Vous êtes sur la page 1sur 26

Wittgenstein and Anthropology

Author(s): Veena Das

Source: Annual Review of Anthropology, Vol. 27 (1998), pp. 171-195
Published by: Annual Reviews
Stable URL: http://www.jstor.org/stable/223368
Accessed: 20/12/2009 05:31

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

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.

Annual Reviews is collaborating with JSTOR to digitize, preserve and extend access to Annual Review of

Annu.Rev. Anthropol.1998. 27:171-95
Copyright? 1998 by AnnualReviews.All rights reserved

Veena Das
Departmentof Anthropology,GraduateFaculty,New School for Social Research,
New York, New York 10003, and Departmentof Sociology, University of Delhi,
Delhi 110007, India;e-mail cdedse@del2.vsnl.net.in



This essay explores the theme of Wittgenstein as a philosopher of culture.

The primarytext on which the essay is based is Philosophical Investigations;
it treats Stanley Cavell's work as a majorguide for the understandingandre-
ception of Wittgensteininto anthropology.Some Wittgensteinianthemes ex-
plored in the essay are the idea of cultureas capability,horizontaland verti-
cal limits to forms of life, concepts of everyday life in the face of skepticism,
and the complexity of the inner in relation to questions of belief and pain.
While an attempthas been made to relate these ideas to ethnographicde-
scriptions,the emphasis in this essay is on the question of how anthropology
may receive Wittgenstein.


I wish to invite reflection in this paper on a certain kinship in the questions that
Wittgenstein asks of his philosophy and the puzzles of anthropology. Consider
his formulation-"A philosophical problem has the form: 'I don't know my
way about"' (Wittgenstein 1953:para. 123). For Wittgenstein, then, philo-
sophical problems have their beginnings in the feeling of being lost and in an
unfamiliar place, and philosophical answers are in the nature of finding one's
way back. This image of turning back, of finding not as moving forward as to-
ward a goal but as being led back, is pervasive in the later writings of Wittgen-
stein. How can anthropology receive this way of philosophizing? Is there

0084-6570/98/1015-0171 $08.00
172 DAS

something familiarin the feeling of being lost in anthropologicalexperience?

Wittgenstein'sfear, "theseed I am likely to sow is a certainjargon"(Diamond
1976:293), is to be respectedso thatthe translationof his ideas into anthropol-
ogy should not be taken as the opportunityfor merely a new set of terms. In-
stead of renderinga systematic account of any one aspect of his philosophy, I
shall try to follow a few lines of thought that might interest anthropologists,
hoping to convey the tones and sounds ofWittgenstein's words. My thoughtis
not thatthis will help us reachnew goals but thatit might help us stop for a mo-
ment: to introducea hesitancy in the way in which we habituallydwell among
our concepts of culture,of everyday life, or of the inner. In this effort I am in-
debtedto the work of StanleyCavell, whose thoughtson several of these ques-
tions have acted like signposts in my own efforts to move withinPhilosophical


In his recent, passionate work on the "anthropography"of violence, Daniel
(1997) is moved to say, "Anthropologyhas had an answer to the question,
Whatis a humanbeing? An answerthathas, on the whole, servedus well, with
or without borrowingsfrom philosophers.The answer keeps returningto one
form or anotherof the concept of culture:humanshave it; other living beings
do not"(p. 194). He goes on to discuss how Tylor's (1974) foundingdefinition
of culturehelped to move it away from the "clutchesof literature,philosophy,
classical music, and the fine arts-in otherwords, from the conceit of the Hu-
manities"(Daniel 1997:194). Let us consider for a moment the actual defini-
tion proposed by Tylor: "Cultureor civilization taken in its widest ethno-
graphic sense, is that complex whole which includes knowledge, belief, art,
morals,law, custom, andany othercapabilitiesandhabitsacquiredby man as a
memberof society" (Tylor 1974:1). Whatis interestingin this definition is not
only the all-inclusive natureof culturebut also the referenceto it as capability
and habit acquiredby man as a memberof society. As Asad (1990) has noted,
this notion of culturewith its enumerationof capabilitiesand habits,as well as
the focus on learning,gave way in time to the idea of cultureas text "thatis as
somethingresemblingan inscribedtext"(p. 171). Withinthis dominantnotion
of cultureas text, the process of learningcame to be seen as shapingthe indi-
vidual body as a pictureof this text, inscribingmemory often throughpainful
rituals so that the society and culture of which the individual is a member is
made present,so to say, on the surfaceof the body (Clasteres 1974, Das 1995a,
Durkheim 1976). The scene of instructionin Wittgenstein (1953) is entirely

Scenes of Instruction
Philosophical Investigations begins with an evocation of the words of
Augustine in Confessions. This opening scene has been the object of varying
interpretations.The passage reads as follows:
When they (my elders) named some object, and accordinglymoved towards
something, I saw this and grasped that the thing they called was the sound
they utteredwhen they meant to point it out. Their intention was shewn by
theirbodily movements, as it were the naturallanguageof all peoples: the ex-
pression of the face, the play of the eyes, the movement of otherpartsof the
body, and the tone of voice which expresses our state of mind in seeking,
having, or avoiding something. Thus as I heard words repeatedly used in
their proper places in various sentences, I gradually learnt to understand
what objects they signified; and after I had trainedmy mouth to form these
signs, I used them to express my own desires. (Wittgenstein 1953:para.1)

Stanley Cavell (1982, 1990), who has given the most sustainedreading of
this passage, senses here the presence of the child who moves invisible among
his or her elders and who must divine speech for himself or herself, trainingthe
mouth to form signs so thathe or she may use these signs to express his or her
own desires. Now contrastthis scene of instructionwith the famous builders'
scene, which follows soon after in Wittgenstein(1953:para.2):
Let us imagine a language for which the descriptiongiven by Augustine is
right.The language is meantto serve for communicationbetween a builderA
and an assistant B. A is building with building stones: there are blocks, pil-
lars, slabs, and beams. B has to pass the stones, and that in orderin which A
needs them. For this purpose they use a language consisting of the words
"block","pillar","slab","beam".A calls them out; B brings the stone which
he has learntto bring at such-and-sucha call.-Conceive this as a complete
primitive language.
If we transposethe scene of instructionin which the child moves among the
adultswith that of the builders,we might see that even if the child were to use
only four words, these may be uttered with charm, curiosity, a sense of
achievement.The child has a futurein language.The builders' languageis, in a
way, closed. Wittgensteinwills us to conceive of this as a "completeprimitive
language."Yet as Cavell (1995) points out, thereis no standinglanguagegame
for imaginingwhat Wittgensteinasks us to imagine here. It has been noted of-
ten enough that Wittgensteindoes not call upon any of the naturallanguages
from which he could have taken his examples: Thus his game in this sec-
tion-whether with referenceto the child or the dreamlikesequence by which
one might arriveat an "understanding"of what the words five red apples mean
or with reference to the builders' language-is in the nature of a fiction
throughwhich his thoughts may be maintainedin the region of the primitive.
But the "primitive"here is conceived as the builders' tribe,which seems bereft
174 DAS

of the possession of its cultureor of an undoubtedsharedlanguage-the lan-

guage the tribe uses is invented language, not to be confused either with the
naturallanguagesfound among people who maintainfull forms of sociality or
with the language of the child.
Wittgenstein's sense of the child who moves aboutin his or her cultureun-
seen by the elders andwho has to inherithis or her cultureas if by theft appears
to find resonancein the anthropologicalliteraturein the registerof the mytho-
logical [for instance, in the bird nester myths analyzed by Levi-Strauss
(1969)]. Despite the studies on socialization, rarely has the question of how
one comes to a sense of a sharedcultureas well as one's own voice in thatcul-
ture in the context of everyday life been addressedanthropologically.If asked
at all, this questionhas been formulatedas a questionof socializationas obedi-
ence to a set of normativerules and procedures.Butjuxtaposingthe child with
the builders seems to suggest that whatever else it may be, the inheritanceof
cultureis not aboutinheritinga certainset of rules or a certaincapacityto obey
orders.As Wittgenstein(1953:para.3) says, "Augustinedoes describe a sys-
tem of communication:only not everythingwe call language is this system."
And then, as if the surest route to understandthis concept is to understandit
throughthe eyes of the child, he points out thatthe words in a gamelike ring-a-
ring-a-roses are to be understoodas both the words and the actions in which
they are woven (Wittgenstein 1953:para.7).
Concernwith childhood in early anthropologicalliteraturehas not been ab-
sent but has been expressedthroughthe intricaciesof age ranking,rites of pas-
sage, attitudestoward someone called "the average child," and the construc-
tion of "childhood"in a given society. Both Nieuwenhuys (1996) and Rey-
nolds (1995) have recently shown how sparsethe ethnographicdescriptionsof
childrenand their agency have been. Reynolds's (1995) work on political ac-
tivism of childrenand youth in the volatile and traumaticcontext of South Af-
rica is special because she shows how tales of folk heroes might have provided
a perspectiveto young people with which to view their defiance of the regime
of apartheideven as they had to negotiate questions of obedience, authority,
and kinship solidaritywithin the domains of family and kinship. I would also
draw attention to the remarkableaccount by Gilsenan (1996) and to Das
(1990b,c) and Chatterji& Mehta(1995) on the complicatedquestionof what it
is for childrento inheritthe obligation to exact vengeance, to settle for peace,
or to bearwitness in a feud or in the aftermathof a riot. Claims over inheritance
arenot straightforwardin these contexts, but even in relatively stable societies,
anthropologicaldescriptionsof cultureas either sharedor contested have ex-
cluded the voice of the child. As in Augustine's passage, the child seems to
move aboutunseen by its elders.
Let me go on to the questionthatthe figure of the child raises here:Whatis
it to say that the child has a futurein language?

