Padmapāda's Illusion Argument Author(s): Stephen H. Phillips Source: Philosophy East and West, Vol. 37, No. 1 (Jan., 1987), pp. 3-23 Published by: University of Hawai'i Press Stable URL: http://www.jstor.org/stable/1399080 . Accessed: 06/08/2013 16:10

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

.

University of Hawai'i Press is collaborating with JSTOR to digitize, preserve and extend access to Philosophy East and West.

http://www.jstor.org

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions Stephen H. Phillips Padmapida's illusion argument

INTRODUCTION

Advaita 's traditionof reflectionon perceptualillusion arises out of a concernwith mukti, a "liberation"or "salvation" thatis conceivedas a radical changeof awareness,thus as a "mysticalexperience." Developing the teachings of the and the BhagavadgTta,Sankara (circa 700 A.D.) and his followerstry to articulatea worldview that would explainthe possibility of this mysticalstate and show it to be the summumbonum (paramapuruasrtha). It is problematicwhether any Advaita philosopher,even thegreat Sanikara himself, is a "mysticphilosopher," that is, one who attemptsto describehis own mystical experienceand to theorizetherefrom. But convincedthat a kindof experience, albeita mysticalexperience, is themost important matter in life,the Advaitins tryto formulatea comprehensivetheory of experience(anubhava) that would conformto theirsoteriological views. Above all, it is the topic of perceptual illusionthat forms the bridge between the Advaitins' soteriology on theone hand and theirtheory of experience on theother. The relationof the salvific experience to our everydayexperience is viewed as analogous to the relationbetween veridicaland delusivesense perception.1 Safikara,in the veryfirst sentence of his Brahmasutrabhdsya,2uses the term 'adhydsa',literally "superimposition" but also "false appearance," to capture the relationbetween the natural and mysticalstate of the self'strue experience on theone hand and worldlyexperience on theother.3 The Advaitinthen asks, "What is adhyasa?"and proceedsto givetwo characterizations(or definitions) of it. The second of these he defendswith referencesto everydayinstances of illusion. Thus in this openingpassage, and in pronouncedfashion in his Upanishadiccommentaries and his UpadesasdhasrTas well,Safikara articulates an understandingof "adhydsa"that is supposed to includeboth everydayand mysticalphenomena.4 Further, in this"adhydsa" sectionof his BSB, he men- tionssome alternativeviews of illusion,apparently for descriptive as opposed to polemicalpurposes: without disputing these views, he puts forthhis second characterizationand claims that it would be acceptable to all disputants.5 Therefore,with these characterizations,Safikara must be seen as launching Advaita "phenomenologyof perceptualillusion." However,it is only afterSafikara that Advaita phenomenologyand indeed criticalreflection overall become advanced. Safikarahimself is so much less concernedwith argument and justificationalissues than withelaborating the soteriologicalteachings of diverseUpanishadic textsthat many of his more properlyphilosophic views are too inchoateand embeddedin otherdiscussions to repayat all easilya scrutiny,6although many modern reconstructions of his positionshave appeared. His followersinherit the soteriologicaland textual StephenH. Phillipsis AssistantProfessor of Philosophyat theUniversity of Texas, Austin. AUTHOR'SNOTE: The author wishes to thankWilhelm Halbfass and RichardLarivierefor commenting on an earlierdraft. PhilosophyEast and West37, no. 1 (January1987). ( by theUniversity of Hawaii Press.All rightsreserved.

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 4 Phillips

concerns,but theyalso tryto be systematicand to refuteopponents on a whole rangeof issues. It is withSafikara's disciples, Padmapada, Suresvara, and Vacas- pati,and theirfollowers that the Advaita polemics and phenomenologicalanal- ysesbecome truly astute. Advaita theoryof knowledgeand relatedphenomenology become increas- inglybroad in scope as, throughthe years, there is a shiftin thefocus of Advaita reflection.In general,more and more attentionis paid to cosmologicaland worldly(vyavahdrika) issues, while Safikara's embedded epistemological posi- tionsare enormouslyfilled out. By thetime of the Veddntaparibhdsd,the famous seventeenth-centurytextbook of Advaita epistemology,the topic of perceptual illusionis takenup not so muchwithin a soteriologicaldiscussion as withina wide-rangingexplanation of "meansof knowledge"() in general.7The Veddntaparibhdsdincorporates centuries of reflection-muchof whichorigi- natesin rivalschools-on topicsranging from particulars and universalsto the foundationsof claims about the nonpresenceof somethingx. (Do I directly perceivethat thereis not an elephanthere?) Thus one should not thinkthat Advaitaepistemology is limitedto a concernwith perceptual illusion, particularly in thelater period. Yet in the Veddntaparibhdsdand in manyintermediate works, theearly phenomenology of perceptualillusion, and in particularthat contrib- uted by Padmapada, informscrucially the theory of knowledge,somewhat like theinfluence of Descartesand Hume in theWest.8 It clearlyforms the mainstay of thecentral Advaita positions. Padmapada is generallyadmitted to have been Safikara'syounger contempo- raryand his student.Along with Suresvara (who mostlikely was also a student of Safikara)and Vacaspati Misra (who was probablyat leasta generationlater), he is theoriginator of a distinctline of Advaita commentary. These threeare the foundersof the prasthdnatraya,"three lines of interpretation[of Safikara]," knownto studentsof Advaita. Padmapada launcheda traditionof commentary and understandingof Safikarato whichseveral prominent philosophers belong, includingPrakasatman (circa 975), the title of whose lucid (but often unfaithful- see note 12 following)commentary Vivarana has becomethe name by which this branchof Advaita is commonlyknown; Citsukha (circa 1275),the logician and polemicistwho purportsto refuteall the late schools; Madhava (circa 1350), who is bestknown for a wide-rangingsurvey of Indian schools;9 and Dharmara- jadhvarindra,the author of the Veddntaparibhdsdpreviously mentioned. There are also severallesser figures who belongto Padmapada's "line,"and thenumber of Sanskritcommentaries written on his workis large.'1 In thisarticle, I wishto examinePadmapada's thoughton illusion-with an eye to evaluatinghis metaphysics.Despite the sophisticationof Padmapada's analysis,the marriage between the metaphysics and theunderstanding of illusion is an unhappyone. Yet thispoint could be turnedaround: although, as I shall show, Padmapada is unsuccessfulin his attemptto show throughan illusion analogythe possibility(sambhava) of his viewsabout theAbsolute (),

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 5

his meticulousphenomenology of the mentallife has meritsof its own. I shall not slightit by jumpingtoo quicklyinto an examinationof the metaphysics. Only afterpresenting with some detail Padmapada's analysisof everydayillu- sionsdo I takeup themetaphysics of Advaitaand demonstratethe failure of his illusionanalogy. Anotheraim of thisarticle, overlapping the evaluative concern, is historical. I contend that Padmapada's view of illusion illumineshis metaphysicsof brahman,the "Absolute," and of brahman'srelation to , the "cosmic illusion."This metaphysicshas been misunderstoodby some scholarsof'." By elaboratingPadmapada's use of theillusion analogy, I intend to show preciselywhat his viewsare on thenature and onticstatus of "mdyd." Whileit is a matterof debate to whatdegree he departson thisscore from Safikara, his views on mdya are not the same as those of Prakasatman,his famous "follower"who livedabout twocenturies later.12 Also, Padmapada's insistence on defendingthe possibility(sambhava) of the truthsrevealed by sruti,"scrip- ture," along with the phenomenology,achieves a quantum leap in Advaitic philosophicprocedure. This typeof reasoning,in additionto the phenomeno- logicalarguments, represents a majordevelopment over Safikara. 3 This article is not directedprincipally to questionsof Padmapada exegesisor of his place in theevolution of Advaita,but it is intendedto investigatefurther what is hardly a fullyworked mine.14

THE DISTINCTION BETWEEN ILLUSION AND MEMORY

Padmapada's principalwork, the Paicapddika, is a commentaryon thatportion of Safikara'sBSB whichexplains the firstfour sutrasof the Brahmasutra.5 Presumably,the work was to take up the firstfive sections of Safikara'sBSB (that is, the entirefirst chapter and the firstsection of the second),but all we have is a commentaryon Safikara'streatment of siitras1.1.1-4. As noted,at the beginningof Safikara'sBSB the key topic is illusionas the appropriateanalogue for the Self's true experience in relationto one's experience of the everydayworld. Padmapada focusesfirst and at lengthon thispart of his teacher'streatise. We remarkedthat Sankara characterizesillusion in two ways. These are, first,"an appearance (avabhdsa) of somethingpreviously experienced-[inthis way] like memory-in somethingelse." 16 Thenafter relat- ingthree alternative views, he givesa second"minimalist" characterization that is presumedto be acceptableto all disputants(as wellas to conformto common usage): to wit,"the appearanceof one thingwith the properties of another."17 As I indicated,it seemsto me thatSafikara, because he givesthe second charac- terizationwithout a wordof disputefor the alternative positions, is notso much concernedwith the precisenature of illusionas withmaking a soteriological point.He wantsmerely to referto illusionsas thephenomena that are-within our everydayexperience-the appropriateanalogues for the relationof the worldto the singletranscendent Self. But Padmapada takesit upon himselfto

