Experiments in Holism Theory and Practice in Contemporary

Transcription

Experiments in Holism
9781444333237_1_pretoc.indd i
8/7/2010 6:40:09 PM
9781444333237_1_pretoc.indd ii
8/7/2010 6:40:09 PM
Experiments in Holism
Theory and Practice in Contemporary
Anthropology
Edited by
Ton Otto and Nils Bubandt
A John Wiley & Sons, Ltd., Publication
9781444333237_1_pretoc.indd iii
8/7/2010 6:40:09 PM
This edition first published 2010
© 2010 Blackwell Publishing Ltd except for editorial material and organization © 2010 Ton
Otto and Nils Bubandt
Blackwell Publishing was acquired by John Wiley & Sons in February 2007. Blackwell’s
publishing program has been merged with Wiley’s global Scientific, Technical, and Medical
business to form Wiley-Blackwell.
Registered Office
John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ,
United Kingdom
Editorial Offices
350 Main Street, Malden, MA 02148-5020, USA
9600 Garsington Road, Oxford, OX4 2DQ, UK
The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
For details of our global editorial offices, for customer services, and for information about
how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com/wiley-blackwell.
The right of Ton Otto and Nils Bubandt to be identified as the author of the editorial material in this work has been asserted in accordance with the UK Copyright, Designs and Patents
Act 1988.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system,
or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or
otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without
the prior permission of the publisher.
Wiley also publishes its books in a variety of electronic formats. Some content that appears in
print may not be available in electronic books.
Designations used by companies to distinguish their products are often claimed as trademarks.
All brand names and product names used in this book are trade names, service marks, trademarks
or registered trademarks of their respective owners. The publisher is not associated with any
product or vendor mentioned in this book. This publication is designed to provide accurate and
authoritative information in regard to the subject matter covered. It is sold on the understanding
that the publisher is not engaged in rendering professional services. If professional advice or other
expert assistance is required, the services of a competent professional should be sought.
Library of Congress Cataloging-in-Publication Data
Experiments in holism : theory and practice in contemporary anthropology / Edited by Ton
Otto and Nils Bubandt.
p. cm.
Includes bibliographical references and index.
ISBN 978-1-4443-3323-7 (hardcover : alk. paper) 1. Anthropology–Philosophy.
2. Holism–Philosophy. I. Otto, Ton. II. Bubandt, Nils.
GN33.E94 2010
301.01–dc22
2010006825
A catalogue record for this book is available from the British Library.
Set in 10/12pt Galliard by SPi Publisher Services, Pondicherry, India
Printed in Malaysia
01
2010
9781444333237_1_pretoc.indd iv
8/7/2010 6:40:10 PM
Contents
Acknowledgments
List of Contributors
1. Anthropology and the Predicaments of Holism
Nils Bubandt and Ton Otto
Part 1
Rethinking Holism in Ethnographic Practice
2. Beyond the Whole in Ethnographic Practice? Introduction to Part 1
3. Holism and the Expectations of Critique in Post-1980s Anthropology:
Notes and Queries in Three Acts and an Epilogue
George E. Marcus
4. Worlding the Matsutake Diaspora: Or, Can Actor–Network Theory
Experiment With Holism?
Anna Tsing
5. The Whole Beyond Holism: Gambling, Divination,
and Ethnography in Cuba
Martin Holbraad
Part 2
Beyond Cultural Wholes? Wholes are Parts,
and Parts are Wholes
6. Beyond Cultural Wholes? Introduction to Part 2
7. The Whole is a Part: Intercultural Politics of Order and Change
Marshall Sahlins
8. Lingual and Cultural Wholes and Fields
Alan Rumsey
9. Deep Wholes: Fractal Holography in Trobriand Agency and Culture
Mark Mosko
9781444333237_2_toc.indd v
vii
ix
1
17
19
28
47
67
87
89
102
127
150
8/7/2010 6:40:37 PM
vi
Part 3
Contents
Beyond Structural Wholes? Encompassment, Collectives,
and Global Systems
10. Beyond Structural Wholes? Introduction to Part 3
11. Louis Dumont and a Holist Anthropology
Bruce Kapferer
12. From Wholes to Collectives: Steps to an Ontology of Social Forms
Philippe Descola
13. Holism and the Transformation of the Contemporary Global Order
Jonathan Friedman
Part 4
Beyond Social Wholes? Holistic Practice: Cosmology,
History, and the Continuity of Life
14. Beyond Social Wholes? Introduction to Part 4
15. Proportional Holism: Joking the Cosmos Into the Right
Shape in North Asia
Rane Willerslev and Morten Axel Pedersen
16. One Past and Many Pasts: Varieties of Historical Holism
in Melanesia and the West
Eric Hirsch and Daniele Moretti
17. Drawing Together: Materials, Gestures, Lines
Tim Ingold
Index
9781444333237_2_toc.indd vi
175
177
187
209
227
249
251
262
279
299
314
8/7/2010 6:40:37 PM
Acknowledgments
The idea “to do something about holism” came up during a board meeting of the
Danish Research School of Anthropology and Ethnography in autumn 2004. Funded
by the Danish Agency for Science, Technology, and Innovation, the Research School
has as its main task to support and enhance PhD training in anthropology and is a
collaborative enterprise between the Departments of Anthropology at Aarhus
University and the University of Copenhagen.
The idea arose from a discussion about what constituted the special characteristics
of anthropological research in comparison to other styles of inquiry within the overall
need to specify how one might best train doctoral students for a future of interdisciplinary collaborative work in a world that increasingly challenges the routines of traditional ethnographic fieldwork.
Ton Otto had just taken over as head of the Research School and organized a series
of 1-day workshops on holism in 2005, to which he invited Bruce Kapferer, Veena
Das, Tim Ingold, and George Marcus as guest professors. Next, holism was chosen as
central theme of the biennial conference of the Research School for 2006 under the
title ”Reinventing the Whole in a Global World”. These biennial conferences – which
with a customary Danish sense of ironic self-aggrandizement are called the Megaseminars
– are the main forum for anthropological debate in Denmark and gather staff, doctoral
students, and guests from abroad. Keynote speeches in 2006 were delivered by Eric
Hirsch, Martin Holbraad, Bruce Kapferer, Emily Martin, and Anna Tsing.
From this grew the idea that the theme of holism could work as a diagnostic device
to rethink the state of the art of anthropology. Ton Otto and Nils Bubandt decided
to join forces to produce an anthology on this issue and collaborated in organizing a
conference, “Beyond the Whole? Anthropology and Holism in a Contemporary
World,” in 2008. The conference was generously funded by The Research Priority
Area Globalisation at the Faculty of Humanities, Aarhus University, as well as by the
Danish Research School of Anthropology and Ethnography. We thank both institutions for supporting this conference. Without the conference this anthology would
not have been possible.
All contributors to the present anthology – with the exception of Daniele Moretti,
who joined the project later on – as well as Fredrik Barth, Andreas Roepstorff, and 10
doctoral students, participated in this conference. In addition Veena Das, Kirsten
9781444333237_3_posttoc.indd vii
8/7/2010 6:40:59 PM
viii
Acknowledgments
Hastrup, Emily Martin, and Marilyn Strathern were also invited but were unable to
attend, and due to various other commitments were also unable to produce a chapter.
We regret this of course but are pleased to note that a number of their ideas are still
part of this book.
We would like to thank Rosalie Robertson and Julia Kirk from Wiley-Blackwell for
their support of this project, and freelance project manager Nik Prowse and copy-editor Cheryl Adam for their excellent cooperation. Finally we would like to thank the
doctoral students and staff of the Danish Research School of Anthropology and
Ethnography for contributing to a productive and challenging intellectual
environment.
The work of editing this anthology and writing the introductions has been shared
equally by Nils Bubandt and Ton Otto.
Nils Bubandt, Aarhus
Ton Otto, Cairns
April 2010
9781444333237_3_posttoc.indd viii
8/7/2010 6:40:59 PM
Nils Bubandt
Afdeling for Antropologi og
Etnografi
Aarhus Universitet
Moesgaard
8270 Hojbjerg, Denmark
Philippe Descola
Collège de France
Laboratoire d’anthropologie sociale
52, rue du Cardinal-Lemoine
F-75005 Paris, France
Jonathan Friedman
Lilla Fiskaregatan 8A
222 22 Lund, Sweden
Aberdeen AB24 3QY
Scotland, United Kingdom
Bruce Kapferer
Institutt for sosialantropologi
Universitetet i Bergen
Post Box 7802
5020 Bergen, Norway
George E. Marcus
Department of Anthropology
3151 Social Science Plaza
University of California, Irvine
Irvine, CA 92697
United States
Eric Hirsch
Brunel University
Uxbridge
Middlesex UB8 3PH
United Kingdom
Daniele Moretti
Department of Social
Anthropology
University of Cambridge
Free School Lane
Cambridge CB2 3RF
United Kingdom
Martin Holbraad
University College London
14 Taviton Street
London WC1H 0BW
United Kingdom
Mark Mosko
College of Asia and the Pacific
Australian National University
Canberra ACT 0200
Australia
Tim Ingold
University of Aberdeen
Ton Otto
The Cairns Institute
James Cook University
9781444333237_3_posttoc.indd ix
8/7/2010 6:40:59 PM
x
PO Box 6811
Cairns Qld 4870
Australia
Morten Axel Pedersen
Institut for Antropologi
Københavns Universitet
Øster Farimagsgade 5, opgang I
1353 København K
Denmark
Alan Rumsey
College of Asia and the Pacific
Australian National University
Canberra ACT 0200
Australia
9781444333237_3_posttoc.indd x
Marshall Sahlins
5629 S. University Ave.
Chicago, IL 60637
United States
Anna Tsing
University of California
Santa Cruz, CA 95064
United States
Rane Willerslev
Moesgaard Museum
Moesgaard Allé
8270 Hojbjerg
Denmark
8/7/2010 6:40:59 PM
1
Anthropology and the
Predicaments of Holism
Why Bother With Holism?
We remember seeing a T-shirt inscription once: “Anthropologists Do It in Context.”
It was part of a string of T-shirt inscriptions – perhaps inspired by the 1990s “Just Do
It” advertising campaign by Nike – that used double entendres to describe professions: “Photographers Do It in the Dark” and “Landscape Gardeners Do It
Horizontally.” For most anthropologists, it is probably obvious what “doing it in
context” means. Context is about locating descriptions of particular phenomena
within a wider setting that throws light on these phenomena. It is about making sense
of observations by connecting them to larger experiential, meaningful, cultural, functional, or social wholes. Context is about grounding data; about methodological,
literary, and political circumspection; and about parts and wholes. Context, in short,
is about holism, one of the hallmarks – along with ethnographic fieldwork and intercultural comparison – of social and cultural anthropology.
As hallmarks go, however, holism is an odd one. For one thing, it is not given that
it means the same thing to all anthropologists – in fact, it is pretty clear that there is
no easy consensus. Second, holism is a highly problematic concept, and has been so
for several decades. The likely gut reaction of many contemporary anthropologists to
a volume on holism is therefore that holism is a fraught term that is best avoided.
Nevertheless, we will argue that in spite of its ambivalence and lack of consensus,
holism is still at the heart of the anthropological endeavor and that contemporary
qualms about the concept are in fact symptomatic of a new emergence and experimental approach to the anthropological tradition of holism. The contributions to this
volume demonstrate the variety and critical depth of current attempts to engage and
rethink anthropological holism.
For heuristic purposes, we will adopt a broad (and admittedly also somewhat vague)
definition of holism. We take holism to mean that a phenomenon has meaning, function, and relevance only within a larger context, field of relations, or “world” (see
Chapters 4 and 8). The term “context” derives from a hermeneutical tradition of textual interpretation and is an important part of a holistic perspective (Dilley 1999). This
tradition that blossomed under the influence of Geertzian interpretative anthropology
Experiments in Holism, edited by Ton Otto and Nils Bubandt © 2010 Blackwell Publishing Ltd except for editorial
material and organization © 2010 Ton Otto and Nils Bubandt
9781444333237_4_001.indd 1
8/7/2010 6:41:39 PM
2
Bubandt and Otto
sees the act of interpretation as the establishment of a relation between parts and
wholes: “Hopping back and forth between the whole conceived through the parts that
actualize it and the parts conceived through the whole that motivates them, we seek to
turn them, by a sort of intellectual perpetual motion, into explications of one another”
(Geertz 1983: 134). Context in this sense became part and parcel of a cultural holism.
Holism is not, however, synonymous with contextualization, and other anthropological traditions have their own kinds of holisms whose genealogies and internal ambivalences this volume explores: functional, structural, social, methodological, and
experiential holisms.
Holism may be said to be foundational for modern anthropology in the early
twentieth century. It is associated with the rise of modern anthropology, characterized by the centrality of ethnographic fieldwork; a variety of theoretical traditions, all
of which aspired to understand other forms of social life as integrated wholes; and a
particular form of realistic representation of these other life forms, typically using
media such as the monograph and ethnographic film. At the same time, however,
holism is notoriously problematic and vague. As a central anthropological cornerstone, as Marcus and Fischer noted already in 1986, holism “is currently undergoing
serious critique and revision” (in Marcus and Fischer 1999: 23). The reason for this
is the seemingly close relationship between holism, wholes, and totalization. This
relationship has implicated anthropology, as Sahlins notes in Chapter 7 of this volume, in a theoretical “scandal” that has become increasingly apparent in recent decades. Anthropological holism, it seemed, came to be a postulate about rather than a
search for wholes, conceived as totalities of culture, society, or ideology. The problem
here was both epistemological-ontological (what anthropological theory was set up
to capture and express) and methodological-practical (how fieldwork was delimited
and conducted). The holism of anthropological theory and ethnographic practice, it
became shockingly apparent, seemed geared toward asserting bounded, static, homogeneous wholes.
These problems are compounded by vagueness. Holism, Parkin notes, “seems to
refer to any approach that embraces an undivided view of society and humanity,
and so has little analytical worth” (2007: 3). Scandalously outdated, theoretically
suspect, and conceptually vacuous, holism also appears to smack of New Age
naïveté – and political correctness to boot – at a time when it seems that every
scientist and their healer are turning “holistic” (Fodor and Lepore 1991; Smuts
1999; Caruana 2000; Diamond 2001; Esfeld 2001; Jackson 2003; Pellegrini
et al. 2003).
Does it make sense to speak about anthropological holism under these circumstances? What insights does such a focus bring to an understanding of contemporary theory and practice in anthropology? We argue that it does make sense. In
fact, we argue that looking explicitly at holism again – its history, its problems, and
its (ab)uses, and the uncomfortable silences that often surround it – is an endeavor
that is long overdue. It is also an endeavor that may tell us something about anthropology that we may not have realized as well as something new about where
anthropology is currently going. Reflecting explicitly about holism provides, we
suggest, a fruitful vantage point from which the state of the art of anthropological
theory and practice can be considered in a new light. We take holism to be a
heuristic concept, a vague but nevertheless useful label that helps us uncover and
make explicit a central but contested concern in the style of inquiry we call
anthropology.
9781444333237_4_001.indd 2
8/7/2010 6:41:39 PM
Anthropology and the Predicaments of Holism
3
Holism in Anthropological Self-Representation
Holism is, in textbooks and in anthropological self-understanding, frequently presented as a central part of “the anthropological perspective” (Eller 2009: 13), and it
is often used to characterize the discipline in contrast to others (Nanda and Warms
2005). Nanda and Warms (2009: 6), for instance, put it this way:
Anthropologists bring a holistic approach to understanding and explaining. To say
anthropology is holistic means that it combines the study of human biology, history, and
the learned and shared patterns of human behaviour and thought we call culture in order
to analyze human groups. Holism separates anthropology from other academic disciplines, which generally focus on one factor – biology, psychology, physiology, or society –
as the explanation for human behaviour.
This understanding of holism as a comprehensive approach to the human condition
is widespread and is – in this formulation at least – closely connected to the American
view of anthropology as comprising four subfields – cultural anthropology, physical
anthropology, linguistics, and archaeology – which together allow a comprehensive,
holistic view on humanity (see Chapter 6; Harris 1993; Kottak 2006). The four-field
approach to holism is, however, only one of many streams of holism within the anthropological tradition. Holism has thus become central to a number of discussions about
what anthropology is and where it is going. Two of these discussions concern the particular strength of applied anthropology – again, in relation to more narrowly focused
disciplines (see Harris 1993: 428–9; Ferraro 2006: 14; Nanda and Warms 2007: 166–
7) and in connection with reflections about the method of participant observation,
often presented as the key characteristic of the discipline (Marcus 1998).
The practice of ethnographic fieldwork during which anthropologists live with and
partake in the lives of their informants in order to get better observational data –
hence the term “participant observation” – is seen as an important reason for the
development of a holistic perspective (Kloos 1974: 169; Blok 1977: 49; Kottak 2006:
262–3; Nanda and Warms 2007: 60). Holism, as Marcus and Fischer note, is “one of
the cornerstones of twentieth-century ethnography” (1999: 22–23). For a long time,
the solidity of this “cornerstone” was founded on the legitimacy of exotic fieldwork.
Because of the small scale of the societies they originally studied, so the argument
went, anthropologists were better able to see the integration between life spheres that
are seen as separate in more complex societies (kinship, religion, politics, and the
economy). These separate spheres or domains should therefore be seen as parts of a
social whole. Once it had been established as a genre, this holistic perspective also
proved useful in other, more complex contexts where face-to-face relationships no
longer are standard. Whether this assumed connection between ethnographic fieldwork and the development of holism is historically correct or not – we will argue that
there are other factors at stake as well – the textbook image of anthropologists employing a holistic perspective is very much part of the anthropological self-image, as that
which makes us unique. As Clifford Geertz has put it,
The specialness of “what anthropologists do,” their holistic, humanistic, mostly qualitative, strongly artisanal approach to social research, is (so we have taught ourselves to
argue) the heart of the matter. (2000: 93)
9781444333237_4_001.indd 3
8/7/2010 6:41:39 PM
4
Bubandt and Otto
This is an image about anthropology that has now been popularized outside of the
discipline, in other disciplines, and among the wider public. Indeed, Geertz has noted
the curious discrepancy that has come with this tendency to describe anthropology as
a particular style of research associated with holism, namely, that it is turning out to
be a highly successful way of promoting the discipline elsewhere while producing
“a certain nervousness, rising at times to something near panic,” within the discipline
(2000: 94). Below we shall seek to explain the history of this nervousness, while also
suggesting that it is pointing the way forward to a kind of anthropological holism that
is constituted by neither light-headed panic nor smug confidence.
Despite the many references to holism, explicit reflections about holism have not been
the rule in anthropology. If holism was a central axis in anthropology, it had more the
character of being a totem pole than an elaborated theoretical dimension. Holism was
an exclamation mark of the discipline, a watchword cried out, as Geertz has noted, at
conferences and in general call to arms (2000: 97). But it was rarely elaborated on at any
length in writing. For all its emotional force as a disciplinary totem, holism was rarely
given theoretical depth. It is thus surprising how rarely holism was mentioned in what
we would now see as its heyday in the early and middle decades of the twentieth century.
Thus, in most classical textbooks, either holism is not mentioned at all or its treatment
is extremely cursory.1 The centrality of holism in anthropological thought and practice
cannot therefore be fully traced in the classic writings of the discipline. Of course, there
are the loci classici of holism, which we deal with below and detail in Chapters 2, 6, 10,
and 14 of this volume, but it is our strong impression that there is another, more hidden
life of holism in the practice of teaching the discipline. Especially anthropology teachers
who are now in their fifties or older remember that holism was one of the key terms used
to characterize the discipline and its methods. But much of this is oral culture.
One of us recently discussed the issue with a well-known visual anthropologist who
had been a student of Margaret Mead. For him the criterion of success of a good ethnographic film or book was the holistic representation of indigenous culture, and he emphasized that this has been one of the key themes in Mead’s teaching. Asked where we could
find this theme unfolded, he referred us to Mead’s introductory text in Principles of
Visual Anthropology (Hockings 1975). To our surprise, however, the word “holism” is not
mentioned at all there. Although Mead refers briefly to holism elsewhere in her writings
(see, for example, Mead 1953),2 it was evidently much more central in Mead’s teaching than in her writing. It was left to others to explicate it. Indeed, in his book on ethnographic film that he dedicated to Gregory Bateson and Mead, Karl Heider, another visual
anthropologist, highlights holism as one of three central principles of ethnography –
along with detailed description based on long-term fieldwork and the ideal of relating
observed behavior to general norms – that are central to ethnographic filmmaking. For
Heider, holism – understood as the need to understand things and events in their cultural
and social contexts (Heider 2006: 6) – is a reminder of the importance of allowing the
camera to capture “whole bodies, whole interactions, and whole people in whole acts”
(Heider 2006: 114). Heider does not explicitly credit Mead for this importance of holism,
but it is clearly part of the credo and tonality of the tradition of Franz Boas and Mead.
Anthropologists write, as Geertz has famously quipped (1973: 19). It would seem,
however, that they do not write everything down. We argue that despite being one of
the central tenets of our discipline, holism has been the object of talk rather than of
writing. Just like anthropologists return from the field with written fieldnotes as well
as unwritten “headnotes” (Ottenberg 1990), it would seem that holism for generations of anthropologists was a “headnote,” a part of the common theoretical imaginary and corridor discourse that were central to a sense of métier and professional
9781444333237_4_001.indd 4
8/7/2010 6:41:39 PM
5
identity but that only rarely and sporadically made it into theoretical writings. This
headnote quality of holism, its existence as a totemic catchword for what anthropologists told their students, each other, and the rest of the world about what they did,
rather than a theoretical term they reflected much upon in writing, was not restricted
to North America.
In the English tradition, too, holism was around already at an early stage. In conventional accounts of the origins of “modernist” British anthropology, the rise of
fieldwork-based (structural-)functionalism was thus associated with a “revolution” in
which armchair anthropology and conjectural history were “rejected in favor of the
discovery of holism and synchrony” (Strathern 1987: 258). The fragmented comparison of myth and ritual by Tylor and Frazer was replaced, so the story goes, by the
insistence of Malinowski and Radcliffe-Brown that “practices and beliefs were to be
analyzed as intrinsic to a specific social context; that societies so identified were seen
as organic wholes, later as systems and structures; and that the comparative enterprise
which modern anthropologists set themselves thus became the comparison of distinct
systems” (Strathern 1987: 254). As Stocking and Strathern have shown, this idea of a
fundamental revolution is a particular disciplinary fiction (Stocking 1983), but it
allowed the rise of a new form of holism centered on fieldwork. This in turn generated
a successful, but also largely fictional, break that rendered the evolutionist tradition,
as Strathern suggests, “unreadable,” while it made possible the discipline of modern
anthropology (Strathern 1987: 269). If the revolution was a fiction, it was a persuasive one, for it recalibrated anthropological self-understanding as a scientific form of
knowledge in all its aspects. Holism was thus related to new theoretical aspirations, to
the new methodological imperatives of fieldwork, and to a new genre convention, all
constituting novel standards of proper knowledge and valid representation.
Despite the importance of this paradigm shift for the self-understanding of modern
anthropology, holism also rarely floated to the surface of written texts in British
anthropology as a theoretical concept.3 Hortense Powdermaker, for instance, mentions that she had been trained in a “holistic frame” by Bronislaw Malinowski at the
London School of Economics; by this she means the new standard Malinowskian idea
of holism described above, namely, that she would always put the issue or group studied in the context of its relations to other issues and groups (1971 [1933]: 8). She
writes this in a later (1971) introduction to her ethnography on Lesu (from 1933), in
which, typically we think, the term is not used at all.
It would seem, in other words, that as a cornerstone of twentieth-century anthropology, holism acquired the quality of an anthropological doxa. It appears to have
been for a long time an aspect of anthropological practice and writing that was rarely
discussed. It was something that did not need reflection in writing because it came
“without saying,” as Pierre Bourdieu has it (1977: 167). In the context of introducing novices to the discipline, it probably had a more prominent role, especially in the
classroom. Holism was an anthropological tradition, and like all traditions it was
mainly “silent, not least about itself as a tradition” (Bourdieu 1977: 167).
The Abject Heart of Anthropology
The 1980s, however, marked a break. Postmodern critique, a concern with pluralism
and multivocality, the attack on totalizing theoretical paradigms, experiments with
new forms of representation, and not least a changing world made holism as
9781444333237_4_001.indd 5
8/7/2010 6:41:39 PM
6
Bubandt and Otto
an unarticulated ideal a problem. The wholes implicated by conventional holism –
culture, society, ideology, social organization, and symbolic system – began to appear
theoretically unsuitable and politically suspect. The 1980s, in other words, provided
anthropology with a new language about itself. This included a hitherto novel way of
being explicit about, and often highly critical toward, anthropological wholes and
anthropological holism. Despite trenchant critique, however, holism lingered. New
forms of holism (like globalization) emerged, and old ones (like culture) refused to
disappear, perhaps because the anthropological style of inquiry depended on such
notions (cf. Marcus, Chapter 3, this volume; Thornton 1988) and perhaps also
because the human worlds studied by anthropologists were replete with them (see
Tsing [Chapter 4], Holbraad [Chapter 5], Mosko [Chapter 9], Pedersen and Willerslev
[Chapter 15], and Hirsch and Moretti [Chapter 16], this volume).
As a result of these changes over the last three decades or so, holism has now
acquired an ambivalent, if still central, position in anthropology. Our own disciplinary
fictions established holism as enduringly central to anthropological theory, practice,
and self-representation, but left it vaguely articulated – part of programmatic selfdefinition rather than concrete analysis, and implicit in anthropological writings rather
than explicitly formulated. The critique of recent decades added new complications.
Critique lifted holism to the surface of our writings but as a problematic – even dangerous – anthropological idea(l). Bringing the concept into the light of reflexive scrutiny revealed its shadowy side, namely, the totalizing assumptions of the theoretical
models and ethnographic representations based on this ideal.
As a consequence, we suggest, holism now has the status of being both central and
abject in anthropology. Julia Kristeva calls abjection one of the “dark revolts of being”
(2002: 389). It describes the process of throwing away or casting aside a part of self
through which the self comes into being. It is by ridding oneself of the abject – a something that fails to be entirely named or captured – that one becomes a self in the first
place. The abject is therefore not an object; rather, it is a something that simultaneously creates the borders of the self as an object and makes possible the self as a subject.
Identity begins, in other words, with abjection. Using the example of food repulsion,
Kristeva argues that it is through disgust, a corporeal vomiting, that identity is established: “I expel myself, I spit myself out, I abject myself in the same motion through
which ‘I’ claim to establish myself” (2002: 390). This claim, however, can be tenuous
at best, and abjection cannot fully succeed. Since the abject is both part of myself
and intolerable, my attempt to expel it from me is both necessary and impossible
(Kristeva 1982). An illustrative pop image of abjection is the creature Sméagol in the
Hollywood motion picture based on J. R. R. Tolkien’s opus The Lord of the Rings. In
the trilogy, Sméagol constantly has to stop midsentence to try to cough up or vomit
out Gollum, his alter ego whose obsessive desire for the ring has consumed Sméagol.
The name “Gollum” is derived from the sound of Sméagol’s abjection, so the more he
attempts to cough Gollum up, the more Sméagol becomes Gollum. Such is the irony
of abjection.
One may disagree with the structural psychology that informs Kristeva’s notion of
abjection (Fraser 1990), but the notion of abjection is nevertheless helpful, we suggest, as a device through which to understand anthropology’s ambivalent relationship
to holism, and the way critique and disavowal of holism have seemingly not led to its
successful expulsion. Instead, it would appear that the critique of the last three decades of the various forms that anthropological holism may take (culture, ideology,
system, and society – abject wholes that have been easy to name but extremely difficult
9781444333237_4_001.indd 6
8/7/2010 6:41:39 PM
7
to handle theoretically) has led not to the rejection of holism but to new forms of
experimentation with and reflection on holism that are contributing to the reinvention and rethinking of the discipline.
Holism, we suggest, is emerging as the abject whose impossible expulsion from the
center of the discipline makes anthropology once again possible, tenuously, in the
contemporary world. The critiques of old forms of wholes, we suggest, have become
part of a redefinition of disciplinary identity, and with this redefinition have come new
forms of theory and practice – new ways of doing anthropological theory, analysis, and
fieldwork and new ways of relating to texts, to informants and collaborators, to readers, and to power. We are now in a moment when critique itself appears inadequate
and when we are concerned to be critical while “getting on with anthropology”
(Metcalf 2002). Under these circumstances, holism has become the abject heart of
anthropology, a problematic part of our self-identity of which we have sought to
purge ourselves but without which our discipline apparently cannot exist.
The current moment is therefore accompanied by an array of experiments with new
forms of holism that remain imbedded in modernist anthropology but that also seek
to be better attuned to new global conditions and new disciplinary practices. This
book brings together a range of different experiments in holism that critically deal with
the anthropological tradition but in new and challenging ways show the continued but
also fraught relevance of a holistic perspective. The contributors discuss the varied
heritage, current challenges, and critical potential of holism by lodging their discussions within concrete ethnographic analyses. In doing so, they suggest a variety of ways
in which holism might be reconceptualized, refashioned, and retooled as part of their
own efforts to get on with anthropology and make it relevant for the present world.
A Brief History of Holism
The anthropological ambivalence concerning holism is at least partly grounded in the
history of Western social, political, and scientific thought and the way the concept of
the whole made its entry into the discipline. We cannot pursue the intellectual history
of Western holism in any detail here, but we would nevertheless like to highlight two
important moments in Western political and scientific thought, namely, the birth of
individualism in the seventeenth century and the rise of holism in the life and human
sciences at the beginning of the twentieth century. At the outset, we should make it
clear that we use holism in a wide sense to indicate theories that emphasize wholeness
or collectiveness. The term “holism” itself was very likely coined or at least popularized by the South African statesman Jan Christiaan Smuts (Harrington 1996: xxii;
Smuts 1999 [1926]) in the 1920s. In the early texts of the founding fathers and
mothers of anthropology, words like “complex whole,” “the whole phenomenon,”
“total social fact,” and “collective representations” are found throughout, but it is
only retrospectively that their theories have been called holistic.
Looking for the intellectual origins of the concept of holism, one could argue that
Western individualism and holism (in the shape of collectivism) were born as a conceptual pair. In the work of the seventeenth-century philosophers Thomas Hobbes
and especially John Locke, individuals are defined as the natural proprietors of their
own persons and capacities, not owing anything to society for them. For this individualism to be able to exist as a social form, however, individuals have to accept a supreme
9781444333237_4_001.indd 7
8/7/2010 6:41:39 PM
8
Bubandt and Otto
state that protects individual property and therefore transcends the rights and powers
of the individual. The individual is therefore necessarily encompassed by a legal and
economic whole that is more than the sum of its parts, namely, all the individuals who
are subjects of the state. In the liberal tradition, “possessive individualism” (Macpherson
1962) and holism, in other words, logically imply each other, although the dominant,
explicitly stated value is that of individual freedom and choice.
Both the individual, a free and equal actor on the marketplace, and the whole upon
which sovereignty is based – the state – are presented as bounded wholes, but the
existence of the latter is legitimated with reference to the existence of the former. This
hierarchical relation supports the central observation by the anthropologist Louis
Dumont (1977, 1986) that individualism is the dominant value in modern Western
culture. Interestingly, Dumont assumes that most non-Western societies are characterized by the opposite value hierarchy, with the whole being dominant over the
individual. This is what he calls “holism” and opposes to Western individualism.
Dumont’s work has inspired a great deal of comparative ethnography supportive of
his thesis,4 as well as fierce criticism aiming to break open this central dichotomy (see
Chapters 11, 12, and 13, this volume), but important for the present argument is that
his comparative framework echoes concepts that are inherent in Western thought and
are informed by Western political practice. In line with this, as Richard Handler
(1988), among others, has pointed out, the political tradition of possessive individualism articulated well with modern ideas of “possessive culture,” a concept developed
to explain the role of culture in the construction of national – and ethnic – identities.
In this view the nation-state, conceived as an individual-writ-large, constitutes its
bounded individuality – a whole – through the ownership of its unique culture (cf.
Harrison 2006). We may thus observe a conspicuous historical convergence between
the modernist anthropological project of understanding culture as a holistic phenomenon and that of nationalist movements all over the world aiming to create holistic
and bounded cultures to establish their separate identities.
The liberal roots of holism notwithstanding, from the inception of modern anthropology the concept of the whole primarily served the function of allowing the discipline to critique this same liberalism. Anthropologists thus often conceptualized the
kinds of wholes in which other people lived their lives in critical contrast to life in the
West, which was understood as being – for better or for worse – institutionally and
existentially fragmented. The wholes – social, cultural, structural, and materialist –
that early-twentieth-century anthropologists saw in the lives of others were all meant
to undermine the taken-for-granted assumption in the West about the relationship
between the individual and society. Indeed, all the major traditions of anthropological
thought since the early twentieth century insisted that Western ideas of the individual
self were inadequate, even misplaced, and that religious systems, social functions,
structures of kinship, and modes of production meant that the analysis needed to begin
at the level of society or culture as a whole. This criticism of Western individualism –
simultaneously epistemological, political, and moral – is arguably inherent in the rise
of individualism as an ideology (as its counterpoint, so to speak) and was developed
by the early modern anthropologists at the beginning of the twentieth century in their
study of non-Western life worlds.
The turn to a holistic understanding of society and culture, accomplished by scholars such as Boas, Malinowski, Durkheim, and Mauss, was not an isolated phenomenon
within the emerging discipline of anthropology. In the relatively young discipline of
sociology, there was a similar concern with the impact of individualism on modern
9781444333237_4_001.indd 8
8/7/2010 6:41:40 PM
9
society as famously expressed by Ferdinand Tönnies’ (1912 [1887]) conceptual opposition between a Gemeinschaft (a holistic community with shared values) and
Gesellschaft (a community of individuals). This conceptual opposition was reinforced
by another contrast that was engendered by the success of modern science in the
nineteenth century and the concomitant industrialization of Western economy and
society. The metaphor of the machine as a model for proper knowledge, not only of
physical nature but also of life, mind, and society, was gaining wide support, due to
the evident successes of physics and engineering. This led, however, to a reaction
among a number of biologists, neurologists, and psychologists at the end of the nineteenth and the beginning of the twentieth centuries. This reaction against the metaphor of the machine was articulated as an emergent fascination with holism, a tendency
that was particularly strong in German-speaking countries even though it was by no
means restricted to this cultural area (Harrington 1996).
Whereas the sociologist Max Weber lamented but ultimately accepted the (for him)
unavoidable disenchantment of modern science, biologists like Jakob van Uexküll,
neurologists like Constantin von Monakow and Kurt Goldstein, and Gestalt psychologists like Max Wertheimer worked to redefine – and reenchant – their science by means
of a dominant concept of wholeness (Ganzheit). In opposition to the pervasive
mechanic worldview, which they found lacking in explanatory power as well as moral
orientation, they introduced concepts like Umwelt (environment or context), organic
wholeness, and psychological Gestalt. Harrington (1996) emphasizes how this new
holistic science not only had an internal scientific motivation and dynamic, but also
became entangled with the political developments of the time. In particular, some of
the holistic ideas of this new science articulated well with the emerging ethos of Nazism
and thus became enmeshed with anti-Semitism. This is not to say that academics who
were fascinated by the notion of wholeness necessarily had leanings toward this political ideology; quite a few, for example the Gestalt psychologist Wolfgang Köhler (Ash
1998), were in open opposition to the fascist regime and were forced to leave German
academia. But the holistic ideas could be adapted to support a totalitarian regime and
its exclusive ideology, thus showing the shadow side of holism: its potential, if used
uncritically, to totalize and exclude alternative explanations and worldviews.
Although cultural and social anthropologists mostly steered clear of Nazi totalitarianism and more often than not were in ideological opposition to it – Boas’ criticism
of the concept of race is a case in point – the genealogy of holism makes it clear that
it is prone to be used in totalizing ways. The recent upsurge of fundamentalist ideas
about religious incompatibility, ethnic cleansing, and the clash of civilizations comprises examples of holistic concepts gone wild as totalizing devices (Stolcke 1995).
Indeed, part of the recent critique of the anthropological concepts of culture, ethnicity, and society concerned the way they have assumed a politically totalizing character
in the world that makes their analytical use politically suspect (Clifford 1988; AbuLughod 1991; Brightman 1995; Fox and King 2002).
The Scope of the Book
Three decades of critique of anthropology – born as a discipline with a notion of
holism that has both a problematic heritage and a critical potential – have forced the
discipline to look at itself afresh. What – if anything – is left of anthropological holism
9781444333237_4_001.indd 9
8/7/2010 6:41:40 PM
10
Bubandt and Otto
as a style of research in the wake of this critique? We think that the contemporary
theory and practice of anthropology demonstrate that a great deal is left. The critique
has deconstructed the totalizing aspects of holistic perspectives but has thereby generated the debris with which anthropology is now experimenting, putting the pieces
together in new ways, joining the holisms of various theoretical traditions in new
ways – bricolage-like within their own practice.
What forms and aspects of anthropological holism were wrongly dismissed or
retained, and what are in dire need of rethinking? Can anthropological holism be salvaged by new kinds of theory, or does holism itself constitute a kind of practical theory?
Does anthropology need to forget holism, or has it not been holistic enough? These
are some of the questions that run through the contributions to this anthology.
If nothing else, we hope this introduction has demonstrated that holism is much more
than cross-disciplinarity (cf. Parkin and Ulijaszek 2007). Holism, as we see it, is rather a
style of research that has been inherent in anthropology since its birth as a modern
discipline in the early twentieth century but has a genealogy that can be traced back to
and is entangled with some of the major traditions of Western social theory. It was
appropriated by anthropology exactly to transcend the ideological limits of these social
theories. And the predicament of this heritage has stayed with the concept even as it was
lodged within a variety of anthropological theories and practices. In this book, we use
holism, therefore, as a diagnostic device. It is a device that allows the contributors not
only to critically engage this complex heritage, but also to look ahead to the future of
anthropology. The contributions to this volume offer a broad – and we think broadly
representative – selection of contemporary experiments in holism. Being anthropologists, the contributors conduct these experiments mostly in nonconformist, often idiosyncratic ways that are concerned less with the production of a new common theory or
solution than with trying to understand their own ethnographic material. This volume,
therefore, seeks to query the state of the art of contemporary anthropology through the
prism of holism, and to debate the future of anthropological forms of holism. It does not
offer a new paradigm of holism but presents a palette of the different ways in which the
paradigms of yesteryear are being deconstructed and reconstructed across conventional
boundaries to critically rethink the problems of old forms of holism while maintaining
and reviving the critical potential that all forms of anthropological holism harbor.
The contributions to this volume suggest that it is time to move beyond the nervous, even panicky, relationship that Geertz felt anthropologists maintained with their
heritage of holism (2000: 94). The contributions are all highly critical of the various
traditions of holism out of which they grew, but they use this critique to launch a
series of experiments in holism that point toward the possibility that anthropology can
be holistic without being totalizing, that there can be holisms without wholes. It is
therefore an invitation to open reflection and experimentation with the holisms that
continue to guide – and haunt – anthropology.
Despite the general willingness to experiment with holism, there is among the contributors little consensus about how this should be worked out in practice. In part this
heterogeneity is related to the diverse traditions of holisms that the contributions
engage. As the volume will show, the four hegemonic kinds of holism within anthropology – the “methodological” holism of ethnographic fieldwork, cultural holism,
structural holism, and social holism – were themselves never uncontested and were
characterized by internal division and dissent. This complexity is further complicated
when the contributors assemble new varieties of holism from this puzzle across disciplinary, national, and philosophical boundaries.
9781444333237_4_001.indd 10
8/7/2010 6:41:40 PM
11
If there is something on which all contributions appear to agree, albeit in their
different ways, it is the critical potential of holism, which makes it both central and
essential to the anthropological endeavor. The contemporary critical potential of
holism is both epistemological and culture-critical or -political. It is a necessary part
of the anthropological style of knowledge, if this involves making sense of other people’s actions and intentions. The critical potential of holism in this sense traverses, as
Marcus suggests in Chapter 3, the entire process of anthropological knowledge production: from the holisms associated with fieldwork, via the holisms involved in
anthropological reflexivity and representation, to the holisms involved in redesigning
anthropological analysis and comparison. At the same time, the holistic ambition to
decipher contrasting and contradictory “life worlds” gives anthropology a unique
vantage point from which to reflect critically on the constitution of these worlds. In
this lies a political ambition to transcend the situatedness of Western knowledge.
Anthropological holism, if it is possible to speak of it in the singular, is in other words
a style of inquiry and a way of putting context into academic play that affords a critical
perspective on the conditions of anthropological knowledge itself.
At the same time though, all contributions are critical of aspects of the way holism
has been used in the past, for example the construction of totalizing or reifying theoretical structures or the pursuit of exclusive identity projects. Holism is thus both an
object and a necessary condition for anthropological critique. As such, it remains
uneasily at the core of anthropological theory and practice – even after three decades
or more of disciplinary self-critique. But, as many of the contributions suggest, holism
should perhaps also be understood as part and parcel of human practice in general.
When human beings act, they imagine – implicitly and explicitly – contexts in which
their actions make sense and in which they and others figure as agents. These practical
holisms vary in scale and scope and can be more or less totalizing in their effects. It is
anthropology’s difficult but socially and academically important task to critically
illuminate, describe, and engage these holisms in the world.
The Structure of the Book: Four Kinds
of Experiments in Holism
The volume is divided into four sections, each of which presents four broad varieties
that current experiments in holisms may take. Contemporary anthropology, we suggest, may be divided into at least four different strings of experiments in holism that
each engages a particular genealogy of holism: the holism in ethnographic representation and practice, cultural holism, structural holism, and social holism, respectively.
This division is not without its problems. A division between textual, cultural,
structural, and social kinds of holisms clearly runs the risk of reviving conventional
distinctions between American, German, French, and British traditions of anthropology and fetishizing national boundaries of anthropological theory and practice.
Insisting on such national divisions would be unfortunate, because it would repeat a
conventional anthropological historiography that ignored contacts, overlaps, and
cross-influences. In fact, the contributions to each section demonstrate the global
scope of contemporary reflections on holism. Tsing in Chapter 4, for instance, brings
together Latour from France and Strathern from the UK to solve a problem of
American and Japanese scholarship. In Chapter 11 Kapferer, an Australian living in
9781444333237_4_001.indd 11
8/7/2010 6:41:40 PM
12
Bubandt and Otto
Norway, uses the theories of Frenchmen like Dumont and Deleuze to engage the
heritage of the Manchester school. Nevertheless, even though anthropological ideas
continually transcend national borders and disciplinary divides, the national histories
of the development of these ideas and divides still matter, and it is evident that the
reflections by our contributors on how the discipline should reorient itself to answer
the challenges of the contemporary world do grow from a variety of different conceptions of holism, each with its own intellectual heritage, characteristics, problems,
and potential.
A concomitant problem of the division we have chosen is that it is in danger of forgetting traditions of holism that fall between or outside these national spaces, either
because they fail to easily slot into the four dominant kinds of holism (and here the
ecoholism of Bateson is a prominent example; see Dudgeon 2008), or because they
belong to “other” kinds of anthropology outside of the hegemonic Global North
(Eades et al. 2004; Ribeiro and Escobar 2006). We have to admit in part to this kind
of forgetting, if for lack of space rather than lack of will. The kinds of holisms the contributions critically engage are clearly hegemonic kinds of holism. It is, however, also
this hegemony that makes a critical engagement with them all the more important. We
think that the selection of contributions to this volume take on those genealogies of
holism that have had a formative effect on anthropology on a global scale. It is, however, by no means exhaustive. There are clearly other possible experiments in holism
than the ones conducted in this volume, experiments that grow more explicitly out of
a critical engagement with ecology (Dudgeon 2008), with Marxism (Jay 1986; Shiell
1987; but see also Friedman, Chapter 13, this volume), with the natural sciences
(Esfeld 2001; Parkin and Ulijaszek 2007), and with phenomenological or narrative
theory (Weiner 1999; but see also Holbraad [Chapter 5], Bubandt and Otto [Chapter
14], and Pedersen and Willerslev [Chapter 15], this volume), to mention but a few.
In Part 1, the focus is on the implicit and explicit holisms that appear in the practice
of ethnographic research and writing. It includes reflections on how an aesthetic of
holism is fundamental to the design and execution of ethnographic projects, how the
conception of what Anna Tsing calls “worlding” (see Chapter 4, this volume) informs
the choices we make while tracing explanatory connections, and how the epistemological premise of holism in ethnography is the methodological conditio sine qua non
for theoretical openness and creativity.
Part 2 takes up the heritage of the predominantly American and German tradition
of cultural anthropology. This heritage has conventionally emphasized cultures as
autonomous and sui generis wholes, a convention that has been under attack for decades. The outcome of what to do with culture is, however, still muddy. As a way of
dealing with the challenges of this critique and incorporating it into a practical research
agenda, the contributions discuss how a focus on these “wholes” might be decentered. One option, advocated by some of the contributors, is to show that “wholes”
can only be understood within greater historical fields of mutually determining
dependencies. Using insights from linguistics, they argue that “wholes” are defined
not only from the theoretical perspective of the researcher (as in de Saussurean linguistic analysis) but also on the basis of political identity projects. An alternative
option for a decentered approach to cultural holism, this section suggests, may be
developed by using concepts from modern complexity theory, in particular the application of fractal holography to anthropological data.
In Part 3, the common ground for reflection and experimentation is the inspiration
from the French structuralist tradition, with Louis Dumont as the centrally important
9781444333237_4_001.indd 12
8/7/2010 6:41:40 PM
13
protagonist defending the relevance of a holistic approach. A thorough analysis of the
Dumontian anthropological program of comparative analysis of value hierarchies and
encompassments is the basis for an exploration of how this program can be developed
to deal with transformations, events, and modern forms of subjectivity. In addition, the
program is extended to map and compare different types of collectives, characterized
by different ontologies concerning the relations between humans and nonhumans and
the relative importance ascribed to interiority versus exteriority. Finally, the structural
consequences of global interdependencies and political inequalities for cultural ontologies and identities are brought within a comprehensive analytical perspective.
Part 4 addresses the heritage of the predominantly English tradition of social
anthropology, which conceives of social systems as organic, mechanical, cybernetic, or
theatrical wholes. The contributors to this section all seek to discard the preconceived
social wholes that come with what Tim Ingold (Chapter 17, this volume) calls the
“painting” approach to anthropological description, an approach that seeks the totality to fill out all the blanks of the ethnographic picture. Instead, the contributions to
this section seek to follow the connective loops of human actions and imaginations
that “draw” the emergent lines of partial, temporary, and emic holisms. In order to
grasp the holisms of human practice in concrete ethnographic situations – whether in
relation to narrating the past (Hirsch and Moretti, Chapter 16, this volume) or in relation to joking the ancestors into their proper place (Willerslev and Pedersen, Chapter
15, this volume) – bold conceptual theoretical experiments are carried out, which give
this brand of anthropology its particular innovative appeal.
Notes
1 There are, of course, exceptions. See, for example, Blok (1977: 49–69). This textbook deals
with holism in a more comprehensive way by devoting a whole chapter to it. It argues that
holism is one of four central perspectives that characterize anthropology. The other three
are participant observation, the comparative perspective, and the development perspective.
2 The concept of holism is very central, but again in an implicit way, in Mead’s monograph New
Lives for Old (1956), which deals with how a culture does not change in a piecemeal way, but
rather as a whole.
3 Again, there are exceptions. See, for instance, Gellner (1956).
4 See Barraud et al. (1994) for a Melanesian example.
References
Abu-Lughod, Lila (1991). Writing Against Culture. In Recapturing Anthropology: Working in
the Present, ed. R. Fox. Santa Fe: School of American Research Press: 137–62.
Ash, Mitchell G. (1998). Gestalt Psychology in German Culture 1890–1967: Holism and the
Quest for Objectivity. Cambridge: Cambridge University Press.
Barraud, Cécile, Daniel de Coppet, and André Iteanu (1994). Of Relations and the Dead: Four
Societies Viewed from the Angle of Their Exchanges. Oxford: Berg.
Blok, Anton (1977). Antropologische Perspectieven. Muiderberg, the Netherlands: Coutinho.
Bourdieu, Pierre (1977). Outline of a Theory of Practice. Cambridge: Cambridge University
Press.
Brightman, Robert (1995). Forget Culture: Replacement, Transcendence, Relexification.
Cultural Anthropology 10(4): 509–46.
Caruana, Louis (2000). Holism and the Understanding of Science. Aldershot: Ashgate.
9781444333237_4_001.indd 13
8/7/2010 6:41:40 PM
14
Bubandt and Otto
Clifford, James (1988). The Predicament of Culture: Twentieth-Century Ethnography, Literature,
and Art. Cambridge, MA: Harvard University Press.
Diamond, John (2001). Holism and Beyond: The Essence of Holistic Medicine. Ridgefield, CT:
Enhancement Books.
Dilley, Roy (1999). Introduction: The Problem of Context. In The Problem of Context, ed.
R. Dilley. New York: Berghahn, 1–46.
Dudgeon, Roy (2008). The Pattern Which Connects: Batesonian Holism and Postmodern Science.
Raleigh, NC: Lulu.com.
Dumont, Louis (1977). From Mandeville to Marx: The Genesis and Triumph of Economic
Ideology. Chicago: University of Chicago Press.
Dumont, Louis (1986). Essays on Individualism: Modern Ideology in Anthropological Perspective.
Chicago: University of Chicago Press.
Eades, J. S., Yamashita, Shinji, and Joseph Bosco, eds. (2004). The Making of Anthropology in
East and Southeast Asia. New York: Berghahn.
Eller, Jack David (2009). Cultural Anthropology: Global Forces, Local Lives. New York: Routledge.
Esfeld, Michael (2001). Holism in Philosophy of Mind and Philosophy of Physics. Dordrecht:
Kluwer.
Ferraro, Gary (2006). Cultural Anthropology: an Applied Perspective, 6th ed. Belmont, CA:
Wadsworth/Thomson.
Fodor, Jerry, and Ernest Lepore (1991). Holism: A Shopper’s Guide. Oxford: Wiley-Blackwell.
Fox, Richard, and Barbara King, eds. (2002). Anthropology Beyond Culture. Oxford: Berg.
Fraser, Nancy (1990). The Uses and Abuses of French Discourse: Theories for Feminist Politics.
Boundary 2 17(2): 82–101.
Geertz, Clifford (1973). Thick Description: Towards an Interpretive Theory of Culture. In The
Interpretation of Cultures, ed. C. Geertz. New York: Basic Books, 3–30.
Geertz, Clifford (1983). Local Knowledge: Further Essays in Interpretive Anthropology. New York:
Basic Books.
Geertz, Clifford (2000). Available Light: Anthropological Reflections on Philosophical Topics.
Princeton: Princeton University Press.
Gellner, Ernest (1956). Holism Versus Individualism. In Readings in the Philosophy of the Social
Sciences, ed. M. Brodbeck. London: Collier-MacMillan, 254–68.
Handler, Richard (1988). Nationalism and the Politics of Culture in Quebec. Madison: University
of Wisconsin Press.
Harrington, Anne (1996). Reenchanted Science: Holism in German Culture From Wilhelm II to
Hitler. Princeton: Princeton University Press.
Harris, Marvin (1993). Culture, People, Nature: An Introduction to General Anthropology. New
York: HarperCollins.
Harrison, Simon (2006). Fracturing Resemblances: Identity and Mimetic Conflict in Melanesia
and the West. New York: Berghahn.
Heider, Karl (2006). Ethnographic Film. Austin: University of Texas Press.
Hockings, Paul, ed. (1975). Principles of Visual Anthropology. The Hague: Mouton.
Jackson, Michael (2003). Systems Thinking: Creative Holism for Managers. Oxford: Wiley.
Jay, Martin, ed. (1986). Marxism and Totality: The Adventures of a Concept From Lukács to
Habermas. Berkeley: University of California Press.
Kloos, Peter (1974). Culturele Antropologie. Assen: Van Gorcum.
Kottak, Conrad Phillip (2006). Anthropology: The Exploration of Human Diversity. Boston:
McGraw-Hill.
Kristeva, Julia (1982). Powers of Horror: An Essay on Abjection. New York: Columbia University
Press.
Kristeva, Julia (2002). Approaching Abjection. In The Feminism and Visual Culture Reader,
ed. A. Jones. London: Routledge, 389–91.
Macpherson, C. B. (1962). The Political Theory of Possessive Individualism: Hobbes to Locke.
Oxford: Oxford University Press.
9781444333237_4_001.indd 14
8/7/2010 6:41:40 PM
15
Marcus, George E. (1998). Ethnography Through Thick and Thin. Princeton: Princeton
University Press.
Marcus, George E., and Michael M. J. Fischer (1999). Anthropology as Cultural Critique: An
Experimental Moment in the Human Sciences, 2nd ed. Chicago: University of Chicago
Press.
Mead, Margaret (1953). National Character. In Anthropology Today: An Encyclopedic Inventory,
ed. A. L. Kroeber. Chicago: University of Chicago Press, 642–67.
Mead, Margaret (1956). New Lives for Old. London: Gollancz.
Metcalf, Peter (2002). They Lie, We Lie: Getting on With Anthropology. London: Routledge.
Nanda, Serena, and Richard L. Warms (2005). Anthropology: A Holistic Approach. Belmont, CA:
Wadsworth.
Nanda, Serena, and Richard L. Warms (2007). Cultural Anthropology, 9th ed. Belmont, CA:
Thomson Wadsworth.
Nanda, Serena, and Richard L. Warms (2009). Culture Counts. Belmont, CA: Wadsworth.
Ottenberg, Simon (1990). Thirty Years of Fieldnotes: Changing Relationships to the Text. In
Fieldnotes: The Makings of Anthropology, ed. R. Sanjek. Ithaca: Cornell University Press,
139–60.
Parkin, David (2007). Introduction: Emergence and Convergence. In Holistic Anthropology.
Emergence and Convergence, ed. D. Parkin and S. Ulijaszek. New York: Berghahn, 1–20.
Parkin, David, and Stanley Ulijaszek, eds. (2007). Holistic Anthropology: Emergence and
Convergence. New York: Berghahn.
Pellegrini, Claudio, Paula Cerrai, Paolo Freguglia, Vieri Benci, and Giorgio Israel (2003).
Determinism, Holism, and Complexity. Berlin: Springer.
Powdermaker, Hortense (1971). Life in Lesu: The Study of a Melanesia Society in New Ireland
[1933]. New York: Norton.
Ribeiro, Gustavo, and Arturo Escobar, eds. (2006). World Anthropologies: Disciplinary
Transformations Within Systems of Power. Oxford: Berg.
Shiell, Timothy (1987). On Marx’s Holism. History of Philosophy Quarterly 4: 235–46.
Smuts, Jan Christiaan (1999). Holism and Evolution [1926]. Chico, CA: Sierra Sunrise.
Stocking, George W. (1983). The Ethnographer’s Magic: Fieldwork in British Anthropology from
Tylor to Malinowski. In Observers Observed, ed. G. W. Stocking. Madison: University of
Wisconsin Press, 70–120.
Stolcke, Verena (1995). Talking Culture: New Boundaries, New Rhetorics of Exclusion in
Europe. Current Anthropology 36(1): 1–24.
Strathern, Marilyn (1987). Out of Context: The Persuasive Fictions of Anthropology. Current
Anthropology 28(3): 251–81.
Thornton, Robert (1988). The Rhetoric of Ethnographic Holism. Cultural Anthropology 3(3):
285–303.
Tönnies, Ferdinand (1912). Gemeinschaft und Geschellschaft [1887]. Leipzig: Fues’s Verlag.
Weiner, James F. (1999). Afterword: The Project of Wholeness in Contemporary Anthropology.
Canberra Anthropology 22: 70–8.
9781444333237_4_001.indd 15
8/7/2010 6:41:40 PM
9781444333237_4_001.indd 16
8/7/2010 6:41:40 PM
Part 1
Rethinking Holism
in Ethnographic Practice
p01.indd 17
8/7/2010 7:21:20 PM
p01.indd 18
8/7/2010 7:21:20 PM
2
Beyond the Whole in
Ethnographic Practice?
Introduction to Part 1
Holism first truly became a hallmark of anthropological inquiry at the beginning of
the twentieth century. At this time various conceptions of the whole informed the
emerging theoretical traditions of modern anthropology: culture as a whole, symbolic
structure as a whole, and society as a whole (see Chapters 6, 10, and 14, respectively).
Interestingly, this embrace of holistic concepts – which also happened in a variety of
other academic disciplines (see Chapter 1) – coincided in anthropology with a major
methodological shift from broad ethnographic comparison to focused and intensive
ethnographic fieldwork. Within two of the major traditions of anthropology – the
American and British traditions – holism thus became closely associated with ethnographic fieldwork as a way of understanding a whole way of life. This link was, as
Stocking has observed, to a large extent “mythopoetic” (1983). It established a heroic
image of fieldwork, pretending it was new when in fact it was not, and creating a utopian ideal about the independent and neutral fieldworker who was able to capture
“the whole” through a form of participant observation that leaned heavily toward
natural science ideas of truth, comprehensiveness, and objectivity.
This “mythopoetic” link was strongest amongst British and American anthropologists, epitomized by Bronislaw Malinowski and Franz Boas, respectively. French
anthropology was different in this regard, since it was characterized by a much more
ambivalent relationship to fieldwork. Claude Lévi-Strauss, the founder of French structuralism, famously declared that he “hated travelling” (Lévi-Strauss 1992), and while
other French anthropologists – like Marcel Griaule and Michel Leiris – spearheaded a
tradition for fieldwork (Stocking 1983; Clifford 1988), this aspect of the French tradition never entered, as Archetti has pointed out, the annals of “international anthropology” (2006). As a result, fieldwork and holism were never mythopoetically linked in
French anthropology, neither in its self-image nor in its outside disciplinary reception,
to the extent that they increasingly were within British social anthropology and
American cultural anthropology. The French contribution to anthropological holism
(deservedly or not) therefore came to be associated with a particular theoretical heritage rather than with ethnographic fieldwork (see Chapter 10).
The implicit link between holism and fieldwork within two of the dominant traditions of anthropology was probably one of the main reasons why holism became
firmly established as an anthropological doxa. Holism was not only clearly implied in
c02.indd 19
8/7/2010 7:15:46 PM
20
Otto and Bubandt
the discipline’s central theoretical perspectives and models, but also deeply grounded
in the new practice of doing and writing ethnography, typically leading to its classical
products: the monograph and the ethnographic film. It is with regard to this modality
of its existence – embedded in the practices and standards of ethnographic research
and representation – that holism came to the center of critical attention in the 1980s
as part of the postmodern critique of anthropological representation.
Holism and Culture Critique
A key moment in this critique was the appearance in 1986 of Writing Culture (Clifford
and Marcus 1986) and Anthropology as Cultural Critique (Marcus and Fischer 1986).
In the latter work, Marcus and Fischer discussed how the classical aim of the ethnographic monograph – “to provide a full picture of a closely observed way of life” (1986:
22) – had to be critically revised in a world that was now understood as being globally
connected and where new forms of cooperation were emerging. Their solution to this
problem was to foreground a function that anthropology in their view had always had –
at least as a potential – namely, to formulate a critique of the anthropologists’ own
cultural background on the basis of and in contrast to the analysis of the cultural ideas
and practices they studied. Techniques that would afford such a critical stance were
identified as “epistemological critique” and “cross-cultural juxtaposition,” which both
appeared to allow for the defamiliarization of established perspectives and implicit
assumptions.
In their new introduction to the 1999 edition of the book Anthropology as Cultural
Critique, Fischer and Marcus noted a number of new ways in which the critical function of anthropology was being reinvigorated after the 1980s. In addition to the new
experiments in writing that characterized the 1980s, the 1990s were increasingly preoccupied by new ways of doing fieldwork within multisited contexts, by novel forms
of cross-disciplinary cooperation that went beyond the humanities, and by a focus on
new kinds of topics that were concerned with the changes and challenges of a global
world. Anthropology was increasingly, so Marcus and Fischer suggested, reinventing
its critical potential by turning in new ways to topics and subjects that mystified all of
us, whether we were Azande or American.
No longer, then, is the project of anthropology the simple discovery of new worlds, and
the translation of the exotic into the familiar, or the defamiliarization of the exotic. It is
increasingly the discovery of worlds that are familiar or fully understood by no one, and
that all are in search of puzzling out. (Fischer and Marcus 1999: xvii)
Anthropologists, they suggested, were approaching these new mystifying topics –
topics such as violent conflict, science, finance, the media, and technology – within a
renewed framework of critique. In such a project of critique, it was less relevant to
produce a “whole” picture of another way of life than to expose the global links that
made local worlds possible and thereby to highlight the connections and contrasts
between cultural worlds. Marcus and Fischer’s new kind of critique therefore entailed
a reorientation of holism. The predominant mode of ethnographic holism in the 1980s
involved the situating of an ethnographically knowable and describable “microworld”
in the context of an encompassing “macroworld” or “system” that could not be
known and described in the same terms. In contrast to this, Marcus foresaw the
c02.indd 20
8/7/2010 7:15:46 PM
Beyond the Whole in Ethnographic Practice?
21
development of a mode of ethnography that emphasized links between local worlds
and “more pluralistically sensitive systems perspectives” (Marcus 1989: 8). In order to
grasp how a site – in which a technological application, financial transaction, violent
conflict, or way of life took place – was connected to other sites, the notion of the
whole would need to be refigured. The refunctioning of anthropology as cultural
critique thus entailed not only a critique of ethnographic forms of representation and
practice but also a critical reexamination of holism.
Holism and Ethnographic Representation
An early key text that began such an examination in light of the 1980s critique of representation is the article by Robert Thornton (1988). Thornton’s text was originally
planned as a chapter in Writing Culture, but it did not make the publisher’s deadline
and was therefore published later as a journal article.1 In his article Thornton makes
the central argument that the holism of ethnographic descriptions is in fact the result
of a rhetorical trope. It is a product of the text that as such has little to do with any
assumed holism “out there in the world”. Thornton argues that holism is an “essential
fiction” of the ethnographic text (1988: 289) – essential because it actually guarantees
the facticity of the ethnographic “facts” for the reader. Ethnographic data become
acceptable as fact only to the extent that they imply a reference to some ulterior entity,
which by necessity is imaginary because it is not directly accessible, neither to anthropologists nor to their informants and readers. Thornton refers here to a central characteristic of ethnography as a style of reasoning or inquiry (see also Hacking 1992),
namely, that for the description to be convincing as valid knowledge, it has to refer to
a larger context or “whole.” According to Thornton, this is achieved in ethnography
through the rhetoric of classification. Thornton contrasts this anthropological style of
holism to that of historical holism, in which the rhetorical tropes of chronological narratives create a similar closure through generic plot structures like in fictional literature
(see White 1973). If ethnography asserts holism through the rhetorical idea(l) of an
encompassing social and cultural structure, history asserts holism through the device
of temporal sequence, causality, and plot.2 In ethnography the structural order of classification generates a sense of comprehensiveness and closure, and thus scientific
authority, so Thornton argues, through the way it is textually laid out in the outline
order of chapters, headings, subheadings, and paragraphs: “Social structure, then, like
plot, is the image of coherence and order that writing creates” (1988: 286).
The problem that Thornton pinpoints is as banal as it is difficult to circumvent: a
textual whole is fundamentally different from a social whole. The whole of the text is
purely mereological, which means that the physical parts together create a whole, like
the slices of a cake. A social whole is of a different order, as it is not physically composed of its parts such as individuals, clans, age grades, nations, or factories. Its wholeness is a matter of imagination that draws on different types of metaphors as we have
seen, but that is never materially realized in the same way as the parts of a text.3
Therefore, the holism evoked by ethnographic texts is the result of a specific rhetorical
device – that of classification laid out in the structure of the text – and this leaves the
purported holism of society basically unexplored.
Recently Alan Rumsey (2004) has provided a more positive reading of this divide
between rhetoric and reality. According to Rumsey “[T]ropes are essential tools of
understanding, and can lead to insights which, however ‘partial,’ would not otherwise
c02.indd 21
8/7/2010 7:15:46 PM
22
Otto and Bubandt
have emerged” (2004: 286). Knowledge is always partial, because it is mediated by
particular ways of knowing, and tropes offer particularly powerful ways of creating
knowledge. While Rumsey agrees with Thornton’s identification of classification as a
central ethnographic trope, he argues that ethnographers have more arrows on their
bow. Important for the way in which a text establishes a sense of whole is not only its
physical division in classificatory elements such as chapters, headings, and subheadings, but also the sustained cross-referencing between its parts. This happens when in
one part of the text the author refers backward or forward to another part. A conventional way of making such connections would be to write something like “In this
chapter on economic organization, we build on our description of the ecology in the
previous chapter and look forward to the treatment of political structure in the next.”
Although perhaps an outdated rhetorical device, it nevertheless links representational
parts into wholes that are not linear. Experimental ethnographies that seek to provide
more fragmentary and particularistic representations are bound by literary conventions about “sustained arguments” to intimate similar kinds of rhetorical linkage.
Rumsey thus argues that different part–whole connections are possible aside from the
one that treats the social elements described in the text as part of a social – organic –
unity.4 Ethnographic texts can alternatively draw on mediation between spheres of
human activity (for example, Evans-Pritchard’s The Nuer [1940]), on whole-to-whole
relations (for example, between symbolic microcosm and social macrocosm, as in
Geertz’s The Religion of Java [1960]), or on the sustained use of a central metaphor
(for example, the trope of “marginality” in Anna Tsing’s In the Realm of the Diamond
Queen [1993]). All these rhetorical tropes create a sense of connection and integration, even in those texts that are intentionally written against classical forms of holism,
as is the case with Tsing’s monograph.
The Chapters
Anthropological reflection on the holism implied in ethnographic practice has thus
taken two different but interrelated directions since the 1980s. On the one hand,
there has been the critical deconstruction of holism as a product of representation
(Thornton, Marcus, Rumsey). On the other hand, holism has been shifted from being
mainly a concern with the description of single life worlds to being the critical contextualization of single worlds in relation to larger systems (Marcus and Fisher, Clifford).
The chapters in this section take issue with these matters as they are currently undergoing experimentation and revision in ethnographic practice. George Marcus reassesses the developments since the 1980s – to which his own work has made a critical
contribution – and looks ahead at the changing conditions for doing ethnography in
a world increasingly characterized by the new media of communication afforded by
the Internet. The responses within anthropology to changes in the world – and in
particular to the changing relations between anthropologists and the people formerly
known as “informants” – have made it clear, Marcus argues, that holism is not merely
a matter of theory or representation. Rather, he suggests, holism is an aesthetic that
traverses the entire production and reception of anthropological knowledge.
Anna Tsing develops the aspect of ethnographic critique, identified as central by
Marcus, through a sustained ethnographic exploration that serves as a critical engagement with actor–network theory (ANT). By zooming in on what happens in the
c02.indd 22
8/7/2010 7:15:46 PM
23
ethnographic encounter, Tsing highlights the need for a soft – and critical – form of
holism which she calls “worlding.” Like Tsing, whose account is informed by an
analysis of the socioenvironmental world of matsutake mushrooms, Martin Holbraad
also uses a sustained ethnographic example – namely, that of gambling in Cuba – as
an experimental platform from which to engage with the holism that is involved in the
practice of ethnography. He argues that ethnographers have to be able to go beyond
their own holisms by allowing themselves to be open to the coincidental, ephemeral,
and surprising nature of other people’s holisms that anthropologists are invited into
during fieldwork. This openness to the unexpected is tuned, so he suggests, toward
an imaginary “whole beyond holism.” In each their way, the three contributions are
all concerned with re-evaluating the relevance of holism for critical ethnography.
George Marcus observes that holism in anthropology primarily has been a “regulative ideal and an aesthetic of practice and discourse, expressive of professional culture,” rather than an explicit theoretical problem or topic of methodological reflection
(Chapter 3: 31–2; see also Chapter 1). It is part of a particular style of thinking that
characterizes what anthropologists do. With this, Marcus is in line with Thornton’s
critique of a particular ethnographic style of writing, but he develops the issue beyond
the problem – and critique – of representation. Holism, according to Marcus, is an
aesthetic that “makes anthropology more than ethnographic expertise.” This suggestion has wide implications at a time when the ethnographic method has become a
label from journalism to marketing, when the outsourcing of ethnographical fieldwork has made anthropology as popular as ever, and when anthropology is venturing
into a host of new fields where conventional forms of holism are inadequate. Under
these circumstances, the need to rethink holism is pushed by broader concerns than
merely worries about how we represent our knowledge in writing. Rather, it is about
the very media of knowledge production in new sites and situations where there are
no obvious and conventional frames to steer our investigations.
To investigate how holism is currently being rethought under these conditions,
Marcus lays out a trajectory of changes within anthropology since the 1980s which he
summarizes under two headings: the rescaling and the refunctioning of ethnography.
The first – the rescaling of the discipline – refers to the rethinking of anthropology in
a context where “the global situation” has become the dominant metanarrative of
contemporary challenges and experiences, issues which Marcus has previously
addressed in a number of publications – including Marcus (1986), the anthology
Ethnography Through Thick and Thin (Marcus 1998b), and the Late Editions series
(e.g., Marcus 1998a). The anthropological answer has been to conceive of fieldwork
as multisited and engaging differently with informants and collaborators. This requires
that the classic holistic frames become more loosely defined and are made continuous
with the globally distributed sites of research. Research is done not on the basis of
established holisms, but in a temporal mode of emergence that deeply affects the
holistic frames being developed by anthropologists, as these engage with other knowledge-making frames in the making.
Extending from this – and from the thesis of his earlier book coauthored with
Fischer – Marcus argues that ethnography is also increasingly being refunctioned as an
enterprise of critique. This refunctioning of “anthropology as critique” proceeds by
contextualizing the holisms of others rather than being guided by holistic theoretical
frames. Therefore, Marcus argues, holism in ethnography is now “a question of technique, of accumulating the range and kinds of material that will support critical arguments that are more than exceptions, or complicating particularities for the claims and
c02.indd 23
8/7/2010 7:15:46 PM
24
Otto and Bubandt
actions of others who invest agency in the realization of particular ideas, narratives,
and generalities” (Chapter 3: 37). While some of the successful examples of
critique he discusses (such as that of Marilyn Strathern) still largely stay within the
ethos of classical ethnographic practice, Marcus argues that the “unboundedness”
of the Internet necessitates an even deeper reflection on and reconditioning of the
basic aesthetic of ethnographic holism as a way of designing the research process
and managing the complex and diverse bodies of materials generated by ethnographic practice.
While Marcus leaves us with an awesome vision of the unchartered terrains that are
currently being explored as anthropologists enter new virtual, political, and technical
worlds with ethnographic holism as the only compass to steer by, Anna Tsing takes up
ethnography’s critical function of contextualizing other people’s holism. The partner
she chooses, Bruno Latour (2005) – as the main representative of the ANT approach –
presents himself as the champion of an alternative way of understanding social life that
explicitly and strongly denies the value of a holistic frame. By following the ANT
strategy, to which she is largely sympathetic, as part of an effort to come to terms with
how different groups of scientists deal differently with the same kind of mushroom,
Tsing shows that a certain form of contextualization is unavoidable to make sense of
different scientific explanations. A “flat” analysis of the kind advocated by the ANT
approach makes “sense” only to the extent that it is able to conjure up a world, a
context, of the kind it is devised to obviate. The ANT denial of context, while theoretically challenging, is thus also disconcertingly a forgetting of the “world” that
makes its own standpoint possible. To make sense, any analysis has to imagine a
“world,” however tenuous, in order to make connections appear. Tsing uses the term
“worlding,” derived from literary criticism, to refer to this kind of contextualization.
Worlding is here an alternative to holism, which Tsing, following Latour, rejects
because she feels it connotes units of analysis that are too rigid and internal divisions
that are too definitive. Tsing defines worlding as “the always experimental, partial,
and often quite wrong, attribution of worldlike characteristics to scenes of social
encounter” (Chapter 4: 54). Worlding happens in the relation between people,
between scientists, and in the ethnographic encounter. It can be seen as ethnography’s central analytical and interpretative mechanism. By following an ANT strategy,
which is founded on a denial of context, a denial that is part of both its strength and
its weakness, Tsing shows that it is precisely the things that are left out or are moved
to the margins in such an approach that give shape to the figures in the worlds in
which human beings find themselves.
For Tsing, the problem of holism is therefore not its call for contextual framing, but
the tendency toward a stabilization of such analytical frames. Latour and the ANT
approach accurately saw such preestablished frames as obstacles to research and proposed a counterstrategy without any worlding at all. The total denial of context however threw out the baby with the reflexive bathwater, as it were, by making it possible
for the ANT approach to remain silent about the worlding that guides its analytical
frames and selections. An alternative and opposite reaction to the problems of wholes
and context came from within anthropology. Instead of denying contexts, this
approach sought to multiply contexts or perspectives and to use contrasting perspectives to destabilize holism. This approach has been spearheaded by Marilyn Strathern
in particular. According to Tsing, Strathern sees it as anthropology’s central task to
understand other people’s perspectives and holisms, thus producing – as it were – an
excess of worldings. Both strategies are perhaps extreme in their execution, but their
c02.indd 24
8/7/2010 7:15:46 PM
25
prototypical character highlights that the problem of holism comes back to haunt
even those who deny it. Tsing shows that worlding is a part of any form of scientific
encounter, whether ANT or Strathernian. She thus argues for worlding – a “weak”
kind of holism that constantly questions and undermines its own tendency toward
stabilization – as a simultaneously orienting and disorienting metaphor for anthropological analysis.
Martin Holbraad argues that there have been two distinct kinds of holism in anthropology. On the one hand is the kind we normally call holistic, which uses a particular
model of the whole – culture, society, structure, and system – to describe and explain
concrete social practices. The key concepts of this type of holism have come under
attack, as we have argued in Chapter 1, because they totalize even when this is not
warranted by the material at hand. On the other hand, Holbraad suggests, there is a
much more implicit form of holism that forms the basis of ethnographic fieldwork, a
holism that has been the ideal of fieldwork since Malinowski (see also Chapter 14). In
the field, ethnographers are ideally open to the unexpected. In this way, they allow
themselves to be surprised by “coincidences” that bypass existing holisms or explanatory models and thus force them to invent new concepts to make sense of the different worlds that are thus serendipitously revealed. As Holbraad metaphorically puts it,
ethnographers have to read the “cosmos” (or world) directly – that is, not through an
existing “cosmology” (theory) – in order to produce elements of new cosmologies
(analytical concepts).
Holbraad develops this contrast between two kinds of holism on the basis of an
ethnographic study of gambling and divination in Havana. At first glance, Holbraad
suggests, gambling and divination both employ “cosmological” schemes to make
sense of events. However, his ethnographic attempt to make sense of gambling practices forces Holbraad to see gambling as a very different kind of practice, in which the
punters try to gain direct access to the cosmos as a whole in order to find clues for
their bets. Holbraad uses the contrast between the gambler and the diviner to illustrate how two dissimilar kinds of holism can be seen to be at work in anthropology.
Conventional forms of anthropological holism are like divination in that both are
oriented toward “ordering” the world according to a pregiven “cosmological” set of
rules, while experimental forms of ethnographic holism, according to Holbraad,
ought to proceed like the Cuban gambler by being attentive to the emergent and
unexpected ways in which the world reveals itself to us during fieldwork. Holbraad’s
ethnographic study thus provides concrete metaphors for theoretical reflection, and at
the same time it is an example of how the holism of ethnographic fieldwork works to
create new concepts that bypass and criticize existing holisms. Holbraad’s (no doubt
provocative) conclusion is that existing forms of holisms fail not because they are
holistic but because they are not holistic enough. Holbraad’s chapter is an elegant
ethnographic experiment that challenges our understanding of ethnographic practice.
But at the same time, it remains strongly grounded within both a realist and a hermeneutical tradition, with its central premises of openness to reality, interpretation, and
sense-making.
In each their own way, all three chapters in this section address and experiment with
holism in the practice of ethnography. Marcus highlights the implicit aesthetic of
holism in ethnographic research, which functions to find an appropriate way of reframing our research questions to the contemporary world and to give direction to the
“composition” of our empirical material under new global conditions. Thus, experiments in ethnographic holism are about creating appropriate contexts, or “worldings”
c02.indd 25
8/7/2010 7:15:46 PM
26
Otto and Bubandt
as Anna Tsing would say, that put things into new and surprising relief, make them
visible as figures against a background, and allow us to be critical about worlds that, as
Marcus and Fischer put it, are “familiar or fully understood by no one, and that all are
in search of puzzling out” (Fischer and Marcus 1999: xvii). Ethnographic practice is,
in all three experiments in holism, about making sense: of ethnographic worlds, of our
analytical worldings, and of the changing conditions for both.
Notes
1 The other chapters in Writing Culture that explicitly address holism, albeit in a less concerted manner, are those by George Marcus and Stephen Tyler (Tyler 1986).
2 For a discussion of historical holism, see Hirsch and Moretti, Chapter 16, this volume.
3 For a discussion of a shift in the conception of social wholes from empirical fact to imaginary
conditions for action, see Chapter 14.
4 The prototypical example is the understanding of the social whole as a segmentary organization (tribes contain clans, subclans, lineages, etc.) that was characteristic of much of British
social anthropology in its heyday (see Chapter 14).
References
Archetti, Eduardo (2006). How Many Centers and Peripheries in Anthropology? In World
Anthropologies. Disciplinary Transformations Within Systems of Power, ed. G. L. Ribeiro
and A. Escobar. New York: Berg, 113–32.
Clifford, James (1988). The Predicament of Culture: Twentieth-Century Ethnography, Literature,
Clifford, James, and George E. Marcus, eds. (1986). Writing Culture: The Poetics and Politics
of Ethnography. Berkeley: University of California Press.
Evans-Pritchard, Edward (1940). The Nuer. Oxford: Clarendon.
Fischer, Michael M. J., and George E. Marcus (1999). Introduction to the Second Edition. In
Anthropology as Cultural Critique, 2nd ed., ed. G. E. Marcus and M. M. J. Fischer.
Chicago: University of Chicago Press, xv–xxxiv.
Geertz, Clifford (1960). The Religion of Java. Chicago: University of Chicago Press.
Hacking, Ian (1992). “Style” for Historians and Philosophers. Studies in History and Philosophy
of Science 23(1): 1–20.
Latour, Bruno (2005). Reassembling the Social. Oxford: Oxford University Press.
Lévi-Strauss, Claude (1992). Tristes Tropiques. New York: Penguin.
Marcus, George E. (1986). Contemporary Problems of Ethnography in the Modern World
System. In Writing Culture: The Poetics and Politics of Ethnography, ed. J. Clifford and
G. E. Marcus. Berkeley: University of California Press, 165–93.
Marcus, George E. (1989). Imagining the Whole: Ethnography’s Contemporary Efforts to
Situate Itself. Critique of Anthropology 9(3): 7–30.
Marcus, George, ed. (1998a). Corporate Futures: The Diffusion of the Culturally Sensitive
Corporate Form, Late Editions no. 5. Chicago: Chicago University Press.
Marcus, George E. (1998b). Ethnography Through Thick and Thin. Princeton: Princeton
University Press.
Marcus, George E., and Michael M. J. Fischer (1986). Anthropology as Cultural Critique: An
Experimental Moment in the Human Sciences. Chicago: University of Chicago Press.
Rumsey, Alan (2004). Ethnographic Macro-Tropes and Anthropological Theory. Anthropological
Theory 4(3): 267–98.
c02.indd 26
8/7/2010 7:15:46 PM
27
Stocking, George W., ed. (1983). Observers Observed: Essays on Ethnographic Fieldwork.
Madison: University of Wisconsin Press.
285–303.
Tsing, Anna Lowenhaupt (1993). In the Realm of the Diamond Queen. Princeton: Princeton
University Press.
Tyler, Stephen A. (1986). Post-Modern Ethnography: From Document of the Occult to Occult
Document. In Writing Culture: The Poetics and Politics of Ethnography, ed.
J. Clifford and G. E. Marcus. Berkeley: University of California Press, 122–40.
White, Hayden (1973). Metahistory. Baltimore: Johns Hopkins University Press.
c02.indd 27
8/7/2010 7:15:46 PM
3
Holism and the Expectations
of Critique in Post-1980s
Anthropology
Notes and Queries in Three Acts
and an Epilogue
George E. Marcus
Orientation
In writing this chapter, I do not want to stay within the confines of the discussions of
holism in which I participated during the 1980s and early 1990s, when holism in
anthropology first became (to my knowledge) a prominent object of theoretical consideration and critique. This would restrict me to discussing holism as a problem of
representation (part to whole) in the conventional media and means by which ethnographic knowledge has been produced. I feel that more is at stake in addressing again
the question of holism at this juncture. So, somewhat uncertainly, I try to cross terrains (in the Epilogue at the end of this chapter) with which I am only haltingly familiar in taking up this topic before returning to familiar conventions but in a different
place, and with a different emphasis, than where the original critical discussions of
holism first directed me. Holism really is no longer only about producing better or
more adequate descriptive-analytic accounts of ethnographic subjects in a changing
world. It is more problematically about the media of knowledge production in ethnography, from inception, without obvious or conventional bounds, rather than about
providing analytic frames for the messy experiences of contemporary fieldwork.
Act 1: How Holism Has Mattered in Modern
Anthropology
Holism in social-cultural anthropology has been less a theoretical problem or a
topic of methodological concern than a regulative ideal and an aesthetic of practice
and discourse expressive of professional culture. Most practically, it encourages
9781444333237_4_003.indd 28
8/7/2010 6:43:02 PM
Holism and the Expectations of Critique
29
anthropologists in a particular style of thinking: to be broad-minded, to contextualize
the particulars in which they specialize in wide scale and scope, and to discover unsuspected connections and make something of them. It also encourages rhetorically a
humanist’s sometimes prideful overreach, leading to an excess of interpretation.
This aesthetic holism contrasts with its explicit topical import in some traditions of
social theory, such as western Marxism (see Jay 1986)1 and its methodological significance in the core protocols of those social sciences committed to a formal positivism.
Now the holistic aestheticism, as a matter of the craft (and ideology) of research in
anthropology, has certainly shared in and reflected the history and ambition of positivist social science. Social-cultural anthropology sought to describe and generalize about
societies and cultures as whole systems (e.g., functionalism as the frame for “good”
ethnography, which still haunts as an aesthetic, and as shaping the deep knowledge
standard and performance of ethnographic expertise), but it would have to have been
invested much more heavily in formal and quantitative techniques – especially modeling – for its commitment to holism to be more than a matter of rhetoric and a guildlike ideal.2 Still, in its more subtle and rhetorical expressions as one of the signifying
dimensions of anthropological research, holism has had powerful shaping influences
on both the research process (its habits, its stages, and the discursive expression of the
results and claims that come from it) and what anthropology’s long-established publics
(especially other social scientists)3 have expected, and continue to expect, of it.
When I came of age in anthropology during the early 1970s, I cannot recall a single
work on holism as such. Yet, in very clear terms and evaluative practices, it pervaded
expectations in the making of individual ethnographic expertise. This was predominantly rooted in areal specialization in peoples and places, on the one hand, and on
the other hand it was expressed distinctively in anthropology’s evolved rhetorics of
generalization: about universals, about the human, and about the importance of ethnographic expertise to ambitious collective projects of comparison (that, with some
notable exceptions, have not materialized as integrative frameworks for the discipline).
So, without ever being an explicit doctrine of pedagogy, theory, or method, holism
was (and still is) a deeply existential and aesthetic matter in practicing social and cultural anthropology. Two questions for this chapter and for this volume are “How are
we to take it more substantively now? What reconditionings of this aesthetic have
occurred, or are occurring?”
While I believe that holistic concern is what makes anthropology more than ethnographic expertise, this broader program (comparison, universals, the human from the
ethnographic) has not prospered as a disciplinary activity (although it might be argued
to have done so as an interdisciplinary one – in the realms of cognitive studies and
evolutionary research, for example). In contrast, the ethnographic centering of the
discipline has prospered. So as it is cut from its moorings in an older disciplinary program that is not practiced much at present, I will limit my concern to how holism has
shifted, and still might further shift, in the way that it is expressed in the pursuit of
very individually centered ethnographic research – this rite of research through which
social-cultural anthropologists are made, then practice in diverse ways, and pass on
canonically to others (Faubion and Marcus 2009).
Like many anthropologists today, I have mixed feelings, and am of two minds,
about this powerful legacy of holistic expectation concerning a systemic (extending
from particular societies to all of humankind), though informal, grasp of particulars,
closely and expertly observed and interpreted. The social solidarity of the discipline
depends on this trope and a few others.4 It seems so sentimental. Yet, I also know that
9781444333237_4_003.indd 29
8/7/2010 6:43:02 PM
30
Marcus
more is at stake in present attitudes about this living legacy than a matter of sentiment,
and that a more viable expression of it has developed in the form that ethnographic
research has taken since the 1980s (see “Act 3,” below).
It is significant perhaps that anthropologists could not come to terms with this
aesthetic of craft until they could develop a means of discussing it. This occurred
ironically during the 1980s in incisive critiques and exposures of how it operated as a
structure of expectations, rhetorics, and discursive performances.5 Holism came to be
reflected on for the first time just as it was acutely exposed. Anthropology in response
might have revisited its formalist ambitions of the 1960s if classic formal positivism in
the social sciences were not also having its own reevaluations (and anthropology was
rather too late to the party, so to speak, by this time). More importantly, the world in
which anthropology materialized its subjects of study was itself being reconfigured, first
by changes on the ground – the media and politico-reflexive terms by which “peoples”
understood themselves – and, second, in academia by the fashion for postmodern ideas,
then by the perception of globalizing systems of change. The mainstream of socialcultural anthropology reconfigured its project (see “Part 3,” below) in which the enduring aesthetic of holism inheres in a more viable, but perhaps less ambitious, form.
Act 2: Holism as a Problem of the Limits
of Conventional Representation
The well-known critiques of the 1980s (Clifford and Marcus 1986) focused on ethnography as a genre of writing and a discursive form, but it had strong implication for the professional culture of method, of which the regulative talk and construction of holism in
ethnography have comprised one key dimension. As such, what the 1980s critiques
exposed was not that anthropology had tried to generalize and represent cultural and
social wholes, or to state rigorous general truths or principles from comparative ethnography, and failed – but in fact, that it had not even made a preponderant effort toward this
project (it would have been perhaps easier to make this case of failed ambition and effort
against projects of formal positivism). Rather, these critiques made explicit just how rhetorical and dependent on tropes of narrative the discourse of holism was in ethnographic
writing (and, by implication, the culture of craft aesthetics and regulative ideals that produced it), not that different from how holistic representation functioned in literary realism. Ethnography depended for its contextualizations on the professional craft labor, in its
diverse manifestations, of producing chronotopes for it (after Bakhtin 1981 [1925]).
Hayden White (1973) provided the key analysis of how narrative holism has operated in the major traditions of historical writing, and his work had much to show
anthropology about the writing of ethnography. Historians were able eventually to
relegate White’s work to the realm of historiography. The related critique in anthropology was not so easily managed. It keyed into and articulated ambivalences that
were well developed among anthropologists by the 1980s. Also, it did clarify, usefully
and in very explicit terms, how they produced explanations through contextualizations. But it also made clear how soft positivism operated in anthropology through
narratives, normative representations, and an aesthetic of practice inculcated through
the attaining of ethnographic expertise.