There are several scenes of instructionin Philosophical Investigations-

those pertainingto completing a mathematicalseries, those pertainingto read-
ing, those pertainingto obeying an order.All raise the issue of what it is to be
able to projecta conceptor a word or a procedureinto new situations."A"writes
down a series of numbers;"B" watches him and tries to find a law for the se-
quence of numbers.If he succeeds, he exclaims, "Now I can go on." Whathas
One powerful way of understandingwhat gives a child the confidence to
say "I can go on" is providedby Kripke(1982) with the example of what it is to
follow a mathematicalprocedure or a rule. He points out that Wittgenstein
shows convincingly thatwe cannotspeakof an innerunderstandinghaving oc-
curred;nor can we say thatthereare some basic rules thatcan tell us how to in-
terpretthe other rules. Here is how the problem appearsto Kripke(1982:17):
Hereof courseI amexpoundingWittgenstein's well knownremarksabouta
"rulefor interpretinga rule".It is temptingto answerthe skepticfromap-
pealingfromone ruleto anothermore'basic'rule.Butthe skepticalmove
can be repeatedat the morebasic level also. Eventuallythe processmust
stop-"justificationscometo anendsomewhere"-andI amleftwitha rule
whichis completelyunreduced to anyother.HowcanIjustifymypresentap-
plicationof sucha rule,whena skepticcouldeasilyinterpret it so as to yield
anyof anindefinitenumberof otherresults?It seemsmy applicationof it is
anunjustifiedstabin the dark.I applytheruleblindly.
Without going into this argumentin any detail, I want to comment on one
formulationthatis proposedby Kripke(1982):, thatourjustificationfor saying
thata child has learnedhow to follow a rule comes fromthe confidence thatbe-
ing a member of a community allows the individualperson to act "unhesitat-
ingly but blindly." Kripke (1982) gives the example of a small child learning
additionand says that it is obvious that his teacherwill not acceptjust any re-
sponse from the child. So what does one mean when one says that the teacher
judges that, for certaincases, the pupil must give the "right"answer?"I mean
thatthe teacherjudges thatthe child has given the same answerthathe himself
would have given.... I mean thathe judges thatthe child is applyingthe same
procedurehe himself would have applied"(Kripke 1982:90).
For Kripke(1982) this appealto communityand to criteriaof agreementis
presentedin Wittgensteinas a solution to the "skepticalparadox"-that if ev-
erythingcan be made out to be in accordwith a rule, then it can also be made to
conflict with it. But this skepticism with regardto justification, says Kripke
(1982), applies to the isolated individual:It does not hold for one who can ap-
ply unhesitatinglybut blindly a rule that the communitylicenses him or her to
apply.As with applicationof a word in futurecontexts, thereis no "innerstate"
called "understanding"that has occurred. Instead, as he says, there are lan-
guage games in our lives that license under certain conditions assertions that
176 DAS

someone means such and such and that his present applicationaccords with
what was said in the past.
My discomfortwith this descriptionarises from the centralitythat Kripke
(1982) places on the notion of rule as well as from the processes he privileges
for bringingthe child in agreementwith a particularform of life thatwould li-
cense such blind and unhesitatingobedience to the rule.
If we take the teacherin Kripke(1982) to be the representativeof the com-
munity within which the child is being initiated, then I am compelled to ask
whetherthe "agreement"in a formof life thatmakes the communitya commu-
nity of consentcanbe purelya matterof makingthe child arriveat the same con-
clusion or the same procedurethatthe adultwould have applied.Rather,it ap-
pears to me thatas suggested by Cavell (1990), this agreementis a much more
complicatedaffair in which there is an entanglementof rules, customs, habits,
examples,andpracticesandthatwe cannotattachsalvationalimportanceto any
one of these in questions pertainingto the inheritanceof culture.Wittgenstein
(1953) speaks aboutordersor commandsin several ways: Thereis the gulf be-
tween the orderand its execution or the translationof an orderone time into a
propositionand anothertime into a demonstrationand still anothertime into ac-
tion. I do not have the sense thatthe agreementin formsof life requiresthe child
to producethe same responsethatthe teacherdoes. To have a futurein language,
the child should have been enabledto say "andafterI had trainedmy mouthto
form these signs, I used them to express my own desires."There is of course
the reference in Wittgenstein(1953:para.219) to following a rule blindly.
"Allthestepsarealreadytaken"means:I no longerhaveanychoice.The
meaning,tracesthelinealongwhichit is
rule,oncestampedwitha particular
to be followedthroughthe whole of space,-But if somethingof this sort
reallywerethecase,howwouldit help?
No; my descriptiononlymadesenseof it was to be understood
cally.-I shouldhavesaid:Thisis howit strikesme [emphasisin the origi-
WhenI obeya rule,I do notchose.
I obeytheruleblindly[emphasisin theoriginal].
And then in paragraph221, he explains, "My symbolical expression was
really a mythological descriptionof the rule."I cannot take up fully the ques-
tion here of what it is to speak mythologically or symbolically, but from the
aurathat surroundsthe discussion of these issues, speaking of obeying a rule
blindly seems to be similarto the way one speaks of wishes, plans, suspicions,
or expectationsas, by definition,unsatisfiedor the way one speaks of proposi-
tions as necessarily true or false, that is, thatthey are grammaticalstatements.
When Wittgenstein(1953) talks aboutrules and agreementbeing cousins, the
kinship between them seems more complicatedthan Kripke's (1982) render-
ing of either of these two concepts allows.

I want to take an ethnographicvignette now to show the entanglementof the

ideas of rule, custom, habit, practice, and example in what might be seen as
constituting agreementwithin a particularform of life. Gilsenan (1996) has
given us a stunningethnographyof violence andnarrativein Akkar,a northern
province of Lebanon,in the 1970s. Fromthe several narrativesin this text, one
can infer the rules by which issues of vengeance and honor are articulatedin
the exchange of violence. Indeed, if one reads Evans-Pritchard(1940) on the
feud among the Nuer, it all seems like a matterof kinship obligations that can
be stated in terms of clear genealogical principlesthroughwhich feuds are or-
ganized. One could imagine thata male child being socialized into such a soci-
ety could be taughthis place in the communityin terms of rules thathe learns,
much as Kripke's child learnsto follow the same proceduresas the adultswho
are initiatinghim if he is to learnhow to add.But here are sketches from a story
from Gilsenan(1996:165-66) of how a boy becomes a man even as he is being
initiated into the rules of vengeance.
... the chosen young man walked, alone and in broad daylight, up the
steep hill separatingthe quartersof the fellahins andthe aghas.... Everyone
could see him, a fact much insisted upon in accounts.At the top of the hill, he
approachedthe small ill provisioned shop owned by Ali Bashir who was
standingat the entrancelooking on to the saha (public space) before him ...
the boy simply said to him: "Do you want it here in the shop or outside?"Ali
ran back inside, grabbedthe gun, and was shot in the wrist, his weapon fal-
ling to the ground.The killer then emptied his revolver into Ali's chest. He
died instantly.
Turninghis back on those fellahin who had witnessed his deed, the kil-
ler-and now hero-walked back down the hill.... All agreed that he pre-
sented his back to the enemies in a granddisregardfor his own safety. No one
This archetypalgeste of agnostic indifferencefilled every requirementof
the heroic act. He was superbin exit as he had been on entry. The aesthetics
of violence were in all respects harmoniouslyachieved.
My informants all remembered that the senior of their number, a re-
nowned hunter, companion of the lords, and also a paternalhalf-brotherof
the wounded man, hailed the young hero when he came down to the lower
mosque at the entranceof the village exclaiming: "Ya 'aish! Reja'it shabb!"
(Long may you live! You have returneda man!). He saluted one who had
gone up the hill a boy and come down a true, arms bearingyoung man.
Some may arguethatthe scene of the instructionin Kripke(1982) bears lit-
tle resemblanceto the scene in which this young man is chosen by the elders as
the appropriateinstrumentof revenge. (But then is the example of learning a
procedurefor solving a mathematicalproblema good analogy for what it is to
obey rules-a particularlyclarifying one, as Kripkeclaims?) As for the young
boy, it is his display of the aesthetics of violence thatmakes him a man.No one
can say thathe acted exactly as the elderwould have acted in his place, for such
178 DAS