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 6 Phillips

show thateach of Safikara'scharacterizations is superiorto alternativeviews, and does not elaboratethe minimalist point. His concernis withdisagreements on illusionamong rival schools. Padmapada begins his discussionby distinguishingan illusorycognition (mithydjndna)from an instanceof memory.He has two goals in thisregard. First,he wantsto show in what sense the term'smrtiripa', "like memory,"is used by Sankara in his firstcharacterization. Second, he wantsto establish- againstan alternativeview of illusion("smrtipramosakhydtivdda," interpreted as "akhydtivdda"by some of thecommentators18)-the positions that illusion has a distinctphenomenological character and, most importantly, that cognition is unitary.According to thealternative view, illusion is predominantlymemory butalso involvestwo cognitions, a sensecognition as wellas a memorycognition, somehow blended together(because the "awarenessof remembering"is ob- scuredor "stolenaway"). The reasonthat Safikara uses theterm 'smrtirupa', "like memory," is, he says, to suggestthat there can be no illusoryappearance of somethingthat has not been previouslyexperienced, just as therecan be no memoryof something whereofone has had no immediateacquaintance. This is a viewthat Padmapada wholeheartedlyendorses. Sensation creates mental (or subliminal)impressions (called "samskdra"and "vdsand"), while both memoryand illusioninvolve, psychologicallyconsidered, the activity of thesesense-created impressions. The way in whichmental impressions inform a presentcognition is complex on Padmapada's view.He elaboratesthis portion of his theorywhile disputing thesmrtipramosakhydti view just mentioned.19 To translatethe text, the position to be refuted(the "purvapaksa")is: The cognitionof somethingx that occurs when the eye is in contactwith a differentthing y isjust memory,but there is an obscuration(pramosa- "stealing away") of theawareness that it is a remembering.Because of somedefect of the instruments[or causes]of cognition,i.e. thesenses, etc., there arises a memory- awarenessof a particularobject, while the capacity of the organto presentthe distinctthing with which it is in contactis vitiatedby the defect.In thisway, solelybecause of the defectof the sense-organ,there is a failureto note the distinctionbetween the memory and thesensation which arise together, and thus the erroris produced that thereis a single [cognition]that in fact has not occurred,as in thecase of two treesin thedistance [perceived as one].20

While accordingto this,the opponent'sposition an illusorycognition is pre- dominantlymemory, Padmapada believesthat an illusionhas littlein common withan instanceof memory except in one respect.This is that,in both,the object cognizedhas been experiencedpreviously. (This is whatPadmapada sees as the pointof Safikara'suse of'smrtirupa',as we noted.)He insiststhat as a presenta- tion(avabhdsa) an illusorycognition differs crucially from an instanceof mem- oryin itsclarity ("spastam") and immediacy("aparoksdrthcvabhdsa").21 Then in additionto thisphenomenological argument, Padmapada presentsa linguisticone. He pointsout thatif theobject of an illusorycognition were the

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 7

same as the object of an instanceof memory,then a person deluded by, for example,a piece of mother-of-pearlappearing as silverwould make thejudg- ment'That silver',not 'This silver',as actuallyoccurs. In ,the term '', "this," is used to referto an object presentedin immediateexperience, whilethe term'tat', "that," is reservedfor objects previously experienced and remembered.Thus thecommon usage tells against the opponent's position. This, to mymind, is Padmapada's bestpoint against the purvapaksa, although surely thephenomenological argument is also a good one. But thisis not all he says. He also attacksby asking,"What is an awareness of remembering(smaranabhimiana)?" According to the opponent,it is this awarenessof rememberingwhich, invariably present in an instanceof correct memory,is obscuredor suppressedby a sensedefect in an instanceof illusion. Padmapada avers that thereis no particularform in whichan awarenessof rememberingpresents itself and by whichit can be discerned.22He is firmly convincedthat cognitionis unitary.And I believe that it is, above all, this suppositionthat rules out forhim the possibility of any suchdistinct awareness of remembering.But he does seemto have somegood phenomenologicalpoints. First,we are not alwaysself-conscious about our rememberingwhen we recall somethingor other.This is particularlyevident in recallingthe meaningof words:in comprehendingthe meaning of a sentenceone mustrecall the meaning of individualwords, but one does not recall theirmeaning as remembered meaning.So thereis clearlyno "awarenessof remembering"in thisinstance.23 Second, memoryis invariably"intentional": one cannot engage in an act of rememberingwithout remembering some object or other. No awareness of rememberingoccurs entirely apart from the objective content of thememory.24 But Padmapada's eschewalof theidea of smarandbhimdnainvolves more than the denials thatthere is any rememberingthat is "nonintentional"and thatin all instancesthere is plainlya self-consciousnessabout theremembering. He says in a sweepingfashion that whereasadmittedly one sometimesdesignates an objectas rememberedby a certainusage-for example,"I remember(such and such)"-still eventhen, as always,no distinguishable"awareness of remember- ing" is involved.25Thus herehe appearsto be guidednot by phenomenological considerationsbut by a peculiaridea of theunity of cognition.Surely we can be awareof engaging in theact of remembering-insome cases withlittle attention to the objects remembered.(Imagine an amnesiac questionedby a physician: "Can you rememberwhat you ate forbreakfast? Can you picturethe entrance to the hospital?"The point in thesecases would be to focuson remembering itself,not its objects.) Some of Padmapada's argumentsin thispassage (Ppk 45-47) appear to be merehand-waving. In facthis eschewalof the notionof self-consciousremembering is one indicationthat his is a "one-dimensional phenomenalism"-a positionthe full implications of which I shallexplain in the last section. Now an everydaycognition is no simple unityaccording to Padmapada.

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 8 Phillips

Judgments(reflecting cognitions) do indeed,he says,involve mental impressions, because judgmentsare expressedin words whose meaningsare remembered. Further,in the inference'The mountainis fiery'(made by one who sees only that the mountainis smoky),mental impressions of past experiencesof fire conjoinedwith smoke clearly have a role.Also, in an instanceof recognition (for example,'This is thatDevadatta whomI metpreviously'), mental impressions are responsiblefor the "that" portion of the judgment. But in all thesecases there is onlya singlecognition in partinfluenced by memory(smrtigarbha), not two cognitions. The referencesto cases of inferenceand of recognitionare presumedto show the possibility(utpddana = sambhava)of a unitarycognition. Thus therealso couldbe a unitarycognition in thecase of an illusion:

... it is evidentthat an inferentialcognition [e.g., as expressedin thejudgment 'The mountainis fiery']arises froma [direct]cognition [i.e. sensation]of the inferentialmark [e.g. particular smoke] together with mental impressions, while an instanceof recognition[e.g. as expressedin the judgment'This is that Devadatta'] arisesfrom eye-contact together with mental impressions. In both thesecases as well [as in illusion],there is just one singlevalid cognitionthat is "impregnated"with memory (smrtigarbha). Without an arisingof mentalim- pressions,both [ofthese two typesof valid cognition]would not occur.... Nor is there,moreover, any [other]reason to believethat in thesecases we have two cognitions.26

The unityof inferentialcognitions and of thoseof recognitionreveals, Padma- pada argues,the possibility of single yet "memory-impregnated" cognitions. That this possibilityis actual in the case of illusion he presumablytakes to be demonstratedby the unitary judgment it provokes, for example, in 'idamrajatam', "This silver,"just as withinference and recognition,although he is not explicit here.27In anycase, a keyidea in thissection is theunity of consciousnessin the presentmoment. Later, we shall reviewthe role of the idea in Padmapada's metaphysics. Also importanthere-though fora differentline of themetaphysics-is that Padmapada sees theforce of his opponent'sposition as derivingfrom the need to explaincausally the contrastof illusionand veridicalperception. He wants to appropriatethat forceinto his own position.Note that accordingto the smrtipramosakhydtivddin,illusion comes about througha break in the causal chain characteristicof veridicalperception. Padmapada agrees.28He too em- bracesa causal theoryof perception(pratyaksa) in general,and as we shall see, muchof his explanation of individual instances of illusion relies on theidentifica- tion of causal factors.His disagreementwith the smrtipramosakhydtivddinis limitedto thecharacterization of themental life.

