½¢tmakap¢lanacarva½avy¢-p¢ro 'nve¼a½¢tm¢, ¢bhoga¡ cint¢santatilak¼a½a¾ pr¢ [read: pra]½idh¢nam).

With some hesitation I correct ºsambhinnasy¢pi MS to ºasambhinnasy¢pi. The latter is46

the reading that Abhinavagupta's remarks (p. 25, l. 23 ! p. 24, l. 4) seem to presuppose. ekam api ca, my conjectural emendation for ekam api tat MS (the prat¤ka p. 26, l. 2547

reads evam api ca, but see the objection which immediately precedes it: [...] tatra v’k¼a itiekabuddhiÀ katham).

sampramo¼aÀ › p‹rvaprak¢¡itatva¾ ca p‹rv¢t¤tade¡ak¢lapram¢tr¢k¢ravi¡e¼¢vacche-denaiva prak¢¡an¢d bhavati › tad ¢ha “svasatt¢k¢la eva ca” iti › tad evam anyath¢par¢mar¡an¢bh¢vaprasaËg¢t paurvak¢likasy¢rth¢bh¢sasya samarthan¢t smara½a-k¢layog¢nupayog¢t p‹rv¢vabh¢sasya smara½ak¢le n¢¡¢n¢¡avic¢ro ni¼phala eva ›tad ¢ha “smara½ak¢le na¼¿asy¢pi” iti › et¢vat¢ ca sm’titvam upapadyate 'nyath¢p‹rv¢rth¢vabh¢s¢nunme¼e tad abhinavam eva jñ¢n¢ntara¾ sy¢t p‹rv¢rtham¢tra-vi¼aya¾ yogina iva › yogino hi yasya prathamaÀ k¢labhedaÀ sarvath¢ na vigalitom¢y¢maye [15r] purya¼¿ak¢dau pram¢tary avasth¢n¢parity¢g¢t tasya smaryam¢½airaparair at¤taiÀ sahac¢r¤ ka¡cid artho 'bhiyogava¡¢d ad’¼¿ap‹rvo 'pi t¢vaty a¾¡elabdhasvar‹ponme¼asya s¢k¼¢tk¢re½¢vabh¢ti, p‹rv¢nubhav¢sa¾spar¡enaiva v¢pram¢tur vikalpa eva ka¡cid at¤tavi¼ayaÀ svecchay¢ ki¾ notpadyate r¢j¢ mamapit¢bhavad ity¢diÀ › atr¢t¤te 'py arthe 't¤tade¡¢disahac¢ri½i jñ¢n¢bh¢sas tad¢n¤nta-nas tad¢n¤ntanapram¢t’sa¾lagnatv¢t, smara½e tu p‹rvapram¢t’mayaÀ › tasm¢t43

p‹rv¢vabh¢sonme¼aÀ sm’t¢v ava¡y¢bhyupagaman¤yaÀ › arthitv¢c ca ¢bhogava-44 45

¡ena sm’t¢v upaj¢t¢y¢¾ gha¿¢dy¢bh¢sasyaiva k¢ñcan¢dy¢bh¢s¢ sambhinnasy¢pi46

yathopayoga¾ p‹rvade¡ak¢l¢bh¢s¢dyavacchedopar¢g¢t svalak¼a½ar‹pasya sma-ra½¢d asphu¿o 's¢v avabh¢so bhavati › tad ¢ha “kad¢cit tu” iti › p‹rvade¡ak¢l¢dya-vacchedavirahe tu svatantravikalpe¼u s¢m¢ny¢k¢rasyaiv¢sy¢sphu¿asy¢vabh¢saÀ ›s¢m¢ny¢ny eva hi prabh‹t¢ni sa¾bh‹y¢nyony¢vacchedena vyavasthit¢ni sphu¿¢-bh¢sat¢m ¢s¢dya de¡¢dis¢m¢ny¢bh¢s¢vacchedena svalak¼a½at¢¾ bhajante › e-47

ka[15v]m api ca s¢m¢nyar‹pa¾ ¡¢kh¢dimattvam¢tra¾ v’k¼atva¾ dhavakhadir¢di-var½asa¾sth¢nas¢m¢nyavaicitry¢bh¢sasa¾mi¡ra¾ de¡ak¢l¢bh¢sas¢m¢nyasahasra-sambhedabhedita¾ c¢nek¢tmat¢m ¢padyate › svayam ade¡ak¢labhedatve 'pi paras-para¾ bhedaÀ paraspara¾ pary¢y¢vaccheda¡ ca s¢m¢ny¢n¢m anekasvalak¼a½at¢-pattihetuÀ, te¼¢¾ c¢nugu½yenaiv¢nyony¢vacchedo bhavati ¡¤tas¢m¢ny¢gnis¢m¢n-


I have accepted the reading ki¾cic ca, quoted by Abhinavagupta, which better fits the48

context; kiñca MS. anyas¢m¢ny¢navacchede, my correction for atyas¢m¢ny¢vacchede MS. anyas¢m¢ny¢-49

navacchede is what the sense requires and is also confirmed by Abhinavagupta's paraphrase(p. 28, ll. 5f.): anyena vyatiriktatay¢ par¢m’¼¿ena ¢bh¢s¢ntare½a avi¼e¡itatve 'pi.

The original reading of MS was hi, then corrected to hy aº. 50

The passage kad¢cit punar [...] na k¼atiÀ is quoted in ªPV I, p. 125, fn. 67: yathoktam51

a¼¿asahasry¢m ) kad¢cit punar atyantaikaras¢dh¢nodyuktasya nirava¡e¼avi¡e¼a½¢vabh¢sa-s¢m¢n¢dhikara½y¢pattivi¡ad¤k’taÀ smaryam¢½o 'rthaÀ s¢k¼¢tk¢ramaya eva sphu¿a¾ puraÀsphurati, tath¢ ca pratyak¼¤bh¢ve 'pi p‹rvad’¼¿atayaiva pratyak¼¤bh¢v¢d anubh‹tavi¼ay¢-sampramo¼aÀ sutar¢m iti sm’tibhedamadhyaga½an¢y¢¾ na k¼atiÀ. The text is the same asin MS, apart from tath¢ instead of tad¢ MS and the omission of the passage tad ¢ha [...]nairmalyam.

ºavadh¢naº, my conjectural emendation for º¢dh¢naº MS. Cf. the following avadh¢-52

nasya. ºk’t¢À MS, then corrected to ºk’taÀ.53

vij¢t¤yacittavyavah¢r¢bh¢vo nairmalyam, my tentative restoration for vij¢t¤yacittay¢-54

vyavah¢r¢bh¢vanairmalyam MS, evidently corrupt. ªPVV II, p. 30, l. 2 quotes, incorrectly, tad¢ na.55

p. 30, ll. 14f. quotes, incorrectly, svatantre 'pi.56

tad¢ hi MS, for which I substitute the more congruous tad¢ ca, quoted ibid. l. 23.57

vikalpat¢, my correction for vikalpaÀ MS. The expected reading vikalpat¢, required by58

the following nirvikalpat¢, is indirectly confirmed ibid. l. 23 tat katha¾ vikalpat¢ ukteti. p. 31, l. 1, v¢¡abdo 'tra na sa¾¡aye, api tu sambhavavikalpe.59

yayos tath¢tv¢bh¢v¢t › ki¾cic ca s¢m¢nyam anyas¢m¢ny¢navacchede 'pi svata48 49

ev¢nekas¢m¢nyamayaikas¢m¢ny¢tmatay¢ svalak¼a½at¢m an¢pannam api sphu¿a¾bhavati yath¢ v’k¼atv¢pek¼ay¢ dhavatv¢di › svabh¢vatvena hy avi¡e¼itam api50

