Perceptual Cognition: A Nyya-Kantian Approach

Author(s): Monima Chadha

Source: Philosophy East and West, Vol. 51, No. 2 (Apr., 2001), pp. 197-209
Published by: University of Hawai'i Press
Stable URL: http://www.jstor.org/stable/1400000
Accessed: 08/10/2009 18:38
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=uhp.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
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 support@jstor.org.

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

http://www.jstor.org

PERCEPTUAL COGNITION:
A NYAYA-KANTIAN APPROACH

MonimaChadha
Departmentof Philosophy,MonashUniversity

We findcertainthingsaboutseeing puzzling because we do not find the whole businessof seeing puzzling
enough.

Ludwig Wittgenstein'

Introduction
Experiencemust find a prominentplace in the philosopher's list of the "rich and
famous" entities of this millennium. Not just the more exotic experiences that the
more fortunateamong us have had the privilege to enjoy, but even the ordinary
perceptualexperiences of the world around us are rich and many splendored. It is
a fact that the human perceptualmechanism is such that it transformsthe meager
input, namely the stimulationof the senses, into a much richeroutput. Interestin
the philosophy of perception lies in understanding,as W.V.O. Quine puts it, in
what way one's theory of naturetranscendsany available evidence. The result is
an epistemologicalurgeto delineate the availableevidence-the pure given-from
the total content of a perceptualexperience that is enriched with concepts. This
preoccupationwith the pristinepurityof the given was shared by philosophersin
ancient India, and the traditioncontinues among contemporaryWestern philosophers in theirquest to identifythe nonconceptualcontent of perceptualexperiences.
Itis commonly believed thatthe given consistsof particularsthatare cognizedas
such in perceptualexperiences. Againstthis popularbelief, I shall arguethatthere is
no coherent notion of perceptualcognition of particulars.Perceptualcognition must
be restrictedto universal features. If particulars-as-suchdo not even qualify as
objects of perceptualcognition, then there is no possibilityof knowing particularsas-such in perceptualexperiences.Thus it seems to follow that perceptioncannot be
a source of knowledge of particulars.This claim is hardto digest. We have always
thought that perception is the primary,perhaps the only, source of knowledge of
particulars.We are forcedto conclude thatparticularscannot be objects of cognition
or knowledge. This conclusion is counterintuitive.I think that the conterintuitive
conclusion can be avoided by reconsideringour intuitivenotion of knowledge of
particulars.
The argumentin this essay drawson discussionsamong ancient Indianphilosophers, especially among Buddhistsand Naiyayikas,on the characteristiccontent of
perceptualexperiences. Althoughthere are considerabledisagreements,the ancient
Indianphilosophersagree on one crucial matter:perceptualexperiences are, at the
very least, cognitive in character.I think they are right.The contents of perceptual

Philosophy East & West Volume 51, Number 2 April 2001 197-209
? 2001 by University of Hawai'i Press

197

experiences are, at least potentially,objects of cognitionor awareness.Afterall, the

whole point of introducingthe notion of the content of perceptualexperiences is to
explain a perceiver'scapacityto gain knowledgeof perceivedobjects. This minimal
claim is beefed up by the Nyaya philosophersby addingthat all cognition, and thus
perceptualcognition, requiresconceptualization.This is the focus of the controversy
between Buddhistsand Naiyayikas.The Buddhist,in contrast,insiststhat perceptual
cognition mustbe totallydevoid of conceptualization.Perceptionis conception-free
This radicalthesis is based on a plausible intuiawarenessof a particular-as-such.2
tion: perceptualexperiences, unlike other cognitive episodes, must be constrained
by causal interactionbetween the sense faculty and the given. For the Buddhist,
conceptualizationrequires"imaginativeconstruction"by the mind that is unconstrainedby the given.
Inthe firstsection, we shall considerthe centralargumentsofferedby the Nyaya
againstsuch a conception-freeawarenessof a particular.Inthis section, I will extend
the Nyaya argumentsto revealthe incoherence in the very idea of a conception-free
awarenessof particulars.Thisdoes not mean that I rejectthe plausible intuitionthat
guides the Buddhist:the content of perceptualexperiences must be constrainedby
the causal interactionbetween the senses and the given. What I do reject is the
strongerclaim that the Buddhistdefends on this basis, namely that the content of
perceptualexperiences must,therefore,be restrictedto particulars.There is nothing
new to this response, it is just a restatementof the famous Kantiandictum:intuitions
withoutconcepts are blind.
In this essay I shall defend the simple Kantianthesis about the natureof perceptual cognition: perceptual experiences require cooperative activity between
sense-facultiesand the mind. Butthis simplicitycomes at the cost of vagueness. In
the second section, I shall put forth a precise claim that gives substance to the
Kantianthesis above: perceptual cognition requiresthe possibility of recognition.
The substantialthesis will providethe argumentfor the claim that perceptualcognition must be restrictedto universalfeatures.We will begin by unpackingthe notion
of "conceptualization"or "imaginativeconstruction"as involved in perceptual
experiences by buildingon some suggestiveremarksby the Naiyayikaphilosophers
and Kant.For them, conceptualizationor imaginativeconstructionis a necessary
ingredientof perceptualexperience, but it is not totally unconstrainedby the given.
As Kantwould say: concepts without intuitionsare empty.Thus the plausible intuitionguidingthe Buddhistcan be saved. The claim thatperceptualcognitionmustbe
restrictedto universalfeaturesis surprising,for it seems to rule out the possibilityof
knowledge of particulars.In the thirdand final section, I shall show that this counterintuitiveconsequence can be avoided by thinkingabout our intuitivenotion of
the knowledgeof particulars.
PerceptualCognitionof Particulars
We begin with the radicalposition, defended by Buddhistphilosophers,according
to which the content of perceptualexperiences is totallydevoid of any proliferation