THE "OBJECT" OF AN ILLUSION

Padmapada, we have seen, in part uses a linguisticargument to refutethe smrtipramosakhydtivddin:we say 'idamrajatam.', "this silver," not 'tad rajatam',

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 9

"that silver,"when presentedwith what we take to be silver,whether the presentationbe illusoryor veridical.What then is thereference of the term 'idam', the "this" used as an ostension?Padmapada sees the thing(visaya = = alambana)29pointed to as thepresented "silver" supported by thereal mother- of-pearl(suktigatamithydrajata). Here he is opposed firstof all by "realists"who espouseanyathdkhydtivdda, "the view that illusion is theperception of something as otherwise(than the way it is in fact)," forexample, certain Naiyayikas or "Logicians." His attackagainst this view reveals the extent to whichPadmapada is a "phenomenalist."My readingis thathe is a phenomenaliston theissue of thecriterion of theultimately real. But althoughhis is a radicalphenomenalism on thatissue, it is hardlythoroughgoing or comprehensive:his ontologicalposi- tionis peculiarlyAdvaitic, having little in commonwith the Western subjectivist systems(such as Berkeley's)that espouse a phenomenalistposition on theissue of thecriterion of thereal. The opponent'sanyathdkhydtivdda is presented as a naiverealism, though one witha peculiarwrinkle. Padmapada's opponentclaims that the illusory cogni- tion not only presentssomething real, the somethingcan satisfydesire. In the stockexample, the deluded person on thebeach would be seeinga real piece of silverin themarketplace through some bizarreextension of hisnormal power of sight.(And he would not realizethat he has such capability-unless he werean astutephilosopher.) Padmapada pokes funat the view so rendered,except for its understanding of the natureof an "object" of a (perceptual)cognition (whether illusory or veridical).This is theidea thatsuch an "object" is presentedas somethingtoward whichpurposeful activity might be directed.The BuddhistDharmakirti and the Naiyayikasargue that "arthakriyd," "causal efficiency".and "capacityto fulfill an aim," is themark of a true"object" (or real thing,vastu) in thata real thing is capable of satisfyingdesire and thusis appropriatelysought in action.30In otherwords, according to Dharmakirtiand theothers, whom I see as "pragma- tists,"something is realjust in case it can satisfya desire.A mirageis not real just because it will not quench thirst.Nor is therea dangerthat one could be bittenby the snake of the "snake-rope"illusion. Thus the Indian pragmatists have a "holistic"notion of experience such that experience would involve activity in the world,and, more precisely,interaction with worldly objects.31 Padma- pada, on the otherhand, does not use the term'arthakriyd', and incorporates only as much of the theoryas suits his purposes: instead of 'arthakriyd'he uses the term'vyavahdrayogya', "suitable (as the object of) [illusory]worldly activity."32The term'vyavahdra', "worldly activity," is used by Advaitinsto designatethe illusoryworld of our everydayexperience in contrastto the transcendantreality of the Absolute, Brahman.Thus this usage shows that Padmapada does notbelieve activity has anythingto do withthe means whereby the real is to be discerned.Although the expressionechoes the pragmatists' position,for Padmapada it means only that a perceptualpresentation is a

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 10 Phillips

presentationof somethingthat appears as thoughit could be an object of purposefulactivity. But "purposefulactivity" is itselfillusory from the sublating perspectiveof thesingle Self. All worldlyactivity is mdya,"illusion," according to Padmapada, and is revealedto be so by the sublatingcognition that simul- taneouslyreveals brahman.33 Consonantly within the "vyavahdrika"perspec- tive,the means whereby an illusoryobject (of the "silver"and "snake" sort)is discernedas such is, he says, not interactionbut a sublationalcognition, for example,as expressedin thejudgment "This is mother-of-pearl."Thus thepoint is thataccording to Padmapada theillusory silver is presentedas an appropriate objectof desire. One does notfind out thatit is incapableof fulfillingdesire until one has a sublatingexperience that shows it to be unreal.Not the capacityof the object to fulfilldesire is thecriterion whereby veridical and illusoryexperi- encesare to be discriminated.Rather, only a furtherexperience can be thebasis forthat determination-at least with illusions of thesilver-mother-of-pearl sort (we shallsee thatPadmapada distinguishesbetween two broad typesof illusion and thatit is onlywith regard to one ofthese that the character of the immediate presentationis all-decisive).Padmapada goes no furtherin his embraceof the notionarthakriyd than to hold thatthe object (alambana) appears as an appro- priateobject of desire. This, in hisview, is onlya matterof phenomenal fact, with no special onticrelevance. The contrastof Padmapada's phenomenalismwith thepragmatism of thisgroup of opponentsis extremelyimportant, and we shall returnto it in evaluatinghis theoryin thelast section. Yet despite his stance on "arthakriyd,"Padmapada's phenomenalismis limited.The "silver-mother-of-pearl"(also "snake-rope")type of illusionis, he says, the appropriateanalogy for the metaphysicsof Brahman,but it differs cruciallyfrom a secondtype that he identifies.Four examplesof this second type of illusionare given:(1) the bittertaste of somethingsweet to a personwith hepatitis("diseased withbile"), (2) the sightof a double moon (dvicandra)by an astigmatic,(3) thered appearanceof a crystalbecause of theproximity of a redflower, and (4) thereflection of an objectin a mirror.These are presentations thatremain false, Padmapada pointsout, even afterone understandsthat they do not presenta truereality, unlike the case of the snake and the rope.34One can knowthat these presentations are false(mithya) while they remain immediate presentations. Padmapada does not give full details about how this is possible, beyond distinguishingbetween the two kinds of illusion.To be sure, he findsother pramdna-s,"means of knowledge,"and it is in particularthe pramdna inference (anumdna)that appears to be operativein this "counter-intuitional"discern- ment.Further, the inferenceshere are drawnbased on causal relations.35But "Whyare theremeans of knowledge in additionto thecharacter of experience?" and "How are thesepramdnia-s related one to another?"are questionsthat Padmapada does not directlyaddress, although, as I shall argue,on thelatter question he appears to have a prettydefinite view.36 There is a tensionin

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 11

Padmapada's position:sense perception (pratyaksa) provides a reasonto believe that,for example, the crystal is red,while inferential reasoning based on causal relationsurges that such an appearanceis false.Why should the character of an immediatepresentation be thecriterion of thereal in some instancesbut not in all? Padmapada's commentatorsgive more thoughtto the questionthan does Padmapada himself,but thegroundwork for a responseis indeedpresent in his distinctionof the two typesof illusion-along with his complex use of the distinctionin explicatingthe metaphysics of Brahman.37 The presentationsof thesecond type of illusion(the red crystal, and so forth) are relational(sopddhika); their very nature invites reflection on therelations of one thingto another.Indeed, experience itself teaches that sense presentations in generalare mediatedby complex causal factors.But all thiscausal interaction is conditionedby thepolemics of thefirst type of illusion,which is relationless (nirupddhika).38Recall thatPadmapada acceptsthat the object of a perception appears as an object of purposefulactivity, while viewing the entirerealm of activityas illusory.In parallel fashion,he sees objects as standingin a causal chainthat involves the sense organs while he also maintainsthat the entire nexus of causal relationsin negatedin theSelf's trueperception. The causal nexus is complexon Padmapada's account. The presenceof the object is a necessarybut not a sufficientcondition for the perceptionof it.39 Many contributingfactors are identified:a causal theoryof perception(em- bracedby manyof theIndian schoolswith variations) is by thetime of Padma- pada alreadyquite developed.The objectperceived, along withmany variables pertainingto distinctkinds of objects, light(or anothermedium), the sense organ, subliminalimpressions, the internalorgan (manas), and the self are all necessaryfactors. (We mightcall this Advaita psychologyof perception, exceptfor the factorsrelating to externalobjects such as the transparencyof a crystal.)40Through identifying the conditions(upadhi) that influencepercep- tion,Padmapada can agreewith many of therival thinkers that the reason that, forexample, a persondiseased with an excessof bile sees white things as yellowis a defectin the sense organ; likewisewith other illusions of the sopddhika, "relational,"type. But the entirecausal nexus is subjectto a sublationof the "snake-rope"(nirupddhika) sort.