tadantapravi¼¿av’k¼atv¢di s¢m¢nyat¢¾ na tu j¢h¢ty eva, de¡ak¢labhed¢vacchede-naiva hi svalak¼a½¤bh¢vaÀ › yad¢ punar arthit¢va¡enaiva sadgha¿ak¢ñcanalohita-tv¢dy¢bh¢sasa¾vedanenaiva p‹rvadikk¢l¢bh¢sasa¾bhedina ev¢rthasya smara½a¾tad¢ sphu¿a ev¢bh¢saÀ › tad ¢ha “anyad¢ tu” iti › kad¢cit punar atyantaika-51

ras¢ vadh¢nodyuktasya nirava¡e¼avi¡e¼a½¢vabh¢sas¢m¢n¢dhikara½y¢pattivi¡ad¤-52

k’taÀ smaryam¢[16r]½o 'rthaÀ s¢k¼¢tk¢ramaya eva sphu¿a¾ puraÀ sphurati › tad53

¢ha “ati¡ayanirantara” iti › nairantaryam avadh¢nasya vij¢t¤yacittavyavah¢r¢bh¢vo54

nairmalyam › tad¢ ca pratyak¼¤bh¢ve 'pi p‹rvad’¼¿atayaiva pratyak¼¤bh¢v¢d anu-55

bh‹tavi¼ay¢sampramo¼aÀ sutar¢m iti sm’tibhedamadhyaga½an¢y¢¾ na k¼atiÀ ›svatantro 'pi vikalpas tattatpram¢½apari¡uddh¢nekavi¡e¼a½avi¡i¼¿¢vabh¢son-56

mukho yad¢ bhavati, tad¢ sphu¿¤bhavaty eva so 'rthaÀ › tad ukta¾ brahm¢dibh¢¼a-½¢kar½anaunmukhy¢d iti › tad¢ ca vikalpat¢ sy¢n nirvikalpat¢ veti n¢sm¢ka¾57 58 59

grahaÀ › par¢mar¡a¡‹nyat¢ tu na kv¢cid api vidyate t¢d¢tmy¢t sarvaprak¢¡¢n¢¾ ›sarvatr¢nusmara½¢dau ca cittattvasyaiva sv¢tantrya¾, na tu svak¢ra½as¢marthy¢d


MS has: ›› 25 ›› 2 ››60

See fn. 3.61

If Utpaladeva decides to start with memory (out of the three ¡aktis under consideration:62

the above mentioned Knowledge (in general), Memory and Exclusion), it is because theanalysis of the phenomenon of memory makes immediately and easily evident that the latterpresupposes a knowing subject assumed as identical to the Lord (ªPVV II, p. 2, ll. 18!21t¢vadgraha½a¾ jñ¢n¢pohan¢der api anantara¾ nir‹payi¼yate svar‹pam iti krama¾ dyota-yati › nanu kuto 'yam eva¾bh‹taÀ kramaÀ › ¢ha suspa¼¿a¾ k’tv¢ ¤¡varasvabh¢vasya ¢tma-naÀ siddhau hetut¢ sm’ter yato 'sti, tato 'ya¾ kramaÀ).

Utpaladeva has in mind the classical definition of memory in Yogas‹tra I.11: anubh‹-63

tavi¼ay¢sampramo¼aÀ sm’tiÀ. It is the differentiated content of cognition to project a corresponding differentiation64

onto the cognizer and the cognition (p. 5, ll. 19f. tata¡ ca arthopar¢g¢d arthagatau de¡ak¢-lau pram¢t¢ra¾ ca anubhava¾ ca ¢vi¡antau lak¼yete).

udbhavat¤ti pratip¢dayitu¾ “p‹rv¢bh¢sitam” iti “¢bh¢sayati” iti ca ½ic¢ nirde¡aÀk’taÀ “prak¢¡itasya” iti ca <I.IV.2>60

Translation

Viv’ti: The author, who, with the present section on knowledge, is going now todescribe the specific nature of memory etc. ) with the aim to demonstrate that61

memory is formed by the power of the Lord, constituting the very nature of the selfas characterized by a free dynamic awareness running through all times ) starts infact his proving argument precisely with respect to memory, since in a very clearmanner memory can serve as a logical reason for the establishment of the identityof the self with the Lord.62

1. The Free One (svair¤), the perceiver of the object formerly per-ceived, continuing also to exist later, has the reflective awareness:`that´: this is what is called `remembering´.

V’tti: The light-perception of the object formerly perceived is not extinguished(asa¾pramo¼a½am) [at the moment of memory], since he [the knowing subject,63

the Lord] also continues to exist later as the perceiver of the object formerly per-ceived. To have here a reflective awareness of an object in terms of `that´ ) asformerly perceived ) on the part of him who is one, the Lord, the agent: this is thefunction called `memory´.Viv’ti: Although the cognizer is permanent in his essential nature, since permanentis the consciousness from which his nature is never divided, the notions of `beforeánd àfter´ can be applied [to the self and his cognitive activity] in common experi-ence on account of the temporal differentiation of the object perceptions, whose64

essence is in fact solely consciousness: this [seeming] temporal differentiation isdue to the influence of the objects, which are various (tattatº), created [by the


That is, as if they were separated from their essential nature as consciousness.65

My translation follows Abhinavagupta's paraphrase (see fn. 18 to the edition).66

According to Abhinavagupta (ibid., p. 6, ll. 5f.), this is meant to exclude the alternative67

view: the object has the form of the perceiver. That is, in some particular cases used by cultivated persons, but it is not to be taken as68

generally valid as this would entail the complete non-applicability of explicitly statedgeneral rules, like, in this case, P¢½ini II.2.15 t’jak¢bhy¢¾ kartari (p. 6, l. 12 `kvacit´ ¡i¼¿a-prayukte › `s¢rvatrika´ iti jñ¢pake `t’jak¢bhy¢m kartari´ ni¼edo 'navak¢¡aÀ sy¢t; apparently`s¢rvatrika´ is a citation from the Viv’ti, but it is not found in the MS). The issue whether ajñ¢paka can be applied to the generality of cases has often been debated among the gram-marians and usually answered in the negative (cf. the well known paribh¢¼¢ “jñ¢paka-siddha¾ na sarvatra”). See e.g. Puru¼ottamadeva's explanation in his Laghuparibh¢¼¢v’tti(p. 160, ll. 8!10), which is also quite close to the wording of Utpaladeva's text: jñ¢paka¾hi n¢ma na v¢caka¾ ki¾ tarhi iËgitena s‹caka¾ kvacid eva i¼¿asiddhav ¢¡r¤yate na sarva-tretity yujyate jñ¢pakasiddha¾ na s¢rvatrikam iti › [...] ki¾tu vi¡i¼¿a[read: ¡i¼¿aº?]prayo-gadar¡an¢t kvacid eva). In other words, as we can read in a marginal note found in the ms.sigled Gh in the KSTS ed. of the ªPV and included in the fn. 44 of the same edition (p. 120),this grammatical `mistake´ is not to be imputed to the carelessness of the author, disregard-ing the prohibition for compounding a kart’- or karma-¼a¼¿h¤ with an agent noun in -t’ or-aka stated in P¢½ini II.2.15, but to his specific intention (yatnataÀ in the Vimar¡in¤) topoint out one of the cardinal tenets of the Pratyabhijñ¢ (`t’jak¢bhy¢m kartari´ iti s‹tre½a hisam¢saÀ atra prati¼iddhaÀ iti parasya pram¢dakathitam etat iti vy¢moha¾ niv¢rayitum ¢ha`yatnataÀ´ iti sam¢sas tv atra vaktavyavastu¡raddhay¢ k’taÀ).