scenes arealso markedby contingenciesof all kinds in which one might end up
not a hero but a buffoon. Yet it is throughthe entanglementof rule, custom,
habit, and example thatthe child has not only been initiatedin the community
of men but has also foundhis own style of being a man.In fact, the aftermathof
the story of this young hero convertshim into a source of danger,always look-
ing for some replication of the originatingmoment of his public biography,
and who finally dies in a quarrelas if he were predestinedto have such a death.
A considerationof thatevent would take us into a differentregion of Wittgen-
stein's thought:the region of the dangersthatthe othernessof this hero posed
for the rest of the community.
Anthropologicalaccounts have suggested that attentionto Wittgenstein's
discussion on rules and especially the distinctionbetween regulativerules and
constitutive rules, as suggested by Searle (1969), may give new direction to
questions of how to distinguishthe natureof prescriptionsin ritualactions and
otherkind of actions. Humphrey& Laidlaw(1994) have not only writtena fas-
cinatingaccountof the Jainritualofpuja (worship),they have also arguedthat
what is distinctive about ritualprescriptionsin general is the constitutive na-
ture of rules that go to define rituals.
Constitutiveritualscreatea formof activity,by definingtheactionsof which
it is composed.Wepointedoutthatritualizedactionis composedof discrete
actswhicharedisconnected fromagents'intentionsandwe saidthatthisfea-
tureof ritualization
dependsuponstipulation. as distinct
Itis thisstipulation,
frommereregulationwhichis constitutiveof ritual.Onlyritualacts (like
validmovesinchess)countashavinghappened, so thecelebrantmovesfrom
actto act,completingeachin turnandthenmovingonto thenext.Thisis un-
affectedby delays,falsemoves,extraneoushappenings, ormishaps.(Hum-
phrey& Laidlaw1994:117)
They use this distinctionthen to show the wide variationin the ways of per-
forming the ritual act ofpuja, which are nevertheless consideredright by the
participantsbecause they may be said to accordwith the constitutiverules. The
importanceof this formulationis not only thatit breaksaway fromthe distinc-
tions between instrumentalaction and expressive action, or from the overde-
terminedview of ritualas a form of communication,but also that it addresses
some puzzling featuresof ritualobservationsthat are often ironed out of final
ethnographictexts. I refer to the kinds of mundaneactivities that may be car-
ried on duringa ritualbut are neverthelessnot seen as constitutiveof the ritual
and hence can be ignored in judgments about "rightness"of a ritualact.
Thereis an explicit analogy in Humphrey& Laidlaw's (1994) discussion of
the constitutiverules of ritualandof chess. Wittgenstein's(1953) observations
on chess may be pertinenthere. He has talkednot only aboutthe rules thatcon-
stitute the game but also customs-for example, the use of the king to decide
by lots which playergets white in drawinglots (para.563) or not stampingand

yelling while makingthe moves (para.200). But Wittgensteinleads us to a dif-

ferent direction, one in which the entanglementof rules with customs, prac-
tices, and examples comes to the fore: "Whereis the connection effected be-
tween the sense of the expression 'Lets play a game of chess' and all the rules
of the game?-Well, in the list of rules of the game, in the teaching of it, in the
day-to-daypracticeof playing"(para. 197). Wittgensteinused the analogy of a
chess game to illuminatewhat it means for language to be governed by rules.
In both language and chess there are rules thathave no foundation,that is, the
rules cannotbe justified by referenceto reality:They areautonomous,andthey
could be different.But thereare limits to this analogy. The most importantdif-
ference, as pointed out by Baker& Hacker(1980), is thatthe rules of chess are
devised to cover every possible situation whereas our language cannot lay
down the rules thatwill cover every conceivable circumstance.Hence there is
always a gap between the rule and its execution. Could we say that the consti-
tutive rules of ritualcan cover every conceivable circumstance?I suggest that
while this is sometimes the ambition of the theoreticiansof ritual, as the mi-
mamsa school of Indianphilosophy claimed (see Das 1983), the embeddingof
ritualin the forms of life do not allow for this. In fact a situationof complete-
ness would make rituallike the inventedlanguagesof Wittgensteinratherthan
the naturallanguages,which arenever complete (Wittgenstein1953:para.18).
Baker & Hacker(1980) suggest thatnaturallanguagegames may be distin-
guished from invented ones by the fact that the former are masteredonly in
fragmentswhile the latterare presentedas complete languages. The feeling in
readingabout the builders' language is thatthey seemed particularlybereft of
culture. I suggest it comes precisely from thinking of their language as if it
were complete.
An anthropologicaltext, we know, is markedby a certainkind of excess or a
certainsurplus.Call it thick ethnography,call it fascinationwith detail. Most
ethnographiesprovide more than the theoretical scaffolding requires. It has
been arguedby some that this excess is embedded in the emplotmentof eth-
nographyas a performance(Clifford 1990). Othershave spoken of the diffi-
culty of portrayingways of life that are "experiencedistant"to their readers
(Scheper-Hughes 1992). I suggest that this excess or this surplus expresses
equally the distrustof formalrules and obligations as sources of social orderor
moraljudgment. If cultureis a matterof sharedways of life as well as of be-
queathing and inheritingcapabilities and habits as members of society, then
clearly it is participationin forms of sociality (Wittgenstein's forms of life)
that define simultaneouslythe innerand the outer,that allow a personto speak
both within language and outside it. Agreement in forms of life, in Wittgen-
stein, is never a matterof sharedopinions. It thus requiresan excess of descrip-
tion to capturethe entanglementsof customs, habits, rules, and examples. It
provides the context in which we could see how we are to trace words back to
180 DAS

theiroriginalhomes when we do not know ourway about:The anthropological

quest takes us to the point at which Wittgensteintakes up his grammaticalin-
vestigation. It seems a naturalpoint to break here and inquire into what are
"forms of life," "criteria,"and "grammaticalinvestigation"in Wittgenstein.


Forms of Life
The idea of forms of life is what has often been takento signal the availability
of Wittgenstein's thoughtfor sociology and anthropology.Wittgensteintakes
languageto be the markof humansociality: Hence humanforms of life are de-
fined by the fact thatthey are formscreatedby and for those who are in posses-
sion of language.As it is commonly understood,Wittgenstein'snotion of lan-
guage is to see it in the context of a lived life, its use within humaninstitutions
ratherthan its systematic aspects. But is this enough? Cavell (1989) has ex-
pressedanguishat the conventionalviews of this text, which in his understand-
ing eclipse its spiritualstruggle.
Theidea[of formsof life] is, I believe,typicallytakento emphasizethe so-
cialnatureof humanlanguageandconduct,as ifWittgenstein's missionis to
rebukephilosophyforconcentrating too muchon isolatedindividuals,orfor
emphasizing theinnerattheexpenseof theouter,inaccounting forsuchmat-
tersas meaning,orstatesof consciousness, orfollowinga ruleetc.... A con-
ventionalizedsenseof formof life will supporta conventionalized, or con-
tractualsenseof agreement. Butthereis anothersenseof formof life thatwill
What Cavell finds wanting in this conventionalview of forms of life is that
it is not able to convey the mutualabsorptionof the naturaland the social-it
emphasizesform but not life. A hasty readingof Cavell on this point may lead
readers(especially anthropologists)to the conclusion thatthe idea of naturalis
taken as unproblematicin this interpretation.Let me dwell for a moment on
this point. Cavell suggests a distinctionbetween what he calls the ethnological
or horizontal sense of form of life and its vertical or "biological"sense. The
first capturesthe notion of human diversity-the fact that social institutions,
such as marriageand property,vary across societies. The second refers to the
distinctionscapturedin languageitself between "so-called 'lower' or 'higher'
forms of life, between say poking at your food, perhapswith a fork, and paw-
ing at it or pecking at it" (Cavell 1989:42). It is the vertical sense of the formof
life that he suggests marksthe limit of what is consideredhuman in a society
and provides the conditions of the use of criteriaas appliedto others.Thus the
criteriaof pain does not apply to that which does not exhibit signs of being a
form of life-we do not ask whether a tape recorderthat can be turnedon to
play a shriek is feeling the pain. Cavell suggests that the forms of life have to