THE "SUPPORT" AND "LOCUS" OF AN ILLUSION

One factorthat is commonto all illusions,whether relational or not,and indeed common to all cognitions,is the presenceof a real "locus" and "support," alambana (or, less ambiguously,"adhisthdna" and "asraya").41 This standsas an empiricalgeneralization, and Padmapada refutesa counterexamplethat purportsto showthat there are presentationsthat have no locus.42He also puts fortha linguisticargument in supportof theclaim: a judgmentthat expresses a sublationalcognition invariably makes reference,at least implicitly,to a locus wherethe change occurs.43

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 12 Phillips

Now in ordinaryexperience, the locus of theobject of a cognitionis either(a) external-a physical(jadaka) object-as exemplifiedin sensorycognitions, or (b) internal,as exemplifiedin dreams. Padmapada in discussingdreams (as examplesof illusion) says that he does notwant to quibbleover the terms 'antar' and 'bahir',"internal" and "external."Though he does viewthe self as internal, likedream objects and unlikephysical things, his chiefconcern is to defendthe positionthat-from the metaphysicalperspective-the presentations both of dreamsand of wakingexperiences have consciousness(caitanya) as theirsup- port(asraya): Even in thewaking state [as well as in dreaming]the immediacy [of conscious- ness-"aparoksatd"] thatinvolves [external] objects is not to be distinguished frominner immediate experience (anubhava)-[at least not]according to [any] cognitionthat is valid-for theyare presentedin thesame way. Thus an object evenin thewaking state is experiencedinvariably (eva) as intimatelyassociated withinner immediate experience. Otherwise the presentationof the material worldwould be impossible.As a pot thatis covered(avagunthta) by darkness is notpresented [to cognition]without the [counter-]covering (avagunthana) of the lightof a lamp,so it is here.44 From themetaphysical perspective, the Self, that is, Brahman,is thelocus and supportof mdyd the cosmic illusion, and thisreality (pdramdrthikatva) is thought to underlieevery worldly appearance. Although the externality of the "locus" in wakingexperience is a dimensionto be sublated,a real locus is said to remain evenafter the ultimate sublation. Padmapada's stresson an invariablepresence of an adhisthdnain all instancesof cognition,to includethe worstillusions, is marriedto his view of the "ground" of all phenomenaand their "materialcause."45 In thisway, Padmapada presentshis understandingof thecentral doctrine of Advaita,namely, the identity of theself with Brahman. And indeedthe famous atman-brahmanequation of theUpanishads appears to be understoodsimilarly throughoutAdvaita, that is, with the brahman-factorunderstood as the "ground" or "locus" of all phenomena.46The connotationsof the termsused in thepassage just translatedreinforce the stress on theidea thatall cognitions involvea locus: both darknessand lightare thoughtof as "coverings"(ava- gunthana).Everything "objective" except the locus and supportin all cognitions is similarlya matterof "covering"and can be removedin a sublationalexperi- enceof the snake-rope sort-so thatonly the bare "locus" and "innerimmediate experience"would remain.47 Padmapada does not presentan empiricalargument for the ultimate identity ofall objectswith a singlelocus-support-namely Brahman-and reliesinstead on sruti,"scripture." But he does arguedefensively, pointing out thatbecause dreamand externalobjects are presentedin the same way48there is no reason (pramana) to believethat external appearances have anythingother than the self for theirlocus-support.49 Thus his point is that this ultimateidentity is possible.50

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 13

A SUMMARY AND AN APPRAISAL

We have seen thatPadmapada distinguishestwo kindsof illusionand thatit is only the relationless("snake-rope") sort that is the properanalogue for the "relation"between Brahman and one's experienceof the everyday world. In this finalsection, I want to focuson the metaphysicsand in particularon Padma- pada's use of an illusionanalogy to explain the relation(or nonrelation)of Brahman,the Absolute,to mdya,the cosmic illusion.I shall also say a word about whatappears to be wrongwith his theory. The centralconception-what makes this "illusionist"world view click in Padmapada's mindand providesthe key to itsstructure-is, I believe,the notion thateverything except the self(or cognitionitself) is a candidatefor sublation (bddha) withinthe invariablepresentation of a "locus"-in otherwords, that onlythe self is nota candidatefor the sort of change in thephenomenal character of experiencethat occurs when one ceases to see the snake and sees insteadthe rope as it is.51The notionthat only (a) the selfand (b) the locus-supportof a cognition(the two are thoughtto be identical)are unsublatableis the linchpin ofthe system. To changemetaphors, it is thisidea thatundergirds the remarkable statementof Safikara'sthat perception and all theother pramdna-s, "means of knowledge,"including scripture (sdstra), are dependenton thepresence of avidyd, "nonawareness[of oneselfas Brahman]";52that is to say, theiroperation pre- supposes the conditionof maya. This statementis remarkable,and a fortiori Padmapada's embraceof it, because of Safikara'sconcern with giving reasons (thatis, "pramdna")for his views,a concernthat Padmapada expandsa thou- sandfold.How can a viewbe rationalthat purports to transcendthe canons of rationality-orcriteria for warranted belief-including, to be sure,scripture? Now thetopic of the foundations of the Advaita world view, whether scriptural revelation,reason, mystical experience, or somecombination of these,has been amplydiscussed in recent,and not so recent,scholarly literature, both in the Westand in India.53My intentionis notso muchto contributefurther reflection on thistopic, though, as I have said, I shallnot hold back myopinion about the meritsof Padmapada's theory,but to questionwhether the ideas about sublata- bilityreally show thepossibility of a singleSelf excluding the world, as Padma- pada claims. First let us summarizehis most abstractpositions. The Advaita doctrine championedby Padmapada is thateverything except the self-luminous character of the self,its intrinsicawareness of itselfas aware (svayamprakasamdna),is sublated-withinthe invariable "locus"-in theultimate "knowledge" (vidyd), even the phenomena of thoughtprocesses, emotions, and desires,even the 54 profoundmystical appearance of an Tsvara,"God." Unlike Vacaspati, who holdsthat Brahman always carries within its supreme self at leastthe "potencies" of worldforms,55 Padmapada holds-in consonancewith his idea of mukti- thatthe ontic status of theworld is simplythat it does not exist;it is onlymaya,

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 14 Phillips

like the illusorysnake.56 All appearancesof "otherness,"including that of an isvara,similarly are not veridical;there absolutely are no otherthings than Brahman,the Self. This radical "illusionism"is the distinguishingmark of Padmapada's interpretationof Safikara. I admit that he also gives,as Karl Potterpoints out,57 an interestingcos- mologyand theoryof how Brahmancomes to appear as maya,and thisinvolves botha notionof an i-svaraand theconcession that it is Brahmanitself that suffers ,"nescience." 58 Prakasatman takes this theory as thepoint of departure forhis own muchmore realistic view of worldappearance. But thepassage in the Pancapddikdpresenting these ideas is exceedinglybrief and, in effect,a digression.59Moreover, as Prakasatmanrightly explains, the key idea in this regardis thatBrahman can undergothe sort of changethat is typicalof illusory presentationswithout relinquishing its nativestate.60 The way thatthe passage (Ppk 98-99) connectswith the main line of thoughtreinforces this emphasis: near the end it is proclaimedthat "ego-sense" (, which is said to be foundationalfor much appearance) is illusion only, and an analogy to the illusionof a redcrystal is made. Now thered-crystal illusion is "relational,"and correspondinglyPadmapada holds thatego-sense is relatedto a more primal "nescience,"avidyd. The pointis thatwhat then follows is furtherdiscussion of relationlessillusion as theway to understandavidyd. Finally, the ontic status of avidyais declaredto be, as alreadynoted, illusion alone. Thismeans that it exists as long as one is caughtup in it. But withliberation, it disappears.61 The metaphysicsis thus a radical "illusionism,"and the quasi-theisticcos- mologybound up withthe notions of "jianasaki" and "kriyasakti,"Brahman's powersof knowledgeand (creative)action, would be betterattributed to Pra- kasatman,the preeminent "Vivartavadin."62 Let us returnnow to the questionof whetherPadmapada demonstratesthe possibilityof this"Brahman." The idea thatthe self and thelocus-support are unsublatablewhile everything else is-conceived on analogyto therelationless sortof everydayillusion-is supposed to securethe possibility(sambhava) of thesingle and absolutereality of Brahmanexcluding the world. Scripture (sruti) indicatesthat the possibilityis actual; or at least,this is theofficial position of bothSafikara and Padmapada. Both of theAdvaitins also appearto hold thata mysticalexperience (brahmasdksatkdra)-revealing brahman as senseexperience revealseveryday objects-is the theory'sconsummation, and even "confirma- tion"; thusunofficially and in a peculiarway bothappear to believethat there is mystic"evidence" for the theory.63 In anycase, thepossibility of thetruth of themetaphysics is to be securedby theideas about sublatability,and thetruth itselfby scripture or extraordinary experience. The priorquestion is thenwhether theseideas do theirjob. Now insofaras Padmapada is correctin understandingthe snake-ropetype of illusionto be "relationless,"he would be able, firstof all, to maintaina pure phenomenalismon thecriterion issue-at leastat itshighest level (pdramirthika-