In this case the jñ¢paka is to be found in the intention of the s‹trak¢ra himself, then69

restated by the v’ttik¢ra, to underline through the model of what happens in nominal com-position (the ek¢rth¤bh¢va of its components) the essential non-separateness of the three ele-ments at issue: artha-anubhava-anubhavit’ (cf. ªPV I, p. 120, l. 7 ! p. 121, l. 3). In the syn-thetic expression of nominal composition (v’tti) the meanings of the single words mergeinto the overall meaning of the v’tti just as dust mixes with water, so that in the end thewater/v’tti does not show any more the `troubling´ presence in itself of the dust/single

Lord's freedom] in such a modality (ºtath¢nirmitaº), separated [from the con-65

sciousness and among themselves] (ºvibhinnaº) and qualified by time and space(ºde¡ak¢lopasarjanaº). This is said [in the V’tti] with the phrase “also later”.66

Precisely because of this ) i.e. the fact that the use in common experience of thenotion of time depends on the objects insofar as they are manifested ) the S‹traand the V’tti have the word “object” (arthaº). But also the condition of perceiverof any content of perception (vi¼ayaº) would be inadmissible were the object sepa-rated [from the perceiver]. Therefore, in order to indicate that there is an innerinterpenetration with the object, [both the S‹tra and the V’tti intend to state that:]“the perceiver has the form of the object” . It is having this intention in mind that67

both in the S‹tra and the V’tti the author compounds àrtha´, which would havethe genitive ending in the analytic sentence, with ànubhavit’´ ending with theagent suffix t’c. Such a compound is to be accepted as correct in some particularcases due to a jñ¢paka . Or else, we can take it as composed by the words68 69


words: ibid. p. 6, ll. 7!9 sam¢se hi v’ttipad¢rtho v’ttyarthe p¢¾sur iva udake mi¡r¤bhavantatk¢lu¼yar‹¼a½a¾ tatra avabh¢sayati (read: n¢vabhasayati; the correct reading is alsofound in the above cited fn. 44 of the KSTS ed. of the ªPV I, which is based on the ªPVV).The above remark, which might be also taken, as I did, as referring to the close interconnec-tion of the terms in a sam¢sa and their merging into the overall meaning of the synthetic ex-pression (v’ttyarthe) is given a much more specific connotation in Abhinavagupta's dis-course in the ªPVV: in spite of the fact that such a compound goes against the grammaticalrules, nonetheless it is provided with a sort of `correctness´ by the urge of the author to ex-press what he aims to express (p. 6, ll. 6f. `t’c¢pi´ iti prati¼iddho 'pi sam¢so ´tra vaktavya-vastu¡raddhay¢ k’taÀ). The `force´ of the overall intended meaning of the compound hasthe capacity to make the imperfection or unfitness of its single components (tatk¢lu¼yar‹-¼a½a¾) unnoticeable.

The suffix t’n produces agent nouns identical to those ending with t’c except for hav-70

ing a verbal regime and shifting the ud¢tta onto the root; see LAZZERONI 1997. The vigrahaform would be: artham anubhávit¢. The possibility for the two words to form a dvit¤y¢-sam¢sa is, however, not allowed, unless (Bh¢skaraka½¿ha says in his Vy¢khy¢, vol. I, p. 149,ll. 2f.) we resort to the yogavibh¢ga device and divide P¢½ini II.1.24 dvit¤y¢ ¡rit¢t¤tapatita-gat¢tyastapr¢pt¢pannaiÀ into two parts. According to Bh¢skaraka½¿ha, yatnataÀ in the ªPVrefers precisely to the use of the yogavibh¢ga; this interpretation can also be found in a mar-ginal note of the ms. Gh of the ªPV, then reproduced in fn. 43 of the KSTS ed. of the ªPV (I,p. 120). For his part, Utpaladeva refers to v¢rttika 2 under P¢½ini II.1.33 (Mah¢bh¢¼ya, I, p.386, l. 13) s¢dhana¾ k’teti v¢ p¢dah¢rak¢dyartham, which, in order to justify compoundslike p¢dah¢raka, states that a factor [whatsoever] can be compounded with a k’danta (wemay add: in the absence of a specific P¢½inian rule to exclude it). Through K¢ty¢yana'smore allowing interpretation of P¢½inian rules concerning the k’danta compounds, also thedvit¤y¢sam¢sa with a t’nnanta as uttarapada comes to be accepted. Interpreting anubhavit’as a t’nnanta instead of a t’janta, says Abhinavagupta in the ªPVV (p. 6, l. 15), would havea further implication: a restatement of the fact that the whole discussion on memory pre-supposes a `m¢yic´ scenario (t¢cch¤ly¢din¢ hi m¢y¢padam avadyotitatara¾ sy¢t; t¢cch¤la,taddharma and s¢dhuk¢ri are the three possible meanings of t’n according to P¢½iniIII.2.134f.). In conclusion, either as ¼a¼¿h¤sam¢sa (incorrect, but allowed due to jñ¢paka) ordvit¤y¢sam¢sa (allowed, but only after applying K¢ty¢yana's v¢rttika or an ad hoc yoga-vibh¢ga operation), the compounds p‹rv¢nubh‹t¢rthopalabdh’- in the S‹tra and p‹rv¢nu-bh‹t¢rth¢nubhavit’- in the V’tti point to a bhed¢bheda view of the object, its perception andthe perceiving subject. More analytically, Abhinavagupta says (ªPV I, p. 120, l. 6 ! p. 121,l. 3), they show that perception comes to be (apparently) qualified by time through itsobject, that is, it `rests´ on the object, while both the perception and the object rest on theknowing subject. Grammatically, p‹rv¢nubh‹ta- insofar as it is a vi¡e¼a½a of -artha rests onit (the past participle suffix kta gives it a temporal qualification); in their turn, they rest onthe perceiver, who is the central element, the pradh¢na as the uttarapada in the tatpuru¼acompound (Bh¢skar¤, I, p. 155, ll. 8!10).

àrtha´ and ànubhavit’´ with the latter interpreted as ending with the agent suffixt’n (t’nnantena). In this manner, the compound comes to be a dvit¤y¢sam¢sa, al-lowed on the basis of [the v¢rttika] “a factor [can be compounded] with a wordending with a primary suffix” (s¢dhana¾ k’t¢).70


According to Abhinavagupta (p. 6, l. 19), the reference is to II.1.4ab kramo bhed¢¡ra-71

yo bhedo 'py ¢bh¢sasadasattvataÀ “Succession is based on differentiation and this in turnderives from the existence or non-existence of a certain manifestation” (TORELLA 2002:154).

According to Abhinavagupta (p. 9, ll. 17f.), the reference is to II.4.21 ittha¾ tath¢72

gha¿apa¿¢dy¢bh¢sajagad¢tman¢ › ti¼¿h¢sor evam icchaiva hetut¢ kart’t¢ kriy¢ “Thereforecausality, agency, action are nothing but the will of Him who whishes to appear in the formof the universe, in the various manifestations of jar, cloth and so on.” (TORELLA 2002: 187).

In the limited subject (p. 10, l. 1).73

That is, in terms of separation from the limited subject (p. 10, ll. 5f. bahir iti tata eva74

sa¾kucitasa¾vidr‹p¢t vedak¢t p’thagbh¢vena). I take anubh‹tabh¢va in the sense of [arthasya] anubh‹tatva (but the easier interpreta-75

tion as anubh‹to bh¢vaÀ is also possible). The additions between brackets come from the thorough gloss and elucidation of this76

very concise passage furnished by Abhinavagupta (p. 10, ll. 1!9 antar iti pratyag¢tmar‹pedehapr¢½apurya¼¿aka¡‹ny¢vacchedini sa¾vittattve yaÀ sthitaÀ p‹rv¢nubh‹taÀ pr¢ktana-prak¢¡apar¢mar¡ar‹pay¢ idantay¢ upalak¼ito 'rthas tasya yo vimar¡as tamo'pas¢ra½enasphu¿¤kara½am, tatra y¢ icch¢ saiva upakramaÀ pr¢rambho yasy¢ bahir iti, tata eva sa¾-kucitasa¾vidr‹p¢t vedak¢t p’thagbh¢vena sphu¿avedyatvena yo 'dhun¢ vimar¡aÀ, sa yataÀpr¢kk¢labh¢vin¢ vimar¡ena prakhy¢[read: pr¢cy¢]nubhavasvabh¢vaprak¢¡aj¤vitena mili-taÀ, sa ity eva¾r‹pe½a ¡abdena ¢vi¼¿aÀ ubhayak¢laspar¡it¢y¢m api p‹rvak¢laspar¡apra-dh¢naÀ prak¢¡ate [...] ).