be acceptedbut that it is in the sensibility of Investigationsto call not so much

for change as for transfiguration.I am going to leave aside, for the moment,the
relevance of this question for or against skepticism.InsteadI want to point to a
directionin which this distinctionbetween the horizontaland the vertical may
also show what happensat the limit of each. Whatis it thathumansocieties can
representas the limit? Here I draw from some of my own work to show how
such an idea may strike a chord on the keys of anthropologicalimagination.
For some years now I have been engaged in tryingto understandthe relation
between violence (especially sexual violence) in everyday domestic contexts
and violence in the extraordinarycontext of riots duringpolitical events, such
as the Partitionof India or the violence against Sikhs following the assassina-
tion of then prime minister IndiraGandhi. In one of my recent papers (Das
1996) I have triedto conceptualizethe violence thatoccurswithin the weave of
life as lived in the kinship universe, as having the sense of a past continuous,
while the sudden and traumaticviolence that was partof the Partitionexperi-
ence seems to have a quality of frozen time to it. In discussing the life of a
woman, Manjit, who had been abducted and raped during the Partitionand
subsequentlymarriedto an elderly relative, I argued that while the violence
she was submittedto by her husbandwas something sayable in her life, the
otherviolence was not (could not be?) articulated.The horizontaland vertical
limits seemed to me to be particularlyimportantin formulatingthis difference.
It is thisnotionof formof life, i.e. its verticalsenseof testingthe criteriaof
whatit is to be human,thatI thinkis implicatedin theunderstanding of Man-
jit's relationto thenon-narrativeof herexperience of abduction andrape.Men
beatuptheirwives,commitsexualaggression,shamethemin theirownself
creationsof masculinity-butsuchaggressionis still"sayable" in Punjabilife
throughvariouskindsofperformative gesturesandthroughstorytelling(I do
not meanto say thatit is thereforepassively accepted-indeed the whole story
of Manjitshows thatit is deeply resented).Contrastthis with the fantasticvio-
lence in which women were strippedand marchednaked in the streets;or the
magnitudesinvolved; or the fantasyof writingpolitical slogans on the private
partsof women. This productionof bodies througha violence thatwas seen to
tearapartthe very fabricof life, was such thatclaims over culturethroughdis-
putationbecame impossible. If words now appear,they are like brokenshad-
ows of the motion of everyday words .... Such words were indeed uttered
and have been recordedby otherresearchers,but it was as if one's touch with
these words and hence with life itself had been burntor numbed.The hyper-
bolic in Manjit'snarrationof the Partitionrecalls Wittgenstein's sense of the
conjunctionof the hyperbolic with the groundless. (Das 1996:23)

I have takenthis example in some detail because it suggests, throughmeans

of an ethnography,thatwhile the rangeand scale of the humanis tested andde-
fined and extended in the disputationsproper to everyday life, it may move
throughthe unimaginableviolence of the Partition(but similarexamples areto
182 DAS

be found in many contemporaryethnographiesof violence) into forms of life

that are seen as not belonging to life proper.Was it a man or a machine that
plunged a knife into the privatepartsof a woman afterrapingher?Were those
men or animals who went aroundkilling and collecting penises as signs of
their prowess? There is a deep moral energy in the refusal to representsome
violations of the humanbody, for these violations are seen as being againstna-
ture, as defining the limits of life itself. The precise range and scale of the hu-
man form of life is not knowable in advance, any more thanthe precise range
of the meaning of a word is knowable in advance. But the intuitionthat some
violations cannotbe verbalizedin everyday life is to recognize thatwork can-
not be performedon these within the burnedand numbedeveryday. We reach
througha differentroutethe questionof what it is to have a futurein language.
I believe thatthe limits of the forms of life-the limits at which the differences
cease to be criterialdifferences are encounteredin the context of life as it is
lived and not only in the philosopher'sreflections on it. These are the times in
which one may be so engulfedby doubtsof the other's humanitythatthe whole
world may appearto be lost.
In his work on violence, Daniel (1997) calls this point the counterpointof
culture:"The counterpointI speak of is something that resists incorporation
into the harmonyof a still higher orderof sound, sense, or society" (p. 202).
Otheraccountsof violence similarlysuggest thatcertainkinds of violence can-
not be incorporatedinto the everyday (Langer 1991, 1997; Lawrence 1995):
But then how is everyday life to be recovered?

EverydayLife and the Problem of Skepticism

In describing what he calls the counterpointto culture, Daniel (1997) inter-
viewed severalyoung men in Sri Lankawho were membersof variousmilitant
movements and who had killed with ropes, knives, pistols, automaticfire, and
grenades.But it is clear fromhis powerfuldescriptionsthatwhatwas traumatic
for Daniel in hearingthese accounts of killings was the mannerin which the
styles of killing and the wielding of words was interwoven.Here are some ex-
Hewashidingin thetemplewhenwe gotthere.... Thisboywashidingbe-
hindsomegod.Wecaughthim.Pulledhimout... Theboywasin themiddle
of theroad.Wewereall goingroundandroundhim.Fora longtime.No one
saidanything.Thensomeoneflungathimwitha sword.Bloodstartedgush-
ing out.... We thought he was finished. So they piled him on the tyre and
thenset it aflame.(Daniel1997:209)
Daniel finds the shifting between the we and the they to be noteworthy,but
what stuns him is the next thing that happened.
Thiswastheearlydaysof myhorrorstorycollectingandI didnotknowwhat
to say. So I asked him a question of absolute irrelevanceto the issue at hand.

HeavenknowswhyI askedit;I musthavedesperately wantedto changethe

subject pretendthat we had been talkingaboutsomethingelse all along.
"Whatis yourgoal in life?"I asked.The replyshot rightback:"I wanta

Wittgenstein's sense of exile of words is what comes to mind here. It is not

that one cannot understandthe utterancebut that in this context when these
words are spoken, they seem not to belong-they seem not to have a home.
Daniel's (1997) turningaway fromthis event is a desperateone. He lurchesto-
ward a hope (p. 211)-the rustle of a hope-wherever it may be found and
wheneverit may be found.And it is foundin a scene of almostquietdomesticity.
He recountsan event in the 1977 anti-Tamilriots in which a Sinhalawoman is
journeyingon a train;she is in one partof the compartment,andon anotherseat
is a retiredTamil schoolteacher.A mob began to drag out Tamils and to beat
them. The Sinhalawoman, recognizableeasily as a KandyanSinhalesebecause
of the way she wore her sari, moved over to his side and quietly held his hand.
Some membersof the mob enteredthe compartment,but the gestureof conjugal
familiarity persuaded them that the gentleman was a Sinhala, so they pro-
ceeded elsewhere. Daniel (1997) thinks of the gesture of the woman as a sign,
gravidwith possibilities.But what arethese possibilities?Froma Wittgenstein-
ian perspective,these seem to be only possibilitiesof recoverythrougha descent
into the ordinarinessof everyday life, of domesticity, throughwhich alone the
words thathave been exiled may be broughtback. This everydaynessis then in
the natureof a return-one that has been recovered in the face of madness.
The intuition of everydayness in Wittgenstein appearsthereforequite dif-
ferent from, say, that of Schutz (1970), who emphasizes the attentionto the
"paramountreality"of the everyday and conceptualizestranscendenceas mo-
mentary escapes from these attentions.It is also different from the many at-
tempts made in recent years to capturethe idea of the everyday as a site of re-
sistance (Jeffery & Jeffery 1996; Scott 1985, 1990). My sense of these ap-
proaches is that there is a search in these attemptsfor what Hans Joas (1996)
calls the creativity of social action. Ratherthan searchingfor agency in great
and transgressivemoments of history, it is in the everyday scriptsof resistance
thatit is thoughtto be located. Thereis nothingwrong with this way of concep-
tualizing the everyday, for it has the advantageof showing society to be con-
stantlymade ratherthan given. The problemis that the notion of the everyday
is too easily secured in these ethnographiesbecause they hardlyever consider
the temptationsand threatsof skepticism as partof the lived reality and hence
do not tell us what is the stake in the everyday that they discovered.
In Cavell's (1984, 1988, 1990) renderingof Wittgenstein's appeal to the
everyday, it is found to be a pervasive scene of illusion and tranceand artifici-
ality of need. This, to my understandingand experience, is because both the
temptationsand threatsof skepticism are taken out from the study of the phi-
184 DAS