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 15

sat)-and exclude causal factors.Without relations (other than the identity relationexemplified in theproposition that dtman is brahman),there can be no basis for the discernmentof causal factorsconditioning the veridicalityof an immediate(mystical) presentation, in theimagined experience of "mukti." Similarly,there would be no basis fora suppositionof "otherness,"since this would requirea presentationof thingsin a relationother than that of identity. Indeed,the claim about possibilityis verystrong. It is, again, thaton analogy to thetype of sublationthat occurs when the snake is seen to be not therope it was formerlytaken to be, it becomesimaginable-however remotely-that I am deludedright now and thatnothing that I take to be real on thebasis of my presentexperience (the chair here, the piece of paper, and so forth)is indeedreal, exceptmy own subjectivity.With all candidatesthat involve relations excluded, the only thingnot possibly sublated would be the subject's immediateand unreflectiveawareness of itself(svayamprakasamdna). The snake-ropetype of sublationis takento show thatit is possiblethat some futureexperience of my own could revealthe illusoriness of everythingin mypresent experience, except thatdimension which is theawareness's awareness of itselfincluding, to be sure, thepresentation of a real thingas its "support."64 But, let us ask, is therenot in the sublationalcognition revealing the rope a presentationof relationsamong things-betweenthe rope and our bodies,for example?The ropeis usefuland wouldnot bite were we to pickit up. Padmapada himselfresponds to a similarquestion.65 He tells us not to take the specific analogy too seriously.Of course, he has to have an analogy, otherwisehe would have no argumentand no demonstrationthat the identity of theself with Brahmanexcluding the world is possible.So he offersus anotheranalogy: space and thespace insidea pot. Thereare in realityno relations(other than identity) obtainingbetween universal space and the space thatwe arbitrarilydelimit as thatinside a pot.66This analogyis takento illustratethe idea thatany presump- tion,and perception,of thingsin relationcould be false(mithyd). And I must say thatI see no reason whywe should not agree. The chiefdifficulty I find is not thatall presentationof relationcould appear to be negatedexperientially, but thatthe austere self-experience that he imaginescould not. Let me explain. The polemicsof sublatabilityare veryprofound. In modernWestern thought, thishas been broughtout by Descartesand his thoughtexperiment of possible deceptionby a devil. He imagines,that is, deemsit possible,that a devilcould be deceivinghim with illusorysense presentations.(Nowadays philosophers bringout thesame idea withtalk of"experience tanks" or thelike: "Superduper neuropsychologistscould stimulateyour brain so thatyou would thinkand feel you were writinga greatnovel, or makinga friend,or readingan interesting book. All thetime you would be floatingin a tank,with electrodes attached to yourbrain." 67) Padmapada's illustrationsdo not appear to me to be as vividas Descartes',but thishardly matters. He has a similarargument. What presenta- tion of experiencecould not possiblybe sublated,as an illusoryappearance of

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 16 Phillips

silverby a veridicalappearance of mother-of-pearl,by somefurther experience? In respectto whatdimension of experienceis it impossibleto imaginethat one is deceived?Surely not anyparticular presentation of thingsin relation.Is it not imaginablethat therebe a cognitionthat involvesa self-awarenessbut not anythingexternal to oneself?(Yogic trance,or "deep sleep," mightbe an exampleof this.) As I see it, the problemis not withimagining an apparentdisappearance of theexternal world in a cognitionof "self-absorption,"but withthe notion that theself-absorption is unsublatable. One of Padmapada's greatthemes is, as we have seen,the unity of cognition. It is evidentin his notion of "memory-impregnated"cognitions, which we reviewed,and in manyother details of his theoryof experience.His claimis that thereis a possibilityof a cognitionthat is unitaryto theultimate degree. What would it be like?It would be the barestbones of a cognition.It would be an awarenessof itselfand a sense of its own realityas its own "locus" and "sup- port."I findall theseideas intelligibleand thisstate of "mukti"imaginable-so long as the conceptionis restrictedto the psychologicaland no notion of brahmanis implied.More about thisin a moment.But ifone could "pop" into thissupreme self-absorption, why couldn't one also pop out?There is no end to theregress of possibilitiesof experientialsublation. Padmapada takesthe unsublatabilityof the selfto show thepossibility of a trueultimacy of experience.But theidea showsonly the trap of the"solipsistic" ultimacy68of an extremeand "one-dimensional"phenomenalism. The meta- physicswould hoist itselfby its own petard: the verylogic of sublatability requiresthat in thenegation of worldappearance in a unitarycognition of self all means of knowledge(pramdna) would be inoperative.The pictureis thusa potenttrap. Does the objectionthat hinges on thesublatability of "self-absorption"pre- suppose a view of temporalrelations among cognitivemoments, in thatit is thoughtthat a furthersublational cognition couldfollow the salvific experience? Yes, it does. Well, the salvificexperience is presumedto transcendtemporal relations,and thusthe objection is notcogent. No, itis cogent.Every dimension of everydayexperience that is imaginableas sublatablein the supremeself- absorptionis also imaginableas again presentin some ensuingexperience- imaginablenot, to be sure,on thepart of the"liberated" person, but on all our parts.Neither Padmapada nor any of his commentators,so faras I am aware, considersthis point-absorbed as theyare in thepicture of self-absorption. But has not Padmapada identifiedfactors that are essentialto all cognition, namely,self-awareness and thepresentation of a locus,and wouldnot a cognition limitedto theseessentials thus indeed be a stop in theregress of sublatability? No. Oftenwe are absorbedin objectivepresentations, and have no awarenessof ourselves.Also, our everydayself-understanding is far richer than that which is reflectedin Padmapada's notionof intrinsicself-illumination. All thatPadma-

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 17

pada has shownis thatwe are potentiallyself-aware at anymoment, or perhaps thatthere is no possibilityof any othertype of awarenesswithout this potenti- alityof self-awareness(the centraldoctrine of Kant's Critiqueof Pure Reason, accordingto some interpreters).The selfmay be self-illuminating,svayampra- kasamdna.69And the potentialityof thisprofound intuition of selfmay be as constantas theAdvaitins think it is. But thereis nothingin theidea thatsuggests thatsuch an intuitioncould not be followedby a fullerexperience of self-such as we have of our own bodies and our "being-in-the-world."Thus thereis no reasonto suppose thata cognitionthat has onlyitself as its objectas well as its "locus" and "support"is not also a candidatefor sublation. This is thefunda- mentalfailure of thetheory.70 There is also an equally grievousinstance of incoherence,related to this phenomenologicalmisconception. Padmapada's notionof the unityof the self and Brahmanbelieved to be evidentin a mysticalexperience is unintelligible, that is, not imaginable,because Brahman is consideredthe adhisthdnaand asraya,"locus" and "support,"of all things.In otherwords, Padmapada cannot show thata cognitionof the selfand Brahmannegating the worldis possible because Brahmanis conceivedas the locus-supportof all worldlythings. For Brahmanto be presentedas identicalwith a state of self-absorptionwould requiresome kind of referenceto worldlythings, and in the state of muktias Padmapada conceivesit thereis none such. Thus we can constructa "pragma- tist" objection(invoking the traditionof Dharmakirtiand others)that would runas follows.With everyday illusions, even of theradical "relationless" sort- bestexemplified, I would say,not in a snake-ropesublation but in wakingfrom a vividdream-a judgmentof illusorinesscan be securedonly by an experience involvinginteraction or interrelation.Since thereis imaginedno interactionor interrelationin thestate of self-absorption,it is unimaginablethat Brahman, by definitionthe locus-supportof all things,could be revealedtherein. Thus Pad- mapada mustneeds fail in his attemptto show that a cognitionrevealing all thingsto be mdydyikacould occur.

NOTES

1. In developingan illusionistmetaphysics and soteriology,the Advaitinsdraw deep froma commonwell of Indian religiousconceptions. See theMajjhima-Nikdya, Sutta 26, foran examplein theBuddhist tradition-outside Vedantic "scriptures." Then foran earlyand particularlystriking examplewithin Vedanta, see BrhaddranyakaUpanisad 4.3.9-22. 2. Brahmasitrabhdsya(hereafter cited as BSB), in Brahmasftra-Sdhkarabhasyam,ed. J. L. Shastri(Delhi: Motilal Banarsidass,1980; originallypublished (2d ed.) Bombay: Nirnaya Sagar, 1909), 1.1.1,pp. 4-9. 3. The choice of theterm 'adhydsa', "superimposition," to designateillusions reflects gafikara's metaphysicalviews. Illusions are moreneutrally referred to as "viparyaya"and "bhrdnti."(Later, in fact(BSB 1.1.4,p. 69), Sankarauses the term 'bhrdnti': "sarpabhrdnti," "the illusion of the snake.") 4. E.g., UpadesasdhasrT2.2.51-55(in thetranslation by Sengaku Mayeda, A ThousandTeachings (Tokyo: Universityof Tokyo, 1979),pp. 235-236.)