[Objection:] But later on the text will put the matter in a different way, i.e. that71

it is instead through the perception that the contact of the object with time occurs,in the form of “[the object] has been perceived, is being perceived, and so on”.[Reply:] This is true. However, to this Consciousness-principle, which is indeed initself without beginning or end and possesses all powers, belongs m¢y¢, that is, thepower of the agent subject whose activity is the manifestation of the temporalsuccession, and then this very power, with the denomination of `power of Time´,appears in the form of various and multiform activities like `making manifest´ andso on, qualified by temporal succession insofar as they depend on their objects,that is, the various and multiform things to be manifested, and so on. This is theaction named `manifestation´ (¢bh¢sanaº), belonging to Parame¡vara; it beginswith the will, consisting in the internal manifestation of the various things, andends with their external manifestation. It is this and nothing else, as we shall see.72

Also the act of memory ultimately belongs to Him. In fact, memory starts withthe wish to regain the reflective awareness of an object perceived in the past andcurrently stored internally (antaÀ) , and ends externally with the [present] reflec-73 74

tive awareness [of the object] in terms of `that´ ) i.e. the reflective awareness ofthe object as formerly perceived (ºanubh‹tabh¢vaº) , `coloured´ by [the reflective75

awareness belonging to] that former time. He means this when he says [in the76

V’tti]: “ the light of the object perceived”.


Instead of the ghañanta word sampramo¼a, as found in the classical definition of mem-77

ory in Yogas‹tra I.11. While the nominal stems are generally considered èstablished´ (siddha), namely78

frozen items with respect to verbs, whose core is an action in progress (s¢dhya), instead theaction nouns belong, so-to-speak, to an intermediate sphere (cf. Mah¢bh¢¼ya ad P¢½iniV.4.19, vol. II, p. 432, l. 18, k’dabhihito bh¢vo dravyavad bhavati). Utpaladeva, and thenAbhinavagupta, seem to introduce a further distinction of their own (or, at least, I havefailed to find their vaiy¢kara½a source): the verbal action expressed by a k’danta word mayhave a higher `dose´ of s¢dhyat¢ when expressed by a lyuÎanta with respect to a ghañanta.To this topic, and more in general to the very interesting siddha-s¢dhyat¢ issue in Indiangrammatical thought, I am devoting a separate study.

Abhinavagupta in ªPVV (p. 10, ll. 19!21) adds that the neuter gender of a lyuÎanta79

word is also significant with respect to the masculine gender of the ghañanta word: theword in the neuter, though possessing the qualification of being siddha, does not dismiss itsalso being s¢dhya, just like adverbs do.

This is an implicit reply to the opponent who might object (p. 11, ll. 11!14) to the attri-80

bution to the Lord of an activity which obviously points to a state of spatio-temporallimitedness: how could the Lord, whose knowledge perpetually embraces everything (sar-vataÀ prak¢¡ar‹pa¾ sarv¢rthakroÎ¤k¢ri) perceive a particular thing, then keep it in a sortof dark deposit, then resurrect it.

According to Abhinavagupta, in saying purya¼¿ak¢diº Utpaladeva means to totally ex-81

clude the ¡‹nyapram¢t’, who is not involved in vyavah¢ra activities, like memory (p. 11, ll.16f. ¡‹nyasy¢pi vyavah¢r¢t¤tasya smart’t¢ na bh¢ti); the body, though not mentioned di-rectly, is not to be excluded, due to its close association with the purya¼¿aka. The latter alsoincludes pr¢½a from which it cannot be separated. The mention of their autonomy as sam¢-ropita means that sv¢tantrya naturally belongs to the Lord only, and only through him it is,as it were, extended to other levels of subjectivity; the point is clearly stated in ªPK IV.2.

By using the lyuÎanta word [a]sampramo¼a½a he means to clarify the fact that77

memory is a function of the agent subject, since [in such words] the aspect of78

`being brought about´ is conventionally deemed as predominant, whereas it not sowith words like `jar´ and so on. For this function belongs to the Lord alone, iden-79

tical with consciousness, and takes place due to His freedom, in these terms: it isthe Lord that, having assumed the form of the [limited] knower, identified with80

the purya¼¿aka and other planes to which freedom is superimposed (sam¢ropita-sv¢tantryaº), cognizes, remembers or ascertains. And it is not that by their own81

power (svas¢marthyena) cognition, memory and so on, conceived of as entities(ºvast‹n¢m), separated from [any agent subject], establish among themselves a


In this case, memory as an independent èvent´ (vastu) would be simply the effect of82

the impression left by the former perception, itself also an independent vastu. This is theBuddhist position, clearly summarized by Abhinavagupta (p. 11, ll. 23!25 anubhavaÀsa¾sk¢raÀ sm’tir iti tu jñ¢nam¢laiva up¢d¢nop¢deyabh¢vena bhavi¼yati).

According to Abhinavagupta, this unacceptable (to the ¹aivas) position has been criti-83

cized by Utpaladeva in ªPK I.3.2. In this K¢rik¢ it is strongly underlined that a cognitioncannot be the object of another cognition: this applies also to memory. Also the Buddhistsagree on this, but their alternative explanation of memory is far from being convincing.What the sa¾sk¢ra can cause is the mere fact that memory has a content, but this contentcannot be either the original perception nor the content as having been the object of a formerperception (ªPV I, p. 97, ll. 5!8 sa¾sk¢r¢t para¾ savi¼ayat¢m¢tra¾ sm’ter siddham, na tuanubhavavi¼ayatvam, n¢pi asya vi¼ayasya p‹rv¢nubhavavi¼ay¤k’tatvam). See above, fn. 6.

As Abhinavagupta specifies (ªPVV II, p. 13, ll. 22!24), here vibhu means “ the One84

who exists in various modalities” (vividha¾ k’tv¢ bhavati), and is not to be taken in theusual sense of “pervader” (vy¢paka) because, strictly speaking, this would presupposes theexistence of spatial differentiation.

That is, the difference between fresh perception and memory. Here Utpaladeva is ap-85

parently replying to the opponent who might say that, if memory is defined as anubhav¢-sampramo¼a½a, it is after all itself a perception, or, to be more precise, the `coming forthagain of that very perception´ (p. 14, l. 22 anubhavasyaiva punarunme¼aÀ). But, if this ishow things stand, then, when we experience this coming forth again of the perception, whydo we not say in common usage that we are just `perceiving´ (ll. 22f. tat punarunme¼e 'pianubhav¢m¤ty eva na ki¾ vyavah¢raÀ)? Having this possible objection in mind, Utpaladevaproceeds to clarify what is the specific nature of anubhava ( I follow Abhinavagupta'sunderstanding of the passage).

The fresh perception is felt by the perceiver as the appearance of something totally86

new, as a `creation´. But in fact this is not a creation proper (of something that was not be-fore), since other subjects have already experienced it previously (p. 15, ll. 15f. nirm¢½a-

cause-effect relationship independently, so that this thesis would lend itself to the82

criticism that there is a different object (vi¼ayabhedaÀ).83

For this very reason, the K¢rik¢ says `the Free One´ (svair¤) glossed by the V’ttiwith `[of] the agent´ (kartuÀ). The capacity of the agent subject to embrace boththe time of perception and the time of memory is stated in the K¢rik¢ by the phrase“continuing also to exist later” (sa parato 'pi san), commented on by the V’tti with“since he also continues to exist later as the perceiver of the object formerly per-ceived” (pa¡c¢d api p‹rv¢nubh‹t¢rth¢nubhavit’tv¢d). Then, òf Him who is one´(ekasya) sovereignty, that is, his being the agent subject is indicated with the wordòf the Lord´ (vibhoÀ).84

But there is a difference to specify (kevalam): the object perception consists in85

the entrance [of the knower] ) through the `creation´ of his unification with otherknowing subjects in whom the appearence of a certain object is taking place ) intothe dimension of individual subject, namely, of the perceiver of that same object atthat very time (how this can happen we will explain later on); this perception is ina sense also a creation, since that object is made manifest. Even if that object86


tay¢ cak¢sad api na nirm¢½ar‹paÀ p‹rva¾ pram¢trantarar‹patay¢ bh¢v¢t › iha[read:iva]¡abdo 'nubhavitur abhim¢no 'yam iti dyotayati).

Generally speaking, this act of ascertainment (ni¡caya) ) or, to use a broader term, of87

conceptual elaboration (vikalpa) ) operates on the content of perception, which would bedestined by its very nature to remain in the reign of nirvikalpa, that is, over and above thesphere of ordinary reality (vyavah¢ra). To resort to the nice simile of Abhinavagupta, the vi-kalpa is just like the door keeper of the dancing hall who helps the poor villager in his desireto come out from this too lofty place where he has incidentally found himself and comeback to the more modest reality, the world of m¢y¢, where he is accostumed to live (p. 15,ll. 21!23 tad aya¾ gr¢mya iva l¢syamandir¢t nirjigami¼an dv¢rap¢lena vikalpena anu-g’hyate m¢y¢pr¢Ëga½e nik¼ipat¢); likewise, the vikalpa helps the human subject leave thevery ùncomfortable´ plane of pure perception and come back to ordinary life where hecould make use of the wealth (less shining, of course ...) of nirvikalpa perception. This men-tal act will result in the awareness `this´, when it follows immediately the perception and isclosely linked to it; in the awareness `that´, when it is more or less disconnected from per-ception (see below); or else, in a combined awareness (this-that), which characterizes the actof recognition.

Both the general vikalpa and memory take place after some time from perception and88

produce the awareness `that´. But there is a significant difference between them. While theformer totally disregards the time of any original perception (it can even ignore any definiteperceptual content), the latter is characterized by its connecting itself, through its presentreflective awareness, with the reflective awareness which immediately followed the originalperception: the smart’ resurrects in the present the original perception, leaving intact itshaving occurred originally in the past. As Abhinavagupta puts it, id¢n¤ntan¢vabh¢sana-k¢lapar¢mar¡o ' pi na nim¤lati ) iti etatpar¢mar¡abhittipr¢dh¢nyena p‹rvak¢lapar¢mar¡aÀiti viruddhap‹rv¢parapar¢mar¡asvabh¢va eva `sa´ iti par¢mar¡a ucyate ( ªPV I, p. 119, ll.8!12).

Cf. ªPVV II, p. 17, ll. 11!15 yadi tu sa ity etat sv¤k’tau p‹rvau de¡ak¢lau na asya89

par¢m’¡ati, api tu n¤la¾ p¤tam ity et¢vanm¢tram tat ni¡cayanasmara½apratyabhijñ¢n¢di-vi¡e¼avyapade¡anibandhan¢yog¢d ap‹rvat¢n¢vabh¢s¢c ca ¡uddham eva vikalpanam ityucyate.

ªPK II.3.4, etc.90

shines in perception (avabh¢ty api), it cannot enter worldy transactions without anact of ascertainment (ni¡cayena), consisting in the exclusion of different objectswhich might be wrongly superimposed on to it. And such ascertainment may occurat the very moment of perception, thus producing the notion `this´, or in a differentmoment, thus producing the notion `that´, or with a combination of the two times.87

But if this ascertainment takes place subsequently (k¢l¢ntare), having the [former]reflective awareness as its very essence, then we call it `memory´. This has been88

said [in the V’tti, with the phrase]: “ the non-extinction of the light [...] the reflec-tive awareness”. Instead, if there is a reflective awareness of this or that manifes-tation as not delimited by the space and time that were originally connected withthem, we have a generic vikalpa (vikalpam¢tram). This topic will be dealt with89

later on. ! 1 !90


Ìnternal´ insofar as the object is no more in the range of sensorial pereception; èx-91

ternal´ in the sense that it appears as separated from the I (p. 19, ll. 8!10 antar iti indriya-bh‹my atikramya sthito 'pi bahir iti sa¾vidbh¢g¢t vyatiriktatvena ahant¢vibhaktena).

This obvious remark (obvious in the context of the Pratyabhijñ¢ philosophy) is the92

explicitation of arth¢t às a matter of fact´ in the K¢rik¢ (p. 18, l. 14 arth¢t s¢marthy¢tpar¢mar¡¢nyath¢nupapatty¢).

See below, fn. 102.93

Having modified the text of the V’tti with respect to my edition of it (TORELLA 2002:94

15, l. 9), I have modified my translation accordingly. See above fn. 37 and 38 to the text. My translation agrees only with the latter part of Abhinavagupta's explanation (p. 19,95

ll. 6!8 `niyatar‹patay¢ ´ ava¡yambh¢venety arthaÀ › sm’tau hi pr¢gde¡ak¢layogo 'va¡yamsphurati › anyath¢ vikalpam¢tra¾ tat bhavet, na tu sm’tiÀ).

Viv’ti: When reflective awarenesses in the form of that particular mental constructthat is memory occurs, there is, as a matter of fact (s¢marthy¢d eva), the manifes-tation of the object as being internal, and external as well, with the delimitation91

by place, time, etc. The object may be indicated through various pronouns, like`that´, `this´ and so on. The K¢rik¢ says:

2. As a matter of fact (arth¢t), [he who remembers] must necessa-rily, having a reflective awareness (¢m’¡an) of the particular enti-ty (svalak¼a½am) formerly made manifest, make it manifest at theactual moment of the memory (svak¢le), either as a single mani-festation `jar´ or as the totality of its components (akhil¢tman¢).

V’tti: Thanks to the power of memory the subject, when having a reflective aware-ness as `that´ of the particular entity formerly perceived, does make it manifest [inthe present] (bh¢sayaty eva). Otherwise, with regard to an object made manifest[only in the past] no reflective awareness could take place (prak¢¡itasya par¢-mar¡o na k’taÀ sy¢t); and this manifestation occurs at the very moment of the act92

of remembering. Therefore, the manifestation at the time of the memory also of anobject that has disappeared does not create any shortcoming [for the explanation ofthe phenomenon of memory]. At certain times, then, the object appears in the93

form of a single manifestation, i.e. limited to one of the many manifestations thatconstitute it, such as ) in the case of a jar ) `jar [in general]´ `golden´ ìndividualsubstance´ èxistence´ etc., depending on the subject's intentions. In these cases, itsmanifestation [in memory] is dim (asphu¿aº). At other times, instead, the object94

appears in its totality, since this is the subject's intention: its manifestation isindeed vivid. And, finally, the subject whose mind is intensely concentratedwithout interruption even directly visualizes the object formerly perceived.Viv’ti: In memory, when the subject has a reflective awareness of the object asformerly perceived, endowed with a spatial, temporal etc. delimitation, he is awareof the object as a particular entity insofar as it has a well defined nature. This is95

said [by the V’tti] with the words `thanks to the power of memory´ (sm’ti¡akty¢)


Cf. p. 19, l. 24 ! p. 20, l. 1 sa hi par¢mar¡aÀ prak¢¡anasya svabh¢vabh‹to dharmaÀ96

prak¢¡an¢bh¢ve niravalambhana eva katha¾ sy¢t. If the entity characterized by a certain essential property is absent, how can its essenti-97

al property be present (p. 20, ll. 3f. svabh¢vino hi abh¢ve asya aya¾ svabh¢va iti katham)? Utpaladeva is here probably reminiscent of the passage of the K¢¡ik¢v’tti (vol. III, p.98

128, l. 6 nimantra½a¾ niyogakara½am) on P¢½ini III.3.161 vidhinimantra½¢mantra½¢dh¤-¼¿asampra¡napr¢rtane¼u liË, which lists the possible meanings of liË. The slightly ambi-guous niyogakara½am by which the K¢¡ik¢v’tti explains nimantra½a (vol. III, p. 128) isglossed by the Padamañjar¤ with niyogato 'va¡yambh¢vena yat kara½am anu¼¿h¢nam (cf.Mah¢bh¢¼ya on P¢½ini III.3.161, vol. III, p. 327, yan niyogataÀ kartavya¾ tan nimantra-½am); in sum, nimantra½am means “ inciting one to do something that has to be obligatorilydone, like an obligatory ritual” (Padamañjar¤ ibid. ava¡yakartavye ¡raddh¢bhojan¢dau dau-hitr¢deÀ pravartaneti y¢vat). Abhinavagupta glosses niyogakara½a as follows: (ªPVV II, p.20, ll. 20f. niyogena niyamena yat kara½a¾ s¢dhyam¢natva¾, tat lak¼a½a¾ svar‹pa¾yasya). The possibility that the remembering subject might not make the former objectmanifest is to be excluded; on the contrary, it `does´ make it manifest (`bh¢sayet´ iti vidhi-r‹pe½a niyogena niyamo lak¼yate, na bh¢sayaty etat na, api tu bh¢sayaty eva iti). Theabove explicitation given by Abhinavagupta in the ªPVV shows incidentally that Utpala-deva, in the long debated issue of the meaning of liË, favours the thesis of the optative ex-pressing first of all the verbal action denoted by the verbal root, with an accessory qualifi-cation represented by vidhi, etc. ( l. 19 vidhyup¢dhikadh¢tvarthas¢dhyam¢natodrekacanamliËartha iti). This corresponds to the third of the four theses examined by grammarians (e.g.Padamañjar¤, vol. III, p. 129 aya¾ vidhy¢diÀ prak’tyarthaÀ, pratyay¢rthaÀ, prak’ty-arthavi¡e¼a½am, pratyay¢rthavi¡e¼a½a¾ ceti catv¢raÀ pak¼¢À), who instead generally fa-vour the fourth one; cf. SHARMA 1995: 577f.

The possible objection that Abhinavagupta has in mind is made explicit in ªPVV II, p.99

21, ll. 17f.: If you say that the object shines in the memory, it means that there is no differ-ence between memory and perception (since in both of them the object `shines´).

Though in the absolute sense prak¢¡a has no connection with time, nonetheless in the100

world of m¢y¢ it comes to be associated to the past or the present of the experiencer (in thathe identifies himself with his bodily dimension): the shining of the object in memory isassociated through vimar¡a to the shining that occurred in the `past´ experiencer.

and `the particular entity´ (svalak¼a½am). Just as the reflective awareness is theessential nature of the light of manifestation, so it must necessarily be accom-96

panied by the becoming manifest of the object, without which the act of reflectiveawareness itself would become impossible. Therefore, on this point there is no97

possible disagreement. Precisely for this reason there is [in the V’tti] the emphaticstatement ìt does make it manifest´ (bh¢sayaty eva), by which the meaning of theoptative form (li½) in the S‹tra [¢bh¢sayet] is commented on in the sense of beingcharacterized by òbligation´ (niyogakara½aº). But we should introduce a specifi-98

cation (kevalam): the manifestation of the object depends directly on the modality99

of the reflective awareness concerning it, and in memory the object appears assomething made manifest in the past, since the reflective awareness of it is pre-cisely that of something which has been made manifest in the past. This is said100


What has been said so far proves to be only an explication and reformulation of Pa-101

tañjali's definition: the `non-extinction´ of the object is reached through the `conservationóf its former perception within the present reflective awareness (p. 22, ll. 24 ! p. 23, l. 1anubhavo 't¤to hi atra prak¢¡abh¢vena vi¼ayasya uktaÀ, asampramo¼a¡ ca v¢rtam¢nika¾tath¢par¢mar¡anam).

In other words, if the manifestation of the (past) object does not occur in the present,102

also the question about the persistence or the total loss of the object becomes irrelevant, be-cause the `manifestation´ of the object at the actual time of memory is the very pre-con-dition for the phenomenon of memory to exist. There is no `loss´ for the memory from thedisappearance of the object, just as there is no `gain´ if the object continues to exist (p. 23,ll. 10f. n¢¡e arthena sm’ter na do¼aÀ, sadbh¢ve na po¼aÀ iti t¢tparya¾ v’tti¿¤kayoÀ). Thisdefinitely being what Utpaladeva had in mind, as confirmed now by the Viv’ti, my previoustranslation of the corresponding V’tti passage (TORELLA 2002: 105, “Therefore it is noterroneous to state that there is the manifestation, at the time of the memory, of an object,which, however, is no longer present at that moment.”) has to be revised accordingly (seethe translation above, p. 542).

That is, it would be only a past object, not a past object qualified by a past experience,103

actualized at the time of memory (cf. p. 25, ll. 14!16 p‹rvasya arth¢vabh¢sasya anunme¼e,p‹rvaprak¢¡itvena apar¢mar¡e, anubhavena at¤tena avacchinne [read: anavacchinne] 'rtha-m¢tre yo jñ¢navi¡e¼aÀ, so 'bhinava eva anubhava evety arthaÀ).

Abhinavagupta's comments (p. 23, l. 16 ! p. 24, l. 19) give us useful clues for clari-104

fying this dense passage. A yogin, according to Utpaladeva, can experience an object be-longing to the past independently from its having actually been the object of a past percep-tion. This is possible because of the ìntermediate´ nature of the yogin that has overcome the

[in the V’tti]: “Otherwise, with regard to an object made manifest [only in thepast] no reflective awareness could take place”. And it is precisely in this that the“non-extinction of the object formerly perceived” consists. And we can speak101

of something `made manifest in the past´ in the sense that the manifestation of thatobject was delimited by [the association with] a specific previous past place andtime, a specific subject, a specific form. This is said [in the V’tti]: “and [this mani-festation occurs] at the very moment of the act of remembering”. This is how thematter stands, for otherwise any discussion whether the past manifestation of theobject is extinguished or not at the time of memory would turn out to be vain: itwould be so since there would be no point in stating its connection with the time ofmemory once it has been established that the manifestation of the object has oc-curred in the past [only], for in this case the impossibility of the reflective aware-ness concerning it would necessarily ensue. This is said [in the V’tti]: “[the mani-festation], at the time of the memory, also of the object that has disappeared.”102

Only the above description, taken en bloc (et¢vat¢), can account correctly for thephenomenon of memory, for otherwise ) i.e., in the case that the previous mani-festation of the object does not come into play ) we would have just yet anotherspecific cognition, a new one, concerning the previous object in its generic form ,103

as in the case of the yogin. In fact, in the yogin the original full-fledged temporal104


stage in which he identified himself with the body but, being still involved with the mayicworld, has not yet reached a full identification with consciousness and still identifies himselfwith the stages between the body and pure consciousness, namely, the purya¼¿aka, thepr¢½a, the mind or the void. This enables him still to have a notion of temporal succession:if the latter is full-fledged (¢tyantika) only for the deha-pram¢t’, nonetheless it affects, moreor less dimly, also the purya¼¿aka etc. But, at the same time, the yogin also has a (stilllimited and sporadic) access to the identification with the supreme consciousness, “ that im-maculate consciousness which, though different from the presumptive identification withthe thickest veil represented by the body, is however intimately present in all levels ofsubjectivity (body, purya¼¿aka, etc.), just like the autumnal sun obscured by clouds” (p. 24,ll. 13!15). This immaculate consciousness, once occasionally freed of the veils dimming itby virtue of the intense application of the yogin, enables him to have direct access to pastevents. The events/objects are perceived as past but not through the intermediation of a pastperception; this can happen because these past but essentially `new´ objects are associatedwith other past objects which instead are actually remembered (cf. ll. 17!19 tam [read: ta¾tam] artham ap‹rv¢vabh¢sam eva tata eva smara½¢yogyam api p‹rvasmaryam¢½as¢hity¢tp‹rvatay¢ anubh¢ty eva). A similar case is the yogin's cognition of the past lives.

That is, if I understand well, with regard to the object, which, though not having been105

previously perceived, is associated with other objects which have instead been perceived inthe past and are now being remembered.

Lit.: “made of previous subjects” ( p‹rvapram¢t’mayaÀ). It is not easy to derive the106

expected meaning from the very laconic transmitted text (see fn. 43 to the text). Memory isin fact a combination of a past object shared by past subjects, a past perception experiencedby the (presently) remembering subject and a present memory act concerning the synthesisof both. This is what also Abhinavagupta's comments would suggest: p. 25, ll. 11f. smara½etu p‹rvapram¢t„½¢¾ yo jñ¢n¢bh¢sas tasya smartuÀ p‹rv¢nubhavena saha m¤lito 'bh‹t, saeva id¢n¤m evonmi¼at¤ti vi¡e¼aÀ.

differentiation has not yet totally disappeared, due to his still residing in a con-dition where the mayic subjectivity is identified with the purya¼¿aka, etc.: to him,thanks to the power of a most intense application (abhiyogava¡¢t), a certain [past]object, being associated with other past objects currently being remembered,though not having been itself perceived previously (ap‹rvad’¼¿o 'pi), is manifestedthrough direct perception, since the yogin has experienced a disclosion of his ownessential nature with regard to some definite part [of the content of his cognition](t¢vaty a¾¡e). Or else, why not speak of a mere conceptual construct whatsoever105

concerning some past object, which arises at will in the subject without any contactwith a previous perception, as for example “my father was a king”? In these cases,the manifestation of the cognition with respect to an object which belongs itself tothe past and is associated with a past place etc. belongs to that [past] time, since itresides in subjects of that [past] time, while in the case of memory the cognition ismanifested [in the present] as related to previous subjects. Therefore, it is neces-106

sarily to be admitted that in memory the past manifestation emerges again. Andonce the memory has arisen thanks to an intense mental application depending onthe requirements of the subject, we may have the recollection of a single manifes-


The content of memory is necessarily a svalak¼a½a insofar as it is `localised´ in a cer-107

tain time and space. Even if the object of memory is a single ¢bh¢sa, i.e. a single universal,like `jar´ (in general), this is seen as a svalak¼a½a in that what would make it a s¢m¢nyaproper, namely its embracing a multiplicity of particulars and its permanence, are `contract-ed´ by the very fact that it is `remembered´ (p. 29, ll. 3!6 atra adhuneti g’h¤to ya ekaiko 'pi¢bh¢saÀ, sa vy¢pakatvanityatv¢tmas¢m¢nyalak¼a½atiraskar¢t sa¾kucitaÀ svalak¼a½am uc-yate atreda¾ sad iti). This however is not enough to make it sphu¿a in cases like the one atissue, where the object is a single ¢bh¢sa; its manifestation lacks `vividness´, Abhinava-gupta says, because, due to its not being combined with other ¢bh¢sas, it does not possess afull efficiency (p. 26, ll. 7f. ¢bh¢s¢ntarasa¾parkavirah¢t p‹r½¢rthakriy¢samp¢danavai-kalyena). This apparently conflicts with what Utpaladeva will say some lines later: even asingle s¢m¢nya may be sphu¿a, when it includes a multiplicity of s¢m¢nyas of a progres-sively broader extension. But, if I understand correctly, in memory we have a deliberate actof ìsolation´ of the single ¢bh¢sa depending on the subject's requirements.

The novelty of this criterion for establishing the distinction between s¢m¢nya and sva-108

lak¼a½a (p. 26, l. 14 na anyatra prasiddhaÀ) requires an adequate explanation by Utpala-deva.

The individual ¢bh¢sas do not possess a status of avacchedaka or avacchedya of their109

own, but they can assume either according to the circumstances. They are not associated )Abhinavagupta says (p. 26, ll. 15!18) ) remaining on the same plane, like the members ofa dvandva compound, but in a relationship of principal/secondary or qualifying/qualified.The ¢bh¢sa `cloth´, in a piece of white cloth, may appear predominant to a subject who, at

tation only, like `jar [in general]´ etc., even not combined with other manifestationslike `made of gold´ etc., depending on what may serve [on that occasion] the sub-ject's need; such single manifestation (`jar´, etc.) has the form of unique particular(svalak¼a½ar‹pasya) due to its being `coloured´ by the delimitation by a previousspace and time, etc. As a consequence, the manifestation of the object recollectedwill be a dim one (asphu¿aÀ). This is said [in the V’tti] with “at certain times,107

then [...]”. Then, in autonomous mental constructs, the manifestation of the object,taken solely in its universal form, without being `coloured´ by the delimitation bya previous space, time, etc., is [also] not vivid. In fact (hi) , [a group formed by]108

many universals, when combined among themselves and set in a relationship ofmutual delimitation, become vivid, and, in the end, once delimited by the universalof space etc., they attain the status of particular realities (svalak¼a½at¢¾ bhajante).Treeness (v’k¼atvam), which yet is one (ekam api) in its universal form consistingexclusively of being endowed with branches etc., once it is combined with a fullvariety of universals of colour and shape, such as those of the dhava or khadiratrees, and differentiated by myriads of combinations with the universals of themanifestions of space and time, becomes many (anek¢tmat¢m ¢padyate). Even ifin themselves they are exempt from spatio-temporal differentiations, universalsgive rise to a multiplicity of particular realities by virtue of their being differ-entiated from one another and of their mutually delimiting one another now in oneway, now in another. This mutual delimitation must comply with a criterion of109


that particular moment, is interested in the object as something to cover himself with; on thecontrary, the ¢bh¢sa `white´ will appear predominant, if he is thinking of the capacity thecolour white has to confer serenity to the soul (cf. Pram¢½av¢rttika I.58 and svav’tti).

It is again the Yogas‹tra definition of memory, which had been the starting point of110

Utpaladeva's discussion. The above considerations apply not only to the `dependent´ vikalpas, like ascertain-111

ment or memory, but also to those called àutonomous´ (svatantra), like imagination. The source of this quotation, according to Abhinavagupta (p. 30, ll. 19f.), are two112

commentaries on the ¹ivas‹tra, the Madhuv¢hin¤ and the Tattv¢rthacint¢ma½i, both ascrib-ed to Bha¿¿a Kalla¿a and now lost.

compatibility, and, therefore, cannot take place, for instance, in the case of theuniversals `cold´ and `fire´, which do not possess such compatibility. Then thereare certain universals which, even without entering into relations of delimitationwith other universals, are by themselves (svata eva) manifested in a vivid mannerwithout this entailing their assuming the status of particular realities, but onlythanks to the multiplicity of universals that they, in their unity, encompass. Forexample, this is the case of the universal `dhava´ with respect to the universal`tree´, since the former, though not being particularized as regards its own nature,contains within itself other universals such as `tree´, etc. But this does not meanthat it ceases to be an universal, because to become a particular the universal mustbe delimited by spatio-temporal differentiation.

When, however, depending on the subject's requirement, the object is remem-bered as associated with the manifestation of a previous space and time along withthe perception of many manifestations such as èxistent´ `golden´ `red´ etc., its ap-pearence [in memory] will be indeed vivid (sphu¿a eva). This is said [in the V’tti]with “at other times, instead [...]”. Sometimes, then, when one is intent on remem-bering something with an especially intense concentration, the object being re-membered, rendered clear and limpid by referring to it all its attributes ) excludingnone ) vividly appears before him, amounting in fact to something directly per-ceived (s¢k¼¢tk¢ramaya eva). This is said in the V’tti with “intensely [concen-trated], without interruption”. The absence of interruptions in the concentration isthe absence in it of the flowing of extraneous thoughts, its being pure. And in thiscase, although there is an immediate perception [i.e. a direct visualization] of theobject, the object comes to be immediately perceived [not as present but] as havingbeen experienced in the past. Thus, even more appropriately (sutar¢m), one is al-lowed to speak of non-extinction of the object perceived [in the past] (anubh‹ta-vi¼ay¢sampramo¼aÀ) : therefore, there is no incongruity in including this type of110

experience among the various forms of memory.Also when the autonomous mental construct turns to the manifestation of111

something that has been proved beyond doubt by various means of right know-ledge and is distinguished by many attributes, then the concerned object becomesindeed vivid (sphu¿¤bhavaty eva). This has been said: “Due to the fact of being112


The visualization of deities is a case of memory-like experience, in the sense that pre-113

vious cognitions gained through various means of knowledge, like revelation or inference,enable the devotee to `see´ or evoke a chosen deity so vividly as to entertain direct relationswith it, such as talking to it. But, as Abhinavagupta clarifies, this is a `novel´ manifestation(ªPV I, p. 126, navam eva ¢bh¢sanam), though depending on previous cognitive experi-ences.

The use of the causative shows that the shining of the object in memory is not to be114

viewed as a mechanical or autonomous fact but is to be inscribed within the wider contextrepresented by the free agency of an unitary, supreme consciousness. As Abhinavaguptaputs it (p. 31, ll. 17!19), it is only to the Consciousness-principle, manifesting itself as theinner self-consciousness of all cognitive acts, that pertains the power to freely obscure oneform and highlight another, to connect them or separate them. To express this in grammat-ical terms: the supreme I as the causal agent (hetu, prayojaka) is the ultimate responsible forthe shining of the object, `causes´ it to shine. Due to its complex nature (in fact, one caninduce a certain action in another only on condition that the induced subject has the intrinsiccapacity to perform such action), the model of the causative form is also applied to accountfor the paradoxical status of limited subjects in their delicate balance between autonomy andheteronomy with respect to supreme Consciousness (TORELLA 1987: 157f.).

intent in listening or speaking to Brahman or other deities”. We are not inter-113

ested here in settling whether this experience is to be considered a mental constructor not. On the other hand, at no moment may we speak of the absence of reflectiveawareness, because this is the very essence of all lights [of cognition]. And inevery cognitive experience, such as memory and so on, it is the freedom preciselyof this Consciousness principle to be at work: it is not simply produced thanks tothe power of its specific immediate cause. It is in order to show this that [in theS‹tra and the V’tti] causative forms are used, like “formerly made manifest”(p‹rv¢bh¢sitam), “makes it manifest” (¢bh¢sayati), “[of an object] made manifest[only in the past]” (prak¢¡itasya).114

Bibliography

Texts

Abhinavagupta, ª¡varapratyabhijñ¢vimar¡in¤, ed. by MUKUND R¨M SHASTRI, 2 vols., Bom-bay 1918!1921 (Kashmir Series of Texts and Studies 22, 33). = ªPV.

Abhinavagupta, ª¡varapratyabhijñ¢viv’tivimar¡in¤, ed. by MADHUSUDAN KAUL SHASTRI, 3vols., Bombay 1938!1943 (Kashmir Series of Texts and Studies 60, 62, 65). = ªPVV.

Bhart’hari, V¢kyapad¤ya with the Commentary of Hel¢r¢ja, K¢½Îa III, part 1, critically ed.by K.A.S. IYER, Poona 1963 (Deccan College Monograph Series 21).

Bhart’hari, V¢kyapad¤ya (m‹lak¢rik¢s), ed. by W. RAU, Wiesbaden 1977 (Abhandlungenfür die Kunde des Morgenlandes 42,4).


Bh¢skaraka½¿ha, Bh¢skar¤. A Commentary on the ª¡varapratyabhijñ¢vimar¡in¤ of Abhinava-gupta, 2 vols., ed. by K.A.S. IYER and K.CH. PANDEY, Allahabad 1938!1950 (ThePrincess of Wales Sarasvati Bhavana Texts 70, 83).

Dharmak¤rti, Pram¢½av¢rttika with the Commentary `V’tti' of Acharya Manorathanandin,critically ed. by SWAMI DWARIKADAS SHASTRI, Varanasi 1968.

Dharmak¤rti, Pram¢½av¢rttikam. The First Chapter with the Autocommentary, text and crit-ical notes by R. GNOLI, Roma 1960 (Serie Orientale Roma 23).

Paribh¢¼¢sa¾graha, ed. by K.V. ABHYANKAR, Poona 1967.Patañjali, Vy¢kara½a-Mah¢bh¢¼ya, ed. by F. KIELHORN, 3rd ed., rev. and furnished with

additional readings, references and select critical notes by K.V. ABHYANKAR, 3 vols.,Poona 1962!1972.

Patañjali, Vy¢kara½a-Mah¢bh¢¼ya with N¢gojibha¿¿a's Uddyota and Kaiya¿a's Prad¤pa[...], ed. with footnotes [...] by GURU PRASAD SHASTRI (rev. by BAL SHASTRI), 5vols., Varanasi 1987!1988.

Pra¡¢stap¢da, Pad¢rthadharmasa¾graha with ¹r¤dharabha¿¿a's Ny¢yakandal¤, ed. by DUR-G¨DHARA JH¨ ¹ARM¨, Varanasi 1977 (GaËg¢n¢thajh¢-Grantham¢l¢ 1).

Puru¼ottamadeva, Laghuparibh¢¼¢v’tti (see Paribh¢¼¢sa¾graha).Utpaladeva, ª¡varapratyabhijñ¢k¢rik¢ with svav’tti (see TORELLA 2002). = ªPK.Vai¡e¼ikas‹tra of Ka½¢da with the Commentary of Candr¢nanda, critically ed. by MUNI ¹Rª

JAMBUVIJAYA-JI. Baroda 1982.V¢mana-Jay¢ditya, K¢¡ik¢v’tti with Ny¢sa or Pañcik¢ commentary of Jinendrabuddhi and

Padamañjar¤ of Haradatta Mi¡ra, ed. by SW¨MI DW¨RIK¨ D¨S SHASTRI and K¨LIKAPRAS¨D SHUKLA, 6 vols., Varanasi 1983.

Translations and studies

LAZZERONI, R. (1997), “L'espressione dell'agente come categoria linguistica. I nomi indo-europei in -TÉR / -TOR.” In: BOLELLI, T.; SANI, S. (eds.), Scritti scelti di RomanoLazzeroni. Pisa, pp. 85!95 [originally published in: Studi e saggi linguistici 32(1992), pp. 233ff.].

SHARMA, R.N. (1995). The A¼¿¢dhy¢y¤ of P¢½ini, vol. 3: English Translation of Adhy¢yasTwo and Three [...], New Delhi.

TORELLA, R. (1987), “Examples of the Influence of Sanskrit Grammar on Indian Philo-sophy,” East and West 37, pp. 151!164.

TORELLA, R. (1988), “A Fragment of Utpaladeva's ª¡varapratyabhijñ¢-viv’ti,” East andWest 38, pp. 137!174.

TORELLA, R. (1992), “The Pratyabhijñ¢ and the Logical-epistemological School of Bud-dhism.” In: GOUDRIAAN, T. (ed.), Ritual and Speculation in Early Tantrism. Studiesin Honor of André Padoux. Albany (NY), pp. 327!345.

TORELLA, R. (2002), The ª¡varapratyabhijñ¢k¢rik¢ of Utpaladeva with the Author's V’tti.Critical Edition and Annotated Translation, 2nd ed., Delhi [1st ed., Roma 1994 (SerieOrientale Roma 71)].

TORELLA, R. (forthcoming), “Studies on the ª¡varapratyabhijñ¢-viv’ti. Part I. Apoha andanupalabdhi in a ¹aiva garb.” In: Halbfass Commemoration Volume. Vienna.

Documents

Studies on Utpaladeva's Ishvara-pratyabhijna-vivritti Part II-Torella