losopherandreformulatedas questionsaboutwhat it is to live in the face of the

unknowabilityof the world (for my purposesespecially the social world). Let
me departfor the moment from the kinds of scenes of violence thathave been
describedby Das (1990b, 1995a,b, 1997), Daniel (1997), Langer(1991, 1997),
Lawrence (1995), and many others. These scenes may appearexceptional to
many. Instead,I ask, is the sense of the unknowabilityof the social world also
encounteredin other contexts, in the context of normalsuffering, so to speak?
Some scholars suggest that this unknowability of the social world has been
made more acute by the processes of modernityor globalization(see Appadu-
rai 1996:158-78), whereas my sense is that uncertaintyof relations is part of
human sociality as it is embeddedwithin certainweaves of social life (Das &
Bajwa 1994). But let me take my example from an anthropologicalclassic.
Evans-Pritchard's(1937) account of witchcraftamong the Azande has of-
ten been seen as thatsociety's way of dealing with misfortuneratherthanwith
the essential unknowabilityof other minds. For instance, Taussig (1992) has
To cite the common phraseology, science like medical science, can explain
the "how" and not the "why" of disease; it can point to chains of physical
cause and effect, but as to why I am struck down now ratherthan at some
other time, or as to why it is me ratherthan someone else, medical science
can only respondwith some varietyof probabilitytheorywhich is unsatisfac-
tory to the mind which is searching for certainty and for significance. In
Azande practice, the issue of"how" and "why"are folded into one another;
etiology is simultaneouslyphysical, social, andmoral.... My disease is a so-
cial relation, and therapyhas to addressthat synthesis of moral, social, and
physical presentation.(p. 85)
It is true that Evans-Pritchard(1937) veered in several directions in ac-
countingfor the Azande beliefs in witchcraft,includingquestionsaboutthe ra-
tionality of the Azande. If we pay some attentionto the descriptionsthat he
provides,however, we find not so much a searchfor certaintyand significance,
but rathera shadow of skepticism regardingother minds (Chaturvedi1998).
Moreover,this skepticism seems to have something to do with the mannerin
which language is deployed.
Evans-Pritchard(1937) reportsthatthose who speak in a roundaboutman-
ner and are not straightforwardin their conversationare suspected of witch-
craft:"Azandeare very sensitive and usually in the lookout for unpleasantal-
lusions to themselves in apparentlyharmlessconversation"(p. 111). Very of-
ten they find double meaning in a conversation(p. 116) and assume that harm
would be done to them, as in the following instance recounted by Evans-
An old friend of mine, Badobo of the Akowe clan, remarkedto his compan-
ions who were cleaning up the governmentroadaroundthe settlementthathe

hada founda stumpof woodoverwhichTupoihadstumbledandcuthimself

a few dayspreviouslywhenhe hadbeenreturninglateat nightfroma beer
feast.Badoboaddedto his friendsthatthey mustclearthe roadwell, as it
wouldneverdo for so important a manas Tupoito stumbleandfall if they
couldhelpit. Oneof Tupoi'sfriendsheardthisremarkandrepeatedit to his
fatherwhoprofessedto see a doublemeaningin it andto finda sarcasticnu-
ancein Badobo'swholebehaviour.(pp. 115-16)

A pervasive uncertainty of relations is indicated by many factors: the

Azandeaphorism"Onecannotsee into a manas into an openwoven basket";the
Azande belief that one cannot be certainthat anyone is free from witchcraft;
and the carethata Zandeman takes not to angerhis wives gratuitouslybecause
one of them may be a witch and by offending her he may bring misfortuneon
his head. And although a Zande would not state that he is a witch, Evans-
Pritchard(1937) reports that one may know nothing about the fact of one's
own witchcraft(p. 123). Uncertaintyaboutotherminds here is linked to a cer-
tain alienationfrom the languagethatone speaks, as if the languagealways re-
vealed either more or less than the words spoken. Indeed it is the intimate
knowledge of how Azande converse and interpretone another'smeaningsthat
Evans-Pritchard(1937) considers importantto an understandingof how attri-
butions of witchcraftare made: "Once a person has been dubbeda witch any-
thing he says may be twisted to yield a secret meaning. Even when there is no
question of witchcraftAzande are always on the look-out for double meaning
in their conversations"(p. 131). Here we have the intuitionof the humansas if
one of the aspectsunderwhich they could be seen is as victims of languagethat
could reveal things about them of which they were themselves unaware.
This idea touches upon the Wittgensteiniantheme of language as experi-
ence (and not simply as message). He takes examples of punning, or of a feel
for spelling: If you did not experience the meaning of words (as distinct from
only using them), then how could you laugh at a pun? The sense is of being
controlledby the words one speaks or hears or sees ratherthan of controlling
them. Thereis some similarityto Austin's (1975) concernswith performatives
especially with perlocutionaryforce.
A context that I consider decisive for understandingthese themes is that of
panic rumor.I shall take the example of anthropologicalstudies of rumorto
show how the theme of the unknowabilityof the social world and the theme of
humansbecoming victims to words come to be connected. Although rumoris
not an example that figures in Wittgenstein,I propose that one may find con-
nections in the way in which there is a withdrawalof trust from words and a
special vulnerabilityto the signifier in the working of rumorand the exile of
words under skepticism.
Several historiansand anthropologistshave emphasized the role of rumor
in mobilizing crowds (Rude 1959, 1964; Thompson 1971). Historians of the
186 DAS

subalternschool have seen it as a special form of subalterncommunication,"a

necessary instrumentof rebel transmission"(Guha 1983:256). Othercharac-
teristics of rumoridentifiedby Guha(1983) are the anonymityof the source of
rumor,its capacityto build solidarity,andthe overwhelmingurge it promptsin
listeners to pass it on to others. The excessive emphasis on communication,
however, obscures the particularfeature of language that is often broughtto
the fore when we consider the susceptibilityto rumorduringtimes of collec-
tive violence (Das 1990a,b, 1998; Tambiah 1996). Bhabha(1995) has posed
the question in an incisive manner:What is special to rumoras distinct from
other forms of communication?He goes on to isolate two of its aspects. The
first is rumor's enunciative aspect, and the second its performativeaspect.
"Theindeterminacyof rumour,"he says, "constitutesits importanceas a social
discourse. Its intersubjective,communal adhesiveness lies in its enunciative
aspect. Its performativepower of circulationresults in its contiguous spread-
ing, an almost uncontrollableimpulse to pass it on to anotherperson"(p. 201).
He concludes thatpsychic affect and social fantasy are potent forms of poten-
tial identification and agency for guerrillawarfare and hence rumorsplay a
majorrole in mobilization for such warfare.
Otherviews of rumor,especially those derivedfrommass psychology, have
emphasized the emotional, capricious, temperamental,and flighty nature of
crowds (Le Bon 1960). Something common in these situationsis an essential
grammaticalfeature(in Wittgenstein's sense) of what we call rumor:that it is
conceived to spread.Thus while images of contagion and infection areused to
representrumorin elite discourse,the use of these images is not simply a matter
of the elite's noncomprehensionof subalternforms of communication.It also
speaks to the transformationof language;namely, that instead of being a me-
dium of communication,languagebecomes communicable,infectious,causing
thingsto happenalmostas if they hadoccurredby nature.Inmy own workon ru-
mor in a situationof mountingpanic of communalriots, I have identified the
presenceof an incompleteor interruptedsocial storythatcomes backin the form
of rumorandan alteredmodalityof communication(Das 1998). The most strik-
ing featureof whatI identifyas panicrumors(in which it is difficultto locate any
innocentbystanders)is thatsuddenlythe access to contextseems to disappear.In
addition,thereis an absenceof signaturein panicrumorsso thatrumorworksto
destroyboth the source of speech and the trustworthinessof convention. (This
characteristicseems to distinguish perlocutionaryforce from illocutionary
force. In the latter,trustin conventionand law allows promisesto be made and
marriagesto be contracted.)Cavell (1982) has invoked Othello as the working
out of a skepticalproblematic.The mountingpanic in which the mediumof ru-
mor leads to the dismantlingof relations of trust at times of communal riots
seems to sharethe tempo of skepticism.Once a thoughtof a certainvulnerabil-
ity is lost, as Cavell shows (1982, 1994), the world is engulfed without limit.