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 18 Phillips

5. BSB 1.1.1:sarvathd api tu ... na vyabhicarati. 6. N. K. Devaraja, in An Introductionto Sahkara's Theoryof Knowledge(Delhi: Motilal Banar- sidass, 1962), is able to presentSafikara's views as an intricateepistemological system because he expoundsthe opinionsof a whole host of Safikara'sfollowers-while not always creditingtheir originalityand overvauntingthe achievements of Advaita's"founder." Some othersdo thisas well, apparentlyfollowing the traditionamong the classical thinkersof all schools of achronistically attributingsophisticated positions to the earliestdocuments, and in particularto the ","of each school. Of course, therehas appeared a considerableamount of excellentscholarship on Safikara,including in part,to be sure,this book of Devaraja's. 7. Because ofthe enormous popularity of the Veddntaparibhasd,itsauthor Dharmarfijidhvarin- dra mustbe viewedas thepreeminent systematizer of Advaitaepistemology, although it maybe, as SurendranathDasgupta claims in the "Forward" to Swami Madhavananda's edition(Howrah: RamakrishnaMission, 1972), thathe is heavilydependent upon Ramadvaya's VedantaKaumudi. Cf. Dasgupta's discussionin his A Historyof IndianPhilosophy, vol. 2 (Delhi: Motilal Banarsidass, 1975),pp. 204-214. Note thatin theintroductory verses Dharmarfjadhvarindra does say thatphilosophy should be concernedwith the paramapurusartha.He also elaborates the idea in his book's last chapter. Nevertheless,the book is dominatedby epistemological concerns, not soteriologicalones. 8. See, forexample, Veddntaparibhdsa 1.55. Note furtherwhat Karl Pottersays in the"Introduc- tion" to thevolume of The Encyclopediaof IndianPhilosophies (Delhi: Motilal Banarsidass,1981) devoted to (vol. 3): "It is Padmapada who pioneersthe epistemologywe ... associatewith Advaita. He providesthe bridge between Safikara and laterAdvaita, which is obsessed withepistemology as Safikarawas withthe contrast between knowledge and action" (p. 73). Also interestingis Paul Hacker's generalstatement on Padmapada's originality:"Im ganzenk6nnen wir sagen,dass Padmapada ein schlechterKommentator, dafiir aber ein um so selbstfndigererDenker ist" (Untersuchungeniiber Texte desfrtihen Advaitavdda: 1. Die SchiilerSahkaras (Wiesbaden: Aka- demieder Wissenschaftenund der Literaturin Mainz, Abhandlungender geistes- und sozialwissen- schaftlichenKlasse 1950,no. 26), p. 27). 9. Viz., Sarvadarsanasam.graha. Note thatMadhava's Vivaranaprameyasam.igrahahas been trans- lated by S. S. SuryanarayanaSastri and SaileswarSen, Andhra UniversitySeries, no. 24 (1941), and theyattribute it to "Bharatltirtha." 10. See E. P. Radhakrishnan,"The PanicapadikaLiterature," Poona Orientalist6 (1941-1942): 57-73. 11. E.g., ibid.,pp. 58-59. 12. Prakasatmanhas a "tiered" ontology,and views the sublation of world appearance in Brahman-experienceas showingthat the worldhas no absolute realitybut not-as Padmapada holds-as involvingits disappearance; see thefinal section of thisarticle. Note thatPrakasatman's Vivaranastands in a complexrelation to Padmapada's work,and itis farfrom a sentence-by-sentence gloss. 13. Doubtless,it is Safikarahimself who suggeststo Padmapfda the importanceof thisline of reasoning.He gives(BSB 1.1.1: p. 19) a counterexampleto theapparent general rule () that superimpositionoccurs only on perceptibleobjects: apratyakse pi hy dkdsebdlds talamalinattdy adhyasyanti.Sankara also twiceuses theexpression 'vyavahdrah sambhavati', "the worldlypractice [of ...] is possible [onlygiven ...]" (pp. 20-21). But not only is thisline of reasoningnot at all prominentwith Safikara-especially as comparedwith Padmapfda's use of it-he appears to be confusedabout itsnature. When he says'tasmdt', "therefore," after his statements about possibility (p. 21), he conflatesarguments about thepossible and theactual. 14. There have been, however,some notable effortsin Padmapada scholarship.Many of the generalhistorians of Indian philosophyhave givenserious attention to theAdvaitin. For example, SurendranathDasgupta claimsto base hisexposition of Saikara's viewsin volume1 ofA Historyof IndianPhilosophy (Cambridge: Cambridge University Press, 1969) on Padmapfda's and Prakfast- man's commentaries.Also, Karl Potteroften refers to Padmapada's viewsin his long introduction to the Advaita volumeof The Encyclopediaof IndianPhilosophies. He also providesthere a thirty- odd page summaryof Padmapada's major work, the Pancapddikd. The most significant contributionin Englishto scholarshipon Padmapada is D. Venkataramiah'stranslation of the

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 19

Pancapddikd,Gaekwad's OrientalSeries, vol. 107 (Baroda: Baroda OrientalInstitute, 1948), along withthe accompanying notes. Venkataramiah's notes often elucidate obscure passage, but sometimes theyreflect the views of later Advaitinswho are unacknowledged.They also suffer,as does the translation,from Venkataramiah's lack of commandof the technicalphilosophical vocabulary in English.(Potter's summary on thisscore represents a greatadvance.) But despitethese works, Padmapada has not receivedthe attention he deserves.As Karl Potter's bibliographyshows, Advaita scholarship has focusedprincipally on ganikara,with Padmapada enjoy- ingabout one percentof theamount of attentionthat his teacherhas received:The Encyclopediaof IndianPhilosophies, rev. ed. (Delhi: Motilal Banarsidass,1980), vol. 1, listsmore than four hundred secondaryworks on Safikara,while only seven pieces are listedfor Padmapada. Let me add thatthere is good German scholarshipon Padmapada, and on Prakasatman:most notably,Paul Hacker's Untersuchungeniiber Texte des friihenAdvaitavdda, pp. 1927-1933 and 2014-2061, and Vivarta (Wiesbaden:Akademie der Wissenschaftenund der Literaturin Mainz, Abhandlungender geistes- und sozialwissenschaftlichenKlasse 1953,no. 5), pp. 36-41; and Klaus Cammann,Das Systemdes Advaitanach der Lehre Prakdsatmans (Wiesbaden: Harrassowitz, 1965). 15. The Pancapadikd,ed. S. SriramaSastri and S. R. KrishnamurtiSastri, Madras Government OrientalSeries 155 (1958), part 1, hereaftercited as Ppk. 16. BSB 1.1.1: ko 'yamadhydso ... smrtirapahparatra pirvad.rstvabhdsah. 17. BSB 1.1.1: anyasyaanyadharmdvabhdsatdm. 18. Viz.,by Prakasatman in his Vivarana,which comprises part 2 ofMadras GovernmentOriental Series 155 (p. 129),and by Atmasvarupaand Vijfiiantman,MGOS 155,pt. 1, pp. 41 ff. 19. Padmapfda does not identifyhis opponentsby name. Venkataramiahattributes the view to the Mimam.sakaPrabhakara (p. 19). It seemslikely that Padmapada had a directfamiliarity with Prabhakara,although at least herehe does not assumeany specialunderstanding of theviews that he refutes.In fact,the three opposing views that he takesup arejust thosethat Safikara mentions in givinghis "minimalist"characterization. Prabhakara'sview, by theway, is, as noted,also referredto as a()khydtivdda-andrightly so, sinceas we shall see, a "failureto distinguish"memory from sensory cognitions is one element crucialto thetheory. 20. anyasamprayuktecakus.y anyavisayajiinamsmrtir eva, pramosas tu smarandbhimanasya. indriyddmndmijninakdranandm kenacid eva dosaviesesenakasyacid eva arthavisesasyasmrtisamudbo- dhahkriyate. samprayuktasya ca dosenavisesapratibhdsahetutvam karanasya vihanyate. tena darsana- smaranayohnirantar6tpannayoh karanadosad eva vivekdnavadhdranddduirasthayor iva vanaspatyoh anutpanneeva ekatvdvabhdseutpannabhramah (Ppk 42-43). 21. Ppk 51 and 55; cf. Ppk 40. 22. Ppk 46. 23. Ppk 45. 24. Ppk 45. 25. Ppk 47. 26. drsyate... lihgajinansamskdrayohlihgijcdn6tpddanam, pratyabhijnin6tpddanam ca aksasam - skdrayoh.ubhayatra api smrtigarbhamekam eva pramdnajnfinam.samskdrtinudbodhe tadabhdvat.... na punahjndnadvaye pramdn.am asti (Ppk 53). 27. But see Ppk 52. 28. Ppk 51 ffand manyother places. 29. See note 41 followingon theambiguity in theseterms. 30. Venkataramiahattributes this view to theNaiyayikas (p. 25), but it appearsto originatewith .See MasatoshiNagatomi in "Arthakriya,"Adyar Library Bulletin 31-32 (1967-1968): 52-72, who identifiesboth sensesof theterm in Dharmakirti'swriting. 31. A favorablecomparison to such twentieth-centuryWestern philoso/philosophers as John Deweyand W. V. Quine mightbe made in regardto theircriticism of the phenomenalism of Berkeley and Hume. 32. Ppk 48. 33. Althoughthere is a differencein connotation,the terms'mdyd' and 'vyavahdra'are used synonymouslyby Padmapada, as by mostAdvaitins. 34. Ppk 98-113.