198

PhilosophyEast& West

school hold that

of concepts. Buddhistphilosophers in the Dinnaga-DharmakTrti
is
free
from
construction
or
(kalpana).No seeing
imaginative
perception seeing-truly
of conceptual
involves
the
intervention
latter
since
the
is seeing-as,
necessarily
construction.The Buddhistclaims that a perceptualjudgmentor a verbalizationof
such a conception-freeexperience is, strictlyspeaking, impossible.3Forverbalization invariablyrequiresconceptualization, and that brings in more than what is
given in perceptual awareness. The given, for the Buddhist,consists of distinct,
unique, momentary,self-characterizedparticulars(svalaksanas).
The Buddhistargumentin supportof this radical thesis is based on a simple
intuition.Perceptionis cognitive awareness arisingfrom sense-object contact, and
since an object does not carrya concept or a label on its body, the sense faculty
cannot become aware of a concept as a resultof such causal contact. A sensory
awarenessthat arises from a color can representonly the color. It cannot represent
that color as accompanied by anothersensory particular,for example taste, texture,
and so forth. The Buddhistis suggestingthat the awareness of an object in one's
visual field-as a mango-cannot be called a properperception.Seeing-asrequires
imaginativemental construction.We have discretevisual awarenessesof color and
shape, but there is no mango to be seen. Perceptionresultsfrom sense-object contact and thus must depend solely on the causal powers of particulars.Words and
concepts are not ingrained in particulars,and thus perception can only result in
conception-freeawareness of particulars.If this restrictionis ignored it will lead to
absurdconsequences. Forthen perceptualawarenesscan representthingsthat are in
no way connected with the perceptual occasion, and the imaginativemind can
representanythingand everything.
The Naiyayikarejectsthe claim that constructiveawareness involvingconcepts
is not perceptual,since it is merely a figment of our imagination.Perceptionis a
cognitive episode triggeredby causal interactionbetween sense and object. This
interactiongives rise to a preliminaryawarenessof an unstructuredwhole. The nirvikalpaka,that is, conception-freeawareness, is a necessarystage in the process of
perception. But this unstructuredawareness is only a firststep; it is invariablyfollowed by a structuredawarenessas a necessaryconsequence in the same sequential
series. The firstawareness does not destroythe perceptualcharacterof the second;
rather,it facilitatesit. The nirvikalpakaor the sensory awareness in the immediately
preceding moment is an indispensablecausal factor for the generation of a savikalpaka or constructiveawareness, although memory, conception, and collateral
informationare also required.A cognition that is independent of the preliminary
sensory awareness cannot result in a perceptualjudgment. Forexample, the judgment "Thisis a collection of leptonsand quarks"that is made on seeing a table may
be legitimate(the collateral informationis supplied by physics), but it cannot be a
perceptualjudgment, since the necessary dependence on an appropriatesensory
awareness is missing.
Buddhists reject this analysis of the perceptual process. They argue that
the sequential process of perceptionas described by the Naiyayika,that is, senseobject contact resulting in unstructured,conception-free awareness, followed by