Unlike Cavell, Williams (1996) considers skeptical doubts to be unnatural

doubts.He holds thatthe experiencethatwe know nothingaboutthe realworld
has to arise from a particularlystrikingexperience of error.Yet no experience
of error,he argues, can give us a feel of a total loss of the world. The threatof
skepticism for Cavell lies in our feeling that our sensations may not be of this
world: But for Williams this threatarises in the philosophizing of Cavell be-
cause he has internalizeda contentious theoreticalview. Cavell, on the other
hand, suggests in all his work that skeptical doubt arises in the experience of
living. Skepticism is for him a site on which we abdicateour responsibilityto-
ward words-unleashing them from our criteria. Hence his theme of disap-
pointment with language as a human institution (Cavell 1994). The site of
panic rumorsuggests similarlya subjectionto voice (comparableto Schreber's
subjection to the voices he heard). There seems a transformationfrom social
exchange to communal trance, and if this trance is to be resisted, one has to
lead works back to the everyday, much as one might lead a horse gone sud-
denly wild to its stables.


It might be temptingto suppose thatthe unknowabilityof the social world es-

sentially relates to the unknowabilityof the other. But the question of skepti-
cism in Wittgenstein does not posit an essential asymmetrybetween what I
know about myself and what I know about the other. His famous arguments
against the possibility of a privatelanguage is not that we need sharedexperi-
ence of language to be communicableto one anotherbut that without such a
sharingI will become incommunicableto myself. The innerfor Wittgensteinis
thus not an externalizedouter-there is no such thing as a privateinnerobject
to which a privatelanguagemay be found to give expression. This view is not
to be construedas Wittgenstein'sdenial of the innerbut ratherthat innerstates
are,as he says, in need of outwardcriteria(Johnston1993, Schulte 1993). Thus
what appearoften in our language as intrinsic differences between different
kinds of inner states are basically grammaticaldifferences in disguise. PartII
of Philosophical Investigations begins with the following:
One can imagine an animal angry, frightened,unhappy,happy, startled,
but hopeful? And why not?
A dog believes his masteris at the door. But can he also believe his master
will come day afterto-morrow?And what can he not do here?-How do I do
it?-How am I supposed to answer this?
Can only those hope who can talk? Only those who have masteredthe use
of a language?That is to say, the phenomenaof hope are modes of this com-
plicated form of life. (Wittgenstein 1953:174)
188 DAS

The reference to language here is obviously not to suggest that those who
have masteredthe use of a languagehave acquiredthe logical skills necessary
to express hope but ratherthatgrammartells us what kinds of objects hope and
grief are. Thus the innerstates are not distinguishedby some referenceto con-
tentbut by the way we imagine somethinglike an innerstatefor creaturescom-
plicated enough to possess language (and hence culture).I would like to illus-
tratethis idea with referenceto the discussion of belief and then follow the il-
lustrationwith a discussion on pain.
The questionof belief in Philosophical Investigationsappearsas the asymme-
try between the use of first-personindicative and third-personindicative. Two
observations in the second part of this text are crucial. The first is "If there
were a verb meaning to believe falsely it would not have a first person indica-
tive" (Wittgenstein1953:190). The second, closely relatedto thatobservation,
is "I say of someone else 'He seems to believe. ....' And other people say it of
me. Now, why do I never say it of myself, not even when othersrightlysay it of
me?-Do I not myself see or hear myself, then?"(Wittgenstein 1953:191).
Wittgensteinis asking, Whatdoes a belief look like from the inside? When
he says that it is possible to misinterpretone's own sense impressionsbut not
one's beliefs, he is not referringto the contentof an innerexperiencebut rather
to the grammaticalimpossibility of inferringone's belief (or one's pain) intro-
spectively. That is why he says that if therewere a verb thatmeant "to believe
falsely," it would lack a first-personpresent indicative. Wittgenstein is not
stating a metaphysical truth about belief here, but a grammaticalone. Even
when it is possible to make such statementsas "Itis rainingandI do not believe
it," the grammarof the termbelief does not allow us to make these statements,
for we cannot imagine a context for such statements-they violate the picture
of the inner in the grammarof the word belief.
Anthropologistshave wrestled with the problemof belief in the context of
translationof cultures.The problemhas been persistent:Whenanthropologists
attributebelief statementsto membersof othercultures(i.e. non-Westerncul-
tures), are they making a presumptionthat a common psychological category
of most Western languages and cultures is to be treatedas a common human
capacitythatcan be ascribedto all men andwomen? Such questionshave been
asked of several categories of emotion (see Lutz & Abu- Lughod 1990, Lutz &
White 1986), but the case of belief is special because it has been anchoredto
questionsof universalhumanrationalityon the one hand(Gellner 1970, Lukes
1977) and common human condition of corporealityon the other (Needham
As far as the side of universalrationalityis concerned,the puzzle for many
scholarsseems to be to accountfor the apparentirrationalityof beliefs like that

of witchcraftor of other scandalousstatements:for example, thatthe Nuer be-

lieve thattwins arebirds(Evans-Pritchard1956). In his polemic againstthe an-
thropological tendency to find coherence in such statements, Gellner (1970)
states that only throughan excessive charityof translationcan such beliefs be
renderedintelligible. He seems to suggest that they are either to be taken as
evidence of prelogical thoughtor as ideological devices to hide the power ex-
ercised by privileged classes in society (the latterpoint is made with regardto
the category of barak among Moroccan Berbers). Gellner warns that "[t]o
make sense of the concept is to make non-sense of the society" (1970:42).
Asad (1990) has given a devastatingcritique of Gellner's method, especially
of the mannerin which in his haste to pronounceon the irrationalityof such
concepts he actually manages to evade all questions on their use in everyday
life of the society under consideration.Wittgenstein's general view seems to
be that there are many empirical assertions that we affirm without specially
testing them and that their role is to establish the frame within which genuine
empiricalquestions can be raisedand answered(Cavell 1969, Williams 1996).
If this scaffolding is questioned, then we are not in the realm of mere differ-
ences of opinion. Thus to someone who is offering an explanation of the
French Revolution I will probablynot ask whether she has any proof that the
world is not an illusion. If such a questionis asked,we shall have to say thatour
differences are noncriterialdifferences that cannot be resolved by adducing
more evidence.
Thus, for the Azande there are genuine empiricalquestions abouthow one
is to know whetherone's illness is to be attributedto the witchcraftof a neigh-
bor or a wife. The final empiricalproof of the cause is provided by the post-
mortem of a body to show whetherwitchcraftsubstanceis found in the body.
Obviously if one shifts this kind of questionto the kind of questionin which we
ask a Zandeif he or she believes witches to exist, one is shiftingthe framecom-
pletely. In this revised frame (in which we are certainthat witches do not ex-
ist), one can ask questions only about witchcraftbeliefs, or witchcraftcraze,
but not about superiorityof one kind of witchcraft medicine over anotheror
whether unknowing to oneself one may be a witch-a source of danger to
one's neighborsand friends.Whatdoes this mean for the practiceof ethnogra-
phy? One strategyis thatadoptedby Fevret-Saada(1977), who felt thatto open
her mouthon issues of witchcraftin Bocage was to become implicatedin utter-
ances that constitute the practices of witchcraft. Thus her ethnographybe-
comes an account of the complicated relation that the ethnographercomes to
have with the "bewitched"and the "unwitchers."It does not raise questions
about the rationalityor truthof witchcraftbeliefs because there is no way in
which such questions may be asked from within the language games of the
Bocage. The other strategy is to think of ethnographyas a persuasive fiction
(Strathern1988). I shall returnto the question of translation.For the moment
190 DAS

let me say only thatthe disappointmentin the indeterminateplace of anthropo-

logical knowledge is perhapslike the disappointmentwith language itself, as
somehow naturalto the human.This disappointmentis a greatWittgensteinian
theme and shouldperhapslead us to thinkthatthe reasonwhy so-called contra-
dictions in belief do not paralyzeone in any society is thatone's relationto the
world is not on the whole thatwhich would be based on knowing (Cavell 1969,
1982, 1994, 1995).

Belief and Corporeality

Needham's (1972) enquiryon the status of belief statementsand the problem
of translationis on entirely different lines. He states,
If they [beliefs]are assertionsaboutthe innerstatesof individuals,as by
commonusagetheywouldnormallybe takento be, then,so faras my ac-
quaintancewith the literaturegoes, no evidenceof suchstates,as distinct
fromthecollectiverepresentations thatarethusrecorded,is everpresented.
In thiscasewe haveno empiricaloccasionto acceptsuchbelief-statements
as exactandsubstantiated reportsaboutotherpeople.(p. 5)
Needham goes on to address this problem throughWittgenstein's idea of
grammaticalinvestigationandparticularlythatan innerprocess standsin need
of outwardcriteria.However, his notion of grammaticaldoes not appearto be
that of Wittgenstein's-it is hasty and confuses philosophical grammarwith
the notion of grammarin linguistics (perhapsit is comparableto a case of sur-
face grammarin Wittgenstein,but I am not on sure groundhere). The burden
of Needham's argument is that even when we are convinced that a person
genuinely believes what he says he believes, our conviction is not based on ob-
jective evidence of a distinctinnerstate:"We can thus be masters,as we are, of
the practicalgrammarof belief statementsyet remainwholly unconvincedthat
these rest on an objective foundation of psychic experience" (Needham
Now if I am correctthatthe inneris not like a distinct state that can be pro-
jected to the outer world through language in Wittgenstein but rather like
something that lines the outer, then language and the world (including the in-
ner world) are learnedsimultaneously.Neeedham is right in suspecting that a
grammarof belief in the English languageand in forms of life in which beliefs
are held, confessed, defended, solicited, guarded, and watched over may be
differentfrom the way in which similarconcepts throughwhich the world and
the word areconnectedin the woof andweft of some othersociety's life. How-
ever, the solution Needham (1972) offers to the problem of translation-that
some innerstatesare accompaniedby bodily expressions (such as body resem-
blance, naturalposture, gesture, facial expression) whereas other inner states
(such as belief) have no specific behavioral physiognomy-is to misread

grammaticaldifferences as intrinsic differences in the content of experience.