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 20 Phillips

35. Padmapada manytimes says explicitly that causes maybe inferredfrom effects, e.g., in Ppk 51. Note thatit was again Dharmakirtiwho brought out to theIndian philosophic community the causal underpinningsof manyinferences. It is because fireis thecause of smoke,necessary to itspresence, thatone mayreason validly from the premise that there is smokeon themountain to theconclusion thatthere is firethere too. 36. Sarikarasays at BSB 1.3.20that perception (pratyaksa) is independentin itsoperation, while theother praman a-s, in particularinference (anumcna), are dependenton perception.Padmapada's commentarydoes notextend as faras thispassage, but there is no reasonto believethat he does not also see perceptionas havingthis privileged "foundational" status. Indeed, his phenomenalism with respectto theultimately real dovetailswith this position. But I have foundno explicitstatement in a passage thatis clearlynot a "purvapaksa." 37. Ppk 70-71 and 98 ff. 38. Padmapada does notuse theword 'upcdhi' here, but he does use 'upadhdna'(Ppk 102),a word thatderives from the same verbalroot, 'dha'. (The commentatorsprefer the former term, possibly throughthe influenceof the Naiyayikas,and interpretthe two termsas virtualsynonyms.) Also, and thisis the importantexegetical point, Padmapada clearlyexpresses the idea of conditioning relations.Often he does so throughforms of the verb'apeks', e.g., in Ppk 54-55. In thispassage (Ppk 54-55) and laterin expositingthe metaphysics of the self (Ppk 112-113),he pellucidlyexpresses theidea of"relationless"sublation. (Note thatSafikara himself uses the term 'nirupddhi' in thissense, e.g., in Mundak6panisadbhdsya2.2.4.) 39. Ppk 54 (reading'dloka' forthe misprint'aloka': when "light" is absentone cannot see an objectthat is present). 40. Padmapada's cleareststatement of themechanics of perception occurs at Ppk 114-117. Here we maynote that Sengaku Mayeda's paper,"The AdvaitaTheory of Perception," Wiener Zeitschrift fiir die Kunde Siidasiensund Archivfir IndischePhilosophie 12-13 (1968-1969): 221-240, is an excellentreconstruction of the Advaita psychology(as opposed to epistemologyand phenomen- ology)of perception,although he too concentrateson Safikara'sviews somewhat at theexpense of themore developed theory in laterAdvaita. 41. Ppk 52: mithydlambanamjidnam. mithydjinam, "An illusorycognition is a cognitionthat has an illusoryobject." Thus one mightthink that the term'alambana' invariablymeans "presented object," and would referto that whichthe presentation(avabhdsa or prakdsa) would be of. But Padmapada sometimesuses theterm to mean "locus" and "support"instead. Two otherterms used regularly,'visaya' and 'artha',have potentiallythe same ambiguity. 42. Ppk 63. 43. Ppk64-65. 44. jdgarane pipramdnajnidndd antaraparoksdnubhavdt na visayasthd aparoksatd bhidyate, ekaru- paprakdsantt.ato 'ntaraparoksdnubhavdvagunthitaevajdgarane py arthonubhfiyate. anyathdjadasya prakdsdnupapatteh.yathd tamasd 'vagunthitoghatah pradlpaprabhdvagunthanam antarena na pra- kdisbhavati,evam (Ppk 57-58). 45. Doubtless,some of thisemphasis is due to Padmapada's desireto refutethe Madhyamika Buddhists.These so-called"nihilists" hold thatthere is no ultimatesupport for appearances: the self is reallysunya, "nothing." To contradictthis view Safikarauses the term'paratra', "in something else," in characterizingillusion, Padmapada says. The firsttime he stressesthe necessityof an adhisthdnais in thiscontext (Ppk 61-68), but he stressesit in thelater talk about Brahmanas well. See Ppk 139ff. 46. The identitystatement would not be analyticor epistemicallynecessary, but syntheticand metaphysicallynecessary-to use the termselucidated by Saul Kripke,in Namingand Necessity (Cambridge,Massachusetts: Harvard, 1980), Lecture 1. Cf. Karl Potter'sdiscussion of theidentity statement,in Encyclopediaof IndianPhilosophies, vol. 3, p. 73, and WilhelmHalbfass, Studies in Kumdrilaand Sahkara (Rienbek:Wezler, 1983), pp. 54-60. 47. Sometimesit seemsto me thatthis is a metaphysicsof thelocative case. Note thatSafikara's use of'paratra' in his firstcharacterization of adhydsais brilliantlysuggestive: the "something else" whichthe illusory object appears in is-from themetaphysical perspective-Brahman, and 'paratra', likeall formsof 'para', has theconnotation of "supreme,highest, preeminent" as wellas "remote" and "beyond."

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 21

48. But are they?Dreams do not exhibitthe same degreeof continuityas wakingexperiences. Moreover,the only way we know (or could know) that dream objects have the self as their locus-supportis withreference to wakingexperiences. The factthat Padmapada missesthese points is anotherindication of his radical(and misguided,as I shall show) phenomenalismon theissue of thecriterion of theultimately real. 49. Ppk 57. 50. We can understandnow why Padmapada feelsno need to use differentterms for (a) the presentationalobject of a cognition(e.g., silver)and (b) thelocus and supportof a cognition(e.g., mother-of-pearl):these are ultimatelyidentical. We remarkedthat the terms 'alambana', 'artha', and 'visaya' are all used in bothsenses-see note41 preceding. 51. Sankara expressesthe idea of the unsublatabilityof the self,although not so clearlyas Padmapada, in BSB 1.1.1: aparoksatvdcca pratyagdtmaprasiddheh;and (muchmore explicitly) in Aitarey6panisadbhd.sya,introduction (Ten PrincipalUpanishads with Sdhkarabhdsya (Delhi: Motilal Banarsidass,1978), p. 325): utpannasyaca brahmitmavijynanasyaabddhyamdnatvdt. It is also implicit in the Brhaddran.yakaUpanisad, 4.3.23-31. Padmapada presentsthe idea manytimes, e.g., in Ppk 52: na svatahjidnasya mithydtvamasti, bddhtbhdvdt. See also in particularPpk 65-66. 52. BSB 1.1.1: avidydvadvisaydn.yeva pratyaksddinipramdni sastrdni ca. See also BSB 4.1.3: pratyaks.dyabhavesruter apy 'bhdvaprasahgaiti cet; na, istatvdt. 53. Ram Mohan Roy'schampioning of the rationality of afikara'sVedanta in theearly nineteenth centurylaunched a seriesof similardefenses in India. See WilhelmHalbfass, Indien und Europa (Basel: Schwabe, 1981),chap. 12,and his laterextensive discussion of theentire issue of reasonand revelationin thephilosophy of Safikara,Studies in Kumdrilaand Sahkara,pp. 27-84. 54. Ppk 286-288. 55. Bhdmati1.3.30, in Brahmasutra-Sdhkarabhds.yam,pp. 261-264. Note thatSrirama Sastri in his Prefaceto vol. 155 of the Madras GovernmentOriental Series (in whichthe Ppk appears along with seven commentaries),lists ten principaldifferences between Padmapada's and Vacaspati's philosophies,but he neglectsthis one (pp. xvii-xix).And it shouldcount, I would say,at leastin the top three. 56. Thereare ample indicationsof bothviews in Safikara'sworks. The reasonthat Padmapada's teachermakes cosmological statements that spark Vacaspati's interpretation is that he is committed to an idea ofinvariable perfection for sruti from a world-boundperspective. Of course, he nonetheless believesthat sruti,like everything that is inessentialto theself, belongs to theprovince of mdyd.See BSB 1.3.30. 57. Encyclopediaof Indian Philosophies, vol. 3, p. 86. 58. Ppk 98-99; and 74: anddisiddhdvidydvacchinnanantajTvanirbhdsdspadamekarasam . 59. Ppk 98-99. 60. Vivarana,pp. 653-657. This passage is Prakasatman'scommentary on Padmapada's under- standingof Suitra1.1.2 ("[Brahman is that]from which the beginning, etc. of this[world proceeds]") expressedat Ppk 300: "visva vivartate."Jose Pereira has translatedthis section in Hindu Theology. A Reader(New York: Doubleday, 1976),pp. 201-206. Let me repeatthat the idea of "transmogrification"(Pereira's coinage for 'vivarta')-contrasting with"transfiguration" (parindma)-is not pronouncedin the Ppk. However,that Padmapada is thinkingof Brahmanas not losingits nativestate when he says here(Ppk 300) thatit is the "root cause" is likely:"yadavastambho visvo vivartate prapancah tad eva mflakdranambrahma"-"having thisas itssupport all worldappearance unrolls, this alone, the root cause, is Brahman."Nevertheless, thekey ideas are "avastambha"and "mulakdrana,"not that expressed by the verb 'vivartate' whose grammaticalsubject, moreover, is not "brahma'but 'prapanca',"world appearance." (Cf. Ppk 56: ... aparoksacaitanyasthdvidyasaktiralambanatayd vivartate, "the powerof 'nescience'stationed in immediateawareness 'transmogrifies' as objects.") Let me add thatI findPrakasatman's ideas hereof particularinterest, since it is on the issue of whetherBrahman "transfigures"and/or "transmogrifies" that Aurobindo,that premiermystic Brahmavadinof our century,takes himself to departfrom Safikara. (I do not believehowever that Aurobindohad a directfamiliarity with Prakgastman's work.) For Aurobindo'stheory of Brahman's real"evolution" (and "involution")in worldforms, see TheLife Divine (Pondicherry: Ashram Trust, 1973), pp. 295-321, 439-481, and 834-835. Cf. my "Aurobindo's Concept of

