Bird David Animism
Bird David Animism
Bird David Animism
The University of Chicago Press and Wenner-Gren Foundation for Anthropological Research are collaborating
with JSTOR to digitize, preserve and extend access to Current Anthropology.
http://www.jstor.org
1999 by The Wenner-Gren Foundation for Anthropological Research. All rights reserved 0011-3204/99/40supp-0002$3.00
Animism Revisited
Personhood, Environment, and
Relational Epistemology1
by Nurit Bird-David
S67
cause the logic underlying it is today questionable. Tylor was not as rigid a positivist as he is often made out
to be (see Ingold 1986:9496; Leopold 1980). However,
he developed this representation within a positivistic
spiritual/materialist dichotomy of 19th-century design
in direct opposition to materialist science, in the belief
(and as part of an effort to prove this belief) that only
science yielded true knowledge of the world. Furthermore, the moral implications of this representation are
unacceptable now. Tylor posited that animists understood the world childishly and erroneously, and under
the influence of 19th-century evolutionism he read into
this cognitive underdevelopment. Yet the concept still
pervasively persists.
Equally surprisingly, the ethnographic referentthe
researchable cultural practices which Tylor denoted by
the signifier/signified of animismhas remained a
puzzle3 despite the great interest which the subject has
attracted. Ethnographers continue to cast fresh ethnographic material far richer than Tylor had (or could have
imagined possible) into one or more of the Tylorian categories religion, spirits, and supernatural beings
(e.g., Endicott 1979, Howell 1984, Morris 1981, BirdDavid 1990, Gardner 1991, Feit 1994, Povinelli 1993,
Riches 1994). At the same time, they have commonly
avoided the issue of animism and even the term itself
rather than revisit this prevalent notion in light of their
new and rich ethnographies.4
A twofold vicious cycle has ensued. The more the
term is used in its old Tylorian sense, without benefit of
critical revision, the more Tylors historically situated
perspective is taken as real, as the phenomenon
which it only glosses, and as a symbol that stands for
itself (Wagner 1981). In turn, anthropologys success in
universalizing the use of the term itself reinforces derogatory images of indigenous people whose rehabilitation from them is one of its popular roles.
This paper attempts a solution generally drawing on
a synthesis of current environment theory (insisting
that the environment does not necessarily consist dichotomously of a physical world and humans) and current personhood theory (asserting that personhood does
not necessarily consist dualistically of body and spirit).
These dualistic conceptions are historical constructs of
a specific culture which, for want of a better term, will
henceforth be referred to by the circumlocution modernist. (Modernist signals neither the dichotomous
opposite of primitive nor the equivalent of scientific but ideas and practices that dominated the EuroAmerican cultural landscape from the 17th to the 20th
century. Furthermore, modernist self-concepts will
be used as an objectification of what is often only a fragment of peoples composite identity, a part of their consciousness, while local person-concepts will be used
as an objectification of fragments of todays complex in3. It is regarded one of the oldest anthropological puzzles by Descola (1996:82).
4. An exception coming close to revisiting the notion is Hallowell
(1960); a liminal exception is Guthries recent revisit (1993), Descola (1992, 1996) contrasts totemic systems and animic systems but does not look deeply into animism as such.
1930s) and especially his paper Ojibwa Ontology, Behavior, and World View (1960) are provocative starting
points for our reassessment of theories of animism. Hallowell observed that the Ojibwa sense of personhood,
which they attribute to some natural entities, animals,
winds, stones, etc., is fundamentally different from the
modernist one. The latter takes the axiomatic split between human and nonhuman as essential, with
person being a subcategory of human. The Ojibwa
conceives of person as an overarching category
within which human person, animal person,
wind person, etc., are subcategories. Echoing EvansPritchards account of Azande magic (1937), Hallowell
furthermore argues that, contrary to received wisdom
and in the absence of objectivist dogma, experience itself does not rule out Ojibwa animistic ideas. On the
contrary, he argues (a point reiterated by later ethnographers [see Scott 1989, Feit 1994]), experience is consistent with their reading of things, given an animistic
dogma.
Hallowells contribution is to free the study of animistic beliefs and practices first from modernist personconcepts and second from the presumption that these
notions and practices are erroneous. However, the case
needs to be further pursued. He states that the Ojibwa
sense of personhood is different without exploring its
sense far enough, perhaps because, although the concept goes back to Marcel Mausss work of 1938,9 before
the 1960s research into the person as a cross-cultural
category hardly existed. He argues that Ojibwa engagement in the world does not rebuff their animistic views
but does not explain how the beliefs are engendered and
perpetuated. I shall pursue his insight through ethnographic material largely drawn from my work with
Nayaka, a hunter-gatherer community of the forested
Gir Valley in the Nilgiri region of South India.10 My objective will be to understand the senses of what they
call devaru, a concept which is not just a foreign word
requiring translation but enigmatic to positivistic
thought. Neither spirits (deriving from the spirit/
body dualism of the modernist person-concept) nor supernatural beings (mirroring the Western idea of nature)11 is an appropriate English equivalent, though
these are the common translations of corresponding notions in other studies.12 Hallowells alternative otherthan-human persons escapes these biased notions but
still conserves the primary objectivist concern with
classes (human and other-than-human). I use superpersons (persons with extra powers) as a general reference
9. Mausss work was first translated into English only in 1979 (and
see 1985). For some recent works on the self see Morris (1994),
Carrithers, Collins, and Lukes (1985), and Shweder and LeVine
(1984).
10. The Gir Valley is a fictive name for one of the Nilgiri-Wynaads
valleys.
11. See Durkheim (1915), Lovejoy (1948), Saler (1977), Descola
(1996).
12. See Endicott (1979), Howell (1984), Morris (1981), Bird-David
(1990), Gardner (1991), Feit (1994), Povinelli (1993), and, for a comparison, Mageo and Howard (1996).
terrain would have allowed their dispersal. They contained one, two, or sometimes even three living spaces,
barely separated from each other, each occupied by a
nuclear family. Weather permitting, families rested,
ate, and slept in the open beside outdoor fireplaces only
a few meters apart. They led their domestic lives together, sharing space, things, and actions. They experienced simultaneously what happened to them and to
their fellow Nayaka. This was the case with respect to
most Nayaka in the Gir area, not just the residents of
ones own place, because there was much movement
between sites and people stayed at each others places
for days, weeks, and even months at a time.
The idea that one shared space, things, and actions
with others was central to the Nayaka view of social
life. A Nayaka was normatively expected to share with
everybody as and when present, especially (but not
only) large game, irrespective of preexisting social ties,
criteria, and entitlement. Sharing with anyone present
was as important as if not more important than effecting a distribution of things among people. A Nayaka
was, furthermore, expected to give others what they
asked for, whatever this might be, to preempt refusals
and hence challenges to the felt sense that all of us
here share with each other. The idea and practice of
sharing constituted a habitus within which agentive negotiation, manipulation, and nonconformity took place
(see Bird-David 1990). For example, normally people
shared things requested of them, but when exceptionally they did not want to part with something, rather
than disrupt the ongoing sense of sharingthe rhythm
of everyday social lifethey hid that thing or avoided
people. This way, they preempted chances of sharing requests and refusals. Equally, people excessively requested things from people they wanted to embarrass or
manipulate into persistent giving.
As I understand it, this common experience of sharing space, things, and actions contextualized Nayakas
knowledge of each other: they dividuated each other.
They gradually got to know not how each talked but
how each talked with fellows, not how each worked but
how each worked with fellows, not how each shared
but how each shared with fellows, etc. They got to
know not other Nayaka in themselves but Nayaka as
they interrelated with each other, Nayaka-in-relatedness with fellow Nayaka. Through cumulative experiences, they sensed each other as dividuated personalities, each with a relatively persisting way of engaging
with others against the relative change involved in their
mutual engagement. Nayaka speakers, for example,
commonly described fellow Nayaka by the way they behaved vis-a`-vis themselves, for instance, as Mathen
who laughs a lot, Mathen who listens attentively,
and so on (Mathen being one of a few personal names
in circulation) (see Bird-David 1983).
Nayaka commonly objectified each other not as the
Maussian characterthe locus [in everyday life] of
different rights, duties, titles and kinship names within
the clan (Carrithers, Collins, and Lukes 1985:vii)but
as kin, relatives, ones related with. In everyday social
them and share with them. Their composite personhood is constitutive of sharing relationships not
only with fellow Nayaka but with members of other
species in the vicinity. They make their personhood by
producing and reproducing sharing relationships with
surrounding beings, humans and others. They do not dichotomize other beings vis-a`-vis themselves (see BirdDavid 1992a) but regard them, while differentiated, as
nested within each other. They recognize that the other
beings have their different affordances and are of diverse sorts, which is indicated among other things by
the different words by which they refer to them (hills,
elephants, etc.). However, Nayaka also appreciate that
they share the local environment with some of these beings, which overrides these differences and absorbs
their sorts into one we-ness. Beings who are absorbed
into this we-ness are devaru, and while differentiated
from avaru (people), they and avaru, in some contexts,
are absorbed into one we-ness, which Nayaka also
call nama sonta. The devaru are often objectified by
kinship terms, especially ette and etta(n) (grandmother
and grandfather) and occasionally dodawa and dodappa
(big mother and father). The use of kinship terms for
superpersons, especially grandparents, is common
also among other hunter-gatherers (e.g., see Hallowell
1960:27).
Maintaining relationships with fellow Nayaka but
also with other local beings is critical to maintaining
Nayaka identity because it is critical to maintaining
personhood. They retain immediate engagement with
the natural environment and hold devaru performances
even when they make a living by different means such
as casual labor. This is common among many other
hunter-gatherers, even those well integrated into their
respective states who live by such diverse means as
state benefits or jobs in the state bureaucracy (see, e.g.,
Tanner 1979, Povinelli 1993, Bird-David 1992b). By
maintaining relationships with other local beings to
reproduce their personhood, Nayaka reproduce the
devaru-ness of the other beings with whom they share.
The other beings are drawn into interrelating and sharing with Nayaka and so into Nayaka kinship relationships. These relationships constitute the particular beings as devaru.
To summarize this point of the argument, the devaru
objectify sharing relationships between Nayaka and
other beings. A hill devaru, say, objectifies Nayaka relationships with the hill; it makes known the relationships between Nayaka and that hill. Nayaka maintain
social relationships with other beings not because, as
Tylor holds, they a priori consider them persons. As and
when and because they engage in and maintain relationships with other beings, they constitute them as
kinds of person: they make them relatives by sharing
with them and thus make them persons. They do not
regard them as persons and subsequently some of them
as relatives, as Durkheim maintains. In one basic sense
of this complex notion, devaru are relatives in the literal sense of being that or whom one interrelates
with (not in the reduced modern English sense of hu-
siveness and engagement between things, events, moreover, which prototypically involve the actor-perceiver.
Discriminating devaru is contingent on affordances
of environmental events and things and (as I shall next
argue) on enhanced attention to them through particular traditions of practice.
devaru as performance characters
Devaru performances are pivotal in developing attention to devaru in-the-world and reproducing concepts of
devaru as objectifications of relationships. These performances are complex affairs which, in the modernist
sense, involve spirit-possession by devaru but also a
great deal more, including a communal social gathering,
healing, an altered state of consciousness, communication with predecessors, secondary burial for people who
have died since the previous event, and music and dancing. Each affair spans two days and the intervening
night. Nayaka hold them every year or so in each village, one place after the other, each attended by people
from the whole area who participate in several events
of this sort every year. Nayaka do not seem to refer to
this event by any single name or mark it off from everyday experience. Pandalu, the word I apply to the affair, is sometimes used for the purpose, referring to the
hut which is specially built for the event as accommodation for the visiting devaru.21
In examining one pandalu event, limiting myself to
devaru alone, I adopt a performance-centered approach
influenced by, among others Tambiah (1970, 1985
[1979]). Unlike the Geertzian tradition, this approach
focuses on what the pandalu does rather than what it
means. It focuses on the pandalu as an event in-theworld itself, not a text. It is concerned with the extent to which such events, instead of referring to or
talking about, do something in-the-world. I go farther,
as I cast the pandalu (following Nayaka) right away as
an experience, a performance, a social event in-theworld, which is continuous and coherent with and even
nested within other Nayaka experiences. (I do not cast
it as ritual, as opposed to practice, and then correctively adopt a performance-centered approach to it.) The
examination fills a lacuna in the work of Ingold (e.g.,
1996), who, like Gibson, pays inadequate attention to
interhuman action in-the-world in favor of action
towards other species;22 clearly, action towards fellow
humans constitutes an important part of ones environment.
June 9, 1979, Kungans place23 (where I lived at the
time with his family): People arrive casually during the
day, each family at its own time. They engage with
the local residents in everyday activities, chatting, sharing food, going to the river, fetching firewood, etc. Late
21. Pandalu means temple to neighboring Hindu people.
22. Ingold (1997) questions the autonomy of social relations.
23. Nayaka have no fixed names for places and refer to them by
mentioning a prominent landmark or the name of a central person
living there.
occasionally by names; sometimes only by their dividuated characters (as the one who always requests wild
fowl for food or waves a knife, etc.) and sometimes
just as devaru in general. The most vivid and generally
known devaru are hill devaru, whose existence appears
to go far back into the past. (Among neighboring huntergatherer Pandaram and Paliyan, hill chavu and hill devi
are also singled out [Morris 1981, Gardner 1991].) Other
vivid devaru are elephant devaru, minor Hindu deities
worshiped locally, and a deity of the Kurumba people
who lived in the locality several decades before. Generally, the more devaru appear year after year and are related with, the more vividly they are invoked, the more
they are known, the more, in a sense, they exist.
Hardly anything is said about devaru in myth or other
oral tradition either within the performance or outside
it (Morris reports the same for Hill Pandaram [1981:
208]).
The devaru evoked often improvise on the same repetitive phrases. The saying, the voicing, the gesturing
are important. These principal aspects of their behavior
are, in Batesons term (1979), meta-communication,
namely, communicating that devaru are communicating, because the devaru are present as they move, talk,
make gestures, etc. They are present as they communicate and socially interact with Nayaka. At peak times,
everyone gathers around the visiting devaru, taking an
active part in the conversation or just closely listening
to it. At other times, only a few people do this while the
others busy themselves with their own domestic affairs.
The conversation has to be kept going at all times.
When it slackens, the devaru complain and urge more
people to join in. At the extreme, at dull moments in
the heat of the day and deep in the night, this or that
Nayaka grudgingly comes forward and engages the devaru in conversation. (I became helpful at various points
in this event, letting Nayaka go about their business as
I recorded and listened to the devaru by myself.) Keeping the conversation going is important because it keeps
the Nayaka-devaru interaction and in a sense the devaru themselves alive.
Conversation with the devaru is highly personal, informal, and friendly, including joking, teasing, bargaining, etc. In its idiomatic structure it resembles the
demand-sharing discourse which is characteristic of
Nayaka and hunter-gatherers generally (see Bird-David
1990). With numerous repetitions or minor variations
on a theme, Nayaka and devaru nag and tease, praise
and flatter, blame and cajole each other, expressing and
demanding care and concern. For example, Nayaka
stress that they are taking proper care of the devaru (or
apologizing for not giving more or moaning about not
being able to give more, etc.) and complain that the devaru, in turn, do not take care of them (or not enough
or not as in previous years, etc.). The devaru stress how
much they care for Nayaka and request better hospitality (more offerings, an earlier start for the event, more
dancing, etc.). The Nayaka request cures from illnesses.
The ordinary round of everyday affairs continues during the two days of the pandalu. Domestic chores are
not marginalized on account of the occasion but constitute a significant part of its structure. Throughout the
two days, Nayaka families go on with their domestic
activities, frequently sharing with each other and, in
some ways, with their devaru visitors, too. The devaru
hut resembles ordinary Nayaka dwellings. Some men
occasionally take naps there, sharing the hut with the
devaru. In the morning, when people go to wash in the
river and bring back water, they bow in the four directions, inviting local devaru to join them. Women on
their way back sprinkle water from their vessels in the
four directions, sharing the water with devaru around.
In the course of conversation devaru request betel-nuts
from their Nayaka interlocutor. One elderly Nayaka
woman falls into a trance. She does not utter coherent
words; in her frenzy she only sweeps the ground around
the devaru hut and starts to undress (which bystanders
stop her from doing). A joint meal of rice, cooked by
Kungans daughter and her husband, brings the event to
a close. The food is shared equally among those present,
and some food is spread in the four directions.
The pandalu makes known the Nayaka-devaru relatednesses and at the same time reproduces them. Objectified as kinship relationships, the relatednesses reconstitute all the participants as sonta and each of them
as a person (Nayaka person, hill person, stone person,
etc.). Furthermore, the pandalu constitutes (in the Gibsonian sense) aids to perceiving that put the viewer
into the scene (Gibson 1979: 282, cited above). It educates the attention to perceive and specify the environment (while engaging with it) in a relational way. The
pandalu preserves information (as effectively as
books and even motion pictures); moreover, it encourages the learner to engage interactively with this information and so to experience it socially. The engagement with devaru characters educates the attention
to notice devaru as they interact with oneself. It improves the skill of picking up information about the engagement itself, within its confines, from an engaged
viewpoint.
If Nayaka only subsisted by hunting and gathering in
their environment, they might perceive only its utilitarian affordances: an animal as something edible; a stone
as something throwable; a rock as something one can
shelter under. Within the practice of engaging with devaru characters in the pandalu they are educated to perceive that animals, stones, rocks, etc., are things one
can relate withthat they have relational affordances,
that is, what happens to them (or how they change) can
affect and be affected by what happens to people (or how
they change): an animal-avoiding-me in relation to meupsetting-the-animal, a stone-coming-towards-me in relation to me-reaching-for-the-stone, a rock-securing-me
in relation to me-seeking-a-shelter. Participants learn
from conversing and sharing with devaru characters to
discriminate mutually responsive changes in themselves and things they relate with; they become increasingly aware of the webs of relatedness between themselves and what is around them. From the bargaining
and demand-sharing with devaru characters they learn
other things, making ones awareness of ones environment and ones self finer, broader, deeper, richer, etc.
Knowing, in the second case, grows from and is maintaining relatedness with neighboring others. It involves
dividuating the environment rather than dichotomizing
it and turning attention to we-ness, which absorbs
differences, rather than to otherness, which highlights differences and eclipses commonalities. Against
I think, therefore I am stand I relate, therefore I am
and I know as I relate. Against materialistic framing
of the environment as discrete things stands relationally framing the environment as nested relatednesses. Both ways are real and valid. Each has its limits and its strengths.
Framing the environment relationally does not constitute Nayakas only way of knowing their environment, though in my understanding they regard it as authoritative among their other ways. Nor is it unique to
Nayaka. I would hypothesize that relational epistemologies of this kind enjoy authoritative status in cultures
of peoples we call hunter-gatherers. These peoples normalize sharing with fellow persons. They engage intimately with their environment (if only periodically
while on a break from other economic pursuits [BirdDavid 1992b]). They celebrate animistic performances.
Their performance traditionsfor example, the Cree
shaking tent ritual (e.g., Hallowell 1960, Feit 1994,
Brightman 1993), the !Kung medicine dance (e.g.,
Marshall 1962, Katz 1982), the Hadza sacred epeme
dance (see Woodburn 1982), the Batek fruit-seasons
singing session (see Endicott 1979), and Paliyan and
Pandaram spirit possession (see Gardner 1991, Morris
1981)are functionally similar to the Nayaka pandalu.
These performances involve the visiting of superpersons who appear through trance and dance or make
their voices heard.27 The people regard these superpersons as friends and relatives and often address and refer
to them by kinship terms. They approach them in a personal, friendly, and immediate way. These events are
the central communal affairs of these communities and
often the main celebrational means by which they sustain their senses of identity.28 Each event constitutes a
participation frame (Lave and Wenger 1991) which,
together with the participation frame of hunting-gathering practice itself, nurtures a complex articulation of
skills, a double-bind engagement which co-privileges
27. In some cases devaru are additionally invoked by objects, with
which one talks, eats, sings, dances, etc. This is less common than
their invocation by performance but of considerable theoretical importance.
28. At their respective times of study, these events were frequently
held, for example, weekly among !Kung, monthly among Hadza,
and whenever need arises among Pandaram. They spanned a significant stretch of time, for example, the whole night among
!Kung, two to three nights in succession among Hadza, and
from evening into the night among Paliyan. The events involved
the entire community as active spectators and a considerable proportion as performers, for example, one-third of the men among
!Kung, one-eighth of the men among Pandaram, and 28% of the
adults among Paliyan. In the case of Nayaka, about one-fifth of
the men acted as performers.
utilizing and respecting animated things, self-interest and the cooperation within which that self-interest
can be achieved.29
Furthermore, relational epistemologies function in
diverse contexts where other epistemologies enjoy authority, including Western contexts (to a much greater
extent than the authoritative status of science permits).
When (going back to Guthries examples) we animate
the computers we use, the plants we grow, and the cars
we drive, we relationally frame them. We learn what
they do in relation to what we do, how they respond to
our behavior, how they act towards us, what their situational and emergent behavior (rather than their constitutive matter) is. As Nayaka get to know animated aspects of their environment, so we get to know these
animated things by focusing on our relatedness with
them within the confines of that relatedness from a relational viewpoint. This sort of relational framing is articulated with other epistemologies in complex, variable, and shifting ways that deserve study. (The
example of ethologists mentioned earlier is a case in
point: in regarding as persons the study animals with
which they live, they frame them relationally in addition to making them the objects of their scientific
study.)
As a hypothesis, furthermore, I am willing to agree
with Tylor, not least because Guthrie goes some way
towards substantiating the point, that the tendency to
animate things is shared by humans. However, this
common tendency, I suggest, is engendered by human
socially biased cognitive skills, not by survival of
mental confusion (Tylor) or by wrong perceptual
guesses (Guthrie). Recent work relates the evolution of
human cognition to social interaction with fellow humans. Its underlying argument is that interpersonal
dealings, requiring strategic planning and anticipation
of action-response-reaction, are more demanding and
challenging than problems of physical survival (Humphrey 1976). Cognitive skills have accordingly evolved
within and for a social kind of engagement and are socially biased (Goody 1995). We spontaneously employ
these skills in situations when we cannot control or totally predict our interlocutors behavior, when its behavior is not predetermined but in conversation with
our own. We employ these skills in these situations, irrespective of whether they involve humans or other beings (the respective classification of which is sometimes part of reflective knowing, following rather than
preceding the engagement situation). We do not first
personify other entities and then socialize with them
but personify them as, when, and because we socialize
with them. Recognizing a conversation with a
counter-beingwhich amounts to accepting it into fellowship rather than recognizing a common essence
makes that being a self in relation with ourselves.
29. Compare Briggs (1982) and Guemple (1988) on the teaching of
Inuit children to relate with other people in a double-binding way
and Myerss study (1986) of tenuous articulation of personal autonomy and relatedness among the Australian Pintupi.
Conclusions
How we get to know things is nested within culture and
practice and takes multiple forms. Nayaka relationally
frame what they are concerned about as their authoritative (but not only) way of getting to know things. They
seek to understand relatednesses from a related point of
view within the shifting horizons of the related viewer.
Their relational epistemology, their study of how
things-in-situations relate to the actor-perceiver and,
from the actor-perceivers point of view, to each other,
is embodied in the practices which Tylor christened
primitive animism, articulated with a relational personhood concept and a relational perception of the environment. Previous theories of animism, taking mod30. For example, compare hunter-gatherer animism with premodern Western animism as described in Merchant (1980) and Burke
(1972).
31. I owe the formulation of this question to Ingrid Jordts forthcoming work on the articulation of Buddhist and animist epistemologies in Burma.
Comments
e du a r d o v i v e i r os de c a s t r o
Kings College, Cambridge CB2 1ST, U.K. 10 iv 98
Bird-David rejects modernist understandings but holds
fast to the quintessentially modernist concern with
epistemology. The massive conversion of ontological
questions into epistemological ones is the hallmark of
modernist philosophy. She does not accept the modernist answers, but the question how we come to know
things is taken as a natural one to be put with reference
to the Nayaka, who are thus encompassed by this ambiguous we and expected to provide an answer for
us. The answer is that knowing is relating and the
cogito is relational. The problem remains framed in
terms of knowledge even though the answer could be
taken to imply that knowledge, let alone the cogito, has
little to do with it. Anthropologists persist in thinking
that in order to explain a non-Western ontology we
must derive it from (or reduce it to) an epistemology.
Animism is surely an ontology, concerned with being
and not with how we come to know it. Bird-David falls
into the Tylorian trap and feels compelled to assess the
validity of this epistemology and to justify it on the basis of its cognitive naturalness.
The author has a fondness for scare quotes, but I am
afraid this sort of pocket deconstruction is hardly
enough to keep one safe from essentialization and modernist projection. The notion of hunter-gatherers is a
case in point. Bird-David finds the concept suspicious,
but all the same she attributes to hunter-gatherers a
number of characteristics also to be found in many horticultural societies. There is then a suggestion that the
prevalence of epistemologies of the kind described for
the Nayaka is somehow (causally?) derived from the
fact that [hunter-gatherers] normalize sharing with fellow persons; in other words, sharing is taken as the essence of hunter-gatherers social life. This seems close
to the traditional notion of a metaphoric projection of
human relations onto the environmentan idea which
has been cogently criticized by Ingold. Also, she dislikes
dualisms and dichotomies, but this does not prevent her
from posing a dichotomy between a dichotomous modernist epistemology and a non-dichotomous relational
one. She objects, in particular, to the concepts of subject
and objectbut whence comes the notion of objectification?
I find the attempt to combine Stratherns and Ingolds
theories very problematic. The dividual of the former
shares only its name with Bird-Davids, among other
things because Stratherns notion of relation, as I understand it, has little in common with Bird-Davids notion
being in-the-world has infiltrated ecological anthropology (see also Ingold 1996) without any mention of
the sources of these concepts (see Gooch 1998).
The example of the ethologists coming to view their
animal objects as subjects illustrates how relatedness
is something that people are capable of achieving in particular experiential contexts of some minimal duration.
It is undoubtedly everywhere fundamental to the local
and embedded dimensions of human life. Why exoticize
it into something that theythe Nayakahave?
Once again, the anthropological gaze risks reducing itself to the class perspective of urban cosmopolitans
making careers out of objectifying the rural and the
local.
There is a contradiction between Bird-Davids concluding assumption that the modernist project estranged itself from the tendency to animate things and
her earlier observation that we may animate our computers and cars. Animation is one of Ellens (1988)
criteria of fetishism, and fetishism to Marx was central
to modern capitalism. It is indeed relevant to ask how
animism relates to fetishism. There is a difference
between representing relations between people as if
they were relations between things (Marxian fetishism)
and experiencing relations to things as if they were
relations to people (animation). The former is a
cognitive/ideological illusion, the latter a condition of
phenomenological/experiential resonance. I have suggested (Hornborg 1992) that machine fetishism, at
the ideological level, is as crucial to capitalism as
money or commodity fetishism. Machines can undoubtedly also be animated in a phenomenological
sense, as Bird-David suggests. We probably need to distinguish between the animation of living things such as
trees (animism, more narrowly defined) and that of nonliving things such as stones or machines (fetishism).
Cartesian objectivism and fetishism here emerge as
structural inversions of one another: the former denies
agency and subjectivity in living beings, whereas the
latter attributes such qualities to dead objects. In this
framework, a more strictly defined category of animism
would be reserved for the intermediate and quite reasonable assumption that all living things are subjects.
The epistemological predicament codified by Descartes was not so much an innovative, cognitive shift
from animism to objectivism as the emergenceor unprecedented generalizationof a social condition of
alienation. Rather than a cerebral innovation that has
since diffused, it is a reflection of a set of social circumstances that is continually being reproduced and expanded. Bird-Davids programmatic ambition to articulate environmental relations and personhood is
supremely worthwhile, but where in this text are the
insights on personhood that she wishes to employ? A
highly relevant aspect of personhood which might have
illuminated the relationist/objectivist contrast is the
tendency of non-Western (local?) people to anchor
their selves in concrete rather than abstract referencepoints (see Shweder and Bourne 1984, Hornborg 1994).
It is the long immersion in the concrete and experien-
intended actions: this is what Goody (1995) calls anticipatory interactive planning (AIP).
The kind of responsiveness envisaged in anticipatory
interactive planning, however, is fundamentally different from what Bird-David has in mind when she speaks
of the two-way responsive relatedness to components
of the environment such as trees that comes from a history of intimate engagement with them. To talk with
a tree, as she points out, is a question not of (mistakenly) attributing to it an inner intelligence and then
configuring how it might decide to react to what one
does but of perceiving what it does as one acts towards
it, being aware concurrently of changes in oneself and
the tree. Responsiveness, in this view, amounts to a
kind of sensory participation, a coupling of the movement of ones attention to the movement of aspects of
the world. If there is intelligence at work here, it does
not lie inside the head of the human actor, let alone inside the fabric of the tree. Rather, it is immanent in the
total system of perception and action constituted by the
co-presence of the human and the tree within a wider
environment. To develop this idea further, the first
thing we shall have to jettison is the cognitivist conception of intelligence as a mental computational device
responsible for processing the data of perception and
pulling the strings of action (see Ingold 1993:431). Human beings everywhere perceive their environments in
the responsive mode not because of innate cognitive
predisposition but because to perceive at all they must
already be situated in a world and committed to the relationships this entails.
b ri a n m o r r i s
Goldsmiths College, New Cross, London SE14 6NW,
England. 7 iii 98
I have read Bird-Davids lucid and valuable paper with
great interest and offer the following critical reflections
in the spirit of friendship:
Modernism, as Bird-David defines it, implies a conception of the human person as involving a radical
body/spirit (soul, mind?) split, a radical dualism between humans and nature, and the notion that the human person is an individual thing, a bounded asocial entity (organism). These conceptions, of course, largely
came out of Cartesian metaphysics and the bourgeois
liberal theory of the 17th century and were intrinsic as
ideologies to the rise of capitalism. A critique of these
conceptionswhich Bird-David links to current theory in ecology and personhoodgoes back two centuries to the time of Goethe, Hegel, and Marx. Philosophers, historians, sociologists, anthropologists,
socialists, romantic poets, evolutionary biologists, and
naturalists have long since concluded that humans are
a part of nature and that people everywhere are neither
disembodied egos (Descartes) nor abstract individuals
(the asocial organisms of bourgeois liberalism) nor simply a reflection of the commodity metaphor (Strathern)
referred to an indivisible relational wholesomething that cannot be divided, like the unity of the Trinity. As Gurevich argues (1992:297), in medieval times
man thought of himself as an integral part of the
world. . . . His interrelation with nature was so intensive and thorough that he could not look at it from
without. Nowadays, in contrast, the concept of the individual suggests the very opposite, namely, distinctions and discontinuities. The change in the meaning of
the concept, Williams points out, is a record in language of an extraordinary social and political history
(1976:133). Given this history, references to the dividual person are not particularly illuminating. If dividualism existed in medieval Europe, it probably was an
inversion of the relational view which Bird-David is
suggesting.
It would be wrong, however, to reduce the issue to
etymology. Bird-David, in my view, has not only constructed an eclectic and highly useful theoretical framework but also skilfully applied it to both the general
phenomenon of animism and her Nayakan ethnography.
l a u r a m. r i v a l
Eliot College, Department of Social Anthropology
and Sociology, University of Kent, Canterbury, Kent
CT2 7NS, England ([email protected]). 28 iv 98
Bird-Davids call for revisiting animism is timely.
While challenging Tylors embrace of the Western objectivist view of reality, she rightly points to the contemporary relevance of his seminal study of the belief
that life is produced by a spiritual force. She opposes
Durkheimian, structuralist (Levi-Strauss and Descola),
and cognitive evolutionist (Guthrie) explanations of animism and offers an original contribution to the debate
by focusing on the sociality, religious ideas, and ritual
practices of egalitarian hunter-gatherers whose tendency to attribute life to inanimate objects or mental
states to non-humans stems, she asserts, from a distinctive relational epistemology.
One of the greatest problems with Tylors view of animism, according to Bird-David, is its monolithic character. She argues instead for a plurality of animisms, on
the ground that different belief systems conceptualise
life, non-living, and human in fundamentally
different ways. She goes on to show that Nayaka nature
spirits (devaru) and spirit possession rites have much in
common with those of the Hadza, Hill Pandaram, and
!Kung. Hunting-and-gathering populations seem to attribute similar meanings to nature, life, and personhood. They do not dichotomise the person into
spirit and body or the environment into the physical
and the social but envisage instead a social environment based on the immediate, intimate, and engaged
experience of relatedness between dividual persons.
People like the Nayaka define a person as someone with
whom one shares. Spiritual forces are treated as persons
brought to life with whom space, things, actions, experience, and conversations can be shared.
I find Bird-Davids thesis that the animistic beliefs of
egalitarian hunter-gatherers objectify relations of sharing insightful. In the same way as she was able to recognise something unique in the economic activities of
nomadic hunters, gatherers, and some swidden horticulturists who procure rather than produce, as well as
in their social organisation (they demand-share rather
than exchange), she has now identified something distinctive in their cultures: the principle that to relate is
to know and that to bring to life is to impersonate. Following this principle, and depending on the context, animals may be turned into mere objects, into people, or
into divinities. And when natural kinds or natural
forces are made alive as persons, people relate to
them and communicate and socialise with them exactly as if they were fellow human beings.
Unfortunately, the ethnographic material she cites in
support of her thesis is not sufficiently developed or
clear (I found the examples of elephant devaru particularly obscure and ambiguous). Moreover, too little is
said about local perceptions and experiences of trances
and possession by animal spirits for the reader to decide
whether to agree or not with the author about the distinctiveness of hunter-gatherer animistic performances.
I found even more problematic the theoretical ground
on which Bird-David bases some of her most perceptive
ideas, for example, the idea that stones are given life and
personified as, when, and because of the desire to socialise with them. To reject Kennedys distinction between animation and anthropomorphisation on the
ground that Gibsons ecological psychology better explains why the affordances of natural objects are not essential properties but context-dependent is not, in my
view, satisfactory. Far from saying that ethologists feel
empathy for, hence relate to, the animals they are experimenting upon, Kennedy points to the intrinsically
anthropomorphic nature not only of everyday language
(which could simply be brushed aside as a metaphoric
property) but also of scientific thought. Scientists, like
all of us, and like the primitives Tylor was trying to
understand, tend to ascribe feelings and cognitive processes to living organisms, especially higher ones. What
concerns Kennedy (1992:9394) is that by thinking
about animals as if they had minds like oursthat is,
as if they were conscious and self-aware, as if they
thought, and as if they had purposes and used mental
imageswe confuse functions and causes and wrongly
project the exclusively human mind-body problem onto
other species (Kennedy 1992:168).
The question why humans tend to use human experience to interpret biological processes, in particular animal behaviour, is so fundamental that anthropologists
cannot answer it without entering into dialogue with
other disciplines, including cognitive psychology. I
agree with Bird-David that the main issue at stake is the
attribution of life to the non-living and how such attribution relates to the conferrring of human traits on nonhuman entities (see Rival 1998: 2027). But the matter
cannot be adequately settled without paying serious attention to the mechanisms that connect the intuitive
assumptions of everyday cognition or common sense
(as used by Atran 1990) and the counterintuitive representations that make up the core of complex religious
beliefs such as those informing Nayaka ritual performances (Boyer 1994).
a l a n r. s a n d st r om
Department of Anthropology, Indiana UniversityPurdue University Fort Wayne, 2101 Coliseum Blvd.
East, Fort Wayne, Ind., 46805, U.S.A. 5 iv 98
Bird-David is to be congratulated for writing on an interesting topic, a reevaluation of Tylors concept of animism. Unfortunately, in my view her postmodernist
stance robs the article of much of its potential value not
only in clarifying animism as an analytical concept but
also in evaluating anthropological contributions to an
understanding of animistic thought. Although careful
to avoid complete dismissal of science, she apparently
rejects its uniqueness as a way of knowing.
Bird-David discusses the work of Tylor, Durkheim,
and Levi-Strauss as scientific or modernist but can offer
nothing to replace it but an antiobjectivist, relational
epistemology supposedly practiced by the Nayaka of
southern India. She seems to propose a radical relativism in which each groups conceptions of personhood
replace or at least stand beside scientific attempts to
understand this difficult aspect of culture. Bird-David
uses Tylors 19th-century work as an example of how
science can lead researchers astray, but no contemporary anthropologist follows Tylors program of cultural
evolution. It has been made obsolete by the very
science Bird-David discounts. Moreover, she is forced
to admit that as empirical scientific research has increased knowledge of the worlds cultures the concept
of animism itself has fallen into disuse among ethnographers.
She attempts to explain animistic thinking simply by
placing it in the context of the Nayaka worldview, in
which, not surprisingly, it makes complete sense. The
Nayaka talk with superpersons because they have an
animistic worldview. But the question of what leads
people to develop such a worldview in the first place is
not addressed, and so no real explanation of animism is
offered and no advance is made over the work of Hallowell. Is Bird-David implying that Nayaka animism is
somehow natural and therefore not in need of explanation?
She suggests that relational epistemologies characterize hunting and gathering peoples everywhere but shies
away from further exploration of this intriguing propositionundoubtedly because it suggests techno-environmental causation, a concept from cultural ecology
derived from the scientific tradition within anthropology. The interesting hypothesis that animism may be
an extension of human cognitive skills to nonhuman
persons remains largely unexplored. Nowhere is it
reason that they are evaluated by distinct criteria. Science is a way of publicly presenting and evaluating evidence and contains within its practice a self-correcting
mechanism that addresses the critiques leveled at it by
postmodernists like Bird-David. The spectacular successes of scientific anthropology in expanding our
knowledge of the human condition, of making ones
awareness of ones environment and ones self finer,
broader, deeper, [and] richer, since the days of Tylor
should be acknowledged before being replaced by the
relativistic, antiobjectivist approach suggested in this
article.
Reply
nurit bird-david
Haifa, Israel. 27 vi 98
Critical or supportive, the commentators have taken a
close interest in this papers thesis, and I appreciate
their reciprocity for the work I put into writing it. I shall
first address critiques and misunderstandings and then
the suggestions offered by commentators for pursuing
the argument further.
Science is needlessly defended by Sandstrom. The
spectacular achievements of science are not undermined at all. A graduate in economics and mathematics, I have myself worked with hard data in the objectivist tradition and continue to do so whenever it is
possible and advantageous. Yet, powerful as it is, the
scientific way is neither good for studying everything
nor the only way of studying everything. Thisno
more, no lessis the broadest frame within which the
argument can be situated. The paper does not reject
[sciences] uniqueness as a way of knowing but on the
contrary stresses its being unique among other ways,
which makes it morenot lessintriguing for study
(comparative, sociological, and historical) and precious
as a study tool. Presenting Nayaka animistic practices
as a way of knowing is not to blur the difference between religious and scientific knowledge but rather to
rescue these practices from our pigeon-hole religion,
in which they were formerly placed.
The analytic use of dualisms and dichotomies is
forcefully defended by Viveiros de Castro. I argue that
in animistic perceptions of the environment oppositions are of secondary importance. Therefore, in order
to interpret, to try to get closer to, and to make sense
of their perspectives, the language of dualisms and dichotomies is an obstacle. In no way does this imply
dislike for dualisms and dichotomies in general. Indeed, to view this culture within a broader frame and
try to compare it with other cultureswhich is equally
part of the anthropologists workI myself use dichotomies, including the one between a dichotomous modernist epistemology and a non-dichotomous relational
tions as things, then (with anthropomorphism) attributing human qualities to them, then engaging with them
as with persons. Animism (as I conceptualize it) involves responsively engaging with beings/things, then
perceiving them as persons.
Several ethnographic queries have been raised by Viveiros de Castro, in support of his general critique, of
which for lack of space I respond to only two. He reads
into Nayaka culture an opposition he sees as unavoidable between superpersons and human persons, while
Nayaka, I suggest, primarily perceive both as persons
and the superperson as a person-plus, a person like
the human but with something added. He sees a contradiction between the description of some devaru as
devaru in general and the argument for their particularistic, event-derived nature, which may be due to
the awkwardness of my expression devaru in general.
Appearances in the pandalu are always particular: each
is enacted by a particular performer at a particular
time/place. As they come and engage with Nayaka
in the pandalu, they are devaru, though in some cases
(the ones I call devaru in general) the particular appearances are not immediately recognized as this or
that particular devaru by habitual ways of engaging
with Nayaka (e.g., gestures and sayings) remembered
from previous engagements. Rather, by engaging with
them, Nayaka gradually learn their ways of engagement
and learn to learn about the other within the engagement.
Let me move now to suggestions for pursuing the argument further: I agree with Rival that the thesis needs
to be ethnographically expanded with Nayaka material
and, ideally, material provided by other students such
as Rival herself. I agree with Palsson that it would be
interesting to compare hunter-gatherers animism with
some current environmental thought. Schools such as
deep ecology, social ecology, and eco-feminism envisage an all-encompassing moral community constitutive
of humans and nonhumans. Some radical environmentalists even call for a paradigmatic shift in not only our
view of nature but our view of the self, for example,
from ego to spirit, understood as a self not split but
differentiated from others within relationships (Koval
1988:esp. 300305).
The issue of modernitys emergence was introduced
briefly precisely in order to indicate that, far from being
parochial, a discussion of animism has far-reaching implications. Research on shifts occurring during the Renaissance period, would, as Palsson and Hornborg suggest, be a natural path for further work. Perhaps, as
Hornborg hypothesizes, objectivism emerged and expanded in connection with social relations of alienation. However, seeds of this seem to pre-appear in the
periods particular brand of animism. Nature generally
and earth particularly were personified as a woman and,
furthermore, a mother, but mother was used as a
symbol, dissociated from interactive, dynamic social relationships. Mother was associated with attributes
such as giving life and nurturing, which themselves
were thought of in a social and historical vacuum, as
References Cited
a t r a n, s c o t t. 1990. Cognitive foundations of natural history. Cambridge: Cambridge University Press. [lmr]
b a r n a r d, a. 1981. Universal categorization in four Bushmen
societies. LUomo 5:21937.
b a r n e t t, s. a. 1976. Coconuts and gold: Relational identity
in a South Indian caste. Contributions to Indian Sociology,
n.s., 10:13356.
b a t e s o n, g. 1979. Steps to an ecology of mind. New York: Ballantine Books.
b i r d - d a v i d, n. 1983. Inside and outside in kinship usage:
The hunter-gatherer Naiken of South India. Cambridge Anthropology 7(1):4757.
. 1989. The people and the ethnographic myth: An introduction to the study of the Nayaka, in Blue mountains: The ethnography and biogeography of a South Indian
region. Edited by P. Hockings. New Delhi: Oxford University
Press.
. 1990. The giving environment: Another perspective on
the economic system of gatherer-hunters. current anthropology 31:18396.
. 1992a. Beyond The Original Affluent Society: A culturalist reformulation. current anthropology 33:2547.
. 1992b. Beyond The hunting and gathering mode of subsistence: Observations on Nayaka and other modern huntergatherers. Man 27:1944.
. 1994. Sociality and immediacy, or, Past and present conversations on bands. Man 29:583603.
. n.d. Puja, or sharing with the gods? On ritualized possession among Nayaka of South India. Eastern Anthropologist. In
press.
b l u r t o n j o n e s, n. , a n d m. j. k o n n e r. 1976. !Kung
knowledge of animal behavior (or, The proper study of mankind is animals), in Kalahari hunter-gatherers: Studies of the
!Kung San and their neighbors. Edited by R. B. Lee and I. De
Vore. Cambridge: Harvard University Press.
b o u i s s a c, p. 1989. What is a human? Ecological semiotics and
the new animism. Semiotica 77:497516.
b o y e r, p a s c a l. 1994. The naturalness of religious ideas: A
cognitive theory of religion. Berkeley: University of California
Press. [lmr]