Wittgenstein's way of describingthis idea was to say thatthe body is a picture
of the soul or thatthe soul standsnext to the body as meaningstandsnext to the
We arethus not going to get out of the problemof translationby an appealto
certainhuman capacities that are real and universal, as contrastedwith others
that are artificial constructs of various cultural traditions, as proposed by
Needham (1972). Thatis not to say thatwe do not read the body but ratherthat
we dependon grammarto tell us what kind of an object somethingis. Inserting
the centralityof the body in humansociety is importantnot in inferringinternal
states of mind but in the intuitionof languageas a bodying forth,as in Wittgen-
stein saying, "Sometimes a cry is wrenched out of me." Let us now consider
this question with regardto pain.
Pain and Private Objects
Wittgenstein on pain is a major philosophical and anthropologicalissue, yet
there is no highway of thoughtavailable to traverse.It would have to be from
the side roads and the meanderingin unchartedterritoriesthat one would find
the relationbetween Wittgenstein's thoughts on pain and the anthropological
task of studying forms of sociality. ConsiderCavell (1997), who says,
PhilosophicalInvestigations is thegreatworkof philosophyof thiscentury
whosecentraltopicmaybe saidto be pain,andoneof its principaldiscover-
ies is thatwe will neverbecomeclearabouttherelationof attributions of the
conceptof pain,noraboutanyof the conceptsof consciousness,norof any
unconsciousness-neither of myattributionof painto myselfnorof myattri-
butionof painto others-withoutbringingintoquestiontheendlesspictures
we havein storethatprejudicially distinguishwhatis internalor privateto
creatures(especiallyoneswithlanguage,humans)fromwhatis externalor
publicto them.(p. 95)
In some of the most creative anthropologicalwriting on this issue, we find
the disappointmentwith languageto somehow be integralto the experience of
pain (Good et al 1992). Wittgensteinemphaticallydenies the possibility of a
privatelanguage in this case, as in othercases, thatrefers to what is internalor
privateto creatures.But what this means is thatfor Wittgensteinthe statement
"Iam in pain"is not (or not only) a statementof fact but is also an expressionof
thatfact (Cavell 1997). The internal,as I have stated,is not an internalizedpic-
ture of the outer-nor is the external only a projectionof the internal.In this
context, what is unique about pain is the absence of any standing languages
either in society or in the social sciences that could communicatepain, yet it
would be a mistaketo thinkof pain as essentially incommunicable(Das 1997).
At stake here is not the asymmetrybetween the first person ("I am never in
doubt about my pain") and the thirdperson ("you can never be certain about
192 DAS

anotherperson's pain"),but ratherthatto locate pain I have to take the absence

of standinglanguagesas partof the grammarof pain. To say "I am in pain"is
to ask for acknowledgmentfrom the other,just as denial of another's pain is
not an intellectual failure but a spiritual failure, one that puts our future at
stake: "One might even say that my acknowledgementis my presentation,or
handling of pain. You are accordingly not at liberty to believe or disbelieve
what it says-that is the one who says it-our future is at stake" (Cavell
1997:94). Some passages from The Blue and Brown Books (Wittgenstein
1958) areremarkablein the notion of languageas embodiedor bodying forth.
In orderto see thatit is conceivablethatone personshouldhavepainin an-
otherperson'sbody,one mustexaminewhatsortsof factswe call criterial
fora painbeingin a certainplace.... SupposeI feel thata painwhichon the
evidenceof thepainalone,e.g. withclosedeyes, I shouldcall a painin my
lefthand.Someoneasksmeto touchthepainfulspotwithmyrighthand.I do
so andlookingaroundperceivethatI amtouchingmyneighbour'shand....
Thiswouldbe painfelt in another'sbody.(p. 49)
I have interpretedthis passage (see Das 1995a,b, 1997) to propose that
Wittgenstein's fantasy of mypain being located in your body suggests either
the intuitionthat the representationof sharedpain exists in imaginationbut is
not experienced,in which case I would say that language is hooked ratherin-
adequatelyto the world of pain or thatthe experience of pain cries out for this
response to the possibility thatmy pain could reside in your body and that the
philosophicalgrammarof pain is aboutallowing this to happen.As in the case
of belief, I cannotlocate your pain in the same way as I locate mine. The best I
can do is to let it happento me. Now it seems to me thatanthropologicalknowl-
edge is precisely aboutletting the knowledge of the otherhappento me. This is
how we see Evans-Pritchardfinding out about himself that he was "thinking
black"or "feeling black"thoughhe resistedthe tendencyto slip into idioms of
witchcraft.In the Introductionto this paper,I talkedof Wittgenstein'sidea of a
philosophical problem as having the form "I do not know my way about."In
his remarkson pain, to find my way is similar to letting the pain of the other
happen to me. My own fantasy of anthropologyas a body of writing is that
which is able to receive this pain. Thus while I may never claim the pain of the
other, nor appropriateit for some other purpose (nation building, revolution,
scientific experiment),thatI can lend my body (of writing)to this pain is what
a grammaticalinvestigationreveals.


In the preface to Philosophical Investigations,Wittgenstein(1953) wrote, "It
is not impossible that it should fall to the lot of this work, in its poverty and
darknessof this time, to bringlight into one brainor another-but, of course, it

is not likely" (p. vi). Beam (1998) writes thatthe destructivemomentof the In-
vestigationsthreatensthe fabricof our daily lives, so it is more destructivethan
textbook skepticism of the philosopheror the cafe skeptic. If in life, said Witt-
genstein, we are surroundedby death,so too in the health of ourunderstanding
we are surroundedby madness (Wittgenstein 1980:44). Ratherthan forcefully
excluding this voice of madness, Wittgenstein(1953) returnsus to the every-
day by a gesture of waiting. "If I have exhausted the justifications, I have
reached bedrock, and my spade is turned.Then I am inclined to say: 'This is
simply what I do (handle)' "(para. 217). In this pictureof the turnedspade,we
have the pictureof what the act of writing may be in the darknessof this time.
The love of anthropologymay yet turnout to be an affairin which when I reach
bedrockI do not breakthroughthe resistanceof the other.But in this gestureof
waiting, I allow the knowledge of the otherto markme. Wittgensteinis thus a
philosopherof both cultureand the counterpointof culture.

Visit the Annual Reviews home page at


Literature Cited

AppaduraiA. 1996. Modernityat Large: Cul- philosophy. In Themes Out of School. Ef-
tural Dimensions of Globalization, pp. fects and Causes, pp. 195-234. San Fran-
48-66, 158-78. Minneapolis:Univ. Minn. cisco: North Point
Press Cavell S. 1988. The uncanniness of the ordi-
Asad T. 1990. The concept of culturaltransla- nary. In In Quest of the Ordinary.Lines of
tion in British social anthropology.In Ge- Skepticismand Romanticism,pp. 153-78.
nealogies of Religion. Discipline andRea- Chicago: Univ. Chicago Press
sons of Power in Christianity and Islam, Cavell S. 1989. Declining decline: Wittgen-
pp. 171-200. Baltimore/London: Johns stein as a philosopher of culture. In This
Hopkins Univ. Press New Yet Unapproachable America. Lec-
Austin JL. 1975. How to Do Things with tures After Emerson After Wittgenstein,
Words. Cambridge, MA: Harvard Univ. pp. 29-77. Chicago: Univ. Chicago Press
Press Cavell S. 1990. The argumentof the ordinary:
Baker JP, Hacker PMS. 1980. Wittgenstein. scenes of instruction in Wittgenstein and
Meaning and Understanding,Vol. 1. Ox- in Kripke. In Conditions Handsome and
ford: Blackwell Unhandsome. The Constitution of Emer-
Beam GCF. 1977. Waking to Wonder. Witt- sonian Perfectionism, pp. 64-101. Chi-
genstein's Existential Investigations. New cago: Univ. Chicago Press
York. State Univ. NY Press Cavell S. 1994. A Pitch of Philosophy. Cam-
Bhabha HK. 1995. By bread alone: signs of bridge, MA: HarvardUniv. Press
violence in the mid-nineteenthcentury. In Cavell S. 1995. Notes and afterthoughtson the
Location of Culture, pp. 198-212. Lon- opening of Wittgenstein's Investigations.
don: Routledge In Philosophical Passages. Wittgenstein,
Cavell S. 1969. Must WeMean WhatWeSay? Emerson, Austin, Derrida, pp. 125-87.
Cambridge,UK: CambridgeUniv. Press Oxford: Blackwell
Cavell S. 1982. The Claim ofReason: Wittgen- Cavell S. 1997. Commentson Veena Das's es-
stein, Skepticism, Morality, and Tragedy. say "Language and body: transactions in
Oxford: Clarendon the constructionof pain." See Kleinman et
Cavell S. 1984. Existentialism and analytical al 1997, pp. 93-99
194 DAS