MonimaChadha

199

the structuredconceptual awareness, involves a mistakentransferof the propertyof

perception-hoodfromthe firstawarenessto the second. Udayana,one of the most
famous Navya-Nyaya(thatis, the new or laterNyaya school) philosophers,makesa
surprisingcounterclaimin response to this Buddhistargument.Perception-hoodis
really a characteristicof the second, structured,awareness; it is mistakenlytransferredto the first,unstructured,awareness.This bringsus to the crux of the matter.
In denying the perceptualcharacterof conception-freeawarenessthe Naiyayikais
reallyquestioningits cognitiveaspect. The use of the term "awareness"in this context confusesthe issue, for awarenessis necessarilya cognitive relation.To framethe
issue clearly, I will introducethe term"grasp"to signifya neutralrelationthat is not
necessarily cognitive in character.Following Matilal,we can say that the dispute
between the Naiyayikaand the Buddhistphilosopheris not about the occurrenceof
a pure sensorygraspof a particular,nor is the dispute about the physical concomitantof such a sensorygrasp.The issue is: can we regardthe conception-freesensory
graspof a particularas cognitive?
The Buddhistas well as the Naiyayikaregardsperceptionas a cognitiveepisode.
Rejectingthe cognitive characterof a sensorygraspof a particularis not a problem
for the Naiyayika,since it is merelyan intermediatestep in the process the resultof
which is a structuredperceptualawareness.The Naiyayikaholds that causal interaction between sense-facultiesand an object resultsin a sensory impressionthat is
no more than a mere physiologicalchange. The awarenessof the impressionarises
at the subsequentstage as a resultof the infiltrationof concepts, and only this second
awareness is, strictlyspeaking, a perceptual cognition. However, Buddhistsmust
show that the preliminarysensorygrasp is cognitive in character,since for them the
process of perceptionterminatesat this stage. Thatis, the Buddhistneeds a criterion
for distinguishinga concept-freesensory grasp of a particularfrom a mere physiological change. And the Buddhisthas a criterion:an essential markof awarenesshood is the presence of reflexive self-awareness.We are reflexivelyself-awareof
a conception-freesensory grasp.The intuitiveidea is not only that a sensory grasp
results in the occurrence of a physiological change, but in addition that such a
change is registeredin some partof the mind. The Buddhistis suggestingthat the
sensorygraspof particularsis a cognitive event by virtueof this registration,that is,
reflexiveself-awarenessof the sensorygrasp.
The response sounds plausible, but it falls short of hittingthe mark.Buddhists
cannot offera satisfactoryanswerto the question:what is it that we registeras a result of such a perceptualepisode?4To ensure the cognitive characterof such a sensory graspof a particularthey focus entirelyon the awarenessaspect of the sensory
grasp, at the cost of completely ignoringits intentionalaspect, that is, the objectdirectedness.Butthey do not deny the intentionalaspect of perceptualexperiences,
for perceptual cognition results in an awareness of distinct, unique, momentary,
particulars(svalakSanas).
The real problemfor the Buddhistis: in what sense are we aware of a unique,
distinct particularin a perceptualepisode? The Buddhistdisallows any differentiation or characterizationof particularsas a resultof perceptualcognition.We cannot

200

PhilosophyEast& West

characterizethe particularusing "this"or "it,"for even this minimalcharacterization

counts as misidentification.The Buddhisthas a reason for denying any kind of differentiationor identification,since it necessarily brings in concepts. Concepts are
requiredas a precondition not only for kind-identitybut also for object-identity.
Thereis no identificationin a sensorygrasp.Ifthe object-identificationis absent in a
sensorygrasp, it cannot arise magicallyin the reflexiveself-awarenessof the sensory
grasp. But, then, one may ask, how does the Buddhistaccount for the intentional
aspect, that is, the object-directedness,of perceptualepisodes?
As we said above, the Buddhistdoes not deny the object-directednessof perceptual episodes. In order to account for the intentionalaspect of perceptualepisodes we cannot demand that the cognizer must be in a position to offer necessary
and sufficientcriteriafor identificationof distinct, unique particularsgiven in such
experiences. That is asking for too much. The intentional aspect of perceptual
experiences can be accounted for in terms of the cognizer's acquisition of some
abilityto differentiatebetween distinctparticulars,for that provides us with a minimal basis for regardingthe given in such experiences as consisting of distinct and
unique particulars.Otherwise, it is impossibleto make sense of any talk about distinct and unique particulars.According to the Buddhist,there is no conceptual
interventionin perceptualcognition. As a resultthe cognizer loses out not only on
the abilityto identifya particulardatum(thegiven) as belongingto a certainkind(or
but also on the abilityto identifyit as a
instantiatinga concept or class-characteristic)
distinct,unique particularaltogether.The Buddhistanalysis of perceptualcognition
fails to meet the intuitive requirementthat accounts for the intentionalaspect of
perceptualcognition. To save the cognitive characterof a perceptualepisode the
Buddhistdelinks the awareness aspect from its intentionalaspect of the sensory
grasp. But this comes at a cost, for the Buddhistcannot account for the intentional
aspect of perceptual cognition. The sensory grasp together with its reflexive selfawareness vacuously resultsin an awareness, but it cannot result in an awareness
of a particular.Thus the Buddhistcannot claim that the sensory grasp results in a
perceptual cognition, for perceptual awareness, by his own lights, is essentially
an awareness of distinct, unique particulars.This shows that the very notion of a
"conception-freecognition of a particular"is incoherent.
Beforewe close this section, let us pause to note the statusof conception-free
awareness in Navya-Nyaya philosophy. Firstly,unlike the earlier Naiyayikas,the
later Naiyayikasallowed conception-loadedawareness even in the firstmoment of
sense-object contact, where the sense faculty and the mind cooperate immediately,
as in the case of habituatedand recurringperception.Forexample, on the way to my
office every morning,I am used to looking into my mailbox. I do not firstgraspsensory particularslike color, shape, and so forth;I instantaneouslysee a mailbox in
the very first moment of sense-object contact. In some cases a conception-loaded
awareness may be mediated by a conception-free grasp of a particular,but this
intermediategrasp is not logically requiredfor perceptual awareness, since it is
causally impotent.And this bringsus to the second point:the Navya-Nyayaposition
on the statusof the conception-freegraspof particularsis thatsuch a grasp is or may

MonimaChadha

201

be factuallypresent,but it is not an indispensableelement in perceptualawareness.

By rejectingthe necessityof such a conception-freegraspas a preliminarystep in the
process of perceptionthe Navya-Nyayasecures its position against the danger of
incoherence, which is lurkingin the notion of the conception-freegrasp of particulars.The firstpoint is importantbecause it drawsour attentionback to the Kantian
thesis, which is preemptedby Naiyayikaphilosophers:the mind and sense-faculties
cooperate (immediately)in the act of perception.In the next section, I hope to give
some substanceto this notion of cooperationbetween the mind and sense-faculties.
This, in turn,saves the plausibleintuitionthatguides the Buddhist;that is, perceptual
experiences must be constrainedby causal interactionbetween sense-facultiesand
real particulars.
PerceptualAwarenessof UniversalFeatures
There seems to be an intuitivecontrastbetween seeing and seeing as. Thinkof the
familiarexperience of meetingan old acquaintance,someone you have not seen for
years.You see the personclearly,but fail to recognize him. Suddenlyyou remember
him. You see the old face in the alteredone. Firstyou have the experienceof seeing,
and then the experience of seeing as. The numberof examples of this are numerous,
and the Buddhistwill be thrilledwith them. The firstexperience can be regardedas
that of seeing truly,whereas the second is adulteratedwith recognitionalabilities,in
which a real particularis identifiedas someone, and this invariablybringsin concepts. Butone may ask: is the firstexperience reallyconceptuallyinnocent?Can we
regardit as a case of pure seeing? No! To see why, contrastthis case with another
familiarvisual experience. Suppose you are taking a walk in a desert at dusk. You
are looking at a solitarypalm tree in the distance, but you see it as a pole. As you
walk towardit, you see it as you normallywould, as a tree. Inthis example,too, as in
the firstone, there is the experience of seeing a particularin a new aspect. Unlike
the firstcase, there is no sudden dawningof an aspect; ratherthere is only a gradual
change in aspect as you approachthe tree. Firstyou have the experience of seeing it
as a pole, and then the experience of seeing it as a tree. The second case bringsto
lightthe uncontroversialpointthatvisualexperiencescan presentdifferentaspects;a
particularobject can be seen now as one thing, now as another.Butset againstthe
firstcase, this shows that the sudden dawning of an aspect of a visually presented
particularmerelydramatizesfor us a featurethat is presentin perceptiongenerally,
namely seeing as. Inotherwords,the intuitivecontrastbetween seeing and seeing as
is a false one. All seeing is seeing as. This generalizationis in starkcontrastto the
Buddhistclaim: no seeing is seeing as.
In the previous section, we saw that such a conception-freesensory grasp (or
awareness)of a particularcannot be regardedas a perceptualcognition of a particular.Pureseeing does not resultin perceptualcognitionof particulars,for there is no
coherent notion of a conception-freesensory grasp of a particular.The following
question becomes pressing at this point: what exactly do we need to add to the
sensorygraspso thatit counts as a perceptualcognitionof a particular?No doubtthe

202

PhilosophyEast& West

perceptual experience of particularsis triggeredby a causal interactionbetween

sense-facultiesand particulars,butwe need to go beyond what is immediatelygiven,
that is, fleeting sensory impressions.Thus,we invokethe mind, as Hume would say,
"for the mind looks furtherthan what is immediatelygiven to it." The necessary
interventionof the mind for perceptualexperience forces us to compromiseon the
pristinepurityof seeing and replace it with the conceptually loaded notion of seeing
as. Thatis to say, perceptualcognition requiresthe co-activityof sense-facultiesand
the mind.
Discrete, fleeting sensory impressionscannot be regardedas perceptions. In
Kant'sterms, "A combinationof them [fleetingsensory impressions],such as they
cannot have in sense itself, is demanded [for perceptualcognition]" (A 120; emphasis mine). The interventionof the mind brings in concepts that are vehicles for
providing the synthesis demanded for perceptual cognition. Kant introduces the
notion of a concept as a principlefor unifyingor synthesizingdiscrete sensory impressions. Concepts have application in experience because they serve to link or
combine distinct fleeting sensory impressionsas differentperceptionsof the same
object. The point Kantwants to emphasize is that some concepts are necessaryfor
the very possibilityof perceptualcognition.5This idea of concepts as a principlefor
synthesis contains the germ for a precise formulationof the Kantianthesis that we
shall explore in the restof this section: cognition requiresthe possibilityof recognition. The Kantiannotion of a concept brings in the possibilityof recognition.This
thesis, in turn,providesan argumentforthe claim that perceptualawarenessmustbe
restrictedto universalfeatures.
This kind of reasoningwill be rejectedoutrightby the Buddhistfor two reasons.
First,the thesis that cognition requiresthe possibilityof recognitionis metaphysically
loaded, since it assumes that realityconsists of persistingobjects that endure over
time. Thismetaphysicalassumptionis questionedby the Buddhist,and this formsthe
basis for the outrightrejectionof the epistemologicalthesis. Accordingto the Buddhist, realityconsists of svalaksanas,which are unique, discrete, momentaryparticulars. I don't thinkwe need to get into a metaphysicalquibble, for the thesis above
does not requireany debatable metaphysicalassumption.The thesis, as we shall
make clear at the end of this section, is merely an epistemological thesis that is
neutral regardingany metaphysicalcommitment.And second, the Buddhistdoes
not want to allow minimal-most others would say inevitable-intervention from
the mind because once concepts and words are allowed, there is a tendency for
the obsessive proliferationof unnecessaryconcepts. The point is that as soon as we
allow the mind to intervene,it has the tendency to take over to the extent of completely ignoringthe input from the senses, and then we are forced into a position
where we can perceive anythingand everythingof which the mind can conceive.
The Buddhistis rightin warningus about this danger, but goes too far in restricting
the content of perceptualexperiences to the pristinepurityof the given: real particulars. As a result, the Buddhistfails to offer a satisfactoryaccount of perceptual
cognition of particulars.
In an attemptto offera satisfactoryaccount of perceptualcognition, I begin here

MonimaChadha

203

with enigmatic insights of the Navya-Nyaya philosophers.As we said above, in

denying the causal necessity of conception-freeawarenessof particularsin the process of perception,the Navya-Nyayaphilosophersoffer a novel insight about the
nature of some perceptualexperiences: there are instances where sense-faculties
and the mindcooperate immediatelyin the firstmomentof perception,resultingin a
conception-loadedawareness instantaneously,as in the case of habituatedand recurringperception,for example. There is anotherexample, offeredin the context of
the direct perceptionof simple universals,that is interestingand useful for our purposes here.6Gangesa'sactual example has been modifiedby Matilal;I will stick to
Matilal'smodificationwithout using his terminology(Matilal,Perception,pp. 347348).7 Suppose I am looking at a disc that is changing color. While I am looking at
the disc, it has just changed to blue. Furthersuppose that it is a particularblue tint,
the like of which I have never seen before. I see the particularblue tint as the relevant qualifieror the propertyof the disc. The point is thateven thougha particularthe color in this example-is presentedto the relevantsense-faculty,the awareness
is of a particularas a property-instance,for I see the particularas an instance of a
certaincolor propertyof the disc in frontof me.
In the very firstmoment of perceptionthere is immediatecooperationbetween
the senses and the mind, which resultsin an instantaneousabstractionof structure:
the awareness is that of a particularas a prototypicalinstanceof a simple universal
or property.The immediatecooperation of the mind is essential since what one
perceives or is aware of is beyond what is immediatelypresent. Thus the same
model can be extended to explain our perception of particularsthat exemplify
propertieswith which we have prioracquaintance.The mind relatesthe particular
presentedon a certainoccasion to otherpast and futuresensory impressions,and, in
the case of new properties,to the possibilityof futuresensory impressions.These
impressionsare not presenton the perceptualoccasion; therefore,they are not sensory impressions.Neitherare these totallyfictitious.These are real possibilitiesin the
sense that they may have been actualized in historyor, at the very least, have the
potentialto be actual. Letus call them images. Images,in this sense, are not purely
imaginaryor fictitious.8The introductionof this termbringsus to explain the notion
of imaginativeconstructionas involved in perceptualcognition. Forthis we shall
now turnto Kant.9
To set the stage for discussion, let us begin with a quote from Kant:"Psychologists have hithertofailed to realize that imaginationis a necessary ingredientof
perception itself" (A 120). Kant introduces the notion of "imagination" specifically

for the job of synthesizingthe distinctand fleeting sense impressionsinto coherent

perceptions.This synthesisserves a dual purposefor Kant,since it also shows that
concepts have applicationin experience. As we have alreadysaid, a transientand
fleeting sensory impressioncannot be regardedas a perception.Forit to count as a
perception it must be relatedto other past and futuresensory impressionsof other
Kantinvokesthe facultyof
particularsof the same kind (or of the same particular).10
The
at
this
(or
recognition)of an immediately
cognition
point.
imaginationexactly
a
occasion
on
requiresthat other past and/or future
perceptual
present particular

204

PhilosophyEast& West

images of particularsof the same kind are represented,or broughtalive in one's

imagination.In Kantianterminology,the facultyof imaginationprovidesa "schema"
for the concept so that it can be applied in experience. The activityof invokingother
real images and combining or linking such images with the immediatelypresent
sensory impressionis essential for perceptualcognition. This is what is meant by
"imaginativeconstruction,"which involves concepts as vehicles for bringingalive
other images of the same kind,some of which may even be of the same particular,if
the particularendures over time. The thought here is just anotherexpressionof the
idea floated by Gangesa: "seeing a particularas a property-instance."There is no
notion of cognizing a particular-as-such;one can only cognize universalfeatures
that are exemplified by particulars.These featuresare cognized by a subject in the
sense that the subject acquiresthe ability to recognize these featureson other perceptual occasions. On a perceptualoccasion we are confrontedwith a particular
datum, but what we registeris the possibilityof recognizingother data of the same
kind.
Thoughthe Naiyayikaphilosopherrestrictsthis claim to some perceptualoccasions, I think, following Kant,that we can extend this idea to all instances of perceptual cognition. Thisview is forced by the negativeconclusion of the last section:
there is no coherent notion of cognizing a particular-as-such.No doubt, perceptual
cognition is intentional;it is an awareness of something.The awareness is not of a
particular-as-such;ratherit is an awareness of a particularas a property-instance.
The very notion of experience or perceptualcognition is that of cognizing (or recognizing) the particulardatum as an instantiationof a propertyor a universal. In
otherwords, cognition requiresthe possibilityof recognition.Thus,we can conclude
that perceptualcognition must be restrictedto universalfeatures.The argumentin
this section makes it obvious thatthe plausibleintuitionguidingthe Buddhistcan be
saved. Perceptionis constrainedby the causal interactionbetween sense-faculties
and real particulars,since the direct perceptionof property-instancesis, at the very
least, occasioned by the presentationof a particulardatum, which instantiatesthe
propertyor the universal.
Some of my readersmay object that in discussing Kantand Naiyayikasin the
same vein I have conflated the distinctionbetween propertiesand concepts. This
objection is not justified.It is importantto note thatthe Naiyayikaphilosophers'talk
of cognition of particularsas property-instanceson perceptual occasions is just
another way of describingthe process of ascriptionof propertiesto particulars.In
Kantianterms, such ascriptionof propertiesis no more than the unifyingand synthesizingof data by applyingconcepts to generateperceptualexperience. Infact, the
Naiyayikaphilosophers'account of cognizing new propertieson perceptualoccasions providesa novel account of acquiringand simultaneouslydisplayingthe possession of the concept in its applicationto perceptualexperiences. Inthis sense, the
Nyaya-Kantianthesis is only an epistemologicalclaim. The argumentin the following paragraphwill reveal its metaphysicalneutrality.Thiswill also help in clarifying
the thesis further.
It may seem that the Buddhistworldview, accordingto which realityconsists of

MonimaChadha

205

unique, distinct,momentaryparticulars,conflicts with the substantialthesis, that is,

that cognition requiresthe possibilityof recognition.Ifthe particularexists only for a
moment, it cannot be recognized on a futureoccasion. The substantialthesis does
not requirethat particularsmust endure over a period of time. The requirementis
met, even if cognizer is only confrontedwith momentaryparticulars:the cognizer
sees the momentaryparticularas an instantiationof the some property.The Buddhist
will object to such a cognition of universalfeatureson a perceptualoccasion, for
each distinctparticularis a self-characterized(svalaksana),uniqueparticular,one of
its kind. Since realityconsists of distinct,unique, momentaryparticulars,there is no
futureoccasion on which the cognizer is presentedwith an identical sensory impression.This conflictswith the substantialthesis because it seems thatthe possibility of recognitionis ruledout. The seeming conflict arises because of the ambiguity
in the use of the term "possibility"that infects the substantialthesis. The thesis
claims that cognition is restrictedto universalfeatures that characterizethe particularsgiven in perceptualexperiences, in the sense that they constitutethe basis
for distinguishingamong distinctparticulars.As a resultof perceptualcognition (or
knowledge),the cognizer acquires an abilityto identifyor differentiatea particular
given on a perceptualoccasion. The acquisitionof this ability manifeststhe possibility of recognition. If the Buddhistmetaphysicalview is the true descriptionof
reality,then there is no futureoccasion on which the cognizer is presentedwith an
identical sensory impression.But that just means that the ability acquired by cognizers in perceptualexperiences is not exercised. The point remainsthat there can
be no cognitionof a particular-as-such;
cognition is of a particularas an instanceof
a property.
Knowledgeof Particulars
In the previous section, we concluded that perceptualcognition must be restricted
to universalfeatures.This conclusion seems to have counterintuitiveconsequences.
Ifparticularscannot even qualifyas objects of perceptualcognition,there is no possibility of knowing particularsin perceptualexperiences. Thus it seems to follow
that perceptioncannot be a source of knowledge of particulars.On the other hand,
however, it seems plausiblethat perceptionis the primary,perhapsthe only, source
of knowledge of particulars.This consequence is unacceptable and will force
some of my readersto reject, or at least reconsider,the conclusion of the previous
section. In this section, I will show that the counterintuitiveconsequence can be
avoided, if we pay some attentionto our commonsense notion of the knowledge
of a particular.
In the firstsection, we saw that the notion of a cognition of a particular-as-such
is incoherent.This resultsin a deeper problemaccordingto the Buddhistworldview.
In the absence of any identificationor differentiation,it does not make sense to talk
about the particulardatum. The Buddhisthas no basis for the uniqueness or distinctnessof real particulars,and is thus forced to give up the idea that distinct,real
particularsare given in perceptualcognition. The given in perceptualawareness is

206

PhilosophyEast& West

reducedto an indistinguishableheap or massof realitywhose constituentscannot be

distinguished.The Buddhistnotion of svalaksanasounds mysticalat the very least.
This suggests that our ordinarynotion of a particularcannot be that of a unique,
distinct,momentary,self-characterizedparticular-as-such.
The uniqueness of distinct particularsis a commonsense assumptionthat can
only be accepted as an article of faith. We do not-in fact, I believe, we cannothave a comprehensiveknowledge of whateverit is that accounts for the uniqueness
of distinct particulars.However, as we saw in the last section, the particularsare
characterizedby universal features, which constitute the basis for distinguishing
among distinctparticulars.In cognizing a particularas an instantiationof some specific property(or group of properties),we have some hold on featuresthat can be
offeredas groundsfor distinguishingthe particularin question. Butthese featurescan
never constitutenecessaryand/orsufficientgroundsto identifythe particular(thatis,
distinguishit from everythingelse). Fortunately,our intuitivenotion of the knowledge of particularsdoes not demand a comprehensivegrasp of featuresthat characterize it uniquely. We settle for much weaker criteriain which it suffices to distinguish the particularin question from other objects in a given situation. What
works in a given situationdepends on the context that is determinedby cognizers'
interests,attitudes,needs, and so on. The perceptualcognition, and thus knowledge
of particularsin this weaker sense, is possible in the Nyaya-Kantianframework.
Therefore,I think,the counterintuitiveconsequence thatwe can have no knowledge
of particularscan be avoided.

Notes
1 - LudwigWittgenstein,PhilosophicalInvestigations,p. 212.
2 - Inthe absence of an explicit definitionof a "concept,"I shall be workingwith a
minimalistclaim about concepts that is uncontroversial.Whatever else they
may be, concepts are, at least, vehicles to structurethe inputfromsense-object
contact. This working definition is neutral on the realist/nominalistdebate,
which is centralto the disagreementbetween Buddhistsand Naiyayikas.So I
shall take the libertyto use the notions of "conception-freeawareness"and
"unstructuredawareness"interchangeablyin this essay.
3 - It must be noted that the Indianphilosopherswere workingwith a theory of
language where the demarcationbetween words and concepts is very fine, if
not practicallynonexistent(Matilal,Perception,p. 310). This can be attributed
to the influenceof the Indiangrammarians,notablyBhartrhari.
4 - My positive account in the next section will propose an answer to this
question.
5 - Kantregardedsuch concepts as transcendentaland hoped to have identified
them. I don't agree with the Kantianthesis because it rests on a distinction

MonimaChadha

207

between transcendentalconcepts, pure concepts, and empirical concepts.

Such a distinctionis untenableand can be questionedon variousgrounds.For
the purposesof this essay, I will avoid that debate. I thinkwe can endorse the
broad Kantianclaim that some concepts or other are necessary for the very
possibilityof experience, withoutclaimingthat we can identifysuch concepts
at this stage.
6 - The Nyaya philosophersuphold the obvious distinctionbetween simple and
complex universalsin the hope that this offersus a criterionbetween real universals and mind-dependentones. I am skepticalof the tenabilityof any such
claim, in the same vein as in Kant'sdistinctionbetween transcendentaland
empiricalconcepts. I don't think any useful distinctioncan be made between
kindsof propertiesor universalsand concepts.
7 - Gangesa, Tattvacintamani,
Pratyaksa,vol. 1, p. 87.
8 - It is importantto note that the term "image" as introduced here is not in
the sense of fanciful imagery,which is usually associated with false beliefs,
mistakenmemory, dreams, delusions, and creative original insight.The term
"image" has another use also, in the sense of a mental image or picture in
one's mind. Forour purposeshere, we will restrictourselvesto the latteruse.
9 - It is importantto note that Kanttreats understandingand imaginationas two
distinct faculties. But the distinction is not importantfor our purposes here.
Bothof them can be treatedas mentalfacultiesand thus the distinction,if any,
is irrelevanthere.
10 - This parentheticalqualificationis importantto be fair to Kant,but it will be
dropped henceforth. Kant insists on other sensory impressionsof the same
particularbecause he is tryingto defend the commonsense assumptionthat
perception representsthe particularsas distinct and enduringobjects. As we
have already noted, this will beg the issue againstthe Buddhistphilosophers.
So we will circumventthis metaphysicaldebate by focusing on particularsas
exemplifiersof kinds.

References
Gangesa. Tattvacintamani.Edited, with Didhiti and Jagadisi of Jagadisa, by S.
Nyayaopadhyaya.Varanasi:Chowkambha,1906-1908.
Kant,Immanuel.The Critiqueof Pure Reason.Translatedby Norman KempSmith.
London:Macmillan,1958.
Matilal,BimalKrishna.Perception:An Essayon ClassicalIndianTheoriesof Knowledge. Oxford:ClarendonPress,1986.

208

PhilosophyEast& West

Russell,Bertrand."The Philosophyof LogicalAtomism."In BertrandRussell,Logic

and Knowledge:Essays 1901-1950, edited by RobertCharlesMarsh.London:
George Allen and Unwin, 1956.
Strawson,P. F. "Imaginationand Perception."In P. F. Strawson,Freedomand Resentment,and OtherEssays.London:Methuen, 1974.
Wittgenstein,Ludwig.PhilosophicalInvestigations.Translatedby G.E.M.Anscombe.
Oxford:Blackwell,1953.

MonimaChadha

209