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 22 Phillips

Supermind,"International Philosophical Quarterly 25, no. 4 (December 1985): 410-416, and my Aurobindo'sPhilosophy of Brahman(Leiden: Brill,1986), pp. 124-131. 61. Ppk 66-67. An indicationthat Padmapada does not take his own "cosmology"seriously is thatin thispassage (Ppk 66) he says thatthe occurrenceof the cosmicillusion is "inexplicable," anirvacanTya. 62. Cf. Klaus Cammann'sdiscussion, in Das System,pp. 124-128. 63. I discussthe implicit "mystic empiricism" of Safikaraand various"folk Vedfntins" in "Is Sri Aurobindo's PhilosophyVedanta?" Adyar LibraryBulletin 48 (1984): 1-27, and in my book, Aurobindo'sPhilosophy, pp. 67-73. AnantanandRambachan has attemptedto depreciate"experience" (anubhava) in Safikara'ssoteri- ologyand correspondinglyto appreciate"scripture" or "revelation"(sruti). See "Safikara'sRationale forSruti as the DefinitiveSource of Brahmajfiana:A Refutationof some ContemporaryViews," PhilosophyEast and West36, no. 1 (January1986): 25-40. Now srutiis indeedaccording to Safikara an indispensablemeans both (1) to have a rightintellectual understanding of the selfand (2) to achievean existentialknowledge (vidyd) of brahman.But it is importantto distinguishthese and to see thatthe value of theintellectual knowledge is thoughtof as onlyinstrumental, as derivingfrom its role in the attainmentof brahmasdksdtkdra,"immediate brahman-experience." The intellectual understandingbrought by "scripture"does not,furthermore, appear to be thesole possessorof this instrumentalvalue. Contraryto what Rambachan says, thereoften appear to be yogic "prere- quisites"(adhikdra) as well,although Safikara is notalways consistent about these(see e.g., Upadesa- sdhasrf2.1.2, Mun.dak6panisadbhs.ya3.1.1, 3.1.2, 3.2.4, Ken6panisadbhdsya4.7, and Taittirfyo- panisadbhdsya3.1.1). Whateverbe the preciselist of adhikara-s,understanding the words of the Upanishadsis clearlyheld to be, thoughnecessary, not sufficient. From Safikara's Mundak6panisad- bhdsya1.1.5: vedasabdenatu sarvatrasabdardsir vivaksitah. .abdardsyadhigame 'pi yatndntaram an- tarenagurvabhigamanddilaksanam vairagyam ca na aksarddhigamahsambhavati. But beyondthis confusionof necessaryand sufficientconditions, Rambachan's worsterror is, I repeat,that he conflatesthe ideas ofintellectual and existentialunderstanding and failsto see thataccording to sruti as interpretedby Safikarathe liberating "knowledge" of Brahmanis a "psychologicalevent" (anu- bhava),not an "intellectualcomprehension," (sabdddhigama). See in particularBSB 1.1.2:srutyddayo 'nubhavddayasca yathdsam.bhavamiha pramdnam,anubhavdvasdnatvdd bhiutavastuvisayatvdc ca bra- hmajnfnasya.(The most significantuse of 'anubhava'here is not the first,on whichRambachan focuses,in pp. 35-36, but the second.) Rambachan says that experienceis not a praman.afor knowingbrahman. In theintellectualist sense of "knowing,"he is right.All pramdna give intellectual knowledge,pramd, and thereis, accordingto Safikara,no intellectualknowledge in theliberating experience.There can be no reportof theexperience of brahmansince the liberated person is lostto the world (or more properly,the world is lost to him). Thus not any "mysticreport" but, as Rambachan partlysees, only self-authenticatingsruti teaches the essentialnature of the selfand mukti.But it is an austere"self-experience" that is most highlyvalued, not sruti.Moreover, this "experience"is conceptualizedin such a way thatthe world, including sruti, has to be mdyd.(The inferenceis drawnaccording to theunderstanding of everyday illusion.) See theBSB citationsin note 52 preceding,and Safikara'sBrhaddran.yak6panisadbhdsya 2.1.20: ekasminbrahmani nirupddhike na upadeso na upadestana ca upadesagrahanaphalam.tasmdd upanisaddm ca anarthakyamiti etad upagatameva. The officialepistemology is surelyanti-empiricist, even anti-mystic-empiricist,but not onlydoes Safikarabelieve that a livingmukti is a real possibilityfor a personand fromthat perspective"virtual evidence"; the logic of his metaphysicsis governedby a complexanalogy to everydayexperience, veridical and illusory.See theBSB citationin the followingnote (note 64) as anotherexpression of thisgoverning analogy. 64. Brahmanthus includes both intrinsicself-awareness and all reality-in thatit is thatwhich underliesall worldlythings. This is an extremelyimportant point. The reason that therecan be salvificBrahman-experience is that Brahmanis real, the Advaitinshold, just as the knowledge mediatedby the senses is dependenton realthings. Safikara says this explicitly at Brahmasitrabhdsya 1.1.4: brahmavidyd... pratyaksddipramdnavisayavastujidnavadvastutantrd, "Knowledge (vidyd) of brahmanis dependenton a real thing,like the knowledgeof the real thingsthat are the objectsof suchmeans of knowledgeas senseexperience." 65. Ppk 111-113.

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions 23

66. This appears to be the same typeof phenomenonas thatbrought out by the "duck-rabbit" patternpopularized by Wittgenstein,in PhilosophicalInvestigations (New York: Macmillan,1953), p. 194. 67. RobertNozick, Anarchy,State, and Utopia(New York: Basic Books, 1974),p. 42. 68. Cf. Paul Hacker, "Die Idee der Person im Denken von Vedanta-Philosophen,"in Kleine Schriften(Wiesbaden: F. Steiner,1978), pp. 280-281. 69. The idea is thatwe can be aware of ourselvesdirectly, without mentally reflecting on thefact of our existence;the selfis "irreflexivelyself-illuminating." Compare M. Merleau-Ponty'sunder- standingof the Cartesian"cogito" in Phenomenologyof Perception,trans. Colin Smith(London: Routledge& Keagan Paul: 1962),pp. 369 ff.Paul Hacker illuminesthe Vedantic idea in numerous articles;see forexample "Safikara der Yogin und Sankarader Advaitin" and "ganikara'sConception of Man," republishedin his KleineSchriften. Cf. Aurobindo'snotion "knowledge by identity,"in The LifeDivine, pp. 524-552, and mybook, Aurobindo'sPhilosophy, pp. 103, 117,and 124. 70. The failureis particularlyevident in the obscurityof thenotion of "jTvanmukti,""liberation whilealive." Cf. Potter'sdiscussion of thenotion, in Encyclopediaof Indian Philosophies, vol. 3, pp. 34-35: "The jivanmuktistate seems paradoxical" (p. 34). Indeed,it is, if"living" is takento involve consciousand willfulactivity and notjust a comatosestate.

This content downloaded from 130.65.109.155 on Tue, 6 Aug 2013 16:10:28 PM All use subject to JSTOR Terms and Conditions