ChatterjiR, Mehta D. 1995. A case study of a Fevret-Saada J. 1977. Les mots, la mort, les
communalriot in Dharavi,Bombay. Relig. sorts. Paris:Gallimard
Soc. 42(4):5-26 Gellner E. 1970. Concepts and society. In Ra-
Chaturvedi R. 1998. Witchcraft and other tionality, ed. BR Wilson, pp. 18-49. Ox-
minds. M.Phil. diss. Univ. Delhi, Delhi, In- ford: Blackwell
dia. 120 pp. Gilsenan M. 1996. Lords of the Lebanese
Clasteres P. 1974. Socidet contre I'etat. Paris: Marches: Violence and Narrative in an
Les Editions Minuit Arab Society. Berkeley:Univ. Calif. Press
Clifford J. 1990. On ethnographicallegory. In Good MJD, Brodwin PE, Good B, Kleinman
WritingCulture: The Poetics and Politics A, eds. 1992. Pain as Human Experience.
of Ethnography, ed. J Clifford, GE Mar- An Anthropological Perspective. Ber-
cus, pp. 98-122. Delhi, India: Oxford keley: Univ. Calif. Press
Univ. Press Guha R. 1983. ElementaryAspects of Peasant
Daniel EV. 1997. Charred Lullabies. Chap- Insurgencyin Colonial India. Delhi, India:
ters in an Anthropography of Violence. Oxford Univ. Press
Princeton:PrincetonUniv. Press HumphreyC, LaidlawJ. 1994. TheArchetypal
Das V. 1983. The language of sacrifice. Man Actions of Ritual. A Theoryof Ritual Illus-
18:445-62 trated by the Jain Act of Worship.Oxford:
Das V, ed. 1990a. Mirrors of Violence. Com- Clarendon
munities, Riots and Survivors in South Jeffery P, Jeffery R. 1996. Don't Marry Me to
Asia. Delhi, India:Oxford Univ. Press a Plowman! Women's Everyday Lives in
Das V. 1990b. Our work to cry: your work to Rural North India. Boulder/Oxford:West-
listen. See Das 1990a, pp. 345-99 view
Das V. 1990c. Voices of children. Daedalus. Joas H. 1996. The Creativityof Action. Transl.
Special issue onAnother India, pp. 48-65 J Gaines, P Keast. Chicago:Univ. Chicago
Das V. 1995a. Critical Events: An Anthropo- Press
logical Perspective on ContemporaryIn- Johnston P. 1993. Wittgenstein. Rethinking
dia, pp. 175-97. Delhi, India: Oxford the Inner. London/New York: Routledge
Univ. Press KleinmanA, Das V, Lock M, eds. 1997. Social
Das V. 1995b. Voice as birth of culture. Eth- Suffering.Berkeley: Univ. Calif. Press
nos 60(3-4):159-81 Kripke S. 1982. Wittgensteinon Rules and
Das V. 1996. Violence and the work of time. Private Language. London: Oxford Univ.
Presentedat Plenary Sess. Violence. Univ. Press
Edinburgh,Edinburgh,Scotland Langer L. 1991. Holocaust Testimonies. The
Das V. 1997. Languageand body: transactions Ruins of Memory.New Haven: Yale Univ.
in the constructionof pain. See Kleinman Press
et al 1997, pp. 67-93 Langer L. 1997. The alarmed vision: social
Das V. 1998. Official narratives,rumour,and suffering and holocaust atrocity. See
the social production of hate. Soc. Ident. Kleinman et al 1997, pp. 47-67
4(1):1-23 LawrenceP. 1995. Workof oracles. overconm-
Das V, Bajwa RS. 1994. Communityand vio- ing political silences in Mattakalapu.Pre-
lence in contemporary Punjab. In Puru- sented at SrilankanConf., 5th, Indiana
shartha (Special issue on Violences et Le Bon G. 1960. (1895). The Crowd: A Study
Non-Violences en Inde), ed. D Vidal, G of the Popular Mind. New York: Viking
Tarabout,E Mayer, 16:245-59 Levi-StraussC. 1969. (1964). TheRaw and the
Diamond C, ed. 1976. Wittgenstein'sLectures Cooked. Introductionto a Science of My-
on the Foundations of Mathematics, Cam- thology, Vol. 1. New York:Harper& Row
bridge 1939. Ithaca, NY: Cornell Univ. Lukes S. 1977. Some problemsabout rational-
Press ity. In Essays in Social Theory, pp.
Durkheim E. 1976. (1915). The Elementary 122-74. London:Macmillan
Forms of the Religious Life. London:Allen Lutz C, White GM. 1986. The anthropologyof
& Unwin emotions. Annu. Rev. Anthropol. 15:
Evans-PritchardEE. 1937. Witchcraft, Ora- 405-36
cles and Magic Among the Azande. Ox- Lutz CA, Abu-LughodL, eds. 1990. Language
ford: Clarendon and the Politics of Emotion, pp. 1-24.
Evans-PritchardEE. 1940. The Nuer. A De- Cambridge,UK: CambridgeUniv. Press
scription of the Modes of Livelihood and Needham R. 1972. Belief Language and Ex-
Political Institutions of a Nilotic People. perience. Oxford: Blackwell
Oxford:Clarendon Nieuwenhuys O. 1996. The paradox of child
Evans-PritchardEE. 1956. Nuer Religion. Ox- labor and anthropology. Annu. Rev. An-
ford: Clarendon thropol. 25:237-51

Reynolds P. 1995. 'Not known because not Problems with Womenand the Problems
looked for': ethnographerslistening to the with Societies in Melanesia. Berkeley:
young in SouthernAfrica. Ethnos 60(3-4): Univ. Calif. Press
159-81 Tambiah SJ. 1996. Leveling Crowds: Eth-
Rude G. 1959. The Crowdin the French Revo- nonationalist Conflictsand Collective Vio-
lution. Oxford: Clarendon lence in SouthAsia. Berkeley: Univ. Calif.
Rude G. 1964. The Crowd in History, Press
1730-1848. New York: Wiley Taussig M. 1992. Reification and the con-
Scheper-Hughes N. 1992. Death Without sciousness of the patient. In The Nervous
Weeping:The Violenceof EverydayLife in System,pp. 83-111. New York:Routledge
Brazil. Berkeley: Univ. Calif. Press Thompson EP. 1971. The moral economy of
Schulte J. 1993. Experience and Expression: the English crowd in the eighteenth cen-
Wittgenstein'sPhilosophy of Psychology. tury. Past Present 50:76-126
Oxford: Clarendon Tylor EB. 1974. (1878). Primitive Culture.
Schutz A. 1970. On Phenomenology and So- New York: Gordon
cial Relations: Selected Writings,ed. HR Williams M. 1996. UnnaturalDoubts. Prince-
Wagner. Chicago: Univ. Chicago Press ton: PrincetonUniv. Press
Scott JC. 1985. Weapons of the Weak. New Wittgenstein L. 1953. Philosophical Investi-
Haven: Yale Univ. Press gations. New York: Macmillan
Scott JC. 1990. Domination andthe Arts ofRe- Wittgenstein L. 1958. The Blue and Brown
sistance. Hidden Transcripts.New Haven: Books. Oxford: Blackwell
Yale Univ. Press Wittgenstein L. 1980. Culture and Value, ed.
Searle JR. 1969. SpeechActs. Cambridge,UK: GH von Wright. Transl.P Winch. Oxford:
CambridgeUniv. Press Blackwell
Strather M. 1988. The Gender of the Gift: