The Soul's Body and Its States: An Amazonian Perspective on the Nature of Being

Human
Author(s): Anne Christine Taylor
Source: The Journal of the Royal Anthropological Institute , Jun., 1996, Vol. 2, No. 2
(Jun., 1996), pp. 201-215
Published by: Royal Anthropological Institute of Great Britain and Ireland

Stable URL: http://www.jstor.com/stable/3034092

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.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Wiley and Royal Anthropological Institute of Great Britain and Ireland are collaborating with
JSTOR to digitize, preserve and extend access to The Journal of the Royal Anthropological
Institute

This content downloaded from

186.195.42.1 on Fri, 10 Jul 2020 04:26:02 UTC
All use subject to https://about.jstor.org/terms
 THE SOULS BODY AND ITS STATES:
AN AMAZONIAN PERSPECTIVE ON

THE NATURE OF BEING HUMAN*

ANNE CHRISTINE TAYLOR

CNRS, Paris

This article begins by exploring some of the premisses concerning personhood, sociality and
mortality underlying the experience and representation ofselfinJivaroan Achuar culture. Although
inexplicit and seemingly contradictory, these assumptions combine to produce an intricate though
unspoken theory of what is implied in being a true, live human. Jivaroan sense of self is rooted
in the progressive fusion of a generic, given bodily form and of attributed perception of this same
bodily form; the initially anonymous body image is thus progressively singularized by the memory
of the affective moods experienced in daily social interaction. Achuar selfhood is therefore
susceptible to states ofweakness and uncertainty, categorized as induced illness, as well as to states
of enhancement brought on by communication with a certain category of spirits. The interactive
basis of the set of representations concerning selfhood leads the author to discuss traditional
anthropological ways of dealing with indigenous ideas, and to suggest an approach more attentive
to the contextualization of knowledge.

Among the many 'things that go without saying'1 is how very honoured I feel
by your invitation to deliver this lecture. Honoured, but also, in equal if not
greater measure, horrified. First, because the occasion evokes a host of father
figures, dead or alive, on whom we have all sharpened our teeth in our youth,
and whose benevolence is therefore open to doubt; secondly, because I am a
French anthropologist, and in the skirmishing which is one of the great pleas-
ures of the long relationship between our two scientific communities, I find the
stakes have suddenly been drastically raised. Thirdly, because picking one's way
through the minefield of conflicting paradigms that is today's anthropology is
an inherently stressful affair.
The problem I want to consider here, one of central importance for Malinowski,
is that of the relation between social environment and individual psychology.
Indeed, this lecture is a kind of response to the call he made in the introductory
chapter of Argonauts to study 'what concerns man most intimately, that is the
hold which life has on him', in so far as I shall attempt here to define what is
implied for an individual in being alive and experiencing the selfhood of a socially
constructed body in an Amazonian culture. Much research has been devoted
recently to issues of this sort, so much so that the body has, to a large extent,
come to replace society as our discipline's major focus of analytic exploration;

* Malinowski Memorial Lecture 1993

J. Roy. anthrop. Inst. (N.S.) 2, 201-215

This content downloaded from

186.195.42.1 on Fri, 10 Jul 2020 04:26:02 UTC
All use subject to https://about.jstor.org/terms
202 ANNE CHRISTINE TAYLOR

in fact we have been taught, most

society is nowhere if not in the b
involved in constructing and deconstructing it. This perspective deals effec-
tively with the many difficulties raised by older sociological reifications, but it is
not easy to reconcile it with any plausible view of selfhood, in so far as it is hard
to imagine that people actually experience themselves purely as a succession of
structured concatenations of bits and pieces, and it is even harder, in the ab-
sence of a minimally stable subjectivity, to account for the relative continuity of
tradition. Thus, while taking into account the importance of this work, my aim
is to show how a body image is experienced subjectively, and how a person is
thereby capable of reproducing the social structures that have shaped his or her
selfhiood. This is, admittedly, an absurdly broad question, but its very magni-
tude invites the kind of hit-and-run approach ideally suited to the format of a
short lecture. As Nietzsche usefully put it, great issues are like cold baths: one
must get in and out of them as quickly as possible.
I shall take as my starting point a minor paradox implicit in the ethnography
of the indigenous cultures of lowland South America. On the one hand, an-
thropological accounts of these groups are replete with statements to the effect
that Amazonian Indians do not believe that death can be caused by natural
causes; rather, they view it as due to malignant human agency.2 In such a
perspective, death exists only as a form of homicide, whether overt or clandes-
tine. This conception of mortality is thought to lie at the heart of two highly
important types of social practice: namely, shamanism and various forms of
institutionalized feuding and hostility, sometimes between domestic groups,
sometimes between segments of tribes, sometimes even between tribes or eth-
nic entities. On the other hand, in all these groups we invariably come across
one or several myths about the origin of mortality, in which death seems to be
viewed in a far more 'naturalistic' light; that is, as a feature of the way the world
is and, in short, as an inescapable fact of life.3 This second point of view seems
to imply that Amazonians do, after all, seem to conceive death as some sort of
universal natural phenomenon. The obvious 'rationality' of this belief, from our
point of view, entices many anthropologists into a familiar kind of spontaneous
functionalism which leads them to explain away the vindictive view of mortality
as an ideological device necessary for the continuity of major sociological insti-
tutions, the underlying assumption being that the Indians do not believe in it as
they believe in the naturalistic view of death. After all, if people are destined to
die anyway, it seems unnecessarily redundant to assume that they are always
being murdered; thus, the homicidal approach to mortality tends to be dealt
with as if it were merely a cunning sociological artifice.
This sort of approach is surely unacceptable, as indeed any anthropologist
who stops to think about it immediately recognizes, if only because 'ideology' is
obviously experienced as truth and not as false consciousness. Furthermore, on
empirical grounds, the Indians appear to hold the homicidal view much more
strongly than the naturalistic one. This, however, leaves us with the problem of
accounting for the coexistence of two seemingly contradictory sets of beliefs. It
can of course be argued, and often is nowadays, that the requirement of logical
coherence in the representations particular to a given society is no more than an
anthropological prejudice, and that accounts which present culture as a body of

This content downloaded from

186.195.42.1 on Fri, 10 Jul 2020 04:26:02 UTC
All use subject to https://about.jstor.org/terms
 ANNE CHRISTINE TAYLOR 203

self-contained, logically flawless implicit metaphysics bear absolutely no resem-

blance to the way people actually think and act. In fact, given the ubiquity of
those processes of 'creolization' which many anthropologists stress in their
studies, as well as the compartmentalization of mental processes that cognitivist
psychologists argue for, we should allow for, and indeed expect, contradiction.
Let me state at the outset that, in general terms, I entirely endorse the 'cogni-
tivist' critiques of traditional anthropological views of culture, such as those
developed by Boyer (1990; 1993) or Bloch (1991; 1992; 1993) in their recent
work. It is perfectly true that no people actually think the way anthropologists
seem to think they do, and that people do not refer to a mental blueprint of
their culture, as one would to a text, in order to produce non-contradictory
statements and practices. At the same time, I strongly oppose the view that there
is no durable systematicity in culture beyond that which is produced by anthro-
pological accounts, or beyond that which is locally created by the dialectics of
ethnicity. The fact that culture rests to a large extent on 'that which goes with-
out saying' does not mean that anything goes, and that there is not some degree
of integration between compartmentalized mental models. To assert this is
sheer common sense. Yet the source and precise nature of this relative coher-
ence is not easy to define and account for.4 My own feeling is that it must lie to
some extent in a certain property of circularity inherent in the mental models
shared by people of any given culture; in the fact, in other words, that these
self-evident, discursively unelaborated conceptual clumps, which we might, for
convenience's sake, call premisses, must refer back to each other, and that it is
precisely from this circular process of mutual referral that they gain their quality
of obviousness. In what follows, I will attempt to show how a unique, precise
and intricate definition of person and selfhood emerges from just such chains of
circular unelaborated notions, and to understand how a complex idea comes to
be developed while no one actually thinks and expresses it. And this, of course,
raises the problem of what anthropological accounts can and should describe,
an issue to which I will return later.
The Jivaroan Achuar of lowland eastern Ecuador5 offer a perfect exemplar of
the implicit contradiction I mentioned above. In their view, illness and death
are, invariably, the outcome of an act inspired by deliberate homicidal intentional-
ity, operationalized through recourse to the invisible machinery of bewitchment
caused by invisible blow darts. Furthermore, they make no sharp distinction,
even lexically, between illness and death, the difference between the two states
being a matter of degree rather than of kind. This implies that both are viewed
as a process, linked by a series of metamorphoses, rather than as ontologically
distinct conditions. Moreover, for them illness is a single phenomenon, what-
ever its symptoms, whether psychic or somatic; there are no specific illnesses,
just undifferentiated suffering.
Like most other Amazonians, however, the Achuar also tell a characteristically
brief and concise myth explaining how mortality came to be as a result of an
accidental and entirely trivial act of disobedience.6 I have no intention of elabo =
rating on this myth, which ends with the lapidary statement 'let there now be
mortality', beyond underscoring two of its important features. First, this myth
posits a massive and abrupt change from a time of undifferentiation when there
was 'only life', to a time when there was life and death; that is to say, life as we

This content downloaded from

186.195.42.1 on Fri, 10 Jul 2020 04:26:02 UTC
All use subject to https://about.jstor.org/terms
 204 ANNE CHRISTINE TAYLOR

know it. But the myth says nothin

it. In other words, it focuses on a p
the relation between the terms. Second, by conceptualizing the origin of finite-
ness as the outcome of a trivial deed of transgression, the myth establishes a
huge, indeed a monstrous disproportion between cause and effect, between an
act and its consequence. This property, common to many mythic accounts,
might best be explained by assuming that certain kinds of myth are in fact
anti-causal propositions: in other words, they do not, as used to be thought (not
least by Malinowski), justify the world and explain how it came into being or
how one should behave; rather, they describe the world as it is in a highly
problematic way, and thus make the obvious paradoxical. This is precisely why
no one 'believes' in what myths say in the way one 'believes' in the portrayal of
a spirit. If this is so, what distinguishes the two perspectives on mortality is not a
matter of content but of kinds of discourse: one of them - the homicidal view -
presents death as a gradual process, the extremes of which remain undefined, and
its emphasis lies on exploring the nature of the processual relationship between
the unmarked poles; the other - the mythic and naturalistic one - presents
sharply defined terms in a paradoxical and therefore undefined relationship.
Thus, what appeared initially as a contradiction between contents, that is to say
between two different conceptions of death, may now be viewed as a probably
necessary articulation between two distinct types of representation7 of a relation.
To understand the nature of this articulation, we must begin by taking a closer
look at mortality as a mode of processual relationship; and this, in turn, entails
understanding what being alive is in Jivaroan terms. At a certain level, this is a
simple matter: being alive is to be perceived, and to perceive oneself, as a
person, a notion locally covered by the term shuar. This expression refers t
multi-layered set of relations between contrastive terms: thus, according to
context, the term shuar refers to 'my bilateral kindred' as opposed to others, 'my
local group' as opposed to other territorial groups, Achuar' as opposed to other
Jivaroan tribal units, 'Jivaro' as opposed to Whites or other Indians, and so on.
In short, the term functions as a generalized 'we/they' classifier.
For our purposes, the interesting thing about this classifier is that, in given
contexts, the 'we' it defines include two classes of imaginary beings. One of
these refers to spirits characterized by their normal human appearance and their
entirely non-human and, indeed, inhuman, behaviour: they are solitary, blind,
do not eat, and exist in a state of permanent generalized desire. The other class
includes spirits with a non-human appearance who nevertheless behave like
human beings, in so far as they use language and signs, follow moral rules and
are credited with human emotions. In fact, a large portion ofJivaroan mythol-
ogy is devoted to the description of these latter creatures' acts, and entities or
shuar of this sort also figure prominently in shamanic discourse and practice as
well as in garden and hunting magic. However, if these imaginary beings can in
certain circumstances be defined as persons, that is as part of the 'we' group, in
other contexts they are quite different from real persons, for a number of
reasons: sometimes because they are dead, and sometimes because they are
believed to be like humans only under certain conditions of interaction. What
they do share with real live humans, however, is consciousness and intentionality.
These are properties which for the Achuar are not limited to specific kinds of

This content downloaded from

186.195.42.1 on Fri, 10 Jul 2020 04:26:02 UTC
All use subject to https://about.jstor.org/terms
 ANNE CHRISTINE TAYLOR 205

being, but attributable in given contexts to many different sorts of things, in-
cluding inanimate ones; life, in short, is a postulated state of mind rather than a
state of matter.
It follows from this that being a real, live human implies displaying a special
type of bodily appearance, practising certain types of communicative and social
behaviour, and possessing certain states of consciousness. In order to specify
this combination which defines real, live humanness, we may begin by taking a
closer look at bodily appearance and explore some Achuar notions concerning
the body. The salient features of the mental model that shapes these ideas are as
follows. First, and most surprisingly, the Achuar have remarkably unelaborated
theories of procreation, and they have in fact very little to say about the concep-
tion and formation of a child; questions about these matters clearly strike them
as irrelevant. Further, pregnancy and birth are not ritualized and there are no
myths explicitly concerned with conception and procreation.8 Secondly, if we
examine prohibitions and observances linked to bodily substances and functions
- a set of practices often considered of particular importance for understanding
indigenous ideas about the body's formation and the shaping of the person -
some interesting properties begin to emerge. Prohibitions are most numerous
and stringently observed precisely in those practices and situations which, from
the indigenous point of view, involve a process of transformation: making a
canoe, preparing curare poison, suffering from snake-bite. Thus, the relative
scarcity and laxity of prohibitions attendant on pregnancy and childbirth sup-
port the view that these processes are not seen as similar to other culturally
stressed metamorphoses. In other words, and contrary to death as viewed in
certain contexts, birth is never thought to be a process of transformation, and
there is therefore no parallelism between entry to, and exit from, the state of
being a real live human. It should also be stressed that, with regard to their
effects, bodily substances do not form a class separate from other, non-bodily,
substances: semen, for example has the same kind of properties as curare, as
snake-poison, or as the burning sensation of red pepper; and menstrual blood is
just blood, or if it has any power this comes from non-specific attributes such
as the potency of red, or that of being heavy. Finally, things posited as alive, that
is to say credited with intentionality and consciousness, are all fundamentally
the same in terms of organic attributes and physiological mechanisms: a bat or
a dog, or for that matter a manioc plant, are all believed to be organized in the
same way. They function according to identical biological processes, and their
bodily stuff - appearance apart - is the same. If we humans are not normally
aware of this fact it is for epistemological reasons - because we do not ordinarily
communicate with them - and not because these metabolisms are ontologically
distinct.
We are thus led to the conclusion that what differentiates species is essentially
shape or, more accurately, appearance. As I have shown elsewhere (Taylor 1993),
from a Jivaroan point of view perceived shape refers to a set of differentiated
bodily forms, particularly faces, specific to each class of animate beings. These
outward forms exist in limited numbers and are endlessly recycled, which ex-
plains why there is no natural creation, and why birth is not viewed as a process
of transformation or 'making' that adds something new to the world. Birth is
reappearance, and the Achuar person thus comes ready-made in terms of bodily

This content downloaded from

186.195.42.1 on Fri, 10 Jul 2020 04:26:02 UTC
All use subject to https://about.jstor.org/terms
206 ANNE CHRISTINE TAYLOR

features. It follows from this that

rooted in a sense of singularity of form, and not, as Western post-Freudian
notions would have us believe, in a gradual realization of bodily integrity and
autonomy, since Jivaroan bodies do not possess organic specificity. Yet it would
fly in the face of common sense to assume that the Achuar actually experience
themselves as purely generic singularities. Impersonal particularized shape may
give the self wholeness, but it obviously cannot give it subjectivity.
Subjectivity, however, is primarily a matter of refraction: it takes its source in
the sense one has of others' perceptions of self And that is where we must look
for a solution to the conundrum I have just evoked - of how the experience of
an impersonal, externally created body can be experienced subjectively I believe
such a solution is to be found in the web of notions pertaining to affect and
memory, if by memory we mean the mental image we form of people or things.
As we all know from reading Gow's fine book (1991), memory, for Amazonian
peoples, is intimately linked to kinship. Indeed, in some sense it is kinship itself
Social relations, in this view, are the condensation and memory of the affective
moods built up by daily interaction in nurturing, sharing and working. The
personal mental image one has of others as kin is moulded by this web of
feeling; therefore the image of self, in so far as it is based on the attribution of
others' images of it, is necessarily suffused with the memory others have of you.
It is precisely a representation of this generically singular, yet uniquely individu-
alized, image of the person that is denoted by the vernacular expression wakan,
a term usually translated as 'soul' and which in fact refers to the reflected image
of a thing, the appearance of someone in a dream as well as the dreamer's
consciousness. Above all, it refers to the ghost of a recently deceased person,
that is to say a mutilated memory, in so far as it is a substantivization of the
intersubjectivity which was once fused with the image of a bodily shape that no
longer exists. The constitutive relation between subjectivity and interpersonal
ties allows us to understand why vision, language and, more generally, commu-
nication are such a vital axis in defining selffiood, since the refracted image is to
a large extent an implicit and even, at times, an explicit description of the person.
The Achuar themselves are acutely aware of this fact, as may be inferred from,
amongst other evidence, the structure of their magical love songs. These are
invariably verbal descriptions, addressed to the beloved, of the state he or she is
experiencing on seeing, or rather feeling, a magically induced image of the
sender.9 In sum, the Achuar would certainly endorse Wittgenstein's claim that
the body is the best image we may have of the soul, not least for its reversibility,
since it is equally obvious to them that the soul is also the best image we may
have of the body as a generic personalized form.
Sociality as inherently affective memory has been described with great finesse
and insight by what we, in Paris, call the English school of Americanism.
Nevertheless, our colleagues' accounts have often struck me as somewhat one-
sided and indeed surprisingly angelic, for they tend to minimize a vital
component of social relations, namely hostility or vindictiveness. And while the
English are very good on love, I think we French have the advantage of them in
the matter of hate.10 Hostility is a particularly important aspect of social re
tions and of the psychological configurations inherent to them, above all in a
society such as that of the Jivaroans, which is structured by endemic feuding

This content downloaded from

186.195.42.1 on Fri, 10 Jul 2020 04:26:02 UTC
All use subject to https://about.jstor.org/terms
 ANNE CHRISTINE TAYLOR 207

and interfribal warfare. Learning to

of the social environment, is just as important to them as growing to love. And
of course hostility also feeds into the experience of self; it colours, just as much
as love does, the texture of the body image as singularized appearance which, as
we have seen, lies at the heart ofJivaroan selfhood.
If selffiood as person is a state, it is also by nature a highly unstable one, in so
far as one's inner landscape is shaped by the understanding one has of others'
perceptions of oneself The integrity of one's feeling of self is vulnerable on two
counts. First, it is exposed to the death of others, the shattering of bits of that
mirror on which it is dependent - an all too frequent occurrence in the course
of an Achuar life, and one that provokes, as its first reaction, intense socially
directed anger. Second, it is exposed to the chronic instability of relations in a
society based on loosely-knit kindred groups, the frontiers of which constantly
vary in the flux of endemic feuding and shifting alliances. Traditional life among
the Jivaro thus breeds a kind of rational, latent paranoia, since members of one's
closest family may become suspected of treachery in periods of open conflict.
Suspicion falls primarily on one's affines, of course, but it may include even
same-sex consanguines, such as brothers or sons.11 This pervasive uncertainty
as to the real nature of others' feelings for oneself cannot fail to have conse-
quences for the texture and foundations of selfhood.12 This, I would claim, is
where illness comes in. Sickness, in Jivaroan terms, is the suffering experienced
by individuals when they become overwhelmed by the ambiguity of the social
environment and thereby lose a clear sense of their identity; that is, when their
perception of self is clouded by uncertainty. In fact, I would surmise that it is
the high level of anxiety generated by the extreme unpredictability of social
relations inherent in Jivaroan existence that explains why any affliction, regard-
less of its origin and apparently benign character, turns into a symptom of
bewitchment if it lasts for more than a few days or even hours, and why the
relatively detailed taxonomy of pathologies that Achuar informants develop in
the abstract so rapidly breaks down into a single, massive contrast between
'health' and undifferentiated suffering; that is to say, dying.13
It is this breakdonv of clarity in perception of self, experienced as suffering
and conceptualized as an intangible homicidal onslaught, that justifies recourse
to shamanic therapeutic practice. As Severi (1993) has demonstrated for the
Kuna healing of madness, shamanic therapy is a form of cure predicated on the
build-up of a complex pragmatic structure wherein the shaman produces a
description, generally incomprehensible to the patient, of his communicative
interaction with foreign spirits. The shaman thereby creates an analogue of the
patient's state of confusion, with the crucial difference that he, the shaman,
masters this chaotic world by sight and word, and communicates with its deni-
zens through the medium of his familiar spirits, whereas the patient, for his
part, is trapped in a state of communicative breakdown, both with himself and
with others. Through this transmutation of dumb, internal disorder into clari-
fied, structured, explicit though incomprehensible alterity, people are eventually
restored to a normal state of self-awareness; that is, one in which self and the
world can coexist at acceptable levels of ambiguity.
There is, however, another Jivaroan way of dealing with the fragility of self-
hood. This is by resorting to a ritual experience which is in many ways a mirror

This content downloaded from

186.195.42.1 on Fri, 10 Jul 2020 04:26:02 UTC
All use subject to https://about.jstor.org/terms
208 ANNE CHRISTINE TAYLOR

image of the shamanic cure. I refer here to the quest for so-called arutam
visions, whereby a person may, in the course of a private ritual involving isola-
tion, rigorous fasting and the ingestion of large doses of hallucinogenic drugs,
receive a message or vision relative to his or her future existence. The spirit
responsible for this prophecy, the arutam or 'ancient thing', takes the shape of a
dead Jivaro, who, after a complicated and frightening series of metamorphoses,
briefly appears in person to the seeker and addresses him or her. In the case of
men, this message usually concerns the outcome of an act of war or a revenge
killing, which they then, of course, feel compelled to carry out. The arutam
experience is thus directly linked to those situations and interpersonal relation-
ships most heavily fraught with unpredictability, and it rests on the same logic
that underlies the resort to shamanic healing. I need not add that the pragmatic
structure of the interaction between seeker and spirit is just as complex as that
implied in shamanic curing, indeed in many ways more so (I refer here to the
way in which the stereotyped circumstances surrounding the ritual encounter
are constitutive of the event's meaning). I will therefore limit myself to under-
scoring two salient features of the arutam quest.
First, it centres on the ritual framing of the normal interactions on which
subjectivity is built, as perceived by the Jivaroans. Thus, when the Achuar speak
of the arutam's message as a kind of 'soul' which will become henceforth a part
of themselves, they are evoking a reification, projected into the future, of an
image of self rooted in a special kind of intersubjective relation, that between
themselves - more accurately a modified state of their consciousness - and the
arutam. This hypostasis is modelled on the introjection of an attributed image
of self that underlies normal states of subjectivity; thus, just as the wakan - the
body's soul - briefly survives the recently deceased as a substantivization of the
memory that surviving kin hold of it, so the arutam vision he or she has received
encapsulates the spirit's description, or image, of their future selves. The wakan,
in short, is a reification of attributed memory, while the arutam 'soul' is a reifi-
cation of projected selfhiood. Paradoxically, this is in fact all that ultimately
remains of people, in the guise of the 'ancient thing' or arutam they in turn
eventually become. In other words, the arutam ritual is not linked to an elabo-
rate cosmology or ontological theory. It is based on the same perception of
subjectivity and intersubjectivity that informs the notion of wakan, and its spe-
cific meanings are rooted in the ritual construction of a particular context of
interaction, rather than on an elaboration of content. In the second place, the
effect or result of the arutam quest springs from an event of hypercommunica-
tion, a kind of saturation of certainty and unambiguous meaning. Yet this
glimpse of a destiny cleansed of unpredictability must remain indescribable, in
so far as it is strictly prohibited to speak about the message received from the
arutam. Were one to do so, one would immediately lose the benefit of the vision,
indeed the vision or message itself as a kind of soul-stuff whereby the sense of
self is fortified. And that, after all, is the prime motivation for undergoing the
mystical experience, as well as its final outcome: the acquisition of invulnerabil-
ity, made manifest by forcefulness of speech and manner, face painting of a
certain kind, and heightened anger; that is to say, intensified homicidal drive.14
In sum, just as illness leads to a loss of the capacity to communicate except
through the mute language of symptoms, the state of super well-being brought

This content downloaded from

186.195.42.1 on Fri, 10 Jul 2020 04:26:02 UTC
All use subject to https://about.jstor.org/terms
 ANNE CHRISTINE TAYLOR 209

on by the encounter with an arutam implies a suspension of linguistic exchange

and the manifestation of an array of indirect signs or 'symptoms' expressing a
condition of enhanced selfhood.
Thus, if we compare the two kinds of ritual experience I have briefly evoked,
the shamanic cure and the arutam quest, we have on the one hand a mediated
description of mastered chaos through which an individual may rid himself of
the suffering occasioned by overwhelming ambiguity and the erosion of self-
hood it entails, and on the other hand a mediated secret vision or 'hearing' of
absolute certainty, which dramatically intensifies the force of selfhood. I speak
advisedly of mediation in the case of the arutam quest because, even though the
seeker is alone during the ritual, it is not the seeker himself who receives the
message of the apparition but rather his wakan, his body's soul, the whole point
of the drugs and fasting being to induce a state of dissociated consciousness. So,
in fact, the structure underlying the situation of communication in both cases is
similar. In the curing session, the patient is 'as dead', entirely passive while the
shaman gradually identifies himself with the components of disorder, figured as
foreign creatures and languages, and so orders them as he describes them. In the
arutam ritual, the seeker is also figuratively dead (during his period of isolation
his kin must avoid evoking his image, just as they would avoid thinking about a
recently deceased person), so that his 'soul', his de-intentionalized conscious-
ness, may gradually become similar to the truly dead person who will eventually
appear to him; and this interaction with an entity which is structurally just as
'outside' society as are the foreigners encountered by the shaman begets abso-
lute clarity, not just a healing - which is really a return to the self - but rather
an enhancement of the self or state of hyper-selfhood.
Let me now sum up this sketchy description of Jivaroan notions about the
person and the experience of self they imply I began by showing that the person
is defined negatively, 'en creux', as it were, by the intersection of a certain
number of inexplicit assumptions concerning animation (viewed as imputed
subjectivity), sociality (viewed as ordered communication and therefore implicit
in the indigenous notion of animation), and finally shape, ordered surface, that
principle of speciation dividing an otherwise generalized physiological model or
matter. I went on to show that the surface proper to humans as a class of beings
- that is to say, their appearance - is conceived of as being drawn from a finite
stock of recyclable distinctive forms. Hence, the Jivaroan sense of self is predi-
cated on the fusion of a singular though generic body image and other people's
emotionally-laden perception of this body image, whereby it comes to be expe-
rienced as uniquely personal. I then argued that because selffiood is textured by
intersubjectivity, because intersubjectivity is itself created in the context of social
relations, and because social relations among the Jivaro involve shifting kin ties
and institutionalized forms of reciprocal violence - because, in short, they live
in a world permeated by uncertainty and hostility - the Achuar's sense of self is
highly vulnerable. It therefore alternates between states of uncertainty, erosion
and breakdown, on the one hand, experienced as murderous, undifferentiated
suffering which calls for shamanic healing, and, on the other hand, states of
enhancement brought on by a mystical experience of certainty.
This leads me to the heart of my argument. In analytic terms, a person or self
is not a thing, a specific essence shaped by an explicit or implicit theory, or in

This content downloaded from

186.195.42.1 on Fri, 10 Jul 2020 04:26:02 UTC
All use subject to https://about.jstor.org/terms
 210 ANNE CHRISTINE TAYLOR

other words, an indigenous concept. Being a live human person is not a state
defined as such - there is no canonical discourse about 'the person', and nobody
will ever state 'this is our idea of what a man or a woman is' - yet it is nonetheless
precisely circumscribed by the articulation of a set of non-explicit premisses.
Being a person is thus an array or cline of relational configurations, a set of links
in a chain of metamorphoses simultaneously open and bounded. The chain is
open because death itself is an endless process, as is the shift from 'we' to 'they',
from Jivaro to foreigner; it is bounded, nevertheless, because being a live self
can be defined only by contrast to either a state of being less than alive, in
illness, or a state of being more than alive, through the acquisition of arutam.
And this is why we find both the stark problematic oppositions set up in mythic
narratives - life versus death - and, in ritual and in other fields of practice, life
and death as a continuous process. This is the key to the paradox I started out
with, concerning the coexistence of two seemingly contradictory notions of
death. These are not, in fact, two distinct compartmentalized conceptions of
mortality, one naturalistic, the other persecutory, but rather two mutually impli-
cated perspectives, one focusing on the terms rather than on their relation, the
other focusing on the relation while bracketing the terms. Further, the actuali-
zation of the different occasions in which a notion of the self is evoked forms a
chain: the arutam quest leads to killing, aggression and suspicion, which is the
cause of sickness and disorientation, which in turn requires either shamanic
healing or further arutam quests, and so on. This means that the different types
of relationships and self-creations discussed here are not only related structur-
ally, but also practically
The approach to the question of personhood outlined above has some wider
implications which I would, by way of conclusion, draw to your attention. Most
Jivarologists, myself included, have tended to consider the arutam complex,
because it is intellectually spectacular, and also because of its esoteric aspects, as
the heart of Jivaroan culture, the very basis of its identity for Indians and
ethnographers alike. And yet, while it is true that until recently the vast majority
of Achuar men had experienced arutam encounters, it is equally true that these
mystical quests only concern an extreme state of Jivaroan personhood, not
'Jivaroanness' as such, and most informants would state that one could live a
normal Jivaroan life, and be a Jivaro (though admittedly a second-rate one)
without ever experiencing an arutam encounter. Thus, by positing that the aru-
tam complex lies at the heart, rather than on the boundaries, of Achuar culture,
we have misrepresented not only the arutam complex itself but Jivaroan culture
in general. Nor is this optical skewing a particularity ofJivarology. I suspect that
a great many of our ethnographic accounts are in fact based on a similar confla-
tion between 'culture' and 'extreme states'. For example, any account that
purports to characterize the symbolic culture of this or that society by general-
izing conceptions deduced from any kind of specialized discourse can only be
wrong: wrong because it is blind to the importance of practice and contextuali-
zation, and assumes that culture is a system of language and thought shared,
with due allowance for sex and generation, by all. Semantic premisses may
indeed be shared by all, but pragmatics most certainly are not: in so far as the
conditions of use determine how these premisses are elaborated, expressed and
experienced, it is surely unsustainable to claim that a given representation is

This content downloaded from

186.195.42.1 on Fri, 10 Jul 2020 04:26:02 UTC
All use subject to https://about.jstor.org/terms
 ANNE CHRISTINE TAYLOR 211

common to 'the society' if its meaning is in fact rooted in situations of interac-

tion and forms of contextualization which are not at all collective.
My second point derives in a sense from the former. If my account of the
Achuar person and selfhood as a repertoire of different states of being has more
than local validity, then we will have to review our approach to the problem of
acculturation. In particular, we must cease to think of acculturation as a gradual
erosion and consequent reworking of the central beliefs of a culture. Among the
Achuar, acculturation has in fact always been there, at the very core or 'middle
ground' of the cultural system: it is not a matter of loss so much as the feeling
of being no longer compelled to define the self by experiencing the whole range
of states that it normally implies. Acculturation begins in a condition of being
locked into a state of undefined or unmarked normality by no longer engaging
in the situations of interaction characteristic of the extreme states; thus an
acculturated, or potentially acculturated, Jivaro is simply an ordinary being,
what the Achuar themselves aptly call a nangami shuar, a 'just-so person', the
kind of individual who can move in and out of his and other cultures with ease,
provided he remains in their middle grounds or 'zero states'.
At no point in my account have I claimed to present an indigenous theory, or
even to spell out a set of implicit or unconscious collective representations
about the person, about life or about death; I have simply followed a few of the
links between clumps of 'things that go without saying'. Yet even my superficial
presentation of this chain of relations has, I trust, been sufficient to evoke the
stylistic uniqueness of Jivaroan culture. Taken separately, bits of the conceptual
schemes on which it is founded can be found floating all across Amazonia;
equally, the cognitive mechanisms and, more specifically, the logical structures
underlying the pragmatics of Achuar culture can also be found all over Ama-
zonia, and indeed far beyond. The particularity of Achuar culture must
therefore be rooted in the establishment of a certain kind of logical circularit
between these cognitive and notional bits and pieces. This is a commonsensical
view, and hence a very lame conclusion, but it need not be for all that a down-
beat one. It allows us, in particular, to circumvent that situation akin to the
physicists' uncertainty principle, whereby if we concentrate on cognitive
mechanisms we miss out on culture, and if we describe culture in the tradi-
tional manner we fly in the face of what cognitive research has taught us about
the workings of the mind. Cognitive mechanisms as such are not our province
as anthropologists, first because, ultimately, they are probably invariant, and
secondly because the experimental protocols now being used in cognitive psy-
chology are virtually impossible to reproduce in normal field conditions. But
then neither is culture as we used to see it. Obviously we cannot go on saying
'the Achuar think that ...', because the content of thought we then usually go
on to describe is not in fact what they actually think. And yet we must, in some
sense, go on saying precisely that, and this is really the good news. For if our
aim is to gain, and share with our readers, an insight into the thoughts and
experiences of people whose assumptions and lifestyles are different from ours,
we can only do so, surely, by spelling out and explicating the large part of
culture that 'goes without saying', that escapes indigenous conceptualization
because it is embedded in, and acquired through, practice rather than discourse.
Certainly, the version of culture we are thus led to produce does not in any

This content downloaded from

186.195.42.1 on Fri, 10 Jul 2020 04:26:02 UTC
All use subject to https://about.jstor.org/terms
 212 ANNE CHRISTINE TAYLOR

sense mirror our informants' experiential and mental universe, but I do not see
how we could even imagine this universe unless we invent for ourselves and
our readers a kind of discursive 'stand-in' for a culture we have not been
socialized in, and thereby give ourselves the means of empathizing to some
degree with the lives and thoughts of the people we are studying. We must, in
other words, treat the net of often inexplicit and unelaborate assumptions that
is constitutive of culture as tf it were a metaphysics, because this fictional con-
struct is of necessity our essential, indeed our only, procedure for bringing to
light and verifying the necessary circularity of the combination of premisses
found in any given culture, as well as our only way of allowing for the exercise
of that 'analogical introspection' that lies at the heart of our discipline. And this
means that we can go on happily working away at our socio-cosmological
monographs, provided we respect two sets of conditions. First, we have to pay
much closer attention than previously to the contextual aspects of discourse and
communication, cease assuming that culture is a collective text, and be more
realistic in our description of who thinks what and how in which circum-
stances. Second, we should accept that our ethnographic accounts are elaborate
thought experiments rather than accurate renditions of indigenous systems of
thought, and view them as a conceptual tool inherent to the practice and writ-
ing of anthropology. Cumbersome this tool may be; yet if we take into account
the complexity of the phenomena that, by common consent, anthropology is
supposed to deal with, our monographs must rate as highly economical instru-
ments (in every sense of the term) and the disproportion between our means -
an elaborate controlled fiction with no pretence at realistic description of men-
tal phenomena - and our ends - the discovery and verification of the nature of
the links between mental models - is not as large as might seem at first glance.
The shift of perspective I have suggested is not much to ask for, and it implies
no major overhaul of object or method. Yet it does seem to offer anthropology
some kind of future as a worthwhile intellectual pursuit, which is surely more
than can be said either for the bland objectivism we have long practised, or for the
self-destruct vehicle devised by the more zealous militants of post-modernism.
Moreover, the sort of minor displacement of point of view I am advocating is
entirely typical of our discipline's intellectual tradition; indeed the occurrence
of such slight and often unnoticed shifts is precisely what makes anthropology
appear to outsiders as the endless rehashing of the same problems, while its
practitioners feel it is constantly progressing. Thus, I would like to think that
the banality of my conclusion is proof of sorts that the approach I am arguing
for is not entirely alien to that great canon so signally shaped and illustrated by
the scientist whose memory we honour today.

NOTES

I The expression refers to a paper by M. Bloch published in Kuper 1992, under the title
'What goes without saying. The conceptualization of Zafimaniry society'. Throughout this lec-
ture, I will often refer implicitly to the hypotheses presented in that contribution. I would like
to thank Bloch for his illuminating comments on an earlier draft of this work, and for his help
in preparing this paper for publication.
2 This is a crude simplification. It would be more accurate to say that the intentiqnally evil
agency that causes death is usually anthropomorphized, but not necessarily human. For illustra-
tions of the view of strictly human agency, see for example Capistrano de Abreu (1941: 140-1)

This content downloaded from

186.195.42.1 on Fri, 10 Jul 2020 04:26:02 UTC
All use subject to https://about.jstor.org/terms
 ANNE CHRISTINE TAYLOR 213

on the Cashinahua, and Harner (1972: 152-3) and Descola (1994: 257-70) on the Jivaroan
Shuar and Achuar. Among the Piaroa, the agent may be a divinity, an animal or a foreign sor-
cerer (Overing 1985), among the Yagua, it may be human or vegetal (Chaumeil 1983: 264-311);
among the Guajiro (Perrin 1992: 209-12) and the Tukano (Reichel-Dolmatoff 1971: 80-6), it
may be animal (qua animal or under the guise of the Master of animals).
3 A sample of these myths is analysed by Levi-Strauss (1967).
4 Two points are at issue here. The first is the problem of the integration of mental models.
Among anthropologists, discussion of this question, in so far as it exists, has mainly centred on
the way domain-specific cognition might, through the capacity for meta-representation, be cul-
turally elaborated (Atran 1993 [following Sperber 1990]; Bloch 1993; Sperber 1993). As Bloch
points out, this theme, which is obviously of vital interest to anthropologists, is still very little
explored. The second point is the question of the unity of a given culture. 'Cognitivist' anthro-
pologists who, in the name of psychological realism, refuse to consider culture as the manifes-
tation of an underlying shared script or, afortiori, as the expression of unconscious structures,
find it very difficult to account for this aspect of their object of study. They tend to sidestep the
problem by claiming that coherence or systematicity is simply an artefact of ethnography, or
even that the very notion of culture is meaningless. Culture may be little more than 'a second
rate orchestra playing a half remembered tune' without the benefit of a conductor (Lawson
1993: 206), yet we must still explain how and why everyone remembers the same tune, however
inaccurately, and why the level of cacophony in any given 'culture' is in fact surprisingly low.
5 This lecture is based on data collected by P Descola and myself during 26 months of field-
work between 1976 and 1979 and in 1984 and 1992. It also draws, of course, on the extensive
literature concerning the Jivaro, particularly on the work of Brown (1985), Harner (1972), Kar-
sten (1935) and Pellizzarro (1978; 1980).
6 For some published versions of this myth as it is told among the Shuar of Ecuador, see
Pellizzarro 1980.
7 Implicit in this formulation is the idea that the difference between mythical and ritual dis-
course is inherent in their respective meaning; i.e, part of the sense of a mythical narrative
derives from the fact that it is implicitly contrasted to a ritual pronouncement, and vice-versa.
Although it is perfectly legitimate to isolate such bodies of discourse for analytic purposes, at
some point their interconnexion, even if it appears to be purely negative, must be taken into
account.
8 With the exception of a body of myths relating to a switch from lethal caesarian to 'natural'
childbirth. These myths are common at least to the Shuar, Achuar and Aguaruna. According to
these narratives, the birth of a child used in illo tempore to imply the mother's death, since ba-
bies could only be delivered by cutting open their mothers' bellies. Rats eventually took pity on
women, and struck a deal with them whereby they taught them to give birth normally in ex-
change for a share of their peanut harvest (see Pellizzarro 1980 for some Shuar versions of this
myth).
9 These songs belong to a class of utterances named anent, a kind of soul-speech that tran-
scends normal channels of communication. A brief example of one such invocation, silently
addressed by a woman to her absent husband:

'Go flock to my little father's heart/ make him return to me crying pitifully/ go flock to his
thoughts (and make him cry) 'why does this feeling come to me?'/ fly to his thoughts and
make him awaken in tears/ (saying) 'why do I awaken thus?/ Oh, she's angry at me/ she is
going to leave me!'/ make him awaken with this thought/ crying, crying, go flock to himn my
little wakan, go flock to him' (my translation; the vernacular version, along with linguistic data,
may be read in Taylor & Chau 1983: 118-119).

10 The irenic perspective of some of the British Americanists is linked to their emphasis on
morality, i.e. normative values of sociality, as expressed by their informants and embodied in
their practice; see for example Belaunde 1992; Gow 1991; McCallum 1989; Santos Granero
1991. By contrast, French Americanists, who generally adhere to a structural-Durkheimian
approach, attempt to build up a model of observed social relations highlighting the 'construc-
tive' (in the sociological sense) aspect of conflict. For examples of this, see Albert 1985; Clastres
1972; Carneiro da Cunha & Viveiros de Castro 1985; Combes & Saignes 1991; Erikson 1986
(two are French only by intellectual filiation).

This content downloaded from

186.195.42.1 on Fri, 10 Jul 2020 04:26:02 UTC
All use subject to https://about.jstor.org/terms
 214 ANNE CHRISTINE TAYLOR

11 Virtually the only close kinsperson always to escape suspicion is one's mother. However,
one should not imagine that family life among the Jivaro is persistently conflict-ridden and
fraught with uncertainty. In normal circumstances, relations among members of a household or
local group are easy-going and often tender. Still, no one is surprised or particularly indignant -
as opposed to angry - when close members of a same family fall out and become involved in a
feud. In such cases the group splits and one or the other party joins a different territorial unit.
12 I think it also goes some way to explain the importance in this culture of forms of magical
discourse, such as the anent, meant to shape or modify others' affects.
13 The Achuar distinguish and name a variety of illnesses. Some - usually epidemic patholo-
gies - are labelled as 'white sickness', others are considered 'endogenous' and are initially dealt
with by resorting to plant medicine or to domestic forms of 'magical' curing. These afflictions
are thought of as 'accidental' only in the sense that the agent responsible for inflicting them
may have done so involuntarily, but of course the unintentionality of this imputed intentional-
ity is inherently suspect; if an illness lingers or worsens suspicion soon fades into the certainty
of deliberate mischief Judgements concerning the 'health status' of individuals or indeed whole
communities are thus highly dependent on the perceived shape of social relations: in times of
imminent or open conflict, not only do people tend to become unusually prone to sickness, but
their illness, whatever its taxonomic status, is immediately attributed to shamanic aggression.
14 This is true of men only Women do not systematically seek to encounter arutam, and they
experience them more rarely than men. The hidden 'strength' (kakarma-) - rather than anger -
they gain thereby is generally described in terms of a longer life and greater well-being, privi-
leged relations to the entities that control the appropriated fertility of garden plants, domestic
animals and humans, and secure affective relations to kin. It is also worth noting that female
arutam encounters occur in women's gardens, rather than in the forest, and that they seem to
happen only in times of acute emotional crises.

REFERENCES

Albert, B. 1985. Temps du sang, temps des cendres: repr6sentations de la maladie, sy

espace politique chez les Yanomami du sud-est Amazonie bresilienne. Thesis, University of
Paris, Nanterre.
Atran, S. 1993. Core domains versus scientific theories: evidence from systematics and Itza-Maya
folkbiology. In Mapping the mind: domain spectftcity in cognition and culture (eds) L.A. Hirschfeld &
S.A. Gelman. Cambridge: Univ. Press.
Belaunde, E. 1992. Gender, commensality and community among the Airo Pai ofwestern Amazonas.
Thesis, University of London.
Bloch, M. 1991. Language, anthropology and cognitive science. Man (N.S.) 26, 183-98.
1992. What goes without saying: the conceptualization of Zafimaniry society. In
Conceptualizing society (ed.) A. Kuper. London, New York: Routledge.
1993. Domain specificity, living kinds and symbolism. In Cognitive aspects of religious symbolism
(ed.) P Boyer. Cambridge: Univ. Press.
Boyer, P 1990. Tradition as truth and communication: the cognitive description of traditional discourse.
Cambridge: Univ. Press.
1993. Cognitive aspects of religious symbolism. Cambridge: Univ. Press.
Brown, M. 1985. Tsewa's gtft: magic and meaning in an Amazonian society. Washington: Smithsonian
Institution.
Capistrano de Abreu, J. 1941. Grammatica, textos e vocabulario Caxinauas. Rio de Janeiro: Edigao de
Sociedade Capistrano de Abreu.
Carneiro da Cunha, M. & E.B. Viveiros de Castro 1985. Vinganga e temporalidade: os Tupinambas.
J. Soc. Am. 71, 191-208.
Chaumeil, J.P 1983. Voir, savoir, pouvoir: ke chamanisme des Yagua du nord-estpe'ruvien. Paris: Editions de
l'Ecole des Hautes Etudes en Sciences Sociales.
Clastres, P 1972. Chronique des Indiens Guayaki. Paris: Plon.
Combes, I. & T. Saignes 1991. Alter ego: naissance de l'identite chiriguano. Paris: Editions de l'Ecole des
Hautes Etudes en Sciences Sociales.
Descola, P 1994. Les lances du crepuscuk: relationsjivaro. Paris: Plon.

This content downloaded from

186.195.42.1 on Fri, 10 Jul 2020 04:26:02 UTC
All use subject to https://about.jstor.org/terms
 ANNE CHRISTINE TAYLOR 215

Erikson, P 1986. Alt6ritd, tatouage et anthro

Soc. Am. 72, 185-209.
Gow, P 1991. Of mixed blood: kinship and history in Peruvian Amazonia. Oxford: Univ. Press.
Harner, M. 1972. TheJivaro: people of the sacred waterfall. NewYork: Doubleday, Natural History Pres
Hirschfeld, L.A & SA. Gelman (eds) 1993. Mapping the mind: domain specficity in cognition and cultur
Cambridge: Univ. Press.
Karsten, R. 1935. Head hunters of the westernAmazonas (Commun. Hum. lit. 7:1). Helsingfors: Societas
Scientiarum Fennica.
Kuper, A. (ed.) 1992. Conceptualizing society. London, New York: Routledge.
Lawson, TE. 1993. Cognitive categories, cultural forms and ritual structures. In Cognitive aspects of
religious symbolism. Cambridge: Univ. Press.
Levi-Strauss, C. 1967. Du miel aux cendres (Mythologiques 2). Paris: Plon.
McCallum, C. 1989. Gender, personhood and social organization amongst the Cashinahua of
western Amazonia. Thesis, University of London.
Overing, J. 1985. Images of cannibalism, death and domination in a 'non-violent' society. J. Soc.
Am. 72, 133-56.
Pellizzarro, S. 1978. Nunkui: mitos y ritos. Sucua, Mundo Shuar, Serie Mitologia Shuar.
1980. Tsantsa: la celebraci6n de la cabeza reducida. Sucua, Mundo Shuar, Serie Mitologia Shuar.
Perrin, M. 1992. Les praticiens du r&ve. Paris: Presses Universitaires de France.
Reichel-Dolmatofl G. 1971. Amazonian cosmos: the sexual and religious symbolism of the Tukano Indians.
Chicago: Univ. of Chicago Press.
Santos Granero, F. 1991. The power of love: the moral use of knowledge amongst the Amuesha of central Peru.
London: Athlone Press.
Severi, C. 1993. La memoria rituale:follia e immagine del Bianco in una tradizione sciamanica amerindiana.
Firenze: La Nuova Italia.
Sperber, D. 1990. The epidemiology of beliefs. In Psychological studies of widespread beliefs (ed.) C.
Frazer. Oxford: Univ. Press.
1993. The modularity of thought and the epidemiology of representations. In Mapping the
mind: domain spectfcity in cognition and culture (eds) L.A Hirschfeld & SA. Gelman. Cambridge:
Univ. Press.
Strathem, M. 1992. Parts and wholes: refiguring relationships in a post-plural world. In
Conceptualizing society (ed.) A. Kuper. London, New York: Routledge.
Taylor, A.C. 1993. Remembering to forget: identity, mourning and memory among the Jivaro. Ma
(N.S.) 28, 653-78.
& E. Chau 1983. Jivaroan magical songs. Amerindia (Rev. Ethnoling. amerind.) 8, 87-127.

Le corps de I'ame et ses etats: une perspective amazonienne sur I'etre

humain

Resume'
Cet article s'attache tout d'abord a explorer certaines des premisses relatives a la personne,
au lien social et a la mortalite sous-jacentes a l'exp6rience et a la repr6sentation du soi dans
la culture jivaro-achuar. Ces pr6suppos6s, quoique non explicites et apparemment contra-
dictoires entre eux, se combinent pour former une th6orie implicite pr6cise et complexe de
l'etre humain vivant. La conscience jivaro du soi s'enracine dans la fusion d'une forme
corporelle generique et d'une introjection de la perception d'autrui de cette forme-corps;
ainsi, l'image d'un corps initialement anonyme s'individualise progressivement en s'impr6-
gnant de la memoire des 6motions li6es aux rapports sociaux quotidiens. La perception du
soi est donc sujette tant a des etats d'affaiblissement et d'incertitude - cat6goris6s comme
maladies induites - et a des 6tats d"hypersubjectivite' engendres par la communication
rituelle avec des esprits. Les fondements interrelationnels des representations du soi menent
l'auteur a s'interroger sur l'approche traditionnelle en anthropologie des faits de conscience
et a proposer une perspective d'analyse plus attentive aux contextes et aux conditions
d'enonciation des savoirs traditionnels.

EREA-CNRS, B.P 8, 7, rue Guy Moquet, 94801 V4lejutfC6dex, France

This content downloaded from

186.195.42.1 on Fri, 10 Jul 2020 04:26:02 UTC
All use subject to https://about.jstor.org/terms