Both the value and the excesses of the critique were thoroughly debated, but the
demonstration that the broader frames and objects of anthropology – its much valued
9781444333237_4_003.indd 30
8/7/2010 6:43:02 PM
31
holistic spirit or aesthetic – were realized through tropes and techniques similar to
those of fiction was deeply troubling to many anthropologists. Eventually, the controversy subsided, and anthropologists went “back to business,” so to speak, but in so
doing, the sense of the limits and character of holism as a process of producing representations remained an indelible aspect of doing ethnographic research.
It would have been intolerable to just keep talking about societies, and cultures,
and humankind in specific research projects, if the character of basic ethnographic
research in its mainstream had not also changed after the 1980s. The critiques of the
1980s and the wave of intellectual ferment of which it was exemplary contributed to
this larger shift, but the story of this change is of course much more complex, and no
one has yet written a near-term history of it. For my purposes here, I want to point
to two post-1980s developments in the character of mainstream ethnographic research
that represented a shift in the way that holism was manifested in anthropology. Still
within the problematic of representation, these developments embraced the validity
of this critical argument and offered working resolutions of it. The aesthetic of holism
remains as a character of research practice and writing, but in very different senses.
Both developments relieve anthropology of the burden of constructing wholes as
encompassing contexts for its production of accounts of particularities. They still
involve the production of descriptive-analytic accounts but without embedding them
in claims or constructs of holism that the ethnography itself produces. Rather they
initiate more complex engagements with the holism in the representations of others.
They still operate within the limitations of representation, but retreat from the responsibility to account for wholes from parts. One of these developments has to do with
rescaling inquiry; the other has to do with refunctioning it. I consider the first – one
trend, among others, within the latter, more comprehensive shift in the mainstream
research program of social-cultural anthropology – in this section, and devote the
next section to the latter.
Multisited ethnography as a fix for holism amidst the global
situation and in the paleo-Internet era
During the late 1980s and early 1990s, the “global situation” (Appadurai 1996; Tsing
2000) became a dominant metanarrative, a milieu of discussion and holistic thinking,
in terms of which research projects in anthropology, among several other disciplines
and policy arenas, became defined and oriented.6 The “global,” as opposed to “world
systems,” “world history,” or “world cultures,” destabilized the past holistic framings
of anthropological research, and in line with the new hyperawareness of processes of
representation in ethnography, it also became an explicit object of conceptualization
in the definition of research projects, not only theoretically but also, more importantly, in terms of the research process itself.
I tried to capture this latter concern in my 1995 paper on the emergence of multisited ethnography (Marcus 1998a [1995]). Completely within the conventional
media and procedures of ethnographic research of the time, as defined by its professional craft culture, this paper envisioned shifts in how the traditional norms and regulative ideals that defined ethnographic research might be reconceived by the
circumstance of the rescaling of fieldwork (the language of “scale” has been explicitly
developed by Tsing 2005). A shift in the holistic imagery of research played a large
role in conceiving how multisited research might still also be ethnography. And as
9781444333237_4_003.indd 31
8/7/2010 6:43:02 PM
32
Marcus
such, the rescalings of ethnographic research served as a kind of “fix” or solution to
the problem of the classic holistic enframings of ethnography, exposed by the understanding of them as rhetorics, by making them contingent, open, less structural, and
continuous with the sites of research.
While the expansion of research in spatial terms has been the main dimension in
which the practical and conceptual challenges of multisitedness have been thought
through, it is actually the particular temporality within the construction of the global
as a milieu and frame of ethnographic research that crucially alters the discourse of the
“whole” and the obligation to represent it (see Rabinow et al. 2008). The study of
globalization is associated with the temporality of emergence or the contemporary as
the study of change in a dynamic present that exists at the intersection of the recent
past and the near future (see Rabinow 2008). Thus, the holism that enframes ethnographic research into particulars is only partly known, contingent, an open system, at
best. Ethnographic research produces arguments that are speculative in the most
responsible sense, are experimental in the ethos of science,7 and thrive as both complicating and clarifying supplements or challenges to other knowledges studied in fieldwork and/or as part of the interdisciplinary ecology in which they are produced (this
discursive space as object and reception of ethnography is potentially radicalized by the
potentials of Internet communication, and is where the problem of holism in anthropology shifts – a preview of the Epilogue). Not only is multisited research following
and describing the emergent course of processes in the contemporary world, with the
history of anthropology’s intellectual commitments composing it, but also, by engaging first and most directly with other knowledge-making frames, it challenges anthropology to rethink that history within its own disciplinary confines.8
The production of ethnography since the 1990s in the milieu of the global situation, and the theoretical and methodological discussions informing it, reflect and are
part of a major shift that has occurred in the function of research in social-cultural
anthropology since the 1980s, to which I now turn.
Act 3: Holism as a Different Problem
in the Refunctioning of Ethnographic Research
as an Enterprise of Critique
The research agendas of ethnographic inquiry since the 1980s–1990s are predominantly formed as projects of critique, with a primary commitment to demonstrating
nonreductive results in relation to others’ production of holistic discourses. Anthropological arguments are about multiplicities, contradictions, fluidities, resistances,
“both” rather than “either/or,” ironies, the overcoming of binaries, and so on. Thus,
despite the persistence of an older rhetoric in the promotion of anthropology and of
perceptions within collateral fields about what anthropology of the moment is able to
pronounce authoritatively upon,3 anthropology functions in its main research modality of ethnography as a science of critique9 in which the engagement with holism has
an entirely different character than before.
Anthropology’s own generativity arises from working within or in relation to other
discourses, ways of life, and projects of knowledge making, and commenting normatively on them.10 Rather than holism being a standard or expectation of the descriptive-analytic accounts produced from ethnographic research, in the project of
9781444333237_4_003.indd 32
8/7/2010 6:43:02 PM
33
critique, it emerges as an object of critique or qualification in the relations of research,
broadly conceived, with ethnography’s subjects, partners, and intended audiences and
readerships.
So, in critical projects, ethnography is always a nonreductive account of somebody
else’s holism. This could be the holism of the subject (“worldview” or “native point
of view,” as usual), that of a theoretical grand narrative or a reigning power knowledge (Marxism, neoliberalism) as target of critique, or that of a project within which
ethnography locates itself (environmentalism, development, finance, etc.). Its own
construction of holism is then dependent and referential, and does not provide the
“whole” context for the analysis of a part, but a frame that facilitates a relation of
critique that moves alongside or within power knowledges of broad distribution and
ambitious scope.11
Holism was formerly a burden of ethnographic representation; now it is constitutive of core relations that define critical argument. As such, holism is not about creating conceptual frames for descriptive-analytic accounts but about creating contexts
for translation, or what has replaced this venerable function – critical interpretation
and engagements. The exploration of this changed sense of holism in the critical
enterprise of ethnography is not limited by the character of representation itself, but
by the character of the conventional media of ethnography through which it has been
produced (fieldwork followed by writing up and print publication). It is the glimmer
of changing media in the production of ethnography by the emergence of Internet
forms of communication that suggests yet another shift in the terms of thinking about
holism in anthropological research.
In traditional mode, holism was an expectation of the adequacy of the descriptiveanalytic account, an expectation of expertise and the means by which it was satisfied
in performance and writing. The analyses of the dynamics and limits of ethnographic
representation during the 1980s showed this to be rhetorical since anthropology
rarely adopted the formalist analytics of other positivist or empiricist approaches from
which this holism derived. In the current mainstream enterprise of critique, there has
been a shift from the sense of holism as a problem of representation in writing or
articulating arguments to a sense of it as a problem of organizing inquiry on moving
ground and time according to scales and a temporality appropriate to its objects of
study. Holism is less a problem of the form of argument than a question of technique,
of accumulating the range and kinds of material that will support critical arguments
that are more than exceptions, or complicating particularities for the claims and
actions of others who invest agency in the realization of particular ideas, narratives,
and generalities.
Three Exhibits: Strathern, Rabinow, and Bunzl
What follows are samples or soundings from three sources that resonate with points
that I have attempted to make about the expression of holism in social-cultural anthropology’s shift toward an enterprise of critique. The first two exhibits I take to be
efforts in their very different idioms that exemplify the more general shifts in the rescaling and refunctioning of ethnography that I have described. I take issue with the
diagnosis of the aesthetic of holism today in the third exhibit, but treat it in more
detail because its references and framing allow me to anticipate the current shift into
9781444333237_4_003.indd 33
8/7/2010 6:43:02 PM
34
Marcus
a very different register and challenge of the problem of holism from the enterprise of
critique to the design of ethnographic research as located or assembled communication amid Internet modalities.
Exhibit A: Marilyn Strathern
Social anthropology has one trick up its sleeve: the deliberate attempt to generate more
data than the investigator is aware of at the time of collection … a participatory exercise
which yields materials for which analytical protocols are often devised after the fact.…
[Ethnography allows] one to recover the antecedents of future crises from material not
collected for that purpose … to anticipate a future need to know something that cannot
be defined in the present. (Strathern 2004: 5–7)
While skillfully drawing on concepts from her Melanesian ethnography, Marilyn
Strathern has been deeply involved in doing research and in creating its ethos as
anthropology in the realms of science, technology, policy, and social-cultural implication, centered in England but characteristically engaged in linking discourses of global
scope and emergent vision. She has situated herself and students within these complex
organizations and has tried to reproduce the conditions (and aesthetics) of classic
fieldwork there.
In a set of working papers (Strathern 2004) about her ongoing research on audit
culture and on new collaborative initiatives in the biosciences at Cambridge, she made
the above interesting statements about the distinctive purposes and contributions of
anthropological research. For my purpose here, what I think they exemplify is a statement of the critical project of ethnography trying to resolve the problem of holism as
it manifests itself not as a problem of a governing representation, but rather as a problem of orchestrating the conditions, space, and time of fieldwork itself.12 Analytics are
figured out later; ethnography’s special knowledge is a knowledge of the anticipatory –
the emergent issues not fully articulated in the literal, empirical present and scenes
of fieldwork.
So, yes, for Strathern holism is a problem of research performance in complex
settings which require a rethinking of the special value of what anthropologists do.
But, while very suggestive, Strathern’s remarks favor preserving the aesthetics and
practices of fieldwork in their ethos and craft culture, rather than a focused and
explicit discussion of how these practices might themselves be rethought. This is a
problem that becomes more explicit in the Epilogue and is clear only when an alternative media imaginary for research is available.
Exhibit B: Paul Rabinow
I take the object of anthropological science to be the dynamic and mutually constitutive,
if partial and dynamic, connections between figures for anthropos and the diverse …
branches of knowledge available during a period of time that claim authority about the
truth of the matter; and whose legitimacy to make such claims is accepted as plausible by
other such claimants; as well as the power relations within which and through which
those claims are produced, established, contested, defeated, affirmed, and disseminated.
Thus, taken as an object domain, contemporary anthropology is neither as broad as
traditional ethnography nor as grand as classic philosophic anthropology nor as ambitious as the human sciences. This narrowing of scope has some advantages: there are
9781444333237_4_003.indd 34
8/7/2010 6:43:02 PM
35
good reasons to accept the claim that a restricted form of anthropology refers to an actual
object domain in the present whose recent past, near future, and emergent forms can be
observed. (Rabinow 2008: 4–5)
Rabinow has gone farther than anybody else I know in proposing a restatement of the
ethos and terms of anthropological research to meet its current conditions. I don’t
agree entirely with what he has done, and it has gotten only limited attention among
anthropologists, but I admire the nature of the effort and its sophistication. He also
works within the shift to an enterprise of critique that I described. Influenced by
Foucault, Rabinow has reconceived the aesthetic of holism in anthropology as problematizations, distinctive realms of knowledge making and truth claims broadly distributed in global space and time.
I chose the above quotes because they refer to “a restricted form of anthropology,”
which is similar to the point that I made about how a refunctioned anthropology
operates as a science of critique in which holisms are the burdens of others, but their
nonreductive clarification, metacommentary, or restatement is a task for an enterprise
of “narrowed scope,” as Rabinow provocatively proposes. He thus provides an expression of how the problem of holism shifts from the contextualizing representation of
anthropological explanation to the scene of interaction and conversation between
anthropologists and experts, where the anthropologist works as a critic inside or
alongside a problematization that has empirical form as apparatuses, assemblages, and
events. Set in a temporality of the emergent, anthropological research describes and
actively critiques a contingent holism which it joins.
Rabinow has confidence in his own concept work and while he changes the basic
terms of anthropological research – he replaces talk of ethnography, fieldwork, informants, and so on – he remains true to the traditional scale of research. So holism is a
feature emergent in knowledge forms and truths, and it is their critique that anthropology provides. Yet, this critique reinvokes the holistic tropes on which classic
anthropology depended – not as a higher truth but as an undeveloped implication of
the ambitious and technical holism of, in Rabinow’s case, science projects.
Exhibit C: Matti Bunzl
Contemporary anthropology rejects culture as an essentialized abstraction that betrays
the invariable traces of power/knowledge. But it does so to advance a fundamentally
empirical claim, namely that what has been misinterpreted as homogenized culture is
really a set of complex negotiations and contestations. Sociocultural anthropology may
have rejected a scientistic variant of positivism, but it retains, even augments, a more
immediate form, one that purports that all empirical phenomena are amenable to observation and description. What else, after all, is the demand to eschew false generalizations
in the interest of more accurate representations of complexity? (Bunzl 2008: 55)
Bunzl goes on to quote from a story by Jorge Luis Borges – a parable about the ambition of exact representation:
On Exactitude in Science
“In that Empire, the Art of Cartography attained such Perfection that the map of a single
Province occupied the entirety of a city, and the map of the Empire, the entirety of a
Province. In time, those Unconscionable Maps no longer satisfied, and the Cartographers’
9781444333237_4_003.indd 35
8/7/2010 6:43:02 PM
36
Marcus
Guild struck a Map of the Empire whose size was that of the Empire, and which coincided
point for point with it. The following Generations, who were not so fond of the Study of
Cartography as their forebears had been, saw that the vast Map was Useless.” (Quoted in
2008: 57)
To me, contemporary anthropology is much like Borgesian cartography. Driven by an
unselfconscious demand for “exactitude in science,” it is on a quest to find the perfect
representation of human reality, one that is free of all essentialism and generalizations.
What it does not realize is that such a representation – if it could be had at all (and, of
course, it could not) – would be entirely unwieldy. Even worse, it would be altogether
useless, not because it would be false but because it would be true. (2008: 57)
These quotes are from an article in which Bunzl takes to task those anthropologists
who would criticize the oversimplicity and reductiveness of contemporary major political, policy, and media discourses in the glare of the US public sphere, especially about
peoples and places associated with anthropological expertise. These discourses (like
Samuel Huntington’s much discussed and much derided “clash of civilizations” vision)
are powerful holisms meant to affect and mobilize opinion. Bunzl laments how ineffectual the anthropological critique, presented as the unresolvable complexity of particulars, is in this arena. While acknowledging clearly that the enterprise of critique is
the mainstream preoccupation and mode of anthropological research since the 1980s,
he criticizes the result: a field of excessive empiricism, of a concern with particularities
that are imagined to simply add up to a literal whole, the representation of which can
be achieved, an absurd mapping exercise of the same literal scale of the phenomena
described in toto and in all of their particularities. Bunzl ends his article by a declaration
of faith in the older liberal anthropological posture of critique in the public sphere.
I disagree with Bunzl’s diagnosis about the empiricist ambitions behind anthropology’s
continued pursuit of particularities, since it overlooks how seriously the current enterprise
of ethnographic critique took the lessons about the limits of representation, especially
totalizing description, that were instilled by the 1980s Writing Culture debates. The lack
of standing of anthropological critique in the public sphere and media is more a question
of relative status and power than of strategy. Certainly, it does not arise by a misunderstood extreme empiricist ambition, still at the heart of anthropology. What Bunzl misses
is the dynamic of the relation and function of critique which precisely relieves anthropological analysis of its positivist burden of exacting holistic representation. Whatever truths
anthropology currently delivers, it does so through critique in which it is the nonreductive David to the holistic Goliath of globalizing and naturalizing discourses in all sorts of
venues. Anthropological explanation only works now in this relation.
Still, his allusion to Borges’ absurd cartography is useful in making a transition to an
argument about where the holistic aesthetic in the culture of ethnographic research is
moving. Ethnography (the process from fieldwork to writing in a unified, coherent sense –
this is the ultimate holistic problem of anthropology today, as I suggest in the Epilogue)
is trying to locate itself in the world by different coordinates than the Malinowskian miseen-scène and course. This is what the “global situation” and the multisited fix, so to
speak, began to pose as a problem of metamethod (a different way to think about the
craft of inquiry) in the paleo-Internet moment of the 1990s. But as for example the
recent work of Strathern and Rabinow, among many others, demonstrates, this is far
from a problem in ethnography’s conventional terms of empiricist, holistic desire as Bunzl
argues. While still operating with the end of ethnography as adequate representation
(with the book in mind), they each begin to see problems with the organization,
conception, and management of what has long gone under the term “fieldwork.”
9781444333237_4_003.indd 36
8/7/2010 6:43:02 PM
37
The emergence of Internet forms of communication poses the issue of publics of
indeterminate bounds in the composition of materials that are gathered in ethnography and form the basis for its accounts. How to derive ethnographic accounts from
the wholes of interaction and communication that the Internet elicits in various forms
(blogging, websites, searches, wikis, and now social networking) is the emerging
problem of holism in ethnography. It has similarities to the Borgesian cartography.
But it does not involve an ambition of exacting description at an absurd scale. Nor is
it useless. It is about producing the materials of knowledge making in a regulated
commons that has indeterminate social dimensions. While we are waiting for Internet
forms to become common practice, how does their example help us to reconfigure
the ethnographic research process that we know and practice now?
Web communication mobilizes wholes of unknown dimensions. The holism problem
of anthropology moves from the precincts of representation to the management of communication. The problem of organizing ethnographic inquiry is literal, material, and technical often. But it is not absurd and it is not at base about representation, but about the
management of publics in the forms of research that anthropologists can now use. The
uses of the Internet in knowledge-making projects are by now beyond hype and wonder,
but the skills and operations required to use the still evolving forms of this medium in
ethnographic inquiry are not yet certain. Still, such uses are available now as a resource for
rethinking the wholes as publics and multitudes that the research process that we know
mobilizes and must work within. Internet forms have Borgesian qualities of scale, but the
problem is not cartography or representation but, as Chris Kelty has argued (2009; discussed in the Epilogue), composition, coordination, and sometimes collaboration.
Epilogue
Taken as a formal conclusion to this chapter, “Act 3” ends with a mere gesture toward
making an argument that would require another essay to develop. But rather than just
leave it at that, I want to make use of the epilogue device and the latitude that it
allows,13 not to speculate about or foresee a future regarding the media of ethnographic research, but to provide from pieces of it that are already here resources for
thinking about changes in the post-1980s version of classic, familiar anthropological
research that I have attempted to chronicle in the three sections of this chapter.
Anyhow, I am probably not the most appropriate person to produce such an
essay that both explores the state of the art and speculates about new technologies
and forms of ethnography,14 since I believe that I am typical of my generation (of
anthropologists at least) in relation to the pace and level of attention at which I am
assimilating rapidly changing information technologies for personal use (clearly
these new media in their present uses are beyond merely the promissory and visionary hype about the Internet typical through the 1990s). I continue to write and
teach, finally, in the older media (narrative, notebooks, and print ethnographic
forms), but I can’t help but register alongside these practices as usual, even if
debated and rethought after the 1980s, how the new media reconfigure the problem of holism as one of composing (or engineering) ethnographic materials among
publics of Borgesian scale from fieldwork to authoritative publication. Holism is
now situated in the practical, engineered spatiotemporal dimensions of anthropology’s research process. The problem of multisited fieldwork becomes a problem of
9781444333237_4_003.indd 37
8/7/2010 6:43:02 PM
38
Marcus
the design of networked knowledge spaces stabilized, somehow, for the purposes
of sustained inquiry in the classic ethnographic tradition.
In my most recent research, I have found it most practical to think through this
present challenge of holism to the design and scale of ethnography in terms of the
pedagogy of “first” fieldwork through the choices and strategies of current careerdefining dissertation research.15 So, I want to end this epilogue, in its “notes and
queries” mode, with two exhibits quite within the conventions of longstanding teaching that show in a refracted way how the aesthetic of holism that I have tracked, under
pressure of new capacities of communication, is transforming in the still conventional
precincts of ethnographic research practice.
The first exhibit, a commentary on a dissertation oral defense “ritual,” offers an
example of fractures in the ways that even the kind of revised holism described in “Act
3,” above, in response to a refunctioned-global anthropology of the 1990s, is shifting
in a world of networked knowledges. The problem is defining research amid potentially
unbounded networks and not getting lost in them – giving scale to processes at the
scale at which ethnographic inquiry is effective. And the second exhibit deals with innovation in the teaching of basic ethnography by a younger, prominent scholar of the
“culture” of the Internet; it reflects techniques and concepts appropriate to improvising
ethnographic subjects and media (both at the same time) as indeed problems of holism
that new technologies pose. This exhibit reveals holism to be a problem of managing
publics and overlapping communities of research in giving a form to appropriate ethnographic inquiry.
Exhibit A: Discussions at the oral defense of a dissertation
A memorable discussion of a dissertation at its oral defense several years ago in the
Rice University Anthropology Department focuses issues about alternative frames or
senses of holism at play in contemporary ethnographic research that were elicited by
comments of the student’s three advisers attempting to give her work a context of
anthropological significance.
In discussion was the dissertation of a very able Korean American student who
did fieldwork on the culture of risk at a venture capital firm in Seoul in the wake of
the 1997 Asian currency crisis. This student could have studied this topic in terms
of a more generic contemporary culture of finance, and had she done so, there
would have been less reason to have focused her time and effort in Korea itself. If
there were ever a subject for a multisited design and logic of research, this was it.
But there were strong professional and career pressures to make fieldwork local –
to look for something not just situational and historical of the moment, but also
distinctively cultural about the operations of risk and finance in Korea.
In any case, she settled into a year of research in a Seoul venture capital firm. All did
not go smoothly in this site of participant observation, but she returned with rich material, and she developed interesting things to say about markets, finance, the present
conditions of modernity as “globalization,” and the ways capital operates through the
practices of financial experts.
But there was this nagging expectation in producing the dissertation to “make it
Korean.” So she attempted a core argument about the deeply Korean character in the
use of particular instruments, protocols, and operations of finance by associating them
with a Korean concept, value, and perception pertaining to the speed and pace of life.
9781444333237_4_003.indd 38
8/7/2010 6:43:02 PM
39
This was a potentially interesting connection but not convincingly demonstrated
through ethnographic materials (how it might have been is another issue).
This problem in the dissertation elicited three discussions from her advisers which
were about possible “solutions” to the form that an adequate and identifiably anthropological analytic-descriptive version of her argument might take. In my view, each is
about how holism can be constructed around the particulars of anthropological
inquiry. Two were conventional and familiar, and the third was more radical, suggesting that in projects like that of this student (and there are many like it now, in my
experience), holism is really a matter of the design and controlled porosity of ethnographic research projects in their conduct today – and not so much a matter simply
of the choice of analytic frame and scale. Thus, holism is more a feature of the design
of the research process than of theory, concept, or the discursive product ofresearch.
One solution called for more contextualization in the history of financial institutions in Korea along with the cultural history of its nationalism, a source for the
broader expression of values on speed and pace of change, for example. A second
solution called for an enframing, and systematic analysis of the semiotic field of the
key Korean concepts for tempo, speed, and celerity which identified the conduct of
certain venture capital operations as distinctly Korean. The holism called for here is of
the most traditional sort through defining the wholeness of culture in terms of semiotics and its analytic techniques.
Both of these solutions were reasonable, and neither touched upon the standard process by which training in ethnographic research is undertaken. It was the third solution
that came to terms with the unusual “object” that the student had selected for study and
for which the assumptions and application of cultural, historical, or social explanations
as holistic enframings of particulars were not adequate. And this third also shifted attention to issues of the design, process, and function of ethnographic research projects,
especially in projects of “first fieldwork.”
The third adviser paid particular attention to the written questions that had been
submitted by the outside member of the student’s committee, who was not able to
attend the defense. This fourth member was from the Business School faculty, was
better informed about contemporary ambitions and projects in the qualitative social
sciences than most Business School scholars, and had a more substantive relationship
with the student than is usual for outside members of dissertation committees. While
the fourth member knew nothing about Korea, she had a keen and informed curiosity
about the situated practices of venture capitalists. She had better questions for the
student than any of the other advisers. This might have not been apparent had she
been present. By having to ask her questions in absentia, the third adviser saw that
anthropology had no questions of specificity for the student’s subject matter, while
another expertise did. The issue of making the project distinctively anthropological
might best have been achieved by operating along a cline of translations which first
made the Business School expertise a site of ethnographic inquiry in order to derive
the specific questions that could eventually be pursued in the Korean venture capital
firm. As such, the fourth member of the committee might actually have been the
student’s first major informant, and while the student would still have spent time in
Seoul, the project as field research might better have been imagined to be located in
mediational space, working back and forth along the cline of translations and frames
of questioning between one kind of situated expertise through which ethnographic
questions are formulated and another kind of situated expertise in which such
questions were engaged through the presence of the ethnographer. Along the way,
9781444333237_4_003.indd 39
8/7/2010 6:43:02 PM
40
Marcus
the Koreanness of venture capital would have emerged in comparative and relative
terms within a global culture of theory and practices.
In the third commentary on this project, fieldwork is conceived as no longer site specific in a literal sense. It has become multisited not in the sense of making Malinowskian
scenes of encounter several times over but through a different strategy of materializing
the field of research at certain locations – these are questions of scale making, a located,
situated logic of juxtaposition in which fieldwork literally and imaginatively moves and
is designed as such. The object of study is not a particular cultural structure or logic to
be described, analyzed, and modeled but the exploration of the anthropologist’s relations to social actors who are both subjects and partners in research.
What is of interest here in this proposed third scenario of alternative solution to the
contextualization of the student’s core argument is that it shifts the aesthetic of holism
in anthropological research from its conventions in terms of a question of representation of part to whole to a problem of the management and design of the ethnographic
research process itself. Methodologically, holism is the material, so to speak, out of
which the field of ethnography must be tailored. This anticipates the problem of constituting the field of research out of global relations in their vastness, among “wholes,”
or publics of potentially boundless participation – a problem confronted by every
organized project of Internet communication (such as in the success of Wikipedia [see
Zittrain 2008] or the “recursive publics” of open access movements [Kelty 2008] and
sustained blogs [Saka 2008]). How concepts derived from Internet projects are fashioned for use in teaching is the subject of the second, and final, exhibit.
Exhibit B: Chris Kelty’s experiment in teaching
fieldwork as “composition”
In his first teaching position in anthropology at Rice University, Kelty developed an
experimental course in teaching fieldwork methods that derived from his study (2008)
of how inquiry proceeds on the Internet. He reported (2009) on one iteration of this
experiment that concerned the study of the culture of computer scientists on the Rice
campus by the students of the seminar constituted as a collaborative research team.
Very much a student of post-1980s anthropology that refunctioned itself around the
project of critique (the task of ethnography being to clarify and expand discourses and
practices of reflexive critique found in the scenes, idioms, and practices of the subjects
of fieldwork), he has rethought the terms of this project in the idioms of knowledge
making on the Internet.
For Kelty, the holism in any ethnographic project today, and especially in teaching ethnographic research process to anthropologists-in the-making, is in the
management of the complex and diverse bodies of materials that it accumulates.
This is not a phase of research but its constant activity. The task of composition is
fundamentally social, and where knowledge is made through coordinations and collaborations. Results can be parsed in the individual labor of conventional writing,
which brings us back to the many debates and issues about representation. But the
broader frame of projects continues by other media, and this is where the holistic
aesthetic within the professional craft culture of anthropology might be most satisfyingly served today.
In this course experiment, he has produced a refraction of the organization of network inquiry as a source of innovation in method of current forms of fieldwork
9781444333237_4_003.indd 40
8/7/2010 6:43:03 PM
41
practice. The following pastiche of quotations from his report (2009) appropriately
serves as a conclusion to this epilogue and finally to this chapter.
Kelty asks, “How can we tease apart millennial promises of Internet-inspired radical
transformation from the clearly felt, but dimly perceived, material changes in how we
research, write, disseminate and receive our work, specifically in terms of the practice
of fieldwork?” (2009: 185). His teaching project
was not intended to reach its apotheosis in published articles or books, but to go on living in diverse print and electronic forms as long as there were (or are) graduate students
or faculty interested in pursuing or expanding the materials and/or addressing the problems; and second, it is open to participation by anyone wishing to make use of the existing materials, provided they contribute their work back to the project.…
From the perspective of digital media, the concept and the practice of “writing” neither
exhausts the challenges of conceptual innovation and anthropological analysis, nor is it
sufficient to describe the range of practices anthropologists increasingly engage in, from
field-note writing to blogging, from book-lending to social book-marking, from letterwriting to listening and observing, from qualitative data analysis to collaborative interviewing, from draft articles circulated amongst informants and colleagues to public wikis.…
“Composition” is intended to be a somewhat broader concept that captures this diversity because it is more inclusive than “writing.” Composition is a practice that crosses
between writing understood as an artful craft and research understood as a conceptual
innovation (or more specifically, fieldwork understood as an epistemological encounter
and platform for shared conceptual work).…
Composition, therefore, confronts new challenges that extend (if not alter) the familiar notions of style, audience, idiom, jargon, and readability.… Experimenting with commentary and composition is therefore a way to rethink the kinds of texts and new objects
we produce so that they might fill both needs and respond to the reality of trying to do
so (e.g., articles and books, documents and files, databases software, content and version
management, vocabularies and taxonomies, archives and search engines and so on).…
Coordination is the media-specific material and technical choices that allow a group of
people to work together on similar topics, in the same places, in structured time frames,
and with the same group of subjects. Collaboration is the conceptual and theoretical work
that results, if it results. Coordination does not imply collaboration, but collaboration
entails coordination. (2009: 187)
Kelty describes a circular process of “problem, inquiry, analysis”:
The circular process itself is the basis for experiments in “composition” – experiments
that aim to fix this process in multiple media: as an accessible and navigable website (navigable to both researchers and potential readers); as a glossary of concepts that emerge
from the problem-inquiry-analysis circle; and as papers, articles, and other standard
scholarly forms. Coordination between researchers takes place in all three stages, coordination between researchers and subjects takes place primarily in the inquiry stage, but
overflows into the others to the extent that interviewees are interested in reading or
contributing to the material. (2009: 192)
In this respect, my own work on Free software provided a useful foil and template for
understanding our goals. But Linux does not proceed by strong design criteria; it proceeds only with respect to what contributors produce and what the leaders accept as
“working” – a non-trivial problem that entails a fantastically refined engineering
knowledge and aesthetic sensibility, and which is analogous to the problem of “composition” raised here. Contributors to Linux “compose” new ideas and forms out of the
9781444333237_4_003.indd 41
8/7/2010 6:43:03 PM
42
Marcus
existing platform without regard for, or against, the historical plans, meanings and functions of “operating system.” (2009: 190)
Such an experiment, if successful, amounts to a strongly, perhaps radically “conventionalist” approach: research problems and approaches do not refer back to disciplinary
questions or “theory” as a guiding hand but directly to the practice and outcome of
fieldwork itself – an immanent theory and critique. (2009: 189)
The implication of this experience, for our research group, is that deep within computer science, there is an immanent critique of government – one which Computer
Science researchers find themselves reluctantly and somewhat awkwardly trying to
articulate in the only language they know—that of science. Part of the goal of this
project, then, has been to attempt to articulate, through dialogue and commentary,
the nature and meaning of this immanent critique. (2009: 197)
While it may well be possible to perform critical, deconstructive or oppositional readings of the informants we deal with, our attention has instead focused on finding those
voices where they appear in the field first and trying to amplify them, transcode them, or
simply explain them in our own work. (2009: 197)
Composition, therefore, confronts new challenges that extend (if not alter) the familiar notions of style, audience, idiom, jargon, and readability. Our primary audience will
not always be anthropologists or social theorists, but it is social theorists or anthropologists who will always determine the value of, need for, fundability of and quality of our
work. Experimenting with commentary and composition is therefore a way to rethink
the kinds of texts and new objects we produce so that they might fill both needs and
respond to the reality of trying to do so (e.g., articles and books, documents and files,
databases software, content and version management, vocabularies and taxonomies,
archives and search engines and so on). (2009: 206)
Notes
1 A parallel to the holistic concern in anthropology is the modern tradition in Marxist thought
that concerns itself with totality, as Jay (1986) traces. But holism in anthropology has none
of the absolutism and elaboration of the complex philosophical and theoretical debates about
totality in Marxism. It is probably useful to distinguish the flexibility and fuzziness of the
application of holism in anthropology – these traits are what make it so generally influential –
compared to the ironic reductiveness of totality in theory. Holism in anthropology sometimes evokes totality – all that there is or has been in society and nature, and in life – but most
often it refers to something more modestly (but just as impossibly) functionalist and to the
sorts of claims and expert competence in knowledge that should stylistically come from it. In
social-cultural anthropology today (see “Act 3,” this chapter), being holistic means being
able to contextualize in the broadest possible way and being rigorously nonreductionist
rather than being able to encompass or engage totality – as an abstraction or a truth.
2 In this sense, it was probably easiest or most natural for linguistic anthropologists to be holistic, and why models of language were once in vogue as the template for holistic accounts.
3 For example, here is an inquiry I received recently from a scholar in informatics, who works
on the popularity of games in China.
Hi George,
Is there any recent writing on cultural universals that you recommend? I got to thinking about them as I am observing activities that mimic initiation rites and ritual in
World of Warcraft.
Thanks.
9781444333237_4_003.indd 42
8/7/2010 6:43:03 PM
4
5
6
7
8
9
43
What the anthropologist might “authoritatively” provide is highly marked and stereotypic
in this request, despite the fact that the domain of current ethnographic research of anthropologists overlaps with the inquirer’s own. Indeed there are distinctive disciplinary aspects to
anthropology’s contributions in this research domain that are more subtle than those evoked
in the inquirer’s question. They have to do with the retention of a certain aesthetics from past
craft culture (e.g., “What happens to the trope of holism?” – the question that is examined in
this book) while moving away from its more overt topoi (like “universals,” “rituals of initiation,” etc.) in anthropology’s post-1980s investment in the enterprise of critique of the
emergent or alternatively modern, the contemporary, and the global.
Frankly, I was at a loss to be able to recommend to her any prominent recent writing on
these topics. I could only point her to classics.
I have found that, interestingly, scholars who receive their training as anthropologists in
interdisciplinary programs (most prominently today in science studies graduate programs)
find the holistic aesthetic of anthropology’s professional culture a bit strange, even tribal,
and certainly superfluous, although they perform the same referred to values in their
research. To them, the appeal to holism seems like what it is – an expression of a particular
professional culture and history.
My own explicit treatment of the problem of holism during the era of anthropology’s
critical exploration of ethnographic representation was Marcus (1998b [1989]) and also
Marcus (2001). Robert Thornton (1988), in my view, provided the most important discussion of this aspect of classic ethnography. His paper was to have been part of the volume
Writing Culture (Clifford and Marcus 1986), but it could not be revised in time for its
inclusion.
The other major and overlapping metanarrative that provided a context in which anthropologists could produce and write ethnography after the 1980s was postcolonialism. This
was an essentially historicizing holistic frame that permitted anthropologists to construct
the mise-en-scène of ethnography as they had in the past. It was also a framing that was
deeply and overtly constructed from realist literary chronotopes and the tropes of classic
metahistory that Hayden White (1973) provocatively discussed. Defining ethnography
within the realm of the emergent global – the present becoming the near future – stimulated
more of a rethinking of ethnography beyond its classic area studies pursuits. The fashion for
studying alternative modernities was caught somewhere between these overlapping metanarratives of postcolonialism and globalization.
Hans-Jorg Rheinberger, who has focused on experiment in natural science (see Rheinberger
1997), has done much to bring a convincing association between, rather than an idealization of, what natural scientists do in research and aspects of what social scientists do. The
association is not about the production of results, but the conception of problem in the
conduct of research. In Rheinberger’s view, experimental systems are to be seen as the smallest integral working units of research. As such, they are systems of manipulation designed to
give unknown answers to questions that experimenters themselves are not yet able clearly to
ask. This is a lot like the process of ethnography playing out in a time–space of emergent,
contingent wholes.
The central discursive space in US anthropology at present is occupied by calls for and conceptions of the practice of public anthropology. This in my view is indexical of the current
rethinking and reauthorizing of the forms and function of anthropological knowledge arising from how its projects of research first produce significant descriptive-analytic accounts
in realms of knowledge making and discussion – the “public” – that are part of the sites of
ethnography itself. There is a still submerged sense of holism here that, as I argue in the
Epilogue, is now being defined in the proliferation and development of Internet modes of
communication.
Since the 1980s, much mainstream social-cultural anthropology has become a minor science –
distinctly not an insignificant science – in a sense analogous to Deleuze and Guattari’s discussion of minor literature (1986 [1975]) in which social problems are not solved or
9781444333237_4_003.indd 43
8/7/2010 6:43:03 PM
44
Marcus
explained in holistic terms, but which is rather a medium where conflicts might be articulated in different registers.
10 Of course, this sounds familiar, but it is the valence of this longstanding critical function
that has shifted over the past two decades from minor key to raison d’être, and this shift
has made a considerable difference in how key aspects of anthropology’s strongly held
aesthetic culture of practice have been sustained.
11 Anthropological research as a critical enterprise is not necessarily only the study of knowledge forms, or of experts, or of elites, but the object of such critique often and eventually
engages at this level, even when it is done in villages, at the level of ordinary life, and is used
to comment on, qualify, or make nonreductive arguments about some influential macroperspective or theory, or policy, in play. A taste for how ethnography typically engages with
the holism of other knowledge-making regimes comes in this comment from Douglas
Holmes, who has been studying central banking cross-culturally and transnationally. He
was responding to an e-mail from me in which I mentioned working on this chapter.
June 8, 2008
The ECB [European Central Bank] and the RBNZ [Bank of New Zealand] both
published amazing reports last week and I have been trying to work through them.
I decided to quickly do a piece outlining the economy of words idea.
The punch-line goes something like this: I outline the operation of an economy of
words, which encompasses the technical means for modeling linguistically economic phenomena operating at the limits of calculation and measurement. In this
economy “at large” or “in the wild,” as Callon describes it, words perform the decisive function of creating context – countless contexts – that frame data series, statistical measures, and econometric projections. This is the world, our worlds, where
ceteris paribus does not necessarily obtain, where the rational and irrational co-exist
or may be entirely inseparable, where knowledge is imperfect, and where information
is asymmetrical, alternative heuristics gain articulation. The shifting and fugitive
dynamics of global markets, their operation from innumerable perspectives is made
available to us through the inter-mediation of language, through agile linguistic scenarios that are susceptible to continuous modification and elaboration. The economy
of words is by no means indifferent or antagonistic to the realm of numbers, far from
it; its analytical modalities in the first instance are shaped by the analytical predicaments posed by various forms of statistical measurement and quantitative analysis.
I suppose that contra-holism, my aim is to delineate acute narratives that can
mediate between a dynamic macro-level and countless and diverse micro-level
phenomena.
Anyway, Doug
12 Strathern is thoroughly into the “multisited” solution, discussed earlier, of the classic
problem of holism by contextually and contingently evoking “wholes” as the researcher
goes along in the mode of experiment – her remarks capture this very well, as does her
pioneering monograph on scale, Partial Connections (2005 [1991]).
13 In this, I am inspired by Gregory Bateson’s use of the epilogue form (one in 1936, and
another in 1958) to evoke in some detail implications of his famous monograph Naven
(1958 [1936]) that did not fit in its genre conventions, even given its experimental nature
as ethnography.
14 There are countless discussions, in print publication and online, and in constant production, about Internet forms of sociality. The most influential and current for me in suggesting what the aesthetic of holism is becoming in the tradition of anthropological ethnography
are as follows: discussions and seminars of the European Association of Social Anthropologists’
Media Anthropology Network (2004–10); the blogs and archives of Geert Lovink’s
Institute of Network Cultures (Lovink n.d.-b; see also Lovink n.d.-a), in which projects of
9781444333237_4_003.indd 44
8/7/2010 6:43:03 PM
45
Internet criticism are being forged; Chris Kelty’s Two Bits (2008); Johannes Fabian’s
Ethnography as Commentary (2008); Jonathan Zittrain’s The Future of the Internet and
How to Stop It (2008); Manuel Castells’ Communication Power (2009); and the Asthma
Files, a Web project organized by Kim and Mike Fortun (Fortun and Fortun 2010).
Given the oceanic quality of the Internet, and the unboundedness and instability of the
networks of communication that one can form in it, I find that my continuing knowledge
of and know-how in using it, both personally and as a researcher, are ironically quite local
and localized. In everyday life, the Internet has the character of a Maussian “total social
fact,” but one that cannot rely on conventional part–whole understandings of context to
act on this knowledge and know-how. Rather, for exemplars, advice, and workable concepts in the task of engineering or designing communication environments for contemporary ethnographic research, I need to consult those locally, who know more and whom
I know by conventional association in sustained collegial, student, and friendship relations.
As a holistic phenomenon, the Internet is dauntingly oceanic, but working knowledge of
it is local – at least in my case.
15 This interest is most fully explored in Faubion and Marcus (2009), but the Center for
Ethnography at the University of California, Irvine, founded in 2005, has defined projects
that are devoted to examining emerging changes in forms, practices, and norms of ethnographic practice (see Center for Ethnography n.d.).
References
Appadurai, Arjun (1996). Modernity at Large: Cultural Dimensions in Globalization.
Minneapolis: University of Minnesota Press.
Bakhtin, M. M. (1981). Forms of Time and of the Chronotope in the Novel: Notes Toward a
Historical Poetics [1925]. In The Dialogic Imagination, trans. and ed. Michael Holquist.
Austin: University of Texas Press.
Bateson, Gregory (1958). Naven [1936]. Stanford: Stanford University Press.
Bunzl, Matti (2008). The Quest for Anthropological Relevance: Borgesian Maps and
Epistemological Pitfalls. American Anthropologist 110(10): 53–60.
Castells, Manuel (2009). Communication Power. Oxford: Oxford University Press.
Center for Ethnography (N.d.). [Home page]. www.socsci.uci.edu/~ethnog/ (accessed
January 22, 2010).
Clifford, James, and George E. Marcus (1986). Writing Culture: The Poetics and Politics of
Ethnography. Berkeley: University of California Press.
Deleuze, Gilles, and Felix Guattari (1986). Kafka: For a Minor Literature [1975]. Minneapolis:
University of Minnesota Press.
European Association of Social Anthropologists (2004–10). Media Anthropology Network.
www.media-anthropology.net (accessed January 22, 2010).
Fabian, Johannes (2008). Ethnography as Commentary: Writing From the Virtual Archive.
Durham: Duke University Press.
Faubion, James, and George E. Marcus, eds. (2009). Fieldwork Is Not What It Used to Be:
Learning Anthropology’s Method in a Time of Transition. Ithaca: Cornell University Press.
Fortun, Kim, and Mike Fortun (2010). The Asthma Files. http://theasthmafiles.ning.com
(accessed January 22, 2010).
Jay, Martin (1986). Marxism and Totality: The Adventures of a Concept From Lukács to
Kelty, Christopher (2008). Two Bits: The Cultural Significance of Free Software and the Internet.
Durham: Duke University Press.
Kelty, Christopher (2009). Collaboration, Coordination, and Composition: Fieldwork After
the Internet. In Fieldwork Is Not What It Used to Be: Learning Anthropology’s Method in a
Time of Transition, ed. J. Faubion and G. E. Marcus. Ithaca: Cornell University Press.
9781444333237_4_003.indd 45
8/7/2010 6:43:03 PM
46
Marcus
Lovink, G. (N.d.-a). Geert Lovink, 1997–Present. http://laudanum.net/geert/ (accessed
January 22, 2010).
Lovink, G. (N.d.-b). Institute of Network Cultures. www.networkcultures.org (accessed
January 22, 2010).
Marcus, George E. (1998a). Ethnography in/of the World System: The Emergence of MultiSited Ethnography [1995]. In George E. Marcus, Ethnography Through Thick and Thin.
Marcus, George E. (1998b). Imagining the Whole: Ethnography’s Contemporary Efforts to
Situate Itself [1989]. In George E. Marcus, Ethnography Through Thick and Thin.
Marcus, George E. (2001). The Predicament of Irony and the Paranoid Style in Fin-de-Siècle
Rationality. In Irony in Action: Anthropology, Practice, and the Moral Imagination, ed.
James W. Fernandez and Mary Taylor Huber. Chicago: University of Chicago Press.
Rabinow, Paul (2008). Marking Time: On the Anthropology of the Contemporary. Princeton:
Princeton University Press.
Rabinow, Paul, George E. Marcus, James Faubion, and Tobias Rees (2008). Designs for an
Anthropology of the Contemporary. Durham: Duke University Press.
Rheinberger, Hans Jorg (1997). Towards a History of Epistemic Things. Stanford: Stanford
University Press.
Saka, Erkan (2008). Blogging as a Research Tool for Ethnographic Fieldwork. E-seminar,
EASA Media Anthropology Network. www.media-anthropology.net/saka_blogging.pdf
Strathern, Marilyn (2004). Commons and Borderlands: Working Papers on Interdisciplinarity,
Accountability and the Flow of Knowledge. Oxford: Sean Kingston.
Strathern, Marilyn (2005). Partial Connections [1991]. Walnut Creek, CA: AltaMira.
285–303.
Tsing, Anna (2000). The Global Situation. Cultural Anthropology 15(3): 327–60.
Tsing, Anna (2005). Friction: An Ethnography of Global Connection. Princeton: Princeton
University Press.
White, Hayden (1973). Metahistory. Baltimore: Johns Hopkins University Press.
Zittrain, Jonathan (2008). The Lessons of Wikipedia. In The Future of the Internet and How to
Stop It. New Haven: Yale University Press.
9781444333237_4_003.indd 46
8/7/2010 6:43:03 PM
4
Worlding the Matsutake
Diaspora
Or, Can Actor–Network Theory
Experiment With Holism?
Anna Tsing
Cultural anthropology and science studies have opposite problems these days.
Anthropologists love to put things in context: no cultural trait or social engagement
stands alone. It is the discipline’s reason for existing; it is the key move in its long fight
against utilitarianism. But anthropologists are no longer certain about what might be
appropriate contexts. What are the right units of context: ethnic groups, nations, the
globe? What structures are involved: ecology, symbols and meanings, political economy? In this volume’s terms, anthropologists practice holism without any clear idea of
what wholes to engage. It is a symptom of the problem that many young anthropologists have never heard of holism or imagine it as an obscure practice of the early twentieth century. Yet like Molière’s Monsieur Jordain, who found to his alarm that he had
been speaking prose all his life, anthropologists are practicing holistic arts – by putting
things in context.
In contrast, for many science studies scholars, context is the devil, blocking appreciation of truth-in-the-making. The challenge in studying science is to move beyond
the explanation of innovation as the heroics of scientists. Instead, a prominent branch
of science studies tracks the “actor networks” through which humans and nonhumans
work together, in contingent alliances, to make facts (e.g., Law and Hassard 1999;
Latour 2005). In this project, context gets in the way: context identifies the actors in
advance, making it impossible to attend to how they make themselves through networks. Still, this exorcism of context is performed with a certain amount of unease; a
number of troubling questions refuse to go away. If we stick with connections the
scientists describe, what about networks they downplay or deny? Might ignoring
silences limit where science studies can travel – for example, in regions neglected by
European scientists? And might ascribing networks to nonhumans always already
entangle us with the contextual propositions that allow us to imagine them?
Anthropological holists and actor–network theorists refuse to help each other here.
Context, anticontext: each proposes an impossible clarity in fact clouded by ghosts. In
the awkward relationship between anthropology and science studies, it is easy to want
to flip back and forth between claiming and refusing context. This chapter argues that
such a practice has a use and perhaps deserves a name. I will call it “worlding,” the
9781444333237_4_004.indd 47
8/7/2010 6:43:31 PM
48
Tsing
always experimental and partial, and often quite wrong, attribution of world-like
characteristics to scenes of social encounter.1 And while it would be possible to make
a social theory of this kind of worlding within encounters in general, in this chapter I
limit my focus to exploring how worlding is useful in the study of science. Might
actor–network theory (ANT) experiment with holism through worlding?
Let me take my questions to a classic of science studies, Michel Callon’s (2007
[1986]) “Some Elements of a Sociology of Translation: Domestication of the Scallops
and the Fishermen of St Brieuc Bay.” This is the article students turn to again and
again to understand the key science studies concept of nonhuman agency. Callon tells
of three French scientists who address the potential loss of the scallops fishery of
St. Brieuc Bay by applying a model they learned on a trip to Japan. Japanese resource
managers, they found, have encouraged and protected scallops through allowing their
larvae to anchor on collector lines. The three French scientists hope to use this model
to advance the interests of scallops, fishermen, and science in St. Brieuc Bay. Callon
shows how the three scientists are successful at first in forming the alliance that later
has come to be called an “actor-network.” Later, their work suffers from a series of
betrayals in which first the scallops and then the fishermen refuse to cooperate.
Callon’s article is considered classic for good reasons: it shows how actor networks
involving both humans and nonhumans make things happen. However, there are also
questions Callon does not ask – and perhaps cannot without extending his methods.
Consider the Japanese scientists and scallops hovering in the background. Most readers come away from Callon’s article not remembering that Japan is part of it at all. Yet
Japan starts the story and figures in the crucial translations: “The researchers are
inspired by a technique that had been invented by the Japanese.… [Yet] the number
of larvae which anchored on the collectors never attained the Japanese levels” (Callon
2007 [1986]: 64, 67). Why don’t Japanese scallops and scientists make it into the list
of relevant actors? Did the three French scientists somehow neglect to consult Japanese
scientists in constructing their project? Given that they knew the Japanese scientists
were farther along in building knowledge about scallops, wouldn’t the Frenchmen
have made them integral to their project? Since so much hinged on relationships
between French and Japanese scallops, wouldn’t they also have wanted to arrange a
meeting among the scallops? If not, what contextual constellation of social and natural ecologies – that is, what “world” – were they protecting?
There also seems to me to be a problem of terms. Did the Japanese think they were
“domesticating” scallops – or was this just the awkward and mismatched translation of
the French scientists? The concept of domestication requires a preexisting separation
between domestic and wild, a distinction with a long legacy in French sociality but without such a clear precedent in Japan. Were the Japanese, perhaps, engaged in a different
project of bringing humans and nonhumans into alignment? And in what ways, then,
did the French scientists mangle or dismiss the work of their Japanese colleagues?
These questions are difficult for Callon to address because they transgress his guidelines for the “sociology of translation.” They require attention not just to articulate
leads from the French scientists, but also to gaps, misunderstandings, and omissions.
Using the term I proposed, one could call these gaps, misunderstandings, and
omissions – whether from the French scientists or from Callon – elements of their
“worldings,” that is, their figurings of relevant social worlds. Through worlding,
Japanese scientists, science, and scallops have left the arena of action, dismissed as
irrelevant to the alliances that matter. Someone’s judgment of appropriate “wholes”
has blocked our vision.2
9781444333237_4_004.indd 48
8/7/2010 6:43:31 PM
Actor–Network Theory Experiment With Holism
49
This chapter explores the roads not taken in Callon’s classic analysis. In the next
section, I review ANT in more detail through reference to the work of Bruno Latour.
I then move from scallops to mushrooms for a series of worlding exercises through
which, in close dialogue with Latour’s presentation, I attempt to inhabit the potentially productive boundary between holism and actor networks. I follow ghosts and
gaps: in silences, denials, mistranslations, and refusals I find instances of worlding,
that is, propositions about context.
Studying Science With the Ants
Bruno Latour has done social analysts a favor by explaining ANT in careful terms in
his book Reassembling the Social (2005). Latour argues that, although ANT originated in the study of science, it can be the basis for a general reformulation of sociology. Here are some of the book’s key points:
●
●
●
●
●
The “social” is not a kind of material, to be contrasted with biological, economic,
and so on, but rather a process of association, in which dissimilar things are connected through transformative mediation. Rather than presuming social groups,
social analysts should look at processes of group formation.
Social action should be analyzed not as the result of actors’ intentionality but
rather as the work of nets of association. Nonhumans have agency through such
nets, called “actor networks” to stress the formation of actors through networks
and vice versa. Actor networks are formed through “translations” in which social
agency is forged.
The proper job of social analysts is to follow the actors through their own ways of
tracing networks. This allows a “flatness” of analysis in which the investigator
refuses to impose a “three-dimensional” context that explains away the phenomena that need to be investigated.
Analysts should not assign scales or units of analysis in advance but rather follow
the actors into multiple scalar exercises. This means attending to how actors’
“panoramic” views are themselves made possible – and limited in scope – by localizations. It means attending to how face-to-face interactions bring far-reaching
networks to bear on local situations.
Not everything is part of an actor network. The existence of a mass of stuff
(“plasma”) outside of social association is what makes the action and agency of
association so important and exciting.
I find Latour’s explication in Reassembling (2005) playful and refreshing. Among
other devices, he offers us the figure of that humble social insect, the ant, as a guide
to the meticulous work of ANT in following social trails. I am charmed and enrolled.
I endorse this program wholeheartedly with no objections. My argument here is that
if ANT and its ant keep traveling in the direction they have set, they may find themselves engaged with worlding despite all denials. Since ants smell with their antennae,
perhaps the ant will not be put off by the three-dimensional airborne whiff of “context”; instead, tactile exploration may reveal a useful trail-making practice.
Worlds are also present in ANT; Latour uses the term to refer to the strings of
assemblage his method reveals, but he does not allow world-making claims that
9781444333237_4_004.indd 49
8/7/2010 6:43:31 PM
50
Tsing
override the musings of his most articulate actors. This has the result of limiting the
network to the social assemblages closest to his informants. In contrast, worlding as a
tool asks how informants as well as analysts imagine the relationality of worlds that are
self-consciously unfamiliar – whether across cultures and continents or across kinds of
beings and forms of data.
The need to engage worlding is related to the problem of figuration, a concept
Latour uses to explain the shape of agency (2005: 53). We identify actors through
their figuration, he says; and I agree. But “figuration” is a term, like worlding (in my
use), borrowed from literary criticism, and they work together: Literary figures move
us to the extent that they bring us into semiotic worlds. We can only identify figures
to the extent that we can imagine worlds, that is, the systems of relationality through
which figures emerge. Figures are relational elements of worlds. It is possible for a
world to seem so much like common sense that the analyst “knows” the figures without thinking. But this is just the kind of situation worlding arms us against.
Where Latour is right is that it would be a mistake to imagine worlds as rigid containers into which we can pour our empirical materials. As Latour points out, this
explains away everything we might want to know before we have a chance to explore
it. This is the argument against holism: it has defined much too rigid units of analysis
and given them far too definitive internal divisions. Instead, worlding offers provisional guides both for and against disorientation; when we can’t identify figures, we
grasp at the worlding projects of our informants and start making up our own in
translation of theirs.
In this use of the term, worlding can also be quite wrong. As researchers, and as
ordinary human beings, we constantly misjudge the situations we enter. We fill the
blanks of our ignorance with prejudice or inappropriate “common sense.” At best,
assessments of worlds are partial, tentative sketch maps. Worldings are attempts to
grasp the contours of a situation, to find its figures, but they do not describe what is
actually happening. Even with learning and reevaluation, we advance necessarily
flawed proposals. To participate in worlding should not be to glorify it or raise it to
omnipotence. It may be useful to discriminate between more versus less insidious
worlding projects, but it is never possible to elevate one as the final resting place for
analysis. To be useful, worlding must retain its place within the material encounter
between analyst and unfamiliar material.
All researchers, including both natural and social scientists, use worlding to assess
their research materials, that is, to place them in what seem to be relevant webs of
relationality. Scientists use worlding, for example, to consider the potential relationships among observable bits of data. Formal methods produce data points, but they
do not in themselves transform the data into a story. That work requires setting a
systematic context, a “world,” in which the data form part of a pattern. To understand
scientific storytelling practices, it is necessary to appreciate the work of worlding, even
where it is downplayed or denied in lavish descriptions of formal methods.
Scientists also use worlding to assess their relationships with colleagues in other
laboratories, other disciplines, and other regions of the world. Callon’s French and
Japanese scallop scientists are separated by their perhaps unspoken propositions about
relevant webs of association.
Attention to each of these forms of worlding, I argue, enlivens science studies. The
second is perhaps easiest to grasp: without attention to science and scientists “elsewhere,” science studies becomes limited and parochial. The first, however, is equally
pertinent. We cannot assess nonhuman agency without attention to the scientific
9781444333237_4_004.indd 50
8/7/2010 6:43:31 PM
51
storytelling through which we learn about actor networks. Hidden and implicit worlding practices are the complements to articulate assessments of the scientific process.
In what follows, I explore these forms of worlding through a series of exercises in
which I take the part of a Latourian ant, but an ant open to experimentation. My ant
plays with multiple, partial, provisional contextualizations, each limited to the scene
of those social encounters where a sense of arrangements needs to be made. By chance,
I have a story to tell that recalls Callon’s work, but with mushrooms instead of
scallops.
From Scallops to Mushrooms
The matsutake complex is a group of species of the genus Tricholoma, which grow in
symbiotic relationship with trees in forests across the northern hemisphere. Matsutake
mushrooms are a valued delicacy in Japan, but, because of the decline of the kind of
forests in which matsutake grow, matsutake became rare in Japan by the 1970s. The
rising Japanese economy then stimulated matsutake picking in other countries for
export to Japan. The high value of the mushrooms brought them to the attention of
foresters and scientists, and matsutake science and resource management, once mainly
limited to Japan, have now proliferated in many picking areas. My research has taken
me between Japan and Oregon, in the US Pacific Northwest, where matsutake picking and matsutake science are both active. My fieldwork forms part of a collaboration
that has made central the intercontinental questions Callon put to one side.3
As with Callon’s Frenchmen and scallops, US matsutake scientists in Oregon recognize that Japanese scientists have a much longer history of studying matsutake than
they do. During the buildup to research in Oregon, which began in the late 1980s, some
effort went into having Japanese articles translated into English. Rather to my surprise,
however, Oregon scientists proceeded mainly to ignore these articles, inventing their
own matsutake science as if de novo. When I discussed this issue with scientists in
Oregon, they tended to shrug off the matter or even to deny the gap completely. If
pressed, they allowed that the Japanese articles had turned out to be disappointing. They
were “descriptive,” they said, not tackling analytic questions. They were “old fashioned,”
resembling US science from an earlier period. These dismissals were not well-workedout policy statements; they were offered to me as asides, without the status of science.
Already it appeared that if I wanted to pursue the question, I would have to attend to
little gestures such as laughter, inattention, and irony rather than formal statements.
The situation was even more confusing in Japan. The Japanese matsutake scientists I spoke with were very clear that they belonged to an international scientific
community. The differences that mattered among scientists, they explained, were
not national differences but instead disciplinary differences, such that training as an
ecologist, a botanist, or a bacteriologist might shape divergent research questions.
Yet when I asked them about particular American publications, they had rarely read
them. Their matsutake science did not require these articles. Matsutake science in
Japan is robust enough to be self-centered.
What about the mushrooms? The differences between matsutake mushrooms in
Japan and the US Pacific Northwest have been the topic of considerable investigation.
Here it seems we might escape the shifting bog of innuendos for the firmer terrain of
Latour’s “matters of concern” – that is, facts in the process of being made. Yet here,
9781444333237_4_004.indd 51
8/7/2010 6:43:31 PM
52
Tsing
too, there is trouble for following actor networks. It happens that the handiest method
for studying the differences among nonhuman organisms is itself, like ANT, a denier
of the need for the discussion of parts and wholes. As with ANT, this means that
worlding takes place in the margins, where the analyst can’t really see it, much less
examine it as science.
With scientists and mushrooms, then, I introduce two modes in which consequential worlding might take place within science without anyone noticing. First, why do
my scientists deny national difference despite the fact that it is apparent in their practices? Might this be one kind of worlding? Second, might another kind of worlding be
involved in studying mushrooms through a research method that segregates species as
data points, yet requires wider relational worlds to interpret its results? These questions signal two modes of indirect, underrecognized worlding. They help us to understand why scientists – and those who study them – might not think about worlding
even as they engage in it. I would like each of these two modes to do double duty in
my analysis. On the one hand, they are challenges for the ant that follows actor networks to see how things are assembled. On the other hand, they are allegories for how
ANT necessarily incorporates worlding even in excluding it from analysis. In each
case, my stories comment indirectly on the haunting presence of worlding in Callon’s
analysis of scallops.
Furthermore, by happenstance, the method for studying differences among mushrooms turns out to be a familiar one for anthropologists: it is an extension of the
“comparative method” of the nineteenth century through which the disciplines of
anthropology, linguistics, and biology came into being. To review the problem of
worlding in matsutake phylogeography (the study of the historical development
of present populations) takes us, then, into the history of anthropology, where related
methodological dilemmas eventually gave birth to both holism and modern anthropology as we know it. This seems an auspicious place from which to offer a more
technical description of one wing of matsutake science.
Marginal Worldings
Phylogeography draws from the availability of DNA-sequencing technologies to offer
a lively historical agency to nonhuman organisms. If social analysts are willing to follow phylogeographers into their eon-stretching time scales, we are rewarded with a
whole new arena of social action to consider. Here kinship ties between scallops or
mushrooms across continents are shown to be recalled in the flesh, generating sagas
of both “roots” and “routes” in the making of nonhuman diasporas. Like anthropologists, phylogeographers want to know about the making of diversity and its distribution around the earth. To extend anthropological accounts of global history to include
nonhuman organisms, we must enroll phylogeographers. Yet to participate in collaborative storytelling, we need to learn something about the assemblage of this science.
How should an ant proceed to learn about phylogenetics (the study of evolutionary
relatedness among organisms)? In Reassembling, Latour suggests that the ant begin
by “localizing the global,” that is, offering a concrete geography to projects and
propositions (2005: 173–4). In studying matsutake phylogeography, one might begin
with a research report, such as Ignacio Chapela and Matteo Garbelotto’s (2004)
“Phylogeography and Evolution in Matsutake and Close Allies Inferred by Analyses
9781444333237_4_004.indd 52
8/7/2010 6:43:31 PM
53
of ITS Sequences and AFLPs.” This paper compares short DNA sequences from
matsutake mushrooms taken from a variety of geographical locations. The ant might
be excited to find that the scientists are also looking for associations. Which populations are most closely derived from which others? Furthermore, the ant might be
thrilled that the phylogeographers, like the ant, commit themselves to a “flat” analysis
without three-dimensional impositions. Instead of beginning with holistic ecological
communities, the researchers go straight to their “actors”: base pair sequences on a
stretch of DNA. The researchers mobilize these actors in polymerase chain reactions
through which the actors “express” themselves to the scientists. These expressions can
then be brought into common parliament by the scientists, who draw from their association a family tree. In solidarity, the ant might crawl to one of these family trees,
reproduced here as Figure 4.1. Here, in a vertical list tied by horizontal lines, the ant
can find a web of associations, including both Japanese and Oregon matsutake!
Without a background in this science, however, it is difficult to make heads or tails of
this figure. T. stands for Tricholoma; these are different matsutake, labeled by location.
But what is the purpose of this list, what do those connecting lines mean, and where is
the science in this? We have arrived at my first example of a scene of encounter, where
the analyst is confronted by the material and, confused, grasps at worlding.
Since the point is orientation, the most relevant worlding engages the competencies
of the analyst. How might I learn to identify the figurations in Figure 4.1? One route
is to focus on methods that connect phylogeography and anthropology, my home
discipline. This move follows Latour’s admonition to “redistribute the local” by investigating multiscalar elements in local interactions, in this case in Chapela and
Garbelotto’s report. But the first question to ask about figuration would concern the
appropriate system of relationalities, a concern Latour writes off as a mistake. Keeping
an open mind for the moment, let us consider where this trail might lead.
Worlding exercise 1
Phylogeography is an example of the comparative method, a method that shaped the
formation of anthropology, linguistics, and biology as disciplines in the nineteenth
century. What if we used parallels across the disciplines to explain phylogeography?
Might we learn something about their respective figurations?
Since this sort of comparative method is no longer much practiced in anthropology,
consider the somewhat familiar terrain (to anthropologists) of historical linguistics. In
this field, the analyst uses the comparative method by assembling a potential list of
cognate words, as in Table 4.1, from Hubbard (n.d.).
The next step is to establish sound correspondences among these cognates, and to
consider how one sound might have changed to another. The farther one goes back in
time, that is, the more distantly the languages are related, the more changes one
expects. Working with patterns of sound change in each language allows the analyst to
construct protophonemes, word sounds that could have characterized the common
ancestors of the words in the list. In Table 4.2, asterisks identify the word as part of the
analyst’s evolutionary reconstruction, built from the range of contemporary sound
diversity. These protophonemes are thought of as elements of an ancient language that
could have evolved into the variety of modern words seen on the original list.
Phylogeographies work very similarly. Indeed, since DNA sequences are usually
represented as letters, which indicate their constituent base pairs, DNA sequences
9781444333237_4_004.indd 53
8/7/2010 6:43:31 PM
54
Tsing
Image not available in this electronic edition
Figure 4.1 Matsutake phylogeography
The original caption reads,
Phylogram representing one of the 30 500 most parsimonious trees obtained from maximum parsimony analysis of ITS sequences from matsutake specimens and Clitocybe lateritia as an outgroup. Specimens are identified according to sequential number and
code-name from Table 1. Branch lengths are proportional to the number of changes
connecting each node. Bootstrap values above 50% are provided in italics. (Chapela and
Garbelotto 2004: 734)
“Bootstrap values” – the numbers on the horizontal lines – refer to tests for parsimony in ranking samples. CB = circumboreal; AT = Atlantic; MX = Central America; and WNA = western
North America.
Source: From Chapela and Garbelotto (2004: 734). Reprinted with permission from The Mycological
Society of America.
9781444333237_4_004.indd 54
8/7/2010 6:43:31 PM
Table 4.1
55
Cognates (words descended from a common ancestor)
Sanskrit
Avestan
pita
padam
bhratar
bharami
jivah (‘living’)
sanah
virah (‘man’)
barami
jivo
hano
viro
Greek
Latin
Gothic
pater
poda
phrater
phero
pater
pedem
frater
fero
wiwos
senex
wir
fadar
fotu
brothar
baira
qius
sinista
wair
henee
English
father
foot
brother
bear
quick
senile
were (wolf)
Source: From Hubbard (n.d.). Reprinted with permission from the author.
Table 4.2
Proto-Indo-European reconstructions
father
foot
brother
carry
live
old
man
e
*p ter*ped*bhrater*bher*gwei*sen*wi-ro-
Source: From Hubbard (n.d.). Reprinted with permission
from the author.
Table 4.3 DNA base pairs from varied matsutake
ITS2 : Sequencing of Clones
101
RM17
RMC9
RM21
RMB10
Re36
RE39
RM11
RM16
ATTCCTCAAC
ATTCCTCAAC
ATTCCTCAAC
ATTCCTCAAC
ATTCCACAAC
ATTCCTCAAC
ATTCCTCAAC
ATTCCTCAAC
150
CTTTTCAGCT
CTTTTCAGCT
CTTTTCAGCT
CTTTTCAGCT
CTTTTCAGC.
CTTTTCAGC.
CTTTTCAGC.
CTTTTCAGC.
TTTTGTTGAA
TTTTGTTGAA
TTTTGTTGAA
TTTTGTTGAA
TTTTGTTGAA
TTTTGTTGAA
TTTTGTTGAA
TTTTGTTGAA
TAGGCTTGGA
TAGGCTTGGA
TAGGCTTGGA
TAGGCTTGGA
TAGGCTTGGA
TAGGCTTGGA
TAGGCTTGGA
TAGGCTTGGA
TTTTGGGAGT
TTTTGGGAGT
TTTTGGGAGT
TTTTGGGAGT
TTTTGGGAGT
TTTTGGGAGT
TTTTGGGAGC
TTTTGGGAGT
Source: From Nakai (2005). Reprinted with permission from the author.
look a lot like words. To use the comparative method, the matsutake phylogeographer
sequences a stretch of DNA from related kinds of matsutake. The result looks rather
like a list of cognate words. In Table 4.3, the list is laid out 90° askew from Table 4.1,
so that the samples of various kinds of matsutake (akin to languages) are on the left
side, and the “words” or stretches of DNA from each sample extend horizontally to
the right.
Just as in historical linguistics, the next step is to look for changes in what are represented in each case as letters, that is, sounds in linguistics and base pairs in biology.
In phylogeography, as in linguistics, the number of changes in letters indicates a
9781444333237_4_004.indd 55
8/7/2010 6:43:32 PM
56
Tsing
greater evolutionary distance. If one assumes that base pairs mutate at around the
same rate, then this difference is also an indication of time, through which varied
samples can be compared to each other as having more or less distant common ancestors.4 As in historical linguistics, the analysis of contemporary diversity allows the
analyst to reconstruct an evolutionary past.
Perhaps this excursion allows us to look again at Figure 4.1 with new appreciation.
Figure 4.1 offers one way of organizing the DNA sequences from a set of matsutake
samples from various geographical areas. In each case, a piece of one mushroom represents the population of its area. The length of the horizontal line to the left of the
sample name indicates how many base pairs it differs from those to which it is connected. The shorter the line, the less time has passed since the evolutionary separation
of those two kinds of matsutake. The diagram thus offers us a geographical history of
evolutionary relationships, based on extrapolating back from contemporary variation.
Worlding exercise 2
Phylogeography carries with it the historical legacies of other instances of the comparative method. That’s why it’s more than just mildly amusing to learn to understand
it through historical linguistics. What might we learn by thinking about phylogeography through the experience of anthropology?
In anthropology, the nineteenth-century comparative method allowed analysts
to reconstruct evolutionary histories of culture from the evidence of present-day
practices. Marriage practices, rituals, or kinship rules could each be compared to similarbut-different customs from other groups; from their degree of difference, the analyst
would draw conclusions about the common ancestor from which each evolved.
The comparative method in anthropology achieved its force as an expression of science from the fact that the method could also be employed in linguistics and biology.
Its usefulness across all three emergent disciplines gave each a sense of legitimacy.5
Anthropology departments were established in this period through the charisma of
the comparative method. Anthropology might not exist as a discipline if it had not
been for this joint agenda. Biology similarly profited from the conjuncture.
By the end of the nineteenth century, however, anthropological reactions against the
comparative method could not be ignored. For the making of American anthropology,
the key intervention was Franz Boas’ paper, “The Limitations of the Comparative
Method of Anthropology” (1940 [1896]). Boas railed against the comparative method:
“The comparative method, notwithstanding all that has been said and written in its
praise, has been remarkably barren of definite results, and I believe it will not become
fruitful until we renounce the vain endeavor to construct a uniform systematic history
of the evolution of culture” (1940 [1896]: 280). Instead, Boas introduced holism:
“We have another method, which in many respects is much safer. A detailed study of
customs in their relation to the total culture of the tribe practicing them” (1940 [1896]:
276). American cultural anthropology was founded by following Boas. Evolutionary
explanations were expelled from the center, which was occupied instead by holism.
And while British social anthropology followed a separate course, by relatively early in
the twentieth century, British and Americans were in agreement about studying social
and cultural wholes – and not evolutionary lineages.6
Boas stressed that the comparative method was not empirically grounded, and
American anthropologists still describe the method as “speculative.” What was
9781444333237_4_004.indd 56
8/7/2010 6:43:32 PM
57
speculative about it? The customs being compared were as accurate as reports of the
time allowed. What contemporary readers find doubtful is the evolutionary ordering
that was imposed upon differences of custom. This ordering rested on worldings in
which some customs could be classified as “savage,” at the bottom of the human
evolutionary tree, while others were slightly higher as “barbarian,” and others yet
were “civilized.” The comparison of customs rested on propositions about worlds
that now seem to us purely a product of the European colonial imagination. Critics
argued that, although the comparison of customs did not require colonial prejudice,
the worldings that made sense of comparison did.7
Nineteenth-century evolutionary biology used the comparative method to reconstruct
similarly “speculative” histories through comparing the morphology (form) of organisms. As in anthropology, the morphological measurements were fine within the standards of the time. But contemporary biologists looking back find all kinds of unacceptable
worldings at the margins of the method, which yet controlled evolutionary storylines.8
Twentieth-century advances in genetics and then genomics revamped this comparative method. Instead of comparing morphologies, biologists now compare
sequences of DNA base pairs. Yet the method itself is the same. And – most importantly for my argument – the importance of marginal worldings to tell the evolutionary story remains. To explain this, let me return to Chapela and Garbelotto’s article
(2004). Figure 4.1 is an important result of the comparative method, but to turn it
into an evolutionary story requires a whole new set of exercises. These new exercises
are marked in the article by a new section, which follows “Results,” called
“Discussion.” In the discussion, lots of other species, not tested in the experiment,
show up. Most of these are host trees with which matsutake is associated in various
forest environments. Suddenly we have multispecies worlds – contexts – not just
stretches of DNA. These multispecies worlds are necessary ingredients of the evolutionary story the researchers are able to tell, drawing on those results in Figure 4.1.
It seems, they argue, that matsutake originally evolved with broadleaf trees and only
afterward moved to an association with conifers such as pine. In Japan, matsutake
have come to associate only with conifers, but in western North America, matsutake
still associate with both broadleaves and conifers. This suggests that those Oregon
matsutake have an ancestral status to the Japanese ones.
At last, we have a real story, and an exciting one! But it has taken ecological worlding to get us from the results of the comparative method to an evolutionary story. The
ascription of systems of relations is necessary to place the genomic comparisons into
ecological history. In this way, phylogeography shows the same vulnerabilities as nineteenth-century evolutionary anthropology. Informed as they are by a stack of evolutionary studies involving host trees as well as fungi, we can hope that Chapela and
Garbelotto’s worldings are better than colonial taxonomies. But the status of these
worlding practices outside the scientific method employed in the research is the same.
Worlding makes the story possible, but from the margins.
Worlding exercise 3
What is the status of my own worldings here, and what do they have to do with actor–
network theory? ANT is not the comparative method. The only thing the two have in
common is their refusal of context. The method of phylogeography excises ecological
context, treating each species’ DNA on its own. As a result, I have argued, the
9781444333237_4_004.indd 57
8/7/2010 6:43:32 PM
58
Tsing
mobilization of ecological worlds must take place outside what is identified as method.9
Ecological worlding makes scientific storytelling possible, but it cannot itself be
recognized as part of the science that produced those specific results. This comment
is not intended as a criticism of phylogeography research or Chapela and Garbelotto’s
article, both of which I love. Instead, I offer this description to help me think through
ANT – and my own research methods.
On the one hand, then, I have drawn an allegory. ANT disallows discussion of context. But because it requires the researcher to identify actor figurations, some kind of
worlding is necessary. In the exercises here, I cannot recognize phylogenetic figurations without putting them in some universe of relational parts and wholes. As long
as this worlding is allowed only to occupy the margins of the research, it produces
troubling blind spots, such as the unexplained exclusion of Japanese scallops and scientists from Callon’s analysis.
On the other hand, the use of worlding in my analysis is not to point to the force
of ecological or social wholes. Instead, my analysis requires worlding to mediate the
uncertain zone of encounter between the analyst and her unfamiliar material. A brief
visit with Marilyn Strathern may be helpful in explaining this point. In contrast to
Latour, Strathern is obsessed with context. She tosses contextual frames out at us in
a dizzying display, a back-and-forth relay between ethnographic and analytical framings of her materials (e.g., Strathern 1988). In common with Latour, her goal is to
destabilize the analytic process; like Latour, she does not want to smother social
analysis with preset frames. However, the comparison highlights their very different
strategies. Latour refuses context. Strathern disorients her readers by juxtaposing so
many combinations of contextual relations that it becomes impossible for the reader
to imagine the analysis resting in only one set. What if we imagined scientific papers
on mushrooms as something like Melanesians in Strathern’s analysis? Her Melanesian
worldings theorize the encounter between the analyst and the unfamiliar object of
study. The worlds we chose to engage are those that participate in forming this
contact zone.
In the encounter, I grasp at worlds, across disciplines, and into ecologies. Nor am
I the only one doing this. My worldings can only exist in dialogue with those generated through actor networks not of my own making. Such worldings may not always
be explicit, even at the margins; they may emerge from untheorized assemblages of
practice rather than formal statements. I turn to this problem.
Implicit Worlding
The one kind of worlding recognized in Reassembling is the “panorama,” a totalizing
view that is in fact limited by the conditions of its projection (Latour 2005: 187–90).
Panoramas are actors’ accounts of the big picture. But they mislead us about their
omniscience. The job of the ANT scholar is to point out the special effects that allow
viewers to confuse panoramas with reality, thus cutting the pretensions of panoramas
down to size.
There are many panoramas in matsutake science. The telling of evolutionary sagas
of the development of the matsutake diaspora, as in the “Discussion” section of
Chapela and Garbelotto’s paper (2004), might qualify. By pointing out the tangential
relationship between the saga and the results gained from the comparative method,
9781444333237_4_004.indd 58
8/7/2010 6:43:32 PM
59
I began the resizing work of the ant. Here, however, I turn to a problem for which
the concept of the panorama does not help us. Sometimes worlding emerges from
practices rather than representations.
To explore this problem, I turn from phylogeography to another kind of matsutake
science, one more closely related to the science in Callon’s scallop story. In both Japan
and the US Pacific Northwest, scientists are engaged with foragers (mushroom pickers) and resource managers (foresters and land owners) in addressing worries that
matsutake may not be flourishing as they should. This kind of science, as in Callon’s
case, involves making alliances between scientists, foragers, and the organism that
forms the object of their joint attentions. Callon’s analysis is helpful in considering
how these alliances wax and wane in Oregon and Japan, respectively – but not, however, across these sites, where alliances are not so explicit. To ask questions about the
gaps and connections between Japan and the US Pacific Northwest, we must begin
somewhere else.10
Again, we might start with the “localized global” of a report. Ecology and Management
of the Commercially Harvested American Matsutake tells of an unusual moment of
encounter between Japanese and US Pacific Northwest matsutake science (Hosford
et al. 1997). The first author, David Hosford, went to Japan to learn from a prominent
Japanese matsutake scientist, Hiroyuki Ohara. The first part of the report offers an introduction to Japanese matsutake science. Then comes a crucial transition statement:
American foresters, however, are likely to view the Japanese methods for enhancing matsutake production in a different context. Japan has been densely settled by people for
millennia and has few remaining native forests, so original forest conditions are less
important than desired future conditions. The area of matsutake habitat is much larger
in western North America than in Japan, and forest management goals differ greatly. The
preservation of biodiversity and healthy native ecosystems are important management
goals on public lands in North America. Introducing new matsutake colonies into previously noncolonized habitats raises questions about potentially harmful side effects, such
as displacing other sensitive fungal species. (Hosford et al. 1997: 50)
After this transition, the report changes, turning to the pioneering projects of an emergent US Pacific Northwest matsutake science, with their own logic. The goal of future
research is not to reproduce Japanese science but rather to create new forms.11
Certainly this passage is a panorama, in Latour’s sense, and an articulate instance
of worlding, in mine. Although I find its portrait quite sensible in many ways, I am
not tempted to take up the reader’s time unpacking the content or staging of this
passage. This is because of how this report has been used in all subsequent literature
coming from the US Pacific Northwest. First, all subsequent literature mentions
Japanese science only through citing this report. Second, no subsequent literature
takes up any of the themes of the panorama I quoted. No comparisons between
Japan and the US are offered at all. Instead, subsequent research goes happily into
US-based projects, having taken care of the whole problem of Japan by citing Hosford
et al.’s (1997) report.
The report is both a dead end and a sign of a new beginning. This is not an effect
caused by the report itself. US Pacific Northwest scientists cite the report to facilitate
something it would be easy for them to do and hard for them not to do anyway:
ignore Japanese science. To consider how this becomes so, it is no use staring at the
panorama, with its excuses for not changing who they are. Instead, it makes more
sense to ask US Pacific Northwest scientists about what they in fact do.
9781444333237_4_004.indd 59
8/7/2010 6:43:32 PM
60
Tsing
Worlding exercise 4
Most of the scientists who work on matsutake in the US Pacific Northwest are hosted
and/or funded by, or otherwise in collaboration with, the US Forest Service. The first
thing they tend to explain to you is the relationship of their research to the long-term
agendas of the US Forest Service. From the first, they explain, the Forest Service has
aimed to balance the use and protection of natural resources; this gives rise to management for sustainability. The goal of matsutake research in the US Pacific Northwest is
to promote the sustainability of matsutake.
The US Forest Service was founded in the early twentieth century in dialogue with
the timber and grazing industries in the American West (Steen 2004). In response to
what emergent foresters saw as the waste and destruction of unregulated logging and
grazing, they promoted a particular kind of “conservation,” aimed at protecting natural resources for future users. In forest management, they developed what they
called “sustainable yield,” that is, forestry in which the regrowth of trees could
replace the timber being cut. From the first, government forestry emerged in tense
and tender ties with loggers, and many compromises were made in the formulation
and practice of scientific rules imagined as promoting sustainable yields. Whether the
rules or the practices were the problem is still much under debate. In any case, by the
1980s, forests in the US Pacific Northwest – a central location of US Forest Service
management – were in bad shape (Langston 1995).
Meanwhile, an environmental movement had grown up to promote a completely
different vision of conservation, one in which nonhuman species should be preserved
for themselves, rather than for future human industry. The new environmentalists
argued that forests should be conserved as sites of biodiversity and ecosystem health,
not as tree farms. The environmental movement won some key legislative battles, and
the US Forest Service language of sustainability was stretched to include sustainable
ecological function as well as sustainable yield. Whether this expansion could make
sustainability anything other than an impossible contradiction has been fought in
many battles, in and out of court. Attention to so-called nontimber forest products
such as matsutake seemed auspicious in this climate: they promised an economic use
for forests, but one that did not require cutting down trees.
Configured as they were within this drama, matsutake could not escape sustainability management. Nor was it likely that concepts of sustainability would be imported
from Japan. Matsutake scientists needed to use a language the Forest Service could
understand. Thus, sustainability put matsutake in dialogue with timber. This dialogue
had important corollaries for research practices. First, the most important goal has
been to measure human impact, and the key site in which to measure this impact has
been seen as the harvest itself – as with trees in logging. US Pacific Northwest scientists became obsessed with the question of whether harvesters were destroying their
own resource. A good deal of research went into the question of harvesting techniques, to see if some were worse than others. The larger set of human–nonhuman
interactions that make up the forest beyond harvesting were not, however, included in
this research frame. Second, timber-sized scales were the focus of attention. US Pacific
Northwest scientists asked questions about stands of trees (basic units of timber management) and patch landscapes as timber “systems.” To achieve these scales, US
Pacific Northwest scientists needed statistical sampling methods through which particular observations could be generalized. This stimulated them to use randomly
9781444333237_4_004.indd 60
8/7/2010 6:43:32 PM
61
directed strip plots to monitor fungal productivity, despite the fact that this ignored
fungal patch dynamics. They also used computer programs to generate landscapescale management packages for imagined timber managers. Third, the preferred
method of research was monitoring. Forest Service understandings of sustainability –
whether as yield or as ecosystem services – are measured through species counts over
time. Thus, sustainability measurements require constant monitoring activity. This
became the first task of US Pacific Northwest matsutake science.
So far, I have followed what I was told by US Pacific Northwest scientists and
what I read in their publications. Let me turn to what Japanese scientists say about
what they do.
Worlding exercise 5
Japanese matsutake scientists are also worried about the decline of matsutake, but
rather than identifying this decline as a result of too much human impact, as in the
United States, they identify the problem as too little human impact. This is because
matsutake in Japan grow in forests understood as anthropogenic in their very nature.
Japanese matsutake are creatures of satoyama village forests, in which shifting cultivation and coppicing of broadleaf trees for firewood create the structure of the forest.
Peasants also raked fallen leaves and litter for fertilizer and gathered shrubs, grass, and
herbs. The process created open forests with nutrient-poor soils; Japanese red pine
invades such places as a pioneer species of disturbed areas. Matsutake grows with red
pine. Japanese scientists understand the decline of matsutake as a result of peasant
inattention to village forests (Ogawa 1991). The turn to fossil fuels since World War
II halted firewood collection, allowing broadleaf trees to shade out the pines, which
were also suffering from a wilt disease caused by an imported nematode. To promote
matsutake, scientists suggest a renewed peasant-type forest management, in which
broadleaf trees are cut back to allow the regrowth of pine – and matsutake.
As in the US Pacific Northwest, Japanese matsutake scientists address a larger social
and ecological agenda. Satoyama village forests are objects of interest from several
directions (Takeuchi et al. 2003). They are considered sites of great beauty, stimulating attention to the character of each season. They are also thought of as sites of traditional harmony between humans and nature, and thus places for nature education
and the transmission of values from elders to the young. The decline of satoyama
forests is regretted with self-conscious nostalgia; nostalgia makes satoyama recognizable figures for conservation. In working to revitalize satoyama, Japanese scientists
contribute to multifaceted environmental, aesthetic, and socially associative goals.
Many of the projects of Japanese matsutake scientists are advised by the need to
improve the conditions of matsutake cultivation. Such efforts can take place in a laboratory, where matsutake mycelia can be raised under controlled conditions. They can
also take place in a forest, where the effects of gardening practices at multiple scales
can be studied – from the microecologies connecting soil bacteria, fungi, and roots to
the patterns of soil or sunlight in variously managed forests. But not all cultivation is
“domestication.” Makoto Ogawa, a pioneer in matsutake studies, describes matsutake
as a product of “unintended cultivation,” an oxymoron from the perspective of those
committed to segregations between domestic and wild (personal communication
2005). Matsutake flourishes where human management practices have been conducive to the growth of red pine. To continue or even intensify ecosystems management
9781444333237_4_004.indd 61
8/7/2010 6:43:32 PM
62
Tsing
does not change the not-fully-tamed status of matsutake, which one Kyoto greengrocer described as “a gift from the gods.” To cultivate the forest adjusts human–
nonhuman relationships without necessarily subjecting nonhumans to the rigors of
permanent membership in the human family.
Worlding exercise 6
We have stepped into two quite different arrangements. The matsutake forests recognized by US and Japanese scientists are incommensurable; each is outside the
reckonings of scientists from the other region. As ecosystems, forests are figures,
identified through real-and-imagined, social-and-natural worlds. US national forests
and Japanese satoyama overlap only tangentially as “forests.” No wonder the scientific projects one can imagine in each have trouble communicating.
Consider the problem of incompatible scales. Japanese scientists map historical
configurations of fungi and trees. Japanese matsutake publications are full of site
maps, something that never occurs in a US Pacific Northwest publication; indeed,
it is this attention to site specificity that US scientists imagine as “descriptive.” US
scientists ignore historical configurations entirely to generate statistical probabilities
that might inform management at timber-sized scales. But thus they are completely
incapable of addressing the relational ecologies of particular sites. The results of
each research exercise whiz past the other without any communication.12
There is also the problem of US commitment to concepts of “industrial,” on the
one hand, and “wild,” on the other. Industrial forests need sustainable yields; wild
forests need sustainable ecosystems. The US concept of “sustainability” is starting to
look rather culturally specific, rather than being the universal science its supporters
claim it to be. Japanese scientists use the concept of sustainability, but it means
something entirely different for them. The ability of human-disturbed environments, such as satoyama village forests, to support useful and pleasing forms of biodiversity is sustainability for these scientists. This is not a concept of sustainability
that makes any sense to US foresters, with their focus on the industrial and the wild.
Japanese foresters assess sustainability in relation to aesthetics, education, and harmony with nature, concepts US scientists would dismiss as unscientific. Yet US versions of sustainability are similarly unscientific in Japan.
We cannot understand the figurations of our actors without engaging with their worldmaking dreams and practices, that is, the contexts they make for themselves and in which
they are made. Even where they do not include each other as figures, the worlds they spin
are revealing for what they leave out. The same is true for ANT. Even in denying worlding, its omissions are shaped by the spectral presence of other stories, other worlds.
Worlding at the Encounters
But what is the status of the contrasts I have drawn? If we assume that Americans or
Japanese have internalized national attitudes to nature that determine all their scientific projects – well, just as Latour argues, why do the research at all? It merely confirms our worst prejudices. It is in figuring out where worlding projects are not the
same as cultural, national, or scientific wholes that they become interesting again.
9781444333237_4_004.indd 62
8/7/2010 6:43:32 PM
63
Back with the ant, I refuse premature closure and continue my search in the
research process itself. The contrasts I offer between resource management science
in Japan and Oregon emerge from collaboration with the Matsutake Worlds Research
Group, in this case particularly with Shiho Satsuka.13 Dr. Satsuka was generous
enough to interpret for me in many of my interviews with Japanese scientists, and
my ability to say anything sensible at all about Japanese science comes out of my
stumbling efforts to understand her translations. We offer each other worlds as part
of the research collaboration process, each aiming to make sense of each other’s and
our own conversation. Furthermore, Dr. Satsuka offers her own theories for understanding science: Latour has famously extended the concept of “translation” to
refer to the making of links in human–nonhuman actor networks; it is time to bring
back linguistic understandings of translation in analyzing such networks.14
Cosmopolitan science requires continual translation.
This process of sense making also characterizes my interactions with the scientists
themselves: I map potential figures and worlds in the process of learning to ask good
questions. These worldings are orienting processes, shifting with the flow of the conversation, and never coming to rest in some perfect representation. Worldings, then,
are creatures of research encounters. Ethnographic fieldwork is an often painful process of shapeshifting, in which the ethnographer must turn her whole self into a potential interlocutor by becoming someone new. This process is itself worlding, a figuration
in someone else’s system of relationality. To act as a figure frames one’s performance.
The figuration sets up the possibility of conversation. But maybe this is – that forbidden word – “context”: making meaningful interaction possible.
All researchers develop their work in context-making collaborations, whether
explicit or implicit. Worlding is the only way to take difference seriously in a collaborative research practice.
The Uses of Disorientation
Localizing worlding projects this way – just as Latour would argue – allows us to
consider how unstable, incomplete, and misleading the worldings I have outlined are.
Worlding is simultaneously orienting and disorienting. Worlding is always practiced
in relation; worlds come into being at the encounter – and at best they explain the
encounter. When anthropologists first got excited about this feature of ethnography,
they thought it meant that the anthropologist, like Paul Rabinow (1977) in his
Reflections on Fieldwork in Morocco, should be a representative of the West confronting its Other, thus restabilizing both positions. It took feminist anthropologists to
show that relationality is always multiple and shifting. Marilyn Strathern may have
been the first to show how the interplay of contradictory relationalities is not only
disconcerting but also productive (Strathern 1987). Strathern then proceeded to
make this insight a keystone of her technique. By engaging in an excess of relational
worlding, she disorients the reader even in the act of orientation.
Strathern defines “the ethnographic moment” as what happens when anthropologists juxtapose frames from their own analytic apparatus with those that “dazzle” them
in their fieldwork experience (Strathern 1999). Her writing argues that it is impossible
to know anything about Melanesia without uprooting Euro-American knowledge practices and vice versa. Furthermore, since her position as a Euro-American feminist gets
9781444333237_4_004.indd 63
8/7/2010 6:43:32 PM
64
Tsing
in the way of her position as a Euro-American anthropologist, she is forced to juxtapose
within as well as across these categories (Strathern 1987). Somehow in the going back
and forth between partial and contradictory moments of “getting the point,” she suspects we will see more than if we let ourselves be satisfied with easy explanations.
This kind of excessive, disorienting, and relational worlding takes us rather a long
way from the original hopes of holism, which guided analysts toward a single stable
explanatory frame. But it has a lot to teach us about how holism works. Holism posits
a contextual frame. To stabilize that frame is the problem. How do we know that we
have the one proper frame, the one that explains everything? Latour’s insights are usefully provocative: sociologists in the Durkheimian tradition, he argues, have been so
obsessed with social engineering that they have forgotten to describe what is going on
in the sites they describe. There is something to this. To stabilize the frame as the one
proper frame is always artificial, perhaps even ideological, for better or worse.15 Latour’s
answer is a self-purification that rids analysis of worlding exercises. Strathern’s answer
is an excessive use of worlding. Latour disallows comparison; Strathern pushes comparison beyond ordinary limits. Both intend to shake us out of our habits of knowing.
To see them as having similar goals with different methods might make it possible to
return again to worlding as something different than holism as we knew it.
The gift of worlding is its ability to make figures appear from the mist and to show
them as no more than figures. Worlding of this sort has potential for both orientation
and disorientation, and the interplay between these states continues to be central to
social analysis.
Acknowledgments
I am grateful to Ton Otto, Nils Bubandt, and the members of the Beyond the Whole conference
for their suggestions and inspiration. Ton Otto suggested I engage with Latour, ANT, and
Reassembling the Social. Sasha Welland (2006) introduced me to worlding. Susan Harding,
Shiho Satsuka, Heather Swanson, and Bregje van Eekelen offered perceptive readings of an
earlier draft.
Notes
1 My use of the term “worlding” draws from literary critics. Gayatri Spivak writes of Orientalist
worldings in which Western thinkers impose their own logic of relationality on an imagined
East: “Now this worlding actually is also a texting, textualizing, a making into art, making
into an object to be understood” (Spivak 1990: 1). This use highlights the awkward relationship between worlding and the worlds being described. (For more literary- and artcritical worlding, see Welland 2006; Connery and Wilson 2007.) There are other uses of
the term “worlding.” In science fiction, worldings are practices for making new worlds, as
in terraforming planets. Donna Haraway (2007) uses the term in this sense to discuss the
entangled trajectories of multispecies life. In yet another use, Martin Heidegger (2001)
uses the term “worlding” to discuss his commitment to “being in the world.” These are
not issues I consider. My focus on gaps and interconnections in cosmopolitan science limits my worldings to experiments in perception and analysis within arts of encounter.
2 In his introduction to Laws of the Market, Callon (1998) provides an analysis that might
get us out of this dilemma. Market contracts, he argues, are framing devices; frames can
9781444333237_4_004.indd 64
8/7/2010 6:43:32 PM
3
4
5
6
7
8
9
10
11
12
13
14
15
65
only be established within and by “overflows.” Scallop science works similarly. Worlding
exercises attend to framing–overflow relations.
The Matsutake Worlds Research Group consists of Tim Choy, Lieba Faier, Michael
Hathaway, Miyako Inoue, and Shiho Satsuka (see Matsutake Worlds Research Group
2009a, 2009b). For research in Oregon, I have collaborated with Hjorleifur Jonsson, Lue
Vang, and David Pheng.
Even with this assumption, many evolutionary trees are possible. The analyst uses various
criteria (e.g., “bootstrapping” and “parsimony”) to figure out which makes the most sense.
Harris (1968) discusses the nineteenth-century comparative method in anthropology and
its authority in other sciences.
French Durkheimian sociology, of the kind that drives Latour crazy, played an important
role in all this, as did the German notion of Kultur.
The many twentieth-century attempts to revive the comparative method, for example as
“controlled comparison,” each try to rid it of colonial worlding while maintaining its imagined scientific rigor. The science is seen to reside in the comparison, with the worlding as
a marginal afterthought for interpretation of the results. Some critics, however, argue that
comparison itself requires a colonial gaze (Stoler 2001).
The best known of these now-unacceptable worldings is Lamarckism, in which organisms
pass environmental adaptations to their offspring.
An exception makes the point clearer. Some elements of DNA are now recognized as
results of interaction between an organism and a foreign virus. These “retro-elements,”
themselves inscriptions of a multispecies ecological interaction, become part of the coding
mechanism of the organism itself. Once inside the DNA, ecology enters the scientific
method of phylogenetics; in contrast, host tree–fungi interactions remain outside. This
example also offers a nonhuman extension of what Latour calls “plug ins,” that is, vehicles
through which individuality is made (2005: 204).
This part of the chapter draws from Tsing and Satsuka (2008). Fuller explanations and
citations can be found there.
Hosford and Ohara tried some Japanese-style matsutake science in the US, as reported in
Hosford et al. (1997). This was probably the only time such science was tried.
In contrast to these site descriptions, Japanese resource management publications tend to
introduce forests through national-scale statistics, while US publications introduce forests
through narratives about ecologically specific species associations.
See Tsing and Satsuka (2008).
See Satsuka’s discussion of this point in Matsutake Worlds Research Group (2009a).
It is a mistake, however, to assume that political commitments always block description.
Many social theorists have argued convincingly that politically engaged frames can open
description beyond taken-for-granted common sense. Marcus and Fischer (1999), for
example, claim that “cultural critique” generates the ethnographic richness of American
anthropology.
References
Boas, Franz (1940). The Limitations of the Comparative Method of Anthropology [1896]. In
Race, Language and Culture. New York: Free Press, 270–80.
Callon, Michel (1998). Introduction: The Embeddedness of Economic Markets in Economics.
In The Laws of the Market, ed. Michel Callon. Oxford: Blackwell, 1–57.
Callon, Michel (2007). Some Elements of a Sociology of Translation: Domestication of the
Scallops and the Fishermen of St. Brieuc Bay [1986]. In Technoscience: The Politics of
Interventions, ed. Kristin Asdal, Brita Brenna, and Ingunn Moser. Oslo: Unipub.
Chapela, Ignacio, and Matteo Garbelotto (2004). Phylogeography and Evolution in Matsutake and
Close Allies Inferred by Analyses of ITS Sequences and AFLPs. Mycologia 96(4): 730–41.
9781444333237_4_004.indd 65
8/7/2010 6:43:32 PM
66
Tsing
Connery, Christopher, and Rob Wilson, eds. (2007). The Worlding Project. Santa Cruz, CA:
New Pacific.
Haraway, Donna (2007). When Species Meet. Minneapolis: University of Minnesota Press.
Harris, Marvin (1968). The Rise of Anthropological Theory. New York: Thomas Crowell.
Heidegger, Martin (2001). The Fundamental Concepts of Metaphysics: World, Finitude, Solitude,
trans. William McNeill and Nicholas Walker. Bloomington: Indiana University Press.
Hosford, David, David Pilz, Randy Molina, and Michael Amaranthus (1997). Ecology and
Management of the Commercially Harvested American Matsutake Mushroom. USDA Forest
Service General Technical Report PNW-412. Washington, DC: USDA Forest Service.
Hubbard, Kathleen (N.d.). Everything You Ever Wanted to Know About Proto-Indo-European
(and the Comparative Method), but Were Afraid to Ask! www.utexas.edu/depts/classics/
documents/PIE.html (accessed January 21, 2010).
Langston, Nancy (1995). Forest Dreams, Forest Nightmares. Seattle: University of Washington
Press.
Latour, Bruno (2005). Reassembling the Social. Oxford: Oxford University Press.
Law, John, and John Hassard, eds. (1999). Actor Network Theory and After. Oxford:
Blackwell.
Marcus, George, and Edward Fischer (1999). Anthropology as Cultural Critique. Chicago:
University of Chicago Press.
Matsutake Worlds Research Group (2009a). Strong Collaboration: Toward Mycorrhizal
Relations. In Multi-Sited Ethnography, ed. Mark-Anthony Falzon. London: Ashgate,
197–214.
Matsutake Worlds Research Group (2009b). A New Form of Collaboration in Anthropology:
Matsutake Worlds. American Ethnologist 36(2): 380–403.
Nakai, Takao (2005). Molecular Analysis of Matsutake in Asia and the World. PowerPoint
presentation, Kyoto Forestry Experimental Station, Japan.
Ogawa, Makoto (1991). Matsutake no Seibutsugaku (The Biology of Matsutake) [1978].
Tokyo: Tsukiji Shokan.
Rabinow, Paul (1977). Reflections on Fieldwork in Morocco. Berkeley: University of California
Press.
Spivak, Gayatri (1990). The Post-Colonial Critic. New York: Routledge.
Steen, Harold (2004). The U.S. Forest Service: A History. Seattle: University of Washington
Press.
Stoler, Ann (2001). Tense and Tender Ties: The Politics of Comparison in North American
History and (Post)Colonial Studies. Journal of American History 88(3): 829–65.
Strathern, Marilyn (1987). An Awkward Relationship: The Case of Feminism and Anthropology.
Signs 12(2): 276–92.
Strathern, Marilyn (1988). The Gender of the Gift. Berkeley: University of California Press.
Strathern, Marilyn (1999). The Ethnographic Effect 1. In Property, Substance, and Effect.
London: Athlone, 1–26.
Takeuchi, Kazuhiko, R. D. Brown, I. Washitani, A. Tsunekawa, and M. Yokohari, eds. (2003).
Satoyama: The Traditional Rural Landscape of Japan. Berlin: Springer.
Tsing, Anna, and Shiho Satsuka (2008). Diverging Understandings of Forest Management in
Matsutake Science. Economic Botany 62(3): 244–53.
Welland, Sasha (2006). Experimental Beijing: Contemporary Art Worlds in China’s Capital.
Ph.D. diss., University of California, Santa Cruz.
9781444333237_4_004.indd 66
8/7/2010 6:43:32 PM
5
The Whole Beyond Holism
Gambling, Divination,
and Ethnography in Cuba
Martin Holbraad
Wholes Reduced, or Holism
Bringing the proverbially divergent “schools” of American, French, and British
anthropology under the single rubric of “holism” would probably be either misguided
or messianic. Were such a use of the term to afford practice of what it preached, it
would somehow have to provide an understanding of each of those traditional “parts”
of the discipline with reference to a “whole” to which they may be imagined to belong.
As we and our professional detractors know, however, no such whole exists. Certainly
“holism” hardly denominates something we have in common as anthropologists. If
anything, it is one of the recipes for what looks like our confusion – the fact, a cause
for pride to some of us, that we are an undisciplined discipline of “parts.” Boas, we
say, was a holist, by which we mean that the elements that make up a “culture” only
make sense in terms of each other (Boas 1940). Malinowski was also a holist: to get
the native’s point of view, the ethnographer must learn to appreciate how the various
bits of his world hang together (Malinowski 1953). But Durkheim, too, was a holist;
any given social fact is to be understood in terms of its contribution to the workings
of society taken as a whole (Durkheim 1982). These diverse insights, often in confluence, have spawned so many more wholes in anthropology that one can only admire
the size of a discipline that – whole or not – can fit them all in!
When faced with a mess, the temptation is either to order or to abandon. The latter
impulse seems ascendant in recent developments in the social sciences – in spheres as
disparate as cognitive science and actor–network theory – that spring from a common
wish to go beyond holism and its tendency to reify its objects (though see George
Marcus’ Chapter 3 in this volume for an alternative genealogy). Hence, in place of
“society” or “culture” we now have “schemata,” “modules,” “assemblages,” “multiplicities,” and so on (e.g., Sperber 1996; Law 2004; Latour 2005). Less radically, this
article raises the possibility that holism might be equipped to go beyond itself, leaving
the morass of its many available versions to one side in favor of a new one. Such an
attempt starts from the premise that existing versions of holism are not holistic enough.
In particular, if the opposite of holism is “reductionism” – which suggests that any
This chapter is dedicated to Panos Giannakakis and to Tata and her family.
9781444333237_4_005.indd 67
8/7/2010 6:44:00 PM
68
Holbraad
given phenomenon is best understood in terms of its constituent parts and their
relationships – how are we then to deal with the paradox that current conceptions of
holism are essentially reductionist in a crucial sense?
The central idea here is that anthropologists’ traditional affection for holism, however articulated, turns out to be directed not at the whole as such, but rather at models
of the whole. So, very roughly speaking, the whole in British holism is traditionally
posited as an observable structure – the nuts and bolts of everyday life that Malinowski
encouraged us to chart, and that Radcliffe-Brown (1961), for example, institutionalized as designs that might “function” (social structures). For the intellectualist French
the whole that is society is indeed empirically observable, but only because it comes
about in collective acts of representation. The whole’s reality is representational and
therefore also socially constructed, since representations (including those of society
itself) emerge socially – a circular idea with which Dumont ran all the way to “hierarchy” (Dumont 1970). Finally, for the Americans the whole, which they call “culture,”
is a corollary of the irreducibly meaningful character of human phenomena (e.g.,
Geertz 1973a). So, as with the French, it is a function of human representation,
though in this case the whole is not posited as a content of such representations, but
rather as the total web of internally connected representations that lends each one of
them its meaning – the hermeneutic whole (another circle, of course).
In each of the three styles of thinking, then, the whole features as a model, either
because social reality is itself model-like or because it is constituted in the model making of representation – or indeed a bit of both, as in Giddens (1986), Bourdieu
(1990), and the ethnomethodologists (e.g., Garfinkel 1967), to take respective heirs
of the three national schools. The whole is in that sense a reduction of the plenitude
of the world – a summary of its most important parts. Holism, then, is the injunction
that such a summary serve as a baseline for understanding any given phenomenon
that falls within its encompassing purview.
The reduced character of holistic wholes seems sensible. Holism, after all, is supposed to be an analytical strategy, and it is hard to see how analysis could proceed
from anything other than some kind of reduction of the world. Surely the world taken
as a totality “in itself” is hardly graspable, let alone a good starting point for that discriminatory process we call analysis. Indeed, the power of holism as an analytic
approach would seem to lie precisely in the fact that while recognizing that analysis is
model-like, it posits that models need to be total in order to have a purchase on a
world that is given by definition as a totality.
The problem is of course familiar as nothing less than the backbone of Western
philosophy, taking off from the Greek concern with the mismatch between One
Reality and Many Appearances (e.g., see Lloyd 2006: 303–17). In modern times the
problem has received perhaps its most systematic expression in Kant, the idea being,
broadly, that our apprehension of the world must be a function of its correlation to
the “categories” we bring to it. As Quentin Meillassoux has argued recently (2008:
1–49), this move toward “critique,” posed against the putative dogmatism of thinking that the world could be apprehended as it is “in itself,” is one that Western thought
has been unable to think itself out of ever since.
Certainly anthropology would be typical in this respect – exemplary even, considering the discipline’s basic concern with, precisely, the variety of ways in which the
world may be apprehended by different peoples. Indeed, while this concern with
variety is so basic to modern anthropology that it cuts across the discipline’s national
traditions, the problem posed by the putative mismatch between the “total” reality of
9781444333237_4_005.indd 68
8/7/2010 6:44:00 PM
Gambling, Divination, and Ethnography in Cuba
69
the world and its “partial” appearances from different sociocultural standpoints has
arguably received its most explicit and creative treatment by the French. In particular,
if, as is often remarked, Durkheim sought to translate Kant’s a priori categories into a
posteriori sociological ones (Durkheim 1995; see Schmaus 1994: 187–93), then LéviStrauss most magnificently transposed the relationship between such categories and
the world whose apprehension they organize as the central problem for humanity
and, by that virtue, for anthropology too, namely, the fabled problem of Culture’s
relationship to Nature (e.g., Lévi-Strauss 1964, 1966; cf. Schrempp 1992). Indeed, if
cultural constructivism has become, quite literally, second Nature to much anthropology today, that is partly because Lévi-Strauss’ Kantian problem of how Nature, taken
as a uniform whole, could be assessed through the partial (albeit holistic) structures
of Cultures was deemed insoluble – the tragic condition of classification, one might
say, thinking back on Lévi-Strauss on myth and ritual (1981).
For better or worse, this chapter offers a solution to the problem. Or it seeks to
articulate one, since part of my point is that anthropologists have de facto been working on the premise that the problem is already solved. That is to say that beyond
anthropologists’ explicit proclamations of “holism” in its various senses lies an implicit
conception of the whole as such, and of anthropologists’ capacity to access it.
To prove the point, I shall start with a bit of whole-accessing of my own, in this case
demonstrated through my fieldwork in Havana among gamblers and diviners. In particular, I look at the contrasting ways in which gamblers and diviners in Cuba engage
in something analogous to what Anna Tsing (Chapter 4, this volume) calls “worlding,” that is, the contrasting ways in which the relationship between “parts” and
“wholes” is articulated in their respective activities (the mereological terminology is
Tsing’s, though I shall gloss the problem in terms of the parallel distinction between
“cosmology” and “cosmos” for reasons that will become apparent).
Wholes Produced: Gambling in Havana
As any statistician will tell you, gambling doesn’t pay. So why do so many people
gamble so much of their money away? The question has seemed like a good one to
academics of all sorts – economists, sociologists, psychologists, and, of course, a
number of anthropologists (e.g., Woodburn 1982; Stewart 1994; Papataxiarchis
1999). As indicated by the emblematic character to the discipline of Geertz’s essay on
the Balinese cockfight (1973b), for anthropologists the question of gambling poses a
favorite challenge: that of providing a sense for practices that seem to lack one.
In fact, in socialist Cuba today, the question seems a particularly good one. Considered
counterrevolutionary, gambling in all forms is heavily persecuted on the island, so that
punters risk not only a dent in their pockets but also legal prosecution, with hefty fines
and even time in prison. Marxist distaste for “speculation” is only part of the story, for
even before Fidel Castro declared the Cuban Revolution socialist in 1961, gambling
already stood for what the Revolution stood against. Casinos in particular became an
emblematic target of revolutionary reform since, along with prostitution, they were
integral to the corrupt image of Cuba as a playground for North American tourists and
Mafiosi (Schwartz 1997). But lotteries, and particularly the National Lottery, which
had existed on the island at least since 1812, also had a tainted image, associated with
the corruption of the so-called pseudo-Republic at the turn of the twentieth century
9781444333237_4_005.indd 69
8/7/2010 6:44:01 PM
70
Holbraad
(Thomas 2001: 306–7). So, following the revocations of casino licenses in the early
1960s, the National Lottery was discontinued as part of a “revolutionary offensive”
later in the decade, and gambling of all kinds has been penalized ever since.
The ban has not been very effective. Although official data on the subject are hard
to come by, there is little doubt that a number of forms of gambling have continued
to be practiced under wraps throughout the revolutionary period. The most widespread of these is an illicit lottery, commonly referred to as la bolita (the little ball).
Indeed, while the disruption caused by the discontinuation of the National Lottery
should not be underestimated, and for many people the very word “lottery” (lotería)
has prerevolutionary connotations, punters’ own narratives today often speak of resilient practice in adverse circumstances. Such stories of continuity are historically accurate, inasmuch as unofficial lotteries like la bolita existed long before the Revolution,
even featuring in nineteenth-century travel accounts (e.g., Howe 1969: 102–6; see
also Ortíz 1986: 206–8). Like today, these games were organized privately at a neighborhood level, and often piggybacked on the weekly draw of the National Lottery by
using its six-digit winning number for the sake of transparency.
From the punters’ point of view, one of the main advantages that these side games
had over the National Lottery was that they divided its six winning digits into three
separate numbers, effectively taking bets on three separate two-digit numbers and their
combinations. Thus, for example, if the National Lottery had drawn the number
“123456,” the illicit boliteros, as they were called, would pay bets on “12” (referred to
as fijo, meaning “fixed”), as well as on “34” and “56” (referred to as primer corrido
and segundo corrido, respectively, meaning “first runner-up” and “second runner-up”).
This allowed for a much higher frequency of winning bets, paid at correspondingly
lower prices, as well as a greater degree of flexibility in the kinds of bets that could be
constructed (see Table 5.1). Most importantly for our purposes, cutting the winning
Table 5.1
Bolita bets
Type of Bet
Betting Structure
Fijo
On first number:
xx xx xx
On second or third
number: xx xx xx or
xx xx xx
On last three digits:
xx xx xx
Any two of three
numbers in either
order: xx xx xx,
xx xx xx, or
xx xx xx, (where,
e.g., bets on 12 and
34 win both on 12
34 56 and on 34 12
56)
Corrido
Centena
Parlé
Return per
Peso
Probability
(%)
Bank’s
Profit
Margin (%)
75 (or 70)
1
25 (or 30)
25 (or 20)
2
50 (or 60)
300
0.1
70
900 (or 800)
0.06
46 (or 52)
Source: Data based on fieldwork in Havana in 2004.
9781444333237_4_005.indd 70
8/7/2010 6:44:01 PM
71
numbers down to two-digit size allowed punters to place their bets on the basis of a
code of symbols associated with each number from 01 to 99, turning the lottery into
the guessing game that punters call la charada (the charade), as explored below.
The preference for piggybacking on an established draw rather than organizing
draws locally was, if anything, intensified after the National Lottery was banned by
the Revolution. As punters today explain, the need for transparency became all the
more urgent since the authorities’ persecution made it impossible to hold locally
organized draws in public. Thus, since the 1970s, boliteros have run betting lists (listas) at considerable personal risk by using the draws of foreign lotteries, which can be
heard on the island in the evenings on shortwave radio. In Havana for a long time
the two most popular draws were taken from Venezuelan radio, though currently
most boliteros are using the daily draw of Radio Martí, a Miami-based Cuban American
station devoted to transmitting antirevolutionary propaganda to the island – which
of course only serves to confirm the counterrevolutionary associations of the game in
the eyes of the authorities.
Apart from adding persecution as an extra dimension of risk to bolita gambling, the
revolutionary ban has raised the administrative costs of the outfits that run the game.
Operating clandestinely, these so-called bancos (banks) use Mafia-like security arrangements that take the form of a pyramid of four or more tiers of personnel, all of whom
need to be paid a proportion of the profits, calculated on the basis of the risk involved.
Street-level collectors (recogedores) collect punters’ tickets and bets until about 7
o’clock in the evening. They then distribute the bets to one or more bookkeepers
(apuntadores), each of whom may hold a number of “lists” pertaining to different
banks. After the results of the draw are declared on the radio, later in the evening, the
bookies meet with agents of the banks to settle the accounts of the day for each list. By
this point, the money collected may run into thousands of Cuban pesos. The agents
then deliver profits to and/or draw payments from the owners of the bank themselves,
or meet with a further secret intermediary, and so forth. The eventual destination of
the money is the subject of urban myth-like speculation: “they bury it in a box under
the ground in the countryside,” one recogedora from Havana told me.
The costs of these complex secret arrangements are passed on to the punters in the
form of very expensive bets. As shown in Table 5.1, the bank’s gross margin on the
various types of bet on offer ranges from a minimum of 25 percent (or, in the case of
the least generous banks, 30 percent) to a maximum of 70 percent. From the bank’s
point of view, the fact that profit margins tend to increase as the odds of the bets get
longer is due partly to the high risk of bankruptcy for these relatively low-liquidity
outfits. The bancos’ profits are of course the punters’ losses, so my ethnographic question to the latter was “Why do you do it?”
But this is not a very good question – something I realized after the umpteenth
stilted conversation on the topic with my gambling friends, which is to say most of my
friends, since in the inner-city areas where I work most people gamble on la bolita: the
young and the old, housewives and their husbands, black, white, and even members
of the Communist Party. Sure, people would dutifully discuss with me all manner of
plausible hypotheses that I would come up with – the entertainment value of the
game, the hope of cheap cash when money is as hard to come by as it is in Cuba today,
and so forth. But in each of our conversations, there was a strong sense – familiar to
the ethnographer – that all this was somehow beside the point. Here is what I was
told, by way of getting to the point, by a friend of mine in his mid-thirties who works
at a petrol refinery. “Like everyone else,” he says, he likes to play on la bolita
occasionally:
9781444333237_4_005.indd 71
8/7/2010 6:44:01 PM
72
Holbraad
If you read Fidel’s [Castro’s] History Will Absolve Me you see why the government never
liked the gambling in this country … although if for whatever reason they authorise it
again in the future they’ll probably take an even bigger share than the banks do today!…
But the game will never disappear, you know why? Because we Cubans like to play.
Cubans like three things: drinking, gambling and women. It’s in the idiosyncrasy of the
Cuban, he likes this way of winning, it’s easy! He’s walking on the street and sees a car
with a 27 on the number-plate and he thinks, “this is giving me a cábala [sic], let me play
27,” and then he puts 50 pesos on 27 and loses. Or maybe he wins! And then tomorrow
it’s something else, and so forth. Cubans are like that.
My friend’s folk-psychological appeal to the happy-go-lucky aesthetic of “the
Cuban” is a standard feature of these conversations. But just as typical, and arguably
more telling, is the apparent tautology of his account. The answer as to why people
gamble, in spite of state prohibition and poor prizes, is that they like it. Indeed, the
essentializing overtones here (“the idiosyncrasy of the Cuban,” etc.) only serve to
emphasize the cyclical character of the reasoning: “we gamble because we gamble –
it’s what we do” (or, “it’s what we like” – the ultimate conversation stopper). That
this tautology is the brunt of my friend’s comment is clear by the role of his illustration. The hypothetical story about the man and his number-plate cábala (and more
on this notion below) does little to explain Cubans’ taste for the game. Rather, what
it does is illustrate the form that a taste for the game may take. The implication is that
the game’s attraction is its own explanation: the game explains itself. Indeed, in the
context of this tautology, even the apparently crucial matter of winning or losing is
treated as a contingent externality, almost beside the point. A love of “easy winning”
is stated, only to be swiftly swept aside: whether he loses or wins, “the Cuban” will be
playing again tomorrow.
Taking my friend’s lead, we may attend to the form bolita playing takes. The scenario of the man and his number-plate cábala describes the rudiments of the game as
a search for motivations for playing. At stake here, however, is not a search for a
motive for playing in general (which was my anthropological concern), but rather a
search for motivating a bet on a particular number. This is what the game is about, as
far as the players are concerned: to play you must have a motive to play on a number,
and this is “given” to you, often in the form of a so-called cábala – a term used to refer
generically to everyday events that are in one way or another attention-grabbing
enough to warrant a bet. Now, playing on number-plates or on the figure of a certain
number seen in the mud while taking a walk, as another informant described, and
indeed betting on one’s birthday or some other significant date are perhaps the most
straightforward instances of what people mean when they speak of cábala. Just as
typical, however, are cábalas that are generated by a more arcane system of number
symbolism, to which players refer either as just la charada, or sometimes as la charada
china (the Chinese charade). The latter designation suggests at least a notional connection between the number symbolisms people use in Cuba today and numerological systems promulgated by Chinese immigrants who began to arrive on the island in
the latter half of the nineteenth century (see Caillois 2001: 148–52). But significantly,
as we shall see, when I quizzed punters (including Chinese descendants) for historical
details on this connection I was unable to get much further than the fact that the
present form of la charada works on the same principle as the numerological associations of numbers with symbols that the Chinese introduced. The main difference, I
was told, is that while the Chinese “system” (as informants called it) used only numbers from 1 to 36, the creolized version was expanded to include numbers from 00 to
9781444333237_4_005.indd 72
8/7/2010 6:44:01 PM
73
99 to suit la bolita. Beyond this, no one had much to say either on the provenance of
the number symbolism of la charada, or on the manner of its expansion from 36 to
100 numbers. Providing reasons on this score is not a matter of interest.
What is of great interest, however, is the series of charada symbols associated with
the numbers. As a retired factory worker told me, after I had thanked him for a long
interview on his favorite cábala stories (more of which below), “What you still haven’t
learned is the most important part of the game, which is the meanings that each
number has.” He continued,
It’s complicated, but without it you can’t play. Horse is 1, butterfly is 2, child is 3, etc.
But lion is also 1, so you can play it like that too.… I learned these things when I was still
a child, one gets used to it, because everyone in my family used to play la bolita.… People
know this system and in this way they play, from dreams, or what happened to them on
the street, what someone told them, cábalas.
Table 5.2 shows the entire symbolic series, from 01 to 00, as it was given to me by
a housewife in her mid-thirties. Having listed the symbols on the trot “from memory,” as she said, my informant was keen to emphasize that each number has many
more meanings (significados) than she could remember. She added that although
most numbers had some “classic” associations that everyone knows (e.g., 15 = dog,
and 50 = policeman), had I asked someone else for the whole series I would surely get
a different list. Indeed, over the years informants have given me a plethora of overlapping lists, so for some numbers I have compiled more than 10 symbolic associations.
I shall return to the question of apparent redundancy later.
Whatever the grade of its elaboration, however, charada symbolism serves to amplify
exponentially the field from which cábala motivations can be drawn. To the rather
literal story of the number-plate cábala, I could add dozens of stories told to me by
players in the figurative modality of la charada. Indeed, these are the stories players are
most interested in telling, not just for my anthropological benefit but also, more
generally, whenever la bolita is brought up for discussion. For example, this is one of the
stories the retired factory worker told me, before urging me to “learn the numbers”:
I was getting bread one day and in the queue I met an old comrade of mine who started
telling me how he only had a year left to work before retirement. I told him, “Don’t
worry, a year goes quickly enough.” Just as I was saying it, this man who had had a few
drinks comes up and gives his hand to my friend to greet him. My friend greeted him
alright, but when the man left he said, “Do you know this guy? I don’t!” I told him,
“Look, this is a sign (una señal), you should play 49 today” [49 indicating “drunkard” – see
Table 5.2]. I had already played something else that morning, but when I got home
I put two and one [i.e., 2 pesos on fijo and 1 on corrido] on 49 as well. Next day I’m
waiting for the bus and a comrade says to me, “You see what happened yesterday?” –
I had forgotten about it. “They drew double 49 with 90!” [i.e., 49 fijo, 49 primer
corrido, and 90 segundo corrido]. I made 200 pesos. But I should have played old man
too, I would have got the parlé! [“Old man” should have been suggested by the earlier
conversation about retirement, indicating the number 90.]
The having and telling of such experiences are clearly part of the fun of the game, and
comprise the fabric of which players’ bolita talk is made. As Huon Wardle has put it,
with reference to a similar Jamaican lottery called “drop pan,” the symbolic associations of la charada furnish the imagery of a fleeting poetry of urban experience
9781444333237_4_005.indd 73
8/7/2010 6:44:01 PM
74
Holbraad
Table 5.2
Number symbolisms of la charada
1 Small horse,
lion
2 Butterfly
3 Sailor, child
4 Cat, Changó
26
Doctor, eagle
51
Soldier
76
Female dancer
27
28
29
Wasp
Goat
Mouse
52
53
54
Bicycle
Electric light
Flowers
77
78
79
5 Nun, Ochún
6 Tortoise
7 Shit, Yemayá,
shell
8 Small dead
person
9 Elephant, father
10 President, big
fish
11 Rooster
12 Whore
30
31
32
Shrimp
Deer
Pig
55
56
57
Crab
Meringue, queen
Bed
80
81
82
Crutches
Grave
Carriage,
hanged man
Doctor
Boat, theater
Mother
33
Hawk
58
34
35
Monkey
Spider
59
60
Knife, adultery,
83
picture
Madman
84
Egg, cannon shot 85
Tragedy/
problems
Blood
Clock
36
37
61
62
Big horse
Marriage
86
87
13 Pimp
38
Pipe
Witchcraft, black
hen
Money
63
Assassin
88
14 Cemetery
39
Rabbit
64
89
15 Dog
16 Horn
17 Moon, San
Lázaro
18 Small fish
19 Worm
20 Bedpan
40
41
42
Priest
Lizard
Duck
65
66
67
Large dead
person
Food, prison
Divorce, mare
Stabbing
Hose
Banana/
plantain
Living dead,
spectacles
Water, lottery
90
91
92
Old person
Beggar, tram
Airplane
43
44
45
68
69
70
Big cemetery
Fun/commotion
Coconut
93
94
95
Revolution
Machete
Big dog
21
22
23
24
25
46
47
48
49
50
Scorpion
Coconut shell
Gun, president,
shark
Bus
Bird
Cockroach
Drunk
Policeman
71
72
73
74
75
River
Transport train
Suitcase
Kite
Tie
96
97
98
99
00
Old shoe
Mosquito
Piano
Saw
Car
Snake
Frog
Steamboat
Pigeon
Precious stone
Source: Collected from a single informant, 1999.
(Wardle 2005). As he writes, “The archetypal figures of drop pan reappear in day to
day narratives: a man poisoned while attempting to kill a rat, a dog hit by a car, the
appearance of a local madman are all grist to the drop pan mill” (Wardle 2005: 89).
The same could be said of la bolita.
Here, however, we may focus on how bolita players themselves view the relationship between charada imagery and the day-to-day experiences on which it seems to
feed. In particular, I want to consider in some detail an analogy suggested by players’
own accounts of the role of the number symbols, namely, between la charada and
divination – that other favorite, and much better documented, activity in Cuba (e.g.,
Brown 2003). After all, when my elderly informant told his comrade that the drunken
stranger’s greeting was a “sign” that he should bet on 49, or when he later kicked
9781444333237_4_005.indd 74
8/7/2010 6:44:01 PM
75
himself for missing the significance of his comrade’s comments on aging (90 for “old
man”), was he not exemplifying an essentially divinatory way of thinking? The salience in this context of the notion of cábala seems to corroborate such a view. While
my impression is that the mystical connotations of the word are lost on most of my
informants,1 the term does have more prosaic associations relating to the ordinary
Castellan usage hacer cábalas, which literally means “to guess.”
Indeed, divinatory symbolism forms part of the symbolic series of la charada itself.
As can be seen in Table 5.2, certain numbers are associated with particular deities
from the West African pantheon that forms the basis of the divinatory practices associated with the Afro-Cuban cults of Santería and Ifá: 4 for Changó, the virile patron of
thunder, and for Orula, the patron deity of Ifá divination; 5 for Ochún, the goddess
of rivers and love; 7 for Yemayá, the maternal goddess of the sea; and so on. Consider,
for example, the favorite cábala story of my friend who works at the petrol refinery:
Once I woke up feeling desperate – I didn’t have money at the time and had all sorts of
problems. I went to my San Lázaro [referring to a statue of the saint he keeps in his living-room] and spoke to him. “Please San Lázaro, give me some kind of sign – help me
out!” Then I went to my boveda [a shrine devoted to dead spirits that is kept in the
home] and lit a candle. After a while – I was in the kitchen – the candle fell over and
started a fire. “My God,” I said, “this must be the sign!” … So I went and put all the
money I had on 6 [fire], on 76 [hose] … on everything that has to do with fire. Next day
my mother calls and tells me that 17 had won, which is San Lázaro! So I lost everything
… because I played on my cábala, whereas San Lázaro was telling me to play on him.
That’s what desperation is like [laughing].
La charada, one could surmise, is analogous to divination in that it furnishes a system of “signs” which, when generated in an appropriately nondeliberate manner, yield
predictions of otherwise unforeseeable events. The candle’s falling over is taken as a
“sign” that suggests a particular number (6, 76, or, as it turns out correctly, 17), much
like in Santería divination a particular configuration of cowry shells (or, in Ifá, palm
nuts) cast by the diviner in an uncontrolled manner during the consultation is taken as
a “sign” (signo) that has particular predictive associations (such-and-such a divinatory
signo portends death in the family, another heralds the cure of a particular ailment,
etc.). One could perhaps go as far as to compare my friend’s failure to work out the
correct significance of San Lázaro’s sign to novice or unscrupulous diviners’ “errors”
or “lies,” for which savvy clients are always on the lookout (see Holbraad 2008).
Nevertheless, there is a crucial breakdown in the analogy between gambling and
divination that pertains to the role of “reasons,” the question with which we started.2
When asked to explain the predictive power of their oracle, diviners commonly point
to its ability to provide “the reason for things” (el porque de las cosas). The following
extract from an interview with Lázaro Panfét, a young initiate of Ifá whom I probed
about his enthusiasm for divination, is typical:
Anything you need to know is in Ifá. [Q: In the signos of Ifá?] Yes, in the “paths” of Ifá
[viz., the myths associated with each signo; see below]; the birth of the world, of people,
of the deities, wars, progress. You speak to a babalawo [i.e., Ifá diviner] who knows, and
he’ll be able to locate even the most insignificant detail. Even the Internet is in one of the
signos.… That’s why Orula [the patron deity of Ifá divination] tells you what’s happening
to you, your past, present and future; because everything is there, in the signo that he
gives you [in the divinatory consultation].
9781444333237_4_005.indd 75
8/7/2010 6:44:01 PM
76
Holbraad
Elsewhere, I have discussed the peculiar logic by which diviners bring the mythical
knowledge of the signos to bear upon the everyday concerns of their clients (Holbraad
in press). Here we may note only that this logic is explicitly etiological. Divinatory
prediction is premised on the notion that all worldly affairs (including everyday matters predicted for the benefit of clients) have a “reason” that can be found “in” the
signos. In this sense the signos anchor what one might think of as a “total” cosmology,
described in the myths, that is deemed to explain “everything you want to know,” as
Lázaro Panfét put it. For divinatory purposes the cosmology of the signos – an object
of fascination and study for initiates and anthropologists alike – constitutes a world of
“the reasons for things”: a “logocosmy,” if you will.
Gambling could not be more different. As we have already seen, bolita players are
characteristically short on explanation as to the provenance and significance of charada
number symbolisms, or indeed as to why they play the game at all. Astonishingly to
the ethnographer prepped in diviners’ awe, this indifference extends to what could
reasonably be assumed to be the major premise of bolita number guessing, namely,
that the occurrences that precipitate players’ guesses are somehow connected to the
results they purportedly predict. Even when discussing examples that most resemble
familiar forms of divination (e.g., how a dead relative might “give” a number in a
dream), players would invariably treat my requests for exegesis in speculative tone, as
if the matter was ultimately beside the point. The aforementioned retired factory
worker gave the most direct response. When I asked him to explain how a seemingly
irrelevant occurrence like a drunken stranger’s handshake could serve to predict the
evening’s lottery draw, his reply was nonchalant:
These things have no explanation. They are just coincidences. I bet on my cábalas, but if
you start counting you’ll find that most of the times I lose.
Provided we credit Evans-Pritchard’s famous point that in divination coincidences
are characteristically loaded with significance (1976; and see Holbraad in press), the
contrast between divination and gambling is stark. As their disinterest in exegesis
would indicate, bolita players are happy to treat cábalas as insignificant happenstance,
just as Evans-Pritchard assumed his European readers would treat the occurrences
involved in Zande divination (e.g., 1976: 152). The only difference – equally surprising from the viewpoint of divination and of European “common sense” – is that players should nevertheless base their guesses on such coincidences. In other words,
players bet (and too often lose) good money for what they appear to be admitting is,
quite literally, no good reason.
The characteristics of the “a-rationality” of bolita gambling (which, as we shall see,
can be distinguished sharply from irrationality) become clearer if one pursues the comparison with divination, concentrating on the formal properties of the two practices.
Formally speaking, I want to argue, gambling can be seen as an inversion of divination.
If divination draws on a “logocosmy” to offer reasoned predictions, gambling draws on
the process of prediction itself, to offer, if not “reasons” as such, then the raw cosmological elements on which reasoning can be built. Put differently, while divination
involves the application of a cosmology, gambling pertains to its original inception – a
theme to which I shall return.
As we have seen, gambling and divination are both “predictive” inasmuch as they
are geared toward gauging as yet unforeseen events in terms of events that are already
given – given, therefore, as somehow “significant.” In gambling, the unforeseen events
9781444333237_4_005.indd 76
8/7/2010 6:44:01 PM
77
SIGNOS
(1/256)
IFA
MYTHS
WORLD
WORLD
CHARADA
SYMBOLS
BOLITA
DRAW
(1/100)
DIVINATION
GAMBLING
Figure 5.1 The inversion of prediction in divination and gambling
in question are very specific, namely, the nightly bolita draw of a particular number
from 1 to 100. The given events that serve to predict the draw, on the other hand, can
be just about anything – any cábala that tickles the punter’s fancy. In divination, however, this situation is reversed. Here it is the predicted events that can be anything,
while the given events that ground the prediction are the specific ones, namely, the
diviner’s draw, with the help of cowry shells or palm nuts, of a particular divinatory
signo, from 1 to 256. So, simply put, in gambling anything can be used to guess something, while in divination something is used to guess anything. In both cases the
“something” is a technically induced outcome derived from a set of numbered possibilities, while the “anything” is simply the unfolding of events that comprise life.
As can be seen in Figure 5.1, in both cases the movement from given events to as
yet unforeseen ones – the process of prediction – relies on a symbolic matrix that facilitates a translation. In la charada the matrix consists of the various symbolisms associated with each of the 100 numbers (as in Table 5.2). The matrix, which is relatively
“open,” involving a large degree of redundancy and divergence from player to player,
is used to decipher all manner of everyday occurrences to make predictions within a
“closed” set of 100 numbers. Added to the fact that players may often go beyond the
frame of la charada itself for their guesses (finding cábalas in number-plates, personally
significant dates, or what have you), it is precisely the flexibility of the symbolic matrix
of la charada that allows “anything” to be translated into “something.” Matrix-like yet
“open,” the series of number symbolisms of la charada mediates between life and the
100 numbers of la bolita, rendering the two commensurate, allowing the imponderabilia of everyday life to have a purchase on the guessing game.
The same holds for divination in Santería and Ifá, only the other way around. Here
the role of matrix is taken by the myths with which each of the 256 divinatory signos
is associated – the “paths of the signo,” as Lázaro Panfét explained. Diviners draw on
their knowledge of these myths to “interpret,” as they say, the significance of the signo
cast, so as to arrive at an appropriate prediction,3 taking into account the particular
circumstances of the client, since the mythical associations of each signo are so complex
9781444333237_4_005.indd 77
8/7/2010 6:44:01 PM
78
Holbraad
that a host of alternative interpretations is typically possible (see Holbraad in press).
For example, in Ifá the signo Otura Tikú (one out of 256 possible configurations) is,
among many other things, associated with death, impotence, intellectual prowess,
and ascent in matters of Ifá worship. Depending on the circumstances, diviners may
deem it appropriate to focus on any one of these associations when advising a client
about their prospects. So, once again, it is the relatively “open” character of each
signo’s mythical associations that allows for translation. Only here the mythical matrix
serves to open up the closed set of signos, so that they may have a purchase on the
totality of events that comprise everyday life – events that would otherwise remain
imponderable indeed.
My argument regarding the “a-rationality” of gambling turns on the idea that this
inversion goes deeper than it may initially appear, straight to the question of etiology.
True, a superficial reading of Figure 5.1 would posit an analogy between the etiological rationale of divinatory prediction (from mythical “reasons” to everyday events)
and the process of number guessing in la charada (from everyday events to numbers).
As we have seen, however, gamblers resist this. The cábalas that la charada translates
into numbers are not presented as “reasons” for the draws they purport to identify –
these are “mere coincidences,” as the old man said. But this, I would argue, follows
directly from the inversion we have been examining. Diviners’ ability to gauge the
“reasons for things,” we saw, relies on their having at their disposal a cosmology,
described in the myths they spend a lifetime studying. “Anything you need to know
is in Ifá,” Lázaro Panfét said, meaning that all events, including the most everyday,
can be inferred from the stories of the myths through a process of interpretation. But
without even going into the detail of what such a process might entail (see Holbraad
in press), it is clear that interpretation is only possible – indeed necessary – because the
mythical cosmology contains “everything” only in reduced form: it maps the territory
of life in the manner Eliade called “archetypal” (1991). Indeed, it is in this sense that
diviners consider that myths contain the “reasons” for things, as opposed to the things
themselves: divination can provide a logocosmy only because it provides a cosmology –
a logic of the cosmos.
But gambling, as we have seen, inverts just this picture. If anything is supposed to
provide the reasons for gamblers’ number guessing, it is the cosmos itself. For the
cábalas that motivate the guesses are not drawn from an abstracted cosmology, but
from the minutiae of life in its concrete unfolding. To treat such motivations as “reasons” is a non sequitur, precisely because the very idea of a reason implies a process of
logical manipulation, if you like, of the world – the kind of principled reduction cosmologies provide. In other words, if gambling is geared toward guessing “something”
in terms of “anything,” then its predictions cannot be based on reasoning, because
reasons necessarily discriminate: “anything,” or the cosmos, is too indiscriminate –
too big – to be a reason.
Now, this key point about the a-rational (or just a-proportional) magnitude of the
cosmos, as opposed to its cosmological reduction, may sound like sleight of hand.
The counterargument might go like this. Surely punters’ cábalas do serve precisely to
discriminate particular events from the total ebb and flow of everyday life, rendering
life proportionate to the events that cábalas purport to predict – that is, the result of
the draw. “Anything,” after all, only describes the domain from which cábalas can be
drawn, as appropriately little “somethings.” Given this, cábalas could certainly be
construed as “reasons” in some pertinent sense, by analogy to the role that events play
in inductive reasoning, typical of all manner of empiricism. Indeed, in such a view,
9781444333237_4_005.indd 78
8/7/2010 6:44:01 PM
79
gamblers’ proclivity to dismiss their cábalas as mere coincidences may well amount to
a recognition of a failure of inductive reasoning, misused in this context: the futility of
a game that gets your (inductive) hopes up only to dash them in a wave of randomness
(cf. Malaby 2003).
But this seemingly hardnosed empiricism is as wrongheaded as it is patronizing of
gamblers’ supposed “fallacies.” Not only does it ignore gamblers’ indifference to the
charms of reason, but also, more significantly, it fails to recognize the peculiar form
that this indifference takes. “Mere coincidence,” the old man said, sweeping the
expectation of reason aside (and therefore also the accusation of unreason), only to
make space for a conception of his own. Coincidence captures it: what bolita players’
multifarious appeals to cábalas share is the insistence, and great care in practice, that
these events be given over to life – they must, as it were, “happen.” As my friend’s
story about San Lázaro and the fire aptly illustrates, cábalas are not events one sets out
to look for, but events one merely finds. Only because cábalas distinguish themselves
from the run of everyday life are they deemed significant enough to warrant a bet.
An interjection on the part of the gambler in this process of events’ self-revelation, in
an attempt to “tame chance” as Ian Hacking described the rise of statistical reasoning
(1990), effectively nullifies the cábala. So inasmuch as reason must involve discrimination – singling out something from the totality of everything – then coincidence is
a-rational inasmuch as it discriminates without discrimination, singling out something –
sure – but only qua anything. In mock Kantian, one might say that in coincidence the
cosmos is revealed in itself, from behind the matrix of cosmology. If, as Galileo said,
mathematics is the language the cosmos speaks to us, then coincidence is the language
it speaks to itself.
So this is why the number symbolisms of la charada are anything but a cosmology. In divination, as we saw, myths serve to provide reasons for things by apportioning the imponderabilia of everyday life to the systemic matrix of 256 signos.
Things have “reasons” insofar as they can be organized, interpretatively by the
diviner, in the cosmological scheme of the myths. It follows that by inverting this
picture, as in Figure 5.1, the translations performed by means of the matrix of
charada symbolisms amount to a disorganization, and are by that token a-rational.
Far from serving to predict the lottery’s draw by connecting it to a specific cosmological element, charada symbolisms help gamblers to guess the draw by effectively
disconnecting it from any particular event that may serve to explain it, rendering the
whole cosmos – “anything” as such – relevant to it. Far beyond cosmology, the
matrix of la charada facilitates translations indeed, though not for the benefit of
reasonable men, but for the cosmos itself as well as the gamblers who strive to get
caught up in it.
In view of this a-rational image, the peculiar underelaboration of charada symbolism, discussed earlier, makes sense. If the role of the matrix here is to translate strictly
coincidental events into numbers, then any cosmological accounts regarding the
provenance or import of, or relationship between, the symbols involved could only
compromise the coincidental, “anything”-like character of the events they tag. For
the cosmos to appear, as we saw, cosmology must be rescinded, and that is what these
symbols do.
Nevertheless, as the stories of players kicking themselves for having bet on the
wrong cues would indicate, gamblers’ efforts to insinuate themselves into this process
of cosmic revelation are largely in vain: “if you start counting you’ll find that most of
the times I lose,” the man laughed. This is the comic counterpart to Lévi-Strauss’
9781444333237_4_005.indd 79
8/7/2010 6:44:01 PM
80
Holbraad
The world of
particulars
Cosmology
DIVINATION
GAMBLING
The Cosmos
Cábala
Cosmological
Cosmology
(coincidence
element
apprehended) (comprehension)
Figure 5.2 Cosmological comprehension and cosmic apprehension in divination and
gambling
tragic view of classification – the joke being that coincidences can only be compromised by the gambler’s inevitable crunch-point decision to bet on one instance rather
than another (comedy as self-conscious tragedy, so to speak). For such choices pertain
not to cosmic revelation but to cosmological judgment. I am sure my friends would
love it if they could render coincidental their decisions to bet on cábala coincidences –
I imagine such a scenario as some kind of zombie-like divestment of choice, perhaps
a form of spirit possession, although I admit I have not discussed this with players in
Cuba. As it is, they are left with the predicament of having to efface the anything-like
character of the cábalas they apprehend, by comprehending them as betting “options”–
proto-cosmological “somethings,” if you will, that can be manipulated as reasons
(betting on 49 “because” one met a drunk, and so on). Gambling, then, could be said
to lay bare the origins of cosmology. Coincidences, as the language of the cosmos,
provide the raw materials out of which cosmologies may be built. Indeed, while
I would hardly want to tell a faux Tylorean story about how coincidences originated
savage cosmologies (precisely because Tylor saw awe-inspiring coincidences as objects
of explanation), gambling does serve to make the logical point that the role of coincidence with respect to cosmology (or, plainly, thought) is, if you like, originary (cf.
Tylor 1970).
So, in summary, the comparison between divination and gambling in Cuba yields
two contrasting images of the relationship of cosmology to the cosmos, represented
diagrammatically in Figure 5.2. In divination, the cosmos is – literally – “comprehended” by cosmology, and thus everything in it can – in principle – be explained. On
this image, however, the cosmos in its total plenitude remains opaque – the age-old
philosophical problem. Worldly particulars are posited as explananda only insofar as
they can be brought into proportion with the cosmologies that furnish their explanation, and this presupposes that they are already extracted from the world as “particulars.” In gambling, on the other hand, the picture is inverted. The gambler’s question
9781444333237_4_005.indd 80
8/7/2010 6:44:01 PM
81
is not how to explain things by cosmological recourse to reasons, but rather how the
elements out of which such “reasons” are made might be generated in the first place –
how particulars might emerge from the cosmos. So if, for the diviner, cosmology
conditions the cosmos, for the gambler it is the other way around: the cosmos provides the ground for cosmology by giving itself over as such at moments of coincidence. As coincidences, worldly particulars can recommend themselves, as it were, to
apprehension. And as the gambler’s comic predicament of loss demonstrates, these
cosmic self-recommendations are distinctive precisely in that they do not constitute
reasons – cosmic revelation is demonstrably not about etiology.
Anthropology’s Gambles
By way of conclusion, I want to argue that the contrast between divinatory reason and
gambling a-reason (etiology vs. revelation, so to speak) can elucidate a contrast
between traditional understandings of holism in anthropology, and an alternative one.
It will be clear that divination is holistic in the traditional sense, since myths play the
same role in diviners’ reasoning as “wholes” do in anthropologists’. Much like structure, society, culture, the imaginary, habitus, or what have you, the mythical totality
of the signos provides the larger backdrop against which particular phenomena make
sense. But as with anthropological wholes, this mythic totality is also reductive. It
summarizes all worldly particulars by encompassing them in rudimentary form, so as
to make it possible for diviners to find their reasons. For anthropology and for Ifá,
then, holism is a concomitant of etiology.
Now, the problem with holism is not that it is necessarily a bad way of doing etiology (which is perhaps the usual quip), but rather that etiology itself is not as appropriate an anthropological agenda as is usually assumed. I shall not make this argument
here, however, other than to state the problem.4 If one accepts that anthropologists’
most typical challenge is making sense of practices that appear to lack one (making
sense of “the Other,” as people used to say), then it is unclear how etiology might be
fit for the purpose. Unlike the diviner, who may rest assured that his mythical etiologies are god-given, the anthropologist is left wondering: if making sense of the Other
is a matter of explaining it, then whence might the terms in which such explanations
are cast (holists’ wholes, for example) be derived? If they too are Other (as so-called
relativists would have it), then a debilitating regress of Otherness ensues (the hermeneutic circle). And if they are not Other (the universalistic option), then the object of
explanation is merely shown up not to be as Other as it appeared – that is, its alterity is
just denied, explained away.
Gambling, I think, offers a way of articulating anthropologists’ concern with alterity, suggesting an understanding of what “making sense” amounts to, that goes
beyond etiology. Gamblers’ cábalas, we saw, turn on the idea that it is possible to
apprehend the cosmos as such. Its cosmological reductions are bypassed insofar as the
gambler allows coincidences to suggest themselves to him. But this is arguably a fair
description of the anthropologist’s stance to alterity – or, better, the stance of alterity
toward the anthropologist. Alterity, we say, comes down to the fact that certain practices seem nonsensical to the analyst – gambling is an example, particularly in Cuba as
we saw. Faced with such practices, the common reaction is to search for a reason that
might explain them. I did that in my original fieldwork, as described earlier. Searching
9781444333237_4_005.indd 81
8/7/2010 6:44:01 PM
82
Holbraad
for reasons, however, presupposes that one already has a grip on the perplexing practices, as particulars that require explanation. In other words, it presupposes that these
practices already feature in one’s analytical cosmology – for example, we know what
gambling is, so the question is why Cubans do it. So if the problem is that this approach
effaces alterity, as already shown, then the solution must surely be to bypass our cosmological presuppositions, like gamblers do.
Now, there is an apparently banal sense in which this agenda is already built into the
practice of anthropology, namely, in the method of ethnographic fieldwork. The habitual
justification of fieldwork is roughly that long-term engagement with people allows us to
get a deeper insight into their point of view. But this is arguably a poor justification, since
such appeals to experiential osmosis – verstehen in German – hardly crack the logical
conundrum of alterity (the dilemma of infinite regress vs. obliteration outlined above). A
better justification would advertise the role of coincidence. If there is one thing we know
about fieldwork (though we feel we should keep it a secret), it is that it is the antimethod
par excellence. We go to the field only to expect the unexpected, which is to say the coincidental. In that sense, fieldwork is the exact opposite of lab work: an experiment out of
control, fieldwork is by nature oriented not toward planned eventualities but rather
toward arbitrary coincidences. Happiest when things happen to themselves, the ethnographer is professionally vulnerable to the unexpected. Indeed, arguably the value of pretending that fieldwork could ever be systematic (the writing of proposals, research
itineraries, and even the key – but typically arbitrary – decision as to where to do fieldwork in the first place) is that such pretenses construct a backdrop against which the
coincidental character of what finds us in the field can appear as such. Viewed in this light,
the ethnographer’s investment in fieldwork as a conduit for the unexpected is far from
banal. Rather, fieldwork becomes the experiential correlate, so to speak, of alterity itself:
it is precisely because of its repercussions on the intellectual problem of alterity that the
practical “happenstance”5 of ethnographic fieldwork is anthropologically compelling.
Now, if you take my core argument about gambling – that coincidences are the
language in which the cosmos appears from behind cosmology – it would follow
that ethnographic fieldwork is a gambler’s paradise, and that the ethnographer is, as
it were, a cosmic expert. We know, however, that things are not nearly that simple.
In fact, if the analogy between ethnography and gambling makes a virtue of coincidence as alterity, the breakdown of the analogy is just as revealing. For the chief
difference between the gambler and the ethnographer is that while the former is
happy merely to apprehend the coincidences that surprise him, using a readymade
matrix of symbols to translate them into numbers, the ethnographer is professionally
invested in making sense of his own surprises, and, to boot, has no readymade vocabulary with which to do so. So the question arises: what might this exercise in making
sense of ethnographic alterity look like, and on what resources might it draw?
Understood as a (coincidental) divergence between analytical assumptions and ethnographic data, as we have seen, the problem of alterity issues in a series of misunderstandings. In assuming that gambling could be provided with a rationale, I effectively
misunderstood what kind of activity gambling is. In assuming that the guessing game
of the charada is quasi-divinatory, I misunderstood its peculiar orientation toward
coincidence. So the first point to note about alterity is that the challenge it poses to
anthropological analysis is that of understanding (cf. Viveiros de Castro 2003; Henare
et al. 2007; Holbraad 2008). That is to say, contra holism and other forms of etiology,
the problem that things like gambling pose is not that we do not know why people do
them, but rather that we do not, properly speaking, know what they are doing.
9781444333237_4_005.indd 82
8/7/2010 6:44:02 PM
83
But this way of setting up the challenge also suggests the manner in which it may be
taken up. If our misunderstandings of ethnographic data are a function of the analytical
assumptions we bring to them, then it would follow that overcoming our misunderstandings must involve rethinking the assumptions that led to them. Indeed, this is how
the analysis of the present chapter has proceeded. Making virtue of the apparently nonsensical data that, rather, made nonsense of my initial assumptions, I have sought to
transform the latter in order to make sense of the former. In an exercise of what may
best be described as reconceptualization, I used well-worn analytical procedures such as
logical sequencing, transposition, and inversion to try, for example, to make the expectation that gambling must involve a rationale give way to the notion that gambling
turns on the possibility of bypassing processes of reasoning altogether. Similarly, to
articulate the form of this “a-rationality” (the neologism is meant to connote the analytical departure), I turned the supposed parallel with divination into an inversion,
where coincidences become sources, rather than objects, of cosmological reckoning.
So, in conclusion, we may note that while the gambler and the anthropologist are
similarly concerned with the possibility of bypassing familiar ways of thinking in order
to allow the world to reveal itself – and in this sense they are indeed both “cosmic
holists” in the sense I have sought to articulate –their manner of doing so is rather different. More fascinated, perhaps, than challenged by the possibility of bypassing reason,
the gambler is content to channel the cosmic self-revelations of his cábalas directly into
bets, translating them into numbers by means of the transparent ciphers of the charada:
a chance encounter with a drunk is a “sign,” is “drunkard,” is 49, is worth a punt. By
contrast, the anthropologist’s fascination with the coincidences of ethnography derives
from the problem they pose: namely, the fact that they cannot be “translated,” that is,
they cannot be shown to make sense in terms of concepts that are already familiar to
him. So rather than facilitating “translation,” as the symbols of la charada do for the
gambler, the anthropologist’s initial analytical vocabulary (his familiar “cosmology”)
hinders it. In doing so, however, this initial vocabulary also provides the tools for its
own correction: articulating ethnographic alterity as a series of palpable analytical misconceptions, it provides both the starting point and the direction for the processes of
reconceptualization that are required in order to arrive, finally, at sense. For example,
the ethnographic realization that gambling in Cuba makes nonsense of the assumption
that gambling must involve forms of reasoning has not left us merely in the aporia of
having reached the limits of our way of thinking. It has also set the coordinates for our
effort to change our thinking in a way that meets the ethnographic challenge. What is
it exactly about the notion of reasoning that gambling contradicts? And how might we
rethink this notion to overcome the contradiction? As an answer to just such questions,
the conceptualization of the a-rationality of cosmic holism that I have offered in this
chapter, precipitated by the alterity of Cuban gambling, is intended as a recursive,
proof-in-the-pudding illustration of this kind of intellectual endeavor – one that conceptualizes the potential of anthropology as an originator of concepts rather than a
mere user of them.
Acknowledgments
This chapter is dedicated to Panos Giannakakis and for Tata and her family. It is based on a
keynote speech delivered in May 2006 to the annual Megaseminar of the Danish Research
School of Anthropology and Ethnography, a version of which was published in Suomen
9781444333237_4_005.indd 83
8/7/2010 6:44:02 PM
84
Holbraad
Antropologi 32(2) (2007; see also for further acknowledgments). I am grateful to Nils
Bubandt and Ton Otto for inviting me to participate in the present project, and for their
editorial comments on this chapter.
Notes
1 In fact, in conversations with me players often mispronounced the word as cabula (which is
meaningless as far as I know) or even cabra (which means goat!).
2 I thank Morten Pedersen for encouraging me to explore the limits of the analogy.
3 Santería and Ifá divination is often used to predict future events, but also serves as a diagnostic of present and past circumstances of which the client may be unaware (e.g., explaining a current ailment by pointing to an earlier ritual transgression, identifying the source of
ongoing sorcery, etc.).
4 For a fuller discussion, see Holbraad (2008).
5 I borrow the term from Debbora Battaglia (2007), who has written about the role of “happenstantial” connections in anthropological comparison (see also Holbraad and Pedersen
2009).
References
Battaglia, D. (2007). Happenstantial Authorship: A Post-Postmodern Approach to Writing
Culture. Seminar paper at Department of Anthropology, University College London,
March 21.
Boas, F. (1940). Race, Language, and Culture. New York: Macmillan.
Bourdieu, P. (1990). The Logic of Practice, trans. R. Nice. Cambridge: Polity Press.
Brown, D. (2003). Santería Enthroned: Art, Ritual, and Innovation in an Afro-Cuban
Religion. Chicago: University of Chicago Press.
Caillois, R. (2001). Man, Play and Games, trans. M. Barash. Urbana: University of Illinois
Press.
Dumont, L. (1970). Homo Hierarchicus: The Caste System and Its Implications. London:
Weidenfeld and Nicholson.
Durkheim, E. (1982). The Rules of Sociological Method and Selected Texts on Sociology and its
Method, ed. S. Lukes and trans. W. D. Halls. London: Macmillan.
Durkheim, E. (1995). The Elementary Forms of Religious Life, trans. K. E. Fields. New York:
Free Press.
Eliade, M. (1991). The Myth of Eternal Return: or, Cosmos and History, trans. W. R. Trask.
London: Arkana.
Evans-Pritchard, E. E. (1976). Witchcraft, Oracles, and Magic Among the Azande, abridge.
E. Gillies. Oxford: Clarendon Press.
Garfinkel, H. (1967). Studies in Ethnomethodology. Englewood Cliffs, NJ: Prentice Hall.
Geertz, C. (1973a). Thick Description: Toward an Interpretative Theory of Culture. In The
Interpretation of Cultures. New York: Basic Books.
Geertz, C. (1973b). Deep Play: Notes on the Balinese Cockfight. In The Interpretation of
Cultures. New York: Basic Books.
Giddens, A. (1986). The Constitution of Society: Outline of the Theory of Structuration. Berkeley:
University of California Press.
Hacking, I. (1990). The Taming of Chance. Cambridge: Cambridge University Press.
Henare, A., M. Holbraad, and S. Wastell, eds. (2007). Thinking Through Things: Theorizing
Artefacts Ethnographically. London: Routledge.
Holbraad, M. (2008). Definitive Evidence, From Cuban Gods [special issue, Objects of
Evidence”]. Journal of the Royal Anthropological Institute S:S93–S109.
9781444333237_4_005.indd 84
8/7/2010 6:44:02 PM
85
Holbraad, M. (In press). Truth Beyond Doubt: Ifá Oracles in Havana. In Unveiling the Hidden:
Contemporary Approaches to the Study of Divination, ed. K. Munk and A. Lisdorf. Farnham,
UK: Ashgate.
Holbraad, M., & M. A. Pedersen (2009). Planet M: The Intense Abstraction of Marilyn
Strathern. Anthropological Theory 9(4):371–94.
Howe, J. W. (1969). A Trip to Cuba. New York: Negro Universities Press.
Latour, B. (2005). Reassembling the Social: An Introduction to Actor-Network-Theory. Oxford:
Oxford University Press.
Law, J. (2004). After Method: Mess in Social Science Research. London: Routledge.
Lévi-Strauss, C. (1964). Totemism, trans. R. Needham. London: Merlin Press.
Lévi-Strauss, C. (1966). The Savage Mind. Oxford: Oxford University Press.
Lévi-Strauss, C. (1981). The Naked Man: Mythologiques, vol. 4, trans. J. Weightman and
D. Weightman. Chicago: University of Chicago Press.
Lloyd, G. E. R. (2006). Principles and Practices and Ancient Greek and Chinese Science.
Aldershot: Ashgate.
Malaby, T. M. (2003). Gambling Life: Dealing in Contingency in a Greek City. Urbana:
University of Illinois Press.
Malinowski, B. (1953). Argonauts of the Western Pacific: An Account of Native Enterprise and
Adventure in the Archipelagoes of Melanesian New Guinea. New York: Dutton.
Meillassoux, Q. (2008). After Finitude: An Essay on the Necessity of Contingency. London:
Continuum.
Ortíz, F. (1986). Los Negros Curros, ed. D. Iznaga. Havana: Editorial de Ciencias Sociales.
Papataxiarchis, E. (1999). A Contest With Money: Gambling and the Politics of Disinterested
Sociality in Aegean Greece. In Lilies in the Field: Marginal People Who Live for the Moment,
ed. S. Day, E. Papataxiarchis, and M. Stewart. Boulder: Westview.
Radcliffe-Brown, A. R. (1961). Structure and Function in Primitive Society. London: Cohen &
Percy.
Schmaus, W. (1994). Durkheim’s Philosophy of Science and the Sociology of Knowledge: Creating
an Intellectual Niche. Chicago: University of Chicago Press.
Schrempp, G. (1992). Magical Arrows: The Maori, the Greeks, and the Folklore of the Universe.
Madison: University of Wisconsin Press.
Schwartz, R. (1997). Pleasure Island: Tourism and Temptation in Cuba. Lincoln: University of
Nebraska Press.
Sperber, D. (1996). Explaining Culture: A Naturalistic Approach. Oxford: Blackwell.
Stewart, M. (1994). A Passion for Money: Gambling, Luck and the Ambiguities of Money
Amongst Hungarian Gypsies. Terrain 23:45–62.
Thomas, H. (2001). Cuba. London: Pan.
Tylor, E. E. (1970). Religion in Primitive Culture, vol. 2. Gloucester, MA: Peter Smith.
Viveiros de Castro, E. (2003). And. Manchester Research Papers, Social Anthropology.
Manchester: University of Manchester.
Wardle, H. (2005). A City of Meanings: Place and Displacement in Urban Jamaican SelfFramings. In Caribbean Narratives of Belonging: Fields of Relations Sites of Identity, ed.
J. Besson and K. Fog Olwig. Oxford: Macmillan.
Woodburn, J. (1982). Egalitarian Societies. Man 17:431–51.
9781444333237_4_005.indd 85
8/7/2010 6:44:02 PM
9781444333237_4_005.indd 86
8/7/2010 6:44:02 PM
Part 2
Beyond Cultural Wholes?
Wholes are Parts, and Parts
are Wholes
p02.indd 87
8/7/2010 7:21:27 PM
p02.indd 88
8/7/2010 7:21:27 PM
6
It is probably no exaggeration to say that no concept has been more centrally discussed
in twentieth-century anthropology than the concept of culture (Kroeber and
Kluckhohn 1963 [1952]; Barnard and Spencer 1996: 136; Brumann 1999). Its continued popularity as well as contestation are predominantly linked to the uniquely
influential American tradition of anthropology, with its German antecedents, while
the concept has not had the same central place in British and French anthropology
(Kuper 1999; Parkin 2005). In a simplified way, one could say that the French tradition has been more interested in the notion of symbolic structure (see Chapter 10,
this volume) and the British tradition has largely opted for a focus on society or social
structure (see Chapter 14). The concept of culture is undoubtedly one of the key
holistic notions of anthropology, although we will see that the nature of the wholes
to which it supposedly refers has varied greatly between authors and theoretical
traditions.
The particular anthropological usage of culture, which employs the concept in
the plural to refer to the different life worlds characteristic of different peoples
instead of in the singular to indicate a state of being more or less “cultured” or
civilized, derives from the German Romantic nationalism of the late eighteenth
century. Especially influential were the works of the philosopher Johann Gottfried
von Herder (1744–1803), who emphasized the principle of equality of all peoples
and their cultures, equally deserving of respect for their individual spirit and
design (see Chapter 7). The perhaps earliest and most cited anthropological definition of culture with a holistic connotation to it, E. B. Tylor’s notion of culture
as a “complex whole,” does not follow this German tradition. Rather, it is derived
from the English and Continental notion of “civilization.” Culture, considered
synonymous with civilization in his definition,1 is here used in the singular, as
something that the whole of humanity has a part in but at different levels or stages
of development. The plural notion of culture, as characteristic of different peoples
or nations, entered American anthropology via another route, namely, via the
German migrant Franz Boas, who, originally trained as a physicist and geographer
in his country of birth, became the founding father of cultural anthropology in
the US (Stocking 1968).
c06.indd 89
8/7/2010 7:16:41 PM
90
Otto and Bubandt
Franz Boas’ Heritage
Boas’ understanding of culture created an important alternative to the view on human
difference as based on biology or race. It focused on the historical development of
particular cultures through the fusion of elements or “traits” originating in different
times and places. To what extent Boas had a holistic comprehension of the integration
of culture is still an issue of debate. The dominant view appears to be that Boas
emphasized the integration and thus wholeness of cultures and saw it as a result of the
“geniuses” of the peoples who developed them (Stocking 1968). But Verdon argues
that in his ethnographic practice Boas thought more in terms of cultural fragments
rather than wholes (2006: 297–8). According to Darcy, Boas’ concept of culture
became more structural and holistic toward the end of his career (1987: 5). It is clear,
however, that Boas’ own formulations of a holistic view on culture were connected to
his promotion of the method of ethnographic fieldwork (see also Chapter 1). For
Boas, ethnographic fieldwork enabled a holistic perspective of culture that the comparative methods of diffusionism and evolutionism lacked:
We have another method, which in many respects is much safer. A detailed study of the
customs in their relation to the total culture of the tribe practicing them, in connection
with an investigation of their geographical distribution among neighbouring tribes,
affords us almost always a means of determining with considerable accuracy the historical
causes that led to the formation of the customs in question and to the psychological processes that were at work in their development. (Boas 1940 [1896]: 276; italics added)
This quotation, which also shows Boas’ interest in the historical development of particular cultures, points to a parallel movement in Britain and the US in the early twentieth century, namely, the close link between small-scale, intensive fieldwork and a
holistic vision of culture (see Chapter 14). Boas had a large number of students and
followers, who spread and further developed his pluralistic vision of culture. One of
them, A. L. Kroeber, concentrated in particular on comprehending the nature of culture, which he specified as “superorganic” to underline its difference and independence from human organisms (1917). Culture was thus considered as a sui generis
phenomenon: it could only be explained in its own terms and not be reduced to biological, racial, and psychological factors.
Ruth Benedict was perhaps the student of Boas who most intensely occupied herself with the question of the internal integration of cultures. In her famous work
Patterns of Culture, obligatory reading for many generations of American students of
anthropology, she formulates very clearly the holistic idea that the whole is more than
the sum of its parts:
Cultures … are more than the sum of their traits. We may know all about the distribution of
a tribe’s form of marriage, ritual dances, and puberty initiations, and yet understand nothing of the culture as a whole which has used these elements to its own purpose. This purpose selects from among the possible traits in the surrounding regions those which it can
use, and discards those which it cannot. Other traits it recasts into conformity with its
demands. (Benedict 1989 [1934]: 47; italics added)
Benedict compares the integration of cultures with the way in which a style of
art comes into being and persists as the product of a mostly unconscious collective
c06.indd 90
8/7/2010 7:16:41 PM
91
process. This process of integration, therefore, may also fail, as is the case in some
cultures like in some periods of art (1989 [1934]: 47–8). For Benedict, this internal
coherence of a culture was similar to that of an individual person: “A culture, like an
individual, is a more or less consistent pattern of thought and action” (1989 [1934]:
46). This interest in the human person and its relation to cultural wholes is a persisting theme in American anthropology, and for some, like Benedict’s contemporary
Edward Sapir, the emphasis was on the individual as the vital locus of cultural action
rather than on culture as a whole (Varenne 1984; Handler 1990: 163; Barnard and
Spencer 1996: 139). The influence of Gestalt psychology, one of the new holistic
sciences around the beginning of the twentieth century (see Chapter 1), played an
important role in Benedict’s understanding of cultural and personal integration, and
this understanding became for Benedict one of the ways in which she sought to distinguish her own holism from that of Malinowski because of the latter’s emphasis on
functional integration (Benedict 1989 [1934]: 50–2). For Benedict, the whole is
meaningful and integrated around a few dominant drives, themes, or patterns that
can be found at both the cultural and the personal level (Handler 1990: 172–3).
Benedict’s focus on the integration of culture and the person may also have had a
biographical origin. According to Handler, Benedict was a typical exponent of
modernity who was struggling to maintain her own personal integration in a fragmented modern world. Her modernist project was a quest for integrity that she
achieved by making cultural wholeness her disciplinary idea and “through writing
about – or writing into existence – cultures that could be seen to be whole, holistic,
and authentic” (Handler 1990: 180).
Different Conceptions of Culture and Holism
The further intellectual history of the culture concept in American anthropology is
quite complex and can only be sketched in broad outline here (but see Ortner
1984; Keesing 1974). In the first place, cultural anthropology developed within a
disciplinary field that also included sociology, which in the US led to a “division of
labor” between the study of the cultural (assigned to anthropology) and the study
of the social and its systematics (the domain of sociology).2 Secondly, no doubt
simplifying the complexity within anthropology unduly, one could say that there
were two main tendencies in the understanding of the cultural. On the one hand,
the majority of anthropologists followed the heritage of Boas and Benedict with an
emphasis on culture as an ideational system. This included not only the school of
cultural relativism (Margaret Mead, Clyde Kluckhohn) with its focus on “values” as
the key defining parts of a culture, but also the emerging school of cognitive anthropology (Ward Goodenough). On the other hand, there were those who were more
interested in seeing culture as an adaptive system, whether or not in an evolutionary
perspective; this included the theories of so-called neo-evolutionism (Leslie White),
cultural materialism (Marvin Harris), and cultural ecology (Roy Rappaport and
Andrew Vayda). In all their diversity, these approaches nevertheless shared the
assumption that culture was a system of one kind or another; that culture was an
integrated whole that was more than just a sum of its diverse elements. Holism, in
this way, became part of the shared doxa of American cultural anthropology (see
also Chapter 1).
c06.indd 91
8/7/2010 7:16:41 PM
92
Otto and Bubandt
Before dealing with the more recent perspectives on the notion of culture and its
holistic implications, we should also briefly address another way in which the concept of holism has been used in American anthropology. This was in connection with
the so-called four-field approach – the “sacred bundle” of biological anthropology,
linguistics, cultural anthropology, and archaeology – that developed institutionally
in the wake of Boas (Stocking 1988; Segal and Yanagisako 2005). For Boas, these
four fields – already existing as disciplines in their own right – were to be combined
in anthropology as the study of humankind as a whole. This was in line with the
interest of his time in evolutionary and historic explanations of human variation, an
interest that was shared by all four fields and to which each contributed with its own
methodology and data.3 Whereas the turn to the holistic understanding of society in
connection with ethnographic fieldwork (see Chapters 1, 2, and 14) meant an end
to this close collaboration between different disciplines in England and on the
European continent, in the US the four-field approach became entrenched in the
institutional organization of the discipline of anthropology. Without doubt
the charismatic leadership of Boas – one of the few anthropologists who actually
contributed to all four fields – was an important factor in this institutional development, which was inscribed in the American Anthropological Association and its flagship journal American Anthropologist, as well as in the organization of the
anthropology departments that became established across the country. Holism in
the American context is thus often used to refer to the cross-disciplinary collaboration in the study of humankind, which according to Borofsky in practice was more
an ideal or a myth than a reality (Borofsky 2002; Segal and Yanagisako 2005).4 It is,
however, a myth that has been defended with passion, especially at times of institutional threat and reorganization and that in view of new forays in cognitive science
appears far from outdated (Schiffer 1999; Shore 1999; Wiley 2006) (see Chapter 14
for a recent reemergence of a similar interest in the UK). Although they shared the
same genealogy in Boasian anthropology, it is clear that the holism of the four-field
approach and cultural holism were very different and increasingly opposed kinds.5
The holism of the four-field approach thus rested on the ideals of natural science,
ideals that from the 1960s onward were becoming increasingly problematic for
many cultural anthropologists for whom the study of culture could no longer be
pursued with the methods and truth claims from natural science. By the 1970s,
culture theory was changing dramatically.
Clifford Geertz and Hermeneutical Holism
The two towering figures in this reinvention of anthropological culture theory in the
1970s were Clifford Geertz and Marshall Sahlins – each with his own new take on the
holism of culture. Although both hardly use the notion of holism themselves,6 we
believe that a reading of their theoretical positions as holistic enterprises is both
relevant and illuminating, as we will try to demonstrate in the following. We start with
Geertz, who as no other has come to personify the hermeneutic or symbolic turn in
anthropology in the late 1960s and 1970s. Geertz explicitly wished to position
anthropology as a humanistic or interpretative science in search of meaning, in contrast to an experimental or natural science in search of laws (Geertz 1973a). For
Geertz, culture was a system of meanings embodied in publicly accessible symbols and
c06.indd 92
8/7/2010 7:16:42 PM
93
behavior. Geertz famously compared culture to a text that humans – anthropologists
or not – were in the business of deciphering. The task of anthropologists was therefore a hermeneutical one: to inscribe their reading of cultural “texts” and to produce
a “thick description,” that is, an interpretive rendering, of the culture studied (Geertz
1973b). But what kind of assumption about the nature of cultural integration underlay the idea of thick description? To answer this question, Geertz uses the following
metaphor, which interestingly is not one of a textual whole but of a weakly integrated
biological organism:
The appropriate image, if one must have images, of cultural organization, is neither the
spider web nor the pile of sand. It is rather more the octopus, whose tentacles are in large
part separately integrated, neurally quite poorly connected with one another and with
what in the octopus passes for a brain, and yet who nonetheless manages both to get
around and to preserve himself, for a while anyway, as a viable if somewhat ungainly
entity. (Geertz 1973a: 407–8)
Thus, even though a culture is ultimately joined into a kind of system or loosely
structured whole, it shows “confluences of partial integrations, partial incongruencies, and partial independencies” (Geertz 1973a: 408). Therefore, a “heroic”
holism that operates with a comprehensive, systemic whole available for scientific
capture is not the appropriate strategy for cultural analysis, as it exaggerates the
level of cultural coherence and integration (1973a: 408). Geertz’s interpretative
approach, with its weak form of holism, has been enormously influential within but
also outside of anthropology, where it often has come to epitomize the anthropological approach. Geertz’s focus on culture as primarily an ideational system did not
go unchallenged, though, especially when American anthropology became influenced by neo-Marxist and postcolonial critique in the late 1970s and 1980s (see
Chapter 1). The main objection was that the intricate connection between processes of power and cultural integration – or hegemony – had remained largely outside of Geertz’s analytical perspective (Asad 1973; Keesing 1987; Wolf 1990;
Ortner 1999).
Marshall Sahlins and Symbolic Holism
A different trajectory was taken by Marshall Sahlins, who, inspired by his teacher
Leslie White, was originally mostly associated with the culture-as-adaptive-system line
of thinking (Sahlins 1958). Later, Sahlins became strongly influenced by French
structuralism while also taking in the critique of neo-Marxism. His strongest statement on culture is in his much debated and highly influential book Culture and
Practical Reason (Sahlins 1976), in which he argues for a culturalist tradition stemming from Boas and rejects explanations of culture based solely on adaptation or
economic praxis:
As opposed to all these genera and species of practical reason, this book poses a reason of
another kind, the symbolic or meaningful. It takes as the distinctive quality of man not
that he must live in a material world … but that he does so according to a meaningful
scheme of his own devising.… It therefore takes as the decisive quality of culture – as
c06.indd 93
8/7/2010 7:16:42 PM
94
Otto and Bubandt
giving each mode of life the properties that characterize it – not that this culture must
conform to material constraints but that it does so according to a definite symbolic
scheme which is never the only one possible. Hence it is culture which constitutes utility.
(Sahlins 1976: viii)
The book makes a powerful argument for culture as a symbolic system that – following the discoveries of structural linguistics – has a “fundamental autonomy” (Sahlins
1976: 57) based on the arbitrary character of the sign.7 The analysis of culture is essential to understanding the human relation to the world. Culture is a tertium quid – a
“third something” in between mind and matter, subject and object, human actor and
environment, specific practices and general human praxis – that constitutes the relevant
subjective and objective terms of that relationship (Sahlins 1976: x, 70–1). Therefore
“practical reason,” including Marxism and functionalism, can be understood as a particular cultural form of modern Western culture. In our view this is a very strong expression of cultural holism, since cultures are thus characterized by their particular logic,
scheme, system, or reason which is the basis for their integration and the specific object
of anthropological analysis.8 Sahlins thus develops a systemic concept of culture that
aims to combine two different traditions, both present in American anthropology: one
that sees culture as an adaptive, material system and the other that sees it primarily as a
symbolic system.9 Inspired by French structuralism (see Chapter 10), he tries to transcend the opposition between material and ideational perspectives by theorizing symbolic systems in a comprehensive – holistic – way that is clearly distinct from the
Geertzian hermeneutic project. Whereas Geertz appears to follow Parsons’ division
between the social and the cultural – the latter being the object of interpretative
analysis – Sahlins sees the social as part of the symbolic system.
Practice Theory
Culture and Practical Reason (Sahlins 1976) is a wonderfully polemic book that
takes a provocative stance on culture as a holistic concept. Culture itself is for Sahlins
the “explanans” – a systemic order that explains human action – but not an “explanandum,” something that itself has to be explained. In later works, Sahlins (1981, 1985)
addresses the central question of how cultures change in and through human action.
According to Sherry Ortner, this – together with key works by Pierre Bourdieu
(1977) and Anthony Giddens (1979) – marks the emergence of a new theoretical
paradigm, which she – following Bourdieu – has called “practice theory” (Ortner
1984; 2006: 2). Central to this new frame of thinking is the attempt to mediate
between explanations of structural constraints (functionalism, structuralism, Marxism)
and explanations of human action or agency (interactionism, transactionalism, phenomenology). System thinking is an integral part of practice theory (see Ortner
1984: 148–9), and this may explain why it still clearly contains the specter of holism.
This is a holistic perspective, however, that embraces the notion of human action.
Practice, after all, emerges out of the dialectics of subject–object and agent–structure
interaction, and it is this insight which allows practice theory to open up to larger,
systemic questions about power and cultural change.
Ortner is herself one of the leading American proponents of practice theory.
A student of Geertz, she combines practice theory with an interpretive approach that
c06.indd 94
8/7/2010 7:16:42 PM
95
maintains the holism implied in thick description. Indeed, Ortner explicitly defends
its holism in spite of the “hubris” of the original holistic vision:
I would argue that thickness (with traces of both exhaustiveness and holism) remains at
the heart of the ethnographic stance. Nowadays, issues of thickness focus primarily on
issues of (relatively exhaustive) contextualization. (2006: 43)
Indeed, for Ortner the Geertzian notion of thickness overlaps very much with the
holism that she sees as a conditio sine qua non of anthropology. If thickness in the
sense of ethnographic involvement and contextualization was at the heart of ethnographic inquiry and was designed to give a complex representation of ethnographic
subjects, Ortner is critical of what she calls “ethnographic refusal” – an ethnographic
thinness that “destroys the object (the subject) who should be enriched, rather than
impoverished, by this complexity” (2006: 54). In a critique of resistance studies in
particular, Ortner argues that these fail because they “thin” and “sanitize” culture by
insisting on a dichotomy between resistance and domination, instead of exploring in
more detail the internal mechanisms of power also within the subaltern. The effect is
a refusal of the complexity that ethnographic thickness was supposed to deliver. As
such it was also, to the detriment of anthropology, a refusal of holism:
If the ethnographic stance is founded centrally on (among other things, of course) a commitment to thickness, and if thickness is taken and still takes many forms, what I am
calling ethnographic refusal involves a refusal of thickness, a failure of holism or density
which itself may take various forms. (2006: 43–4)
This theoretical position articulates well with that of Marcus (1998), who also
explored the contemporary conditions of ethnographic thinness and thickness, albeit
with a more positive outlook on the benefits of ethnographic thinness. Our impression is that the acceptance of a soft version of cultural holism is the dominant trend in
contemporary American anthropology, although there have also been strong critical
voices against the culture concept, because of its inherent totalizing tendencies. AbuLughod (1991), for example, argued for “writing against culture” as a feminist and
“halfie” (person with mixed cultural identity), because the notion of culture created
artificial boundaries and hierarchical relations between people (see also Chapter 10).
Particularly from the 1990s onwards there has been a strong critical reaction against
the use of the concept of culture within American anthropology, targeting in particular its holistic – in the sense of totalizing – connotations (see Brumann 1999 for an
overview of the debate and a defense of the culture concept).
The Chapters
The contributions to Part 2 take the discussion of cultural holism into new directions,
which address the (mis)conception of cultural boundedness. Marshall Sahlins and
Alan Rumsey deal with the theoretical assumption of autonomous and sui generis
wholes in cultural and linguistic theory, respectively. To resolve the widespread “scandal” of seeing these “wholes” in isolation (more so in cultural theory than in linguistics), they argue that the theoretical perspective needs to be widened, because such
c06.indd 95
8/7/2010 7:16:42 PM
96
Otto and Bubandt
“wholes” – whether from the perspective of the researchers or of the participants –
only come into being as parts of a larger field of relations. Mark Mosko, meanwhile,
revisits the issue of the internal integration of cultural worlds but does so by using
concepts derived from complexity theory.
In his contribution to this section, Marshall Sahlins radically rejects what he identifies
as the dominant trend within much of twentieth-century anthropology: the tendency
to consider cultures and societies as whole systems or more or less coherent orders, and
to see the main theoretical challenge as explaining this internal coherence. In spite of
felicitous exceptions – such as Claude Lévi-Strauss’ comparative study of myth and
Fredrik Barth’s study of ethnic boundaries – this bias toward a concern with internal
coherence while neglecting external connections is, for Sahlins, evident across a wide
variety of theoretical traditions. This neglect is scandalous, Sahlins argues, because it is
precisely the external relations that create the specific dynamics and apparent coherence
of the units that become identified in the process as societies, tribes, nations, or cultures. Sahlins unfolds this central theoretical point by analyzing a wealth of empirical
examples that demonstrate how “anthropology-cultures” only make sense if they are
studied in the greater fields of historical interaction and change of which they are a part.
The widespread phenomenon of the “stranger-king” is a case in point, as it suggests
that indigenous and exogenous cultural concepts and political forces are connected to
form locally specific – and seemingly internally coherent – constellations.
Sahlins argues that the very processes of mutual articulation and dependency
between cultural constellations that have to be kept in focus in order to avoid succumbing to monadic myopia also explain how the concept of culture became such a
bounded and problematic concept in the first place. Developed in the Romantic
counter-Enlightenment movement in Germany at the end of the eighteenth and the
beginning of the nineteenth centuries, the bounded idea of culture is a product of the
political and cultural constellation at the time. In opposition to the idea of a universal
civilization with its apex in the West that was prevalent at the time in the politically
dominant countries of France and England, German intellectuals crafted a concept of
culture that referred to the identity, individuality, value, and dignity of the traditions
of all groups or nations, independent of their economic development and political
might. The concept of culture came, in other words, to denote the “essence” or particular “being” of a people, defined through their cultural heritage. This national
essentialism, so Sahlins argues, developed within a political contrast to the universalist
claims of the British and French, for whom political nationhood was equated with the
concept of civilization that carried strongly evolutionary overtones but that in principle was available to all. Sahlins expresses the paradox between this bounded idea of
culture and its intercultural and political origin very pointedly: “[T]he idea of a culture as an anthropological object constituted by and for itself developed out of a
relationship between European cultural orders – and, more specifically, out of the
interdependence of their differences” (Chapter 7: 119). As we have discussed in
Chapter 1, the anthropological reception of this concept of culture at the beginning
of the twentieth century was part of a wider movement toward holism across a number
of disciplines – especially biology, neurology, and psychology. Interestingly, a major
drive toward this intellectual development occurred in Germany, but not at all exclusively so (Harrington 1996), and it has probably to be understood also in connection
to other cultural and political-economic phenomena than Germany’s lack of political
centralization and power, such as the emerging dominance of a mechanical worldview
due to the success of physics, engineering, and industry.
c06.indd 96
8/7/2010 7:16:42 PM
97
Another holism that originated at the beginning of the twentieth century was that
of structural linguistics. First formulated by the famous Swiss linguist Ferdinand de
Saussure (1857–1913), this understanding of language as a synchronic system was
to have a great impact on anthropology (especially via structural anthropology; see
also Chapter 10). Language has often been used as the prototypical model for the
notion of culture as a whole, but Alan Rumsey, a linguistic anthropologist, shows in
Chapter 9 that this is problematic in a number of ways. First, languages are not simply “wholes” in the sense of discrete, internally coherent systems. In fact, such a
view of language is an abstraction created by the linguistic method, as stated clearly
by the founding father of structural linguistics himself: “it is the viewpoint that creates the object” (Saussure 1966: 8). The conception of a synchronically coherent
system is based on the methodological “fiction” of a “language state,” which
abstracts from regional and historical variation. From another viewpoint, languages
can be seen as internally heterogeneous at every scale, and their boundaries as being
in fact quite arbitrary. Again this alternative view was already expressed by Saussure:
“Just as dialects are only arbitrary subdivisions of the total surface of language, so
the boundary that is supposed to separate two languages is only a conventional one”
(1966: 204).
Moving from a linguistic, analytical perspective to that of the speakers of a language,
Rumsey shows that whether or not a language is conceived as a whole by the users of
the language varies enormously among speech communities: from a total lack of such a
notion of a bounded language to the ideological image of one shared language for the
whole group (even though actual variation with it can be enormous). The latter is the
case with nation-states such as France and Germany, where there is an assumption of –
as well as active development toward – a standardized common language. But this is
only one way in which the relation between a group and a language can be conceived,
coming close to an imagined one-to-one relationship.10
Rumsey demonstrates that language is used in a variety of alternative ways in the
constitution of groups, quite different from the one language–one nation model. In
two of the non-European examples he discusses, speakership is not a criterion for social
belonging but language is still a factor in group identity. In northwestern Amazonia
(Vaupés), people trace language “ownership” through descent, while Australian
Aborigines see a relationship between a language and the land, to which people are
related by birth. In both cases there is in practice a high level of multilingualism, and
speakership is therefore clearly not a relevant criterion for the constitution of groups.
In the third example, from Highlands Papua New Guinea, there does not appear to be
any relationship between language, social identity, and group formation at all.
On the basis of these linguistic and sociolinguistic considerations, Rumsey argues
that holism in (linguistic) anthropology is still essential, both as a heuristic perspective
to get hold of the holisms employed by the users of a language, and as an analytical
perspective to bring into view the networks of relations that constitute spoken languages in their complex and heterogeneous expansion and transformation. Like
Sahlins, Rumsey prefers to see these networks as “relational” or “dynamic fields,” but
Rumsey rejects speaking of these as part–whole relationships, which indicates the
wrong type of relationality for them. The conception of a relational field, according
to Rumsey, is compatible with Saussure’s understanding of a linguistic system, which
arguably was never as bounded as later generations have made it.
In his contribution (Chapter 9), Mark Mosko takes the discussion of cultural
holism in the opposite direction of Sahlins and Rumsey by revisiting the issue
c06.indd 97
8/7/2010 7:16:42 PM
98
Otto and Bubandt
of – primarily – internal coherence within a cultural order. He argues that modern chaos
or complexity theory can help us to understand part–whole connections in a new way
that avoids the epistemological and ontological problems of earlier holisms. To this end
Mosko draws on fractal holography that, among other things, states that whole and
part, or form and content, are relative: what is form (whole) at one level can become
content (part) on another. Whereas awareness of this form–content relativity is not new
– Mosko mentions a number of antecedents in anthropology, such as Evans-Pritchard’s
model of segmentary opposition – fractal holography adds the dimension of self-similarity or self-scaling, which means that a particular form is retained across levels or
scales. An example of such self-similarity is the leaf of a fern plant, the overall shape of
which is replicated in each of its leaflets and subleaflets. Thus the same form appears
over and over again at different levels and across different domains of the phenomenon
studied. This creates an open-ended coherence that can extend inwardly, to lower levels
or parts, as well as outwardly, to more inclusively wholes. Whereas Chapter 7 by Sahlins
argues that the (cultural) whole is a part (of a larger field), Mosko reverses this argument by pointing out that each part can be a whole in its own right. In this way, cultural
integration can be understood as being realized through a continuous repetition of
similar forms at lower scales. But Mosko’s fractal model also allows us to analyze “external” relations between cultural constellations in the same vein, namely, as repetitions of
self-similar forms at higher scales. This is an interesting complement to Sahlins’ thesis
but lacks the latter’s view on the political dynamics of those relations.
The holistic experiment in which Mosko engages is a reanalysis of Trobriand ethnography, based on an analysis of the extensive literature as well as his recent fieldwork. He
shows how a basic indigenous plant classification is a pervasive cultural model or form
that creates content across many, if not most, levels and domains of Trobriand society.
The form is based on indigenous conceptions of the yam – a centrally important tuber
used for subsistence and exchange on the islands – as consisting of a base, body, and tip,
as well as fruit or offspring. This model provides a spatial and temporal metaphor that
orders both physical objects and social action in the life world of Trobriand people.
Gardens, yam storage houses, and human bodies are all understood as being composed
of the same elements: base, body, and tip, and the fruit. These elements also define a
sequence and therefore a scenario for action, for example in relation to the exchanges
that constitute the basis of social life and personal fame in the Trobriands.
Mosko is the first ethnographer to argue for the systemic pervasiveness and enchainment of this plant classification as a form of action scenario in the Trobriands, which
thus functions to create holographic transformations across multiple spheres of
Trobriand social life. By looking at cultural order in this way, Mosko not only provides
a fresh understanding of the experience of cultural coherence and integration, but also
argues that this view contributes to a better understanding of the theoretical problem
of the relation between agency and structure. Although the plant model is a priori to
human action (“culture-as-constituted,” in Sahlins’ words), the action it gives way to
is open-ended and generative of new sequences and wholes.
The three chapters in this section open new and thought-provoking vistas on how
to understand cultural and lingual constellations in their interrelatedness and internal integration. They all argue for the need to rethink the structural-anthropological
idea of relational systems in ways that do not succumb to the theoretical errors of
earlier cultural holisms, in particular their tendency to institute in anthropological
analysis the pervasive idea that cultural systems by default were bounded, closely integrated, and sui generis.
c06.indd 98
8/7/2010 7:16:42 PM
99
Notes
1 E. B. Tylor (1832–1917) was one of the first English anthropologists. He gave the
following definition of culture at the start of his well-known book Primitive Culture:
“Culture or civilization, taken in its wide ethnographic sense, is that complex whole which
includes knowledge, belief, art, morals, law, custom, and any other capabilities and habits
acquired by man as a member of society” (Tylor 1871: 1).
2 This way of carving up the field was famously elaborated at Harvard University by Talcott
Parsons. We are grateful to Alan Rumsey for pointing this out to us.
3 In the late nineteenth century, there was a greater overlap in the methods used by these
disciplines (see Tsing, Chapter 4, this volume).
4 Borofsky (2002) argues that in spite of a widespread conviction about the opposite, actual
collaboration between the four fields as a fulfilment of the holistic ideal is relatively modest
(only 9.5 percent of all articles published in American Anthropologist are based on collaborative research of this kind). Borofsky still wishes to maintain the ideal of holism, but
interestingly his version of holism moves away from the cross-disciplinary approach into a
broader definition that bases holism on the training of anthropologists and how they cope
with solving problems of the world (2002: 474).
5 Tim O’Meara, for instance, argued that the scientific ideals of the four-field approach
could only be realized if the dominance within anthropology of cultural holism was broken
and substituted by a scientific kind of “causal individualism” (1999).
6 Geertz has used the term only a few times, but Sahlins does not appear to use the word at all.
7 According to Saussure (see below and Chapter 8), the relation within a linguistic sign
(such as a word) between a sound image (signifié) and its meaning (signifiant) is not intrinsic but conventional. See Sahlins (1976: 62–4, 214–15).
8 Victor Li (2001) also characterizes Sahlins’ perspective on culture as holistic.
9 One could perhaps say that Sahlins continues and completes the theoretical project of his
teacher Leslie White, for whom the material and symbolic perspectives were not fully
integrated (Sahlins 1976: 103–5). He articulates a strong allegiance to White’s central idea
that the symbol is the origin and basis of human behavior: “In all its dimensions, including
the social and the material, human existence is symbolically constituted, which is to say,
culturally ordered” (Sahlins 1999: 400).
10 There are the German- and French-speaking groups that are not part of the nation-state,
and there are also nation-states that do not have their own language or that have several
ones (such as Switzerland, Canada, and Belgium).
References
the Present, ed. R. Fox. Santa Fe, NM: School of American Research Press, 137–62.
Asad, Talal, ed. (1973). Anthropology and the Colonial Encounter. London: Ithaca.
Barnard, Alan, and Jonathan Spencer, eds. (1996). Encyclopedia of Social and Cultural
Anthropology. London: Routledge.
Benedict, Ruth (1989). Patterns of Culture [1934]. Boston: Houghton Mifflin.
Boas, Franz (1940). The Limitations of the Comparative Method of Anthropology [1896]. In
Race, Language and Culture, ed. F. Boas. New York: Macmillan, 270–80.
Borofsky, Robert (2002). The Four Subfields: Anthropologists as Mythmakers. American
Anthropologist 104(2): 463–80.
Bourdieu, Pierre (1977). Outline of a Theory of Practice. Cambridge: Cambridge University Press.
Brumann, Christoph (1999). Writing for Culture: Why a Successful Concept Should Not Be
Discarded. Current Anthropology 40: S1–S27.
c06.indd 99
8/7/2010 7:16:42 PM
100
Otto and Bubandt
Darcy, Anthony (1987). Franz Boas and the Concept of Culture: A Genealogy. In Creating
Culture: Profiles in the Study of Culture, ed. D. J. Austin-Bross. Sydney: Allan & Unwin,
3–17.
Geertz, Clifford (1973a). The Interpretation of Cultures. New York: Basic Books.
Geertz, Clifford (1973b). Thick Description: Towards an Interpretive Theory of Culture.
In The Interpretation of Cultures. New York: Basic Books, 3–30.
Giddens, Anthony (1979). Central Problems in Social Theory: Action, Structure and
Contradiction n Social Analysis. Berkeley: University of California Press.
Handler, Richard (1990). Ruth Benedict and the Modernist Sensibility. In Modernist
Anthropology: From Fieldwork to Text, ed. M. Manganaro. Princeton: Princeton University
Press, 163–80.
Keesing, Roger M. (1974). Theories of Culture. Annual Reviews of Anthropology 3: 73–97.
Keesing, Roger M. (1987). Anthropology as Interpretative Quest. Current Anthropology 28(2):
161–76.
Kroeber, A. L. (1917). The Superorganic. American Anthropologist 19: 163–213.
Kroeber, A. L., and Clyde Kluckhohn (1963). Culture: A Critical Review of Concepts and
Definitions [1952]. New York: Vintage.
Kuper, Adam (1999). Culture: The Anthropologists’ Account. Cambridge, MA: Harvard
University Press.
Li, Victor (2001). Marshall Sahlins and the Apotheosis of Culture. CR: The New Centennial
Review 1(3): 201–87.
Marcus, George E. (1998). Ethnography Through Thick and Thin. Princeton: Princeton
University Press.
O’Meara, Tim (1999). Causal Individualism and the Unification of Anthropology. In
Anthropological Theory in North America, ed. E. L. Cerroni-Long. Westport, CT:
Greenwood, 105–42.
Ortner, Sherry B. (1984). Theory in Anthropology Since the Sixties. Comparative Studies in
Society and History 26: 126–66.
Ortner, Sherry B. (1999). Thick Resistance: Death and the Cultural Construction of Agency in
Himalayan Mountaineering. In The Fate of “Culture”: Geertz and Beyond, ed. S. B. Ortner.
Berkeley: University of California Press, 136–63.
Ortner, Sherry B. (2006). Anthropology and Social Theory: Culture, Power, and the Acting
Subject. Durham: Duke University Press.
Parkin, Robert (2005). The French-Speaking Countries. In One Discipline, Four Ways: British,
German, French, and American Anthropology, ed. F. Barth, A. Gingrich, R. Parkin, and
S. Silverman. Chicago: Chicago University Press, 155–253.
Sahlins, Marshall (1958). Social Stratification in Polynesia. Seattle: University of Washington
Press.
Sahlins, Marshall (1976). Culture and Practical Reason. Chicago: University of Chicago
Press.
Sahlins, Marshall (1981). Historical Metaphors and Mythical Realities: Structure in the Early
History of the Sandwich Islands Kingdom. Ann Arbor: University of Michigan Press.
Sahlins, Marshall (1985). Islands of History. Chicago: University of Chicago Press.
Sahlins, Marshall (1999). Two or Three Things That I Know About Culture. Journal of the
Royal Anthropological Institute (N.S.) 5: 399–421.
Saussure, Ferdinand de (1966). Course in General Linguistics (trans. Wade Baskin).
New York: McGraw-Hill.
Schiffer, Michael Brian (1999). Return to Holism. Anthropology Newsletter, March, 62–4.
Segal, Daniel A., and Sylvia J. Yanagisako, eds. (2005). Unwrapping the Sacred Bundle:
Reflections on the Disciplining of Anthropology. Durham: Duke University Press.
Shore, Bradd (1999). Strange Fate of Holism. Anthropology News 40(9): 4–5.
c06.indd 100
8/7/2010 7:16:42 PM
101
Stocking, George W. (1968). Franz Boas and the Culture Concept in Historical Perspective. In
Race, Culture, and Evolution: Essays in the History of Anthropology, ed. G. W. Stocking.
New York: Free Press.
Stocking, George (1988). Guardians of the Sacred Bundle: The American Anthropological
Association and the representation of Holistic Anthropology. In Learned Societies and the
Evolution of the Disciplines, American Council of Learned Societies Occasional Paper no.
5. New York: American Council of Learned Societies.
Tylor, Edward B. (1871). Primitive Culture. London: John Murray.
Varenne, Hervé (1984). Collective Representation in American Anthropological Conversations:
Individual and Culture. Current Anthropology 25(3): 281–99.
Verdon, Michel (2006). Boas and Holism: A Textual Analysis. Philosophy of the Social Sciences
36(3): 276–302.
Wiley, Andrea (2006). The Breakdown of Holism: And the Curious Fate of Food Studies in
Anthropology. Anthropology News 47(1): 9–12.
Wolf, Eric R. (1990). Distinguished Lecture: Facing Power – Old Insights, New Questions.
American Anthropologist 92(3): 586–96.
c06.indd 101
8/7/2010 7:16:42 PM
7
The Whole is a Part
Intercultural Politics of Order
and Change
Marshall Sahlins
It is no exaggeration to say that anthropology, by virtue of its traditional concepts of
societies and cultures as self-organized monads, has been implicated for centuries in a
major theoretical scandal. It all began in the latter part of the eighteenth century with
the German counter-Enlightenment ideas of national cultures and national characters. The scandal is that while cultures were thus conceived as autonomous and sui
generis, they have always been situated in greater historical fields of cultural others
and largely formed in respect of one another. Even autonomy is a relation of heteronomy. But our major theories of cultural order, based one and all on insular epistemologies, presuppose that societies are all alone and that cultures, as it were, make
themselves. (Or at least such have been the assumptions until very recently, when
these theories got knocked around by globalization and postmodernism.) As Fredrik
Barth put it some decades ago, “Practically all anthropological reasoning rests on the
premise that cultural variation is discontinuous: that there are aggregates of people
who essentially share a common culture and the interconnected differences that distinguish such culture from all others” (1969: 9). Consider the paradigmatic perversity
entailed in Malinowski’s foundation of modern ethnography as a study of the internal
engagements of different aspects of a Trobriand culture that was participating in a vast
trade network involving, among other intersociety relations of order, complementary
modes of production and a politics of external renown. On the basis of his analysis of
the kula, Malinowski bravely prophesized the coming of a new era of anthropological
understanding devoted to the “influence on one another of various aspects of an institution,” together with “the study of the social and psychological mechanism on which
the institution is based” (1961 [1922]: 516).1
But Malinowski’s functionalism was hardly the only program of anthropology-in-onecountry. Durkheimian sociology, British structural-functionalism, French structuralism, White’s and Steward’s evolutionisms, Marxism of base and superstructure,
cultural ecology, cultural materialism, and even poststructuralist epistemes, discourses,
and subjectivities: all these paradigms assumed that the cultural forms, relations, or
configurations they were explicating were within a more or less coherent order, and
that the articulations and dynamics of that order were the theoretical matters at issue.
The diffusionism of early-twentieth-century American anthropology was an exception, but it was largely recuperated by the patterns of culture of Benedict et al.
9781444333237_4_007.indd 102
8/7/2010 6:45:14 PM
The Whole is a Part
103
Otherwise the main anthropological theories fairly ignored that the cultures they took
to be singular and distinct were involved in relations of alterity as a condition of their
internal consistency, their status as some sort of entity, and their identity.
“Human Societies are Never Alone”
This phrase is from Lévi-Strauss’ Race and History, now more than half a century
old, a text that has been farmed for a number of seminal ideas, but not fruitfully
enough for the observation that the diversity of cultures “is less a function of the
isolation of groups than of the relationships which unite them” (1952: 9). It was a
bold move in a time when “cultural contact” virtually meant diffusion – not to mention the going “despondency theory,” the lugubrious sense of the disappearance of
anthropology’s others by domination and acculturation. The prevailing wisdom was
that intercultural contacts bred homogeneity, or at least syncretism (aka “hybridity”).
Lévi-Strauss, however, was launching a long career of structural studies, particularly
in mythology, based on the contrary principle that the cultural differences of interacting peoples were dialectically related. Each would be itself by not being the other,
that is, by exploiting the possibilities of opposing, transforming, and/or transcending the order presented by neighboring peoples.2 This dialectic has a celebrated history, beginning perhaps with the Aristotelian dictum that contraries are the sources
of their contraries – stated, interestingly enough, in the Politics. And if you are
reminded of its Batesonian version under the description of “schismogenesis,” I will
try to show you have good reason.
Again, homage should be paid to the work of Fredrik Barth and colleagues on Ethnic
Groups and Boundaries, published in 1969. Although this book also received much attention, it seems that postmodern wallowings in indeterminacy have obscured its central
thesis to the effect that the creation of ethnic identities cum cultural distinctions entails the
interaction of peoples in “embracing social systems.” As Barth amply demonstrated for
Pathans and their neighbors, “ethnic distinctions do not depend on an absence of social
interaction and acceptance, but are, quite the contrary, often the very foundations on
which embracing social systems are built” (1969: 12). Even the recent victims of postmodern terror, such as cultural boundaries and essentialized traditions, were to be understood as systematic effects of intercultural relations, rather than prematurely dismissed as
anthropological illusions. Then again, as Karl Izikowitz argued in the same volume, the
very concept of cultures as separate and sui generis, hence their analytic treatment as isolates, is at least in part due to “the influence of the old national romantic movement which
sought to give prominence to each people’s national characteristics and particular system
of values” (1968: 135).
The accusation obviously pointed to Hamann, J.G. von Herder, and their successors in the Germanic tradition of “culture,” and more precisely of “national cultures,”
which under the influence of Franz Boas and colleagues became dominant in American
anthropology and thence broadly influential. Paradoxically, the concept of culture as
an autonomous whole was predicated on its historical position as a heteronomous
part, for it was produced in and as the very opposition of Germanic Kultur to French
civilisation. The anthropological notion of sui generis cultures was itself generated in
an eighteenth-century historical relationship with cultural others. But where sui
generis is an oppositional value, it is not sui generis.
9781444333237_4_007.indd 103
8/7/2010 6:45:14 PM
104
Sahlins
“National Cultures, Where are You?”
Ye men of all quarters of the Globe, who have perished in the lapse of ages, ye have
not lived and enriched the soil with your ashes that at the end of time your posterity
should be made happy by European civilization.
– J. G. von Herder, Outlines of a Philosophy of the History of Man (1803)3
In fundamental respects the Germans of the late eighteenth century and the indigenous peoples around the world who were suffering an oncoming Western European
imperialism, notably British and French imperialism, shared a common historical
situation. This helps explain Herder’s defense of the “cultures” of the one and the
others: all of them worthy of respect as autonomously fashioned, uniquely organized, and inherently valuable expressions of the human spirit. (How many times
must it be repeated that the anthropological concept of “culture” had its source at
least as much in resistance to colonialism as in its service?) The last thing we should
do, said Herder, is to take European culture as a universal standard, an achievement
devoutly to be desired for all humankind; only a real misanthrope could wish this
fate on the rest of the species. Each nation, bearing within itself the standard of its
own perfection, develops its own cultural character, independently of and without
comparison to any other.
Indeed, absent a united national state, what else besides sharing a common culture
could it mean to be German at this time (Elias 1978: 6)? Relatively backward economically as well as politically, their boundaries vulnerable and their unity questionable, the Germans could only oppose in cultural terms the civilisation that the lumières
posed as a measure for all humankind. Their culture was all the more precious insofar
as the alternative conceits of civility and perfectibility gave the French and British
ground to find others wanting on a scale that declined into the vilified degrees of “barbarism” and “savagery.” The German bourgeois intelligentsia in particular were under
a double threat of French civilisation: not only from the power of France itself but also
from their own ruling aristocracy, whose francophone and francophile tastes were too
evident proof of that power. It is not for nothing that Herder saw the state as an external imposition, foreign to the inherited cultural tradition conveyed in the authentic
language of the Volk. Hence the common plight with the Völker everywhere:
The princes speak French, and soon everybody will follow their example; and then,
behold, perfect bliss: t\he golden age, when the world will speak one tongue, one universal language, is dawning again. National cultures, where are you? (Herder 1969: 209)
Well, if anywhere, they were destined to be alive and well in the anthropology of
the twentieth century. The anthropological idea of “a culture” was largely distilled
from this history of complementary opposition to a civilizing process that threatened
to make good on its pretensions to universality. The anthropological genealogy from
Herder to Boas has been traced by others, and I will not rehearse it here. Suffice it to
say that the binary contrasts drawn by the Germans in ideological combat with
Franco-British “civilization” became predicates of “a culture” as a scientific object.
Perhaps most significantly, this included the idea that culture was generated from
within, in contrast to the civilization the French and British were laying on people.
9781444333237_4_007.indd 104
8/7/2010 6:45:14 PM
The Whole is a Part
105
Culture was the essence of a people’s being; civilization was a process of becoming.
Civilization was a matter of degree, whereas culture was a difference in kind. With its
own center of gravity, its own standards of value, and its own direction of development, the culture of a people defined their individuality and identity, whereas civilization presented a cross-cultural measure of sophistication. Think of the echt Germans
and the supercilious French, hence all those things that make culture something
essential and civilization superficial: inner being, honesty, true nature, true appreciation, and indeed truth as such, as opposed to outward show, dissemblance, mere
politesse, fashion, and humbug.
To repeat the paradox: the idea of a culture as an anthropological object constituted
by and for itself developed out of a relationship between European cultural orders –
and, more specifically, out of the interdependence of their differences. Indeed, for all
the Herderian insistence on cultures growing from within and below, the actual formation of natural cultures in dialogue with cultural others did not escape observation
in the German anthropological tradition. Still, if there is one thing I learned from
Althusser, it is that recognizing something is not the same as knowing its correct theoretical place. Given the parochialism of the Germanic anthropology, an observation to
the contrary, even by the likes of a Wilhelm von Humboldt, would have limited theoretical effect:
How can one possibly know completely a nation’s character without having also studied
other nations with which it is in close relation? It is in contrast with them that this character actually came into being and it is only through this fact that it can be fully comprehended. (In Dumont 1994: 120)
Actually Existing Cultures
The prevailing tradition of separate and sui generis anthropology cultures notwithstanding, the actually existing cultures at the time of European colonization were
joined in mutual relations of determination with neighboring societies. These relations included assimilation as well as differentiation, and often both at once, as in
the German case of an opposition (to France) founded on acculturation. Yet beyond
relations to proximate others, specifiable cultures the world around were situated in
wider dynamic fields of mutual influence: regional ecumenes of the kind once
quaintly known (and now largely forgotten) as “culture areas.” Dynamism was
inherent in the ordering of these regions by gradients of cultural authority, power,
and value emanating from one or more strategically situated centers. Although our
current social sciences are pleased to discover core–periphery configurations in the
imperial ambitions of modern capitalist metropoles, most of the planet had already
been mapped for ages in culture areas of that hierarchical form – involving also
invidious distinctions of cultural sophistication.
In Kroeber’s classic treatise, the culture areas of native North America were largely
constituted by influences radiating from certain centers of “cultural climax”; hence
their boundaries, although intergrading, represented “something comparable to political spheres of influence” (1947: 6). Kroeber could make that claim even for relatively
egalitarian regions and “tribally organized” centers such as those of the Hopi Pueblos
in the American Southwest or the Haida and Tsimshian on the Northwest Coast.
9781444333237_4_007.indd 105
8/7/2010 6:45:14 PM
106
Sahlins
Similar observations have been made of locally dominant Melanesian peoples: among
others, the Tolai, Orokaiva, Iatmul, and Marind-Anim. Writing of the Sepik area dominated by Iatmul, Simon Harrison (following Deborah Gewertz) argues that “societies
of the middle Sepik can only be analyzed adequately as a regional political-economic
system in which, intentionally or otherwise, they shaped each others’ histories” (1990:
18; see Gewertz 1983). Of the Manambu people in particular, Harrison observes that
they “seem to have imported throughout their history very many elements of Iatmul
culture, particularly ritual, magic, totemism and myth. To the Manambu, the cultural
forms of the Iatmul are surrounded by an aura of especially dangerous power, and
therefore valuable to acquire” (Harrison 1990: 20). Moreover, a similar hierarchy
obtains among Manambu villages, with the most powerful of which, Avatip, having “a
kind of metropolitan status among them” (17). (Nonstate galactic polities, anyone?)
Regarding the Marind-Anim as famously described by J. Van Baal (1966), no
doubt their large population and their headhunting prowess have similarly carried
some cultural conviction to neighboring groups, but coercion is not the only reason
the latter have been inclined to adapt their own cosmologies and ideas to the Marinds’.
To trespass on Geertz’s conception of Balinese kingship, there is also an attractive
theatricality in the Marinds’ power: spectacular rituals, displays of wealth, and artistic
achievements that spread their authority as demonstration effects. Note, however,
the reciprocal movement in the cultural circulation. For their part, the Marind incorporate the wild things of the hinterland peoples, including myths, cults, heads, living
persons (women and children), and names for their offspring, even as these peoples
gain access to the sophisticated products of the Marinds – which likewise embody
agentive powers useful in their own local politics. And in speaking of outlying peoples who become Marind, are we not reminded of Kachin Hills leaders who “become
Shan” (Leach 1954: 223)?
These intercultural relations of order and change are amplified, however, when they
are politically organized by royal centers holding greater or lesser sway over other
ethnic groups in their hinterlands. Many areas of Asia, Africa, and the Americas were
so configured at the time of European expansion, as was much of the European continent in times before. The world knew many kingships of universal pretensions that
generated the kind of “galactic polities” described by Stanley Tambiah (1976, 1985)
for Southeast Asia. Here were graded systems of power and cultural authority focused
on dominant apical states from which the others derived much of their own political
form, if in reduced versions in proportion to distance from the center, and often
enough their own ruling dynasties of stranger-kings – while retaining a certain ethnic
distinctiveness. Common the world around were political genealogies tracing the
descent of outlying ruling groups to ancestral names to conjure with, some of them
purely legendary: Teotihuacan, Toltec, Sina (China), Kahiki (Polynesia), Kachaw
(Micronesia), Rum (Rome, Turkey, or Persia), Aeneus, Ashoka, Alexander, Ife, Luo,
Lunda, Tikal, Troy, Mycenae, Zande, Majapahit, Bunyoro, Kalinga (India), Kitara,
Jukun, and so on, and so forth – although, as Janet Hoskins remarked of the similar
claims of certain eastern Indonesian aristocrats, “it could hardly be that all the peoples
of the outer islands were descended from exiled Javanese princes” (1993: 35).
My object here is to “get real” about anthropology-cultures by studying them in
the greater fields of cultural value and historical change by which they are actually
generated. In particular, I note some of these processes of order in the overlapping
realms of East Asia, Southeast Asia, and Oceania. But first, some high-flying talk
about the cultural politics of alterity.
9781444333237_4_007.indd 106
8/7/2010 6:45:14 PM
The Whole is a Part
107
The Cultural Politics of Alterity: Depending
on the Kingness of Strangers
What seems truly internal about a cultural existence is the necessity of incorporating
the external. Recent archaeology has shown that even the scattered Polynesian islands
are not as isolated from one another as had long been assumed: for example, the discovery of Tuamotu adzes made from Marquesan and Hawaiian basalt, the latter source
over 4000 kilometers distant (Collerson and Weisler 2007). In any event, by their
own lights the Polynesians are situated historically and spatially in a world of other
islands beyond the horizon or in the sky, and the evidence is there in the growth of
their crops, the passage of the seasons, and the birth of their children. Homeland of
great gods and chiefly ancestors, these celestial islands are conditions of the possibility
of people’s well-being. They prove that if external life-giving and life-taking powers
did not exist, they would have to be invented.
I take Viveiros de Castro’s point that “[i]f humans were immortal, perhaps society
would be confounded with the cosmos. Since death exists, it is necessary for society
to be linked with something that is outside itself – and that it be linked socially to this
exterior” (1995: 190). Ranging from beasts, spirits, and gods to ineffable forces, by
the way of the generic dead or the ancestors and of other peoples with their remarkable gifts, the extraordinary agents that control the human fate live outside the space
of human control. More precisely, the lack of control translates as being-in-otherspace. I am speaking of the so-called and unfortunately named supernatural.
Misleadingly named, I believe, because the term supposes ethnocentric concepts of
“nature” and “natural” – an autonomous world of soulless material things, of Cartesian
res extensa – that are not pertinent to peoples who are engaged in a cosmic society of
interacting subjects, including a variety of nonhuman beings with consciousness, soul,
intentionality, and other qualities of human persons.
Admittedly my notions of the so-called supernatural rest on old-fashioned premises.
I take the rather banal, positivist, and Malinowskian view that people must in reality
depend for their existence on external conditions not of their own making – hence and
whence the spirits. (Recall Malinowski’s observations on the integration of “magic” in
Trobriand activities of gardening, voyaging, or fishing when the outcome depends on
conditions beyond human control [1948: 10 ff].) The going anthropological alternatives argue that divinity is some misrecognition of humanity. For Durkheim (1947
[1914]), god is the misplaced apprehension of the power of society, a power people
surely experience but know not wherefrom it comes. For a certain Marxist anthropology, god is an alienated projection of people’s own powers of production and reproduction, an unhappy consciousness that has transferred human self-fashioning to the
deity (e.g., Sangren 1990, 1991). Such theories may address the morphology of divinity, whether as projection or mystification, but they do not tell us why society is set in
a cosmos of beings invested with powers of vitality and mortality beyond any that
humans themselves know or control, produce, or reproduce. Neither sense of false
consciousness takes sufficient account of the generic predicament of the human condition: this dependence on sui generis forces of life and death, forces not created by
human science or governed by human intentionality. If people really were in control of
their own existence, they would not die. Or fall ill. Nor do they control the biology of
sexual or agricultural generation. Or the weather on which their prosperity depends.
9781444333237_4_007.indd 107
8/7/2010 6:45:14 PM
108
Sahlins
Or, notably, the other peoples of their ken: peoples whose cultural existence may be
enviable or scandalous to them; but in either case, by the very difference from themselves, strangers who thus offer proof of a transcendent capacity for life. It is as if nothing foreign were merely human to them. Endowed with transcendent powers of life
and death, the foreign becomes an ambiguous object of desire and danger. Hence the
ubiquity – and ambiguity – of the aforementioned stranger-king formations.4
If I called the following notes on the distribution and character of stranger-kingship
around the world an “ideal-typical description,” would I be allowed some Weberian
immunity from the charge of essentializing? In any case I generalize – surely at the risk
of oversimplifying.
The rulers of a remarkable number of premodern kingdoms and chiefdoms around
the world have been strangers to the places and peoples they rule. By their dynastic
origins and their inherited nature, as rehearsed in ongoing traditions and enacted in
royal rituals, they are foreigners – who on that ground must concede certain powers
and privileges to the native people. Such immigrant dynasties have been widespread
since early times in Southeast Asia and the Pacific Islands, including the Cambodian
kingdoms of Brahmin ancestry, the Siamese Ayudhyan dynasty founded by a Chinese
merchant-prince, the dominant Balinese dynasty of Klungkung that originated in a
Javanese warrior-prince, the Hawaiian ruling chiefs whose ancestors arrived from legendary Kahiki, the Malay sultans descended from Arabian sayyids or Alexander the
Great, and the Sinhalese realm founded by an errant prince descended from a lion in
India. The like is to be found on the lowland peripheries of the major eastern Asian
empires and kingdoms, as among the Shan, Yi, or Malay principalities; and then again,
on their own peripheries, among the upland tribal peoples whose leaders have been
known to adopt the cultural trappings of dominant lowland societies. Take note for
further reference that native rulers can become stranger-kings, just as stranger-kings
may become native rulers.
The historical Alexander the Great of Macedonia was a stranger-king in his own
right, as we shall see, as were many other ancient Indo-European rulers from Ireland
to India (Sahlins 1985, in press). From Celts to Aryans, one encounters similar narrative charters of immigrant rulers constituting a dualistic polity with their underlying
indigenous or autochthonous subjects. Dynastic founders of Peloponnesus kingships
from the eponymous Pelops and Lacedaemon to Agamemnon and the Heraclid rulers
of Sparta were all Zeus-descended strangers who gained their realms through marriage with the daughters of their native predecessors. Pelops had notoriously conspired with the King of Pisa’s daughter to cause her father’s death, upon which he
married her and succeeded him, a tale of violence and cunning intelligence not unlike
the assumption of rule over the native Latins by Aeneas the Trojan or the subjection
of the Sabines by the fratricidal Romulus. Commenting on many of these same traditions, J.-G. Préaux has written, “Every foundation of a city, every conquest of royal
power becomes effective the moment the stranger, charged with sacredness by the
gods or the fates, endowed moreover with the force of the warrior, symbolically gains
possession of a new land either by receiving peacefully, or by conquering valorously or
by ruse, the daughter of the king of the land” (Préaux 1962: 117).5
In the manner of their great god Lugus, an archetypal stranger-king as Krista Ovist
(2004) has demonstrated, the Celtic societies of the Continent and the Islands were
dominated by kingly clans of foreign derivation, military orientation, and aristocratic
pretension. Ovist speaks of “the self-understanding of Gaulish societies as groups constituted by autochthonous and foreign elements, the former regarded as feminine and
9781444333237_4_007.indd 108
8/7/2010 6:45:14 PM
The Whole is a Part
109
the latter as masculine” (382). Indeed, a set of traditions traces the origin of the Gauls
to Hercules, who in the course of one of his exploits fathered the heroic ancestor of the
people by the daughter of a native Acquitanian king; or in another version, he established the several Gaulish groups by unions with a number of indigenous noblewomen.
In a more historical vein, Britain has been the site of stranger-king formations from the
Norman conquest all the way down (like certain famous turtles) – through the Angles,
Saxons, Romans, and probably immigrant Brythonic Celts. Likewise, the Welsh knew
Irish and British ruling houses; and the Highland Scots were constituted by Irish princes
who married the daughters of native Pictish rulers. In Ireland itself, the kings were ritually married to their realms, which is to say the goddess or princess thereof, a practice
that continued in modified form until the nineteenth century. The epic Irish Book of
Invasions tells of a series of conquests from abroad culminating in the advent of the
dominant Goidelic stock, the Sons of Mil, who purportedly came from Scythia via
Egypt and Spain (Lebor Gabala Erren c. 1150 AD). These Milesians had already learned
to unite with the daughters of local kings during their legendary peregrinations, including a marriage with the eponymous Scota, the pharaoh’s daughter. By the Middle Ages
they were able to integrate other stranger-rulers of Irish kingdoms in the Goidelic narrative and genealogy, incorporating aristocratic clans such as the Eoganachta of Munster,
who had likewise dominated the original native peoples (Eriann and Cruithin).
To continue this rapid tour du monde:
Africa has also been the site of numerous dualistic polities consisting of indigenous
or autochthonous “owners” of the land and stranger-rulers of different ethnic origins,
broad cosmic powers, and notably wild dispositions. Referring generally to West and
Central Africa, Luc de Heusch writes, “Everything happens as if the very structure of
a lineage-based society is not capable of engendering dialectical development on the
political plane without the intervention of a new political structure. The sovereignty,
the magical source of power, always comes from elsewhere, from a claimed original
place, exterior to society” (1982b: 26; see also Heusch 1962, 1982a). The description
is easily extendable to East African kingdoms, such as those dominated by immigrant
Nilotic groups. Well-known examples from around the continent include Alur, Benin,
Shilluk, Nupe, Mossi, Kongo, Luba, Lunda, Zande, Ruwanda, and others, not to
mention the many lesser kingdoms and chiefdoms that are effectively satellites in
galactic orders centered on the greater ones – including the Tallensi.
The major American empires of the Aztecs and Inca were ruled by kings of foreign
derivation. The classic period Mayan dynasties of Tikal and Copan were initiated by
“the arrival of strangers,” as the local inscriptions relate, and similar kingly origins
were known in postclassic cities of the Quiché as well as in Mayapan and Chichen Itza,
among others (e.g., Sabloff 2003; Sharer and Traxler 2006; Stuart 2000).
In a common permutation of stranger-kingship – known in Japan, Korea, Polynesia,
Micronesia, Natchez, and other locations – the founding prince descends from the
heavens, always a good address for outsiders of royal pretensions. Then again, as was
just noted regarding Ireland and Britain and is also commonly observed in Indonesian
and African societies, the coming of strangers is a recursive formation, with one foreign kingship succeeding another without entirely erasing the tradition of the previous dynasty, so the polity is in effect a multiethnic palimpsest of temporally and
politically stratified groups. Still another variation involves the return of the ancestorcum-stranger: an immigrant hero claiming derivation from original chiefs or autochthonous gods who some time ago had left the land, whether voluntarily or by reason
of banishment or usurpation. Indeed, this topos of the ancestral stranger has often
9781444333237_4_007.indd 109
8/7/2010 6:45:14 PM
110
Sahlins
been adapted to the construal of colonizing white men. Before they were cruelly disabused of the idea, many indigenous peoples took Europeans to be long-lost sons of
their own progenitors, if not lovers of their main goddess or forms of their original
god – like Captain Cook, for example.
Summarizing – and still risking: in these stranger-kingships, two forms of authority
and legitimacy coexist in a state of mutual dependence and reciprocal incorporation.
The native people and the foreign rulers claim precedence on different grounds. For
the underlying people, it is the founder-principle: the right of first occupancy, and in
the maximal and common case, the claim of autochthony. Former rulers of the land
and still its “owners” – as they are known in many Austronesian as well as African
societies – the aboriginal people belong to the soil as much as it belongs to them.
Hence within the dualistic polity, they not only are the main subsistence producers
but also detain a privileged spiritual relation to the land and its products. (As Sather
[1990: 30] says of the Borneo Iban, rice is the transubstantiation of the ancestors; or
as Harrison [1990: 63] says of Manambu, their own work and knowledge
notwithstanding, they do not create the crops, they receive them.) People of the
Central African Luapula Valley put the relation of the original people and their rulers
this way: every land has two “owners,” they say, “the original settler, who is owner
through the fact of being first, and therefore has ritual authority, and the political
‘owner’ there by right of might or cunning or duplicity; and each respects the other
for the particular attributes for which he is owner” (Cunnison 1951: 15).
Still, the stranger-kings trump the claims of the native people, characteristically by
aggressive and transgressive demonstrations of superior might, allowing them to gain
the sovereignty. Peaceful versions of the takeover may be recited for politically interested purposes, but as the royal violence of the stranger-hero often has an ongoing
function, so it also has an enduring narrative and ritual presence. Commonly in African,
Indo-European, and Southeast Asian traditions, the immigrant hero is the son of a
powerful king who fails to obtain the succession in his native land, perhaps for some
fault that results in his exile. Even more generally reported, the advent of the stranger
on the local scene is marked by feats of incest, murder, or other crimes against kinship
and morality. This “exploit,” as Luc de Heusch calls it, by its transcendence of the local
native society, is a sign of the foreigner’s power to organize it (Heusch 1958, 1962).
But then, in the ensuing time, the stranger-rulers are domesticated by the indigenous
people, thus sublimating their terrible powers in the general interest. The ritual means of
this transformation are typically rehearsed at the installations of kings of the foreignderived dynasty. They range from the privilege of leaders of the native people to inaugurate the foreign-derived pretender, through various rites of humiliation inflicted on the
new king, to sham battles or other symbolic enactments of the death of the prince as an
outsider and his rebirth as ruler of the people. The stranger is indigenized – indeed, the
foreign dynasty is often linguistically and culturally assimilated by the native people, without however erasing its external origins and powers. In the event, the royal violence of the
immigrant ruler is redefined as the gift of victory, to be turned outward toward expansion
of the realm; while internally, the king initiates a kind of mission civilisatrice, bestowing
benefits on the native people that lift them from a more rudimentary state. There is thus
a reciprocal incorporation of the people by the king and the king by the people. Accordingly,
the alien character of rule is not simply understood as the estrangement of power, the
projection of its antisocial nature to an external source – “society against the state,” to
reprise Pierre Clastres’ well-known formula – since the foreign derivation of the kingship,
transmitting the powers of alterity, has in itself a positive domestic value.6
9781444333237_4_007.indd 110
8/7/2010 6:45:14 PM
The Whole is a Part
111
Nor is victory the only royal gift. Sacralized or divinized, the stranger-king is a rainmaker, both in the Frazerian sense that he fertilizes the bearing earth of the aboriginal
people and in the current colloquial sense that, by contrast to their identification with
the land, he is the source of the society’s artifactual, exchangeable, and spiritual riches,
its mobile and transmissible wealth, including the life-enhancing foreign valuables
distributed in kingly largesse or consecrated as royal regalia. It is not for nothing that
a number of Southeast Asian kingdoms were founded by merchant princes: as it is said
in the Annals of the Melakan sultanate, “Where there is sovereignty there is gold”
(Brown 1952: 187). Providing foreign wealth and fertilizing the land are parallel
sovereign functions, insofar as both convey the vitality of alterity and comprise a necessary complement of active means for realizing the fixed earthly powers of the indigenous people. The material counterpart on the part of the underlying people is the
tribute owed the stranger-king: typically produce of the earth, given in requital for the
cultural riches and transcendent reproductive powers he conveys to them.
This mutually beneficent conjunction is almost invariably sealed by the union of the
immigrant prince with a daughter of the indigenous ruler, thus making the formation
of the polity a mix of contract and conflict. Characteristically, then, there is some
continuing tension between the foreign-derived royals and the native people.
Invidious disagreements about legitimacy and superiority may surface in partisan renderings of the founding narratives. More than political, however, the conjunction
here is cosmological, which is what helps it endure. The foreign rulers are to native
people in some such encompassing relation as the Celestial is to the Terrestrial, the Sea
to the Land, the Wild to the Settled; or in abstract terms, as the Universal is to the Particular, a ratio that also holds for their respective gods. We see, then, why the narratives
of the advent of the stranger-hero function as all-around cultural constitutions. The
union with the other, which is also an elemental combination of an external masculinity and an internal femininity, gives rise to the society as a self-producing totality – and
the permanent contradiction that this autonomy is dependent on alterity.
Elementary Forms of the Political Life
Yet the stranger-king is hardly the only form of the cultural politics of otherness. The
same configuring of power, the same project of prospering the local society by incorporating potent foreign agencies, with the analogous effect of enhancing the social value
of those who accomplish the feat, describes a variety of border-transcending exploits
ranging from headhunting, cannibalism, and other modes of predation, through trading and raiding for foreign valuables, to vision quests, shamanism, and other such means
of incorporating and domesticating the spiritual virtues of external subjects.
Consider the headhunting of Southeast Asian hinterland peoples, for example.
Here the structural resemblances to stranger-kingship are right on top. Indeed, de
Josselin de Jong explicitly likened the initiation rites of Toradja warriors of Sulawesi –
which entail bringing home an enemy head – to the foundation legend of the Negri
Sembilan kingdom of Malaya by a border-crossing, enemy-overcoming Minangkabau
prince from Sumatra (de Josselin de Jong 1975; cf. Downs 1955). Janet Hoskins
(1996: 237) relates an East Sumba (Indonesia) myth which makes headhunting a civilizing mission wherein the hero, by planting the head of a brother who had incestuously
married his sister on the grave of the latter, thereby instituted proper marital rules of
9781444333237_4_007.indd 111
8/7/2010 6:45:15 PM
112
Sahlins
exogamy, an ordered society of nobles and commoners, and the means of wet-rice
cultivation. As famously reported for many Southeast Asian hinterland peoples –
among them, Ifugao, Iban, Kayan, Buaya, Naga, Kenyah, East and West Sumba, and
so on – the powers of the victims whose heads are ritually domesticated in sacrifices
and head-feasts are thus turned to sustaining the lives and livelihood of the victors
(McKinley 1976). The heads might then be as ceremonially honored as they were
initially reviled. By such means, Kayan people say that “those who were once our
enemies thereby become our guardians, our friends, our benefactors” – with the benefits including bountiful harvests and immunity to illness. At the end of the Ifugao
head-feast, the quondam enemy is enjoined to combat sickness, sorcery, famine, evil
gods, and the Ifugao’s own enemies – “For you,” they say, “have become one of us”
(McKinley 1976: 115). Suggestively (of stranger-kingship), the name “Ifugao” means
“earth dweller,” and like other inland Southeast Asian groups the Ifugao live at the
center of a cosmos that includes other peoples, upstream and downstream, who are
something less and more than human. They are less because they are beyond the
Ifugao pale, culturally as well as spatially: they are orang bukai, meaning approximately “people who are not people.” Yet they are more than ordinary humans because
they are closer to the gods of the still more distant celestial and chthonian realms.
Ideally, it is in these far worlds of gods that heads should be sought, as being the most
potent. But in practice heads are taken from the enemy “people who are not people”
on the principle that they “are similar to spirits residing in strange and wild places.”
In a Toradja tradition of the origin of the head-feast, the hero indeed undertakes an
arduous trip to the Upperworld to exact revenge on the killer of his parents, and then
descends to the Underworld to take the heads of his victims’ ghosts (Downs 1955).
Corollary Toradja narratives tell that the village of the headhunter had been dead during his absence but revives upon his triumphal return, and that the hero also brings
home the magical daughter of his victim and marries her after the head-feast – thus
transforming enemies into affines and marking their equivalence as reproductive
agents. In sum, the Toradja warrior returns from a cosmic exploit with a foreign subject (the head) and enhanced reproductive virtue (the wife) in order to give life to
(revive) the whole society. Allowance made for the inversion of stranger-king formations – the local hero who captures foreign power as opposed to the foreign prince
whose power is captured locally – here is another modality of the same relationships.
By an analogous principle, the Iban warriors of Sarawak could not marry until they
had taken a head, although nineteenth-century accounts already indicate that amassing heirloom wealth in the form of large Chinese jars acquired in journeys abroad –
the well-known Iban bejalai – was an acceptable alternative (Low 1848: 215; Keppel
1848: 1, 35). For like heads, these valuables had the agentive capacities of potent
foreign subjects (as Marcel Mauss might have taught). “Many of the goods acquired
through bejalai,” explains the historian John Walker, “were themselves sources of
potency. Antique jars, for example, were credited with supernatural powers and healing virtues and would thereby contribute to the potency of the community to which
they were taken. Moreover, the successful accumulation of prestige goods and other
wealth would indicate, in itself, an increase in spiritual powers, status and strength”
(2002: 20). But then, given these practices of acquiring the potency of alterity in the
form of heads, valuables, and their own persons (through bejalai experience), it is not
surprising that the Iban knew how to welcome stranger-kings, fiercely egalitarian as
they were themselves: not only immigrant Malay aristocrats but also notably James
Brooke, the famous “White Rajah of Sarawak.”
9781444333237_4_007.indd 112
8/7/2010 6:45:15 PM
The Whole is a Part
113
For the Iban and other indigenous peoples of Sarawak, James Brooke was endowed
with an extraordinary soul power – semangat was the Malay term – which made him
capable of mana-like effects on their fertility and mortality (see Keppel 1848; Low
1848; McDougall 1854; Mundy 1848; Pringle 1970; St. John 1879; Walker 1998,
2002). Brooke found it “highly gratifying” when the leaders of Dayak communities
came to Kuching to obtain some of his body fluids, like the Bidayuh headman who
“brought me a young cocoa-nut for me to spit into, as usual: and after receiving a little
gold dust and white cloth returned home to cultivate his fields” (in Walker 2002: 116).
On the occasion of a visit to a Bidayuh longhouse by Brooke and the Anglican Bishop
McDougall, the people begged the Englishmen to spit on some cooked rice, whereupon the Bidayuh ate the rice, “thinking they would be the better for it” (McDougall
1854: 57). Brooke noted that as a matter of course when he visited them, Dayak people
would wash his hands and feet, and afterward sprinkle the bathwater on their houses
and gardens. Likewise, for the gold and white cloth he presented – gifts they particularly desired of him – these too they planted in their gardens (Mundy 1848: 2:43).
Many of Brooke’s European officials and companions were similarly treated, sometimes
with rites that invoked the semangat of Brooke along with their own ancestors.
Among other classic features of stranger-kingship, note in these descriptions that
“where there is sovereignty, there is gold.” Rajah Brooke & Co. are solicited for the
mobile wealth that potentiates the reproductive capacities of the native people – such as
the gold dust that could be showered like rain on the swidden fields of the indigenous
owners. (Not that the exchange value of gold in the marketplace, of which the Borneo
people were probably well aware, works in ways any less mysterious.) Implying direct
contact with Brooke’s person, the gifts of cloth would have the same fertilizing virtue.
The White Rajah was also known to present Malay officials and Dayak leaders with costumes of office, sometimes including decorated loincloths. The nakedness of the native
rulers thus encompassed and sublimated by the foreign king, they were at once civilized
and empowered. In a related cosmic register, that of the founding union of the strangerking with the daughter of his native predecessor, rumors circulated about the liaisons of
Rajah Brooke with important Malay and Dayak women. According to one (unconfirmed) report, Brooke married the niece (brother’s daughter) of the Malay noble who
originally ceded Sarawak to him (Brown 1972). The Iban had even less verifiable and
more marvelous stories of Brooke’s relations to their own cosmogonic deities. Some
comprehended him by the topos of the returning ancestral-stranger, saying he was the
son of their primordial mother and father, Kumang and Keling. Others said Brooke was
not Kumang’s son but her lover, and that he was in the habit of climbing Mt. Santugong
to make love to her – a magnified feat of union with the ranking woman of the people.
Finally, then, is not marriage itself the ordinary and extraordinary representation of
the vital powers of alterity, the gains and losses of reproductive powers that are functions of the incest taboo in the form of spouse taking and spouse giving?7 Inasmuch as
the incest taboo renders the self-organized socius incomplete of necessity, marriage
becomes the experiential social archetype of life-from-without. As a relation of insiders
to outsiders, roughly speaking consanguines to affines, producing offspring who unite
these endogenous and exogenous complements, marriage is a manifest quotidian form
of the same generation of vitality by the integration of external subjects that we have
been discussing in other, mainly political, modalities. Opposing the potencies and gifts
of spouse givers or spouse takers to the constituted authority of elders or ancestors,
involving the negotiated appropriation of the beneficial virtues of potentially dangerous
others, the relations of affines to consanguines amount to something like
9781444333237_4_007.indd 113
8/7/2010 6:45:15 PM
114
Sahlins
stranger-kingship in nuce. The marital alliance of insiders with outsiders is similarly a
synthesis of being and becoming, while the accompanying transactions in gifts and
regards likewise confer value and identity on the issue of the union (the mission civilisatrice). Also like the life-and-death potency of kingship are the powers of blessing and
cursing that may be detained by key affinal relatives. Recall Edmund Leach’s famous
distinction between the we-group of consanguines, defined as relations of common substance, and relations of alliance that “are viewed as metaphysical influence” (1961: 21).
Galactic Dynamics of Culture in East Asia
In his well-known Political Systems of Highland Burma (1954), Leach documented the
aspirations of certain Kachin chiefs to “become Shan,” so it is not surprising that their
houses are spoken of in the same terms as Shan palaces. But how did the establishments
of these hinterland chiefs come to be styled “the cave of the alligator,” referring to a fabulous monster-ancestor whose symbolic associations bear “striking resemblance” to the
kingly Chinese dragon, and how did they come to function like the Temple of Heaven in
the Chinese imperial cult (Leach 1954: 112–13)? Leach makes a considerable case for the
ritual and architectural parallels. The site of important ceremonies undertaken by the
Celestial Emperor for the welfare of the state and the fertility of the crops, the Temple of
Heaven consisted of two parts: a domed building dedicated to Heaven and an open circular platform dedicated to Earth. Just so, in gumsa Kachin, the chief makes offerings to
the principal sky spirit (Madai), a being who ensures the general welfare, at a shrine in his
own roofed house. And he makes offerings to the earth spirit, who controls the fertility
of the soil and of people, at a circular open space at the entrance to the village. Normally
bounded by stones, this site of the earth spirit in a few prosperous villages “becomes a
stone platform decorated in Chinese style.” Hence Leach’s conclusion that the Kachin
chief’s ritual role “has a definitely Chinese flavor,” and his house, more than a dwelling
or a palace, is a kind of temple dedicated to the great celestial deity.
No doubt it would reduce the enigma to know the detailed history of overland trade
routes between China and Assam – which I have not investigated. But something
should be said for the mediated relations of the Kachin Hills to Chinese dynasts at least
since the eighth century AD. The Kachin were more or less subject to neighboring
states – occasionally much less, they were militarily dominant – that were themselves
tributary to China, beginning with the Nan-chao Kingdom of Yunnan during the
T’ang dynasty and continuing into recent times with the Tai-speaking Shans and the
Burmese kingdom. (At an earlier period, the predecessors of the Nan-chao kingdom
traced their ancestry to a purported descendant of the great Indian king Ashoka.) The
tributary relations and cultural assimilations of ruling groups that marked the Chinese,
Burmese, and Indian impact on major lowland states of Southeast Asia were replicated
again in the interactions of the latter with hinterland peoples such as the Kachin. As
noted earlier, we have to do with extensive galactic systems involving potent cultural
influences radiating out from their “civilizing” centers, with reciprocal, attractive
effects on the peoples and products of the “barbarian” peripheries – whence developed
a certain dialectics of assimilation and differentiation (Tambiah 1976, 1985, 1987).
The apical states of the galaxies were universal kingdoms whose rulers conflated
their polis with the cosmos and their hegemony with world order – cosmocrats, if
you will. Such were the Buddhist Chakkavatti King of Kings, the Devaraja Hindu
9781444333237_4_007.indd 114
8/7/2010 6:45:15 PM
The Whole is a Part
115
monarchs, as well as the Chinese ruler of All Under Heaven. (In Southeast Asian
practice, however, Indianized states such as Funan and Chakkavati kings as those
of Ayuddhya, for all their universal sovereignty, sent tributes to the Chinese
emperor.) Although their dynastic founders may have been foreign in origin, these
were not stranger-kingships of the kind I discussed before, marked by the complementary sovereign claims of immigrant rulers and indigenous owners. Rather, the
cosmocrats synthesized the ontological and political dualisms of stranger-king polities. They detained the powers of both heaven and earth. One is reminded of the
famous text of the Han dynasty official Dong Zhongshu, “The Way the King
Penetrates Three”:
Those who invented writing in ancient times drew three horizontal lines and connected
them vertically through the middle, calling the character “king.” The three horizontal
lines are Heaven, Earth and Man, and that which passes through the middle joins the
principles of all three. If he is not king, who can do this? (Granet 1968: 264)
Although it is commonly described as a centralization of power, which it was, the
transformation initiated by the Qin founder of the Empire (third century BCE) was
also a cultural synthesis of this sort, overcoming the dualism of heaven and earth
structurally by undermining the regional lineage authorities, and ritually by undertaking the sacrifices to local spirits.8
Thenceforth, according to imperial theory and practice, world order is created as
the effect of the Celestial Emperor’s personal virtue (te). By a double impulse, centrifugal and centripetal, the imperial virtue extends outward to pacify the barbarian
peoples and attracts them inward on missions of fealty and tribute. Both movements
thus entail a politics of presencing. On the one hand, the civilizing virtue of the
emperor is distributed outward in the persons, objects, and acts of the imperial
bureaucracy. Or else, if need be, it is spread by imperial armies and garrisons – especially since the Yuan (Mongol) dynasty, whose Chinese ministers were able to deduce
that “there was no real contradiction between te and force so long as force was applied
by a ruler possessing te” (Wang 1968: 49). On the other hand, the emperor draws the
world into his own existence. He represents and absorbs the totality of beings and
things in his palaces, his gardens, his hunting parks, his regalia, his diet, and, not least,
the collections of wonders and monsters that come to him as tributes from the
untamed borderlands of the Middle Kingdom.
In official parlance, the tributary missions represent the barbarians’ desire to “come
and be transformed,” that is, transformed to culture by ritual participation in the
emperor’s presence and the splendors that manifest it. The barbarian rulers who sponsor these embassies, however, may have an eye singular to their material or political
advantages. They may be interested in lucrative trading rights. Or else, in coming
away with Chinese titles, seals, surnames, and regalia of office, the hinterland rulers
are invested by imperial authority with sovereignty in their own realms as well as a
subordinate rank in the Chinese world order. This was especially desirable when the
barbarian chief was granted a realm contested by others. Of course, the Chinese also
worked the tributary system for pragmatic benefits. As is well known, since the Han
dynasty if not before, the policy of installing barbarian leaders as Chinese bureaucratic
or military officials had the value of providing a defensive shield of “tame” or “cooked”
barbarians against the “wild” or “raw” barbarians of the regions beyond. However,
my argument is that the utilities of the tributary system are embedded in a larger
9781444333237_4_007.indd 115
8/7/2010 6:45:15 PM
116
Sahlins
exchange of mutually empowering values whose asymmetries generate lasting relations
of hierarchy and the conditions of the possibility of material and political advantage.
Recall the earlier discussion of the cultural politics of alterity, involving the harnessing of external life-giving powers to the well-being of the local society. Just so,
the parties to exchange in galactic systems are engaged in transactions of vital
powers – essentially reciprocal if characteristically hierarchical. Speaking of such
exchanges between Tai states and the subordinate hill peoples they call kha or
“serfs,” Andrew Turton observes that the center “desires the resources, the potency
and the potentiality, the ‘alien powers’ of the periphery, the wild, the forest. Both
center and periphery seek to restore ‘vitality’ in the exchange of powers” (2000:
25–6). Similarly, the tributes of the Chinese borderlands, ensouled with the wild
potency of these peripheral regions, empowered the Celestial Emperor and his officialdom. Indeed, the necessity of sustaining a “Middle Kingdom” by means of
exotic powers betrays the contradictions of the classical ideology of the barbarians’
transformation to civilization. As Magnus Fiskesjö (1999b) points out, for all the
notions of the Celestial Emperor’s civilizing mission, the Chinese always maintained a wild barbarian frontier and, in periods of imperial weakness, a nostalgia for
its lost marvels.
Consider the circulation of power-effects by means of the aromatic plants of
Southeast Asia, such as the camphor or sandalwood sent as tributes to the Celestial
Emperor from as far as Malaya, Sumatra, Borneo, and Java. For the Chinese the aromatics and other exotica from Southeast Asia, observes Edward Schafer (1967: 193),
“partook of the godly and the beneficial, and at the same time the deadly and the
devilish.” This godly and deadly potency, moreover, was equally evident at the distant
barbarian sources of camphor crystals and their aromatic like. In his useful compendium of Malay Magic, W. W. Skeat (1900: 212–8) notes that camphor was controlled
by indwelling spirits who had to be propitiated so that it could be discovered, and that
a special language had to be used – for communicating with the spirits? – when searching for camphor, as ordinary Malay was taboo.
Transported then as tributes to China and sent wafting through the T’ang emperor’s court, incense from Southeast Asian aromatics “marked the presence of the
royal afflatus, breathing supernatural wisdom through the worlds of nature and
human affairs” (Schafer 1963: 156). In formal levees of the T’ang ministers, a table
of aromatics was placed before the Son of Heaven. When inhaled by the court officials, the scent of camphor from Malaya or sandalwood from Borneo insinuated the
ordering presence of the emperor into their own persons, whence it was realized in
imperial statecraft and spread through the world. Like the capitalist surplus value
that returns in a fetishized form to rule its producers, the fetishized products of the
barbarians return as the Chinese imperial power that aimed to control and civilize
them. Responding to the news that the imperial protector general had been chased
out of Annam by a rebellion, a T’ang poet and court official lamented,
Remember when the North was on good terms with the Yüeh,
For a long time both were nourished by the southern fragrance.
Conversely, the potency of the Chinese gifts of regalia, titles, and surnames gave
long-lasting legitimacy to barbarian chiefs, indeed centuries of power resources,
which helps explain how the Chinese empire could have more cultural influence than
it had coercive force. In 255 AD, Zhuge Liang, the famous Shu Han minister and
9781444333237_4_007.indd 116
8/7/2010 6:45:15 PM
The Whole is a Part
117
strategist, led an invasion into (what is now) Yunnan in order to reinstate the imperial
authority that had lapsed with the decline of the Han dynasty. Convinced that he did
not have the power to directly govern his conquests, Zhuge then withdrew, leaving
the area to be kept in order by indigenous leaders who had been endowed with
Chinese offices and surnames. For centuries after, however, many of the native
peoples continued to erect temples to Zhuge and give him a place among their own
deities. The widespread Min-chia (Bai) peasants, reputed aborigines of Yunnan, still
in the twentieth century attributed their particular form of carrying-yoke to Zhuge,
saying their ancestors learned this style when serving in Zhuge’s armies. Other longstanding traditions of local peoples told that “their chieftain’s bronze drums, the
universal symbol of authority in the southwest, were originally bestowed by Zhuge
Liang.” Richard von Glahn (1987: 15) relates that when one of the local chiefs was
forced to surrender such a drum in 1573 to a victorious Ming army, he measured his
loss by saying that “with two or three of such drums, one could proclaim himself
king. Striking the drum at the summit of a hill will cause all of the tribes to assemble.
But now, all is lost.” Zhuge’s drums, in other words, had internal power-effects in
the native society. Which perhaps helps explain why such signs of his conquests,
including the temples honoring him, have been preserved into modern times even by
peoples of western Yunnan – a region his army never got to, according to recent
Chinese scholarship. Moreover, many of the chieftains in the areas he did pacify – as
far as a bit west of central Yunnan – proceeded to convert the Chinese surnames he
gave them into claims of Chinese ancestry, though of course they were not Han
Chinese people. In fact, similar warrants of Chinese ancestry have long been asserted
by grandees large and small among the non-Han peoples of the Sino-Southeast Asian
borderlands, including ruling families of Shan states who functioned as Chinese officials (tusi) in the Ming and Qing periods. It is as if the attractions of the emperor’s
virtue, as the Chinese were pleased to consider it, for the “barbarian” rulers consisted
more effectively in the long-term legitimation and enduring competitive advantage
that imperial recognition afforded them over local rivals. Yet then, by encoding their
own authority and identity in Chinese terms, the native rulers effectively manifested
the powers of the Celestial Emperor’s virtue.
Local rulers claiming Chinese ancestry they invented themselves, worshipping victorious Chinese generals whose campaigns never reached their territories, cherishing
as palladia of their realms precious objects reputedly bestowed by powerful Chinese
officials a thousand years before, sending tributes to the Chinese emperor from lands
too distant for him to rule or perhaps even threaten: everything suggests that, for all
its apparent ideological air, there is an element of truth in the Celestial Emperor’s
power to civilize the world by the force of his wisdom and his virtue. No doubt there
have been compelling demonstration-effects, that is, military and naval campaigns
that could induce compliance even beyond the range of actual conquests. But the
sphere of a given emperor’s influence, for one thing, is not just a real-politics of the
moment. Embedded and diffused in objects with agentive ruling powers, and in traditions of regal magnificence and spiritual efficacy, the cultural cum political authority of the Son of Heaven extends thousands of years in time and thousands of miles
in space. Even the reported military conquests of Chinese armies may be effective
fantasies. C. P. Giersch (2006) tells of Shan state chronicles of the eighteenth and
nineteenth centuries that relate how a fourteenth-century ruler who defied the Ming
emperor thus provoked a punitive Chinese invasion. According to Giersch, however,
there is no evidence of any such military action in Ming historical records, and he
9781444333237_4_007.indd 117
8/7/2010 6:45:15 PM
118
Sahlins
concluded it never happened. This is a hierarchically organized, spatiotemporal
field of ordering influences, involving certain dynamics of acculturation and
differentiation.
Intercultural Dynamics of Order and Change:
The Real-Politics of the Marvelous
But is it not always true, even among us, that good neighborly relations require of
the partners that they become the same up to a point in remaining different?
– Lévi-Strauss (1971: 126).
Differentiation by assimilation – this dialectic that A. L. Kroeber called “antagonistic
acculturation” – ripples through the secondary and tertiary centers of the galactic
system, and out to the tribal hinterlands. One might even recognize a Chinese pattern
in Islamic guise as far off as the coastal waters of western New Guinea, where Biak
islanders return from rendering tribute to the Sultan of Ternate (in the Moluccas)
with honorary titles, prestige goods, and semangat-like power they absorbed in the
ruler’s presence (Rutherford 2003). The successful voyagers could thus earn the
respectful Biak title of amben, “foreigner.” Their quasi-foreign identity, however, did
not make them less Biak, only more appreciated by their countrymen. The same for
the Kachin chiefs who take on “sophisticated” Shan ways and enter into subordinate
wife-taking relations with Shan aristocrats. They “do not thereby surrender their status as Kachin chiefs,” Leach reports, but on the contrary “their chiefly status as
Kachins is enhanced.” The Kachin leader becomes a stranger-prince, not by a foreign
imposition on the locals but by a local appropriation of the foreign – at least until a
gumlao revolt.
The differentiation by assimilation I am discussing would fall under the general
heading of “schismogenesis” as Gregory Bateson designated it, in the first instance by
reference to certain interdependent cultural oppositions in Melanesia (1935; see also
Bateson 1958). This is “complementary schismogenesis” in Bateson’s terms, rather
than “symmetrical schismogenesis,” a competitive dynamics also involved in the historical ethnographies at issue here – and to which I will return momentarily.
Complementary schismogenesis refers to the generation of structurally opposed forms
among interacting peoples: symmetrical and inverse forms that at once match and
negate each other. It is as if each of the competing groups strives to be the same as and
different from, equal to and better than, the other. Indeed, I began this chapter with
a classic example of complementary schismogenesis in the way the Germans opposed
French civilisation with their own Kultur. If civilization is a matter of degree, cultures
differ in kind; if civilization may be imposed from without by an imperial state, culture
grows only from within as an expression of the people. Civilisation versus Kultur
remained fighting words for the French and Germans right into World War I.
Rather than opposition by negation, symmetrical schismogenesis is like an arms
race in which each side tries to outdo the other by doing more of the same, on the
principle of “anything you can do, I can do better.” At the extreme, however, competition in quantity is exchanged for competition in quality: one goes beyond the existing terms of the opposition, trumping the adversary by shifting the contention to
means of another kind and a superior value – like introducing a new, devastating
9781444333237_4_007.indd 118
8/7/2010 6:45:15 PM
The Whole is a Part
119
weapon into the arms race. This is the kind of competition we saw in the external
sectors of galactic polities: ambitious persons or groups taking on prestigious foreign
identities of the center in order to outdo their local rivals, the way Yi tribal leaders
acquire status by claiming Chinese ancestry. The strategy is to win out in the native
competition by engaging transcendent powers in one’s own cause: powers above and
beyond the society and its customary bases of authority.
Allow me to exemplify by way of Polynesia, where, I would argue, a certain sort of
symmetrical schismogenesis, strikingly manifest in the contact with Europeans, was
already inscribed in the indigenous lineage politics. During the early colonial period
in these islands, local ruling chiefs became in effect stranger-kings by assuming foreign
identities on their own part – especially British identities and particularly that of “King
George.” The sons and heirs of three prominent Hawaiian chiefs in the late eighteenth century were named “King George” (Sahlins 1981b, 1992, 1995). This form
of symmetrical schismogenesis – or should we not call it “transcendental schismogenesis”? – was practiced notably by ambitious chiefs who could not claim by ancestry the
authority to which they now aspired by force and wealth. Here were charismatic,
upstart rulers who sought to lend the divine potencies of the foreign to their local
ambitions of sovereignty. Kamehameha of Hawaii for example, whose rise to power
over the archipelago was marked by the sacrifice of his close senior kinsman, legitimate heir to the rule of Hawaii Island, and the raising of the British flag over his house
and war canoe even before he ceded the islands to his “brother,” King George of
Beretania. In the early nineteenth century, John Adams Kuakini was ruling Hawaii
Island, Cox Ke’eaumoku was in charge of Maui, and Billy Pitt Kalaimoku was the socalled prime minister of the kingdom. Analogous observations could be made of
Iotete of the Marquesas, who declared himself an Englishman; King George Tupou
of Tonga; the Pomare rulers of Tahiti; and others. Nor was their behavior without
precedent in the native Polynesian tradition, which knew many legends of the usurpation of ruling chiefs by younger brothers or junior kinsmen: warriors whose demonstrable mana trumps the ascribed powers of the senior line. One might judge that
transcendental schismogenesis is an inherent structural disposition in lineages that are
ranked by seniority of descent and that succession passes by primogeniture, inasmuch
as junior kinsmen and their descendants are then destined to decline in rank as their
senior kinsmen increase in numbers. Hence the many tales of younger brothers who
overcome this fate by acts of violence that manifest a divine power (mana) stronger
than that inherited by descent.
Gregory Schrempp (1992) brilliantly analyzed this politics of transcendence as it
appears in the well-known cosmogonic narrative of the Maori of New Zealand concerning the godly sons of Rangi (Heaven) and Papa (Earth). The younger-brother
Tu– , ancestor of mankind, is able to defeat his elder sibling and celestial enemy Tawhiri,
and by his victory Tu– gains superiority over all his older brothers as well as control of
the earthly species descended from them. Tawhiri was the only one of the several
divine siblings to join the Sky Father (Rangi) when the others forcefully separated
their heavenly parent from the Earth Mother (Papa). In retaliation for this crime,
Tawhiri let loose devastating squalls, whirlwinds, and hurricanes on earth, scattering
the brothers hither and yon – which accounts for the distribution of natural species
respectively sprung from them. Only Tu– stood up to Tawhiri, battling him to a standstill. “Tu– alone was brave,” Maori say. And because his brothers had fled before
Tawhiri, Tu– then turned upon them, defeated them, and consumed them, rendering
them inferior to himself. He became the senior brother. As Schrempp concluded,
9781444333237_4_007.indd 119
8/7/2010 6:45:15 PM
120
Sahlins
bravery is thus able to dominate as a political value, not merely as a physical force,
because it is expressed and legitimated on a cosmic plane. Because Tu– was victorious
in this celestial sphere, his human descendants are able to control the earthly species,
including the staple foods, that instantiate his defeated brothers. The earthly privileges of the Maori are backed ultimately by higher cosmic values.
Some of the same sort of cosmic legitimation of sovereignty is entailed in the claims
of descent from Alexander the Great on the part of certain sultans in the Malay
Peninsula, Sumatra, and Borneo. The historical Alexander, incidentally, was already a
member of a stranger-king dynasty in Macedonia, as these rulers claimed descent from
the Heraclid kings of Argos – not to mention the native kingships and the daughters
of kings that Alexander, following the lead of his father Philip, acquired in his military
campaigns (Arrian 1976; Curtius 1946; Diodorus of Sicily 1963; Plutarch n.d.).
Alexander once had occasion to remind his Macedonian soldiers that his father had
transformed them from small groups of poorly clad nomads following their few sheep
in the mountains to prosperous, well-clad dwellers of cities in the Macedonian plains.
Among the fanciful romances of Alexander, however, the Arabic and Persian versions
identified him with Iskandar D’zul-karnain of Koranic fame, the militant propagator
of the Faith from the setting to the rising sun (the same direction as Alexander’s conquests). Thus the appearance of Alexander’s descendants as local ruling kings in the
Sejarah Melayu, the “Malay Annals” written on behalf of the sultans of the flourishing
fifteenth-century state of Melaka. Apparently redacted shortly after the fall of Melaka
to the Portuguese in 1511, the schismogenic argument is directed rather at the rival
Sumatran state of Melayu-Jambi. At issue was which fifteenth-century sultanate was
the legitimate successor of the ancient powerful kingdom of Srivijaya, centered in
Palembang, Sumatra.
Without going into the fine details of the Melakan tradition, it tells of three handsome descendants of Iskander/Alexander who appear in royal garb on a mountain
above Palembang and turn the rice fields into gold and silver – “where there is sovereignty there is gold.” The youngest and most prominent of the three, Sri Tri Buana,
marries the daughter of the local raja and succeeds him as ruler. Among other classic
aspects of stranger-kingship in the “Annals” account is the neutralization of the harmful aspects of the hero’s potency by this union. Until his marriage to the raja’s daughter, the effect of Sri Tri Buana’s liaisons with the many native women was to infect
them with a disfiguring skin disease. But when the daughter of the raja was unharmed
by sleeping with the Alexandrian prince, her father consented to enter into a pact of
mutual regard in which he effectively surrendered the rule to him. In a few generations, the lineal descendants of Sri Tri Buana converted to Islam and founded the
prosperous commercial state of Melaka.
Still, Melayu-Jambi may have had the last word here. On an edict issued in the late
eighteenth century by the sultan of Minangkabau, successor of the Melayu-Jambi rulers, were affixed three seals representing three sons of Alexander: the sultan of Rum
(Istanbul), the sultan of China, and his own seal as sultan of Minangkabau. The
youngest of the brothers, he was again the most powerful; indeed, as the edict read,
the Minangkabau ruler was “king of kings … lord of the air and clouds … possessed
of the crown of heaven brought by the prophet Adam” (Marsden 1811: 339).
The Alexandrian tradition in Indonesia is linked to many more narratives of the marvelous, some of epic proportions, of which I mention only a couple by way of conclusion. Sri Tri Buana’s mother was the daughter of the King of the Sea whom his father
had married when he went under the ocean and thereby succeeded to the rule of that
9781444333237_4_007.indd 120
8/7/2010 6:45:15 PM
The Whole is a Part
121
underwater realm. One is reminded of the Queen of the South Sea, the mystical bride
of legitimate Javanese kings, herself descended in the ruling line of the ancient Sundanese
kingdom of Pajajaran. Another Javanese hero, Baron Iskandar – whose moniker, as
Anthony Reid (1994: 92) points out, combines a European title with the Javanese
name for Alexander the Great – had a brother who also married a Pajajaran princess,
and among their offspring was J. P. Coen, the governor general of the Dutch East
Indies Company who established its headquarters at Batavia in 1619. Baron Iskandar
himself, who in his career was stranger-king of Spain before conquering the lands from
Arabia to China, eventually winds up in the service of the Islamic ruler of Mataram –
thus proving the superiority of that Javanese state to the intruding Dutchmen (Ricklefs
1974: 373 ff). Like the Polynesian chiefs’ identifications with European rulers or the
Chinese ancestry of “barbarian” rulers, there is a real-politics of the marvelous in these
Javanese traditions. We would lose a lot of anthropology if we followed the lead of too
many historians and were content to ignore or debunk them.
But there is another lesson to be learned from the anthropology of the marvelous,
for perhaps nothing so much as these transcendental representations of Hawaiian or
Indonesian political authority can demonstrate the insufficiency of our received paradigms of sui generis cultures. Plainly it would be impossible to generate such global
concepts and practices of power out of the internal relations of production or coercion
in the societies concerned. Rather than the simple expression of an endogenous
dynamic, the local system – in political, economic, and doctrinal dimensions – is to a
large extent a function of relations in a greater cultural-historical field. The peoples
situated in a regional network of center-periphery relations often engage in a kind of
upward political mobility, insofar as their own rulers model their sovereignty on neighboring galactic superiors, and perhaps ultimately on distant legendary sources. “Galactic
mimesis,” one could call it. Kachin chiefs take on the ruling style of Shan princes,
retreating from active life in the same way that Shan rulers withdraw into their palaces
and delegate affairs of state to underlings. Yet then again, the Shan princes fancy themselves divine beings and absolute monarchs, precisely on Burmese or Chinese models.
We cannot explain these developments from within, on the Herderian model or any
such program of cultural autonomy. The scandal is that the human sciences should
have had a more encompassing vision of culture and history from the beginning.
Notes
1 The theoretical esprit de clocher got even more involuted when Malinowski (1960 [1944])
went on to develop an abstract model of the components of a culture according to their
functions in serving basic human (i.e., precultural) needs.
2 In this connection, one might distinguish a “young Lévi-Strauss” from a “mature LéviStrauss” involving a move from the closed, synchronic, Saussurean-like system of a given
culture, where tout se tient, to an open, transcultural, and dynamic sense of “structure” as
an indefinite set of transformations.
3 The bibliographic background to this section on Germanic Kultur and French civilisation,
together with the reflex of the former in American anthropological concepts of culture, may
be found in Sahlins (2000: 164 n). See also Muthu (2003).
4 The following synopsis of stranger-king formations recapitulates two previous versions more
completely annotated than space will allow here (Sahlins 2009, in press; see also Sahlins 1981a).
For an excellent recent analysis of stranger-kingship in Sri Lanka, with comparative reflections,
see Strathern (2009).
9781444333237_4_007.indd 121
8/7/2010 6:45:15 PM
122
Sahlins
5 The Frankish kings acquired an Alexandrian descent in the ninth century; but with the
Roman emperors, they also claimed descent from Aeneas, as did their imperial and royal
successors down to the Hapsburgs in the sixteenth century (Tanner 1992).
6 I am arguing against one-sided interpretations of stranger-kingship as an external
imposition, a sovereign exception, or a projection of the king as a foreigner by virtue of
the alien social character of power. For a good ethnographic antidote, see Southall
(2004 [1956]).
7 Perhaps we should jettison or at least complement the Augustinian-Tylorean theory of the
origin of the incest taboo, the marry-out or die-out formula that invokes the utilitarian advantages of expanding the kindred, by a theory of exogamy as a quest of the potency of alterity.
Also, given the prestige (cum vital) virtues of foreign things, one might note that, contrary to
the common assumptions of economists, here scarcity is a function of value rather than
vice versa.
8 The following discussion, primarily on the relations of the Chinese imperium and its southern borderlands, is based on the works of Backus (1980), Fiskesjö (1999a, 1999b), Giersch
(2006), Glahn (1987), Granet (1968), Herman (2006), Hevia (1994), Horstmann and
Wadley (2006), Izikowitz (2001 [1951]), Kasetsiri (1976), Leach (1960), Lehmann
(1963), Promboon (1970), Robinne and Sadan (2007), Rock, (1948), Schafer (1963,
1967), Taylor (1993, 1999), Took (2005), Turton (2000), Wang (1968), Whitmore
(1985), Wolters (1970, 1986), and Wyatt (1984).
References
Arrian (1991). Anabasis of Alexander, 2 vols. Cambridge, MA: Harvard University Press, Loeb
Classical Library.
Backus, Charles (1980). The Nan-Chao Kingdom and T’ang China’s Southwestern Frontier.
Cambridge: Cambridge University Press.
Barth, Fredrik (1969). Introduction. In Ethnic Groups and Boundaries, ed. Frederik Barth.
Bergen-Oslo: Universitets Forlaget, 9–38.
Bateson, Gregory (1935). Cultural Contact and Schismogenesis. Man 35:178–83.
Bateson, Gregory (1958). Naven, 2nd ed. Stanford: Stanford University Press.
Brown, C. C. E., ed. and trans. (1952). The Sejarah Melayu or “Malay Annals”: A Translation
of Raffles MS 18. Journal of the Malaya Branch of the Royal Asiatic Society 25:7–276.
Brown, D. E. (1972). Another Affair of James Brooke? Brunei Museum Journal 2:206.
Collerson, K. D., and M. I. Weisler (2007). Stone Adz Compositions and the Extent of Ancient
Polynesian Voyaging. Science 317:1907–11.
Cunnison, Ian (1951). History on the Luapula: An Essay on the Historical Notions of a Central
African Tribe. Capetown: Oxford University Press for the Rhodes-Livingstone
Institute.
Curtius (1946). Histories. Cambridge, MA: Harvard University Press, Loeb Classical
Library.
de Josselin de Jong, P. E. (1975). The Dynastic Myth of Negri Sembilan (Malaya). Bijdragen
tot de Taal-, Land- en Volkenkunde 131:277–308.
Diodorus of Sicily (1963). Library of History. Book 17. Cambridge, MA: Harvard University
Press, Loeb Classical Library.
Downs, R. E. (1955). Headhunting in Indonesia. Bijdragen Tot de Taal-, Land- en Volkenkunde
111:40–70.
Dumont, Louis (1994). German Ideology. Chicago: University of Chicago Press.
Durkheim, Emile (1947). The Elementary Forms of the Religious Life [1914] (trans. Joseph
Ward Swain). Glencoe, NY: Free Press.
Elias, Norbert (1978). The Civilizing Process: The History of Manners. New York: Urizen.
Fiskesjö, Magnus (1999a). The Fate of Sacrifice and the Making of Wa History. Ph.D. diss.,
University of Chicago.
9781444333237_4_007.indd 122
8/7/2010 6:45:15 PM
The Whole is a Part
123
Fiskesjö, Magnus (1999b). On the “Raw” and “Cooked” Barbarians of Imperial China. Inner
Asia 1:139–68.
Gewertz, Deborah B. (1983). Sepik River Societies: A Historical Ethnography of the Chambri
and Their Neighbors. New Haven: Yale University Press.
Giersch, C. Patterson (2006). Asian Borderlands: The Transformation of Qing China’s Yunnan
Frontier. Cambridge, MA: Harvard University Press.
Glahn, Richard von (1987). The Country of Streams and Grottoes: Expansion, Settlement and
Civilizing of the Sichuan Frontier in Song Times. Cambridge, MA: Council on East Asian
Studies, Harvard University.
Granet, Marcel (1968). La pensée chinoise. Paris: Albin Michel.
Harrison, Simon J. (1990). Stealing People’s Names: History and Politics in a Sepik River
Cosmology. Cambridge: Cambridge University Press.
Herder, Johann Gottfried von (1803). Outlines of a Philosophy of the History of Man, 2 vols.,
2nd ed. (trans. T. Churchill).
Herder, Johann Gottfried von (1969). J. G. Herder on Social and Political Culture (ed. and
trans. F. M. Barnard). Cambridge: Cambridge University Press.
Herman, John E. (2006). Amid the Clouds and Mist: China’s Colonization of Guizhou 1200–
1700. Cambridge, MA: Harvard University Asia Center.
Heusch, Luc de (1958). Essais sur le symbolisme de l’inceste royal en afrique. Brussels: Université
Libre de Bruxelles.
Heusch, Luc de, ed. (1962). Le pouvoir et la sacré. Annales du Centre d’Étude des Religions I.
Brussels: Institut de Sociologie, Université Libre de Bruxelles.
Heusch, Luc de (1982a). Rois nés d’un coeur de vache. Paris: Gallimard.
Heusch, Luc de (1982b). The Drunken King: or, The Origin of the State. Bloomington: Indiana
University Press.
Hevia, James Louis (1994). Cherishing Men from Afar: Qing Guest Ritual and the Macartney
Embassy of 1793. Durham: Duke University Press.
Horstmann, Alexander, and Reed L. Wadley, eds. (2006). Centering the Margin: Agency and
Narrative in Southeast Asian Borderlands. New York: Berghahn.
Hoskins, Janet (1993). The Play of Time: Kodi Perspectives on Calendars, History, and Exchange.
Berkeley: University of California Press.
Hoskins, Janet (1996). The Heritage of Headhunting: History, Ideology, and Violence in
Sumba, 1890–1990. In Headhunting and the Social Imagination in Southeast Asia, ed.
Janet Hoskins. Stanford: Stanford University Press, 216–48.
Izard, Michel (1985). Gens du pouvoir, gens de la terre: les institutions politiques de l’ancien
royaume du Yatenga. Cambridge: Cambridge University Press and Éditions de la Maison
de Sciences de l’Homme.
Izikowitz, Karl Gustav (1968). Neighbours in Laos. In Ethnic Groups and Boundaries, ed.
Fredrik Barth. Bergen-Oslo: Universitets Forlaget, 135–48.
Izikowitz, Karl Gustav (2001). Lamet: Hill Peasants in French Indochina [1951]. Bangkok:
White Lotus Press.
Kasetsiri, Charnvit (1976). The Rise of Ayudya: A History of Siam in the Fourteenth and Fifteenth
Centuries. Kuala Lumpur: Oxford University Press.
Keppel, Henry (1848). The Expedition to Borneo of H.M.S. Dido … With Extracts from the
Journal of James Brooke, Esq. of Sarawak. New York: Harper.
Kroeber, A. L. (1947). Cultural and Natural Areas of Native North America. Berkeley:
Leach, Edmund (1954). Political Systems of Highland Burma. Cambridge, MA: Harvard
University Press.
Leach, Edmund (1960). The Frontiers of Burma. Comparative Studies in Society and History
3:49–68.
Leach, Edmund (1961). Rethinking Anthropology. London: Robert Cunningham.
Lebor Gabala Erren (c. 1150 AD). Irish Texts Society. www.ancienttexts.org/library/celtic/
irish/lebor.html (accessed January 23, 2010).
9781444333237_4_007.indd 123
8/7/2010 6:45:15 PM
124
Sahlins
Lehmann, F. K. (1963). The Structure of Chin Society: A Tribal People of Burma Adapted to a
Non-Western Civilization. Urbana: University of Illinois Press.
Lévi-Strauss, Claude (1952). Race and History. Paris: UNESCO.
Lévi-Strauss, Claude (1971). Rapports de symétrie entre rites et mythes de peuples voisins. In
The Translation of Culture: Essays for E. E. Evans-Pritchard, ed. T. O. Beidelman. London:
Tavistock, 161–78.
Low, Hugh (1848). Sarawak; Its Inhabitants and Productions: Being Notes During a Residence
in That Country With H. H. the Rajah Brooke. London: Richard Bentley.
Malinowski, Bronislaw (1948). Magic, Science and Religion and Other Essays. Boston: Beacon.
Malinowski, Bronislaw (1960). A Scientific Theory of Culture and Other Essays [1944].
New York: Oxford University Press.
Malinowski, Bronislaw (1961). Argonauts of the Western Pacific [1922]. New York: Dutton.
McDougall, Harriette (1854). Letters From Sarawak: Addressed to a Child. Norwich: Thomas
Priest.
Marsden, William (1811). The History of Sumatra. London: J. M’Creery.
McKinley, Robert (1976). Human and Proud of It! A Structural Treatment of Headhunting
Rites and the Social Definition of Enemies. In Studies in Borneo Societies, ed. G. N. Apell.
DeKalb: Center for Southeast Asian Studies, Northern Illinois University, 92–126.
Mundy, Captain Rodney (1848). Narrative of Events in Borneo and Celebes … from the Journal
of James Brooke, Esq., Rajah of Sarawak and Governor of Labuan, 2 vols. London: John
Murray.
Muthu, Sankar (2003). Enlightenment Against Empire. Princeton: Princeton University Press.
Ovist, Krista (2004). The Integration of Mercury and Lugus: Myth and History in Late Iron Age
and Early Roman Gaul. Ph.D. diss., University of Chicago Divinity School, UMI
Microform 3116988.
Plutarch (N.d.). Alexander. In Lives of the Noble Grecians and Romans (trans. John Dryden).
New York: Modern Library, 801–54.
Préaux, J.-G. (1962). La sacralité du pouvoir royal à Rome. In Le pouvoir et le sacré, ed. Luc de
Heusch. Brussels: Institut de Sociologie, Université Libre de Bruxelles, 103–22.
Pringle, Robert M. (1970). Rajahs and Rebels: The Ibans of Sarawak Under Brooke Rule, 1841–
1941. Ithaca: Cornell University Press.
Promboon, Suebsaeng (1970). Sino-Siamese Tributary Relations, 1282–1853. Ph.D. diss.,
University of Wisconsin.
Reid, Anthony (1994). Early Southeast Asian Characterizations of Europeans. In Implicit
Understandings: Observing, Reporting and Reflecting on the Encounters Between Europeans
and Other Peoples in the Early Modern Era, ed. Stuart B. Schwarz. Cambridge: Cambridge
University Press.
Ricklefs, M. C. (1974). Jogjakarta Under Sultan Mangkubumi, 1749–1792: A History of the
Division of Java, London Oriental Series no. 30. London: Oxford University Press.
Robinne, François, and Mandy Sadan (2007). Social Dynamics in the Highlands of Southeast
Asia, Handbook of Oriental Studies, Section 3, Southeast Asia, no. 18. Leiden: Brill.
Rock, Joseph (1948). The Ancient Na-Khi Kingdom of Southwest China, 2 vols., HarvardYenching Institute Monograph nos. 8–9. Cambridge, MA: Harvard University Press.
Rutherford, Danilyn (2003). Raiding the Land of the Foreigners. Princeton: Princeton University
Press.
Sabloff, Jeremy A., ed. (2003). Tikal: Dynasties, Foreigners and Affairs of State. Santa Fe, NM:
School of American Research.
Sahlins, Marshall (1981a). The Stranger-King: or, Dumézil Among the Fijians. Journal of
Pacific History 16:107–32.
Sahlins, Marshall (1981b). Historical Metaphors and Mythical Realities: Structure in the Early
History of the Sandwich Islands Kingdom. Ann Arbor: University of Michigan Press.
Sahlins, Marshall (1985). Islands of History. Chicago: University of Chicago Press.
Sahlins, Marshall (1992). Anahulu: The Anthropology of History in the Kingdom of Hawaii,
vol. 1: Historical Ethnography. Chicago: University of Chicago Press.
9781444333237_4_007.indd 124
8/7/2010 6:45:15 PM
The Whole is a Part
125
Sahlins, Marshall (1995). How “Natives” Think: About Captain Cook, for Example. Chicago:
Sahlins, Marshall (2000). “Sentimental Pessimism” and Ethnographic Experience. In Biographies
of Scientific Objects, ed. Lorraine Dastin. Chicago: University of Chicago Press, 158–202.
Sahlins, Marshall (2009). The Stranger-King: or, Elementary Forms of the Politics of Life.
Indonesia and the Malay World 36:177–99.
Sahlins, Marshall (In press). Alterity and Autochthony: Austronesia Cosmographies of the
Marvelous. In Proceedings of the European Association of Oceanists. London: Berghahn.
St. John, Spenser (1879). The Life of Sir James Brooke, Rajah of Sarawak. Edinburgh: Blackwood.
Sangren, Steven P. (1990). Dialectics of Alienation: Individuals and Collectivities in Chinese
Religion. Man (n.s.) 26:67–86.
Sangren, Steven P. (1991). Power and Transcendence in the Ma Tsu Pilgrimages of Taiwan.
American Ethnologist 20:564–82.
Sather, Clifford (1990). The One-Sided One: Iban Rice Myths, Agricultural Ritual and Notions
of Ancestry. Contributions to Southeast Asia Ethnography: Rice in Southeast Asian Myth and
Ritual 10:119–49.
Schafer, Edward H. (1963). The Golden Peaches of Samarkand: A Study of T’ang Exotics.
Schafer, Edward H. (1967). The Vermillion Bird: T’ang Images of the South. Berkeley: University
of California Press.
Schrempp, Gregory (1992). Magical Arrows: The Greeks, the Maori and the Folklore of the
University. Madison: University of Wisconsin Press.
Sharer, Robert J., and Loa P. Traxler (2006). The Ancient Maya, 6th ed. Stanford: Stanford
University Press.
Skeat, Walter William (1900). Malay Magic. London: Macmillan.
Southall, Aiden W. (2004). Alur Society: A Study in Processes and Types of Domination [1956].
Münster: LIT with the International African Institute.
Strathern, Alan (2009). The Vijaya Origin Myth of Sri Lanka and the Strangeness of Kingship.
Past and Present 203:3–28.
Stuart, David (2000). The Arrival of Strangers: Teotihuacan and Tollan in Classic Maya History.
In Mesoamerica’s Classic Heritage: From Teotihuacan to the Aztecs, ed. David Carrasco et
al. Boulder: University of Colorado Press.
Tambiah, Stanley (1976). World Conqueror and World Renouncer: A Study of Buddhism and
Polity in Thailand Against a Historical Background. Cambridge: Cambridge University
Press.
Tambiah, Stanley (1985). The Galactic Polity in Southeast Asia. In S. Tambiah, Culture,
Thought, and Social Action: An Anthropological Perspective. Cambridge, MA: Harvard
University Press.
Tambiah, Stanley (1987). The Buddhist Conception of Universal King and Its Manifestations in
South and Southeast Asia. Kuala Lumpur: University of Malaya.
Tanner, Marie (1992). The Last Descendent of Aeneas: The Hapsburgs and the Mythic Image of
the Emperor. New Haven: Yale University Press.
Taylor, Keith (1993). The Birth of Vietnam. Berkeley: University of California Press.
Taylor, Keith (1999). The Early Kingdoms. In The Cambridge History of Southeast Asia, ed.
Nicholas Tarsky, vol. 2. Cambridge: Cambridge University Press.
Took, Jennifer (2005). A Native Chieftaincy in Southwest China: Franchising a Tai Chieftaincy
under the Tusi System of Late Imperial China. Leiden: Brill.
Turton, Andrew, ed. (2000). Social Identity in Tai States. Richmond, UK: Curzon.
Van Baal, J. (1966). Dema: Description and Analysis of Marind-Anim Culture. The Hague:
Martinus Nijhoff.
Viveiros de Castro, Eduardo (1995). From the Enemy’s Point of View: Humanity and Divinity in
an Amazonian Society (trans. Catherine V. Howard). Chicago: University of Chicago Press.
Walker, J. N. (1998). James Brooke and the Bidayuh: Some Ritual Dimensions of Dependency
and Resistance in Nineteenth-Century Sarawak,” Modern Asian Studies 32: 91–115.
9781444333237_4_007.indd 125
8/7/2010 6:45:15 PM
126
Sahlins
Walker, J. N. (2002). Power and Prowess: The Origins of Brooke Kingship in Sarawak. Honolulu:
University of Hawaii Press.
Wang, Gungwu (1968). Early Ming Relations with Southeast Asia: A Background Essay,” in
John King Fairbank, ed. The Chinese World Order: Traditional China’s Foreign Relations.
Cambridge, MA: Harvard University Press, 34–62.
Whitmore, John K. (1985). Vietnam, Ho Quy Ly, and the Ming (1371–1421). New Haven: Yale
Center for International Studies, Council on Southeast Asian Studies.
Wolters, O.W. (1970). The Fall of Srivijaya in Malay History. Ithaca: Cornell University Press.
Wolters, O.W. (1986). Restudying Some Chinese Writings on Srivijaya. Indonesia 42:1–41.
Wyatt, David (1984). Thailand: A Short History. New Haven: Yale University Press.
9781444333237_4_007.indd 126
8/7/2010 6:45:15 PM
8
Lingual and Cultural
Wholes and Fields
Alan Rumsey
Introduction
Especially in its structuralist and culturalist varieties, anthropology has often drawn on
ideas about languages as totalities for the modeling of other presumed ones, including
“cultures” and “societies” (e.g., Boas 1911; Lévi-Strauss 1963: 55–80, 206–31;
Sahlins 1976). In other more “processual” and “practice-oriented” approaches, such
modeling has often been vehemently rejected, but usually on the basis of similar ideas
about languages – that they comprise integral, relatively autonomous formal systems,
and for that very reason are inappropriate as prototypes for understanding other less
totalizing and/or more concretely “embodied” aspects of human social life (e.g.,
Barth 1975; Bourdieu 1977a).
As the only linguistic anthropologist among the contributors to this volume, I feel
called upon to examine the assumption about language that is common to both of these
positions, and to complexify the picture by considering some of the nontotalizing ways in
which the phenomena of speech and language have been treated by linguists, developing
some of their possible implications for our understanding of aspects of culture and social
life. In common with other contributors to the volume, I believe that it is important to
hold open a place for holism in our anthropological accounts of how people get on in the
world, but that we should treat the issue of holism as an open, empirical one to be
addressed with respect to its place in their various ways of life. Accordingly, in this chapter
I will consider that issue specifically in relation to language, asking,
1. To what extent do languages as posited and studied by linguists actually comprise
discrete, internally coherent systems?
2. To what extent do various local, culturally specific understandings of the kind known
to linguistic anthropologists as “language ideologies” treat languages as discrete?
3. To the extent that they do, what is understood to be the relation between language
differentiation and social differentiation?
4. To what extent and in what ways do these various local understandings involve
notions of “whole” and “part”?
9781444333237_4_008.indd 127
8/10/2010 11:00:03 AM
128
Rumsey
5. Irrespective of question 4, to what extent is the concept of part–whole relations
useful for comparative understandings of sociocultural formations and historical
processes?
6. Must “holistic” approaches necessarily operate in terms of part–whole relations,
and if not, what else?
The ethnographic and historical evidence on which my discussion of these issues is
grounded will come from Aboriginal Australia, the New Guinea Highlands, northwestern Amazonia, France, and Germany. In the course of the discussion I will introduce (or adapt from Sahlins’ Chapter 7, this volume) a notion of “relational field”
that, I will argue, provides a better way of understanding many of the sociocultural
phenomena I consider here than does the idea of part–whole relations.
Turning to the first of the above issues – the relative internal coherence of languages per se – in line with the intended emphasis of this volume on concrete
demonstrations from the contributors’ own research, I will begin with an account
of the regional linguistic variation I have been confronted with in my two main
field areas: the northern Kimberley region of Western Australia, and the Ku Waru
region in the Western Highlands of Papua New Guinea.
Language Differentiation in Aboriginal Australia
Before European settlement, the continent of Australia was fully occupied by
Aboriginal people who spoke many different languages – somewhere between 250
and 400 depending on where you set the threshold between dialect difference and
language difference. The map in Figure 8.1 shows a subset of them from the
Kimberley region of Western Australia. Figure 8.2 shows a more detailed map of the
northern part of that region, including the so-called Worrorran family, a group of
languages which McGregor and Rumsey (2009) show to be historically related to
each other at the maximal level which is demonstrable within the region (and which
are about as similar to each other as, say, the Romance languages). The language
that I set out to study as a PhD student in 1975 was Ngarinyin, the Worrorran language that is shown near the center of Figure 8.2. That map is based in part on a
linguistic survey of the region that was carried out in the 1930s by Arthur Capell
from the University of Sydney. An example of the linguistic data on which it is based
is shown in Table 8.1, which displays the regional variants that Capell collected
pertaining to a particular area of grammar – the prefixes that are used on the words
for some body parts to indicate the person, number, and gender of the person or
creature to whose body they belong: ŋi-ambul, ‘my eye’, nj-ambul ‘her eye’, and
the like. From my own work in the area, I concluded that the range of differences
among regional speech varieties and the set of terminological distinctions made
among them by Ngarinyin speakers were a good deal more complicated than could
be shown on any such map.
On the one hand, there was variation within each of the named varieties. On the
other, many of the names shown on the map within contiguous regions were not
mutually exclusive with each other: Ngarnawu, for example, was for most purposes
regarded as a variety of Wurla, and Guwij more or less the same as Wurla. At a more
inclusive level, the term “Ngarinyin” (or “Ungarinyin”) was sometimes used as a
9781444333237_4_008.indd 128
8/10/2010 11:00:04 AM
9781444333237_4_008.indd 129
Unggumi
GREAT SANDY DESERT
Juwallny
Jarrakan
Pama-Nyungan
Wirrimanu
Kukatja
Gregory Lake
Ngardi
Halls Creek
Nyininy
Jaru
Kija
Walgi?
r
ive
dR
Or
Jaminjungan
Pintupi
Wanyjirra
Gurlndji
Malngin
Ngarinyman
Warlpiri
Timber Creek
Jaminjung
Gajirrabeng
Doolboong
Wyndham
Mirlwoong
Kununurra
Yiji
Andajin Kulu warrang
Wurla
Ngarnawu
Bunuban
Gooniyandi
Walmajarri
Wangkajunga
River
Ngarinyin
Wolyamidi
Worrorran
Munumburru
Daly River
Wadeye
Murrinh-patha
JOSEPH BONAPARTE GULF
Gunin/
Kwini
Wilawila
Guwij
Wunambal
Gulunggulu?
Bunuba
Worrorra
Fitzroy Crossing
Nyikina
Mangala
y
Nyangumarta
ro
Yulparija
Karajarri
Yawuru
Ngumbarl
Jukun
Nyulnyulan
Derby
Warrwa
tz
Bldjadanga
Broome
Umilda
King Unggarrangu
Sound
Jawi
Nimanburru
Nyulnyul
Jabirrjabirr
Bardi
Yawijibaya
Winyjarrumi
Fi
Figure 8.1 Map of Kimberley languages
OCEAN
INDIAN
AUSTRALIA
Kalumburu
Gambera
Miwa
Victoria River
8/10/2010 11:00:04 AM
130
Rumsey
Miwa
Kalumburu
Gunin/
Gambera
Kwini
Wilawila
Wunambal Northern Worrorran
Yiiji
Gulunggulu?
Winyjarrumi
Munumburru
Yawijibaya
Guwij
Worrorra
Wyndham
Ngarinyin
Miriwoong
Worrorran
Wolyamidi
Ngarnawu
Wurla
Eastern Worrorran
Umiida
Western Worrorran
King
Unggarrangu
Sound
Andajin
Kija
Derby
Warrwa
Bunuban
Jarrakan
ls
al
H
Bunuba
y
C
ro
tz
Fi
Nyulnyulan
Unggumi
ek
re
Fitzroy Crossing
River
Nyikina
Gooniyandi
Figure 8.2 Detailed map of Worrorran languages and dialects
cover term for a range of regional variants including most or all of the ones shown
on Figure 8.2 as “Eastern Worrorran.” This was especially the case among people
from the western part of that region – the area designated on the map as
“Ngarinyin.”
The difference between the eastern and western parts of the region in that respect
was something I had been prepared for by Capell (1939a). Having worked with
speakers from across the region (i.e., the one labeled “Eastern Worrorran” on Figure
8.2), Capell concluded that all the varieties spoken within it were similar enough to
be considered dialects of a single language. So in line with the practice of Aboriginal
people associated with the western part of that region, he used the term “Ungarinyin”
for the whole “language.” Apparently assuming that each such “language” in
Aboriginal Australia was spoken by a single “tribe,” Capell found to his chagrin that
speakers of the dialects in the eastern part of that region apparently did not recognize
that fact about themselves:
[The Ngarinyin “tribe”] is bounded on the west by the Worora, on the south-west by the
Unggumi, on the south by the Bunaba and Gidja and on the east by the Djerak tribes.
The Ungarinyin language is spoken over all this large area, though in the eastern parts of
it the blacks do not use this or indeed any one tribal name. (Capell 1939b: 382)
Capell noted a similar contrast between the eastern and western portions of the adjacent region, which on Figure 8.2 is labeled “Northern Worrorran.” There, too, he
found that
9781444333237_4_008.indd 130
8/10/2010 11:00:04 AM
9781444333237_4_008.indd 131
8/10/2010 11:00:04 AM
ja-
nj-
i-
njarwm-
Source: From Capell and Coate (1984: 84).
wm-
ŋŋarnjarŋun-
ŋanŋararŋunnjir-
First singular
Plural included
Plural excluded
Second singular
Plural
Third
I (combined)
I (masculine)
II (nonpersonal)
II (feminine)
III
IV
V
VI
Windjar
Worora
njjawm-
i-
ŋŋarnjarŋunnjir-
Jawdjib.
njarwm-
i-, ju-
ŋŋarjarnjunnjir-
Unggumi
w-~wugum-~mugu-
nju-~njug-
ju-
ŋŋarjarŋun-~nugunjir-
Unggarangi
nj-~njugujaw-~wugum-~mugu-
ja-
ŋŋarnjarŋun-~nugunjir-
Umida
Possessive prefixes for body part nouns in some Worrorran languages and dialects
Person
Table 8.1
nj
bı˙rwm-
a-
ŋiŋarnjarŋaŋagur-
Ngarinjin
njajabiriwm-
aja-
ŋiŋarnjarnjiŋigur-
Munumburu
wm-
njaja-
oja-
ŋŋarnjarnjiŋigur-
Woljamidi

bı˙rwmn-
bı˙rwmn-
a-
b-
ab-
ŋŋarnjarg
gı˙r-
Wunambal
ŋŋarnjarg
gı˙r-
Wilawila
bı˙rwmn-
a-
b-
ŋŋarnjarg
gı˙r-
Gı˙nan
bı˙rwmn-
a-
b-
ŋŋarnjarg
gı˙r-
Forrest
River
132
Rumsey
[t]hese eastern tribes lack national names and national feeling. All who have worked
among them bear witness to the extreme difficulty of getting any tribal names from
them, and in this paper I have had to suffice myself in several instances by calling the
whole after a name of the part. In the west the national feeling is strongly developed.
One can hardly be a day in a Wunambal camp without hearing the word “Wunambal”
used a number of times in one connection or another. (Capell 1939b: 398)
I will return below to the matter of Capell’s frustration at these people’s failure to
recognize the whole to which they belonged, thereby forcing him to recognize it for
them by “calling the whole after a name of the part.” First I turn to an account of
regional language variation at the other locale where I have the most fieldwork experience, in Papua New Guinea.
Language Differentiation in the Western Highlands
of Papua New Guinea
Papua New Guinea is a land of enormous language diversity, more so even than
Aboriginal Australia, with around 800 languages spoken by a total of, nowadays,
about 6 million people. But the core of the highlands region, extending from Chimbu
westward to Enga Province, is exceptional not only in being the most densely populated part of the country but also in being an area of proportionally lower linguistic
diversity, at least as measured by the size of the populations who speak mutually intelligible dialects across contiguous regions. The dialects collectively known as Enga, for
example, are spoken by approximately 250,000 people. And to the east, in the Western
Highlands Province in and around the Ku Waru region where Francesca Merlan and
I have worked, there is a single dialect continuum with a combined speakership of
about 200,000. The map in Figure 8.3 shows the region associated with that larger
dialect continuum, and the Ku Waru dialect region within it. There is no name for the
continuum as a whole, so on the map I have named it rather arbitrarily after dialects
and regions on its periphery, the best known of which is probably Melpa, spoken in
the provincial capital of Mount Hagen and to the north where Marilyn and Andrew
Strathern have worked.
Although the dialects near the outer edges of this region are not mutually intelligible, within it there is nothing but continuous gradation among them. Nor is
there any single set of mutually exclusive names for dialects or “languages” within
the region, or for the continuum in toto. The term “Ku Waru” (literally, “steep
stone” or “cliff”) that we have fixed on for ethnographic purposes is one that is
used to highlight what is common to the locales and people on either side of the
Tambul Range, from Kailge across to its western slopes in the Upper Kaugel Valley
(who are heavily intermarried with each other and share a common dialect which
differs from that of their neighbors to the east and west, respectively). But for other
purposes, the people and dialects on the western side of the Tambul Range are differentially classified as Kakuyl (Kaugel), and the eastern-side ones as Napilya
(Nebilyer), after the major rivers that flow through their respective valleys as shown
on the map.
9781444333237_4_008.indd 132
8/10/2010 11:00:05 AM
Lingual and Cultural Wholes and Fields
133
145⬚E
144⬚E
5⬚S
SC
HR
5⬚S
Ra
AD
m
u
Ri
ve
E
r
R
R
A
N
Sim
E
Jim
TOMBEMA
ENGA
iver
Ba
ive
BIS
MA
R
r
rR
CK
R
NG
E
iver
La
iR
iR
MARING
iR
iye
Wabag
iv e
r
G
MAE
ENGA
ba
H
A
E
N
RA
SE
PIK - W
AH
G
N
NORTHERN
MELPA
Gumant River
B
R
Mt. Hagen
GE
Pabarabuk
u
Ka
Mendi
ge
l
E
Nondugl
Minj
CHIMBU
Kundiawa
NEBILYER
VALLEY
Mt. Giluwe
WAHGI
(KUMA)
Korgua
Kailge
6⬚S
Wa
hg
ver
AN
GI D
IVID
i Ri
L
r
Rive
M
CENTRAL
MELPA
r
lye
ab
TA
U
Tambul
Banz
AN
GE
Ri v
KUBOR
RANG
6⬚S
E
KAMBIA
e
r
Mt lalibu
Ialibu
River
Tua
l
ar
o
River
Er
a
ve
ver
Ri
Papua
New
Guinea
W
Indonesia
E
S
Map Area
7⬚S
7⬚S
0
25
50 km
144⬚E
Figure 8.3 Map of the Imbonggu–Tambul–Nebilyer–Melpa dialect continuum
The overall area of the continuum is shaded in gray. Within it, the Ku Waru region is unshaded. The
boundaries shown are approximate only.
9781444333237_4_008.indd 133
8/10/2010 11:00:05 AM
134
Rumsey
Unitary Language as Analytical Construct
Given patterns of regional variation such as the Ngarinyin and Ku Waru ones that I have
described, what are their implications for the notion of languages as discrete, bounded
totalities? Before I turn to that question, I will first review some basic propositions of
structural linguistics in what I think many anthropologists take to be their canonical form,
as propounded by Ferdinand de Saussure in his Cours de linguistique générale (1955
[1916]). For linguists, Saussure’s most important achievement in that work was to
develop a systematic account of paradigmatic opposition and the mutually defining character of syntagmatic and paradigmatic relations. In some readings of the Cours, Saussure
is claimed to have taken the position that languages are systems which are so tightly unified that the value of every single term within any one of them depends entirely and
equally on its relations of contrast to every other term within the language. But nowhere
in Cours does Saussure say this. Indeed, even the quote which is famously attributed to
him along those lines, that language is “un système où tout se tient,”1 does not actually
occur in the Cours; it seems first to have been used in print by Saussure’s student Antoine
Meillet. The closest that Saussure comes to such a formulation is his statement, “La
langue est un système dont toutes les parties peuvent et doivent être considérées dans leur
solidarité synchronique”2 (Saussure 1955 [1916]: 124). But Saussure was fully aware,
and stated as a matter of principle in one of the less often cited sections of the Cours
(“Part Four: Geographical Linguistics”), that such systems have “no natural boundaries”
(Saussure 1966: 203). With reference to familiar European dialect continua that are similar in many ways to the Highland Papuan one I have described above, Saussure says,
It is impossible … to set up boundaries between dialects. The same applies to related languages. The size of the territory makes no difference. We would be unable to say where
High German begins and Low German ends, and would find it just as impossible to draw
the line between German and Dutch, or between French and Italian. There are extreme
points where we may assert “Here French predominates, here Italian,” but in the intermediate region the distinction would disappear. We might imagine a compact, more restricted
zone of transition between two languages – e.g. Provençal between French and Italian –
but such a zone simply does not exist. How can we possibly depict an exact linguistic
boundary on territory that is covered from one end to the other by gradually differentiated
dialects? The dividing lines between languages, like those between dialects, are hidden in
transitions. Just as dialects are only arbitrary subdivisions of the total surface of language,
so the boundary that is supposed to separate two languages is only a conventional one.
(1966: 204)
How then were linguists to create an object of study for themselves? I use the term
“create” advisedly because that is exactly how Saussure understood the matter. He
has a lot to say about this question in the Cours, and makes it clear that he regards
linguistics as somewhat unusual in this respect. Other sciences, he says, “work with
objects that are given in advance, and can then be considered from different viewpoints,” but in linguistics “it is the viewpoint that creates the object” (Saussure
1966: 8). And how does it do that when faced with the fact of gradually differentiated dialects and languages? Through the standard scientific procedure of “conventional simplification,” which it uses to posit as a methodological fiction something
called a “language state” that allows linguistic science to abstract away from regional
variation in the same way that its abstracts away from change over time:
9781444333237_4_008.indd 134
8/10/2010 11:00:05 AM
135
In short, a language state can only be approximate. In static linguistics as in most sciences, no course of reasoning is possible without a conventional simplification of the
data. (Saussure 1955 [1916]: 143; my translation)
In the case of dialect variation, the simplification that Saussure recommends is to
“focus on a single geographical point” (Saussure 1966: 207) and to study the speech
forms used there as if they comprised a single, bounded system. But he fully acknowledges that in some respects they do not, and suggests that a continuum of dialects
that are spoken within a continuous region can comprise a single system, since
change which [springs] up at one place in the territory [meets] no obstacle in spreading
and gradually [extending] far beyond its origin point. Nothing impedes the action of
intercourse in a linguistic mass within which there are only imperceptible transitions.
(Saussure 1966: 206)
In short, Saussure allows for different points of view that can be taken on the phenomena of speech and language, thereby creating different kinds of objects: that of the
practitioner of “static linguistics,” that of the dialect geographer, that of the social historian, and so on. I cite Saussure here because of the emblematic status of his work, but
his methodological perspectivism in this respect is fairly standard among linguists, for
whom the issue of holism has never become as highly charged as it did for anthropologists in the 1980s. Even among linguistic anthropologists it is not uncommon to find
people treating languages for some purposes as relatively cohesive and autonomous
formal systems associated with given locales, and for other purposes as gradient, highly
permeable ones which are, as Bill Hanks (1996: 140) puts it, “saturated by context.”
For example, the former approach was the one I took in my study of Ungarinyin
(Rumsey 1982), for which I took as my reference point and comparator for regional
variation the western dialect of the language which was the most commonly spoken
one at the settlement where I did most of my initial fieldwork (Mowanjum), and the
approach that Francesca Merlan and I have taken in our work on Ku Waru, for which
we have focused upon the main dialect spoken at Kailge where we did our fieldwork.
But in both locales we have also found that the dialects associated with contiguous
regions comprise the sort of open, continuous field of the kind discussed by Saussure.
Indeed, in the Australian case, this is raised to a higher power by something that
Saussure does not take account of, namely, multilingualism. To this day, in mainland
areas where Aboriginal languages are spoken, almost every fluent speaker is fluent in
at least two of them, and it is not uncommon for one person to speak four or five,
even in areas where the languages differ greatly in grammar and vocabulary. Nor does
speakership cluster in such a way that, for example, languages A and B are spoken by
all and only the members of some particular “tribe” (Sutton 1978). Unlike in many
parts of the world, in Aboriginal Australia language boundaries were not significant
communicative boundaries, because people tended to be able to speak the languages
of neighboring regions as well as their own.
This being the case, as far as Aboriginal languages are concerned, there is probably no place anywhere in Australia that corresponds at all closely to Saussure’s
notion of “a single geographical point” at which we can find a single “language
state.” What one finds instead are complex patterns of code switching which are
indexically related to various kinds of contextual factors including the social setting,
9781444333237_4_008.indd 135
8/10/2010 11:00:05 AM
136
Rumsey
discourse topic, and aspects of the relationship between speaker and addressee
(Sansom 1980; McConvell 1988; McConvell and Meakins 2005).
Nowadays, across much or all of the area in Australia where Aboriginal languages are
still spoken, the languages that figure in this complex language ecology include not
only ancestral Aboriginal ones but also forms of English, often including creolized
ones. These are especially relevant to consider in connection with questions of “holism”
and Saussurean langue, because creoles are another phenomenon, like dialect variation,
that has to be understood in terms of continua rather than discrete grammars or codes.
Sophisticated models along those lines have been developed by linguists such as Derek
Bickerton (1975) and John Rickford (1987). A basic concept is that of the “postcreole
continuum” with, at one end, an “acrolectal” speech variety deriving from an introduced exogenous language; at the other end, a “basilectal” or maximally creolized
variety; and between them, a continuous range of “mesolectal” varieties. Speakers who
participate in such a continuum shift their speech varieties across a range of different
levels within it, in ways that are indexically related to aspects of the speech situation, as
for code switching more generally.
Language Difference and Group Identity in Australia,
New Guinea, and Northwest Amazonia
So far in this chapter, I have been discussing the question of holism as it pertains to language per se. But in the remark I quoted from Capell about the eastern Ngarinyin people’s seemingly obtuse nonrecognition of their status as a “tribe,” one can see an instance
of an assumption that is commonly made, sometimes even by anthropologists, that languages are associated – normally one to one – with discrete groups of people who speak
them. But from what I have said already about multilingualism, it is obvious that, across
much of northern Australia at least, that is not the case. Nor is Australia unique in this
respect. Another well-attested case is in the Vaupés region of northwestern Amazonia,
where multilingualism is not only common but also actively maintained by a principle of
linguistic exogamy which in effect guarantees that children will grow up speaking at least
two languages and usually more, since the longhouse groups in which they are brought
up normally host speakers of three or more languages (Jackson 1983; Chernela 1996).
Both the Australian Aboriginal case and the Vaupés one show that it is wrong to assume
that speakership of a given language is necessarily or normally associated with a discrete
population or group. But in cases where it is not, one cannot necessarily assume that language difference plays no part in the definition or constitution of social groups or categories of people. The Vaupés case provides a prime example of how this can be done on a
basis other than speakership, namely, through a one-to-one relation of ownership between
what Jackson (1983) calls “language descent” groups and given languages. Indeed, some
such relation other than speakership is a necessary condition for a system of linguistic
exogamy in such a region, since if it were based on speakership, given the rate of multilingualism relative to the number of languages in the region, the resulting restriction of the
field of allowable marriage partners would be too severe to allow the population to reproduce itself without most people marrying outside the region (to say nothing of the perhaps less unusual problem of marriage partners not speaking a common language!).
In the Australian Aboriginal case also, language differences play a part in social
differentiation without regard to speakership, but on a different basis than in Vaupés.
In the Australianist literature there are, as far as I know, no reported instances of a rule
9781444333237_4_008.indd 136
8/10/2010 11:00:05 AM
137
of linguistic exogamy, though a high rate of actual instances of it is well attested in many
regions (Sutton and Koch 2008: 495–7). Rather, languages are associated with persons
and groups in an indirect way, as a product of the two more direct relationships – one
between people and country, and the other between languages and country. In the case
of people, the relevant relationship is established through facts of spirit conception
and/or filiation to particular land-linked groups or social categories (clans, semimoieties, etc.). Languages too are seen as being directly linked to particular tracts of country.
By “directly linked,” I mean that the link between language and land does not depend
on the mediating presence of a set of people who speak the language and occupy the
territory, as it frequently does, for example, in ethnonationalist ideologies.
Rather, the language is seen as inherent in the country itself – often with an associated charter myth which accounts for how it became associated with that particular
region. In the case of Ngarinyin, for example, the region associated with that language
(as shown on Figures 8.1 and 8.2) is not called that because it is or was occupied by
people who speak the Ngarinyin language. Rather, it is called Ngarinyin country because
of the founding acts of a mythological creator figure Andarri, “Possum,” who is said to
have spoken the language at Gulemen “Beverley Springs,” and to have carried it from
there all over present-day Ngarinyin country, up to the limits of other language regions
in the area (Rumsey 1996: 5). Myths which in similar ways posit and account for links
between land and language are widely attested across northern and central Australia.3
For example, a thousand kilometers to the east, at Doomadgee in northwestern
Queensland, the territories of the Ganggalida, Garawa, and Waanyi people are
where particular languages are said to “belong,” implying that they fit there appropriately with other features of the landscape.…
The concept of language as a fundamental characteristic of landscape is also evident
from mythic accounts where travelling totemic figures change their language on reaching
the boundary of a linguistic territory [as, for example, in the story of a snake who switches
from Jingalu to Waanyi at the present border between the two]. (Trigger 1987: 217–9)
In central Australia, T. G. H. Strehlow (who grew up among Aranda people at
Hermannsberg Mission) was told of how a pack of ancestral native cats,
after travelling from Port Augusta [on the southern coast of Australia] through the territory of the Jankuntjatjara and Matuntara, entered the Aranda area at Ilbirla – a series of
springs.… As soon as they crossed the Palmer River their ears were deafened by the chirping of crickets in the river grass; and, in their confusion, they began to speak in a mixture
of Aranda and “Loritja” [Western Desert Language] after speaking only “Loritja” during
their previous travels over many hundreds of Western Desert Country. (1965: 133)
This myth establishes a connection not only between the Loritja language and its
country, but also between the neighboring Palmer River region and a particular mix
of languages.
In an insightful comparative study of such myths, Peter Sutton comments that, in
general,
The Aboriginal theories start from regional populations rather than from alleged units
within them. They stress the cultural valorization of difference as well as unity, they
attribute local linguistic diversity to acts of interaction rather than to isolation, and stress
consciousness of kind rather than kinds of consciousness. (Sutton 1997: 223)
9781444333237_4_008.indd 137
8/10/2010 11:00:05 AM
138
Rumsey
I agree with this reading of the relevant principles, and would add that, with respect to
language differences in particular, they take on an added significance in view of the fact
of widespread Aboriginal multilingualism. For one of its consequences is that complementarity can be enacted within a field of mutual intelligibility across linguistic difference.
Within such fields, the choice of which language to use in a given speech situation is
underdetermined by the need for intelligibility as such, since the parties share more than
one language in common. This greatly increases the scope for an etiquette of language
switching which is sensitive instead to aspects of social identity, or what Sutton aptly refers
to as “consciousness of kind.” The ethnographic sources on multilingual interactions in
Vaupés (Jackson 1983; Chernela 1996) suggest that the same is true there.
Before leaving the question of the relationship between language differentiation and
social differentiation, for another case to place alongside the Aboriginal and Vaupés
ones, I want to return to the Ku Waru one that I have referred to above. Here I can be
briefer than I was with respect to the Australian Aboriginal case, because language difference does not play a salient role in that respect, at either the local-dialect level or that
of the entire dialect continuum. With respect to the latter, there is, as I have said, no
name for the continuum in its entirety. Indeed, speakers of the dialects within it seem to
have little or no awareness of its existence. To be sure, at the outer edges of it, people
are quite aware that the degree of difference between their language and the neighboring ones outside of the continuum (Enga, Wahgi, Kewa, etc.) is far greater than between
their dialect and the neighboring ones within the continuum. But people are generally
unaware of, or uninterested in, the fact that their dialect is more similar to ones at the
opposite end of the continuum than it is to other languages outside of the continuum,
or that there is an unbroken chain of dialects in between which shade into one another.
On the other end of the scale, at the local level, as I have explained above, the label
“Ku Waru” is just one of a number of multiple, cross-cutting ways in which people
identify themselves and their dialect for certain purposes, which Francesca Merlan and
I have fixed upon in order to delimit the region that our ethnography pertains to. It
has little or no cachet as a marker of social identity. The political units which are still
of overriding importance in that respect within the region are the local-segmentary
ones known as talapi and regional blocks of them which play a large part in electoral
politics, competitive business ventures, and military alliance and hostility (Merlan and
Rumsey 1991; Rumsey 1999). In formal terms, these units in many ways comprise a
classical segmentary lineage system, but with other principles besides descent as the
basis on which they are constituted (Merlan and Rumsey 1991: 34–45 and passim).
Unlike in the Vaupés and Australian Aboriginal cases, those other principles do not
include language. Per contra, it is not uncommon to find a greater degree of dialect
diversity within one of the highest-level named segmentary groupings in the region
(Ndika, Munjika, Kulka, etc.) than across them at the outer limits of their territory.
Language Differentiation and National Identity
in France and Germany
We have already seen in a quoted passage from Saussure above that, at least in their spoken vernacular forms, the “languages” which we call French and German actually comprise “arbitrary subdivisions of the total surface of language” within larger regional fields
which also include, for example, Provençal in the former and Dutch in the latter. Although
9781444333237_4_008.indd 138
8/10/2010 11:00:05 AM
139
made in summary form in the Cours, this claim by Saussure is supported by detailed studies which were being made around that time under the rubric of “dialect geography”
(e.g., Wenker 1876–87; Wenker and Wrede 1927; Gilliéron and Édmont 1902–10).
As Saussure of course realized, alongside these dialect continua, there are more or
less “standard” varieties of the German and French languages which have developed
over the past several centuries. The course of this development in the two cases has
been similar in some ways and different in others. In both cases, as elsewhere in Europe,
the varieties of vernacular that first came to be used as “standard” ones were those
associated with royal courts. Already by the mid-sixteenth century in France, this meant,
in effect, the speech of a single region – the Île de France, centered on Paris (Brunot
1927). Another, related factor in the ascendancy of this variety of French was the high
degree of the centralization of the publishing industry in Paris, which was enforced by
government intervention during the reign of Louis XIV (Burke 2004: 108).
By contrast, in Germany, where political power long remained much more regionally dispersed, no single royal court provided a model for far-reaching standardization. For a time, from about 1500 to 1700, this role was assumed to some extent by
the Saxon dialect centered on the imperial chancery at Meissen, near Dresden (Twadell
1959; Barbour and Stevenson 1990: 47), but there as elsewhere in Germany, far more
than in France, the process of standardization involved the deliberate incorporation of
elements from other regional dialects, and the suppression of some of the more
regionally specific ones, so as to make the resulting written form of the language more
widely intelligible (Burke 2004: 101). The same was true of the language used in
Martin Luther’s translation of the Bible – which is widely regarded as the single most
important model for the development of German literary language (Arndt and Brandt
1983; Baeumer 1984). In other words, standard German has to a much greater extent
than French emerged as what linguists call a koine, an amalgam of different regional
vernaculars which is used as a lingua franca (or Lingua Communis, as it was called at
the time) across a larger region.
To what extent and in what ways does either of these or any other such national
language comprise a “whole”? This is a question on which a large amount of relevant
research has been conducted over the past few decades by sociolinguists, historians
and sociologists of language, and linguistic anthropologists. Two points have emerged
very clearly from this work. The first is that a basic feature of every such language is
variability along several different dimensions, including region, register, and socioeconomic class. The kinds of regional variation that one finds in this respect are different
from those of regional dialect, or on a “vertical” continuum with them, being, at the
top end of the vertical continuum, more a matter of regionally specific pronunciations
of the standard language than of differences of grammar and vocabulary such as are
found among regional dialects. At the other end of the vertical continuum, especially
in Germany, many people are in effect bidialectal, speaking the local dialect in some
kinds of speech situations and a more-or-less regionally inflected version of the standard language in others (Twadell 1959: 7) – a pattern known to linguists as “diglossia.”
The same is also true of some areas in France, especially in the south (Maurand 1981;
Grillo 1984: 80–1).
Cross-cutting variation of the kind discussed above, there are differences of what
linguists call “register,” that is, different ways of speaking which are characteristically
associated with particular kinds of subject matter and sociocultural context (Halliday
1978; Agha 2007). Variation of this kind even cross-cuts “whole languages” to a
considerable extent, in that, for example, the linguistic registers associated with
9781444333237_4_008.indd 139
8/10/2010 11:00:05 AM
140
Rumsey
criminal jurisprudence, Roman Catholic liturgy, or psychoanalysis as practiced in
Germany and France in some respects have more in common within each of these
registers than they do across the three.
Finally, among the forms of speech broadly designated as “French” and “German,”
there is, as elsewhere across the “developed” world, variation along the dimension of
socioeconomic class. In the case of French, where the process of standardization is of
longer standing, there is historical evidence for an equally longstanding, simultaneous process of differentiation between “cultivated,” “official,” “bourgeois” speech
associated with the emerging standard written form of the language and “popular,”
“vulgar,” “low” speech – these being terms “applied not only to the language but the
people who employ it” (Grillo 1984: 164). This distinction is no less current today,
notwithstanding the rise of mass education and literacy over the past two centuries –
as poignantly captured by Pierre Bourdieu’s account of the French working-class
“linguistic habitus,” which is characterized by an “objective tension” between people’s awareness of the norms of cultivated speech and the limitations on their capacity
to meet them (Bourdieu 1977b; Bourdieu and Passeron 1990). The situation in
Germany is somewhat different in that linguistic variation by socioeconomic class is
more closely tied with regional variation, with middle-class speakers being more fully
“diglossic” in the sense discussed above (Barbour and Stevenson 1990: 106–7, 183–4;
Stellmacher 1977).
In short, at the level of observable speech behavior, neither French nor German is
a “whole” in the sense of being a sharply bounded, internally unified system. Nor
should this come as a surprise after I have shown the same to be true even of an
Australian Aboriginal language with only a few hundred speakers, and a dialect continuum in the New Guinea Highlands with about 200,000. But just as in the Australian
Aboriginal case, this does not preclude a certain range of speech varieties from being
treated as an integral whole for certain purposes. As I have shown above, Saussure
presented linguists with a method for doing so for analytical purposes. But from a
more anthropological perspective, we must also take account of the fact that other
users of languages also construe them as more or less distinct, well-bounded wholes
for reasons of their own.
This brings us back to the other issue that I have discussed above in connection
with the Ngarinyin, Vaupés, and Ku Waru cases, namely, the relation between linguistic differentiation and social identity. As is well known, over the past two or three
centuries language difference has come to play a central part in the imaginary constitution of “nations” as distinct groups of people with a common social identity and
historical destiny. This is certainly true in the case of both France and Germany, but
in rather different ways as between the two. As one of the earliest politically unified
states in modern Europe, France under the ancien régime was conceived primarily as
a territorial entity, the polity unity of which was based on the trilogy of “one king, one
faith, one law” (“un roi, une foi, une loi”; Burke 2004: 163). From the seventeenth
century on, French was promoted as the language of state, but there was no idea that
linguistic unity had to be imposed on peoples speaking different languages who lived
under the same sovereign, for as long as his subjects paid their taxes and remained
subservient, the state had no interest in promoting lateral communication among
them (Grillo 1984: 28; cf. Peyre 1933; Gellner 1983: 10).
This changed radically during the French Revolution, as was forcefully expressed in
the following terms in a report by the Abbé Gregoire that was presented to the
National Convention in 1794:
9781444333237_4_008.indd 140
8/10/2010 11:00:05 AM
141
With thirty different local dialects, we are still, as regards language, at the Tower of
Babel, whilst as regards liberty we form the avant-garde of nations. [We must] make
uniform the language of a great nation, so that all its citizens can without hindrance
communicate their thoughts to each other.… Unity of language is an integral part of
the revolution. If we are ever to banish superstition and bring men closer to the truth,
to develop talent and encourage virtue, to mould citizens into a national whole … we
must have a national language. (From Abbé Gregoire, “The Need and the Means to
Eradicate the Patois and to Universalize the French Language,” as quoted in Grillo
1984: 24)
Based in part on the replies to a questionnaire on this topic that the abbé had circulated throughout the country, his report captures what must have been a broad consensus at the time among the literate bourgeoisie.4 Based on a thorough archival study
of these responses, Grillo says,
One word sums up these replies: “enlightenment.” Access to the French language will
give access to “instruction” … and through instruction enlightenment. The people will
become informed citizens, able to participate on equal terms, without intermediaries, in
the political process. Intellectual, moral and political improvement will all come from
speaking and writing the national language. But critically the language which will bring
out these improvements is French. (1984: 32)
This view was not without its opponents, especially within the Church and in the
southern “Langue-d’Oc”-speaking region of France (Grillo 1984: 39, 63–83).
They even included Napoleon Bonaparte, whose mother tongue was Corsican and
who learned French only as an adult. There is a well-known saying attributed to
him which, even if apocryphal, apparently summed up his views on the subject
accurately, to wit, “Let those brave fellows have their Alsatian dialect. They saber
well enough in French” (Grillo 1984: 39–40). But in another context, when it also
suited the interests of imperial France, Napoleon was willing to play the ethno-linguo-nationalist card as well, urging Hungarians to break with Austria by telling
them that “you have a national language … take up then once again your existence
as a nation” (quoted in Fishman 1972: 9). But notwithstanding such countercurrents, the position that favored the “universalization” of French (in line with Rivarol
1930 [1784]) was in effect the one that prevailed over the course of the next century and a half, beginning with a decree in 1794 forbidding the use of regional
vernaculars in judicial documents, and proceeding apace with the development during the nineteenth century of a highly centralized universal education system, conducted entirely in French.
As many have pointed out (e.g., Elias 1978 [1939]; Meyer 1952; Berlin 1976;
Brubaker 1992; Sahlins 2000, and Chapter 7, this volume), a very different understanding of nationhood developed in Germany, in which it was seen to be based in a
historically specific, shared Kultur which was expressive of the Volksgeist or “national
character” of those who were its bearers. In Chapter 7, Sahlins discusses these developments in some detail. I will not go over the same ground here, but merely draw
attention to two points which are especially germane to this chapter. The first is that
in the Herderian tradition that Sahlins reviews, language difference figures centrally as
an aspect of cultural difference, and therefore as part of the basis of nationhood.
Johann Gottlieb Fichte took this even further than Herder, arguing that “wherever a
separate language exists, there is also a separate nation which has the right to manage
9781444333237_4_008.indd 141
8/10/2010 11:00:05 AM
142
Rumsey
its affairs and rule itself” (quoted in Burke 2004: 164). Conversely, Friedrich Schlegel
maintained that “a nation which allows herself to be deprived of her language …
ceases to exist” (quoted in Burke 2004: 164).
The second relevant point to note is that the “culture” in question – the strain of
German ideology which gave rise to the notion of Kultur as autonomous whole – was
by no means universally embraced within what later became Germany. Sahlins points
out that its main milieu was the “German bourgeois intelligentsia,” who “were under
a double threat of French civilisation; not only from the power of France itself, but
also from their own ruling aristocracy, whose francophone and francophile tastes were
too evident proof of that power” (Chapter 7, this volume). Partially cross-cutting this
class-inflected difference, there was also a deep divide between the form of nationalism that prevailed in Prussia versus elsewhere. As Rogers Brubaker has put it,
The Prussian reformers … were untouched by the incipient ethnocultural nationalism of
the period. Awed by the French triumph and the Prussian collapse, they wished to create
a Prussian nation to regenerate the Prussian state.…
Romantics and reformers understood the relation between nation and state in completely different terms: the former in quasi-aesthetic terms, with the state as the expression of the nation and of its constitutive Volksgeist; the latter in strictly political terms,
with the nation – the mobilized and united Staatsvolk – as the deliberate and artificial
creation of the state.
Thus was engendered the characteristic dualism and tension between ethnonational
and state-national ideologies and programs – a dualism that has haunted German politics ever since. (1992: 10)
Repulsed by the excesses of the Nazi period, we tend to think of earlier German
nationalism as having always been thoroughly ethnonationalist. But for most of the
period during the nineteenth century when Prussia was rebuilding itself after the
French occupation, and through the early years of the second German Reich –
centered as it was on Prussia and Brandenburg – the “state-national” form of nationalism was at least as much of an operative force. In this respect,
Bismarck … set the tone by emphatically rejecting not only further territorial expansion in
Europe but also all demands for state intervention on behalf of Volksdeutsche abroad; his
successors adhered to the same policy. Only the collapse of the Hapsburg empire reopened
the fundamental question of the fit between nation and state. (Brubaker 1992: 126)
In the meantime, not only did Bismarck refuse to accept German ethnicity or language as sufficient conditions for inclusion within the Reich, but also at some points
during the course of his Polenpolitik he was willing to waive them as necessary conditions for it (Brubaker 1992: 129).
Conclusions
In this chapter I have considered the question of whether and to what extent languages
comprise distinct “wholes,” and the related issue of the role that language difference
does or does not play in the constitution or construal of social groups and entities as
9781444333237_4_008.indd 142
8/10/2010 11:00:05 AM
143
“wholes,” especially at the level of “tribe” and nation-state. With respect to the first
question, I have shown that the quasi-units we call “languages” are internally heterogeneous at every scale, whether in the case of Ngarinyin with its few hundred speakers,
the Imbonggu–Melpa dialect continuum with its approximately 200,000, or French
and German with their many millions. Thus, as has been forcefully argued by Irvine
and Gal (2000) and Silverstein (2000), we cannot assume that languages are simply
“out there” to serve as bases for the constitution of bounded social units, either as
actual speech communities or as “imagined communities” in the sense of Anderson
(1983). Rather, a well-bounded, homogeneous language is as much an “imagined”
entity as is “community.” As I have shown, Saussure fully acknowledged this in the
very work through which he founded the modern linguistic concept of language
(langue) as a synchronic system.
As I have also shown, the extent to which any such imaginary abstraction of a linguistic “whole” has taken place within actual speech communities is widely variable,
in ways that are not predictable from other facts about the speech varieties themselves
or the social settings in which they are spoken. Drawing partly on historical evidence,
I have shown that even within the Worrorran language area in northwestern Australia,
there was wide regional variation in this respect. The Imbonggu–Melpa dialect continuum in Highland New Guinea presents a more extreme case in which there is no
indigenous abstraction of a linguistic whole from anywhere within the continuum.
The French and German cases are extreme in the other direction, in that, notwithstanding the gradience of local dialects across both countries, there are clear ideas of
what constitutes la langue française and die deutsche Sprache – ideas which, as I have
shown, have been given real effect by extensive, more or less problematical work of
language standardization in both cases.
As for the other, related issue of the role that language difference does or does not
play in constituting social groups and entities as “wholes,” one can see across the
range of cases I have discussed that the kinds of relationships one finds between linguistic and social differentiation are diverse. In both of the European cases, the relations that matter most between people and languages in this respect are ones of
speakership. In none of the other three, non-European cases does the criterion of
language speakership figure in the constitution of social groups or identities. In two
of those cases – Vaupés and Aboriginal Australia – language difference does figure
strongly in their constitution, but on a basis other than speakership, namely, “language descent” and linguistic exogamy in Vaupés and land–language links in the
Aboriginal Australian case. In both of these cases, language difference figures not only
as a basis of social distinction but also as a principle of interrelatedness. Indeed, as
elegantly demonstrated by Peter Sutton (1997) regarding the Australian case, these
two principles are really two sides of the same coin, neither of which would be possible without the other. In both the Australian case and the Vaupés one, an enabling
condition of this mutually constitutive interplay between differentiation and relatedness is extensive multilingualism.
Both of those cases differ greatly from the two European ones in both respects. While
there has always been some German–French bilingualism within both countries – for
example, among the aristocracy eighteenth-century Germany, as discussed by Sahlins –
this has never been widespread across the socioeconomic spectrum except at some of
the border areas between the two language areas, such as Alsace (in German, Elsass),
Lorraine (Lothringen), and the Moselle (Mosel) Valley. Second, unlike in the Vaupés
and Australian Aboriginal cases, language difference in Germany and France has not
9781444333237_4_008.indd 143
8/10/2010 11:00:05 AM
144
Rumsey
figured as a principle of indigenously valorized, systematic interrelatedness between
the two associated sociopolitical entities, but only as a marker of difference between
them – or what Michael Silverstein (personal communication) calls “emblems … of …
ideologized social differentiation.”
Rather, what Sahlins and others have shown to be systematically interrelated
between Germany and France – through “complementary opposition” or “complementary schismogenesis” – are the concepts of Kultur and civilisation that have been
developed within German versus French intellectual milieux, and associated, inter
alia, with the German and French languages, respectively. This development figures in
Sahlins’ argument as a prime example of the way in which cultures “have always been
situated in greater historical fields of cultural others and largely formed in respect
of one another” (Chapter 7: 116).
With this case in mind, I turn now to the issue of what we mean by “holism,”
how it does or does not relate to the concept of the part–whole relation, and the
uses of either or both of these concepts for anthropology. In this connection I am
particularly interested in the concept of “historical field” that Sahlins invokes in the
quote above. I find this idea to be a very useful one in relation to the Australian
Aboriginal case that I have discussed, in that it allows us to posit systemic relationships without requiring us to treat them as part–whole ones. For that seems to be
exactly what is called for with respect to the relational field among Aboriginal language identities – since that field is an unbounded one, extending right across
Australia, and indeed across the Torres Straits into Papua New Guinea, as I have
argued elsewhere (Rumsey 2001).
Of course, in some cases a notion of the whole is exactly what is needed, as for example
in the Ngarinyin case I have described above, where, whatever differences Capell,
McGregor, and I may have found among what are classified as distinct dialects, such as
the ones represented in columns 7–9 of Table 8.1, the territory with which they are associated is for some purposes bounded off as whole by the movements of a single ancestral
hero. But by identifying their country in this way, Ngarinyin people do not thereby
bound off a whole “culture” or way of life which they identify exclusively with themselves. I suggest that in general, anthropologists should follow their lead in that respect
when trying to model relational fields of the kind discussed above and in Sahlins’ chapter
(Chapter 7, this volume), except insofar as the participants in those fields do posit such
totalities for themselves,5 as they clearly do in some of the cases that Sahlins discusses.
As I have shown above in respect to the relational field in which Australian Aboriginal
language identities are constituted, there is good reason not to characterize such fields
in terms of part–whole relationships, so as to be able to take account of the difference
between them and, for example, the clearly bounded whole that comprises the country associated with the Ngarinyin language, which in turn comprises a closed set of
patrilineal clan countries, and so on (Rumsey 1996). In the same way, notwithstanding the title of Sahlins’ chapter in this volume (“The Whole is a Part”), I would argue
that it is useful to keep the notion of relational field distinct from that of part–whole
relations in the European case he discusses, and to see the relational field of Kultur
versus civilisation as not consisting of a whole and its parts. Rather, it has a kind of
virtual existence, at two levels: as a pair of interrelated, opposing ideas and as a complex social field in which those ideas were generated and have been variously embraced,
elaborated, opposed, or exploited.
At the ideational level, Sahlins (1995: 10–13; see also 2000, this volume) and others (cited above) have made clear how the concept of culture has been developed in
9781444333237_4_008.indd 144
8/10/2010 11:00:05 AM
145
relation to that of civilization. Insofar as the two are opposed to each other at this
level, they are not parts in a whole but terms in a relation.6 As for the social processes
through which the ideas of culture and civilization have been propagated or resisted,
and the social domains with which they have thereby been associated, these cannot be
identified exclusively with Germany and France respectively, as they are neither universal within either country nor limited to the two. On the first count, as discussed
above, there have always been countercurrents to such an alignment, for example in
Bismarck’s Polenpolitik and refusal to treat German ethnicity or language use as necessary or sufficient conditions of German nationality, the Catholic Church and
Napoleon’s opposition to the “universalization” of the French language, or the latter’s willingness to encourage Hungarian ethnonationalism. As for national consistency in this respect, the most that can be conceded is to agree with Norbert Elias that
although the concepts of culture and civilization may have been formed by particular
individuals or given new meaning by them, the concepts “took root” and “established
themselves” to a differential extent on either side of the border because people’s differing “collective history has crystallized in them and resonates in them” (Elias 1994
[1939]: 6).
But, as Sahlins fully appreciates, the relevant differential “resonances” have operated far beyond the borders of Germany and France, not only in that they have been
strong on the “civilization” side in, for example, England, but also insofar as they
have been strong on the “culture” side in, for example, nineteenth-century Russia,
and more recently on a global scale among Third and Fourth World peoples who
“have been self-consciously counterposing their ‘culture’ to the forces of Western
imperialism that for so long afflicted them,” as treated in detail in Sahlins’ classic essay
on “Why Culture is Not a Disappearing Object” (2000: 163). For understanding this
and other instances of cultural (schismo)genesis such as the ones discussed in Sahlins,
the concept of “dynamic fields” (Chapter 7, this volume) proves eminently suitable as
demonstrated both by his chapter and by the examples of relational fields discussed in
this chapter from Aboriginal Australia, Northwestern Amazonia, and Western Europe.
I would argue that all these examples also demonstrate the need to distinguish such
fields in principle from part-whole configurations, which I would claim are not to be
found in any of them.
To be sure, there are some important differences among the fields that I have discussed in this chapter. One difference is that some of them – in common with most or
all of the ones discussed by Sahlins – are asymmetrical and hierarchical, while others
are not. The former include the German-French case, with France as the active, civilizing center and Germany as the reactive, Kultur-positing periphery; the Australian
creole continua, which, like creole continua everywhere, by definition involve a highstatus “acrolect” at one end and a low-status “basilect” at the other; and diglossia in
Germany, where higher social status is associated with the greater ability to move
between standard German and the local vernacular than to speak only the latter. No
such asymmetry is evident with respect to the case of Vaupés linguistic exogamy, the
regional dialect continua of Europe or Highland New Guinea, or the range of ancestrally inscribed language countries in Aboriginal Australia. But notwithstanding this
difference, all of these cases are better understood in terms of a notion of relational
field rather than of parts and wholes.
The distinction I am advocating in this respect is, I believe, consistent with
arguments made by several of the other contributors to this volume; see, for example,
Pedersen and Willerslev, who approvingly quote the work of another contributor:
9781444333237_4_008.indd 145
8/10/2010 11:00:05 AM
146
Rumsey
[W]e can only agree with Tim Ingold that, for the concept of holism to have any theoretical purchase, it is necessary “to dissociate [it] from a concern with wholes. Holism is one
thing, totalisation quite another, and [the] argument for holism [must be] … an argument
against totalisation.” (Chapter 15: 301, quoting Ingold 2007: 209)
The concept of “field” that I have applied is also, I believe, compatible with the
kind of structuralist holism that Jonathan Friedman endorses in Chapter 13, the
aim of which is to demonstrate that phenomena that appear to be autonomous and
even wholes-in-themselves are in fact manifestations of larger processes.7 More generally, it is compatible with an idea that I take to be at least implicit in many of the
chapters, that there is a distinction to be made between holism as a theoretical
stance – or, in George Marcus’ terms, an “aesthetic” – and the understanding of
human social life in terms of part–whole relationships. The phenomena of speech
and language have a special significance in this regard because the holistic stance
toward them which was pioneered by Saussure has proven to be so highly productive, and therefore has often been taken as a model for the understanding of other
sociocultural phenomena. But Saussure’s holism was not a holism of parts and
wholes, but of networks of relations.8 Furthermore, he was what I have called a
methodological perspectivist, with a healthy appreciation of the way in which “the
point of view creates the object.” The point of view that he himself developed most
fully was the one that treats languages as sharply bounded and internally homogeneous systems. But he also developed an account of dialect continua, in which he
treats those too in terms of networks of relations, of much the same kind as the
ones I have discussed here under the Sahlins-inspired rubric of “relational fields.”
By doing the same with respect to language identities and other sociocultural phenomena in this chapter, and expanding on the way in which Sahlins does so in his,
I hope to have shed a little more light on what might be meant by “holism” in
anthropology, and on the network of relations in which that term itself is embedded –
including its relation to concepts of “the whole.”
Acknowledgments
For their helpful comments on various drafts of this chapter I would like to thank Nils Bubandt,
Nick Evans, Francesca Merlan, Mark Mosko, Ton Otto, Marshall Sahlins, and Michael
Silverstein. Thanks also to the participants at the Beyond the Whole conference who discussed
the original version with me there, and to Ton and Nils for organizing such an interesting conference and inviting me to it.
Notes
1 In English, “a system in which everything holds together.”
2 In English, “Language is a system in which all the parts can and should be considered from
the viewpoint of their synchronic interrelatedness.”
3 In addition to the two other examples discussed below, see Merlan (1981: 144) on
Mangarrayi and Alawa; Merlan (1998: 125–6) on Jawoyn, Wardaman, Nangiomeri, and
Wagiman; and Sutton (1997) on the Wik languages of Western Cape York. Sutton (1997)
also includes comparative notes on myths of this kind from many other regions. See also
Rumsey (1993) for further discussion of the land–language nexus in Aboriginal Australia.
9781444333237_4_008.indd 146
8/10/2010 11:00:05 AM
147
4 Grillo (1984: 31) reports, “A survey [of Gregoire’s] correspondents showed that they consisted in the main of professional people (doctors, lawyers, teachers) and clergy. There was
also a substantial number of collective replies from branches of the Société des Amis de la
Constitution.”
5 For discussion of the ways in which such totalities are modeled in ethnography, see Rumsey
(2004, 2009).
6 Compare Ruben: “relations and sets thereof have no parts” (1983: 226).
7 I disagree with Friedman only insofar as he identifies these manifestations as “parts,” and
holistic explanations tout court as “hypotheses concerning the wholes that account for the
observable in terms of the relation between them and their parts” (Chapter 13: 280).
8 Saussure’s model of langue certainly included in it a place for part–whole relations, in the
form of syntagms, the basic combinatorial units of syntax and phonology (noun phrases,
syllables, etc.). But he did not see the “whole” of langue as comprising any such unit.
Rather, the syntagmatic relations in terms of which such units are defined are themselves
embedded within and defined by a more encompassing set of paradigmatic (or, as Saussure
called them, “associative”) relations, which are relations not of co-presence (as are all part–
whole relations) but of significant absence, or “difference that makes a difference.”
References
Agha, Asif (2007). Language and Social Relations. Cambridge: Cambridge University Press.
Anderson, Benedict (1983). Imagined Communities: Reflections on the Origin and Spread of
Nationalism. New York: Verso.
Arndt, Erwin, and Gisela Brandt (1983). Luther und die Deutsche Sprache. Leipzig:
Bibliographisches Institut Leipzig.
Baeumer, Max (1984). Luther and the rise of the German Literary Language: A Critical
Reassessment. In The Emergence of National Languages, ed. Aldo Scaglione. Chapel Hill:
University of North Carolina, 95–177.
Barbour, Stephen, and Patrick Stephensen (1990). Variation in German. Cambridge:
Cambridge University Press.
Barth, Fredrik (1975). Ritual and Knowledge among the Baktaman. New Haven: Yale University
Press.
Berlin, Isaiah (1976). Vico and Herder: Two Studies in the History of Ideas. New York:
Vintage.
Bickerton, Derek (1975). Dynamics of a Creole System. Cambridge: Cambridge University
Press.
Boas, Franz (1911). Handbook of American Indian Languages. Washington, DC: Smithsonian
Institution.
Bourdieu, Pierre (1977a). Outline of a Theory of Practice. Cambridge: Cambridge University
Press.
Bourdieu, Pierre (1977b). The Economics of Linguistic Exchanges. Social Science Information
16: 645–68.
Bourdieu, Pierre, and Jean-Claude Passeron (1990). Reproduction in Education, Society and
Culture. London: Sage.
Brubaker, Rogers (1992). Citizenship and Nationhood in France and Germany. Cambridge,
Mass.: Harvard University Press.
Brunot, F. (1927). Histoire de la langue française des origines à nos jours, vol. 2: Le Seizième.
Paris: Armand Colin.
Burke, Peter (2004). Language and Communities in Early Modern Europe. Cambridge:
Capell, A. (1939a). Languages of the North-West Australia. Mankind 2: 169–75.
9781444333237_4_008.indd 147
8/10/2010 11:00:05 AM
148
Rumsey
Capell, A. (1939b). Mythology in Northern Kimberley, North-West Australia. Oceania 9(4):
382–404.
Capell, Arthur, and Howard H. J. Coate (1984). Comparative Studies in Northern Kimberley
Languages. Canberra: Pacific Linguistics.
Chernela, Janet M. (1996). The Wanano Indians of the Brazilian Amazon: A Sense of Space.
Austin: University of Texas Press.
Elias, Norbert (1994). The Civilizing Process [1939]. Oxford: Blackwells.
Fishman, Joshua (1972). Language in Sociocultural Change. Stanford: Stanford University
Press.
Gellner, Ernest (1983). Nations and Nationalism. Oxford: Basil Blackwell.
Gilliéron, Jules, and Édmond Édmont (1902–10). Atlas linguistique de la France, 9 vols. Paris:
Champion.
Grillo, R. D. (1984). Dominant Languages. Cambridge: Cambridge University Press.
Halliday, M. A. K. (1978). Language as Social Semiotic: The Social Interpretation of Language
and Meaning. Edward Arnold: London.
Hanks, William F. (1996). Language and Communicative Practices. Boulder, CO: Westview.
Ingold, T. (2007). Movement, Knowledge and Description. In Parkin, D. and Ulijaszek, S.
(eds.) Holistic Anthropology: Emergence and Convergence. Oxford: Berghahn Books.
pp. 194–211.
Irvine, Judith, and Susan Gal (2000). Language Ideology and Linguistic Differentiation. In
Regimes of Language: Ideologies, Politics, and Identities, ed. P. Kroskrity. Santa Fe: School
of American Research Press, 35–84.
Jackson, Jean E. (1983). The Fish People: Linguistic Exogamy and Tukanoan Identity in
Northwest Amazonia. Cambridge: Cambridge University Press.
Lévi-Strauss, Claude (1963). Structural Anthropology. New York: Basic Books.
Maurand, G. (1981). Situation linguistique d’une communauté rurale en domain occitan. In
International Journal of the Sociology of Language, ed. A. Tabouret-Keller (special issue,
“Regional Languages in France”), 991–119.
McConvell, Patrick (1988). Mix-Im-Up: Aboriginal Code-Switching, Old and New. In
Codeswitching: Anthropological and Sociolinguistic Perspectives, ed. Monica Heller. Berlin;
Mouton de Gruyter, 97–149.
McConvell, Patrick, and Felicity Meakins (2005). Gurindji Kriol: A Mixed Language Emerges
From Code-Switching. Australian Journal of Linguistics 25(1): 9–30.
McGregor, William, and Alan Rumsey (2009). Worrorran Revisited: The Case for Genetic
Relations Among Languages of the Northern Kimberley Region of Western Australia.
Canberra: Pacific Linguistics.
Merlan, Francesca (1981). Land, Language and Social Identity in Aboriginal Australia. Mankind
13: 133–48.
Merlan, Francesca (1998). Caging the Rainbow: Places, Politics and Aborigines in a North
Australian Town. Honolulu: University of Hawaii Press.
Merlan, Francesca, and Alan Rumsey (1991). Ku Waru: Language and Segmentary Politics in
the Western Nebilyer Valley. Cambridge: Cambridge University Press.
Meyer, Alfred G. (1952). Historical Notes on Ideological Aspects of the Concept of Culture in
Germany and Russia. In Culture: A Critical Review of Concepts and Definitions, ed. C. L.
Krober and Clyde Kluckhohn. New York: Vintage, 403–13.
Peyre, H. (1933). La royauté et les langues provincial. Paris: Les Presses Moderne.
Rickford, John (1987). Dimensions of a Creole Continuum. Stanford: Stanford University
Press.
Rivarol, A. (1930). De la úniversalité de la langue française [1784] (with notes by Thomas
Suran). Paris: Didier/Privat.
Ruben, David-Hillel (1983). Social Wholes and Parts. Mind 92: 219–38.
Rumsey, Alan (1982). An Intra-Sentence Grammar of Ungarinyin, Northwestern Australia.
Canberra: Pacific Linguistics.
9781444333237_4_008.indd 148
8/10/2010 11:00:06 AM
149
Rumsey, Alan (1993). Language and Territoriality in Aboriginal Australia. In Language and
Culture in Aboriginal Australia, ed. M. Walsh and C. Yallop. Canberra: Aboriginal Studies
Press, 191–206.
Rumsey, Alan (1996). Aspects of Native Title and Social Identity in the Kimberleys and Beyond.
Australian Aboriginal Studies 1: 2–10.
Rumsey, Alan (1999). Social Segmentation, Voting, and Violence in Papua New Guinea.
Contemporary Pacific 11: 305–33.
Rumsey, Alan (2001). Introduction. In Rumsey, Alan and Weiner, James (eds.) 2001. Emplaced
Myth: Space, Narrative, and Knowledge in Aboriginal Australia and Papua New Guinea.
Honolulu: University of Hawaii Press, 1–18.
Rumsey, Alan (2004). Ethnographic Macro-Tropes and Anthropological Theory. Anthropological
Theory 4: 267–98.
Rumsey, Alan (2009). Rhetoric, Truth, and the Work of Trope. In Culture and Rhetoric, ed.
Ivo Strecker and Stephen Tyler, Studies in Rhetoric and Culture no. 1. Oxford: Berghahn,
117–49.
Press.
Sahlins, Marshall (1995). How “Natives” Think: About Captain Cook, for Example. Chicago:
Sahlins, Marshall (2000). Ethnographic Experience and Sentimental Pessimism: Why Culture
is Not a Disappearing Object. In Biographies of Scientific Objects, ed. L. Daston. Chicago:
University of Chicago Press, 158–293.
Sansom, Basil (1980). The Camp at Wallaby Cross. Canberra: Australian Institute of Aboriginal
Studies Press.
Saussure, Ferdinand de (1955). Cours de linguistique générale [1916]. Paris: Payot.
Saussure, Ferdinand de (1966). Course in General Linguistics (trans. Wade Baskin). New York:
McGraw-Hill.
Silverstein, Michael (2000). Whorfianism and the Linguistic Imagination of Nationality. In
Regimes of Language: Ideologies, Politics, and Identities, ed. P. Kroskrity. Santa Fe School
of American Research Press, 85–138.
Stellmacher, D. (1977). Studien zur Gesprochenen Sprache in Niedersachsen: Eine Soziolinguistische
Untersuchung. Marburg: Elwert.
Strehlow, T. G. H. (1965). Culture, Social Structure and Environment in Aboriginal Central
Australia. In Aboriginal Man in Australia, ed. R. M. and C. Berndt. Sydney: Angus and
Robertson.
Sutton, Peter (1978). Wik: Aboriginal Society, Territory and Language at Cape Keerweer, Cape
York Peninsula, Australia. PhD thesis, University of Queensland.
Sutton, Peter (1997). Materialism, Sacred Myth and Pluralism: Competing Theories of the
Origin of Australian Languages. In Scholar and Sceptic: Australian Aboriginal Studies in
Honour of L. R. Hiatt, ed. Francesca Merlan, Alan Rumsey, and John Morton. Canberra:
Aboriginal Studies Press, 211–42.
Sutton, Peter, and Harold Koch (2008). Australian Languages: A Singular Vision. Journal of
Linguistics 44: 471–504.
Trigger, D. (1987). Languages, Linguistic Groups and Status Relations at Doomadgee, an
Aboriginal Settlement in North-West Queensland, Australia. Oceania 57: 217–38.
Twadell, W. F. (1959). Standard German. Anthropological Linguistics 1: 1–7.
Wenker, Georg (1876–87). Sprachatlas des Deutschen Reichs. www.libraries.iub.edu/index.php?
pageId=400&resourceId=3175132 (accessed January 24, 2010).
Wenker, Geort, and Ferdinand Wrede (1927). Deutscher Sprachatlas. Marburg: Elwert.
9781444333237_4_008.indd 149
8/10/2010 11:00:06 AM
9
Deep Wholes
Fractal Holography in Trobriand
Agency and Culture
Mark Mosko
It is perhaps not surprising that the conception of wholes which for long shaped much
anthropological thought lost some of its attraction when it did. From the 1980s
onward, many of the phenomena that those totalizing models were designed to
describe came to be seen as dissembled and fragmented. Consequently, in modernity’s
wake the boundaries that earlier had set cultural and social wholes off from one
another have appeared to dissolve. Of course, Clifford (1988; see also Sahlins, Rumsey,
and Willerslev and Pedersen in Chapters 7, 8, and 15, respectively, this volume) and
others have argued that the totalities that previously dominated earlier social anthropology never existed historically. But even if we accept that the wholes are gone, what
about the parts (cf. Strathern 1991b: 107–11 and passim; 1991c)? If we are left with
parts only, these are parts of what? What sense does it make to think of parts if the
wholes that defined them are gone? Or might the residual parts of cultural systems be
viewed as emergent totalities with their own internal parts? Regardless of the recent
anthropological attention to parts and the corresponding neglect of the wholes that
once contained them, the two inescapably imply one another.
This kind of mutual insinuation has enjoyed a certain canonical status in much of
Western thought. In more pedestrian conceptualizations, any aspect of reality can be
composed of phenomena at two fixed levels or scales of observation: forms (i.e.,
wholes) and their respective contents (i.e., parts). But in more complex formulations
including formal logic, the relation between forms and contents is itself relative and,
consequently, dynamic; that is, every part of some encompassing whole may be
viewed as a form with its own internal contents, and any form can be similarly construed as an item of content within some larger whole. Accordingly, despite anthropology’s recent rejections of cultural holism, widely accepted claims that the parts of
earlier imagined wholes are now being rearticulated in the globalizing flows of people, ideas, objects, and media (e.g., Appadurai 1996; see also Friedman and Tsing,
Chapters 13 and 4, respectively, this volume) hint that new, more encompassing
wholes are in the process of emerging. Sahlins’ argument in Chapter 7, that earlier
envisioned cultural wholes were engaged in interdependent relations of which they
were only parts, points as well to the inevitability of whole–part relativity in anthropological analyses. Less often, though, have contemporary investigators considered
the reverse implication, that the constituent parts of now-fragmented cultural systems,
9781444333237_4_009.indd 150
8/7/2010 6:58:24 PM
Fractal Holography Among Trobriand
151
such as “subsystems,” might be usefully construed as wholes constituted of their own
internal contents or parts. Consequently, neither whole nor part should be viewed as
necessarily fixed at just a single level of observation. Paradoxically, I suggest, the
further we pursue the parts of cultural systems, the more wholes we might actually
encounter, and vice versa.
In this chapter, I explore a certain novel twist of this relativity of parts and wholes
from the perspective of chaos theory, or the science of complexity, which has emerged
over recent decades from the natural sciences (Gleick 1987; Stewart 1989). This new
quirk is known among practitioners in terms of “fractals” or “fractal holography.”
Fractal holography is more or less congruous with form–content relativity, as illustrated by several other contributors to this volume (e.g., Sahlins, Rumsey, and Tsing
in Chapters 7, 8, and 4, respectively), except that it presupposes the additional feature of self-similarity or self-scaling – the retention of form across scales. This means,
simply, that the parts of any form iterate or incorporate holographically the properties of that whole, and vice versa. The manner in which a fern leaf’s overall shape is
replicated in each of its leaflets, and again in the subleaflets of each leaflet, is a familiar
illustration of fractal relationship. Another is the irregularity of coastlines – margins
separating land from water – which are self-similar regardless of the altitude of
observation.
In the aftermath of chaos theory’s emergence, fractals are taken to be a surprisingly
common feature of reality due to the claim that nature is overwhelmingly complex, or
nonlinear. Space does not allow me to describe fully the relation of fractals to complexity and nonlinearity or other components of chaos theory as they pertain to social
anthropology, but I can refer interested readers to the book Fred Damon and I recently
edited, On the Order of Chaos (Mosko and Damon 2005). In the “Introduction”
(Mosko 2005), I describe how fractal self-similarity has long been an important, albeit
implicit, feature of some of the most influential of classic anthropological theorizing,
as for example in Radcliffe-Brown’s notion of organic analogy, Evans-Pritchard’s
model of Nuer segmentary opposition, Dumont’s concept of hierarchical opposition,
Lévi-Strauss’ canonic formula of myth, and Van Gennep’s pivoting of the sacred. The
explicit use of fractal self-similarity in contemporary anthropological studies can be
attributed to some of Roy Wagner’s (1991) and Marilyn Strathern’s (1991a, 1991b),
among others’, recent works. To this extent, the present chapter’s demonstration of
fractal whole–part iteration with regard to a single ethnographic case builds on wellestablished anthropological evidence that holographic self-scaling is a widespread
attribute of social and cultural systems.
This experiment, however, also offers insight into a related contentious issue that
has arisen in recent Pacific ethnology concerning the relation of cultural structures to
personal agency. I refer here to the juxtaposition of two hugely influential perspectives: Marshall Sahlins’ structural history program, which he has deployed mainly in
Polynesia (e.g., 1981, 1985, 1991), and Nancy Munn’s (1977, 1986, 1990) phenomenological treatment of Gawan culture in Melanesia.
Whereas Sahlins and Munn both incorporate in their theorizing cultural wholes and
the corollary presuppositions of parts and relations among the parts that are constitutive of those wholes, they are at radical odds with respect to the precise character of
the wholes and the role therein of human agency. Following Sahlins’ Saussurean
structuralism, a given cultural structure consists of a “culture-as-constituted” composed of the various value relations among the signs as parts of the system. Agency
enters in the course of putting those signs, or at least some of them, at risk in practice,
9781444333237_4_009.indd 151
8/7/2010 6:58:24 PM
152
Mosko
which leads either to the reproduction of the preexisting value relations or their
transformation. Therefore, although a given culture-as-constituted may be a product
of prior historical practices, when agents act by putting sign-values into practice, they
do so in the terms of the culture as summarily constituted at that moment.
By contrast, in Munn’s phenomenology, drawing upon Giddens, Sartre, Husserl,
and Schutz (1986: 6 ff; 1990), people’s immediate experiences of their practices rather
than preexisting cultural classifications create the culture structures – the expansible
space-times – in which the practices take place. This is quite dissimilar to Sahlins’ recognition of the historical conditioning of the culture-as-constituted, with a fundamentally different relation between the cultural structure or whole and people’s practices.
As Munn (1990: 13) puts it, Sahlins’ cultural structure consists of a priori concepts that
are anterior to practices, but for her the cultural structure is generated post hoc from
people’s experiences of their own practices in the world. My point is that both frameworks link structures to agency through the trope of wholes and parts, but for the one
it is the wholes which are prior to parts, whereas for the other it is the reverse.
Now Sahlins (1985) has stressed his aim of transcending such dichotomizations as
structure and event, culture and history, and langue and parole. But given the influence of Munn’s phenomenology as measured by the frequency with which poststructuralists and postmodernists have cited it, those dichotomies are not quite dead yet
and betray a noteworthy theoretical impasse.
In this chapter, I attempt to break this stalemate by inserting into the Sahlins–Munn
juxtaposition an ethnographic illustration of four notions: (1) that the cultural structures and practices on which both scholars focus are fractally self-similar; (2) that
part–whole relations recur at levels or scales additional to the two that are incorporated in their models; (3) that fractal self-similarity of this order possesses distinctive
implications for the character of personal agency and cultural process in the Pacific,
and possibly beyond; and (4) that fractal holography offers to anthropologists a new
perspective from which phenomena perceived to be unrelated or existing at different
scales might be connected.
With these purposes in mind, I develop a notion of “action scenario” utilizing ethnographic materials on the Trobriand Islanders – some from extant sources, and some
from my own recent fieldwork at Omarakana, the site of Malinowski’s pathbreaking
studies. Specifically, I outline sequences of activity consisting of certain parts and partaking of a recurrent form recognized in terms of botanical imagery – the widespread
Austronesian classification of plants in terms of “base,” “body,” and “tip.” In the
Trobriands, the corresponding plant parts consist of the u’ula (‘base’, ‘foundation’,
‘origin’, ‘cause’, ‘reason’), tapwala/tapwana (‘body’, ‘trunk’, ‘branch’, ‘main part’,
‘back’, ‘middle part’), and doginala (‘tip’, ‘end’, ‘final part’, sometimes substituted by
dabwala ‘head’ or matala ‘eye’). The several action scenarios I illustrate below covering a wide range of contexts include these three parts as sequential – hence temporal
as well as spatial – actions. But in the Trobriands and probably elsewhere, there is a
characteristic fourth stage whereby, from the doginala tip, keyuwela ‘fruit’ or ‘offspring’ emerge with the capacity of either reproducing the beings or things which
generated them or being transformed into yet other self-similar entities. These cultural
recurrences across contexts and scales, I argue, constitute a key template of, and for,
meaningful social action.1
Traces of these action scenarios are evident in Malinowski’s and subsequent
Trobriand ethnographers’ writings (Mosko 2009: 684–5), but their systemic pervasiveness and enchainment as holographic substitutions or transformations across
9781444333237_4_009.indd 152
8/7/2010 6:58:24 PM
153
multiple spheres of Trobriand human endeavor have yet to be documented. Munn’s
treatment of value transformations among Gawa Islanders, however, has added significance for the present study. Gawans are closely related neighbors to Trobrianders
in terms of language, culture, and history, and Munn’s phenomenological treatment
of the expansibility of space-time parallels numerous contexts of fractal base–body–
tip–fruit recursion which I discuss below.2 Also, it should be mentioned that the presence of cognate botanical metaphors of base, body, and tip has been amply demonstrated
for societies across the Austronesian sphere by James Fox and his collaborators (Fox
1988, 1995; Fox and Sather 1996). However, those investigations have tended to
concentrate on the spatial inferences of base, body, and tip (also eliding considerations
of “fruit”) for clarifying the nature of “origin structures” in Southeast Asia rather than
the metaphor’s temporal implications for semantic and social domains additional to
rank and precedence.
Yam (Taitu) Base, Body, Tip, and Fruit
I begin with the humble yam, or taitu. This is not an arbitrary choice. Taitu yams are
the chief subsistence and exchange crop of Trobriand Islanders, and similarly to other
parts of the Austronesian world the many terminological extensions I describe below
are consciously recognized by villagers as rooted in their classifications of plants generally and yams specifically. Trobriand culture can thus be described with little distortion as a “yam culture,” much in the same way that Evans-Pritchard (1940)
characterized the Nuer as a “bovine culture.”
In Coral Gardens, Malinowski (1935a: 140; 1935b: 89) describes taytu (i.e., taitu)
yams as consisting of a tapwana middle ‘body’, a matala ‘eye’ end, and a sibula ‘end’.
Today’s villagers explain that the base (u’ula) of the word sibula is itself u’ula ‘base’
(Figure 9.1, and below).3 Malinowski also describes how yam seed tubers (yagogu)
produce plants with matala or doginala growing tips, but his terms for trunk, roots,
and new tubers differ from tapwala and u’ula (1935a: 140–1; and see Figure 9.2).
Omarakanans nowadays confirm, however, that Malinowski’s terms for stems (e.g.,
yosila, kari lala’i, and tamna) are subtypes or parts of a vine’s tapwala trunk, and the
various types of roots he describes (silisilata, gedena) refer to components of the plant’s
u’ula base. But Malinowski’s account reveals two further anomalies: villagers insist that
shoots emerge from a seed’s planted base end, not its eye end; and a plant’s new tubers
or fruit (keyuwela), although harvested from underground, are considered extensions
of its doginala tip, not its base. I shall return to these important aberrations below.
According to Malinowski, u’ula, tapwala, and doginala/matala appear in several
additional contexts. Magical spells (megwa), for example, consist of such a “typical
tripartition” (Malinowski 1992: 196, 199–215; 1922: 428–63; 1932: 182; and see
Mosko 2007). The u’ula opening segment wherein the names of ancestral performers
of that magic are invoked is Trobriand magic’s “most universal feature” (1932: 328).
In the middle tapwala part, the magician’s words express the actions, processes, or
qualities desired. The concluding doginala segment enunciates the desired effects.
Human sexual functions also follow base–body–tip sequencing. For males,
[The eyes] are the basis or cause (u’ula) of sexual passion. From the eyes, desire is carried
to the brain … and thence spreads all over the body to the belly, the arms, the legs, until
9781444333237_4_009.indd 153
8/7/2010 6:58:24 PM
154
Mosko
GROWTH OF TAYTU : THE SPROUTING OF THE OLD TUBER
OF
LEVEL
SOIL
MATALA END
YAGOGU
SIBULA END
KOGA
SOBULA
TAPWANA
Figure 9.1 Taitu yam tuber
Source: After Malinowski (1935a: 140).
GROWTH OF TAYTU.THE NEW VINE
DOGINA = MATALA
MATALA
YOSILA
YOSILA
YOSILA
YOSILA
KARI LALA’I
KARI LALA’I = KARI SALALA
KALU POSEM
TAMNA
TOWABU
KALU POSEM = YAWILA
LEVEL OF
SOIL
SILISILATA
GEDENA
GOWANA
KOGA
SILISILATA
BWANAWA
GEDENA
GEDENA
SILISILATA
Figure 9.2 Taitu yam plant
Source: After Malinowski (1935a: 141).
9781444333237_4_009.indd 154
8/7/2010 6:58:24 PM
155
it finally concentrates in the kidneys. The kidneys are considered the main or middle part
or trunk (tapwala) of the system. From them, the other ducts (wotuna) lead to the male
organ. This is the tip or point (matala, literally eye) of the whole system. (Malinowski
1932: 141–2)
From my inquiries, the process is identical for women except that the external vaginal
tissues constitute the doginala ‘tip’ or matala ‘eye’ of sexuality. For both genders, however, the matala ‘eye’ figures at both ends – as the base end of sexual passion but also
at the tip ends of the genitals – similar to how yam shoots with eyes at their growing
ends emerge from the base end of yam seeds, not the eye ends as Malinowski reported
(see above). Thus yams and people are analogous, incorporating a second sexual pathway or road reversing the base–body–tip orientation of the total body otherwise.
These data illustrate the recurrence of the botanical metaphor in a diversity of
contexts, but only once does Malinowski reveal any appreciation of the significance
of this fact. In Sexual Lives, he considers the “three cardinal points” of the sexual
system just outlined to be “characteristic of native canons of classification” (1932:
143; emphasis added) but he refrains from developing this point further, perhaps
reflecting his antiholistic pragmatism (Tambiah 1990; Young 2004: 140, 238–41,
433–4). Moreover, in his scattered illustrations there are implications that matala
eyes or tips have the additional capacity to generate keyuwela fruit which can holographically become the u’ula bases to further base–body–tip sequences (cf.
Malinowski 1992 [1916]: 198). As current Paramount Chief Pulayasi Daniel put it
to me recently, “Base–body–tip is the base [u’ula] of Kilivila custom [gulagula].”
In other renderings, Malinowski characterizes the tripartite (or quadripartite, with
keyuwela fruit added) botanical metaphor as a keda ‘path’ or ‘road’ (1935a: 76;
1935b: 140, 147). Of course, keda roads have received considerable attention from
subsequent investigators, mainly as regards kula exchange, but without explicit reference to the botanical sequencing of growth and reproduction (cf. Leach and Leach
1983; Irwin 1983; Campbell 1983a, 1983b, 2002; Munn 1977, 1986; Damon
1983b, 2005; Weiner 1992). The various illustrations of base–body–tip–fruit recursion I outline below, however, map out the internal structuring of keda roads regardless of context.
Returning to plants, drawing on my own data, in fruit-bearing trees such as coconut, areca, and banana, growth is perceived as unfolding from base to body to tip, and
from bottom to top; hence, fruit emerge at the temporal-vertical tip. And like the
children of human parents, the fruit of these plants are regarded as their gwadi ‘children’ (see below). As mentioned above, taitu yams might here appear to be anomalous,
since their fruit (kabinela tubers) mature underground, the space otherwise associated
with a plant’s base. However, as I have already noted, a yam-child is generated from
its parents’ (i.e., a seed yam’s) base end, just as human children are procreated from
their parents’ base ends where the tips or eyes of sexual passion are located. Thus the
production of yam children from the base of yam tubers and yam plants is not in that
respect anomalous, at least with respect to human beings, and for this reason the relation between yams and people in Trobriand culture is especially marked (see below).
Harvested yams are of two types, subsistence and/or exchange tubers (i.e., taitu
proper) and seed yams (yagogu), each becoming the base of further, but divergent,
keda roads. At the start of the next agricultural season, yam seeds are sown horizontally
in the ground to generate whole yam plants patterned according to the generic image
described above. Both the roots and the stems develop from shoots that emerge from
9781444333237_4_009.indd 155
8/7/2010 6:58:25 PM
156
Mosko
the seed’s base end. As stems grow upward, they produce the main foliage (tamu,
Malinowski’s tamna) of the plant’s tapwala body which culminates in the plant’s
uppermost matala tendrils. Roots growing downward generate young child-tubers
which, although underground, are conceived as emerging at the parent-plant’s doginala tip also. As each new tuber (bwanawa) grows along a root, it swells first at its
base end, next in its body, and finally at the tip. Young tubers’ spatial segmentation
thus recapitulates the keda road of its developmental sequencing.
The luxuriant, moist, green foliage of living yam plants contrasts dramatically with
their brown, desiccated leaves and stems at harvest (cf. Malinowski 1935a: 168). As
explained to me, villagers harvest tuber-fruit only after the rest of the parent plant
dies. This association of the emergence of fruit with the death of the parent form that
generated it recurs in many areas of the culture, as for example in the exit of baloma
spirits from human corpses (see below). In this light, it is perhaps no coincidence that
leaping from the top of coconut palms is a preferred manner of suicide (cf. Malinowski
1932: 399, 424, 475–6). As regards yams, my point is simply that the emergence of
young tubers as fruit coincides with the temporal tip or death of parent plants, and
that the two are self-similar.
Yam Gardens
Trobrianders view yams as temporally and spatially self-similar also to the gardens that
produce them. Clearing, burning, planting, and staking constitute the u’ula base
phase of a gardening season. Subsequently as tapwala activities, a gardener and his
wife will tend the young, emerging shoots, twisting them carefully around vertical
yam stakes, weeding the plot, and pruning the immature tubers. Harvest corresponds
with the garden’s temporal tip. Trobrianders position their rectangular gardens spatially with the two base corners (notatala) closest to the village and the doginala end
corners furthest away (Malinowski 1932: 89–109; cf. Damon 1990: 141 passim,
261n.15).4 The contained area, or lopola ‘belly’, consists predominantly of the garden’s tapwala body. Community magicians (towosi) perform their rites moving from
the base corners through the middle body to the ends. Harvests proceed similarly
from base through body to tip, so that the yams unearthed from the dead plants are
viewed as emerging from the garden’s death also.
Yam Parents and Children
Just as a yam plant is considered the parent of the tuber-children, garden plots are
regarded as the parents of the yams grown there (cf. Malinowski 1932: 263). But
human gardeners and their wives are also fathers (tama) and mothers (ina) of the
yams they grow. And like human children, yams are gendered. Seed yams, capable of
reproducing further yams, are like daughters, and exchange yams, used for feeding
humans, are likened to their human parents’ sons.
The manner of sowing yam seeds is itself suggestive of parental relations and agency.
During the u’ula phase of readying his plot, a gardener turns the soil with his digging
stick held in his right hand, pulverizing the clods (kimimisi pwapwaya) with his left
hand. In the softened soil, he forms a cavity for the seed and covers it up with a mound
of soil, inserting a yam stick nearby. My interlocutor, Pakalaki, likened the soil to a
9781444333237_4_009.indd 156
8/7/2010 6:58:25 PM
157
womb and the yam stick to an erect penis. Another Omarakana resident, Molubabeba,
described the near-erotic passions men experience when, especially at dawn or dusk,
they view the young vines protruding from the uniform, horizontal black, moist softened soil and beginning to form around the hard, dry, white, vertical yam sticks.
Pruning and Heaping Yams
During the intermediate tapwala gardening season, gardener-fathers carefully
prune (basi) the clump of tubers growing under each plant so that it assumes conical dimensions, with a few large tubers at the base, one at the apex, and a limited
number of smaller ones in between (Ralph Lawton, personal communication;
cf. Malinowski 1935a: 151–2). In the absence of pruning, supposedly, taitu plants
produce large numbers of small tubers suitable only for feeding pigs. The gardenerfather’s pruning also forms (ikuli) and differentiates feminine seeds from masculine
exchange yams. Unlike the former, which cyclically regenerate future yams, the latter can be converted through exchange into a variety of other self-similar entities,
that is, exchange yams can be directed into additional base–body–tip–fruit action
scenarios or roads (cf. Munn 1986).
As Malinowski (1935a: 171–4) describes, newly harvested yams are first displayed
in a temporary shelter (kalimomia) for passersby to admire. Exchange yams are
heaped into a conical pile ( gugula) at the shelter’s center, while seed yams are sorted
into smaller piles around the perimeter. Like newborn humans, recently harvested
yams are regarded as white and weak. The shelters thus protect the young tubers
from the sun’s darkening and aging light, analogous to mothers’ birth cloaks
(saikeula). The base of exchange yam heaps, typically circumscribed by a short ring
fence (lolewa; see Figure 9.3), is assembled containing the largest, often irregularly
shaped tubers. More consistently proportioned tubers are added as the heap’s conical
tapwala body, culminating in the pile’s doginala peak. The outermost layer, however, is also regarded as an element of the heap’s tip, especially when decorated with
paint and pandanus streamers similar to the adornment of young children, adolescents, kula traders, and recently deceased kin (e.g., Weiner 1976: 36, 69, 127, 237–8).
The conversion and assembly of exchange yams into heaps thus fractally recapitulate
the base–body–tip–fruit scenarios of the now-dead plants and gardens that grew
them and the clusters in which they were formed.
After several days, the large yam heaps are dismantled and, with great fanfare,
transported and reassembled at villages as the annual harvest payments which
Malinowski mistakenly labeled “urigubu” (Weiner 1976: 179–83). In this process,
the initial yam heaps are destroyed and then, as fruit, reconstituted in new
base–body–tip forms (see Figure 9.4; cf. Malinowski 1935a: 179). This indicates
further transferences from a dead garden’s and heap’s tip to new life at the village.
Each new heap is erected in front of the yam house (bwaima) of its new father (i.e.,
the man for whom the garden couple labored, assuming his sons and/or in-laws
have erected one for him), for days, sometimes weeks, before they are deposited
inside. With much less flourish, the seed yams are placed in small piles near their
gardening parents’ dwelling or stowed in small storage houses (sokwaipa) nearby
(see Figure 9.5). Recapitulating the spatial layout of garden shelters, storehouses
for seed and exchange yams stand in peripheral feminine and masculine regions,
respectively, of circular villages (Malinowski 1935a: 242–3, 392–3).
9781444333237_4_009.indd 157
8/7/2010 6:58:25 PM
Figure 9.3 Garden shelter (kalimomia) and yam heap (gugula)
Source: Photo by Mark Mosko, 2007.
Figure 9.4 Yam heap and storage house (bwaima)
9781444333237_4_009.indd 158
8/7/2010 6:58:25 PM
159
Figure 9.5 Seed yam storage house (sokwaipa)
Yam Houses
Both kinds of storehouses are holographs of the yams they contain. Typically, seed
yam houses stand upon an u’ula base of four stones or wooden posts supporting a
small rectangular tapwala storage cavity framed by horizontal log cabin-like sticks and
topped by shade-providing coconut fronds. Men’s bwaima for storing exchange yams
are virtually identical to those Malinowski (1935a: 255ff.) described, with large foundation stones and two imposing horizontal beams bearing the rectangular log cabin
sides topped by a thatched (and, in the case of chiefs, decorated) roof as base, body,
and tip, respectively. Not insignificantly, both kinds of contained yams are jettisoned
as fruit from the tips of their storehouses for recursive deployment in further action
scenarios (cf. Malinowski 1935a: 257).
Yams Exchanged for Valuables
Whereas seed yams retained by their gardening parents are replanted at the onset of the
next gardening season to regenerate further yams, exchange yams are redirected by their
adopted parents along divergent action scenarios or roads for conversion into self-similar
items of male and female wealth and, eventually, into food. My 2006–9 field inquiries
confirm my earlier argument (Mosko 1995) that there are two prototypical categories
9781444333237_4_009.indd 159
8/7/2010 6:58:28 PM
160
Mosko
of “payment” linking the yam growers and recipients. On the one hand, every adult
male is expected to plant annually a kaimata main garden for a senior male, ideally his
father or another man who stands in relationship to him as a metaphorical or adoptive
father, such as the village leader or chief on whose land he gardens. On the other hand,
the father of a married daughter is expected annually to give yams yielded from a separate plot to his daughter’s husband. If the yams from a man’s main plot are given to his
daughter’s husband, they are kaimata; if his main plot is directed to his father, leader, or
chief, the yams for his daughter’s husband are termed kaimwila (cf. Weiner 1976: 140;
Hutchins 1980: 23). Where Malinowski and others have gone astray in claiming that
“urigubu” (sic) is mainly given by a man to his sister’s husband is in assuming that the
brother does so as a representative of his dala. Instead, my informants insist that when
a brother gives annual yams to his sister’s husband, he acts on behalf of, or in place of,
his father (cf. Weiner 1976: 140–53, 197–98; 1988: 82–84). Generally, a son gardens
for his father as long as the father lives, and it is the father who designates the person for
whom the son will garden on his behalf, either with kaimata payments to the local
leader or chief or as kaimata/kaimwila gifts to the father’s daughter’s husband.
As already noted, the recipient of yams and his wife become their adoptive father and
mother. Periodically that man is expected to return (takola/sasova) to the son or other
man who gardened for him an article of male wealth (veigua) (cf. Malinowski 1935a:
190, 372; Weiner 1976: 179–83; Munn 1986). The loss of a harvest heap of masculine
exchange yams, in other words, yields fruit for the gardener in the form of an article of
masculine wealth. Axe blades, the prototype of these fruit, are envisaged as possessing an
u’ula base at the hafting end, a matala eye at the sharpened end, and an intermediate
tapwala (Ralph Lawton, personal communication). This item of male wealth becomes
the adoptive child of the man who receives it, which as fruit he can convert into the base
of further exchange roads (cf. Weiner 1976: 179–85; Munn 1986: 156–7).
Opening, Intermediate, and Closing Kula Gifts
Kula transactions constitute the anthropologically best documented of these roads. If
the takola/sasova a man receives is a shell valuable that is not already moving along an
established kula path, it becomes his kitomu, his child (cf. Leach and Leach 1983), and
the potential base (u’ula) of a new kula road when given as an ‘opening’ (vaga) gift to
a kula partner. Subsequent ‘intermediate gifts’ (basi) constitute that road’s tapwala
middle body, and the ‘closing gift’ (kudu) at the road’s doginala end kills it
(cf. Malinowski 1922: 98, 353–7; Campbell 1983a; Damon 1990: 92). However, as
elsewhere in the culture, a second vaga opening shell given to accompany the closing
gift enables the initiation of a new road from the death of the prior one (cf. Campbell
1983a: 215–6; Damon 1983a: 315, 325–6; Munn 1977: 44–46; Weiner 1992: 143).
Yams Exchanged for Banana Leaf Bundles
Alternatively, Trobrianders convert exchange yams into “women’s wealth” – banana
leaf bundles (nununiga) and colorful skirts (doba) – as further recursions of the
botanical metaphor. As Weiner noted, leaf bundle and skirt transactions are affected
9781444333237_4_009.indd 160
8/7/2010 6:58:29 PM
161
Figure 9.6 Banana leaf bundle (nununiga)
not only during lisaladabu mortuary ceremonies (1976: 91–120; 1988: 116–38; cf.
Mosko 1985: 220–8; 1995: 773–4), but also in their manufacture and bartering.5
Lisaladabu ceremonies are regularly performed during the annual milamala celebrations after yams have been harvested, transported, and stored, as described
above. In one of the principal categories of lisaladabu exchange termed kaimelu,
women assemble small heaps of bundles for each man who presented them and/or
their husbands with raw exchange yams at either harvest time or previous mortuary
celebrations. The yams men grow for other men thereby elicit female-wealth bundles from the latter men’s wives paralleling the takola/sasova gifts of male wealth
(cf. Weiner 1976: 110–12; 1988: 120, 131). The receiving man’s wife takes possession of the bundles as the two become those bundles’ adoptive father and
mother. The bundle-children thus replace the lost and dead (i.e., now cooked
and eaten; see below) yam-children as the fruit of the latter (cf. Weiner 1976: 79,
80, 104–5).
Banana leaf bundles, moreover, are self-similar to the yams that generate them and
that they replace. The end where a bundle’s separate leaf strips are tied is its u’ula
base; the undecorated, fluffed-out middle portion is its tapwala body; and the fringe
of loose decorated strips is the doginala tip (see Figure 9.6). These bundle zones
recapitulate the process of their manufacture. Each bundle is made from a single
banana leaf, with strips taken from opposite sides of the central stem. After cutting the
strips, bundle mothers scrape them from leaf center to leaf edge against an engraved
wooden board provided by bundle fathers (who also planted the banana tree) which
impress designs onto the strip tips as instances of paternal ‘forming’ (ikuli) agency
(Mosko 1995: 767–70). Like seed yams, old “dirty” bundles can be retied to reproduce
9781444333237_4_009.indd 161
8/7/2010 6:58:29 PM
162
Mosko
Figure 9.7 Decorated young women wearing banana leaf skirts (doba)
“clean” ones, which can be given in exchange for men’s prior gifts of yams. New leaf
strips can also be made into colorful skirts (doba) which signify the creation of new
reproductive relations among humans; but here, only new clean leaf strips can be
regarded as suitable for participation in the conversion path of exchange yams (see
below; cf. Weiner 1976: 94, 95). Unsurprisingly, villagers manufacture and conceptualize skirts in three parts: a waistband and upper flange of white, flat leaves as the u’ula
base; a middle tapwala layer of short, multicolored shredded leaves; and the bottom
swoop of long, red, frayed leaf tips as doginala (see Figure 9.7).
At the start of lisaladabu proceedings, the fractal form of a bundle is reenacted in
the shape of huge bundle piles (pela sepwana) given to the deceased’s father’s and
spouse’s representatives as the most important lisaladabu transactions. Here the tapwala body of bundles is piled up and tied atop a flattened coconut leaf basket as the
u’ula base with skirts, mats, folded pieces of calico cloth, and/or paper money attached
to the doginala top (see Figure 9.8; cf. Weiner 1976: 105; 1988: 125). These latter
items along with clay pots, stone axe blades, and other valuables which are transferred
at the same moment in the proceedings have the capacity to be diverted into other
paths for regenerating other new relations.
Cooking and Alimentation
The conversion path of exchange yams leads ultimately to their transformation into
food which, when eaten, sustains human growth and life. Taitu are the prototypical
‘bodybuilding food’ (kaula) for Trobrianders (Montague 1985). But they must be
9781444333237_4_009.indd 162
8/7/2010 6:58:30 PM
163
Figure 9.8 Mortuary heaps (pela sepwana)
cooked by women before they can be eaten, normally at morning and evening on
self-similar hearths (see Figure 9.9). The three or four coral stones (kailagila) a
woman receives from her husband’s mother at marriage along with the wood fire
they enclose compose her hearth’s u’ula base. A clay (or nowadays aluminum) pot
and contained food and water ingredients constitute its tapwala body. In my observation, women prepare yams by slicing off the coarse skin and, with quick knife
strokes, detaching the base and eye ends, placing only the inner portion of tubers
into the middle portion of the hearth, or pot. A large leaf or aluminum cover serves
as the hearth’s doginala top, from which the cooked food is removed as fruit.
Cooked food is a substance fundamentally transformed by the proper spatial and
temporal ordering of all hearth elements. Cooking kills the raw yams, but unlike
seed yams, which also die when they are planted in gardens, exchange yams cannot
generate new yam plants. They nonetheless contain the vital capacity to feed and
sustain human bodies.
The feeding of cooked yams to family members thus recapitulates the conversions
of harvested yams into garden heaps, storage houses, male wealth, banana leaf bundles, and so on. In all of these transformations, there is a discernible pattern wherein
fractal recursions of base, body, and tip proceed in terms of life stages leading to death
from which fruit and new life sequences are generated. That human parents feed and
nourish their living human children with the bodies of their dead yam-children
9781444333237_4_009.indd 163
8/7/2010 6:58:31 PM
164
Mosko
Figure 9.9 Hearth stones (kailagila) and hearth, here with fish tins substituted
suggests that human beings and yams are fractal holographs of one another – an
analogy widely appreciated across the Massim.
Childhood, Adulthood, Elderhood, and Baloma
Trobrianders’ identifications between yams and people are complex, however, involving
additional temporally enchained self-similar recurrences and conversions. Omarakanans
claim that human life consists of four phases of personal growth – childhood, adulthood,
elderhood, and existence as a baloma spirit after death – corresponding with the base,
body, tip, and fruit of a person’s total life. Childhood, lasting from conception to marriage, constitutes the foundation (u’ula) of a person’s life wherein parents’ devotions are
most concentrated. In adulthood following marriage, the tapwala phase of human life,
people become parents themselves, procreating and growing children through sexual
activity and various gender-defined labors (giving birth, gardening, cooking, fishing,
etc.). The doginala tip of human life consists in old age or elderhood (male toboma,
female naboma), when parents’ contributions to adult children emanate chiefly from
their minds in the form of magical and other secret knowledge. Elderhood ends with the
death of the body and the exit of a baloma soul or spirit as a person’s fruit.
9781444333237_4_009.indd 164
8/7/2010 6:58:32 PM
165
Bodies and Souls
These relationships have direct implications for indigenous notions of personhood
and human agency. Through cooking, the inassimilable peula (‘power’, ‘strength’)
contained in raw yams and other foods is transformed so it can be released and incorporated by humans. Similar to the way prepared tapwala bodies of yams enter a
hearth’s body, eaten cooked food is deposited into a person’s lopola ‘belly’, and from
there its peula is carried by blood to the rest of the body. Partly for this reason, villagers do not regard cooked food as the u’ula base of their lives or bodies; the lower
limbs including the genitals are attributed that status. The trunk above the genitals
including the upper limbs is the literal tapwala middle body, culminating in the
dabwala head or matala eye at the top.
This classification of bodies from lower to upper extremities, as noted above, inverts
the base–body–tip sequence for human sexuality and yam propagation, suggesting the
mutual exclusiveness and inverse directionality of sexual and alimentary roads. Sexual
intercourse is forbidden in gardens, for example, and cooking is prohibited within the
central portion of villages where exchange yams are stored (cf. Malinowski 1932: 60).
However, classification of the lower extremities and genitals as the bodily u’ula base
implies that human as well as other children are fractal recursions of their parents.
A child’s father and mother are said to be the u’ula base from which he or she is produced as the parents’ fruit.
The fate of people’s dead bodies and spirits parallels the divergent roads of seed and
exchange yam conversion (cf. Malinowski 1932, 1992 [1916]; Weiner 1976, 1988).
Released baloma travel initially to Tuma, the world of the dead, eventually to return
to the world of the living as waiwaia ‘spirit children’ and regenerate new humans
much as seed yams reproduce new yam plants (see Mosko 2007). During precolonial
times, although people’s dead bodies were initially buried in the ground, exhumed,
and variously redeposited, they lacked the analogous capacities of seed yams and
baloma spirits to generate new humans. People’s bodies, however, much like exchange
yams, are still effectively converted into other self-similar valuables – pela sepwana
mortuary heaps (see above). Once received by the deceased’s father’s and spouse’s
relatives, pela sepwana are dismantled, with the items composing the tip redeployed as
fruit to create new relations along new reproductive paths.
Agency and Fame
The final expression of the botanical metaphor I shall consider here coincides with the
indigenous theory of personal agency. For just as human beings are the products of
the other self-similar entities and action scenarios I have described, human intentions,
efforts, and results are understood in terms of base–body–tip–fruit recursion.
As my Omarakana associates explain, the u’ula base of a person’s action, as distinct
from the action itself (see below), consists of simuli, which they gloss in English as a ‘plan’
or ‘scheme’. A simuli in a person’s ‘mind’ (nanola) includes his or her plans for a future
action. They insist, however, that a simuli is not synonymous with the English meaning
of ‘plans’, ‘intentions’, or ‘motivations’; it goes much deeper. A person’s simuli consists
of the totality of ‘images’ or elements of ‘knowledge’ (kaikobu or kaikwabu) in his or her
9781444333237_4_009.indd 165
8/7/2010 6:58:32 PM
166
Mosko
mind that is relevant to a specific course or road of action. It is understood that people
acquire their simuli images mainly through the medium of ‘advice’ (guguya) given primarily by parents, especially what fathers give to sons and mothers to daughters.
People can receive images from many sources, as when children receive advice from
relatives and elders besides parents, and any of this advice can contribute to the formation of simuli in later life. But advice received from relatives and elders besides parents
can only point you in the right direction; it cannot guide you step by step on a given
path. For that, people rely on parents, and especially the advice they proffer as kalisam,
or ‘whispers’. The advice parents convey to all their children uniformly is not kasilam.
Kasilam is the much finer and exclusive advice that a parent prototypically whispers
from the tips of his or her lips directly into the chosen child’s ear. Magical spells, for
example, are usually transmitted as kasilam to one child only, although sometimes
they are partitioned among different children. Whispers thus contain the truly important and distinctive images with which children formulate the simuli that truly guide
their lives. Moreover, several of my interlocutors likened parental whispers to fruit,
the personal detachments that parents are expected to transmit to children only in
their dying moments.
To the extent that children’s simuli replicate and regenerate their parents’ images,
children are self-similar to their progenitors. Parental advice gives form to their simuli
as the base (u’ula) of their lifelong personal characters. The images thereby organized
into one’s simuli set the parameters for everything else one senses, experiences, and
accomplishes in subsequent experience.
Now if the images constituting a person’s simuli are complete and nothing from
parents or others has been distorted or left out, it is expected that he or she will be
effective in dealing with the world as presented. When people fail in anything, others
assume that they did not receive appropriate images from their parents or, perhaps, that
the parents possessed similarly impoverished simuli in the first place. For this same reason, villagers regret that fatherless children have no adult male to give them appropriate
paternal images and advice (cf. Malinowski 1932: 166).
The tapwala body of action to which a person’s simuli gives rise qualifies as his or
her kepwakari. In keda terms, one’s kepwakari actions are the actual steps one takes in
life, consisting mainly in corporeal labors (paisewa). The distinctive ways a man works
to create a garden, carve a canoe, carry his children, organize his lineage’s sagali
exchanges, and practice kula are his kepwakari, based in and substantiating his simuli.
Similarly, the manner in which a woman cooks food for her family, manufactures and
accumulates bundles, engages in gossip, and cares for her children are kepwakari
instantiations of her simuli. And it is on the evidence of one another’s kepwakari that
people estimate the character of their respective simuli and those of their parents
before them.
The realized ends or outcomes of a person’s kepwakari actions become his or her
karewaga – variously defined in the literature thus far as ‘authority’, ‘decision’, ‘power’,
‘autonomy’, ‘responsibility’, and so on (Lawton 2002; Munn 1986: 68). Each of
these glosses carries some element of indigenous understanding, but villagers assert
also that the realization of karewaga is most people’s main goal in living, and that
one’s karewaga exists or is realized only in its recognition by other people. Thus karewaga is not the same as the kepwakari actions from which it is produced. Successfully
conducting a kula career so as to acquire numerous renowned shells is an element of
a man’s kepwakari actions. The karewaga of becoming a kula master, however,
involves others’ additional acknowledgment of that success. Only with this recognition
9781444333237_4_009.indd 166
8/7/2010 6:58:32 PM
167
do the karewaga man’s or woman’s decisions have agentive effects on other persons
in that sphere of activity, or more properly in relation to that road or action scenario. Seen in this light, karewaga is the end product – the doginala tip – of the
body of one’s kepwakari actions as performed in accordance with the images of
one’s simuli base.
Persons are understood to enjoy the karewaga they have constructed through their
actions for as long as they live, but with death their karewaga disintegrates and dies. Yet,
as with other action scenarios informed by the botanical metaphor, the death of one’s
karewaga generates fruit containing alternative capacities for regeneration and reproduction paralleling the roads of exchange and seed yams. Upon death, a man’s disintegrated
karewaga in kula, say, will be converted, on the one hand, into a new form – his butula
‘fame’ or, literally, ‘noise’ – which persists and circulates far and wide around the Massim
in the absence of his bodily person; on the other, the dying man’s karewaga effectively
regenerates itself or himself in the form of the whispers released from his lips that become
incorporated in the simuli base characteristics of his children, who accordingly assume his
personal form (see Mosko 1995). A woman’s karewaga is similarly converted into the
fame arising from successful transactions with bundles and skirts or the human children –
waiwaia spirit children reincarnated from Tuma – to whom she gives birth.
Conclusion
From Malinowski’s and subsequent investigators’ ethnography, fragments of the panAustronesian botanical metaphor have been recorded for numerous Trobriand cultural
contexts. In this chapter, drawing upon the notions of fractal self-similarity and recursion, I have illustrated how Omarakanans’ employment of base, body, tip, and fruit is
much more thoroughgoing and systematic than has been noted so far and sheds new
and different light on a wide range of familiar ethnographic imponderabilia. Following
from the immediately preceding discussion, for example, my interpretation of Trobriand
karewaga and its relation to butula fame or renown contrasts with Munn’s interpretation of Gawan kareiwaga as the “ultimate basis [wouwura, cognate to Trobriand u’ula]
of action” (1986: 69; emphasis added). For Trobrianders and I suspect for Gawans
also (with whom they are closely related), the u’ula basis of a person’s actions
(kepwakari) is instead his or her simuli. One’s karewaga is rather the tip or end point
(doginala) of one’s efforts (kepwakari), and the fame (butula) which results is their
fruit (keyuwela) (see also Munn 1986: 55–56, 125). This point underscores the significance that categorical schemata such as base–body–tip have for illuminating indigenous
understandings of personal agency along with other cultural contexts.
Furthermore, the complementary bifurcations of both men’s and women’s agencies
in terms of the specific fruit they produce and contribute add considerable refinement
and clarity to the gendered distinction of historical and cosmic time posited by Weiner
(1976: 20–21, 230–1). Women’s agency in the creation of fame and men’s in the
procreation of children, I suggest, are, as fruit, comparatively short-lived, whereas
men’s agency in the production of fame and women’s in reproducing children are
relatively long-lived. Through their differential fruit, in other words, both women and
men contribute differentially to historical as well as cosmic time.
In the main, I have concentrated on action scenarios or roads additional to the sorts
of botanically inflected origin structures of Eastern Indonesia on which Fox and others
9781444333237_4_009.indd 167
8/7/2010 6:58:33 PM
168
Mosko
have focused. On the kind of evidence presented here, it is likely that those structures
are supplementary to other self-similar dimensions of indigenous classification and
practice. But these recurrences are not merely synchronic replications across diverse
cultural contexts. Every path of base–body–tip–fruit I have retraced contains within
itself the implication of temporal progression in that order, and every fruit generated
by a particular tip has the potential for fractal extension either for reproduction or for
transformation into the base of some further form.
These two kinds of sequential phasing – from base to body, tip, and fruit, and from
fruit to additional base–body–tip–fruit successions – effectively map out the numerous and diverse action scenarios that conform to the bifurcating organization of people’s
cultural classifications and activities. Some recursions are cyclical or continuous – as
when yam seeds as fruit regenerate the base of yam plants whose fruit are yam seeds
once again, when genital bases of adults reproduce human children similarly endowed
for reconstituting further children in their own right, or when children incorporate in
their simuli the advice whispered by parents. Alternatively, the fruit of a given base–
body–tip sequence sometimes initiates conversions into other forms – as when yam
seeds produce self-similar yam plants, when people labor to produce self-similar gardens or yam heaps, or when men’s kula careers are converted into fame that extends
beyond the immediate circumstances of their lives.
Not coincidentally, perhaps, divergences analogous to these have already been formalized for the region in some of the key works of the New Melanesian ethnography –
for example, in Strathern’s (1988: 225–305) contrast between “forms which
propagate” and “relations which separate,” and Munn’s (1986) distinction between
value “replications” or “continuities” and value “transformations.” The analytical
merit of the botanical metaphor in the context of the Trobriands and possibly elsewhere in the Austronesian sphere is that these two complementary roads or action
scenarios are connected by virtue of their fractal self-similarity and recursion.6
Of previous Massim ethnographers, it is probably Munn’s (1986) treatment of
Gawan reproductions and transformations of value which most closely approximates
that which I have attempted here. Those familiar with her materials will have recognized that some of the Trobriand conversions which I have presented as base–body–
tip–fruit recursions parallel Gawan reproductions and transformations of value which
originate from their own wouwura ‘foundations’ or bases. However, from the present
perspective, Munn’s analysis largely elides Gawan understandings of body, tip, and
fruit in virtually all contexts, the only exception being her brief description of the
segmentation of magical spells (1986: 82–84). My concern is that Munn’s otherwise
impressive treatment is heavily burdened by recourse to a body of exogenous conceptualization – namely, the phenomenology of Western philosophy – which has developed independently of the Massim cultural universe. At one juncture, Munn (1986:
11) even concedes that Gawans do not speak about themselves and their experiences
in terms approximating her abstractions. Of course, the same can be said of my use
of “fractal self-similarity” drawn from chaos theory. However, while it is possible that
the success of Munn’s analysis might be due to her phenomenological insights, one
can legitimately ask whether indigenous counterparts such as the Austronesian action
scenario of base–body–tip–fruit, when available, might not yield understandings
which more closely approximate the experiences of Gawans and Trobrianders
themselves.
Realization of this goal leads me back to the conundrum between Munn and
Sahlins over the character and source of the structural wholes in their respective
9781444333237_4_009.indd 168
8/7/2010 6:58:33 PM
169
theorizing. Because Munn’s analysis of Gawan agency and value transformations
engages with only part of the full botanical metaphor (i.e., wouwura bases or foundations to the exclusion of bodies, tips, and fruit), she is freed to proclaim that it is from
people’s experiences of their actions and practices that cultural categories descriptive
of the world (including people themselves) are generated, rather than the reverse.
For Munn (1990: 13), then, “cultural structure” consists phenomenologically in the
formation of space-time through the experiences of relations among events. But if
the full sequence of recursive base, body, tip, and fruit action scenarios are taken into
account in the manner I have outlined in this chapter, there is a certain verification
of Sahlins’ conception of cultural structure consisting of Saussurean values or, as
Munn puts it, in neo-Kantian a priori concepts and “a logic of received cultural categories” – not forgetting that in Sahlins’ fuller framework, such cultural structures
are themselves the historical products of structures-of-the-conjuncture instantiated
in antecedent events. The key difference between Sahlins and Munn is that, according to the former, systemic cultural values exist for actors at the moment of acting,
whereas for Munn they are precipitated in the immediate aftermath of acting. My
main point, however, is that, in light of the pervasiveness of the botanical metaphor
I have outlined, it is difficult to affirm Munn’s emphasis on the generative capacities
of experience alone. If Gawan wouwura bases are systematically followed by bodies,
tips, and fruit as they are among Trobrianders, as seems likely, her value transformations across diverse contexts of space-time expansion accordingly partake of a discernible a priori cultural structure.
While on the evidence presented here Sahlins’ cultures-as-constituted may be
deemed relevant at given movements of agentive action, the dynamic character of
those cultural wholes in their own might also requires some reconceptualization or
refinement. The parts of Trobriand culture that I have described (i.e., the relevant
culture categories-as-constituted at moments of gardening, heaping yams, performing lisaladabu, and so on) are internally structured in terms of base–body–tip–fruit
paths and recurrently, or externally so to speak, conjoined with one another in those
very terms. Trobriand culture-as-constituted in the contexts I have described consists,
in other words, of a fractally structured totality wherein parts and wholes mutually
implicate one another in both their metaphorical form and their metonymical qua
historical enchainment.
Returning to my opening argument, the evidence I have presented here suggests
that the recent eschewal of cultural wholes in favor of the investigation of parts alone
does not, and perhaps cannot, eliminate holism entirely. For all the denial of cultural
wholes and holisms that has overtaken the discipline in recent years, it remains the
case – at least in the Trobriands and I expect elsewhere in Melanesia if not beyond –
that the parts presented to us ethnographically and historically hint at the emergence
of new wholes, on the one hand, while being readily perceivable as wholes in their own
right with their own-self-similar parts on the other.
It is necessary, finally, to situate this conclusion in relation to the findings of some
of the other contributors to this volume. Along with Sahlins, Tsing, Rumsey, and
Friedman, I am critical of the simple assumption held by earlier generations of anthropologists that cultural systems can be usefully conceived in terms of two distinct levels –
cultural wholes and the cultural elements of which they are composed. Whole–part
relationships, or “worlding” in Tsing’s phrasing, discernible at a variety of scales, are
an inherent (if often only implicit) aspect of our anthropological practice and, very
likely, that of the peoples we study. But the observation that those cultural wholes
9781444333237_4_009.indd 169
8/7/2010 6:58:33 PM
170
Mosko
have been open to and regularly engaging with external factors and relationships over
the course of history, as stressed by Sahlins particularly, offers only part of the solution. As I see it, there is also the inverse-like recognition that cultural systems may be
equally dynamic through processes of self-scaling within and beyond their supposedly
external margins. The historical permeability of cultural wholes to which contributors
to this volume have drawn our attention, in other words, can also be seen as an effect
of scaling – if what is conventionally viewed as a “total culture” can be reconceived as
merely a part of some larger, encompassing whole such as modernity, the West, the
Third World, and so on. Trobrianders’ engagements with peoples foreign to themselves – Dobuans, or Europeans – at different junctures in their history are conceived
in terms not dissimilar to those pertaining to yams and yam heaps, pots of cooked
yams and people’s full bellies, or people’s strategies and their ultimate fames. At least
from the viewpoint of Omarakanans, Northern Kiriwina where their village is situated
is the base (u’ula) of the Trobriand universe, the remainder of Boyowa Island is its
body (tapwala), the surrounding islands and communities (Vakuta, Sinaketa, Kitava,
Gawa, Muyuw, Dobu, etc.) are its tips (doginala), and Tuma, and the invisible world
of baloma spirits which infiltrates everywhere and which provides the source of human
reincarnation, is the world’s fruit (keyuwela).7 The limitations of earlier holistic formulations, in other words, of neglecting dynamic relations beyond the bounds of
cultural wholes as well as within them, are one and the same: a failure to appreciate
the significance and agentive potentialities of shifts of scale including, when warranted, self-scaling. Viewed from this perspective, the widening of many contemporary anthropologists’ gaze from discrete cultural systems to ever-expanding external
globalizing tendencies is part and parcel of the sorts of shifts of focus I have pursued
within Trobriand culture. The further beyond wholes we explore, just as the deeper
within them we dig, the more wholes (and parts) we are likely to encounter.
Acknowledgments
This chapter is based on eight months of ethnographic research conducted during July–October
2006, May–July 2007, May–June 2008, and April–June 2009 at Omarakana village in Northern
Kiriwina of the Trobriand Islands. This research was generously funded by the Wenner-Gren
Foundation for Anthropological Research, the Australian Research Council, and the Department
of Anthropology in the Research School of Pacific and Asian Studies at the Australian National
University. I am greatly indebted to support from the Institute of Anthropology, Archaeology
and Linguistics at Aarhus University arranged by Ton Otto which enabled me to develop an
earlier draft of this chapter. Besides Paramount Chief Pulayasi Daniel, I obtained a large share
of the information contained herein through relations with Molubabeba Daniel, Pakalakai
Tokulapai, Tobi Mokagai, George Masulua, Mairawesi Pulayasi, Tokulapai Tamwanesa, Modiala
Daniel, Togugua Tobodeli, Yogalu Vincent, Kevin Kobuli, Mutabalu Tokwasemwala, John
Kasaipwalova, and members of the Kiriwina Council of Chiefs. For other kinds of support, I am
greatly indebted also to Brian Pulayasi, David Mitakata, Nora Youwalela, Babaidou Kweulabuma,
Boyogima Tomuyala, Sanubei Tobodeli, Migiponu Mokagai, and Mary Kasaipwalova. Ton
Otto and Nils Bubandt along with other participants in the July 2007 Sandbjerg conference
provided many valuable comments and criticisms, as have colleagues Ralph Lawton, Gunter
Senft, Fred Damon, Stephen Gudeman, Steffen Dalsgaard, Margaret Jolly, Robert Foster,
Shirley Campbell, Susan Montague, Kathy Lepani, H. E. Charles Lepani, Courtney Handman,
and Simon Coleman. Previous versions of this chapter were presented to the Anthropology
Seminar at the Australian National University and at the 2008 conference of the European
Society for Oceanists in Verona. Several passages contained herein appeared in their original
9781444333237_4_009.indd 170
8/7/2010 6:58:33 PM
171
form in “The Fractal Yam: Botanical Imagery and Human Agency in the Trobriands” (Mosko
2009). I am grateful to Wiley-Blackwell for permission to reprint those materials. Any deficiencies of data or analysis are, of course, my own.
Notes
1 In this chapter, I have focused on seemingly “traditional” Trobriand action scenarios which
some might consider to be irrelevant to the purpose of explicating holism in the contemporary world. However, many if not most of the action scenarios I outline here are still
quite relevant to the experiences and thoughts of Trobrianders of my acquaintance. But
also, I suggest, from the information that I have gathered so far, Omarakanans’ adoption
of Christianity, commoditization, electoral politics, and other forms of modernity have
proceeded in conformity with the bases, bodies, tips, and fruit that I have outlined here. In
other words, the contemporary world presents Trobrianders with new knowledge and
images for incorporating into their ongoing plans, actions, and pursuits. Although I do not
investigate those historical transformations here, I hope to do so in future publications.
And, needless to say, I could not attempt such analyses without a firm understanding of the
action scenarios that have historically preceded them and on which they are based.
2 Gawa is an island to the east of Kiriwina in the Northern Massim. In Omarakanan oral traditions, the residents of Gawa emerged at the same time and place of creation as Trobrianders
(Mosko 2007). Despite later immigrating to their current locations, Kiriwinans regard
Gawans as “Trobrianders” like themselves.
3 According to linguist Ralph Lawton (personal communication), Malinowski’s use of the
apostrophe /’/ as in /u’ula/ signified an extended vowel utterance /uula/, not a glottal
stop. The other main linguist of Kilivila, Gunter Senft (personal communication), insists
that a glottal stop is present in u’ula. I shall employ the conventional spelling, /u’ula/,
leaving the linguists to resolve the exact linguistic status. Also, in the remainder of this
chapter I employ the northern Kiriwinan dialect spellings and, with reference to plant and
body parts, the third person inalienable possessive /-la/.
4 For some dala groups, as reported by Malinowski (1935a: 99–100), the u’ula base is
marked instead at the points where the path from the village intersects the closest border.
5 See Weiner (1976: 79, 80, 104–5) for further indications of the significance of mortuary
bundle exchanges as conversions of yams.
6 Crook’s (2007) recent description of Bolivip culture suggests that the botanical base–body–
tip metaphor may extend beyond the Austronesian sphere.
7 Many Trobrianders I have interviewed reiterate the information given to Malinowski (1992
[1916]), that upon death the released spirits of dead humans enter Tuma, the spirit world,
located on the small island of Tuma to the north of Boyowa. However, ritual specialists and
others including the current paramount chief affirm that this is the standard explanation of
the afterlife that parents give to children. To those who are regarded as knowledgeable about
such matters, the nonsubstantial spirit world, Tuma, is conceived as coinciding with or interpenetrating the material world occupied by visible beings and entities. For these Trobrianders,
Tuma and the spirits of the dead are invisibly all around them (see Mosko 2007).
References
Appadurai, A. (1996). Modernity at Large: Cultural Dimensions of Globalization. Minneapolis:
University of Minnesota Press.
Campbell, S. (1983a). Kula in Vakuta: The Mechanics of Keda. In The Kula: New Perspectives
on Massim Exchange, ed. J. Leach and E. Leach. Cambridge: Cambridge University Press,
201–28.
9781444333237_4_009.indd 171
8/7/2010 6:58:33 PM
172
Mosko
Campbell, S. (1983b). Attaining Rank: A Classification of Shell Valuables. In The Kula: New
Perspectives on Massim Exchange, ed. J. Leach and E. Leach. Cambridge: Cambridge
University Press, 229–48.
Campbell, S. (2002). The Art of Kula. Oxford: Berg.
Clifford, J. (1988). The Predicament of Culture: Twentieth-Century Ethnography, Literature,
Damon, F. (1983a). What Moves the Kula: Opening and Closing Gifts on Woodlark Island. In
The Kula: New Perspectives on Massim Exchange, ed. J. Leach and E. Leach. Cambridge:
Damon, F. (1983b). Muyuw Kinship and the Metamorphosis of Gender Labour. Man (n.s.)
18:305–16.
Damon, F. (1990). From Muyuw to the Trobriands: Transformations Along the Northern Side of
the Kula Ring. Tucson: University of Arizona Press.
Damon, F. (2005). “Pity” and “Ecstasy”: The Problem of Order and Differentiated Difference.
In On the Order of Chaos: Social Anthropology and the Science of Chaos, ed. M. Mosko and
F. Damon. New York: Berghahn, 79–107.
Evans-Pritchard. E. (1940). The Nuer: A Description of the Modes of Livelihood and Political
Institutions of a Nilotic People. Oxford: Oxford University Press.
Fox, J. (1988). Origin, Descent and Precedence in the Study of Austronesian Societies. Public
lecture, De Wisselleerstoel Studien, March 17, Leiden.
Fox, J. (1995). Austronesian Societies and their Transformations. In The Austronesians:
Historical and Comparative Perspectives, ed. P. Bellwood, J. Fox and D. Tryon. Canberra:
Department of Anthropology, Research School of Pacific and Asian Studies, Australian
National University, 214–28.
Fox, J., and C. Sather, eds. (1996). Origins, Ancestry and Alliance: Explorations in Austronesian
Ethnography. Canberra: Department of Anthropology, Research School of Pacific and
Asian Studies, Australian National University.
Gleick, J. (1987). Chaos: Making a New Science. New York: Penguin.
Hutchins, E. (1980). Culture and Inference: A Trobriand Case Study. Cambridge, MA: Harvard
University Press.
Irwin, G. (1983). Chieftainship, Kula, and Trade in Massim Prehistory. In The Kula: New
Perspectives on Massim Exchange, ed. J. Leach and E. Leach. Cambridge: Cambridge
Lawton, R. (2002). Dictionary: Kiriwina to English. Department of Linguistics, Research
School of Pacific and Asian Studies, Australian National University.
Leach, J., and E. Leach, eds. (1983). The Kula: New Perspectives on Massim Exchange.
Malinowski, B. (1922). Argonauts of the Western Pacific. London: Dutton.
Malinowski, B. (1932). The Sexual Life of Savages, 3rd ed. London: Routledge.
Malinowski, B. (1935a). Coral Gardens and Their Magic: A Study of the Methods of Tilling the
Soil and of Agricultural Rites in the Trobriand Islands, vol. 1: The Description of Gardening.
New York: American Book Company.
Malinowski, B. (1935b). Coral Gardens and Their Magic: A Study of the Methods of Tilling the
Soil and of Agricultural Rites in the Trobriand Islands, vol. 2: The Language of Magic and
Gardening. New York: American Book Company.
Malinowski, B. (1992 [1916]). Baloma: The Spirits of the Dead in the Trobriand Islands. In
Magic, Science and Religion and Other Essays. Prospect Heights, IL: Waveland, 149–274.
Montague, S. (1985). Infant Feeding and Health Care in Kaduwaga Village. In Infant Care
and Feeding in the Pacific, ed. L. Marshall. New York: Gordon and Breach.
Mosko, M. (1985). Quadripartite Structures: Categories, Relations and Homologies in Bush
Mekeo Culture. Cambridge: Cambridge University Press.
Mosko, M. (1995). Rethinking Trobriand chieftainship. Journal of the Royal Anthropological
Institute (n.s.) 1: 763–85.
9781444333237_4_009.indd 172
8/7/2010 6:58:33 PM
173
Mosko, M. (2005). Introduction: A (Re)turn to Chaos: Chaos Theory, the Sciences, and Social
Anthropological Theory. In On the Order of Chaos: Social Anthropology and the Science of
Chaos, ed. M. Mosko and F. Damon. New York: Berghahn, 1–46.
Mosko, M. (2007). The Magical Power of Baloma. Paper presented at the annual meetings of
the American Anthropological Association, Washington, DC, November.
Mosko, M. (2009). The Fractal Yam: Botanical Imagery and Human Agency in the Trobriands.
Journal of the Royal Anthropological Institute (n.s.) 15:679–700.
Mosko, M., and F. Damon, eds. (2005). On the Order of Chaos: Social Anthropology and the
Science of Chaos. New York: Berghahn.
Munn, N. (1977). The Spatiotemporal Transformations of Gawa Canoes. Journal de la Societies
des Oceanistes 33:39–52.
Munn, N. (1986). The Fame of Gawa: A Symbolic Study of Value Transformations in a Massim
(Papua New Guinea) Society. Cambridge: Cambridge University Press.
Munn, N. (1990). Constructing Regional Worlds in Experience: Kula Exchange, Witchcraft
and Gawan Local Events. Man (n.s.) 25:1–17.
Sahlins, M. (1981). Historical Metaphors and Mythical Realities: Structure in the Early History
of the Sandwich Islands. Ann Arbor: University of Michigan Press.
Sahlins, M. (1985). Islands of History. Chicago: University of Chicago Press.
Sahlins, M. (1991). The Return of the Event, Again; With Reflections on the Beginnings of the
Great Fijian War of 1893 to 1855 between the Kingdoms of Bau and Rewa. In Clio in
Ocean, ed. A. Biersack. Washington, DC: Smithsonian Institution Press, 37–99.
Strathern, M. (1988). The Gender of the Gift. Berkeley: University of California Press.
Strathern, M. (1991a). One Man and Many Men. In Great Men and Big Men: Personifications
of Power in Melanesia, ed. M. Godelier and M. Strathern, 197–214. Cambridge:
Strathern, M. (1991b). Partial Connections. Lanham, MD: Rowman and Littlefield.
Strathern, M. (1991c). Or, Rather, on Not Collecting Clifford. Social Analysis 29:88–95.
Stewart, I. (1989). Does God Play Dice? The Mathematics of Chaos. Cambridge: Blackwell.
Tambiah, S. (1990). Malinowski’s Demarcations and His Exposition of the Magical Art. In
Magic, Science, Religion, and the Scope of Rationality. Cambridge: Cambridge
Wagner, R. (1991). The Fractal Person. In Great Men and Big Men: Personifications of Power
in Melanesia, ed. M. Godelier and M. Strathern. Cambridge: Cambridge University Press,
159–73.
Weiner, A. (1976). Women of Value, Men of Renown. Austin: University of Texas Press.
Weiner, A. (1988). The Trobrianders of Papua New Guinea. New York: Holt, Rinehart and
Winston.
Weiner, A. (1992). Inalienable Possessions: The Paradox of Keeping-While-Giving. Berkeley:
Young, M. (2004). Malinowski: Odyssey of an Anthropologist. New Haven: Yale University
Press.
9781444333237_4_009.indd 173
8/7/2010 6:58:33 PM
9781444333237_4_009.indd 174
8/7/2010 6:58:33 PM
Part 3
Beyond Structural Wholes?
Encompassment, Collectives,
and Global Systems
p03.indd 175
8/7/2010 7:21:35 PM
p03.indd 176
8/7/2010 7:21:35 PM
10
Structural wholes, one might say, began with Marcel Mauss. This is of course wrong
in both a logical and a chronological sense, for before Mauss there was his uncle,
Durkheim, and beside Mauss there were many other sources for French structuralism:
de Saussure, Jakobson, Lévi-Strauss, and Trubetzkoy, to mention but a few. At the
same time, it is striking how Mauss is represented as an explicit and decisive inspiration by many of the central figures of structuralism. Claude Lévi-Strauss, the founder
of structuralism in anthropology, wrote an introduction to Mauss that hailed him as
the founder of “modern ethnology” and a precursor to his own structuralism (LéviStrauss 1987). Louis Dumont authored a short essay in which he did the same
(Dumont 1986). And in his book The Enigma of the Gift, Maurice Godelier remembers his reaction to reading Mauss’ most famous essay, The Gift (1990 [1924]), for
the first time: “I felt I had suddenly emerged onto the bank of an immense tranquil
river bearing along a mass of facts and customs plucked from a multitude of societies”
(Godelier 1999: 6). The first chapter of Godelier’s book is more than 100 pages long
and is entitled “The Legacy of Mauss.”
This explicit influence of Mauss’ thinking on not only Claude Lévi-Strauss, Louis
Dumont, and Maurice Godelier, but also Mary Douglas, Pierre Bourdieu, and even
Jacques Derrida (1992), appears to a large extent to have either obviated the figure
of Émile Durkheim or at least passed from Durkheim through Mauss. Even though
Lévi-Strauss, for instance, paid homage to Durkheim’s Elementary Forms of the
Religious Life (Durkheim 1971 [1912]) by entitling his first book The Elementary
Structures of Kinship, the substance of the analysis was a sustained application and
extension of Maussian exchange theory (Lévi-Strauss 1969). Indeed, Gane has suggested that Lévi-Strauss “always makes a detour round Durkheim by appealing
directly to Mauss” (1990: 128). Similarly, Dumont’s inspiration from the group of
people around the journal L’Année Sociologique, founded by Durkheim in 1898,
came primarily through Mauss (as well as from Hertz). This was in many ways obvious, since Durkheim had died in 1917 and other members of this circle, including
Robert Hertz, had been killed during World War I. But Mauss was not merely a
broadcaster of a Durkheimian vision. He was a charismatic figure who made his own
distinct contribution. Indeed, as a biographer of Dumont notes, Dumont’s roots in
the Durkheimian school were “based chiefly on his discovery of Mauss, whose courses
c10.indd 177
8/7/2010 7:17:38 PM
178
Bubandt and Otto
he followed and who seems to have inspired him more than anyone else in Parisian
intellectual circles at that time” (Parkin 2003: 2).
So why this enthusiasm for Mauss and partial forgetting of Durkheim in France? As
Kuper notes, there are two strands in the Durkheimian legacy: “one which Mauss followed, while Radcliffe-Brown took up the other” (1983: 52). Radcliffe-Brown, one
might say, brought Durkheim to England where it turned into his variety of “organic
holism” (see Chapter 14), while Mauss became the tipping point for the transformation of Durkheimian theory into “structural holism” in France. It would no doubt be
wrong to overdraw this difference. Mauss’ book The Gift was introduced to British
anthropology by Evans-Pritchard, who also provided a very positive foreword to its
first English translation in 1954 (Mauss 1974; see James 1998).1 And Mauss’ own
thinking was of course closely intertwined with the legacy of Durkheim. Mauss both
co-published a book on classification with Durkheim and edited a posthumous work
of the latter on socialism (Durkheim and Mauss 1928, 2007 [1903]). Nevertheless,
structuralism in France entailed, it would appear, a certain defocusing on Durkheim in
favor of Mauss. The reason for this may be both political and topical. The entire group
of scholars around Durkheim leaned toward socialism and saw their scholarly work as
also a matter of public policy. But Mauss’ sociology was arguably more radical than
that of Durkheim (Gane 1990). This was no doubt related to the topics of research he
chose. While some of the topics were very much in line with those of Durkheim, such
as magic, sacrifice, and classification, others – the body (Mauss 1979 [1935]), reciprocity, and the person (Mauss 1985 [1938]) – introduced something novel.
Critique of Modernity
The Gift, the most influential of Mauss’ works, explained reciprocity as a system of
“total prestations” or gifts. But enmeshed in this analysis was not just a social theory
but also a vocal political statement. The theory of the gift, as Mary Douglas has
pointed out, “is a theory of human solidarity” (1990: x). By showing that social solidarity was organized through the obligations of reciprocity that were at once economic, religious, political, and moral and that seemed to create solidarity even as
individual actors strived strategically to make their own best of these exchanges,
Mauss, it appeared, had found a strong empirical argument against utilitarian economics and liberal individualism. Economics was not, it seemed, a struggle built into
human nature for individual betterment over others, and humans were not an “economic animal” (Mauss 1990 [1924]: 76). Mauss’ analysis of reciprocity seemed to
provide a powerful argument for the anti-utilitarian position. As Douglas has pointed
out, Mauss and others in the group around Durkheim had inherited their anti-utilitarianism from French moral philosophers such as Rousseau and Tocqueville who
accused utilitarianism of having an impoverished conception of the person, social relations, and liberty (1990: x). As a consequence, “individualism” had a bad name
amongst French socialists like Mauss and Durkheim whose research projects aimed at
discovering those social mechanisms through which individual thought and action
were organized – mechanisms such as ritual, religion, classification, and reciprocity.
Mauss’ analysis of reciprocity seemed to deliver such an attack on utilitarian ideas of
human nature and the iron laws of economics by suggesting that economics needed
to be integrated with the social, the political, and the moral to be understood in
c10.indd 178
8/7/2010 7:17:38 PM
179
earnest, and that human strategies were organized by obligations that engendered
social integration even as they allowed for individual maximization (Caillé 1998).
In this sense, Mauss’ notion of exchange as “a total social fact” (Mauss 1990
[1924]: 78) – an organization of obligations that were “at the same time juridical,
economic, religious, and even aesthetic and morphological” (79) – was both an
attempt to identify a human universal and an explicitly rebellious political statement
against a particular conception of modernity and the notions of the individual, human
nature, and value it entailed. Essentially, it was modern capitalism that was barbaric:
[I]t is not in the calculation of individual needs that the method for an optimum
economy is to be found.… The brutish pursuit of individual ends is harmful to the
ends and the peace of all, to the rhythm of their work and joys – and rebounds on the
individual. (Mauss 1990 [1924]: 77)
The structural holism that Mauss in particular made possible entailed, in other words,
a transcendence of the limits of modernity itself, a transcendence that was at once
political and epistemological. This holism was thus analytically necessary. Phenomena
are important to sociological analysis only to the extent that they are understood as
contextual wholes:
They are whole “entities,” entire social systems, the functioning of which we have
attempted to describe.… It is by considering the whole entity that we could perceive
what is essential, the way everything moves, the living aspect, the fleeting moment when
society, or men [sic], become sentimentally aware of themselves and of their situation in
relation to each other. (Mauss 1990 [1924]: 79–80)
Holism was, however, also politically and epistemologically necessary in order to transcend the limits of modern individualism and the impoverished notions of the person,
of social relations, and of liberty it entailed:2
The victory of rationalism and mercantilism was required before the notions of profit and
the individual were given currency and raised to the level of principles. One can almost
date – after Mandeville’s The Fable of the Bees – the triumph of the notion of individual
interest.… It is our western societies who have recently made man an “economic animal.” But we are not yet all creatures of this genus. (Mauss 1990 [1924]: 76)
The political implications of this deconstruction of modern Western ideology were
obvious in France. Mauss’ invitation to trace the “triumph” of the notion of individual interest and Homo economicus was thus one that Louis Dumont would later take
up and pursue in detail (Dumont 1977, 1986). But no such attention to the politics
of Mauss (and Durkheim) accompanied their influence in the rest of the world. As
Hart has noted, the political aspect of Mauss was largely ignored in the AngloAmerican world (2007). In contrast, it was integral to the intellectual heritage of
Mauss in France (Fournier 2006). This difference between the French and the AngloAmerican reception of Mauss might help explain two distinct genera of holism that
grew out of the Durkheimian tradition. In Britain, Radcliffe-Brown’s “social wholes”
were academic rather than political. They were empirical entities organized around a
specific “functional unity” (see Chapter 14, this volume). In French anthropology, on
the other hand, the structural holism made possible by Mauss was part of a thought
experiment in political and epistemological critique.
c10.indd 179
8/7/2010 7:17:38 PM
180
Bubandt and Otto
Structural holism, one might say, incarnated a kind of critical politics that rebelled
against modern individualism at the same time as it opened up to a research agenda
for an investigation into domains and regions of the world where the modern constitution of rationalism and utilitarianism did not match social practice. According to
Mauss, one would be as likely to find such domains of exception among the elites and
working classes of the Western world as amongst the “archaic societies” he himself
had analyzed and compared. French structuralism as well as its heir, poststructuralism,
pretty much followed this research agenda of reexamining the genealogy of Western
self-understanding and its exceptions. Structuralists and structural Marxists (of which
Claude Lévi-Strauss, Louis Dumont, and Maurice Godelier could be cited as examples) concentrated mainly on discovering the exceptions to modern individualism in
other places of the world, while “poststructuralists” like Pierre Bourdieu, Jacques
Derrida, Gilles Deleuze, Michel Foucault, and Bruno Latour deconstructed the genealogy and micropractices of Western self-understanding and individualism at home.3
Poststructural Critique
Structuralism entailed, it would seem, a particular kind of holism that had critical
potential. It also, however, had obvious problems. The well-known criticism of structuralism is that it turned structures into transcendental agents (thereby ignoring
human agency), that it was rationalist (and logocentric), and that it peddled in exoticism, unnecessary abstraction, and mentalism (e.g., Bourdieu 1977; Appadurai 1988).
Much of this critique has come from poststructuralism. But the irony of poststructuralism is that it tried to eliminate structuralism while ensuring in many ways the continuation of structuralist thinking. In a sense all anthropologists today are structuralists,
if by that we mean the analytical tendency to see phenomena as the emergent and
changing properties of relations rather than as sui generis. However, it is unclear to
what extent the strong criticism leveled at structuralism also eradicated (the possibility
of) structural holism. It is thus an open question to what extent poststructuralists are
also post-holists. On the one hand, poststructuralist and postmodernist thinkers like
Foucault, Derrida, Deleuze, Lyotard, and Latour called on philosophy and social
theory to abandon the holistic and totalizing thought of Western ontology. Their
effect on anthropology was a “deconstructive turn” of the discipline that insisted on
making analysis one of emergence rather than becoming, to focus on particulars and
exceptions rather than on general rules, and to be interested in fragments rather than
wholes. As a result, poststructuralist concepts have often been employed in a critique
of the anthropological holisms of the past (see also Tsing, Chapter 4, this volume).
On the other hand, poststructuralists – despite their critique of structuralism – still
operated with forms of structural holism, and critics of poststructuralism often accused
its concepts and analytics (of objective structures, power, discourse, or habitus) of
implying new kinds of totalizing and deterministic wholes (Jenkins 1992; Sahlins
2002). This ambivalent oscillation between the rejection and reinvention of holism is
evident, for instance, in Abu-Lughod’s call for “ethnographies of the particular” that
seek to trace the “discourses” and “practices” of our interlocutors rather than trying
to portray their “culture” (Abu-Lughod 1991). Poststructuralist concepts such as
discourses and practices, Abu-Lughod suggests, will – because they highlight the
emergent, situated, and fragmented – work against the boundedness and idealism of
c10.indd 180
8/7/2010 7:17:38 PM
181
cultural holism. At the same time, however, these concepts still intimate a holistic
perspective. Abu-Lughod in this sense attempts to solve the inherent problems of
cultural holism (see Chapter 6) through a cautious engagement with (post)structural
holism. Abu-Lughod’s appeal to a “new” kind of ethnography – which like other calls
beside it was inspired by some version of poststructuralist thought – therefore poses
the question of how the critical potential of structural holism can be made relevant to
the contemporary world without subscribing uncritically to its politics, or to the kinds
of wholes it implies. Might there, in short, be (post)structural forms of holism that
imply neither deterministic nor fetishizing wholes?
Louis Dumont and the Chapters
The three contributors to this section offer their reflections on how this might be
possible. They all tread the path of structural analysis warily. On the one hand, they
follow in the tradition of Mauss by employing the critical potential of structural holism
to extend anthropology beyond the limits of modern individualism. On the other
hand, they also remain critical of its heritage. The contributions by Bruce Kapferer,
Philippe Descola, and Jonathan Friedman thus hint at the breadth and versatility of
structural arguments in contemporary anthropology, which they each in their own
way try to reframe and expand.
One way to discuss their differences is to look at how they view Louis Dumont.
Dumont is discussed in all three chapters, albeit with different outcomes. Dumont is
interesting in this regard because no other anthropologist – in the French tradition of
anthropology or in anthropology in general – is associated so closely with the concept
of holism. Based on his fieldwork in India, Dumont argued he had found there a truly
different and nonmodern way of handling value and organizing social life (Dumont
1980). He proposed that this way of organizing oppositions was not compatible with
the conventional dichotomies of structuralism. In fact, he suggested that the binary
oppositions that Lévi-Strauss argued were the elementary relationships of the nonmodern or “savage mind” were themselves instances of a modern form of thought
because these oppositions were seen as elements on the same level of value. What if,
Dumont asked, opposites were not related on the same level of value but on different
levels? The result would not be opposition, but encompassment. Instead of terms
linked by a dividing line, you would have two concentric circles, the outer circle representing the higher-valued level that opposed and at the same time encompassed the
element of lower value. This, Dumont suggested, was how social life and values were
organized in India, the main example being the caste system. India is, according
Dumont, a prime example of a holistic society, because every element relates and
conforms to the dominant value of social hierarchy and encompassment. Western
society is characterized by the opposed dominant value of individualism, which puts
the highest value not on the whole but on the human individual and its relations to
its possessions (Dumont 1977: 4–5). The values of a society constitute its ideology,
which remains implicit for its members. It is the task of anthropologists through systematic comparison to expose the central values of other societies as well as their own
(1977: 27). This method of comparing ideologies gives anthropology its critical
potential to transcend the ideological constraints of the home society, in Dumont’s
case Western individualism.
c10.indd 181
8/7/2010 7:17:38 PM
182
Bubandt and Otto
Jonathan Friedman points out that Dumont’s holism itself operated on two levels.
Firstly, holism referred to the hierarchical organization of value in societies such as
India. In this sense, it was opposed to Western individualism. Secondly, holism for
Dumont was also a theoretical argument that went beyond particular societies. Holism
in this sense refers to the existence of structured social orders in the world that can be
discovered, whether they have the form of Western individualism or Indian holism.
Friedman uses this distinction between “emic” and “etic” forms of holism as a point
of departure for his own “etic” use of holism to analyze the systemic properties of the
global world order. Friedman argues forcefully for the need to employ such an “etic”
kind of holism to understand the current global situation, including various political
identity projects such as ethnicity as well academic theories used to explain them. To
clarify this use of an “etic” or theoretical conception of structural holism, Friedman
returns to the father of it all: Claude Lévi-Strauss. A structural relationship for LéviStrauss was not a relationship between two (op)positions but the relationship between
relations. In this vein Friedman suggests – thereby continuing an endeavor that he has
pursued for several decades – that the world is systemically related in economic and
political terms, and that these relations need to be understood in conjunction with the
changing relations in anthropological theory, popular ideas of identity, and notions of
globalization. The global situation is a kind of structural whole in which politicaleconomic relations and discursive relations are themselves related systematically in a
traceable historical pattern. Friedman calls for a “global systemic anthropology” that
can trace these relations between relations dynamically without being held captive by
a particular bourgeois or Homo academicus perspective on the world. At a time when
anthropology has all but forgotten Lévi-Strauss (but see Sahlins, Chapter 7, this volume), Friedman provocatively argues that global systemic anthropology begins with
the structuralist arguments of Lévi-Strauss. Just like the elementary structures of kinship produced socially different roles and moral norms of behavior, the global differences of identity, wealth, or culture are the emergent properties of larger structural
processes of the global system seen as a whole. Friedman concludes his analysis of
these global structures by leveling a powerful critique at the “anti-holism” of the last
three decades. If it is true, as this volume argues, that holism has a particular historical
trajectory, Friedman suggests that the same also holds true for anti-holism. The celebration of the fragment, the particular, the ambivalent, the “individual,” and the
hybrid – concepts that often are employed within a critique of holism – so Friedman
suggests, are ideological products of a particular cosmopolitan outlook on the world
at a time of Western hegemonic decline. As such, the “anti-holistic” critique of holism
as being a particular modernist construct is itself firmly entrenched within, rather than
transcending, the limits of Western self-understanding.
Philippe Descola takes up Friedman’s distinction between “emic” and “etic” conceptions of holism in his treatment of Dumont toward the end of his chapter. Descola’s
chapter is a wide-ranging and magisterial elaboration of the four ways in which humans
and nonhumans can interrelate and can be imagined as “collectives.” For Descola,
each of these four kinds of “collectives” (animism, totemism, naturalism, and analogism) constitutes a complete ontology and epistemology for organizing and knowing
the world and “for passing judgment about identity.” The last one of these models,
analogism, overlaps with Dumont’s notion of hierarchical holism. For Descola,
Dumont’s analysis of value does not represent all of humanity outside the modern
West, but only a logical third. Apart from the naturalism that characterizes modern
Western thought (if not necessarily all practices in the West), Dumont’s hierarchical
c10.indd 182
8/7/2010 7:17:38 PM
183
or “holistic” way of organizing thought and society (what Descola prefers to call
“analogism” and sees as an ontological logic that characterized Europe up until the
scientific revolution in the Renaissance as well as China, India, Western Africa, and the
South American high civilizations) shares the non-Western world with two other
ontological constructions of life, namely, animism (found in Amazonia, in North
American Indian and Siberian societies, as well as in Southeast Asian and Melanesian
societies) and totemism (found amongst Australian Aboriginal people). In other
words, the holism of Dumont is entirely “emic” in Descola’s view. It is an ontology
that characterizes India – and here Dumont’s work has made an important contribution. But while the hierarchical way of organizing social life and social ideas about the
world can also be found elsewhere, it is not a universal way of thought. Dumont’s
holism in this sense is not universal, but situated. Although he thus disagrees with
Dumont about the universality of holism, Descola nevertheless continues the project
of both Dumont and Mauss by seeking to transcend the limits of Western thinking
through a combination of systematic empirical comparison and sustained epistemological critique. Descola, one might say, “out-transcends” Dumont. This is somewhat
of a feat, since Dumont argued that he was transcending the implicit modernism of
Lévi-Strauss’ notion of oppositions. Lévi-Strauss, in turn, saw the notion of oppositions as the elementary form of his structural theory that also was devised as a way of
transcending modern thought and gaining access to universal thought. Thus French
structural anthropology can be seen as a continuous intellectual struggle to transcend
the limits of Western ideology, found not only in Western society at large but also in
the works of its own founding fathers and practitioners. If Dumont’s theory of hierarchy criticized Lévi-Strauss for remaining modern when he thought he had found
the universal model of thought in oppositions, Descola argues that Dumont’s theory
of hierarchy, rather than being a universal model of thought and social organization,
is representative of just one kind of society.
Bruce Kapferer in his contribution begs to differ. For Kapferer the holism of
Dumont is both an empirical descriptor of a certain social order and an analytical
approach relevant to all social realities. It is this very tension between being an empirical designator and an analytical tool that for Kapferer gives Dumont’s holism its theoretical value and critical potential. The strength of Dumont’s holism is, according to
Kapferer, that it insists on being a social analytic that clearly continues Marcel Mauss’
project of critically exploring the social conditions of humanity through empirical
ethnography without being analytically blinded by modern preconceptions. It is these
preconceptions – notably, the tendency to categorize the world in terms of simple
oppositions rather than through encompassment and hierarchical levels of value – that
haunt much science, including most other forms of holism. For example, this oppositional thinking undergirds the scientific and institutional dichotomies between the
social and the biological, the individual and the structural, the modern and the traditional, the whole and the part. Most other kinds of holism that seek to unite the social
with the biological (such as the scientism of the four-field approach; see also Parkin
and Ulijaszek 2007), structure with agency (such as the practice approach; see Chapter
6), or the social whole with the social part (such as British structural-functionalism;
see Chapter 14) are thus all trapped within modernity, despite their own disclaimers.
Dumont’s holism is for Kapferer thus a social analytic that employs the logic of
encompassment to critique modernity’s tendency to insist on the separation of morality and logic as the basis for a “universal rationality.” It is this separation that gave us
modern society but that now blocks an anthropological view of other social forms of
c10.indd 183
8/7/2010 7:17:38 PM
184
Bubandt and Otto
life because it begins and ends in opposition – oppositions between modernity and
tradition, rationality and emotion, “us” and “them” (see also Kapferer 2003). Kapferer
therefore argues against those critics of Dumont who accuse him of Orientalism, of
exoticizing Indian thought, and of rigidly opposing holism to individualism. Instead,
Kapferer insists that for Dumont, Western individualism is merely a transformation of
a general hierarchical principle of social differentiation, which finds its perhaps most
complex form in India but is not restricted to it. For Kapferer, these critics are themselves caught up in the modern logic of opposition when they insist on relegating
holism to India only to then dismiss it.
One might say that Dumont’s notion of modernity in Kapferer’s reading comes very
close to that of Bruno Latour in his book We Have Never Been Modern (1993). For
both Dumont and Latour, modernity is built upon a logic of opposition between itself
and the rest, between nature and culture, between subject and society, oppositions that
are foundational to modernity’s understanding of itself, its “constitution,” but that
constantly collapse in daily practice. For Latour, oppositions constantly collapse into
the social production of “hybrids” (exemplified by phenomena such as pollution or
terrorism which are as moral as they are political), while modernity for Dumont is a
constant and awkward struggle to deal with social forms of value encompassment
(such as when the value of freedom is encompassed by security as a superior value that
protects freedom or when racism is produced only to be repressed by the value of
equality). But for both Latour and Dumont, modern ideology is based on a particular
denial of the logic of social practice. Kapferer insists that Dumont’s notion of holism is
an analytical device that allows anthropology to pry open this space between ideology
and practice and to allow an escape from the strictures of Enlightenment thinking and
its egalitarianism, while he rejects Latour for his flat, nonholistic approach.
Kapferer has pursued the Dumontian analytical agenda through ethnographic
investigation and analytical contrast for several decades (Kapferer 1988, 1997). In this
chapter, he takes a reflexive step back to look at the analytical potential of Dumont’s
holism for recapturing anthropology as a “major” rather than a “minor science” (cf.
Chapter 3: 49, note 9). For Kapferer, Dumont’s holism holds out the promise for an
anthropological investigation of social “being-in-the-world” as well as of the different
conditions of power that is not restricted by modern ideology – an anthropological
science of what it means to be human that insists on seeing modernity as the exception.
In this sense, Kapferer describes Dumont as “one of the last Moderns” who insisted
on a science that was not predetermined by Enlightenment thinking. Anthropology,
Kapferer would argue, is a “major science” because it holds on to the Enlightenment
project of establishing knowledge of what it means to be human in all its diversity,
without being swayed by the ethnocentrism of Enlightenment thought and its pretensions to representing the universal. Kapferer ends his chapter by suggesting that
some of the global transformations within modern ideas of power, statehood, and
governance entail that the holism of Dumont needs to be augmented by new kinds
of holism, in particular the nonessentialist, emergent kinds of holism that grow out of
poststructuralist thinking. The strength of these poststructuralist kinds of holism –
proposed by people like Gilles Deleuze, Manuel Castells, and Alain Badiou – is that
they represent forms of “open holism” that are antiessentialist and critically reflexive
about their own modernity but still have pretensions to global applicability. Kapferer’s
chapter is thus a strong argument for the need for continued experiments in holism –
experiments that simultaneously seek to capture life in a changing world and to
recapture anthropology.
c10.indd 184
8/7/2010 7:17:38 PM
185
Notes
1 One can read the change in the reception of Mauss in Britain through the way his essay The
Gift was translated. The subtitle of the first translation, to which Evans-Pritchard wrote the
introduction, was “Forms and Functions of Exchange in Archaic Societies” (Mauss 1974).
However, when the second translation appeared in 1990, with an introduction by Mary
Douglas, the word “function” had been replaced by “reason,” thus marking a shift from a
functionalist to a structuralist reading of Mauss (Mauss 1990 [1924]). It was arguably only
with this reading that Mauss moved out of the shadow of Durkheim (as read through
Radcliffe-Brown) in Britain (see also Chapter 14). We will use the latest translation to highlight this structuralist reading of Mauss.
2 From a longer historical view, this idea of holism as the opposite of individualism was made
possible by particular conditions of Western thought and grew to a certain extent from
liberal thought itself. See Chapter 1 for an account of the close relationship between liberalism and holism.
3 Pierre Bourdieu, of course, also conducted research in the Western colonial periphery,
among Berbers in Algeria.
References
the Present, ed. R. Fox. Santa Fe, NM: School of American Research Press, 137–62.
Appadurai, Arjun (1988). Putting Hierarchy in Its Place. Cultural Anthropology 3(1): 36–49.
Bourdieu, Pierre (1977). Outline of a Theory of Practice. Cambridge, Cambridge University
Press.
Caillé, Alain (1998). Neither Holism nor Individualism in Methodology: Marcel Mauss and the
Gift Paradigm. Revista Brasileira de Ciencias Sociais 13(38): 5–37.
Derrida, Jacques (1992). Given Time: I. Counterfeit Money. Chicago: University of Chicago Press.
Douglas, Mary (1990). Foreword. In The Gift: The Form and Reason for Exchange in Archaic
Societies, ed. M. Mauss. London, Routledge, vii–xviii.
Dumont, Louis (1980). Homo Hierarchicus: The Caste System and Its Implications. Chicago:
Dumont, Louis (1986). Essays on Individualism: Modern Ideology in Anthropological Perspective.
Durkheim, Emile (1971). The Elementary Forms of the Religious Life [1912]. London: Allen
and Unwin.
Durkheim, Emile, and Marcel Mauss (1928). Le Socialisme: Sa Definition, ses Débuts, la Doctrine
Saint-Simonienne. Paris: Retz.
Durkheim, Emile, and Marcel Mauss (2007). Primitive Classification [1903]. Chicago:
Fournier, Marcel (2006). Marcel Mauss: A Biography. Princeton: Princeton University Press.
Gane, Mike, ed. (1990). The Radical Sociology of Durkheim and Mauss. London: Routledge.
Godelier, Maurice (1999). The Enigma of the Gift. Chicago: University of Chicago Press.
Hart, Keith (2007). Marcel Mauss: In Pursuit of the Whole. Comparative Studies in Society and
History 49(2): 473–85.
James, Wendy, ed. (1998). Marcel Mauss: A Centenary Tribute. New York: Berghahn.
Jenkins, Richard (1992). Pierre Bourdieu. London: Routledge.
Kapferer, Bruce (1988). Legends of People: Myth of State: Violence, Intolerance, and Political
Culture in Sri Lanka and Australia. Washington, DC: Smithsonian Institution Press.
c10.indd 185
8/7/2010 7:17:38 PM
186
Bubandt and Otto
Kapferer, Bruce (1997). The Feast of the Sorcerer: Practices of Consciousness and Power. Chicago:
Kapferer, Bruce (2003). Introduction: Outside All Reason: Magic, Sorcery and Epistemology
in Anthropology. In Beyond Rationalism: Rethinking Magic, Witchcraft and Sorcery, ed.
B. Kapferer. New York: Berghahn, 1–30.
Kuper, Adam (1983). Anthropology and Anthropologists: The Modern British School. London:
Routledge.
Latour, Bruno (1993). We Have Never Been Modern. New York: Harvester Wheatsheaf.
Lévi-Strauss, Claude (1969). The Elementary Structures of Kinship [1949]. Boston: Beacon.
Lévi-Strauss, Claude (1987). Introduction to the Work of Marcel Mauss. London: Routledge.
Mauss, Marcel (1974). The Gift: Forms and Functions of Exchange in Archaic Societies. London,
Routledge.
Mauss, Marcel (1979). Body Techniques [1935]. In Sociology and Psychology: Essays, ed.
M. Mauss. London: Routledge, 95–123.
Mauss, Marcel (1985). A Category of the Human Mind: The Notion of Person; the Notion of
Self [1938]. In The Category of the Person. Anthropology, Philosophy, History, ed.
M. Carrithers, S. Collins, and S. Lukes. Cambridge: Cambridge University Press, 1–25.
Mauss, Marcel (1990). The Gift: The Form and Reason for Exchange in Archaic Societies [1924].
London: Routledge.
Parkin, Robert (2003). Louis Dumont and the Hierarchical Opposition. New York: Berghahn.
Sahlins, Marshall (2002). Waiting for Foucault, Still. Chicago: Prickly Paradigm.
c10.indd 186
8/7/2010 7:17:38 PM
11
Louis Dumont and a Holist
Anthropology
Bruce Kapferer
Holism has at least three distinct uses in anthropology: (1) anthropology as a holistic
discipline in which potentially all branches of human knowledge may be engaged to
understanding the specific practices of human beings, (2) the study of human society
and communities as wholes in which all practices are interconnected and mutually
influential, and (3) holism as a search for the principles whereby assemblages or
forms of human social realities take shape. I will chiefly focus on the last sense,
although the other usages will be addressed, and mainly with reference to the work
of Louis Dumont. His anthropology is distinguished by an attempt to go beyond
holism as a mode of description (e.g., the presentation of societies as institutionally
integrated wholes) and to make it into a methodology for anthropological analysis.
For Dumont, a holist analytical orientation is vital to anthropology as a comparative
project through which anthropology is able to develop both general and specific
theoretical understanding.
Although I will concentrate on Dumont’s holism, my discussion will also place his
approach in the context of more recent holistic approaches upon which Dumont’s has
some bearing. Here it must be stressed that Dumont is committed to the idea that
social and cultural anthropology has a distinct contribution to make among the social
sciences. He is what might be called a “social constructionist” in that he focuses on
the principles which are engaged in the conception, formation, and practice of social
relations. Central to his social constructionism and holism is the understanding that
all orientations of human beings toward existence are in some way or another social
constructions, formed irreducibly in their social relations.1 Above all, this is the case
concerning the action of human beings and their interrelations with other human
beings and with other material objects (animate and inanimate). Dumont’s holism is
sociocentric and, as well, grounded in the concern for the values embedded in relations, integral to their conception and practice. There is some similarity in spirit, if not
in execution, with other holisms recently achieving attention in anthropology extending from the influence especially of the poststructuralist Deleuze (e.g., see Latour
2005; DeLanda 2006). They share with Dumont a focus on relations, an attack on
concepts of the individual and the subject, and their antirelativism, which are aspects
to which I will return briefly in later discussion. However, the major distinction of the
new holism from that of Dumont is its decentering of concepts of the cultural and the
9781444333237_4_011.indd 187
8/10/2010 10:54:08 AM
188
Kapferer
social in anthropology and a corresponding attempt to overcome a disciplinary
narrowness which Latour (2005: 9), for example, sees as primarily a legacy of
Durkheim’s “science of society” perspective. The new holism expresses a concern to
open anthropology to the insights of other disciplines, particularly the sciences, and
to conceive of the social, in a relational sense, as not limited to human beings (hence
Latour’s use of the concept “actant”)2 and tied to human consciousness. The idea of
anthropology as a discipline with a distinct subject matter, the social, and a particular
methodology specific to the social would be anathema, I suspect, to some of the new
holist directions. Dumont, whom I will describe as in many ways the last modernist
(behind his arguments are the specters of Durkheim, Marx, Toennies, and Weber),
would have disagreed profoundly with aspects of these current directions, especially
their flattening out of the concept of the social and their anticonstructionist realism
(Harman 2009). Moreover, it is likely that he would have seen in them shades of the
limitations of what might be called an individualist presentism born of their particular
modernity, which Dumont would have seen as undermining the potential of the
anthropological contribution founded in its comparativist orientation and recognition
of the theoretical significance of human diversity.
Dumont’s Holism: Beyond Marcel Mauss
Dumont’s anthropology takes much of its inspiration from Marcel Mauss rather than
from Emile Durkheim, who is widely conceived as the main influence on social anthropology (in France and Britain) and to a significant though lesser extent North
American cultural anthropology, distinctions which have some historical importance
but are of little worth nowadays. Mauss’ appeal, for Dumont, is his commitment to
ethnographic detail and that anthropological concept and theory should be grounded
in ethnography. Theory is to emerge from the complexities of ethnographic material,
with such theory to be rigorously tested through comparison. Durkheim tended to
establish his conceptual and theoretical orientations prior to the material. Moreover,
Durkheim assumed the universal validity of logically deduced concepts that refused a
thorough connection to a particular historically constituted reality. The comparativism of Durkheim took the character of sorting what on the surface appeared to be
differences into conceptual boxes on the basis of which he noted a range of correspondences leading to various generalizations (e.g., suicide). However, Mauss, as in
his celebrated study The Gift, took up a particular relational principle as it appeared in
one context and then pursued its different permutations in a diversity of other ethnographic contexts. Through his comparisons he devised a general theory arising from
the different understandings that diverse ethnographic knowledge facilitated. It is
Mauss whose method of comparative analysis is central to Dumont’s own development (see Dumont 1986a, 1986b). Most crucial is Mauss’ emphasis on practice and
value in relation to the whole whereby through the whole – the totality, the gift as the
total pre-station – various value relations are defined.
Dumont’s holism is one which asserts that the understanding of particular human
social phenomena should be grasped in relation to the larger totality or whole in terms
of which they are defined. “Holism” in Dumont’s usage refers to the relational value
that encompasses all others that can be conceived of as part of its set. The whole or
totality is relative to issue and problematic, which may be relevant only to a particular
9781444333237_4_011.indd 188
8/10/2010 10:54:08 AM
Louis Dumont and Holism
189
segment of practices in a named territory (e.g., America or India) and may be seen,
through comparison, to apply to practices in otherwise separate political and social
territories. Dumont’s holism is not expressed in the concepts of Society or Culture as
bounded, enclosed totalities.
As I have already implied, in Dumont’s vision social science (and especially anthropology, for it to be worthy of the name) must be founded in the fact that human being
is through and through a being formed out of social existence, which is the premise
upon which any general or even universal claims to understanding can be built. The
unity of human being is in the social nature of human being per se from out of which
human consciousness in all its manifest diversity in thought and practice is formed.3
As such, the social is not reducible as an essence within human beings nor is the social
to be derived from their consciousness but rather the other way around. The social is
the condition or plane of human existence (the presence of others is the sine qua non
of any biological individual human being) so that human beings are first and foremost
oriented to other human beings and come to act in relation to them. In other words,
Dumont stresses human beings as originally relational through which the social foundation of human existence is enduringly affirmed.
Here I note that Dumont would be opposed to reductionism but, I suggest, in a
way that attempts to avoid dualism. Dumont’s holism is avowedly antidualistic and is
engaged to conceive of the parts as being defined and produced out of the principles
that govern the whole as an encompassment of its parts. Thus a common understanding of reductionism can be described in the following string of oppositions:
holism is opposed to reductionism::whole:part::society:individual::structure:agency
The distinction of Dumont’s holism is in his avoidance of such contrasts or his subsumption and differentiation of them within a relational structure of value encompassment. His argument indicates his distinction from other avowedly anti-reductionist
and overtly holist approaches in the social sciences which, nonetheless, he shows are
caught in a paradox of dualism grounded in an oppositional logic whereby the parts
have values independent of the whole which otherwise unites them. This kind of
holism is founded on reduction.
Dumont’s holism finds some of its demonstration in a critique of Marx and
Durkheim. It constitutes a powerful discussion of their modernism which Dumont
argues defeats their generalizing scientific aims. The roots of this failure are the
grounding of the social either in Universal Man (e.g., Marx) or in the individual being
of humankind or in a transcendental notion of the social or Society (e.g., Durkheim)
that logically places the individual as being prior to the social (indeed, presumes the
question of how human being became a social being). These are conceptions that are
grounded in a European and North American modernity which, Dumont argues,
must limit the generality of the sociologies that stem from Marx and Durkheim. This
is so in relation to not only what might be grasped as modern societies (those of
Europe and the Americas, for example) but also what are called “traditional” societies
(i.e., societies whose value configuration is not that of European and North American
realities), often the provenance of anthropology. Here I should stress that although
Dumont would appear to be asserting a dualism (traditional versus modern), this is
not necessarily the case, for his position is that both (what are defined as traditional
and modern) in effect are distinct value configurations with the traditional giving
expression to a hierarchical scheme of value encompassment that is suppressed in
9781444333237_4_011.indd 189
8/10/2010 10:54:08 AM
190
Kapferer
modernity.4 The traditional cannot be reduced to the terms of the value configurations of modernity. Moreover, the analysis of modernity in terms of its own value
scheme simply affirms taken-for-granted assumptions without, in effect, sufficiently
deconstructing their emergence in a historical process and hence their limitation
despite being accorded generalizing and even universal worth.
Dumont’s (see 1977) critique of Marx’s perspective is precisely that it takes the
material and the relation of human beings to the material as being primary (see Sahlins
[1976] for a similar critique that Dumont acknowledges approvingly). The material is
not constituted in the relations between human beings; rather, it is the conditionality
of these relations. The full realization of the social nature of Human Being (which for
Marx is in individual human being rather than in the relations between human beings)
is, Dumont claims, a projection for the future. (For example, The Communist Manifesto
[Marx 1969 (1848)] may be read as linking an ideal prehistorical past – primitive
communism – with an ideal communist society of the future mediated by a desocialized historical material process.) The understanding of social processes is not in and
through the nature of the social. Marx’s concern for the freedom of the individual in
society, the achievement of the formation of a communist society, not only is an ideal
founded in the modernity of a particular history but also expresses dimensions of the
individualism of modernity. Marx’s social holism reveals, for Dumont, individualist
underpinnings that are linked also to its materialism.
Another holism that distinguishes Dumont’s is that of Durkheim. Durkheim’s
famous distinction between mechanical and organic solidarity is, in Dumont’s perspective, an expression of individualism for all the emphasis on the transcendent force
of the social, Society. Durkheim (in The Elementary Forms of the Religious Life; 1995
[1912]) conceives of the origin of society to be in the collectivity of individuals who
have the spark or effervescence of the social (society as the sacred) ignited among
them in the course of intense contact. His notions (in The Division of Labour in
Society; 1933 [1982]) of mechanical and organic solidarity are for all their holism
individualistic conceptions.
Thus the homogeneous small-scale society of mechanical solidarity establishes its
cohesion on the basis of the similarity between individuals. The individual is submerged in the social. However, the social itself is conceptualized as a totalizing individual, the society of mechanical solidarity being the individual writ large. Organic
solidarity is one in which the specialization of tasks places individuals into interdependence within the larger whole of society, in effect liberating individuals within a
heterogeneous order. It is this heterogeneity and the differentiation as well as complementarity in roles and tasks among individuals that are, for Durkheim, the principle of
cohesion in societies of large-scale modernity. One implication of the individualism of
modern systems that Durkheim describes is what might be understood as the partibility of individuals through a diversity of tasks as a key dimension of differentiation
which functions as the cohesive force of society in its entirety.
Overall Dumont’s critique of Durkheim (and Marx) is that his holism and especially
the distinction between tradition and modern are equally grounded in modernity.
The individualism integral to Durkheim’s position realizes the whole as an expansion
of the part. Furthermore, in Durkheim and in Marx there is an implicit simple-tocomplex opposition or progress, whereas in Dumont there is rather a shift between
different orders of complexity. Dumont’s holism eschews that simple/complex
dualism that bedeviled much anthropology and is still implicit in more recent anthropological discourse concerning globalization.
9781444333237_4_011.indd 190
8/10/2010 10:54:08 AM
191
I should add that in Dumont’s holism relative to Durkheim especially, the force
of the shift from the complexity of the traditional into the complexity of the modern is effected historically through reconfigurations of value embedded in social
relations. The social relational and value are inseparable and are in continual historical process, though not along a single trajectory (the modern and the traditional are
not unitary invariant phenomena). In Dumont’s (1986a) discussions, the movement toward modernity is already apparent in what can be regarded as premodern
societies in Europe where the political, for example, moves into a value-encompassing
relation over the religious or status or else enters into unification with it, as in the
warrior papacy of the Renaissance. Moreover, notions of the individual as value –
once evident among marginalized groups (e.g., among the Epicureans or the
Sophists) – begin to emerge in greater dominance as a function of Christianity (that
expands, of course, within an individualism already apparent in ancient Greece) and
schisms developing within it (Eastern Orthodox, Roman Catholicism). All this
occurs well before the major industrialization or urbanism of modernity vital to
Durkheim’s distinction between mechanical and organic solidarity. Dumont’s
holism, more historically sensitive than that of Durkheim, sets out the ground for
the emergence of a particular modernity, that of Europe, from within the hierarchy
of tradition.5
Here I should underline, via recapitulation, Dumont’s assertion that the holism
that he champions is founded in a particular universalism that is foundationally social.
Holisms that are based in notions of Universal Man, or the essential unity of humankind, effectively discover such unity at the level of the individual, usually the biological
individual (conflating the biological with the social). Such an approach is implicit in
the perspectives of Marx and Durkheim, for example, despite the overriding emphasis
they place on the social and upon the individual conditioned in relation to the whole.
Although Dumont is in agreement with their emphasis on a holism relative to a reduction, for example, he discovers in them what might be called a disguised reductionism/individualism which not only reveals their modernity but also further constrains
any claim toward universal understanding. Thus Dumont distinguishes his approach
not only because of its insistence on the foundational nature of the social but also
because his orientation asserts the indivisible unity of the social with value, a value that
is always relational.
Dumont’s holism is thoroughly holist in the sense that human beings both at the
apex and at the base of the systems that he envisages are embraced and conditioned
by what can be grasped as a relation of social/value. The Part is through and through
encompassed by the Whole, and the Whole, as well, is generative of the Parts which
are infused by the Whole. The difficulty with other holisms in Dumont’s perspective
is that many, as with Durkheim and Marx, contradict their otherwise holist positions
by founding them upon a nonholistic base and by conceiving the Parts in a way as
independent of the Whole or as segments of the Whole, as in organismic notions of
society, society as a Body.
I stress a discomfort in Dumont with the common anthropological concepts of
Society and Culture on the grounds of their holistic inadequacy. That is, they are
formed within the assumptions of the individualism of modernity (e.g., Dumont
1994: 3–16). They operate as superficially holistic. Coherence and unity are asserted
rather than demonstrated. The assumption is often of an organismic kind: the social
as a body, or an order of parts in which the whole is greater than the sum of its parts.
Radcliffe-Brown’s structural-functionalism based in his interpretation of Durkheim is
9781444333237_4_011.indd 191
8/10/2010 10:54:08 AM
192
Kapferer
of this kind. Of course, the notions of Society and Culture involve a separation of the
social from value and in anthropological relativism involve their bounding in an individualistic sense (i.e., cultures and societies take on the order of individualized totalized wholes).
Dumont’s Universalist Orientation
Dumont’s orientation is expressly civilizational. That is, what might be conceived of
as particular societies and cultures are diverse formations or assemblages6 within larger
encompassing processes, which inform dimensions of their differentiation. The concept of encompassment in Dumont’s conception is a value principle where one term
in a relation is valued above another. In so far as such value principle is dominant in a
set, it constitutes that which is encompassing. Thus, in Dumont’s (1980, 1986b)
work on caste in India, the Brahmin is valued over the king/warrior (in relation to
outcastes) and is the encompassing principle in the organization and practice of caste
relations: it is the relation of value which governs the relations of caste as a whole.
Dumont’s concept of encompassment should not be confused with other kinds of
holism of a more closed, bounded, and system integration kind such as commonly
attaches to anthropological notions of culture and society. The holism that Dumont
pursues is open to differentiation and variation and is relative to the issue or problem
addressed. It is not directed to the characterizing or essentializing of culture or society
as orders of social relations independently of specific issues or questions addressed.
Dumont is concerned, in his usage, with the relations of encompassing value which
he develops through both an analysis internal to the particular social phenomena that
he is concerned with (e.g., kinship, caste, and the relations between the religious,
political, and economic; see Dumont 1980) and a broader understanding of their
significance through comparison. It is the manner of encompassing processes that is
Dumont’s focus. Thus he locates different logics of encompassment in India rather
than in Europe, which Dumont’s critics would detect as being at the root of his
Orientalism7 in connection with the shift in value that emerges in the context of the
emergence of the individual as value. In this discussion, Dumont, as I have already
intimated, conceives of India and Europe, for example, as themselves variations within
more general processes of hierarchy which he considers gives his approach ultimate
legitimacy as the central method for an anthropology with universal scientific claim.
By hierarchy Dumont refers to what he takes to be virtually a universal whereby human
beings rank their values in relation to that value(s) that commands the totality.8 He
does not mean by hierarchy what is frequently understood in commonsense and sociological literature to be inequality or stratification (see Iteanu 2009; Tcherkezoff 2009).
While hierarchy is general, it is the most overtly expressed in realities where the religious is dominant; hence, he uses a term thoroughly cognizant of its origin in ancient
Greek etymology (hierarches), where it has strong religious connotation.
Europe and India, for Dumont, constitute an underlying unity. His central empirical interest is how they diverged. Thus he concentrates on the fulcrum or point of
differentiation where the idea of the Orient and of the Occident and, as well, where
the modern (i.e., the configuration that achieves its unity through the individual as
value) began to separate from the traditional, or where individualism (the individual
as value) began to emerge from one kind of ranking of value into another. Dumont in
9781444333237_4_011.indd 192
8/10/2010 10:54:08 AM
193
Genesis I and Genesis II in Essays on Individualism (1986a), for example, explores an
argument that has strong resonance with much intellectual discourse in Europe over
the long term. The point of divergence is Greece/Turkey, where Europe becomes
Asia, and this like so many other scholars is where Dumont concentrates on demonstrating that what overtly appears as a contrastive or oppositional difference is a transformational differentiation within a more embracing universality. Perhaps as an overall
project, Dumont’s approach has some parallel with that of André Malraux who starts
his Voices of Silence (1974) with the observation of the similarity and difference
between, on the one hand, the stained glass windows of a Gothic cathedral and, on
the other hand, the patterns and color of Persian carpets. They express as difference
what is underlying both, or as in holist Gestaltist psychology different figurations
upon a common ground. There is an affinity, I think, between Dumont’s position and
that of the so-called perennialists (e.g., Schuon 2007; Burckhardt 2003) who explore
the motifs and practices of the major civilizational religions as transformations upon
underlying unifying themes.
I stress then that Dumont conceives of Individualism in its Western or Occidentalist
form as a transformation or shift in value from within general traditional or hierarchical holistic processes, which India continues to express as a complex, if diverse or
highly differentiated, example. The reason I emphasize this is to underscore
Dumont’s intention to avoid a dualism that his critics routinely accuse him of, possibly because they are dualists which, from a Dumont perspective, is born of their
own modernity. The criticism of Dumont is one from within a modernist position
of dualism, as Dumont might see it. Here, incidentally, Dumont differs from other
“traditionalists,” as they might be called, such as the perennialists to whom I have
just referred and, of course, Hegel. These do see a major cutoff point marking tradition from the modern. In relation to aesthetics, for example, Hegel conceives of
art to be in radical decline in the circumstances of modernity. Civilization is lost.
The perennialists mark the Renaissance as the beginning of a decline with its initiation of secularism (modernism), mourning the secularizing separation of art from
the religious even in the most apparently religious of paintings. There are certainly
traces of a similar devaluation of modernity in Dumont. However, this would take
away from his prime concern, which is to comprehend the full nature of what may
be conceived as the modern as a social-value configuration which realizes its truths
as particular complexities of distortion (see Rio and Smedal 2008). Thus, Dumont
argues that a feature of the individualism of modernity is that it actively suppresses
the hierarchy of value and its relations, yet upon inspection is founded in its persistence. His example is his interpretation of Marx, whose concept of production – as
one of the four basic elements of Marx’s system (consumption, distribution,
exchange, and production) – is encompassing (i.e., the commanding value) of both
the whole and also a part (Dumont 1977).
The Hierarchy of Individualism
Dumont distinguishes what he calls “hierarchical systems” from those that are marked
by individualism, for example, traditional civilizations such as precolonial India as
distinct from the modernity of Europe. Dumont’s comparative discussion involves
their examination as different formations within a universal logic of hierarchy. Societies
9781444333237_4_011.indd 193
8/10/2010 10:54:08 AM
194
Kapferer
of individualism are no less constituted in hierarchical processes as is India, except the
former (individualist societies) overvalue the part at the expense of the whole (or realize the part at the expense of the whole) and are oriented to totalize at the level of the
part as in the relative homogeneity of Durkheim’s societies of mechanical solidarity or
in the imagined societies of modern nationalism.
In the instance of state nationalism, Dumont (1980) indicates, hierarchy insists
itself in a logic of exclusion whereby the unity or holism of the part is achieved by the
rejection or devaluation of difference, which is typically exiled to the margins as in
racism. The systematic exclusion practiced in the apartheid regime of South Africa is
the outstanding example. The racism of apartheid and elsewhere in the West is a paradoxical property of the egalitarian suppression of hierarchy, as it were, that manifests
at the boundaries of its order (see too Dumont’s essay, “The Totalitarian Disease,”
1986a; also Kapferer 1988).
The assertion of equalitarianism in recent nationalisms, in policies of multiculturalism that argue for the equivalent social position of different ethnic communities (conceived as homogeneous unities; Kapferer 1988: 205–7) as well as contemporary
identity politics, is among many other features of modern societies that exemplify
individualist assumptions. They are thoroughgoing shifts in value ingrained in the
production of relations, and are not to be grasped as merely surface expressions of the
overt liberalism of much modernity. In Dumont’s terms they are more deeply embedded in the very logics of the social dynamics of modernity. In particular, they subordinate (flatten and suppress) the hierarchy of value to the equalitarianism of
individualism, which effects its emergence in a diversity of morbid forms, for example
in racism and prejudice (see Rio and Smedal 2008; Kapferer 1988). The exclusionism
of apartheid is an instance of hierarchical value emergent at the boundaries of difference. Much human rights discourse driven in an overt desire for individual equality
disguises hierarchical value and produces it as exemplified in many current criticisms
of Muslim minorities in Western Europe. It may, in fact, be instrumental in the establishment of boundaries it otherwise refuses.
Agamben’s (1998) discussion of shifts in state processes in European and North
American contexts – the continuity of authoritarian, totalitarian dynamics in modernity, and the subversion of social contract notions of the liberal nation-state – has
resonance with a Dumontian holist understanding. In Homo Sacer (1998), Agamben
addresses a process evident in the invasion of Iraq and Afghanistan and the carceral
situation of Guantánamo Bay (itself an excellent example of a Foucauldian surveillance state) in which it appears that power (sacral power?) has achieved a value outside
of and superior to the law. Agamben sees in contemporary circumstance a repetition
with a difference of that which occurred in the ancient Roman Imperium with the
sacralization of the power of the emperor, the realization of the emperor as a god both
above the law and institutive of the legal regulations integral to the production of
political society. In Agamben’s argument, society and its order (the structuring of its
values) are encompassed by two externalities in dynamic tension, on the one hand the
emperor (who is outside and above society), and, on the other hand, those beings
who are effectively below and outside society, without value or the epitomes of negative value, and are not protected by the laws of society, whom he calls “bare life.”
Those who are regarded as bare life are hardly even conceived as being human: they
have no rights in relation to those who are members of society and exist at its perimeters. They are abused, scorned, and exposed to torture and death at the will of those
who, in contrast to those defined as “bare life,” are members of the political order of
9781444333237_4_011.indd 194
8/10/2010 10:54:08 AM
195
society and governed by its laws. This conceptual understanding of the continuation
of the past in the emergent present (as a reinvention and not necessarily as a historical
extension) parallels, if it does not replicate, the hierarchical structure of the India that
Dumont discusses. Brahmin (sacred and pure) and Outcastes (the impure) are externalities whose relation defines the nature of the relations of the Parts within the Whole
and which achieve a particular morbidity in the circumstances of Indian modernity, as
exemplified by the situation of Dalits.
What Agamben describes is the emergence of a rearrangement of the relation of
values within the social order of contemporary states. This is acutely exemplified by a
US presidency (and perhaps neoimperial moves of a Western origin) that appears (in
the hands of George W. Bush) to be outside the law that the president protects, a
president who defines the order of the social through his opposition to “bare life,”
that is, terrorists who are intensely epitomized by the carceral situation of Guantánamo
and who have no or little protection of the law that governs everyday life. In other
words, changes in the processes of allegedly democratic states and their social orders
reveal shifts in the hierarchy of value that is embedded within them that a holist analysis of the kind that Dumont urges may reveal.9
Ideology, Value, and Comparison
In some ways Dumont’s approach – especially with his usage of ideology and his claim
for the universality of hierarchy – parallels Heidegger’s (1962) notion of Being as that
which lies behind practice or, for instance, the ordinary realities of human existence
(Existenz) which conceal Being. Heidegger’s project is to overcome philosophical
metaphysics and to reground philosophy as the investigation of Being or that which
lies behind human practices and which is always implicated within them. This is at the
crux of Heidegger’s deconstructions which inform such followers as Derrida. I should
stress here that I do not wish to suggest that Dumont is doing the same as Heidegger
(I think Dumont would have distanced himself from Heidegger’s project) but to
draw attention to the strong ontological and ontic themes in Dumont’s approach.
In my interpretation the social and the hierarchy of value are virtually ontological in Dumont. These are the universal ground of all human existence. It is this
assumption that underpins Dumont’s indication of the general validity of his holism
as a methodology appropriate to anthropology as a comparative discipline. This is
so both in anthropology’s exercise of ethnographic description and in its concern
to build more general abstract understanding on the basis of its ethnographic
descriptions. Human existence is differentiated as the grounded fact of human
existence and is constantly differentiating in history. Integral to this differentiation, also as a fact of human existence, is the diversity of the social and the various
arrangements of value hierarchy with which the social is inseparably linked.
The differentiation and diversity of social-value arrangements are what Dumont’s
methodological holism is oriented to explore and present. Thus, what he describes as
the individualism of Western modernity and the hierarchical value arrangement of tradition in India are the ontic – historically produced – schemes that emerge from ontological ground. Individualism is a particular forming of the hierarchy of value (even as
this may be refused, for example, in the ideological constructions of individualism and
egalitarianism concerning the nature of the social as primordially equalitarian).
9781444333237_4_011.indd 195
8/10/2010 10:54:08 AM
196
Kapferer
I note that for Dumont, in my interpretation, there is no specific/universal
opposition. Human Being is at once universal, and all humans are social beings
founded in the hierarchalization of value, out of which distinct arrangements of social/
value are historically constituted. Ideology in Dumont’s explications comprises ontic
expressions upon ontological ground. That is, ideology is created in historical processes: Western individualism and, as well, the specific expressions of the hierarchical in
contemporary India. However, a distinction is evident in that Western individualism
rhetorically asserts itself to be fundamental, and the flat and homogeneous realities it
calls forth (as in recent modern state nationalism; see Kapferer 1988) asserted to be
natural and foundational. This is contra Dumont’s argument whereby he sees hierarchy as fundamental and Indian hierarchical ideology to be no less ideological than
Western individualism in the ontical sense though more continuous or extensive from
a hierarchy of value that is present among all human populations. In this sense, in
Dumont’s sociology, the development of theory from historically formed traditional
societies (i.e., relatively marginal to Western history, once the provenance of anthropology) is likely to be more generally valid than the philosophies and sociologies
developed from a highly specific Western and relatively recent historical experience.
Furthermore, I should comment that Dumont’s notion of ideology as this
develops historically is embedded as value in social relations and their ongoing
production. His usage of the concept is different from that of Marx. It is thoroughly ingrained in practice and is not the inverse of the real or its superstructure;
rather, it is integral as the materiality of social/value and vital in organizing the
action of human beings. The ideological in Dumont’s orientation assumes particular configurations of relations in practice that are defined and differentiated in
relation to the dominant value that embraces and infuses the whole. He engages
this holist idea as a method whereby he both analyzes and explores ethnography in
the course of which he establishes the terms for comparison and the limits of the
theoretical understandings that he develops. What defines the whole, or the limits
of the whole, is effectively where his dynamic logic of Whole encompassing Parts
(that are differentiations of value through the Whole) reaches a limit as the organizing and generating principle(s) of social relations.
This is to be analytically discovered and demonstrated rather than assumed or
asserted, although Dumont’s critics might claim that this is precisely what he does not
do. The method of defining the parameters of the whole is effectively hierarchical.
That is, the dominant or key principle or value – that which defines or determines the
whole, including that which appears contrary or oppositional to it – is the encompassing principle. In this sense, the operation of the logic of hierarchical encompassment
can be conceived as leading to the definition of an individualist society where the part
is the whole, having implications for all value and social relations formed within it.
Marilyn Strathern’s (1988) discussion of Melanesian systems could be said to follow
a Dumontian orientation, although it more overtly expresses features of the orientation of Dumont’s critic, McKim Marriott (1976).
Here it might be noted that Dumont’s definition of society is that which is encompassed according to his principle of hierarchical value. Society is not defined independently of this, and thus, for example, the India of his argument in Homo Hierarchicus
(1980) excludes certain societies or communities or practices within the subcontinent. India is not totalized in his conception and, as his own analyses show, is highly
differentiated. All that is Indian is not included in the hierarchical scheme wherein he
understands the relations of caste. Not only are many contexts within India excluded
9781444333237_4_011.indd 196
8/10/2010 10:54:08 AM
197
from his analytical understanding, but also so are some regions such as Hindu Nepal
and Buddhist Sri Lanka that are engaged to the cause of some of his critics. More
broadly through the logic of his hierarchical holism directed to comparison, Dumont’s
method is oriented to detect different configurations of encompassing hierarchical
value. So different configurations of the relation of status to power in Indonesia,
where power seems to encompass or precede the former (although Geertz in Negara
[1980] would appear to present a case close to Dumont’s India example), are not
necessarily contradictions of Dumont’s general approach but perhaps confirmation of
its utility. That is contra those critics who base their work on Indonesian ethnography,
Dumont can be seen as presenting an argument that would exclude the Indonesian
materials from the kind of analysis he has developed for contexts in India. To put it
another way, while Dumont’s method of concerning himself with the hierarchy of
value and the determination of encompassing principles may be relevant to Indonesia,
it is so as a method through which the distinction of specific processes in various
Indonesian contexts can be discovered. Dumont would thus not discount the ethnographically grounded claims of anthropologists of Indonesia but might assert that his
method should establish the analytical grounds of such a difference or distinction.
In Buddhist Sri Lanka power might be regarded as being in a more ambivalent relation to the sacred than in India. The sacred and power are not in complementary
opposition, in the same way as in India with power as the contrary being encompassed
by the sacred. The representatives of the sacred, the Buddhist clergy (Sangha), are in
tension (opposition) with the representatives of power (the kingship, or Raja). They
are so, however, on the plane of power or the political. Both the Sangha and the King
may be conceived of as in complementary opposition, and effectively the King appropriates (as a form of Divine Kingship) the sacrality of the Buddha ideal into himself.
The Sinhala hero-king Dutthagamani is described in the ancient chronicle, the fifthcentury CE Mahavamsa, as a World Renouncer (see Tambiah 1976) who effectively
transcends both the kingship and the Buddhist clergy entering into the position of a
bodhisattva. This he does as his forces conquer the Tamil overlord, Elara, reestablishing Sinhala political and religious hegemony (see Kapferer 1988: 63–4). In the
Dutthagamini story, power is conceived to be in unity with the religious and is the
critical agency in the coming to dominance of Buddhism. It is the value-encompassing
principle, the conception which conditions the order of sociopolitical relations as a
whole. Other principles are also relevant – notions of purity and pollution, for example –
but these are subordinate to the divine power of the king.
Undoubtedly, principles other than those alluded to here might be detected in
terms of a Dumontian holistic orientation. Thus, major village rites in Sri Lanka
project a vision of the whole. That is, their dynamic is in relation to a virtual reality
(i.e., a mythical real within the bounds of ritual practice) that is not the reality of
ordinary nonritual existential circumstances but nonetheless relevant to them.10 One
important such rite (known as Suniyama; see Kapferer 1997) expresses, as well as the
logic of divine kinship discussed above, a thoroughgoing human-centric schema consistent with the idea of the Buddha. The rite I refer to involves a passage from the one
extreme (human being as thoroughly given over to self and self-interest) to another
wherein all such interest and concern with self are exhausted, the key participant in
the ritual occupying a position of self-transcendence at the axial center of world existential realities.11 Within the span of this progress (from absolute self-interest to absolute selflessness) is the space for the ritual emergence of differential multiplicity of the
orders of existence, including human social existence.12
9781444333237_4_011.indd 197
8/10/2010 10:54:08 AM
198
Kapferer
Egregious misconceptions of Dumont’s perspective are those which seem to assume
that he is presenting a universal and undifferentiated view of hierarchical systems, the
Indian contexts he explores being ideal-typical of others (e.g., see Appadurai 1986;
Berreman 1971; Parish 1996). While there is debate based on Indian materials that
questions the particular logic he expounds in relation to the understanding of caste
(the complementary opposition of Brahmin and Kshatriya, or the encompassing value
of the pure as opposed to the impure), such does not generally invalidate his methodological holism as a strategy for anthropological analysis. A great advantage in the
method is that it enables the establishment of the limits of an ethnographically based
argument. It is sensitive to variation indicating the grounds for exception, for example, within a hierarchical universe. His appendices in Homo Hierarchicus provide
demonstration.
Clearly, his understanding of the principles underpinning sociopolitical relations,
specifically caste, concerning India do not apply to Buddhist Sri Lanka, a universe,
nonetheless, where hierarchical logics are at work and at some ideological depth. As
Dumont himself argued (personal communication), his own analysis indicates the
limit of the Indian case and its logical distinction from that of overtly similar processes
in Sri Lanka.
One common misconception of Dumont’s approach should be dismissed immediately. This is that it can be countered by ethnography from other contexts which
on the surface appear to be similar. This is only so after an analysis of the ethnographic
evidence using the strategy of methodological holism has been conducted and the
case shown to be of the same sociological order. The strength of Dumont’s methodological holism is that it indicates the specificity of his Indian case and indeed the
nonrelevance of his particular ethnographic argument to contexts in Indonesia, Sri
Lanka, and elsewhere (Fox 1994). The established difference of these ethnographic
contexts does not invalidate what I have termed his analytical strategy of methodological holism, one great advantage of which is its capacity to determine the logic
of difference and similarity without engaging in prejudgment – a characteristic of so
much anthropological comparison which continually hampers the anthropological
ability to arrive at more general understanding. Dumont’s methodological holism
enables the delineation of ethnographic distinction while simultaneously pursuing
a more generalizing rather than relativizing course.
Holism and Potentiality
Dumont’s holist methodology is not necessarily static, or dependent on notions of
homogeneous or consensual value. The approach that Dumont recommends is distant from others with which it risks being confused such as those based in the study
of systems as normative or as organized around a coherent set of agreed beliefs.
Meaning, so central in the North American cultural anthropological or hermeneutic
senses, is not critical to his approach. Rather, value is relational and its meaning,
interpretation, and the way it operates through practice are open. Dumont expands
a structural approach certainly extending from Lévi-Strauss’s structuralism. However,
his distinction, as should be evident from the foregoing, is to focus on the ranked
encompassing relations of value – a methodological orientation that assumes that
values in relations are never balanced or equivalent (implicit in structuralism) but
9781444333237_4_011.indd 198
8/10/2010 10:54:08 AM
199
hierarchical when conceived through and defined in relation to the whole. The plane
of immanence that is the universality of human existence and upon which the different constructions of human existence arise is not flat but always uneven and originally so. It is hierarchical, even essentially so, as Dumont expresses this with reference
to Hertz and the universal dominance of the right over the left in relation to the
human body as a whole or totality. What I now wish to stress is that the holism of
Dumont based on the hierarchy of value specifies a system which is open rather than
closed, fixed, or static, thus defining a space of multiple potentiality. Here we can
return to the Indian example, where Dumont’s perspective is most vulnerable to
being criticized as depicting a stagnant, traditionalist, conventionalist approach
ignoring the facts of an obviously complex, changing, and diverse reality (Appadurai
1986; Parish 1996).
Much has been written on modernization processes in India that demonstrate the
transforming force of, for instance, egalitarian energies often of a class and ethnic
kind. But the fact of this does not necessarily contradict Dumont’s approach oriented
to uncovering a particular hierarchical scheme and dynamic. Thus, the phenomenon
of Sanskritization in India has been widely discussed as a response to modernization
whereby the inequalities of caste are objectively realized and circumvented by
depressed castes appropriating the Sanskritic practices and value normally associated
with Brahmins. Sanskritization, if an attempt to offset inequalities, in effect also does
so in accordance with hierarchical value. In other words, what can be seen as an
equalization of caste, or negation of a hierarchical caste order through Sanskritization,
is not against hierarchy in Dumont’s conception but is, rather, a potentiality within
the hierarchical schema he proposes.
Much has been written of the apparent resilience of caste in India despite the inroads
of capital, as if the holism of hierarchical value integral to caste is necessarily contradicted and erased by contemporary processes. What may be called the hierarchical
holism that Dumont describes for Indian caste is capable of absorbing a diversity of
changes or new relations and repositioning of relations. It is a scheme that is not
merely oriented to totalities of the past but also one that can gather within its holistic
rubric dimensions of a future that is continually in formation in the course of human
action. Undoubtedly class forces connected to capitalist production (mediated through
colonialism and postcolonial forces of the market, for instance) are crucial in contemporary dynamics. But there are indications that the rearrangements of caste relations
and the modern sociopolitical importance of caste as identity do not necessarily contradict the hierarchy of value associated with precolonial configurations of caste (the
precolonial appearance of things itself being an expression and realization of a potential). The effects of contemporary economic forces are legitimated through the continuing dominance of the holistic value framework (of the dominance of the Sacred
over Power) as implicit in processes of Sanskritization and even in modern anticaste
movements such as the Dalit protests where the appeal to Buddhism rather than
Hinduism indicates, nonetheless, the prevailing encompassing potency of the Sacred.
The holism of the hierarchy of value in which caste relations are defined can be
conceptualized as always in process, opening constantly to new modes of expression.
This is exemplified in the caste and kinship dynamics that surround teyyam rites and
their relevant communities in Malabar in the Indian state of Kerala. Here I draw on
the research of Vadakkiniyil Dinesan (2009). Malabar, of course, has historically manifested the more extreme examples of the kind of hierarchical holism with which
Dumont is concerned. The pure/impure opposition is very much in observance as is
9781444333237_4_011.indd 199
8/10/2010 10:54:08 AM
200
Kapferer
the sacred encompassment of the political. But in the teyyam rites, which are situated
at the shrines of matrilineages within defined caste communities, forms of political
domination are challenged. That is, sacred potency is dislocated from Brahmanic possession and becomes engaged by outside, impure castes. What may be conceived as a
part within an encompassing whole itself takes on an authority and a potency as if it
itself were the whole – which it is for the duration of a teyyam festival.
Thus in Malabar the shrine festivals of matrilineages within specific caste communities transform the communities from parts into wholes expressed in the local apotheosis of deities who manifest the whole (or, in this case, the cosmic totality). That is,
those who gather to a shrine of a specific caste matrilineage, who may come from a
range of castes within the pure/impure scheme, will in certain instances through the
course of the rites be repositioned in relation to the caste who owns the shrine. The
principles of the hierarchical scheme are not altered but for a moment at least the caste
that owns the shrines is effectively in a position of preeminence (which is facilitated in
that the relevant caste is in possession, for the duration of the festival, of brahmanic
potency – the sacrificial fire).
Teyyam are deities that attach to specific matrilineages and may be conceived of as
manifestations of more encompassing divine forms associated as well with dominant
religio-political communities. They are often divine–human hybrid forms that are
highly volatile, capable of shifting allegiance relatively easily. They can move across
territory, often transgressively, and their myths describe them as frequently shifting
allegiance from the dominant to the subordinate or affiliating themselves to the powerful via the mediation of caste members low in the hierarchical scheme of things.
Teyyam can be dislodged from their shrine affiliations by acts of human devotion and
ally their power with other matrilineages. They can shift allegiance not by upsetting
the hierarchical scheme of things, or the relations of value that organize it in reference
to a sacred totality that they can also express, but by effecting or manifesting different
accents upon reality or unfolding possibilities of relations and shifting sociopolitical
positioning continually emergent within the sociopolitical contexts in which a hierarchical scheme of value is in operation.
The context of the teyyam is one in which the complementary opposition and
encompassing relation of the sacred to power (as a Dumontian analysis conceives the
order of hierarchy) describe a relation that is unstable, that is, the encompassing
relation of the sacred to power is at any particular moment not unquestionably
established or set. In Malabar, Kshatriya groups have been in contest with Brahmanic
dominance and this is sometimes reflected in the pattern of teyyam territorial movement and changing shrine allegiance. This tension can be observed as being motivated in the hierarchy of value of the totality in itself. That is, the challenge to
Brahmanic dominance and the conflicts between Brahmin and Kshatriya groups are
motivated in accordance with the principles defining the whole, the conflicts sustaining the whole even as they may appear at any point in a continuing process to
subvert it.
What I have described does not threaten Dumont’s holistic argument as much as
insist on a greater flexibility in its interpretation. Teyyam asserts not only the value of
the encompassment of power by the sacred but also the tension of this encompassment. The sacred can move out of such a relation and remanifest as the potency of the
part (the part as an emergent whole), threatening the entire sociopolitical order and
becoming an agency in the repositioning and revaluing of particular castes within the
larger configuration of castes. In Vadakkiniyil Dinesan’s description during the teyyam
9781444333237_4_011.indd 200
8/10/2010 10:54:08 AM
201
for a specific caste shrine, the members of the caste capture the essence of Brahmanic
sacred and sacrificial potency (they force the gift of fire) which is crucial for the performance of the festival. They effectively take what the Brahmins possess, gain control
of their sacred potency, and by so doing achieve a ritual autonomy in the Brahmanically
dominated hierarchy to which they might ordinarily be seen as subordinate. Through
this action, the performers of teyyam effectively alter their position in the wider sets of
caste relations, at least during the period of the teyyam festival. They can be seen as
circumventing an inferior ritual status by, nonetheless, working with the potentiality
of the system of which they are a part.
Overall, teyyam performances in Malabar express flexibility which resists an overly
rigidified view of hierarchical logic that a Dumontian approach may otherwise indicate.
The terms of the scheme – the encompassing of power by the sacred, the pure/impure
opposition – do not necessitate a view of the whole that is closed and rigid but rather
one that is open and labile. The system does not need essentially to transform into an
individualistic one (the view from outside) for reconfigurations of caste relations and
repositionings of status to occur.13 The kinds of process that teyyam rites describe suggest a diversity of potential trajectories in practice and understanding, some of which
may actively subvert conventional interpretations of Brahmanic domination. Indeed,
teyyam has long been valued by communities in Malabar in this subversive sense. Teyyam
has been an agency in the resistance to state control supported by a Brahmanic,
Sanskritic ideology, and was a ritual institution engaged in political resistance to rigid
views of caste hierarchy used during colonial rule and interpreted through European
ideology involving stratificational understandings of the caste system.
One reason for concentrating on Indian material in the context of a discussion of
Dumont’s holism is because Indian caste has conventionally been taken as emblematic, from a Western perspective, of stasis, stagnation, and dehumanization. His perspective enables a contesting of such a stereotype as well as a fuller understanding of
the forces that gives contexts of the relations of caste resilience even where they appear
to be changing. There is a tendency for his approach to be tainted by the frequently
unexamined assertions concerning caste in India, his method infused with the same
kind of stultification of attitude as the conservative antimodern reality to which it
appears to be applied. But as I have indicated, Dumont’s approach not only can comprehend another modernity, not necessarily reducible to the Western kind, but also
can engage with its diversity and constant creative possibilities. Furthermore, as I have
indicated, Dumont’s holism and stress on the hierarchy of value can also comprehend
dimensions of egalitarian and individualist realities and their potentials. What are frequently opposed are united through a common methodological orientation that is
not locked within the prejudices of one or the other ideological persuasions.14
Dumont the Last Modernist?
Dumont pinned Hegel as being on the cusp of modernism whose profound works
intermingled a traditional holism defined by religious value with a modernist individualism. Hegel’s emphasis on the state as the absolute was an instance of the emergence
of power and the political as separate from the religious and in dominant relation to
it. If Hegel expressed in his own work a turning point in thought, so, in a similar way,
does Dumont. Dumont bridges what is characterized as a modernist approach to
9781444333237_4_011.indd 201
8/10/2010 10:54:08 AM
202
Kapferer
social thought with what is often conceived of as a poststructural movement aimed at
overturning modernism and opening up novel ways of comprehending social realities.
Thus Dumont, critical of much social and philosophical theory vital within anthropology, is yet committed to the kind of grand narrative schema of those modernists of
whom he is critical (Marx, Weber, Durkheim, Freud, and Lévi-Strauss). His vast methodological holist project for anthropology is an alternative within modernism. But
there are intimations of poststructuralism in his approach. I refer to that poststructuralism that in itself is indicating a holism but in a spirit of critique of modernism and
certain postmodern trends that distinguishes poststructuralism. However, I emphasize
that while Dumont anticipates certain aspects of poststructuralism, nonetheless he
would see current developments as bound in modernism (see Ranciere 2006) and as
compounding its difficulties for a comparative anthropology. Regardless, it is worth
considering some of the directions in a poststructural anthropology-sociology to highlight Dumont’s distinction and his continuing relevance in an exciting climate of new
critical debate affecting anthropology and the social sciences.
The major figure in poststructuralism is the philosopher Gilles Deleuze, whom
I will briefly discuss here, and increasingly Alain Badiou (2000), his sympathetic critic.
Deleuze advocates a holism of a kind that attempts to create an approach to the existential processes in which human beings are located that draws together perspectives
from diverse areas of human knowledge involving the arts, humanities and philosophy, and the sciences (e.g., Deleuze and Guattari 1994). Deleuze aims at a synthesis
of a plethora of perceptions and orientations to human realities, including the questions asked as well as the practices engaged. This holism tries to synthesize different
orientations to phenomena and simultaneously to overcome erstwhile antagonisms
and divisions. Effectively a new kind of anthropology of a holistic kind is being
invented that is vastly different from that of the late European Enlightenment idea of
anthropology (see Kapferer 2007), although we are in a time when there is some
attempt to revive even its spirit (see Parkin and Ulijaszek 2007).
In poststructuralist approaches, evolutionist and rationalist assumptions embedded
in past orientations are discarded. Most vitally, human beings are displaced from their
erstwhile centered, privileged position as dictating the terms of discourse. There is an
assault on social constructionist positions, and this would be a major point of disagreement with Dumont. Above all, this is so in the case of some poststructural directions where the primacy of value is denied if not excluded altogether. I will return to
this shortly. However, I note here that the disregard for value is not the course of
Deleuze, whose concepts of coding and overcoding attend to value in the nonnormative usage of Dumont.
There are other parallels between Dumont and Deleuze’s radical poststructuralism.
Among these is avoidance of subject–object dualisms and of the concept of the individual as the elemental unit for analysis, and associated with this is a concern with
relational processes whereby different potentialities of structures (or assemblages in
poststructuralist conceptualization) are emergent at any particular point, moment, or
event. In connection with this approach, the concept of “singularity” is frequently
employed, which comes instead of an emphasis on overarching structural forms or
paradigms (such as those of Lévi-Strauss or Dumont), or the concept of society. Events
in this perspective are singularities taken as a complexity of different dynamics and
potentialities with many trajectories that cannot be reduced to a determinant preexisting structure. This is an idea that is carried to an extreme in poststructuralism (associated with a stress on constant creativity) and that I have indicated is not necessarily
9781444333237_4_011.indd 202
8/10/2010 10:54:09 AM
203
opposed to Dumont’s holism, if the relational scheme is seen not to be fixed but rather
as a domain of potential realized differentially in practice. As I have described, the
ritual events of teyyam provide some illustration, as they can be effectively seen as singularities realizing innovative potential. The practice of the value scheme always opens
to new directions that are not reducible to a fixed interpretation of the value scheme
independent of its practice as event. Dumont’s holism, I suggest, allows for creativity
and innovation and the production of the new. (Deleuze’s two general relational
frames of the arboreal or tree-like hierarchical and the lateral spreading networks of
the rhizome, which are given foundational and almost primordial qualities by Deleuze,
bear some similarity with Dumont’s schemes of hierarchy and individualism.)
With Deleuze, Dumont does not place human consciousness, interpretation, and
meaning as being linchpins from which general understanding is to be developed (clear
in his work on cinema; see Deleuze 1986). The hermeneutic turn in anthropology
(underpinning much North American postmodernism) is not Dumont’s concern or that
of Deleuze. Rather the aim is to explore how the relational processes from within a variety of consciousnesses and perceptions of reality are generated. In both thinkers there is
an anti-organicism, for Dumont particularly that flowing largely from Durkheim’s focus
on the institutional order of society. Moreover, in Deleuze especially, there is a shift away
from the body as image or metaphor of either totalities or their parts – a concern with
“organs without bodies” (see too Zizek 2003) that can form an abundance of different
connections of a nonembodied character or breaking beyond bodily limitation.
Deleuze and his followers have departed from the notions of contradiction and dialectic and the powerful dualism that they demand. For Deleuze, his concept of the fold
is designed to overcome a dualism (implicit in his twin processes of the arboreal and
the rhizomic). Thus the arboreal and the rhizomic in different ways are conceived of
interpenetrating, enfolding each the other, without either realizing a synthesis or necessarily being in contradiction although their dynamics are irreducible to the other.
The idea of the fold is crucial, I think, to the kind of poststructural holism that Deleuze
announces. This holism, of course, is distinct from that of Dumont, but the latter’s
concept of encompassment merits comparison. Through it, Dumont departs from a
Hegelian or Marxian dualism, for example, demonstrating how relational opposites
(the pure and the impure, for instance) are intertwined constituting the dynamics of
the whole (the manifold differentiations and variations of its practical working out).
Overall, the poststructuralism of Deleuze and the orientation of Dumont develop on
the basis of a radical critique of Western modernist approaches, although the former
does so from within its context (revolution from inside), whereas Dumont goes outside
the Western context in line with his firm commitment to anthropology as comparison
and a transcendence of Western value.15 My brief comparison of Deleuze and Dumont
is intended to show some of the parallels and, more importantly, to highlight some of
the significance of Dumont’s project and the fact that he is breaking outside many of
the terms of conventional anthropological discourse that many of his contemporary,
often postmodern, critics still appear to be embedded in, even despite their counterclaims. This observation is not designed to shield Dumont from such criticism but
rather to clarify the grounds of engagement. Often they seem to be at cross purposes.
While there may be similarities between Dumont and certain aims in poststructuralism (more in spirit than in substance), Dumont, I am sure, would be thoroughly critical of certain directions in poststructuralism as this applies to anthropological and
sociological practice. Poststructuralism aims at a thoroughgoing reinvention but as
radical difference of perspectives on the human situation in general and, for
9781444333237_4_011.indd 203
8/10/2010 10:54:09 AM
204
Kapferer
anthropology and sociology, a complete shift away from the established conventions
and traditions of the past. Major figures in such a redirection are Bruno Latour and
Manuel DeLanda, whose similar, if not the same (see Harman 2009), approaches
extend from the Deleuzian project. Dumont, I imagine, would be deeply critical of
them, conceiving of their approaches as subverting the anthropological contribution
born of comparison. But more than anything else, he would see in much of their work
a demonstration of the faults of a Western modernism and, moreover, an extreme
individualism that can only lead to distortion and a systematic failure to comprehend
differentiating processes, indeed the creative capacity of humankind to generate forms
of life subject to distinct as well as changing relations within wholes that themselves
are continually transforming (a potential, in fact, of Deleuze’s program).
Both Latour and DeLanda start their reinventions with reality conceived, at least
initially, as a horizontal plane (perhaps overdetermined in Deleuze’s rhizome). The
concept of value along Dumont’s lines is excluded; there is little notion of a variegated
plane of hierarchical difference in a Dumontian sense to begin with. In other words,
general understanding must start with the fact of comparison and difference, although
the nature of such is to be demonstrated and not assumed – hence the significance of
Dumont’s holism. Latour (2005: 16) is quite explicit on the suggestion of an antiholism in such a perspective in his statement that his actor–network theory (ANT) “has
tried to render the social world as flat as possible” in order to explore the formation
of new social assemblages. DeLanda is even clearer on the matter, who, with Latour,
posits an empirical realism.
DeLanda’s realism, which borrowing from Deleuze he calls assemblage theory,
aims “first of all [to] account for the synthesis of the properties of a whole not reducible to its parts” (2006: 4). DeLanda’s assemblages are “wholes whose properties
emerge from the interactions between parts, [and] can be used to model any of these
intermediate entities: interpersonal networks and institutional organizations are
assemblages of people; social justice movements are assemblages of several networked
communities” (2006: 5). His work, he declares, involves taking
the reader on a journey that, starting at the personal (and even subpersonal) scale, climbs
up one scale at a time all the way to territorial states and beyond. It is only by experiencing this upward movement, the movement that in reality generates all these emergent
wholes, that a reader can get a sense of the irreducible social complexity characterizing
the contemporary world. This does not imply that the ontological scheme proposed here
is not applicable to simpler or older societies: it can be used in truncated form to apply it
to societies without cities or large central governments, for example. (2006: 5)
DeLanda closes off his claim for his social realism (the target being a postmodern
moralism) with the following:
I make no effort to be multicultural: all my examples come from either Europe or the
USA. This simply reflects my belief that some of the properties of social assemblages,
such as interpersonal networks or institutional organizations, remain approximately
invariant across different cultures. (2006: 6)
I won’t bother to deconstruct the foregoing. But in DeLanda’s development from
Deleuze, it looks like modernist individualism has transmogrified with all the dimensionality and more that Dumont critically investigated. Notwithstanding, in my opinion, many of the exciting directions for anthropological analysis that a Deleuzian
9781444333237_4_011.indd 204
8/10/2010 10:54:09 AM
205
direction does proffer, DeLanda outlines a potentiality that smacks of an expansion of
European and North American hegemonic sociological thought together with many
of the gross limitations that Dumont highlighted. It is ironic that the cover of
DeLanda’s book is illustrated by a hive of bees. Mandeville’s The Fable of the Bees, or
Private Vices, Publick Benefits (1988 [1732]) was examined by Dumont as an expression of the individualist turn at the beginning of northern European modernism that
continued through into contemporary anthropology, derailing its potential. DeLanda
might appear as a Dumontian nightmare. In this regard, Dumont’s direction out of a
modernist anthropology and his still radical critique may be sustained both as a sobering consideration of certain poststructural developments and enthusiasms as well as
still offering potential for a comparative anthropology.
A concluding thought. There is a renewed orientation in contemporary anthropology, often highly influenced by poststructural realism and, as well, a certain continuity
of Enlightenment scientific rationalism, to the models of science as ways into the
investigation of sociocultural realities, a direction that in various ways has always been
with anthropology. Much may be gained in such an orientation to science and it may
be increasingly relevant in realities of possible ontological shift, where the separations
between human being and machine, for instance, might be becoming less distinct or
in zones where the boundaries between the human and nonhuman are being blurred
(Hayles 1999). There is no doubt that attentions to the worlds of science and its ways
of conceptualizing reality are likely to be valuable in anthropology – a rich source of
conceptual metaphors as with much of Deleuze. But it is possible that some of the
attention is not innocent. That is, as Dumont himself noted, the orientation to human
beings through scientific models (as the models themselves) is likely to be driven in
the cosmological and ideological (in Dumont’s sense) realities of human social and
value relations. For all the objective truth they may seem to reveal in the areas of
human action, they nonetheless demand reflective attention to the very human realities within which they are produced. It is entirely conceivable that some of these
models and their apparent veracity (as well as the attraction of anthropologists to
them) comprise a dimension of the very modernism and individualism of the worlds
in which they are invented.
Acknowledgments
I have benefited from the criticisms of Nils Bubandt and Ton Otto. Andrew Lattas has assisted
with critical discussion. Roland Kapferer has given this chapter a close reading and has suggested important improvements.
Notes
1 Here I should stress that Dumont is not a subjectivist in his social constructionism (e.g.,
Berger and Luckmann 1966) but treats the social as a given. It is the sine qua non of
human existence. Through the concepts and values that are embedded in social relations
and are continually being formed in their process, human beings come to both reflect
upon and comprehend the various dimensions of their existence. The social is integral to
the shapes of human comprehension and understanding of their realities.
2 Deleuze (see Deleuze 1986) employs the concept of actant in the context of his poststructuralist development.
9781444333237_4_011.indd 205
8/10/2010 10:54:09 AM
206
Kapferer
3 In other words, social being determines or is the ground upon which human consciousness
develops.
4 Dumont in my interpretation uses the concept of “tradition” as a value configuration that
precedes modern, contemporary value configurations. In my understanding of his work,
the traditional can also be a value configuration in contemporaneous (i.e., modern) circumstances. In this sense, the traditional is produced in history and its import can be
thoroughly contemporary even though the schema of its value configuration can be established as being of great historical depth.
5 Dumont’s perspective could be compared with that of Max Weber, who treats Europe and
China, for example, as ideal-typical polar opposites. They manifest a different rationality.
Dumont conceives them as distinct configurations of hierarchical encompassing value. His
approach is holistic rather than fixing on a distinct value element and effectively essentializing and Orientalizing it.
6 Here I use the concepts of “formation” and “assemblage” together; that is, the modernist,
largely Marxist notion of formation that breaks out of the notion of society as an integrated
totality while maintaining the idea of commanding centers and, the poststructuralist concept of assemblage, which is thoroughly outside the idea of society as an enclosed, bounded
totality in which there is no notion of center at all. Whether the concept of formation or
assemblage is used, should, I think, be relative to empirical context.
7 If Dumont’s approach is Orientalist, it is not of the romantic idealist or negative kind (see
Said 1995). Dumont has attempted to explore India not from within a Western, usually
moralist, standpoint. The concepts that he develops, especially that of hierarchy, are
intended to be of general applicability and to be devoid of a judgmental moral kind.
Dumont attempts to keep his own moral values out of his analysis, values which he indicates are sympathetic to Western socialist ideals.
8 Dumont illustrates his universal claim for the hierarchialization of value in relation to the
whole with reference to Hertz’s (1909) study of the privileged value of the right hand over
the left defined in relation to the body as a whole.
9 Buck-Morss (2007) has made an important criticism of Agamben in regard to his ahistorical logic upon which he asserts that democratic systems are doomed to degenerate into
dictatorial and totalitarian systems. Buck-Morss debates this, arguing that in democratic
systems sovereignty ideally rests with the people and that the recent attempt by oligarchic
interests to represent the people’s sovereignty, to be the icons of populism, is the danger.
It is not a logical inevitability. Dumont (1986a) does not argue for a logic to individualism
independent of an understanding of its historical production.
10 Rituals such as the Suniyama can be understood as projecting an imaginal totality or what
I (Kapferer 2005) have described elsewhere as the virtual, developing from Deleuze. In this
sense the virtual, or the totality as the virtual, operates as the whole in terms of which the
differentiated relations of existence in their actuality and potentiality can be (re)generated.
11 It is more accurate to conceive of the passage of the patient as going to a point external to
existence, or at the threshold of nonexistence, from which point the patient can reconstitute self and world. The patient effectively comes to occupy the position of the World
Ruler or Mahasammata – the world-originating Cosmic King.
12 In the Sinhala Buddhist universe to which I am referring, it is possible to see the concepts
of self and nonself that are integral to the ritual process as supplanting the impure/pure
opposition in Dumont’s Indian cases. It is a similar hierarchialization of value that changes
register. As many have often noted regarding Buddhism, including Durkheim and Mauss,
there is a human-centric individual as value thematic. But this is not an individualism that
necessarily reduces to the Western individualism kind, but rather one that is systematic
with the Hindu/Brahmanic universe of Dumont yet markedly distinct from it. Alternatively,
Buddhism demonstrates a potential within the caste dynamics of India of breaking free
from the oppressive determinism that often seems to mark it. This is a message of the current anti-caste Dalit movement in India.
9781444333237_4_011.indd 206
8/10/2010 10:54:09 AM
207
13 There is a widespread sociological assumption that individualist societies are creative and
innovative because of their stress on the individual. Certainly it is individuals who are creative, but this is likely to be true for everywhere and is not necessarily a consequence of
individualist ideology. The view of individualism as being creative is, in itself, an ideologically fueled opinion. This may also be so with the concepts of change and stasis. The very
idea of change, and change as a particular problematic of individualistically self-conceived
societies, also may be linked to individualist ideology. All situations of human existence are
in constant change and transformation. What may distinguish them is the way this is conceptualized either, for example, as reproduction of the same or as a difference.
14 Although Dumont may not have succeeded (e.g., hierarchy as a commonsense and ideologically loaded term in Western egalitarian individualist discourse), Dumont’s analytical
concepts for anthropological comparison are intended to be independent of moralizing
judgment.
15 This should not overlook Deleuze’s own attempt to go outside Western frameworks of understanding. Of all philosophers, Deleuze (see Deleuze and Guattari 1994) makes prodigious use
of ethnographic materials often collected by anthropologists in an effort to demonstrate the
widespread applicability of his concepts as well as some of the grounds for their derivation.
References
Agamben, Giorgio (1998). Homo Sacer: Sovereign Power and Bare Life (trans. Daniel HellerRoazen). Stanford: Stanford University Press.
Appadurai, Arjun (1986). Is Homo Hierarchicus? American Ethnologist 13(4): 745–61.
Appadurai, Arjun (2006). Fear of Small Numbers: An Essay on the Geography of Anger. Durham:
Duke University Press.
Badiou, Alain (2000). Deleuze: The Clamor of Being. Minneapolis: University of Minnesota Press.
Berreman, Gerald (1971). Review of Louis Dumont: Homo Hierarchicus Man 6(3): 515.
Buck-Morss, Susan (2007). Visual Empire. Diacritics 37(2–3): 171–98.
Burckhardt, Titus (2003). The Essential Titus Burckhardt: Reflections on Sacred Art, Faiths, and
Civilizations. Bloomington, IN: World Wisdom.
DeLanda, Manuel (2006). A New Philosophy of Society: Assemblage Theory and Social Complexity.
London: Continuum.
Deleuze, Gilles (1986). Cinema 1: The Movement Image. London: Athlone.
Deleuze, Gilles, and Felix Guattari (1994). What is Philosophy? London: Verso.
Dinesan, Vadakkiniyil (2009). Teyyam: The Poiesis of Rite and God in Malabar, South India.
PhD diss., University of Bergen.
Dumont, Louis. (1980). Homo Hierarchicus: The Caste System and Its Implications. Chicago:
Dumont, Louis (1986a). Essays on Individualism: Modern Ideology in Anthropological Perspective.
Dumont, Louis (1986b). A South Indian Subcaste: Social Organization and Religion of the
Pramalai Kallar. Dehli: Oxford University Press.
Dumont, Louis (1994). German Ideology: From France to Germany and Back. Chicago:
Durkheim, Émile (1933). The Division of Labour in Society [1892]. New York: Macmillan.
Durkheim, Émile (1995). The Elementary Forms of the Religious Life [1912] (trans. Karen
Fields). New York: Free Press.
Fox, James J. (1994). Reflections on “Hierarchy” and “Precedence.” History and Anthropology
(special issue, Transformations of Hierarchy: Structure, History and Horizon in the
Austronesian World, ed. M. Jolly and M. S. Mosko): 87–108.
9781444333237_4_011.indd 207
8/10/2010 10:54:09 AM
208
Kapferer
Geertz, Clifford (1980). Negara. Chicago: University of Chicago Press.
Hayles, N. Katherine (1999). How We Became Posthuman: Virtual Bodies in Cybernetics,
Literature, and Informatics. Chicago: University of Chicago Press.
Harman, Graham (2009). Prince of Networks: Bruno Latour and Metaphysics. Melbourne: re.
press.
Heidegger, Martin (1962). Being and Time, trans. John Macquarie and Edward Robinson,
New York: Harper.
Iteanu, Andre (2009). Hierarchy and Power: A Comparative Attempt Under Assymmetrical
Lines. In Hierarchy: Persistence and Transformation in Social Formations. Oxford:
Berghahn.
Kapferer, Bruce (1988). Legends of People, Myths of State. Washington, DC: Smithsonian
Institution Press.
Kapferer, Bruce (1997). The Feast of the Sorcerer. Chicago: University of Chicago Press.
Kapferer, Bruce (2005). Ritual Dynamics and Virtual Practice: Beyond Representation and
Meaning. Social Analysis 48(2): 33–54.
Kapferer, Bruce (2007). Anthropology and the Dialectic of the Enlightenment: A Discourse on
the Definition and Ideals of a Threatened Discipline. Australian Journal of Anthropology
18(1): 72–96.
Latour, Bruno (2005). Reassembling the Social: An Introduction to Actor-Network Theory.
Malraux, André (1974). Voices of Silence, trans. Stuart Gilbert. Princeton: Princeton University
Press.
Mandeville, Bernard. (1988). The Fable of the Bees, or Private Vices, Publick Benefits [1732].
Indianapolis, IN: Liberty Fund.
Marriott, McKim (1976). Hindu Transactions: Diversity Without Dualism. In Transaction and
Meaning: Directions in the Anthropology of Exchange and Symbolic Behavior, ed. B. Kapferer.
Philadelphia: Institute for the Study of Human Issues, 109–42.
Marx, Karl (1969). The Communist Manifesto [1848] (trans. Samuel Moore). Moscow: Progress,
98–137. http://www.marxists.org/archive/marx/works/1848/communist-manifesto/
Parish, Steven M. (1996). Hierarchy and Its Discontents: Culture and the Politics of Consciousness
in Caste Society. Philadelphia, University of Pennsylvania Press.
Parkin, David, and Stanley J. Ulijaszek (2007). Holistic Anthropology: Emergence and Convergence. Oxford: Berghahn.
Ranciere, Jacques (2006). Film Fables, trans. Emiliano Battista. Oxford: Berg.
Rio, Knut, and Olaf Smedal, eds. (2008). Hierarchy: Persistence and Transformation in Social
Formations. Oxford: Berg.
Press.
Said, Edward (1995). Orientalism: Western Concepts of the Orient [1979]. London: Penguin.
Schuon, F. (2007). Art From the Sacred to the Profane: East and West. Bloomington, IN: World
Wisdom.
Tambiah, S. J. (1976). World Conqueror, World Renouncer. Chicago: University of Chicago
Press.
Tcherkezoff, Serge (2009). Hierarchy is Not Inequality – in Polynesia for Instance. In Hierarchy:
Persistence and Transformation in Social Formations, ed. Knut Rio and Olaf Smedal.
Oxford: Berghahn, 299–330.
Zizek, Slavoj (2003). Organs Without Bodies: On Deleuze and Consequences. London: Routledge.
9781444333237_4_011.indd 208
8/10/2010 10:54:09 AM
12
From Wholes to Collectives
Steps to an Ontology of Social Forms
Philippe Descola
As is the case with most of our creeds and opinions, the ambivalent attitude toward the
question of holism that this chapter reflects is largely the product of the personal history
of its author, both as an anthropologist educated in a specific intellectual tradition and
as an ethnographer with fieldwork experience among Amazonian Indians. Although
confessions are misplaced in theoretical writings, a few initial words about the context
within which I developed some of the ideas that are put forward in this chapter may help
to throw a light on its main argument. In the 1970s, during my early training first as a
philosopher and then as an anthropologist – both domains, in France, are closely connected – I came to accept without much questioning the idea that society exists as a
whole, distinct from, and superior to, the sum of its parts, a transcendent entity that is
mainly responsible for most of our behavior through mechanisms that it is the duty of
the social sciences to elucidate. Together with the sociological vocabulary that sustained
it, this received wisdom had even percolated into the common sense of laypersons in
such a way that it was normal for people then, as it still is now, to blame “society” in
general (not their boss, their wife, or the ruling class) for their specific predicament, or
otherwise to claim that it was “society” that had made them behave the way they did. I
will come later to the historical roots of this conception, but suffice to say here that it
was reinforced at the time by our enthusiastic endorsement of a version of Marxism
wherein the form of the State, the nature of the relations of production, and the movement of history were the products not so much of the individual and collective actions
of real people, but rather the outcome of abstract forces and structural contradictions;
as for the desired outcome of the revolutionary process that we were fervently attempting to trigger, it took the appearance of this most holistic of wholes, a global classless
community of humans wherein each one would be dependent upon everyone else.
And then I went to Amazonia. There, I lived with a bunch of people who had no
chiefs, no villages, no descent groups, no history, no religion, and no ritual to speak
of; a bundle of individualities who were uncertain as to who they were collectively and
not particularly keen to ascertain it, who never acted willingly as a corporate group,
and who spent a great deal of their time trying to kill each other. Society, as I had
learned to recognize it, was conspicuously absent among the Achuar, as were absent
a few other conceptual props that I had brought in my anthropological tool kit: no
9781444333237_4_012.indd 209
8/7/2010 7:09:18 PM
210
Descola
nature and no culture, no economics and no politics, no ancestors and no memory.
I found instead what I had not been prepared to deal with as an ethnographer: a scatter of forceful individualists and strong personalities, proud in their bearing and
prompt to disagree, lovers of solitude and enemies of authority. I was baffled and not
a little disheartened at first, but soon found solace in the challenge of trying to make
sense of an aggregate of people who managed perfectly well to interact – and, on the
whole, to be happy – within the framework, so to speak, of a Hobbesian state of
nature. However, this utter lack of wholeness did not show in my initial writings,
largely because of the holistic effect of the ethnographic endeavor, a feature to which
I shall return. Also, it soon became clear that there was another, more embracing,
kind of wholeness that deserved attention, namely, the process of totalization by the
means of which Achuar individuals managed, in spite of their lack of integrative institutions and formal devices of cultural inculcation, to endow their actions with a markedly distinctive style, the effect of which was to present to the observer all the
components of a quite homogeneous ethos, and one which became after a while
rather predictable. The paradoxical combination that I encountered among the
Achuar, between what appeared as a remarkable conformism in behavior (but conformity to what?) and a remarkable paucity of formal structures susceptible to impose
conformity, thus led me to explore the processes through which people schematize
seemingly in unison their experience of the world without necessarily being aware of
how they operate. Before expounding on this topic, however, and in order to throw
some light on the shortcomings that we have to supersede, the first part of my chapter
will deal with some of the conditions and consequences of anthropological holism.
The Coherence of Experience
A much discussed peculiarity of the ethnographic method, a legacy from the time when
it was mainly practiced in societies devoid of written records, is that it draws almost
exclusively upon the direct information that can be gathered by a single person, most
often in face-to-face encounters. Since that person, barring exceptional circumstances,
cannot interact meaningfully with all the members of the group that she observes, she
will infer from the behavior and the statements of the few individuals with whom she is
best acquainted the norms and practices of the whole community. And since not everything in what she hears and watches seems relevant to her descriptive and interpretive
project, she will tend to select, in the massive quantity of observations that she accumulates daily, the events and utterances that appear to be most congruent with what she
progressively grasps as being the main fault lines of the group she studies. These “data,”
already sifted and purified, will form the core material of the description that she writes
upon coming home. Although ethnographic fieldwork has become the trademark of
anthropology (rather unduly, for anthropology has more to offer than erudite travel
accounts and covert autobiographies) and although it has yielded interesting results
time and again, it is a bizarre proceeding nevertheless, in that it aims at the most sweeping generalizations out of the most minute observations of a very reduced number of
persons. This is achieved thanks to two totalizing devices, seemingly contradictory but
quite complementary: expansion by inference and downsizing by reduction in scale.
Expansion is the process thanks to which scattered information, very often gathered
informally, acquires progressively the status of a template as it is reorganized in the
9781444333237_4_012.indd 210
8/7/2010 7:09:18 PM
211
mind of the ethnographer so as to provide her with a clearer picture of where she
stands in the community that she has chosen to study. In the same manner as one of
the first tasks of the ethnographer upon landing in a strange village or neighborhood
is to draw a rudimentary map and make a preliminary census, so as to identify who is
who and who lives where, the general information gathered by the ethnographer
becomes willy-nilly a vade mecum that allows her to find her way about in social space
and operate as a moderately competent actor in the group she elected. By necessity,
then, expansion draws on the unifying character of one’s own experience, as fluxes of
information and sensations are processed reflectively to provide a framework, not so
much for interpretation, at least initially, as for basic actions; a framework that progressively evolves from a series of scripts allowing an efficient handling of situations in a
foreign environment (“Who can I trust?” “Who can give me shelter?” “Where can I
find food?”) to the implicit notion that the small group of people with whom one
interacts day after day forms the fractal image of a coherent whole at a wider scale. The
only way to escape from this implicit subsumption would be to focus exclusively on
the mechanisms thanks to which situations are constructed and speech acts are produced (i.e., paying exclusive attention to the pragmatics of daily life and trying to
ignore the wider context). But this is almost impossible: when analyzing an interaction
between a group of persons, the observer cannot put into brackets the assumption
that some persons act the way they do because they have a vested authority entrusted
to them by the ancestors, the State, or the CEO, while others appear to behave submissively because they belong to a minor lineage or to an exploited minority. A reference to the whole (i.e., to the grand picture of institutions, contrasting roles, status
differences, and cosmological predicates) always forces its way back into the interpretation because, in our daily lives as in our job as social scientists, we tend to make sense
of unexpected situations by subsuming them under more general patterns.
Reduction in scale is the natural outcome of this process when it comes to the
writing-up stage: since the great whole cannot be described in its complexity, it is
reduced to a scale model wherein some features of the purported prototype will be
given prominence while others will be played down or will altogether disappear. What
is aimed at here is that the monograph should become a plausible microcosm of the
unattainable macrocosm, a vivid picture in which the relative proportions and structure of the whole are faithfully depicted so as to provide an opening into the experience of this whole that the ethnographer has undergone. In that respect, the much
heralded mutation in the style of ethnographical accounts over the past two or three
decades amounts in fact to very little: in spite of the advocacy for a fragmented, kaleidoscopic ethnography, the whole has not subsided but has moved from the foreground to the background, where it provides the vista upon which the ethnographer
opens the window of her soul. True, the whole is not depicted as it was in older, more
conventional, monographs; it is suggested by clues interspersed in the ongoing conversations with informants, much as Balzac and Proust used to do when they gave a
picture in filigree of the social stage on which their characters evolved. But the massive
system of norms, rules, and statuses is there nevertheless, albeit more discreetly, much
as the codes of the aristocratic society of the Faubourg Saint Germain are cleverly
suggested in all their intricacy in Le Côté de Guermantes, to provide the meaningful
context within which the main characters evolve.
A paradox of the ethnographic method is that it was invented by the British rather
than by the Germans (although Malinowski, by his initial intellectual upbringing, was
probably more German than English). For ethnography as participant observation
9781444333237_4_012.indd 211
8/7/2010 7:09:18 PM
212
Descola
gives more readily access to culture as defined in the German tradition than to society
in the Durkheimian sense that became the standard object of British social anthropology. That each nation, people, or tribe conforms a unique and cohesive assemblage of
material and intellectual features, solidified by tradition, typical of a certain mode of
life, rooted in the idiosyncratic categories of a language, and responsible for the specificity of the behavior of its members, is an idea which became prominent among
some German-speaking authors of the nineteenth century as the core concept leading
to national unity and ethnic emancipation. And this idea of culture is better attuned
to the kind of experience that an ethnographer grapples with when he is immersed in
a foreign system of values, than to the notion of society as an intricate and cohesive
mechanism the components of which can be described independently, a notion which
came to be seen as the result of the kind of fieldwork advocated by British functionalism. As Jonathan Friedman argues (Chapter 13, this volume), “social functionalism”
refers to the interlocking of parts in the definition of the whole, while cultural holism
refers to a semantic closure which renders difficult to distinguish the parts. In other
words: society is observed, culture is experienced. This is why the processes subsumed
under the latter will always appear as more coherent to the observer than the devices
making up the former; they will also lend themselves more easily to their perception
as a whole the parts of which are difficult to disentangle. It is because the living body
of society can be, and has been, decomposed fairly easily on the dissecting tables of
monographs (a material basis, a social organization, a system of collective representations) that it was necessary to insist, as a sort of afterthought, upon its organic cohesiveness and functional unity. No such thing can be done with culture, as evidenced by
the unsuccessful efforts made in the United States by the cultural materialists to dissociate an ecological and technical infrastructure from the rest of the cultural whole.
However, the experiential holism derived from the practice of ethnography does
not constitute the main substrate of holism as a paradigm of the social sciences, merely
a phenomenological confirmation. For ethnography is the experience of a cultural
whole that, by definition, is irreducible to any other. And anthropology, as the example of Radcliffe-Brown a contrario testified, cannot be based upon inductive generalizations from ethnographic particulars (Radcliffe-Brown 1958: ch. 1). True, this form
of “butterfly collecting” (to follow Leach’s cruel jest; see Leach 1961: 2) has regained
favor in the past decades, as the heady rejection of master narratives began to clash
with the selfish desire to endow one’s own ethnography with a nobler destiny than the
propagation of local gossip, resulting in a haphazard mosaic of minor narratives united
by superficial resemblances, a tedious combination that has kept the number of readers of books of anthropology steadily falling down. But anthropology, it must be
emphasized time and again, is not ethnography. The kind of anthropology that ignited
my vocation, inspired by the writings and teachings of giants such as Lévi-Strauss,
Merleau-Ponty, Foucault, Bateson, and Dumézil, is not at all descriptive, it is not
interpretive, and it is not even holistic properly speaking. It purports to render less
obscure the human condition, that is, the manifold relations that Homo sapiens entertains with the world. Such an objective cannot be achieved by merely describing the
infinitely varied ways in which humans give meaning to their lives in the hope that this
addition of specifics will result in fruitful generalizations; it can only be attained by
making learned guesses about the properties of specific human institutions and practices, and checking how these hypotheses fit with the human record by experimenting
with a few well-chosen case studies that appear to contrast with one another in a systematic way. As Lévi-Strauss aptly put it, “[I]t is not comparison which founds
9781444333237_4_012.indd 212
8/7/2010 7:09:18 PM
213
generalization, but the reverse” (1958: 28). If there is indeed, as George Marcus
argues (Chapter 3, this volume), an “aesthetic of ethnographic holism,” a disposition
to which I yielded myself once as a literary experiment (Descola 1996), it cannot be
an aesthetic of the anthropological project as such. For, however entertaining they
may be, collage, induction, and free association provide a very impoverished and
unconvincing form of “trans-ethnographic” holism. The idea of a whole with a distinct identity from that of its components would never have taken the hold it has over
the social sciences if it were merely an outcome of the experiential dimension of ethnography; the latter only confirms and reinforces the deceptive obviousness of the
former. So we have to look elsewhere for the persistence of this notion.
A Nostalgia for Ordered Wholes
The holist paradigm is mainly a product of the institution of modern sociology by
Comte and Durkheim, itself a distant outcome of the Kantian notion of the state as a
persona moralis, that is, a nonhuman entity endowed with human attributes – in particular, the ability to freely exercise rights and duties – which distinguished this body
politic from a mere thing (Kant 1795: VIII, 344). For Durkheim, society is to be seen
as a sort of person, but “qualitatively different from the individual personalities which
compose it” (Durkheim 1996 [1924]: 53), the relationship between parts and whole
being somewhat metaphysical: each individual hosts a fragment of society internalized
through education, and it is this very fragment which makes him or her not so much
an individual (since individuation, for Durkheim, results from the possession of a
specific material body) but a proper person, that is, a refraction of the “spiritual principle” that serves as the soul of the collectivité (Durkheim 1960 [1912]: 380).
Paradoxically for what came to be regarded as the moral foundation of a secular
republic, this conception of the relation between individual and society appears to
draw some of its inspiration from early Christian representations of the sociocosmic
polity (Paulinian and Augustinian, in particular) and more generally from the holistic
conceptualization of collectives that is typical of what I have called elsewhere “analogist ontologies” (Descola 2005), a theme to which I shall return later. Durkheim’s
notion of society can thus be seen as a nostalgic echo of a period when the acid of
individualism had not yet dissolved the cohesion and solidarity of the social whole, a
period when the effervescence of collective rituals fueled the emotional adhesion of all
fellow citizens to something greater than themselves.1 But this notion is also a discreet
advocacy for a more enlightened future when, in the midst of a new eruption of “creative effervescence,” new ideals will surge to provide humanity afresh with a guidance
(Durkheim 1960 [1912]: 548). This messianic prediction, very much in line with the
socialist inclinations of the Durkheimians, builds a bridge between the mechanical
solidarity of the hierarchical wholes of the past and the true cohesion and integration
of the egalitarian wholes to come, a bridge that crosses over the selfishness of possessive individualism, and its translation into the neoclassical economics of Menger,
Walras, and Pareto.2 This is why Tarde was so vilified – and misunderstood – by the
Durkheimians: although he shared with Durkheim a superficially Leibnizian conception of the relation between parts and whole, his monadology posits individuals and
their ontological differences at the core of the system, not the individual as a refraction of the whole (see, e.g., Tarde 1999 [1893]).
9781444333237_4_012.indd 213
8/7/2010 7:09:19 PM
214
Descola
While he had the huge merit of granting the social sciences a domain, a method,
and a legitimacy, Durkheim has also rendered hopelessly aporetic the sociological
treatment of the relation between parts and whole, impeding as a consequence any
original approach of the individual within the confines of his sociology. For
Durkheimian sociology is based on the society as a “collective person,” from which
are deduced the individual persons who are its diffracted emanations, thus stripping
them of their agency and capacity for change. Society is given prior to any human
presence and individual action, compelling the sociologist to invent a new regime of
existence for this transcendent entity, and a new system of forces to explain how it
acts upon the individuals and how the individuals manage to sustain it. There is an
element of magic in this process, and it comes as no surprise that Durkheim recurs to
the magical notion of mana to designate this suprapersonal agent that constitutes
society and to which all individuals contribute (Durkheim 1960 [1912]: 378–82).
Mauss and Lévi-Strauss have retaken this mysterious notion under a different guise,
not as a substantial binding force but as a “symbolic function,” that is, as a set of differential relations which constitute, notably through language, the social order of
culture. As society is, genetically, a prerequisite of social action for Durkheim, symbolism is for Lévi-Strauss a given that, so to speak, predates its actualization by
humans as it provides the framework for their actions.3 Both postulates seem highly
unlikely, especially in the light of what we know now about the very gradual nature
of the hominization process.
Although Lévi-Strauss has retained from Durkheim the idea that the expression of
culture requires a template that, at least phylogenetically, appears to exist independently from, and previous to, its actual implementation by humans, he has also made a
gigantic step toward the elimination of empirical wholes by focusing on the study of
structures rather than bounded social groups. As Jean Pouillon neatly sums it, “[S]
tructuralism, properly speaking, begins when one admits that different sets can be
brought together, not in spite, but because of their differences, which are then susceptible to be patterned” (1975: 15–16). This general principle that systematic differences
matter more than haphazard resemblances is what is most distinctive in structural
anthropology, with the result that the object of study is not anymore a “thing out
there,” but a constructed combinatory that must account for all the states of a set of
phenomena based on the chartering of the ordered differences between its elements.
In structural anthropology, wholes are never given or even presupposed. They result
from the operation by the virtue of which provisionally isolated entities – be they social
groups, cultural features, fragments of myth, ritual sequences, norms, social positions,
techniques, or images – are constituted as variants of an analytically constructed totality, which, of course, has no proper empirical existence, no more, say, than a cladogram. Of course, it could be argued that converting into variants within a group of
transformation certain phenomena that have been rendered discrete for the sake of
analysis constitutes the basis of a distinctively structuralist brand of holism (see, e.g.,
Jonathan Friedman, Chapter 13, this volume). But it should be pointed out also that
such a method in fact dissolves the very idea of a whole as a fixed combination of parts,
since there are as many totalities as there are ways of defining the variants and their
9781444333237_4_012.indd 214
8/7/2010 7:09:19 PM
215
salient properties (i.e., the types of contrast between them that are meaningful). For
Lévi-Strauss a myth hardly exists in itself, within a coherent whole that it would reflect
and of which it would give a partial account; rather, it exists at the point where it meets
other myths of other cultures and thus acquires its meaning by opposition. As Marshall
Sahlins shows convincingly (in Chapter 7, this volume), this intercultural schismogenesis that is played out on a number of boundaries applies to much more than mythical
structure: to political systems, to definitions of identity and status, to the values ascribed
to different types of wealth. There is nothing holistic in such an attitude, except if one
takes the word as meaning a process of disentanglement from locality, of treating phenomena at the level of contrast where they can be most productive, and of abstracting
properties and regularities from the flow of narratives and practices. But rather than
label that “holism,” it is better to call it by its proper name: science.
One of the main objections that have been made to structural anthropology is that
if the model is not a mirror image of a discrete empirical phenomenon (if such a thing
exists), then there is no way to ascertain a possible congruence between the model
that the anthropologist constructs and the processes and patterns by the means of
which peoples tacitly organize their experience and knowledge of the world. In other
words we are incapable of tracing how structures could migrate from the plane of the
intelligible to the plane of the conditions of their empirical existence. Lévi-Strauss
himself devoted little attention to this practical problem, except by surmising that the
integration of the intelligible with the empirical must result from the operation of
“conceptual schemes,” here understood in a Piagetian, and ultimately Kantian, sense
(Lévi-Strauss 1962a: 173).
Now this notion of schema has been rejuvenated in the past 20 years thanks to a
series of work in cognitive psychology that has given it a sounder experimental basis
and has made of it a promising candidate for a nonpropositional structure organizing
habitus, guiding inferences, and filtering perception.4 Schemata are not internalized
through formal inculcation; they do not exist in the heaven of ideas, ready to be
captured by consciousness; they are progressively constructed within communities of
practice by individuals who share comparable experiences. In that respect, schemata
help to render less obscure the relation between a structural model and a cultural
pattern: one may think that what the former brings to light is the general pattern
according to which members of a collective schematize their experience, a pattern
which thus becomes the organizing principle of the system of explicit codification to
which they adhere. The adequacy between the model and the underlying characteristics that it purportedly reveals would thus accrue from the fact that these characteristics, rather than revealing universal properties of the mind (except at a very abstract
level), express the very frames and processes of tacit objectivation thanks to which the
actors themselves organize their relations to the world. Midway between praxis and
structure, a schema is an interface that is at the same time concrete (it is embodied in
individuals and implemented in practices), particularized (it actualizes specific
affordances), and quite abstract (it can only be detected through its effects, without
being for that the emanation of mysterious entities such as the “symbolic function”
or the “collective unconscious”). Two such schemata appear to play a prominent role
in the process of totalization by the means of which we structure our relations with
the world. One is predication, the detection of qualities among existents which generates ontological identifications; the other is binding, the unfolding of a dominant
pattern of interaction with humans and nonhumans alike. I will examine each of
them in turn.
9781444333237_4_012.indd 215
8/7/2010 7:09:19 PM
216
Descola
It is implausible to assume that affordances alone are enough to stabilize our judgments about the ontological attributes of objects in our environment: although context and pragmatics are instrumental in passing specific judgments of identity (whether
what we perceive is an effigy, a living animal, or a spirit), they do not provide as such
the patterns that organize the ontological categories to which we refer what we perceive. These are the products of cognitive dispositions which allow us to predicate
identities by lumping together, or dissociating, elements of the lived world that appear
to have similar or dissimilar qualities. I have argued elsewhere that one of the universal
features of the human mind upon which such dispositions rest is the awareness of a
duality between material processes (hereby called “physicality”) and mental states
(hereby called “interiority”) (Descola 2005: secs. II–III; for an English summary, see
Descola 2006). By using this universal grid, humans are in a position to emphasize or
minimize continuity and difference between humans and nonhumans. Thus, on the
physicality axis, one view may maintain that all physical bodies are essentially ruled by
identical “natural” principles, while the opposing one stresses differences and postulates that what marks out different kinds of entities is, precisely, the bodies they
inhabit. Similarly, on the interiority axis, the emphasis may be on continuity (the souls
of all beings are the same) or on discontinuities (humans form a race apart because of
their soul or mind). These dispositions thus generate four major ontologies, that is,
systems of qualities ascribed to things in the world, which I have respectively labeled
“animism,” “totemism,” “naturalism,” and “analogism.”
Animism is an ontology wherein humans and nonhumans reputedly have the same
kind of interiority but are differentiated by the bodies they inhabit. It is most common among native populations of Amazonia, northern North America, northern
Siberia, and some parts of Southeast Asia and Melanesia, who maintain that animals,
plants, and even inanimate objects have a human-like intentionality, lodged within a
mobile bodily clothing which nevertheless determines, because of its anatomical features, the type of world they have access to and how they see it. Totemism is taken here
not in the sense rendered common by Lévi-Strauss of a universal classificatory device
using natural discontinuities to signify social segmentation (Lévi-Strauss 1962b),
but rather as an ontology that stresses the continuity between humans and nonhumans both on the physicality axis (common substances) and on the interiority one
(common essences). It is best exemplified by Australian Aboriginal cultures where
specific plant and animal species are believed to share with particular sets of humans
an identical complex of essential qualities, but one that is absolutely different from
other similar groupings. The reference for the totemic class is not a specific natural
object to which one identifies (Frazer), nor a relation between natural objects used as
a template for a relation between human groups (Lévi-Strauss), but a bundle of precisely defined moral and physical qualities usually subsumed under the name of an
overarching property that serves as a taxon for naming the totem. Naturalism is the
mirror opposite of animism and characterizes the modern world and Western thought.
It insists on the differences between humans and nonhumans on the interiority axis:
humans alone are supposed to have a meaningful selfhood whether individual (mind,
capacity for symbolism) or collective (Volksgeist, cultures). By contrast, clearly since
Descartes, and more neatly even since Darwin, humans and nonhumans are linked by
their shared physicality: they belong to a continuum where the same laws of physics,
biology, and chemistry apply. Finally, analogism assumes discontinuities on both
axes, postulating microdifferences among the components of the world at an infraindividual level, but setting up various kinds of correspondences (hence “analogism”)
9781444333237_4_012.indd 216
8/7/2010 7:09:19 PM
217
between these heterogeneous elements so as to weave them in a seemingly seamless
continuum. Analogism was the dominant ontology in Europe from Antiquity to the
Renaissance, and it is still extremely common elsewhere: in China and India, in
Western Africa, or among native cultures of Mexico and the Andes.
This fourfold schema should not be taken as a taxonomy of tightly isolated “worldviews,” but rather as an unfolding of the phenomenological consequences of four
different systems of inferences about the identities of things in the world. According
to circumstances, each human is capable of making any of the four inferences, but will
most likely pass a judgment of identity according to the ontological context (i.e., the
systematization for a group of humans of one of the inferences only) where he or she
was socialized. Also, historical processes of internal transformation and diffusion will
result very often in a combination between a dominant ontology and a recessive one,
so that this typology should be taken as a heuristic device rather than as a method for
classifying societies. A useful device, however, as it brings to light the reasons for
some of the structural regularities observable in the ways the phenomenological
world is instituted (cultural “styles”) and for the compatibilities and incompatibilities
between such regularities, both of these endeavors being basic anthropological tasks
that have been too quickly discarded and thus left open to crude naturalistic
approaches.
We can now move to “binding.” The four ontological archipelagos become internally differentiated by various elementary patterns of relation that shape interactions
between humans and nonhumans alike. These relations can be divided into two main
groups: on the one hand, exchange, predation, and gift (wherein a value moves
between potentially reversible terms that have an equivalent ontological status), and
on the other hand, production, protection, and transmission (wherein the relation is
oriented between hierarchical terms). Although this set of relations can be said to
form the basic toolkit of the social sciences, they have to be qualified. For instance, by
contrast with Mauss’ position, I posit that gift must be differentiated from exchange,
as the latter operation always calls for a double movement of give and take, while the
first one excludes by principle a counterpart (otherwise, it would not be a gift).
Similarly, production is in no way the universal process it passes for in Marxist and
constructivist approaches: the idea of an intentional agent imposing a form on matter
according to a mental blueprint is a model of action that is unknown in most nonEuropean cultures.
These patterns of relations are partly based on cognitive processes such as schematic induction or the analogical transposition between domains, but are not inbuilt categorical imperatives; rather, they should be treated as objectified properties
of collective life which come to be embodied in physical and mental dispositions, and
are thus stabilized as habitus. Giving (including oneself) to others, taking or receiving from them, exchanging with them, but also appropriating others, protecting
them, producing them, or placing oneself in their dependence form the basic set of
interpersonal actions that humans have inherited from their phylogeny; it should
come as no surprise that they provide a repertory from which each collective will
draw a favorite mode of relating to others. However, a dominant ontology may be
affected by various patterns of relations, thus causing variations of ethos within an
apparently homogeneous cosmological framework. This is the case with animism in
native Amazonia, for instance: the three main styles of interactions between humans
and nonhumans alike which can be found there are exchange, predation, and sharing, each human collective favoring one option to the detriment of others. Within a
9781444333237_4_012.indd 217
8/7/2010 7:09:19 PM
218
Descola
cosmological pattern which renders native Amazonia a distinctive cultural area,
various sets of relations operate in the shaping of differentiated ethos, and thus
account for internal diversity (in particular, the contrast between bellicose and more
pacific tribes).
We are now in a position to get back to wholes. Although I share with advocates of
actor–network theory (e.g., Latour 2005) the premise that the notion of “society” is
not an explicantia but an explicandum, I differ from them in that I am deeply convinced that there are structural constraints to the ways humans and nonhumans
assemble. For each ontology in a way prefigures a specific type of collective more
appropriate to the gathering within a common destiny of the kinds of being that this
ontology distinguishes. By “collective,” a concept I borrow from Latour (1991),
I mean a way of assembling humans and nonhumans in a network of specific relations,
by contrast with the traditional notion of society which is the result of a specific process of filtering, proper to late naturalism, thanks to which sets of humans have been
arbitrarily detached from the web of relations they maintain with nonhumans. Far
from being a founding prerequisite from which everything else is derived, sociality
proceeds rather from the process of collecting and assembling into a common set that
each mode of identification predetermines. Thus, the property of being social is not
what explains, but what must be explained. And the totalization of experience to
which every actor constantly proceeds, and that the analyst tries to decipher, amounts
to recapitulating a pattern of distribution of beings into collectives: who or what is
assembled with whom or what, in what way, and for what purpose? But each of the
“wholes” which result from this patterning process is in every respect different from
the others, and none of them is exactly equivalent to the “whole” of Durkheimian
sociology. This is why we must now turn to a description of the characteristics of
these collectives.
A Typology of Collectives
I shall begin with animism. In such systems, all the classes of beings endowed with an
interiority similar to that of humans live in collectives that possess the same kind of
structure and properties: they all have chiefs, shamans, rituals, dwellings, techniques,
artifacts; they assemble and quarrel, provide for their subsistence, and marry according to rules. But these collectives, that are all integrally social and cultural, are also
distinguished from one another by the fact that their members have different morphologies and behavior. Each collective is equivalent to a sort of tribe-species that
establishes with other tribes-species relations of sociability of the same type as those
that are held legitimate within the given human collective that ascribes its internal
organization, its system of values, and its mode of life to the collectives of nonhumans
with which it interacts. The so-called natural and supernatural domains are thus peopled by nonhuman collectives with which human collectives maintain relations according to norms that are deemed common to all. For although humans and nonhumans
may exchange perspectives (Viveiros de Castro 1996), they also and above all exchange
signs, that is, indications that they understand each other in their practical interactions. And these signs can only be interpreted by all parties concerned if they are
predicated on common institutions that legitimate them and give them a meaning,
thus warranting that the misunderstandings in interspecific communication will be
9781444333237_4_012.indd 218
8/7/2010 7:09:19 PM
219
minimized. This is why all the isomorphous collectives of humans and nonhumans
take as their model a specific human collective.
The concept of species which provides a template for animic collectives can hardly
be equated with a sociological whole in the Durkheimian sense. It is a collection of
exemplars that bear a family resemblance (i.e., a living kind in an inductive sense). The
prototypical homogeneity of the members of the species rests on broad morphological and ethological similarities (toucans have distinctive beaks and plumage, have a
specific system of communication, and live in couples; Achuar have distinctive facial
paintings and feather ornaments, have a specific language, and marry their cross cousins), not on any overarching properties intrinsic to the class that would reverberate or
trickle down into its components. There is no superiority of the class over its parts,
especially as the specificity of the species is based here on the fact that its members
experience themselves as such because the very existence of the other species with
whom they interact, be they human or nonhuman, allows them to ascertain that they
are different from them, since the members of these other species hold of them a different point of view from the one they hold themselves. In other words, the perspective of the putative classifier must be absorbed by the classified in order for the latter
to see himself as entirely specific, an absorption that is often realized quite literally in
the animic world, through cannibalism, for instance.
I will not dwell on the sociological formula of naturalism, since it is the one that is
most familiar to us and that we deem, mistakenly, to be universal: humans are distributed within collectives differentiated by their languages, beliefs, and institutions – what
we call “cultures” – which exclude everything that exists independently from them,
namely, nature and artifacts. The paradigm of collectives is here human society, by contrast to an anomic nature. Humans associate freely, they elaborate rules and conventions that they can choose to infringe, they transform their environment and share
tasks in the procurement of their subsistence, they create signs and values that they
exchange – in sum, they do everything that nonhuman animals do not do. And it is
against the background of this fundamental difference that the distinctive properties
attributed to human collectives stand out; as Hobbes says with his robust concision,
“To make Covenant with bruit Beasts, is impossible” (Hobbes 1914 [1651]: ch. 14,
p. 71). The idea of a social contract, so central from Grotius to Rousseau as a philosophical artifact to legitimate sovereignty, points to the very limited scope of the whole
that is created by the sum of its components when they decided to forfeit a portion of
their free will to create something greater, and above all more powerful, than themselves. In early modern political philosophy, collectives are seen properly speaking as
collections, mere gatherings of human individuals who pull together their moral assets
so as to foster their personal interests. Even Rousseau, familiar though he was with the
rustic democracy of his place of birth, did not envision the gathering of fellow-citizens
as a society in the conventional sense of Durkheimian sociology, that is, as an autonomous entity animated by a specifically collective interest which would be more than,
and qualitatively different from, the sum of particular wills. Durkheim himself stated
the difference quite clearly when he compared his own view of collective utility, determined as a function of the social being considered in its organic unity, with common
interest conceived by Rousseau as “the interest of the mean individual” (l’intérêt de
l’individu moyen) finding its way into the general will through the addition of what is
useful to everyone (Durkheim 1918: 138–9). It is a paradox of naturalism that a holist
conception of society emerged when its collectives were no longer organic wholes but
conflictive aggregates of supposedly isonomic individuals, while the contractualist
9781444333237_4_012.indd 219
8/7/2010 7:09:19 PM
220
Descola
political philosophers who laid the theoretical foundations of possessive individualism
and of an atomistic conception of polities did so at a time when holistic and hierarchically organized collectives, typical of the analogist regime, were in full bloom.
The question of totemic collectives is more complex as its history is fraught with
controversies. Traditionally, totemism has been conceptualized as a form of social
organization in which humans are distributed in interlocking groups that borrow their
characteristics from natural kinds, either because these groups are said to share certain
attributes with a set of nonhumans, or because they take as models for patterning their
internal differences the contrasts between eponymous species. Now, this broadly sociocentric definition has the disadvantage of introducing an analytical dichotomy
between social categories and natural categories that appears to be absent from the
ontological premises of those paradigmatic totemists, the Australian Aborigines.
There, specific sets of humans and nonhumans grouped within named classes share a
core of neatly defined moral and physical attributes owing to the fact that they are
issued from the same ontological prototypes – the “Beings of the Dreamtime.” These
attributes that crosscut species boundaries are not derived from what is improperly
called an “eponym entity,” since the word designating the totem in many cases is not
the name of a species (i.e., a biological taxon) but rather the name of an abstract quality which is present in this species as well as in all the beings subsumed under it in a
totemic grouping (von Brandenstein 1982: 54). For instance, the Nungar of southwest Australia had two totemic moieties, respectively called maarnetj, that can be
translated as “the catcher,” and waardar, which means “the watcher,” these two terms
also being used to designate the totems of these moieties, the White Cockatoo and the
Crow (von Brandenstein 1977). Here, the names of the totemic classes are terms that
denote qualities that are also used to designate the totemic species, and not the reverse,
that is, names of zoological taxa from which would be inferred the typical attributes of
the totemic classes. These contrasted qualities in turn function as labels for bundles of
more precisely defined attributes. To return to the Nungar, humans and nonhumans
who belong to the moiety of “the catcher” are lightly colored, rotund, and impulsive,
while those of the moiety of “the watcher” are dark, short, massive, and vindictive.
Such qualities cannot be directly elicited from the observation of the White Cockatoo
or the Crow; they express a repertoire of more abstract, contrasted predicates for
which these two emblematic species are the best exemplars.
Rather than defining totemism by the transposition to the social domain of natural
differences or natural properties, then, it is more appropriate to view it as a system
wherein sets of humans and nonhumans are jointly distributed in collectives (the
totemic classes) which are isomorphous and complementary, by contrast with animism wherein humans and nonhumans are distributed separately in collectives (tribesspecies) which are also isomorphous, but which remain autonomous in relation to
each other. To return to the example of the Nungar, in the moiety of “the getter”
iconically represented by the White Cockatoo, one does find cockatoos, as well as the
human half of the Nungar tribe, but one finds also eagles, pelicans, snakes, mosquitoes, and whales, in short an apparently ill-assorted aggregate of species that cannot
be matched with any of the groupings of organisms that the environment spontaneously offers to observation. Such a mode of aggregation offers a marked contrast with
the norms of composition of animic collectives. For although these differ from
one another because of the monospecific recruitment of their members, they are
nevertheless homogeneous as regards their principles of organization: for the Makuna
of Colombia, the tapir tribe has the same type of leader, shaman, and ritual system as
has the peccary tribe, the toucan tribe, and, of course, the Makuna tribe (Århem
9781444333237_4_012.indd 220
8/7/2010 7:09:19 PM
221
1996). This is not so with totemic collectives, which are also all different as regards
the composition of their members, but which are furthermore hybrid in their contents
and heterogeneous in their principles of assemblage.
This is particularly the case in Australia, where there exists a great variety of totemic
groupings (recruitment being defined according to gender, generation, site of conception, descent group, etc.) and where humans can belong simultaneously to several of
them. Here again, the contrast is notable with animic collectives which are on the
contrary predicated on a species-specific physicality, since the affiliation to each “society” is based on the fact that all its members share the same physical appearance, the
same habitat, the same diet, and the same mode of reproduction. It is in animism, not
in totemism, that the biological species provides a model for the composition of collectives. And this is so because animic collectives, like biological species, are never
integrated into a functional whole at a higher level: above the Achuar tribe-species, the
toucan tribe-species, and the peccary tribe-species, there is nothing in common, except
this abstract predicate that anthropologists who try to make sense of these arrangements call “culture.” No such thing with totemism, where the integrative whole
formed by the juxtaposition of the different totemic classes cannot be represented on
the basis of the groupings that the natural world proposes: the only available model
would be the species, since the genus is a taxonomic fiction, but the species is precisely
not liable to be decomposed in contrastive segments that would be analogous to
totemic classes. While animism and naturalism take human society as the paradigm of
collectives, totemism thus mixes in hybrid sets, humans and nonhumans that use one
another in order to produce social linkage, generic identity, and attachment to places.
But it does so by fragmenting the constitutive units so that the properties of each of
them become complementary and their assemblage dependent upon the differentials
that they present.
Finally, in collectives that function under an analogist regime, humans and nonhumans always appear as constitutive elements of a wider set, coextensive with the
universe: cosmos and society become indistinguishable, whatever the types of internal
segmentation that such a totality requires in order to remain efficient. For the analogist collective is always divided into interdependent constitutive units which are structured according to a logic of segmentary nesting: lineages, moieties, castes, and
descent groups prevail here and expand the connections of humans with other beings
from the underworld to the heavens. Although the exterior of the collective is not
entirely ignored, it remains an “out-world” where disorder reigns, a periphery that
may be feared, despised, or predestined to join the central core as a new segment that
will fit in the slot that has been allocated to it long before. All analogist collectives,
even those that ignore political stratification and disparities of wealth, have in common that their parts are ordered hierarchically, if only at a symbolic level. The hierarchical distribution is often redoubled within each segment, delimitating subsets which
are in the same unequal relation one to the other as the encompassing units. The classical illustration is the Hindu caste system as described by Louis Dumont, where the
general schema of encompassment is repeated within each of the successive levels of
subordination: in the subcastes composing the castes, in the clans composing the
subcastes, and in the lineal groups composing the clans (Dumont 1966). This is
because hierarchies that encompass too many elements become difficult to manipulate
practically, so that devices are required to structure their lineal gradation and to
attenuate the range of the discontinuities which fragment them.
It is even highly probable that what Dumont defines as holism, that is, a system of
value which subordinates the position of each existent in a hierarchy, and the cohesion
9781444333237_4_012.indd 221
8/7/2010 7:09:19 PM
222
Descola
of the latter, to a whole that is transcendent to its parts, is not so much, as he contends,
a general criterion which distinguishes nonmodern societies from modern individualism (Dumont 1977), but rather a means employed by the sole analogist ontologies in
order to render manageable an enormous accumulation of singularities. In that respect,
as Jonathan Friedman argues (Chapter 13, this volume), Dumont’s notion of holism
is indeed emic in that it uses a model, or perhaps an intuition, provided by a specific
culture in order to account for a more general feature of certain social formations.
However, and by contrast with Mauss explaining the power of the gift with the vernacular notion of the hau, the fact that the idea of holism retains a distinctive Indian
flavor has not prevented Dumont from successfully transforming it into a methodological procedure defining a much admired and much contested kind of anthropological orientation (this is the argument put forward by Bruce Kapferer in Chapter 11 of
this volume). India was a very efficient springboard for Dumont in his ambition to
decolonize European concepts and modes of analysis – more effectively, one should
add, than most of what has been done recently under the label of “postcolonial studies” – in that it provided the inspiration for what is certainly his major achievement, to
wit, the discovery that many oppositions are implicitly hierarchical, and that the terms
which compose them are in a different relation to the whole which encompasses them –
Adam is opposed to Eve both as the prototype of the male members of the human
species and as the very prototype of this species, thus encompassing Eve in her position as the female prototype.5 But this concept of hierarchical opposition which is
universally used as an intellectual tool to organize internal differences in analogist collectives, or so I surmise, is quite uncommon in the structuring of other types of collectives, be they totemic (components of totemic classes are ontologically equivalent
and the classes themselves are not hierarchical), animic (tribes-species are isometric),
or, obviously, naturalist. Dumont’s holism is thus a rare case of a concept originally
identified within a specific civilization, although based on a probably universal feature
of the human mind (hierarchical opposition as a classificatory device), but one that is
not universally implemented in the structuring and conceptualization of collectives.
By contrast with the egalitarian and monospecific collectives of animism and with
the egalitarian and heterogeneous collectives of totemism – collectives that are bound
in both cases to enter into relation with others – the analogist collective is unique,
divided into hierarchized segments and in almost exclusive relation with itself. It is
thus self-sufficient, for it contains within itself all the relations and determinations
that are necessary to its existence and adequate functioning; in that respect, it differs
from the totemic collective, which is indeed autonomous at the level of its ontological identity, but which requires other collectives of the same kind in order to become
functional. For, in an analogist collective, the hierarchy of the parts is contrastive: it
is defined exclusively by reciprocal positions. And this is why the segments do not
constitute independent collectives as the totemic classes do, since the latter draw
from within themselves, from specific sites and prototypical precursors, the physical
and moral foundations of their distinctiveness. The segments of an analogist collective are thus thoroughly heteronomical in that they only acquire a meaning and a
function by reference to the autonomous whole that they jointly form. One can say
that animic collectives also admit a degree of heteronomy, but of an entirely different
kind since the external specification obtains through a series of identifications to
individual and intersubjective alterities of various origins, not through an overdetermination of the elements by the structure which binds them. The enemy whose
alterity I absorb by capturing his head or consuming his body proceeds indeed from
9781444333237_4_012.indd 222
8/7/2010 7:09:19 PM
223
a different collective; however, his capacity to singularize me is not linked to traits
that would be specific to his own collective, but simply to his position of exteriority
regarding myself. This is why, in the case of animism, there is no predetermination as
to the type of collective liable to serve this function of external specification: it may
be, according to context, individuals proceeding from one or various tribe-species of
animals, from one or various tribe-species of spirits, from one or various tribe-species
of humans, or from a combination of all of these. In an analogist collective, by contrast, the members of segment A differentiate themselves collectively from members
of segment B in that A and B are elements of the hierarchical structure which encompasses them all; in philosophical language, one would say that their positions and
relations are the product of an expressive causality. The dependence of the analogist
segments on the collective that defines them is constitutive of their being; with
elements that are intrinsically internal to the collective, they must strive to produce
an illusion of exteriority.
Conclusion
Aside from its intellectual genealogy and epistemological function in the social
sciences, the idea of whole is also a prism by the means of which analogist collectives
reflectively apprehend themselves, be it in medieval Europe or in China, in Polynesian
kingdoms or in peasant villages of southern Mexico. Such types of collectives have
even been branded “totalitarian,” with particular references to the Inca Empire and
Western Africa.6 This qualification is probably a way to express the extraordinary
imbrications of elements in these holistic but much compartmentalized polities,
where the possibility of nonconventional action is quite restricted, and where the
control of conformity exercised by the whole on its parts appears to us unbearable.
The term “totalitarian” also conveys the idea that nothing is left to chance in the
distribution of existents between the various strata and sections of the cosmos, that
every being is allocated a situation that corresponds to the role it is expected to play.
This is why the ideology of an analogist collective is necessarily functionalism (i.e.,
the idea that each of its constitutive elements contributes in its specific position to
the perpetuation of a stable totality). Without making too much of this, one may
even wonder if the vogue of functionalism in British anthropology may not have
been partly fostered by the fact that British anthropologists devoted a great part of
their time and energy to studying analogist collectives, especially in West Africa and
Southeast Asia, with functionalism as a local ideology thus meeting the expectations
of functionalism as a method of description and analysis.7 Be that as it may, and if my
previous propositions are to be taken seriously, then we have to admit that ontological predicates determine sociological arrangements in the sense that the specific
combinations of humans and nonhumans that ethnologists and historians describe
are not best understood by reducing them to projections of human institutions, but
by tracing the principles that allow, in each combination, a form of compatibility and
interaction between particular kinds of existents that would appear impossible, or
ludicrous, in other cases. So, rather than a comparative sociology of wholes, a project
which has now exhausted its credibility, what we urgently need is a comparative
ontology of parts, that is, an analysis of how the various components of collectives,
actual and potential, are jointly and, necessarily, instituted.
9781444333237_4_012.indd 223
8/7/2010 7:09:19 PM
224
Descola
Such an endeavor might appear as the sort of experiment that the editors of the
present volume have asked us to engage in, at least if “holism” is meant here as an
attempt at understanding human life which would be defined by contrast with the
kaleidoscopic, inductive, narcissist, and rather parochial kind of ethnographic depiction that has passed for anthropology in some quarters during the past two decades.
There is perhaps an even better term with which to qualify the type of holism corresponding to the structural ontology I am trying to develop, a term now so derided
that I almost hesitate to write it down: universalism. But the kind of universalism
I have in mind is quite distinct from the common version, the one which Bruno
Latour has labeled “particular universalism” (Latour 1991: 142) and which maintains that we, Moderns, are the only ones to have reached a true understanding of the
natural world of which the nonmodern only have “representations,” thus opening a
huge playing ground for charitable translations by anthropologists. Rather than being
“particular,” my own brand of universalism would be “relative” – not out of an
immoderate fondness for oxymoron, but because I take relative as that which pertains to a relation. Relative universalism is not grounded in nature and cultures, in
substance and mind, in the distinction between primary and secondary qualities; it
stems from the multiple relations of continuity and discontinuity, of identity and difference, of resemblance and dissimilarity that humans everywhere discern in their
environment thanks to the tools they have inherited from their phylogeny: a body, an
intentionality, a capacity to perceive distinctive contrasts, and the ability to establish
with a human or nonhuman alter relations of attachment or antagonism, of domination or dependence, of exchange or appropriation, of subjectivation or objectivation.
Relative universalism does not require a world distributed between a uniform materiality and contingent significations; its only precondition is the acknowledgment of
the salience of discontinuity, in things perceived as in the ways they are perceived,
and the willingness to admit that there exist a limited number of formulae to deal
with this saliency, either by ratifying it or by inhibiting its perception. A modest universalism, then, but one which would help to put anthropology back in business on
a grander scale.
Notes
1 There is an influence here of the interpretation of Roman religion by Fustel de Coulanges
as mostly secular, that is, composed of rites and representations that do not require faith or
beliefs; see Héran (1987).
2 See the general assessment of the theory of utility by Schumpeter (1954: 1033–73).
3 See Lévi-Strauss, “Le langage n’a pu naître que tout d’un coup,” in Lévi-Strauss
(1950: xlvii).
4 For psychology proper, see Mandler (1984) and Schank and Abelson (1977); for applications in anthropology, see D’Andrade (1995), Shore (1996), and Strauss and Quinn
(1997).
5 Houseman (1984) remains the best discussion of this complex issue.
6 The word is used by F. Héritier-Augé in her foreword to a book by Zuidema (1986), and
by M. Duval (1986) in his monograph on the Gurunsi of Burkina Faso.
7 By contrast, functionalism was ill equipped to make sense of animic collectives, and this in
turn may explain in part why British anthropology almost completely ignored until the end
of the 1960s some regions of the world where animism prevails, such as the South American
Lowlands or Subarctic Native America.
9781444333237_4_012.indd 224
8/7/2010 7:09:19 PM
225
References
Århem, Kaj (1996). The Cosmic Food Web: Human-Nature Relatedness in the Northwest
Amazon. In Nature and Society: Anthropological Perspectives, ed. Philippe Descola and
Gísli Pálsson. London: Routledge, 185–204.
D’Andrade, Roy (1995). The Development of Cognitive Anthropology. Cambridge: Cambridge
University Press.
Descola, Philippe (1996). The Spears of Twilight: Life and Death in the Amazon Jungle (trans.
Janet Lloyd). London: HarperCollins.
Descola, Philippe (2005). Par-delà nature et culture. Paris: Gallimard, Bibliothèque des Sciences
Humaines.
Descola, Philippe (2006). Beyond Nature and Culture: The 2005 Radcliffe-Brown Lecture in
Social Anthropology. Proceedings of the British Academy 139: 137–55.
Dumont, Louis (1966). Homo hierarchicus: le système des castes et ses implications. Paris:
Gallimard, Bibliothèque des Sciences Humaines.
Durkheim, Émile (1918). Le “contrat social” de Rousseau. Revue de Métaphysique et de Morale
25: 129–61.
Durkheim, Émile (1960). Les formes élémentaires de la vie religieuse: le système totémique en
Australie [1912]. Paris: Presses Universitaires de France.
Durkheim, Émile (1996). Sociologie et philosophie [1924]. Paris: Presses Universitaires de France.
Duval, Maurice (1986). Un totalitarisme sans état: essai d’anthropologie politique à partir d’un
village Burkinabé. Paris: L’Harmattan.
Héran, François (1987). L’institution démotivée: de fustel de coulanges à durkheim et au-delà.
Revue Française de Sociologie 28(1): 67–97.
Hobbes, Thomas (1914). Leviathan, or the Matter, Form, and Power of a Commonwealth,
Ecclesiastical and Civil [1651]. London: J. M. Dent.
Houseman, Michael (1984). La relation hiérarchique: idéologie particulière ou modèle général.
In Différences, Valeurs, Hiérarchie: Textes Offerts à Louis Dumont, ed. Jean-Claude Galey.
Paris: Éditions de l’École des Hautes Études en Sciences Sociales, 299–318.
Kant, Immanuel (1795). Zum Ewigen Frieden: Ein Philosophischer Entwurf. Königsberg:
Friedrich Nicolovius.
Latour, Bruno (1991). Nous N’avons Jamais été Modernes: Essai D’anthropologie Symétrique.
Paris: La Découverte.
Latour, Bruno (2005). Reassembling the Social: An Introduction to Actor-Network Theory.
Leach, Edmund Ronald (1961). Rethinking Anthropology, London School of Economics
Monographs on Social Anthropology no. 22. London: Athlone.
Lévi-Strauss, Claude (1950). Introduction à l’œuvre de Marcel Mauss. In Marcel Mauss,
Sociologie et Anthropologie: Précédé d’une Introduction à L’oeuvre de Marcel Mauss par
Claude Lévi-Strauss. Paris: Presses Universitaires de France, xlvii.
Lévi-Strauss, Claude (1958). Anthropologie Structurale. Paris: Plon.
Lévi-Strauss, Claude (1962a). La Pensée Sauvage. Paris: Plon.
Lévi-Strauss, Claude (1962b). Le Totémisme Aujourd’hui, Collection “Mythes et Religions.”
Paris: Presses Universitaires de France.
Mandler, Jean M. (1984). Stories, Scripts and Scenes: Aspects of Schema Theory. Hillsdale, NJ:
Erlbaum.
Pouillon, Jean (1975). Fétiches sans Fétichisme. Paris: François Maspero, Bibliothèque
d’anthropologie.
Radcliffe-Brown, Alfred R. (1958). Method in Social Anthropology: Selected Essays, ed. M. N.
Srivinas. Chicago: University of Chicago Press.
9781444333237_4_012.indd 225
8/7/2010 7:09:19 PM
226
Descola
Schank, Roger, and Robert Abelson (1977). Scripts, Plans, Goals and Understanding: An
Inquiry Into Human Knowledge Structures. Hillsdale, NJ: Erlbaum.
Schumpeter, Joseph A. (1954). History of Economic Analysis. London, Allen and Unwin.
Shore, Bradd (1996). Culture in Mind: Cognition, Culture and the Problem of Meaning. New
York: Oxford University Press.
Strauss, Claudia, and Naomi Quinn (1997). A Cognitive Theory of Cultural Meaning.
Tarde, Gabriel (1999). Monadologie et Sociologie [1893]. Paris: Les Empêcheurs de penser en
rond.
Viveiros de Castro, Eduardo (1996). Os Pronomes Cosmológicos e o Perspectivismo Ameríndio.
Mana 2(2): 115–44.
von Brandenstein, Carl Georg (1977). Aboriginal Ecological Order in the South-West of
Australia: Meanings and Examples. Oceania 47(3): 170–86.
von Brandenstein, Carl Georg (1982). Names and Substance of the Australian Subsection System.
Zuidema, Tom (1986). La Civilisation Inca au Cuzco. Paris: Presses Universitaires de France.
9781444333237_4_012.indd 226
8/7/2010 7:09:19 PM
13
Holism and the Transformation
of the Contemporary
Global Order
Jonathan Friedman
Introduction
The theoretical aim of the present chapter is to argue for the importance of holistic
interpretations in the understanding of the contemporary global order and its most
recent transformations. The current effort and interest in the restoration or at least
maintenance of a holistic perspective are perhaps more obvious today when it appears
necessary in order to grapple with a contemporary world that social sciences have
relinquished to the joys of chaos, parts, “partial truths,” and various forms of assemblage. It is still evident in quasi-prophetic calls for
a shift in mode, from a holistic approach, focused on the study of a culture, toward a
contemporary approach focused on assemblages, problematizations, rationalities, and so
forth. (Rees in Rabinow et al. 2008: 79)
In order to counter the above vision and even the particular use of the term, however, it is important to be clear concerning that to which we refer since the word itself
has a long and varied history. It is neither necessary nor possible given the constraints
of the current chapter to discuss the issue of holism at length. For our purposes we
can note that holism has two broad usages. The first is simply the assumption that the
social realities that we deal with can be understood as wholes, as structures, as orders
of different elements linked to one another in comprehensible ways. Holism, in this
sense, refers to the assumption that social reality can be understood in terms of the
interrelation of its different parts or phenomena, and that the latter are connected in
some systematic way. Functionalism represents the empiricist variant of this approach
insofar as it is assumed that wholes are tangible, concrete, and observable phenomena,
while structuralism assumes that wholes are theoretical constructs that account for or
explain phenomena that are otherwise disparate and apparently unrelated to one
another. The structuralist point of view is identical to that of the natural sciences,
where fundamental properties of reality are not themselves observable but rather posited as underlying features that account for our observations. You cannot see gravity
but you can see its effects. It is this that resonates with Marx’s dictum,
9781444333237_4_013.indd 227
8/7/2010 7:09:42 PM
228
Friedman
All science would be superfluous if the outward appearance and the essence of things
directly coincided. (Marx and Engels 1967: 797)
This holism is simply about the act of understanding itself, the reduction of the phenomenal to its underlying principles, a set of systemic interconnections.
However, there is another usage of the term “holism” that is emic rather than etic
and that follows from the work of Dumont on India and on its comparison with the
West. This holism in the most general sense refers to a particular social order in which
fundamental categories encompass all others, which become subsets of the former.
Thus the well-known contrast between Homo hierarchicus and Homo aequalis (the
titles of Dumont 1967 and 1976, respectively) is an argument for a distinction between
a society within which the individual is encompassed within the larger social order and
one in which the individual is the dominant value or, in other terms, category so that
the social order either is generated via interaction or exists as an imposed order with
respect to the individual. But in the latter case as in the former, the individual is a
socially determined category and individualism as such is not a sign of real autonomization with respect to the social order – not, at least, in etic terms. This is a crucial distinction, one between holism as a theoretical approach and the proposition that there are
social orders in which social encompassment is the dominant organizational principle.
Dumont represents both usages of holism in his work on India and on the history
of individualism. “Holism” refers, on the one hand, to a holistic general understanding of the social order, one that assumes that if social life were random there would be
nothing to understand. As my professor of linguistics once taught me, the purpose of
theory is to account for departures from randomness. On the other hand, as stated
above, holism also offers a particular view on the relation of the individual to society,
one in which the social is always primary, even in individualistic societies, insofar as the
individualized subject is itself a product of a particular social order. Actors are, of
course, autonomous subjects in one sense or another, but if men do have free will, the
will is not free. At least this is a working hypothesis of such an approach.
Global Systemic Anthropology
The term “global system” might appear contrary to what has been conceived as an
abandonment of what is disdainfully referred to as “modernist scientism,” not least as
the abandonment is understood by those involved, as an intellectual evolution or even
a transcendence. The term has, thus, a function in denoting something that I feel is
important, not least in this era of fantasy concepts such as “millennial capitalism,”
“fear of small numbers,” creolizations, hybridizations, assemblages, and the like,
terms that have filled the vacuum left by scientific thinking. The terms are enmeshed
in the discourses of globalization which came, not from anthropology and certainly
not from research in our field, but from business economics, economic geography,
and, more recently, postcolonial cultural studies. In the latter versions that have been
appropriated by certain anthropologists, we have arrived at a new era, and certain
anthropologists have even gone over into that other, global world and beckon us, as
prophets, to join them. This is not merely an anthropological delusion. Even quasiphilosophical works such as Empire (Hardt and Negri 2000) are full of biblical quotations and allusions, almost all to the prophets. And all of it is, it might be suggested,
9781444333237_4_013.indd 228
8/7/2010 7:09:42 PM
229
based on personal experience of a particular position in the world, that of the traveling
elite scholar, and here the real source of the new discourses is the presumably the
“class consciousness of the frequent flyer,” since in fact the figure that has become a
sacred object of adoration, the nomad (completely misconceived from an anthropological perspective), is not a more representative phenomenon today than it was a
century ago, to say nothing of previous centuries. While not dwelling on this problem, there are particular features that are important to note. First is the implicit evolutionism based on the assumption that we have moved from the local to the global,
understood as caused by technological change. Second is the tendency, especially
among certain postmodern-oriented writers such as Appadurai, to deny the value of
holistic accounts in general. The latter has done this from the start, most clearly in his
critique of Dumont.
There is, I would suggest, a fundamental set of misconceptions in Appadurai’s
appreciation of Dumont. First, for a critique that pretends to discover the way in
which external categories are applied to falsely characterize populations, his own
assumption that Dumont’s association of caste with India is a form of “incarceration”
as are all classical anthropological categorizations that link geography to particular
cultural traits is a reductionist misconception of the latter’s work, which deals with the
nature of caste as a kind of social logic and not as a geographical container. Dumont
does not try to essentialize India but to understand the workings of a particular structure. And he discusses all kinds of empirical cases that are at variance with the “pure”
model as well as engaging in debate with other anthropologists on this issue. Appadurai
offers no critique of Dumont’s conclusions or his model but, instead (something that
is all too common), seems to attack him by mere labeling (i.e., a guilt by association,
in this case with “essentialism”). The net message is that we should be wary of holism
as a dangerous colonial legacy. In the same article Appadurai praises diffusionism as
liberating (1988: 39) without considering its basis in European racialism. The agenda
is clear: fragments are good, wholes are bad, locality is incarceration, and movement
is freedom.
This is not so much a critique of emic holism but of what is conceived as the essentializing tendency inherent in holistic models (i.e., that there are specific structures or
cultural forms associated with specific populations). His own view of the global as
consisting of divergent scapes and riddled with chaos is a prime example of an antisystemic and antiholistic perspective. To say, as Appadurai does (1990), that technoscapes, financescapes, ethnoscapes, and so on are divergent in the era of globalization
is to simply deny the strongly systemic relations that connect the phenomena so artificially divided. To call them scapes at all is also a peculiar reification of phenomena
that are deeply intertwined in broader social logics.
Global systemic anthropology begins at the other end, with an attempt to grasp the
significance of the global with respect to whatever local phenomena are under study.
It begins, in one sense, with a foundational argument of Lévi-Strauss’ work on kinship. The elementary structure (or atom) of kinship is an atomic global system and the
expression of the all-encompassing nature of such social relations1. No societies have
ever been isolated except by acts of isolation, by either self or others, and such acts are
themselves expressions of the existence of the larger relational context. This does not
imply, of course, a universal uniformity. On the contrary, the global in this argument
is simply the larger space of total social reproduction within which differences are
maintained and themselves reproduced. If the global has properties of its own, they
are emergent properties of the larger reproductive processes themselves. There is no
9781444333237_4_013.indd 229
8/7/2010 7:09:42 PM
230
Friedman
contradiction between practice and structure here since “structure” refers not to
concrete manifestations, like institutions, buildings, and the like, but to the properties
of social practices. This is also an interpretation that Lévi-Strauss properly understood
in his debate with Sartre in La Pensée Sauvage (1962). In other words, structure refers
to structural properties of social process and not to their products, for example, the
kinds of behavioral relations that were the ultimate reality for Radcliffe-Brown. For
structuralism it is about the generation of phenomena and not the phenomena themselves. This implies that global systems are not geographically delimited entities but the
properties of the global processes that both produce and transform such entities in time and
space. The issue of boundaries is an internal issue of the practice of bounding within
particular social orders and not of objective boundaries, independent of practice.
It is in this kind of argument that “globalization” was conceived not as an evolutionary stage but as a phase in a pulsating system of expanding and contracting centers. This implies a reconceptualization of all phenomena that fall within the discourse
of globalization. No understanding of the so-called scapes, for example, is possible
without an understanding of the categories that constitute such phenomena.
Ethnicity is a particular phenomenon that is culturally variable in the way it is constructed. The notion of ethnoscape implies a homogenization of what are very different realities, thus the very essentialization that Appadurai abhors. Ethnicity in
whatever form it is practiced cannot be simply combined with a spatial-demographic
notion of scape. The fact that large parts of the crisis-ridden world are experiencing
what is referred to as “ethnicization” can only be comprehended by placing practices
of ethnicization in a structured articulation with demographic movement, the collapse of social mobility, and the increase of situations of potential conflict, all of
which can be linked with the condition of “hegemonic” decline in which processes
of homogenization turn into their opposites, processes of fragmentation and heterogenization in the zones of decline, not for example in China, where the inverse is the
case, however strained.
Global systems are not systems somehow out there, above the local, and that produce the local. They refer to the structural properties and/or logics, of the processes
of reproduction that link and constitute-by-articulation localities, and the latter are
local in the social sense rather than a geographically reductionist sense. The global
does not come after the local, but is always, from the Mother’s Brother and up,
constitutive in the interactive sense. It is not an empirical (as in world systems literature) but a structural concept, one closer to the structuralist notion of “order of
orders.” In the following, I shall make a go of integrating the holism of global systemic analysis with the holism of structures of representation and organization that
are fundamental to the contemporary period of sociocosmological transformation.
Hegemonic Decline and Cultural Transformation
Horizontal/vertical polarization
The transformation referred to here is one in which modernist-based identities have
declined. The latter were founded on movement away from cultural roots toward a
more abstract sense of development, of both self and society. The cultural was reduced
to lifestyle, a question of roles, of clubs, of voluntary association, rather than a deeper
9781444333237_4_013.indd 230
8/7/2010 7:09:42 PM
231
sense of belonging. The overriding tendency of this earlier period, one that transformed
the cultural order of Western societies during the nineteenth and twentieth centuries,
was a distancing from the fixed and culturally embedded, one that was often expressed
as a liberation from “tradition” (even though, as Marshall Sahlins has so well put it,
this liberation process is a particular Western tradition in its own right; Sahlins 1999).
It was underpinned by real social mobility, not primarily geographical (although
urbanization is certainly part of it) but also an increase in life chances often based on
the gains of bottom-up class struggle. And it was all dependent upon the increasing
economic expansion of the West and its achievement of world hegemony. As the limits
of this development approached, modernism began to lose its punch; the evolutionary future began to fade and the past loomed as the more concrete “real” future,
hence the rise of the practice of rooting, of religiosity, ethnicization, and indigenization. The decline of hegemony has led to four large-scale tendencies to collective
reidentification:
1. Regional: Regional minorities began to assert themselves from the mid-1970s
(Bretons, Occitanists, Catalans, Scots, Welsh, Cornish, and numerous others).
This is documented in numerous texts and studies of the new ethnic movements
and in classical works such as Tom Nairn’s The Break-Up of Britain (1977). It has
continued even today at the highest level with the ambivalence of English identity
in relation to “Britishness.” Even the English can be said to have been forced to
deal with themselves as a regional identity within Britain, where the former imperial identity is no longer dominant. Regionalism is an expression of the fragmenting of the larger national unity.
2. Indigenous: Indigenous populations (according to the UN definition) were hardly
an issue until the mid-1970s, but the category has undergone an explosive development since then in large parts of the world. However, it is most drastically pronounced within the declining hegemonic spheres, in the West and Japan and in
their internal and external peripheries. From a situation in which such groups were
disappearing and by and large ignored, there has emerged an increasing self-identification (according to some sources, almost 500 million) and the institutionalization of global representation within the UN system.
3. Immigrant: In periods of expansion, there is a strong tendency to vertical integration within host countries, as was the case during a large part of the twentieth
century where both labor migration and refugee-based migration fed into a substantial assimilation or at least ordering of populations within receiving states.
The reversals spoken of above are expressed in the declining capacity of states to
integrate immigrants and a consequent marginalization, enclavization, and
diasporization of identity (i.e., the maintenance, revival, and/or reinforcement of
long-distance ethnic identities).
4. National: There is a shift in this period from nationality-as-political-citizenship to
nationality-as-cultural-citizenship. The culturalization of nationality in Europe is
also an expression of the ethnicization process. This is not a mere reaction, as some
have suggested, to globalization or to mass migration. It appears to have occurred
in the wake of a broader transformation.
I have suggested elsewhere (Friedman 1994) that the reason that these “new” identities work is because they are rehabilitating for those who participate in them. This is
an important aspect of the process that should not be forgotten. It is not merely about
9781444333237_4_013.indd 231
8/7/2010 7:09:42 PM
232
Friedman
Cultural Identity
HEGEMONY
centralization
decentralization
resistance
traditionalism
post-modernity
cultural decline
pluralism
modernism
Civilizational
Cycle
fragmentation
“HOMOGENY”
Figure 13.1 Hegemonic and cultural cycles in global systems
fragmenting, but from the individual subject’s point of view about rebirth. Thus fragmentation at the state level implies simultaneous integration at the subnational or
substate level. It is important to note that these are always partial statistical tendencies
that are themselves filled with variation and even conflict. The relation between the
cycle of hegemony and the changing practices of identity can be depicted in terms of
the graphic in Figure 13.1, one that was first suggested in the 1980s but still appears
to be suitable as a rendering of this process.
I have referred to this process in terms of horizontal fragmentation or even polarization, since the former often leads to the latter in periods of increasing competition
and crisis. Multiculturalism or more generally pluralism can be more or less deduced
from this process. It is simply the breakup of larger modernist political spheres into
smaller self-identified fragments that compete for “recognition” within those spheres.
This is reflected in the intellectual discourses that emerge in the period (from the late
1970s onward) in which all larger formations, whether scientific paradigms or nationstates, are attacked as Foucauldian sources of homogenizing power. Thus postmodernism and its postcolonial variants perfectly express this change in vision without
being able to account for it. In anthropology, for example, this is the period of declining structuralisms and Marxisms and rising culturalisms. The Left that had once identified itself in terms of the overturning of capitalism became something quite different.
From having protested against American imperialism, understood in structural and
material terms, it turned to the critique of American and then all of Western culture
as the very basis of this imperialism, now understood as cultural imperialism. This
was an era of class mobility for many academics, including those from the upper
classes of the Third World (Dirlik 1997), notably in a period in which real workingclass wages were on the decline. The mobility of these populations is an essential
aspect of what is referred to as “vertical polarization.” The struggle became relocated
within university departments, especially in the United States, and thus began the era
known for the use of the term “political correctness.” Former leftists who refused to
follow suit (such as erstwhile SDS leader Todd Gitlin [1995], and Russell Jacoby,
who has made an interesting contribution to the issue with his The End of Utopia:
Politics and Culture in an Age of Apathy [1999]) were reclassified by many as reactionary or at least regressive in relation to the new liberal multiculturalism (represented by, e.g., Stanley Fish). It is interesting to follow the fate of particular authors
9781444333237_4_013.indd 232
8/7/2010 7:09:42 PM
Table 13.1
233
The inversion of left ideology
1968
The national
The local
Collective
Social(ist)
Homogeneous
Monocultural
Equality (sameness)
1998
The postnational
The global
Individual
Liberal
Heterogeneous
Multicultural
Hierarchy (difference)
as markers of larger changes in the state of the world system. The latter change in
state, also described in part by Jacoby, is what can be called the inversion of ideology,
represented in part as shown in Table 13.1.
Jacoby states quite simply what he sees as an inversion of left ideology: “We are witnessing not simply the defeat of the left, but its conversion and perhaps inversion”
(1994: 11). After noting the rise of multiculturalism and its politics – going back to
Horace Kallen, who was more pessimistic about the idea because he saw mass migration creating so much opposition – Jacoby argues that pluralism has become a substitute for radicalism: “The rise of multiculturalism correlates with the decline of utopia,
an index of the exhaustion of political thinking” (1994: 33).
The linkage between global transformation and the emergence of multicultural
ideology connects declining hegemony with the decline of modernism, the inversion
of leftist ideology, and the replacement of material critique by the critique of Western
culture, of the national as essentialist and homogeneous. In this process, hierarchy
also returns as a figure of an alternative future. Among those who have been involved
in the recent Balkan conflicts, there are those on both Left and Right (I shall return
to this interesting fusion) who have begun to think of something like the Habsburg
Empire as a solution to such ethnic strife. And among many “postcolonial” authors,
the Ottoman Empire has been celebrated for its successful multiculturalism and is
defined in opposition to that awful phenomenon the Western nation-state, all the
while forgetting the actual social order of that empire and its violently oppressive
character (which ought, of course, to be expected from any imperial order). Even the
caste order, at least under the Raj, has come in for celebratory praise, for compared
to contemporary Mumbai with its Hindu nationalism things were better in colonial
India, which was, perhaps, the true multicultural paradise lost where there was a
place for everyone, including the poor. Opposing the violence of contemporary
Hindu nationalist rule, Appadurai harks back to a past (that turns out to be the
British Empire) where things were much better, even for those at the bottom of the
social order.
The truly destitute were always there, but even they fit into a complex subeconomy of
pavement dwelling, rag picking, petty crime, and charity. (Appadurai 2000: 629)
Appadurai then turns to the recent change that upset this ordered hierarchical life:
“Sometime in the 1970s all this began to change and a malignant city began to
emerge from beneath the surface of the cosmopolitan ethos of the prior period”
(Appadurai 2000: 629). The implied contrast is that between an exclusionist
9781444333237_4_013.indd 233
8/7/2010 7:09:42 PM
234
Friedman
nationalism and a cosmopolitan pluralism based on a cultural division of labor, all the
more peaceful since everyone knows their place. It might be suggested that Hindu
nationalism, not a fixed entity, is primarily the expression of a strong nationalization
process, typical for periods of hegemonic expansion. It should not be assumed, then,
that it is indicative of the kind of ethnic fragmentation discussed above.
In any case, the fact that the increasingly obvious identity conflicts that characterize much of the world are seen as a deplorable error by many “cosmopolitan intellectuals” provides an important linkage between that fragmentation and a vertical
polarization that has produced a vigorous new cosmopolitan identity. Since the
1980s, there has been a rapidly increasing class stratification in the West. It is expressed
in the divergence between rising middle- and upper-class incomes and the downward
mobility of the lower-middle and lower classes. Gini Indexes, the major statistical
measure of income inequality, provide some idea of this process which has increased
since the mid-1980s and 1990s, not least in central Western countries such as the
UK, the US, and even Sweden, once famous for its extreme equality and where the
Index increased by 25 percent over a 4-year period in the beginning of the 1990s.
And it is not merely an issue of the astronomical incomes of certain financial capitalists. There is a new breadth to this process that includes various elites: media, cultural, academic, and not least political. In our research on Sweden and France, we
have noted that this is a question not just of national class identity but also increasingly of a cosmopolitanization of the elites, who identify more as citizens of the
world than of their particular countries. The former CEO of media giant Bertelsmann,
Thomas Middlehoff, said concerning his role in the American media world:
We’re not foreign. We’re international.… I’m an American with a German passport.
(Quoted in McChesney 1999: 4)
But even in the case of a Swedish minister of integration, who when asked if he was
Swedish replied, “NO definitely not!” what can be detected here is an emphatic identification out of the nation-state and, more specifically, out of the nation. This is not
a quirk but a systemic effect of the inversion of Western ideology coupled with the rise
of a new “postnational” elite. This is the distinct expression of the cosmopolitanization of the elites of the world system and not the moral imperative of a new class of
do-gooders.
Cosmopolitanization
Cosmopolitanization is a cultural identification process that tends to create a rather
closed group over time, in spite of the usage of words like “open to the larger
world,” citizen of the world,” and the like. Members of this category often associate
only with their own “kind” in positional or class terms and are often part of increasingly endosocial and even endogamic networks. This is not a particularly new phenomenon. On the contrary, it can be suggested that cosmopolitan identity is a
structure of the long run, perhaps even longer than modern capitalism, since similar
representations of self and world can be found in a great many previous eras, going
back at least to the Hellenistic states. What has been referred to as the “cosmopolitan bourgeoisie” (Jones 1987) of the pre-corporate period in Europe exemplifies
precisely this pattern:
9781444333237_4_013.indd 234
8/7/2010 7:09:42 PM
235
Those traditional firms were operated by families with Liberal inclinations and a relaxed
attitude to nationality. Many of them had moved from one state to another within the
past few decades. Now as far as local cultural and political circumstances would permit,
they inclined to mix freely with families of their own sort, regardless of origin. (Jones 1987:
91–2; emphasis added)
Recent research on foreign elites in France (Wagner 1999) reveals the closed nature
of this openness, and that it is founded upon a genealogical mode of identification
which is a crucial aspect of closure itself:
I have expatriate blood.… I am American by passport and nationality but my family and
that of my wife have a great many branches in many countries, which entails that one
always has one foot in the United States and one abroad. (Informant quote in Wagner
1999: 116; my translation)2
My father was something of a vagabond, we have it in our veins. My brothers are the
same: I have a brother in Austria, one in Finland, a sister in Spain. My father moved often
and I must have learned it from him. (Informant quote in Wagner 1999: 116; my
translation)3
Here the cosmopolitan is certainly not hybrid as a social category even if it crosses
all sorts of national boundaries. It defines a status that is quite distinctive with respect
to national categories, whether they are mixed or not. It is noteworthy that it is transformed into terms of blood as is the traditional practice among most aristocrats of the
past. In order to share hybrid blood it is necessary to maintain a significant degree of
endogamy, thus producing a closed group, no matter how it announces itself to the
world. The curious contradiction in all of this for many multiculturalists and selfproclaimed hybrids is the fact that mixture is here associated with roots and blood,
which appears on the surface to be an impossible combination of openness and closedness, of essentialism and hybridity. Now aside from the fact that this is simply a
problem of faulty thinking, it demonstrates a deeper confusion. The fact that people
network across national boundaries does not in any way contradict the fact that they
live well within other social boundaries. This may sound trivial, but it is exactly the
error committed by those who celebrate cosmopolitan hybridity as a New Age,
a world without boundaries. On the contrary, cosmopolitanism is yet another example of diaspora (in social terms) formation in which ethnicity may well be transnational, but is nevertheless an expression of social closure. One might add that this
extreme embodiment of mixture is a strong form of precisely what the hybridists
abhor in essentialism. But this essentialist identity represents itself as the epitome of
ouverture:
Curiosity, openness, tolerance are terms used to designate the qualities of this position.
(Wagner 1999: 142; my translation)4
And opposed to this are the Others, the people who inhabit the bottom end of the
social order and who are represented as “terrestrials”:
The earthling, then, is someone who occupies a limited space. His activities are concentrated on the lands that he possesses. If another person is seen on his lands he will not
accept him. He is attached to his family, his children, whom he wants to keep at home,
because it is his family that cultivates the land.…
9781444333237_4_013.indd 235
8/7/2010 7:09:42 PM
236
Friedman
I think that the popular classes are more attached to their origins. The English in
France are usually middle class. In England the popular classes are more nationalist than
others, less open. (Wagner 1999: 189; my translation)5
The nationals, the “people,” the popular classes, the dangerous classes, and the indigenes are all part of a vast process of categorization, a cosmology within which particular categories are positioned but where there is also a great deal of variation.
Concentric Dualism, Hierarchization,
and the Remaking of Political Space
While the above discussion is concentrated on self-identification and identification of
others within a larger population, this process dovetails with a broader transformation of
the political order. Cosmopolitanism implies a potential cosmopolitics. The we–they
opposition enunciated by these elites might seem to contradict their self-image as world
encompassing, but this is not a logical contradiction, simply a confusion of categories.
Dumont noted long ago, in relation to the logic of caste, that encompassment was a
form of asymmetrical (concentric) dualism. This was also a crucial argument of LéviStrauss when analyzing the difference between what he called diametric and concentric
dualism in kinship-based societies. The former is typical for a balanced opposition in
which there can only be two groups or multiples of two, whereas the latter represents an
opening up toward third, fourth, or n groups, but now organized in a hierarchical continuum. This particular relation might be used to tackle the political logics of the contemporary state order in transition. If the political elites are a crucial element in this
process of cosmopolitan expansion, if they tend to climb to a higher order of identification, European or global, then there are a number of elements that can be combined into
a new order of governance in this part of the world. These elements are as follows:
1. The emergence of new public management as an alternative to democracy.
Interestingly, it is Bertelsmann, one of the world’s largest media concerns, that
was most deeply involved in this development with their international project on
better local government in which prizes were given primarily for increasing efficiency of the governmental process via downsizing and outsourcing.
The past decades have witnessed the emergence of a new public management discourse, not least in Washington and London think tanks. New managerial philosophy is
linked in international networks referred to as transnational discourse communities like
the Network for Better Local Government, financed by media giant Bertelsmann. The
strategies developed are carry-overs from private sector strategies of effective administration which use techniques of marketing in the reorganization of government. This implies
the replacement of such clumsy practices as voting, political debate based on a plurality
of ideologies[,] by more efficient market based mechanisms. Governance, in this
approach, is no longer an issue of specific political goals but of mere practical solutions
to immediate problems.…
NPM discourse is constantly confronted with a powerful counter-discourse of
“public sector values” stressing democracy, equality, accountability, participation etc. In
9781444333237_4_013.indd 236
8/7/2010 7:09:42 PM
237
People
Iniquitou
s
T he
Elites
Left
Right
The Virtuous
Figure 13.2 Diametric to concentric dualism
NPM discourse, the contradiction does not exist; management techniques are said to
provide the same benefits as classical democratic institutions: to be responsive to customers is equivalent to democratic control, measuring performance is the essence of accountability, choice is pluralism, and so on. (Bislev and Salskov-Iversen 2000: 27)
2. This has entered the debates concerning development of what has been called
“organic” democracy, that is, democracy without the demos whose function is
reduced to legitimizing the organic process, one that is based on negotiations
between principal actors, MNCs, ethnic and other corporate actors, and government itself.
3. In this process of management, the categories of Left and Right largely disappear,
replaced by praxeology. And the latter is expressed by the Right in some countries,
and the Left in others, as la voie unique giving rise to New Democracy in the US,
New Labour in Britain, and, even better, the Neue Mitte in Germany. In this there
is a transition away from dependence upon a “people.” That is why Blair could be
referred to as “Thatcher without the handbag,” and Giddens, referred to by
Bourdieu and Wacquant as “le messager du prince,” can argue that sometimes
“progressive” can mean something quite different than what one usually assumes.
Sometimes the only way to increase welfare is by dismantling the welfare apparatus.
This can be represented as in Figure 13.2.
The people become the dangerous classes in this transformation, and quite popular
leaders often become the opposite of what is deemed truly democratic, whether they
be Rod Livingstone (“Red Rod”), who became mayor of London against the wishes
of the Labour Party, or Joerg Haider of the Austrian nationalist right, who won 30
percent of the vote in a national election. It makes no difference, as long as they are
external to the respectable core. And being not respectable is equivalent to being
undemocratic. But in this process the term “democratic” becomes an essence, an
embodiment of virtue, defined as respectability itself. Thus, when the Swedish foreign minister was assassinated some years ago, it was said that she was murdered in
the midst of carrying out her democratic duties. She was out shopping at an expensive department store. Is that democratic practice? This would imply that no matter
what a certain minister who is an elected representative of the people happens to be
doing, it is automatically defined as the practice of democracy. Sweden actually had a
9781444333237_4_013.indd 237
8/7/2010 7:09:43 PM
238
Friedman
minister of democracy, a position held by the daughter (Britta Lejon) of a previous
minister of justice (Anna Greta Lejon) who said in an interview that it was through
her mother that she was awarded the position. Is this part of a redefinition of the
political field in which all actions by the “correct” people are necessarily of a democratic nature?
This transition is a mere tendency and is surely contested and incomplete. But if
realized it would imply something that is partially accomplished on a fairly broad scale
in a set of discourses and practices that elevate the state above those who it represents.
This would seem to imply a historical retransfer of sovereignty from the people to the
state. The tendency is evidenced in the recent elections and postelection activities in
France, where a new centrist “democratic” party has been created and where the new
conservative president, Sarkozy, has successfully recruited several famous socialists to
participate in his government. In Italy the new “democratic” party is a strange fusion
of Christian Democrats and leftists.
Vertical polarization is a decisive factor in the logical transition to multicultural
states. The distance created by upward mobility toward the cosmopolitan stratosphere
recreates a positioning similar to the cosmopolitan Freemason elites of the nineteenth
century, and is one in which a quite particular relation emerges between the governors
and the governed.
The cosmopolitan bourgeoisie in the eighteenth century came to adopt a perspective on
society as if it were a foreign one[,] a target for “colonial” exploitation. Freemasonry
provided a cover for developing the new identity on which the exploitation of members
of one’s own community is premised. By entering the masonic lodges, merchants and
those otherwise involved in the long-distance money economy[,] such as lawyers and
accountants, realised the primordial alienation from the community which is the precondition for market relations, exploitation of wage labour, and abstract citizenship.
(Rosenstock Huessy 1987 [1961]: 364, quoted in Van der Pijl 1998: 99)
This “foreignness” is the basis of the liberation of the elites from the role of representative and embodies the possibility of a reconfiguration of sovereignty. This is
exemplified by the attempt in Sweden to constitutionally introduce a multicultural
state in place of the former nation-state. Sweden is an extraordinary example
because of its particular history of being extremely homogenizing and then suddenly declaring itself to have a new organization based on an increasingly multicultural or even multiethnic foundation. In the government proposition of 1997–8
that is the basis for this transition, it is stated that Sweden no longer has a common
history and that it instead consists of a number of different ethnicities, including
Swedes. This is a formula for the transformation of the nation-state into a plural
society, one that has been partially implemented in recent years. The vertical polarization which leads to the emergence of a political class identified as outside or
beyond the nation transforms the latter into an ethnic group along with others and
negates the notion of peoplehood. The state thus becomes a neutral actor above
society. And this combined with new public management creates a power thrust
toward an absolutist order of governance. The proposition contains the following
revealing statements:
The point of departure for a new politics according to the government: Society’s ethnic
and cultural plurality should be the basis of the formulation of general policy and its
implementation in all the domains and levels of society.…
9781444333237_4_013.indd 238
8/7/2010 7:09:43 PM
239
As a process, integration concerns the way in which different parts are united in a
larger unity.…
A country’s history often serves to integrate individuals in a larger unity. As a large group
of people originate from other countries, the Swedish population now lacks a shared history. (Government of Sweden 1997–8: 19, 22, 23)
The upward mobility of elites has also been described for France, although here in
a situation in which republicanism (i.e., a strong sense of nationality) has been maintained, albeit with numerous conflicts. Work by Julliard and Joffrin has tried to show
the way in which a Left elite substituted cultural politics for class politics and in which
the people have again become a dangerous class. In this process, the move to pluralism is again part of the same logic.
The exhaustion of the social democratic model has transformed militants of the revolution, then of reform, into militants of cultural liberalism. (Julliard 1997: 201; my
translation)6
This is expressed in the following shift in politics:
For workers it has substituted immigrants and has transferred to them the double emotion of fear and compassion that the proletarian once inspired. However, the immigrant
is not only the victim of social exclusion, but also of ethnic exclusion, in other words of
racism. (Julliard 1997: 201, 105; my translation)7
Indigenization
Hybrid Cosmopolitans
National
Elites
National
Elites
Migrants
National
population
Diaspora formation
Ethnic Minorities
National
population
Indigenization
Cosmopolitanization
At the lower end of the social order, this polarization provides a mirror image of the
cosmopolitanism of the elites. Here also we see the ambivalent relations that develop
in an arena in which there is both extreme ethnic fragmentation and class polarization.
The indigenization referred to in Figure 13.3 is a product of the collapse of the recognition of class identity and class politics.
Ethnic Minorities
Indigenes
Indigenes
Rooted ethnicities
Figure 13.3 The dialectic of cosmopolitanization and indigenization
9781444333237_4_013.indd 239
8/7/2010 7:09:43 PM
240
Friedman
The downwardly mobile find refuge in national identity or in a stronger localism
in which rootedness is a central figure. The most extreme examples of this process
are revealing with respect to the larger structure. There is the example of the
Washitaw Indians (self-identified) who inhabit several states of the US South and
who are black. This is a new tribe whose name is related to the Wichita, who appear
to have adopted runaway slaves in the nineteenth century, but who today represent
an emergent indigenous identity. They claim to have come to the New World when
Africa was still joined to that continent (i.e., before the arrival of the Red Indians).
They demand sovereignty, have a home page and empress, and are allied with the
Republic of Texas, a well-known white racist militia group. How can this be, one
might ask? The answer lies in their common goals and common enemies. They
demand autonomy and control over their own territory, as do militias. They do not
seek to take over the state. Their main enemies are Washington, The Vatican, the
Jews, and all other cosmopolitans and their agenda of world rule. The head of the
KKK in St. Petersburg, Florida, was once a local head of SDS. He sports a poster
of Che Guevara in his office. But he trains together with a local black power group.
Again, the two unlikely partners agree to be separate from one another and they
have the same enemies as the previous group. These are indigenizing localist movements that cannot be confused with the state-focused ideologies of fascism and
Nazism. They display similarities with the European New Right which is multicultural in some ways, supporting the rights of all culturally identified groups while
rejecting mixture. It is worth noting that indigeneity can leave the framework of
the nation-state at the bottom, just as cosmopolitans leave it at the top.
Internationalist racism just as the international organizations of indigenous populations share a common structural position.
This process is also related to the general ethnicization of lower-class mixed zones
in the world, where turf wars abound and identities are tribalized. There is a fragmentation, grassroots as well as state implemented, and this occurs in a situation in which
locals or nationals are equally part of the process, leading ultimately to a state of
generalized tension and conflict. In the fascinating book Redneck Manifesto, this discourse from the bottom is quite explicit:
I tried living in the big-city multicultural thingie for twelve years, only to realize that
most of multiculturalism’s proponents – rich white people – didn’t want me. So I moved
to a neighborhood that is redneck, blue-collar, white trash. Low rent. Low class. Lowlife.
Truckers, welders, meth dealers, pit bulls, rotted picket fences.
St Johns is most notorious for its high white-trash quotient. Yet more blacks and
Mexicans live here than in most parts of the city. For economic reasons, the trash – be
it black, brown, or white – have always lived side by side in America. It’s the Gold Card
whites who’ve always paid to segregate themselves, leaving the rednecks, niggers and
spics to fight over day-old cookies. (Goad 1997: 35)
Here the consciousness that the fragmentation is strongest at the bottom of the social
order is salient as well as the fact that the bottom is itself a class phenomenon. This
generates an explicit aggression against those “cosmopolitans” who are held responsible for multicultural ideology while they themselves live entirely segregated lives
with respect to the real problems of these dangerous zones.
The polarization addressed here is a local-global articulation within the world
system that has had a massive effect on the reconfiguration of territorial states, in
9781444333237_4_013.indd 240
8/7/2010 7:09:43 PM
241
this case nation-states, producing rising cosmopolitan elites who identify with a unified globalized world and declining middle and lower classes who engage in a process of indigenization. The polarization generates a complementary opposition
between the two so that they each identify themselves by contrast with the other.
These are not, as we have stressed, new oppositions, new categories. What have
changed are the sharpness of the conflict and the saliency of the categories themselves. The multicultural representations of the elites divide and fragment the earlier
representation of “peoplehood” and even of class, while the indigenizing ideologies
of the lower reaches of society invest in precisely this representation, now turned
into either national-ethnic identification or subnational secessionist identities.
Geopolitical Fracture in the Global Order
Above, we discussed the way in which hegemonic decline was expressed in the form
taken by fragmentation as identity politics and the tendency to the formation of
minority communities of different kinds: regional, diasporic, indigenous. The
diasporic, we argued, was the product, not of demographic movement but of the
way identity is practiced in such movement. In Paris, where there were only a handful of mosques in 1970, the number rose to well over a thousand in the 1980s and
has continued to increase. Among North Africans who were once economically
integrated into the French economy and even “the Republic,” the long economic
decline led to increasing long-term unemployment. In an interview, a woman from
one of the northern banlieues of Paris recounted how her life in La Courneuve had
changed since the 1970s. She was herself white and had been married first to a North
African and then a black African, and had children in both marriages. In the early
days, she said, there was a strong sense of class solidarity and generosity, and ethnic
differences were marginal phenomena. Today things have become quite the opposite. She is afraid for her children, who are often attacked by members of other
groups, and she longs to live elsewhere with her family. There is a high level of
gang violence within the area, and any sense of local solidarity is all but gone. In
this period, many North Africans who were brought up without knowing Arabic
are now (re)learning the language, and Islam, which was a minor affair in the past,
has now become a dominant form of identity, not least political identity.
Here there is a further connection to a geopolitical fracture, that which has become
evident since the attack on the World Trade Center, and one that has again made
Huntington’s The Clash of Civilizations (1996) into a hot issue of discussion and
debate. There are numerous accounts of this phenomenon, and without discussing
them here, it might be suggested that the model of hegemonic decline that we have
suggested is a useful frame of analysis. There is an interesting interpretation from
within the Muslim world, that of Zaïki Laïdi, who links the current conflict to what
might be called the “Ottoman blues”:
This idea of loss of status, of demotion of the Muslim world is central. In the Muslim
imaginary, belonging to a powerful civilizational complex that rivaled and even surpassed
the Occident is essential. Without this historical fact going back more than ten centuries,
the feeling of demotion would probably not have the same force. (2001: 14; my
translation)8
9781444333237_4_013.indd 241
8/7/2010 7:09:43 PM
242
Friedman
There is a geopolitical vision here, one that is similar to that of the West, one that
is typical of all imperial civilizations as well. But the fact that it has been mobilized
at this point is significant, since it would seem to coincide not with the formation of
(American) Empire, but with its decline and failure. This does not necessarily mean
that there is a real contention for world power involved, although the idea of an
alternative world government based on Sharia is perhaps the only fully developed
example of “alter-globalization.” The diasporic connection itself clearly plays an
important role, since many of those who were earlier recruited to the cause of
Islamism were themselves immigrants to the West – and often economically privileged, to say the least – but who, at least in some cases, apparently snapped, realizing
their ultimate marginalization by what they saw as a decadent civilization whose
values were suddenly entirely rejected. The support for acts of terror in the ghetto
areas of the urban zones was often based on a similar set of interpretations among a
population that was increasingly self-enclavizing. Muslims cannot be said to have
been the least well off in Europe and the US in the years previous to the escalation
of conflict, and even today those most discriminated against belong to other groups,
such as Africans (non-Muslims included) and Roma.
What is specific in this “clash” is the mobilizing potential linked to precisely the
geopolitical imaginary that is today invoked by those intellectuals and others engaged
in Islamist struggle. To deny this and to account for their activities in terms of US
aggression, Israeli aggression, or internal corruption in the Middle East are totally
inadequate. It is a typical “Occidentalist” posture of many Western intellectuals who,
ensconced in the inversion of ideology referred to above, do not understand the
degree to which they, in this way, totally pacify the Islamist actors who are reduced
to mere re-actors. They are in this way robbed of any project that they might in reality possess, and closer examination reveals that there is indeed a real project involved
here. In identity terms, Islamism can be related to the process of indigenization, but
it is certainly not a localizing strategy. The similarities are primarily related to the
strong form of unitary identity and cultural closure which has made it attractive for a
number of actors on the Right and especially the extreme Right which share the
opposition to the cosmopolitan elites.
The highly orchestrated reaction to the publication of the Mohammed caricatures
in Denmark is further evidence of the consolidation of this identity. It is not so much
about the different issues that have occupied so many intellectuals, although there
are interesting issues involved. Images of Mohammed are not strictly forbidden, and
there are plenty of historical examples of such images. Some of the caricatures may
well have been insulting but certainly not all or even most of them, and some were
instead critical of Pia Kjærsgård, leader of the nationalist Danish People’s Party,
Dansk Folkeparti. Among the images that were taken by a particular Danish Islamic
representative to ignite the issue in Egypt and elsewhere were a number of misrepresentations, including one drawing from a “pig feast” in the south of France.
Caricatures in Middle Eastern media of both Christians and Jews have proliferated
at least throughout this century, long before the establishment of the state of Israel.
The issue is not, then, simply a question of insult, but of the political orchestration,
quite selective, of insult in a historical conjuncture in which Occidentalism can be
maximally exploited.9
The role of the inversion of ideology is crucial here, the notion that all world
problems emanate from the expansionist power and the culture of the West. This
“Occidentalism” (Buruma and Margalit 2004) is not new, of course, but its periodicity
9781444333237_4_013.indd 242
8/7/2010 7:09:43 PM
243
Universalism (Occidental)
Universalization I
Hybrid Cosmopolitans
National
Population
National
Elites
Migments
West
National
Population
Diasporiazation
Ethnic Minorities
Ethnic Minorities
Indigenous
Indigenous
Rooted ethnicities
Rest
Universalization II
National
Elites
Indigenization
Cosmopolitanization
Cosmopolitanism
Islamist world Sharia
Global Fundamentalism (Sharia)
Figure 13.4 Cosmology, class, and geopolitics at the end of hegemony
is interesting in global terms. While the West is totally occupied by the issue of terrorism, China is busy expanding at an enormous pace in former peripheral zones, such as
Africa and South America, where both markets and raw materials are a crucial goal,
just as in previous periods of Western imperialism. In this case the material process of
global reproduction extends beyond the logical wholes that are emergent in this
period which are organized around a former geopolitical scheme, including the
Middle East and Europe (and the West, more generally). The latter can be represented as shown in Figure 13.4.
This figure, which can be understood as an extension of Figure 13.3, depicts a
polarization between an emergent cosmopolitanism and indigenization of the downwardly mobile within the nation-state. The extension here is toward a more complete
geopolitical encompassing of potential global oppositions. As we have suggested
above, this opposition is based on the history of shifting imperial power as it exists in
the political imaginaries of the West and of the Middle East. This is not necessarily a
vindication of the views of Huntington, but it does support the ideological shift that
is implied in his analysis, even if the mechanisms involved are quite different.
Huntington, as we suggested above, has been attacked on numerous grounds, not
least his lack of consideration of the great variation within the Muslim world. In this
analysis, however, it is not a question of local variation but of potential mobilization,
and here the model does, I would argue, make sense. This implies that the only really
existing alter-globalization movement is Islamism, insofar as it proposes an alternative
model for the organization of society on a world scale. This schema is meant to delineate a field of identification and of intentionality that encompasses more limited
projects which may, of course, be partially contradictory with respect to the larger
field. On the issue of the support for terrorism, many Islamists and moderate Muslims
9781444333237_4_013.indd 243
8/7/2010 7:09:43 PM
244
Friedman
would clearly be critical at the same time as they might be understanding with respect
to the motives of terrorists based on a reaction to the imperial designs of the West.
While particular political positions within this field are variable, I would argue that
there is a common logic of interpretation. The exploration of this field is the subject
of a forthcoming study and cannot be elaborated on here. What is crucial here is the
suggestion of an encompassing logic that frames subjective experience and practice.
Conclusion
Holism is not a mere assumption. It is a research proposal or direction in which it is
assumed that there are structured orders of social reality that can be discovered and
understood, and that the world is not the mere result of interaction of individual
subjects. These orders are not overt but consist of the properties of social life that are
by and large inaccessible to those who participate in it. The wholes are in this approach
equivalent to the underlying logics of the social order. They are not immediately
observable, unlike their effects. They are the products of hypothetical thinking, which
is always asymptotic insofar as hypotheses are inadvertently limited and as such
falsifiable. The holistic program is nothing more than the search for adequate explanations. And the latter are hypotheses concerning the wholes that account for the
observable in terms of the relation between them and their parts.
This might seem drastic to many who have been critical of the modernist scientific
paradigm, but if this is admitted as a possibility, then there is no reason to deny, out
of hand, the possible existence of holistic accounts. Our analysis has linked cyclical
patterns of expansion, contraction, and global shifts in hegemony with large-scale
economic political and cultural crises and consequent sociocultural transformation.
The holism that we think is required to grasp these complex relations is opposed from
the start to current antiholistic trends in the field, and we suggest that the latter can
themselves be understood as an aspect of the collapse of modernism which is the
substrate within which holism was able to propagate as a scientific approach. Our
diagnosis of much of current postmodernist discourse, as well as much globalization
discourse, is that it is primarily an ideological effect of the reconfiguring of the global
system. This might be turned around, of course, and our own approach criticized for
merely being the expression of a previous era of modernism connected to the expansionist phase of Western hegemony. Here it is important to distinguish between the
nature of scientific argument, or any argument about the nature of reality, and the
social conditions within which such arguments emerge. I have suggested that postmodern and globalization discourses are ideological products of a particular positioning in a period of hegemonic decline in which there is an emergence of new elites. But
I have not suggested that the content of such ideologies comprises immediate products of such social conditions, only that certain conditions are conducive to their
development. The argument for the existence of global systems within which globalization is a particular historical process is a hypothesis that is open to falsification.
Globalization discourse itself is bereft of hypothetical statements. On the contrary, it
is a set of propositions about the nature of the world, expressed in descriptive terms.
In the absence of hypothetical language, such descriptions are easily conflated with
their interpretive assumptions, hence their ideological character. Global systemic
analysis surely contains ideological elements as well, in its propositions concerning
9781444333237_4_013.indd 244
8/7/2010 7:09:43 PM
245
cyclicity, hegemony, hybridity, and the like. But the latter are hypothetical proposals
rather than mere descriptions. Thus the approach as such is not ideological. It is
surely the case that cyclical models of historical process might be related to particular
social conditions. Ibn Khaldun, Spengler, and Toynbee are all representatives of eras
of crisis, of hegemonic decline and shifting power relations, even if their explanations
for such processes vary in the extreme. What is important here is an attitude, a necessary distance to propositions about the world and an acceptance of a principle of
falsification.
Is this Popper? Is it modernism? Is it therefore historically limited and perhaps
outdated in today’s world? It is outdated if we assume that whatever dominant
ideology prevails, it is an expression of historically progressive change. Modernist
science is, in our approach, an extended metaphor on simple trial-and-error learning. It is not just the socially determined set of representations particular to modernism or even modernity. We need to deal with the broad repertoire of modes of
thinking which are quite universal but which are historically selected. This, as
Lévi-Strauss argued, is that which relegates particular kinds of thought to dominant
positions in particular social orders. The so-called savage mind is universal in this
approach. It is not a specific mentality of the “primitive.” Lévi-Strauss argued as
we interpret it here, that in the circumstances of modernist culture, the scientific –
that is, the trial-and-error approach to reality – becomes a dominant paradigm,
while the paradigm that is dominant in kinship-based societies becomes the marginal, if no less important, stuff of dreams, myths, and their analysis. We suggest
that political orders are also part of this mythical universe which is why we can
analyze them as we have done. The cosmological unconscious and its imaginary
logic provide the form within which power structures are configured and reconfigured in this modern world. Thus holism, even in the emic sense, is alive and well
in a modern social order that would deny its existence.
Notes
1 It is global in the sense that the larger reproductive relation, i.e. alliance, is necessary to
account for the formation and maintenance of the local kin group.
2 The original is as follows: “J’ai le sang ex-patrié.… Je suis américain, de passeport et de
nationalité mais ma famille et celle de ma femme aussi, ont un grand nombre de ramifications dans beaucoup de pays, ce qui fait qu’on a toujours eu un pied aux États-Unis un pied
à l’étranger.”
3 The original: “Mon père était un peu vagabond, et on avait ca dans les veines. Mes frères,
c’est pareil: j’ai un frère en Autriche, un en Finlande, une soeur en Espagne. Mon père se
déplacait beaucoup, et j’ai dû apprendre ca.”
4 The original reads, “La curiosité, l’ouverture, la tolérance sont des termes souvent employés
pour désigner ces qualités.”
5 The original: “Alors le terrien, c’est quelqu’un qui a un espace limité. Son activité se concentre sur la terre qu’il possède. Si l’autre va sur sa terre, il ne l’acceptera pas. Il est attaché
à sa famille, à ses enfants, qu’il veut garder chez lui, parce que sa famille cultive sa terre.…
“Je crois que la classe populaire est plus attachée à ses origines. Les anglais en France sont
plutôt des gens des classes moyennes. En Angleterre les classes populaires sont plus nationalistes que les autres, moins ouvertes.”
6 The original reads, “Le constat de l’épuisement du modèle social-démocrate a transformé
les militants de la révolution, puis de la réforme, en militants du libéralisme culturel.”
9781444333237_4_013.indd 245
8/7/2010 7:09:43 PM
246
Friedman
7 The original is as follows: “Aux ouvriers elles ont substitué les immigrés et ont reporté sur
ceux-ci le double sentiment de crainte et de compassion qu’inspire généralement le prolétaire. Or l’immigré n’est pas seulement victime de l’exclusion sociale, mais aussi de
l’exclusion ethnique, autrement dit du racisme.”
8 The original reads, “Cette idée de déclassement du monde musulman est centrale. Dans
l’imaginaire musulman, le fait d’appartenir à un ensemble civilisationnel puissant qui rivalisa,
et parfois dépassa l’Occident est essential. Sans ce fait historique, remontant à plus de dix
siècles, ce sentiment de déclassement n’aurait probablement pas cette force auratique.”
9 See the excellent yet rarely cited book by Favret-Saada (2007) on the caricature crisis.
References
Appadurai, A. (1988). Putting Hierarchy in Its Place. Cultural Anthropology 3(1): 36–49.
Appadurai, A. (1990). Disjuncture and Difference in the Global Cultural Economy. Theory,
Culture and Society 7: 295–310.
Appadurai, A. (2000). Spectral Housing and Urban Cleansing: Notes on Millennial Mumbai.
Public Culture 12(3): 627–51.
Bislev, S., and D. Salskov-Iversen (2000). Globalization and Discursive Regulation: New Public
Management. Paper presented at 16th Nordiska Företagsekonomiska Ämneskonferensen,
Uppsala 16–18.
Buruma, I., and A. Margalit (2004). Occidentalism: A Short History of Anti-Westernism.
London: Atlantic.
Dirlik, A. (1997). The Postcolonial Aura: Third World Criticism in the Age of Global Capitalism.
Boulder, CO: Westview.
Dumont, L. (1967). Homo Hierarchicus: Essai sur le Système des Castes. Paris: Gallimard.
Dumont, L. (1976). Homo Aequalis. Paris: Gallimard.
Favret-Saada, J. (2007). Comment Produire une Crise Mondiale: Texte Imprimé Avec Douze
Petits Dessins. Paris: Les Prairies ordinaires.
Friedman, J. (1994). Cultural Identity and Global Process. London: Sage.
Gitlin, T. (1995). The Twilight of Common Dreams: Why America is Wracked by Culture Wars.
New York: Metropolitan.
Goad, J. (1997). The Redneck Manifesto. New York: Simon & Schuster.
Government of Sweden (1997–8). Sverige, framtiden och mångfalden – från invandrarpolitik
till integrationspolitik [Sweden, the Future and Diversity: From a Politics of Immigration
to a Politics of Integration]. Regeringens Proposition (Swedish Government Bill).
Stockholm: Government of Sweden.
Hardt, M., and A. Negri (2000). Empire. Cambridge, MA: Harvard University Press.
Huntington, S. P. (1996). The Clash of Civilizations and the Remaking of World Order.
New York: Simon & Schuster.
Jacoby, R. (1994). Dogmatic Wisdom: How the Culture Wars Divert Education and Distract
America. New York: Doubleday.
Jacoby, R. (1999). The End of Utopia: Politics and Culture in an Age of Apathy. New York:
Basic Books.
Jones, C. A. (1987). International Business in the Nineteenth Century: The Rise and Fall of a
Cosmopolitan Bourgeoisie. New York: New York University Press.
Julliard, J. (1997). La Faute des Élites. Paris: Gallimard.
Laïdi, Z. (2001). Rebonds: Il est Urgent de s’Interroger sur les Prochains Conflits de Civilisation
et Guerres. La Planète Cannibale. Libération, October 12, p. 14.
Lévi-Strauss. C. (1962). La Pensée Sauvage. Paris: Plon.
Marx, K., and F. Engels (1967). Capital: A Critique of Political Economy, vol. 3. New York:
International.
9781444333237_4_013.indd 246
8/7/2010 7:09:43 PM
247
McChesney, R. W. (1999). The New Global Media: It’s a Small World of Big Conglomerates.
The Nation, November 29, p. 4.
Nairn, T. (1977). The Break-Up of Britain: Crisis and Neo-nationalism. London: NLB.
Rabinow, P., G. E. Marcus, and Tobias Rees (2008). Designs for an Anthropology of the
Contemporary. Durham: Duke University Press.
Rosenstock Huessy, E. (1987). Die Europäischen Revolutionen und der Charakter der Nationen
[1961] (foreword Karl Deutsch and bio. note Rudolph Hermeier). Moers, Germany:
Brendow Verlag.
Sahlins, M. (1999). Two or Three Things I Know About Culture. Journal of the Royal
Anthropological Institute 5(3): 399–421.
Van der Pijl, K. (1998). Transnational Classes and International Relations. London:
Routledge.
Wagner, A. C. (1999). Les Nouvelles Élites de la Mondialisation: Une Immigration Dorée en
France. Paris: Presses Universitaires de France.
9781444333237_4_013.indd 247
8/7/2010 7:09:43 PM
9781444333237_4_013.indd 248
8/7/2010 7:09:43 PM
Part 4
Beyond Social Wholes?
Holistic Practice: Cosmology,
History, and the Continuity of Life
p04.indd 249
8/7/2010 7:21:42 PM
p04.indd 250
8/7/2010 7:21:42 PM
14
In Thomas Barfield’s Dictionary of Anthropology, the entry for “holism” is laconically short. It reads, “Holism. See Functionalism” (1997: 240). Turning to functionalism, one learns that functionalism – apart from being an “ethnographic
methodology distinctive of anthropology,” “a school of sociology,” and “a philosophy of the social sciences” – is a “historical school of anthropology also known as
‘British social anthropology’ ” (1997: 209). Holism, functionalism, and British
social anthropology, one is led to understand, are synonyms. The reality is more
complicated, of course.
The Holisms of Social Anthropology
To trace the links and disjunctures between functionalism, social anthropology, and
(its various kinds of) holism, we need to begin in a locus classicus, and here Barfield
rightly points to the influence of Durkheim. The early functionalism of British social
anthropology was informed, so we learn from Barfield, “by Durkheimian sociological
theory, particularly the notion that the social was a level of organization sui generis
that could not in any simple way be reduced to the motives and intentions of individuals” (1997: 210). While this is certainly accurate, it is also the case that this heritage
from French functionalism was much stronger in some parts of British social anthropology than in others. While the functionalism of Malinowski was mainly focused on
the functions that institutions performed in the fulfillment of biological needs, the
structural-functionalism of Radcliffe-Brown was – as is well known – much closer to
that of Durkheim. Radcliffe-Brown was interested in the function that customs had in
maintaining the integrity of the social system conceived as an organism (Layton
1997: 34). Although British functionalism was thus based on holism, it would appear
that there was more than one holism at stake in British anthropology. The holism of
Malinowski was tied more to the method of fieldwork – the functionalism identified
by Barfield as “an ethnographic methodology” – a methodology of which Malinowski
became the iconic beacon (see Chapter 1). The holism of Radcliffe-Brown and the
c14.indd 251
8/7/2010 7:18:15 PM
252
Bubandt and Otto
later generation of structural-functionalists, on the other hand, was more closely
linked to a theory of social wholes than to fieldwork as a method.
What may be called “the Malinowskian brand of holism” associated with fieldwork, whose ambition it was, as Marcus and Fischer put it, “to provide a full picture
of a closely observed way of life” (1999: 22), was itself never really a theory, and
Malinowski’s explicit theories of biological needs and cultural responses never
became influential – neither within nor outside British anthropology (Barnard
2000: 68). Paradoxically, one could say that this was part of the strength of the
holism that Malinowski gave to anthropology, namely, the holism embedded in conventional ideals of participant observation and fieldwork – a holism that became one
of the cornerstones of the discipline but that rarely was made explicit (see Chapter
1), and that only began to be challenged in earnest in the 1980s and 1990s (see also
Chapter 2).
If Malinowski’s holism was methodological and implicit, the holism of RadcliffeBrown was explicit and framed by metaphors borrowed from biology. For RadcliffeBrown – following Spencer and Durkheim – society was best conceived of as an
organism. Like organisms, societies were held together by a functional unity “in
which all parts of the social system work together with a sufficient degree of harmony or internal consistency, that is, without producing persistent conflicts which
can neither be resolved nor regulated” (Radcliffe-Brown 1935: 397). The legitimacy
of biological metaphors, which Radcliffe-Brown insisted should not be taken literally, was nevertheless informed by an epistemic faith in the knowledge project of the
natural sciences. Radcliffe-Brown’s idea of functional unity and the specific kind of
holism it entailed, however, also had another and partly related source. This source
was Jan Christiaan Smuts, who in 1926 coined the word “holism” in his book Holism
and Evolution (1926; see also Chapter 1). For Smuts, holism was “a creative tendency” in natural evolution toward “the development of ever more complex and
significant wholes” (1926: 105). These wholes – animals and plants – were empirically observable in nature but hinted at a larger process of “whole-making,” which
Smuts held to be the driving principle of evolution, a principle that could also be
found in the way that human Society, human Personality, and the human Mind (all
capitalized) were organized. The holism of Smuts was a rebellion against scientific
reductionism as well as a “spiritualist” call for a holistic science capable of capturing
the “organic” structure that Smuts asserted characterized the entire “Holistic universe” (1926: 345). The holism of Smuts may in places have been far too mystical
for Radcliffe-Brown, but it nevertheless fitted hand in glove with his views about the
“organic” structure of society. Smuts and Radcliffe-Brown were more than passing
acquaintances. Jan Smuts was prime minister of the Union of South Africa between
1919 and 1924, and it was he who in his capacity as an honorary fellow of Christ’s
College in Cape Town had arranged for Radcliffe-Brown to be appointed as the first
professor of anthropology in South Africa in 1920 at the behest of Haddon (Leach
1984: 7). Their relationship seems to have been close, and, indeed, several family
members of Smuts would later pursue careers in anthropology with Radcliffe-Brown
(Kuper 1983: 46). Although most German holists at the time studiously ignored the
work on holism by Smuts (Harrington 1996: 269n94), it clearly belonged to an
early-twentieth-century interest in holism that was to engulf much of the European
academy and its colonial outliers. The patronage of Radcliffe-Brown by Smuts and
their common interest in organic holism suggest both that this holism did not rest
c14.indd 252
8/7/2010 7:18:16 PM
253
entirely comfortably within the limits of the episteme of natural science and that the
colonial outliers may at times have been more central to the formation of holisms at
the center than is usually imagined.
Critique of Organic Holism
Whether this genealogy is correct or not, one thing is certain. Compared to the
implicit acceptance of the Malinowskian holism associated with the mythology of
fieldwork, the “organic” holism of Radcliffe-Brown led a much more contentious life
in British anthropology. Its controversy was no doubt an effect of its initial theoretical
hegemony not only within Britain but also in other parts of the Commonwealth such
as Australia, India, and South Africa (Barnard 2000: 77). When Evans-Pritchard, who
was otherwise regarded as a loyal structural-functionalist, assumed the position as
chair of anthropology at Oxford University after Radcliffe-Brown in 1946, he came
out strongly against the argument of Radcliffe-Brown (but also of Malinowski) that
anthropology was a branch of the natural sciences. Rather, anthropology should be
seen, so Evans-Pritchard insisted, as a branch of the humanities in general and history
in particular. If anything, Evans-Pritchard argued, the synchronic approach and onesided focus on structure over topics such as belief and historical change prevented
functionalism from providing a full picture of a closely observed life. “Organic holism,”
it began to become apparent, prevented the kind of holism that fieldwork was designed
to give access to. This criticism was echoed in a number of ways in other parts of
British anthropology. Edmund Leach and Mary Douglas were inspired by French
structuralism, whose founding father, Claude Lévi-Strauss, was scathing about the
notion of structure that Radcliffe-Brown employed. For Radcliffe-Brown, social
structure was an empirical fact. Like the visible natural wholes of Smuts, social wholes
were for Radcliffe-Brown the total sum of social relationships within society that were
as observable to the naked eye as the structures of a seashell (Kuper 1983: 53). For
Lévi-Strauss and his followers in Britain, however, structure was an analytical model,
the way anthropologists (and humans generally) made sense of the world. Whereas
the legacy of Durkheim is clear in both Radcliffe-Brown and Lévi-Strauss, they spun
from it, as Adam Kuper has noted, two divergent threads. French structuralism was
developed through an engagement with Marcel Mauss (and structural linguistics),
while Radcliffe-Brown remained wedded to the organistic functionalism of Durkheim
(Kuper 1983: 52). This also made for two very different kinds of holism. The latter
was empiricist and positivistic; the former was rationalist and universalistic (see also
Chapter 10).
Other critics of Radcliffe-Brown were even harsher. The Manchester School around
Max Gluckman and Victor Turner as well as the transactionalism of Fredrik Barth
rebelled against the felt strictures of structural-functionalism and its socio-organic
holism, because they prevented careful analysis of power and conflict, of the contribution of individual actors to social change, and of the interplay between ecological
conditions and social structure. The Manchester School reoriented the study of society
away from a focus on functional coherence toward ritual, emphasizing (particularly in
Turner’s case) the performative and conflictual basis of social action and reproduction,
while attempting to combine (in Gluckman’s and Peter Worsley’s case) the functional
c14.indd 253
8/7/2010 7:18:16 PM
254
Bubandt and Otto
holism of the Durkheimian tradition with the conflictual holism of Marxist analysis
(for an analysis of Marx’s holism, see Shiell 1987; see also Chapter 13).
Experiential Holism
The performative turn in particular was to be important for attempts to reinvent
holism in newer British anthropology. The introductions to anthropology by Wendy
James (2003) and Kirsten Hastrup (1995) may be said to be two recent examples
of such attempts to rethink the heritage of British anthropological holism through
theories of performance and experience. James is concerned to restate the importance of the Durkheimian heritage of holism when she defines the object for a new
anthropology to be “social form.” James phrases this focus on social form – very
much in the tradition of Durkheim and Mauss – within an explicit critique of (neo)
liberal methodological individualism (i.e., the theoretical conviction that one should
start analytically from the intentions, needs, or actions of the individual) (James
2003: 17). Human actions, speech, and consciousness are for James inherently performative, and a comparative anthropology should therefore be grounded in an
analysis of the performative dimensions of sociality that are central to the production of social forms (for a different use of the concept of “social form,” see also
Chapter 12).
Hastrup has a similar approach to performance that also grows out of an engagement with British (and American) anthropology, if one inspired more by phenomenology and practice theory. And like James, Hastrup feels that holism “must be
reinvented but not necessarily abandoned” (Hastrup and Hervik 1994: 2). But where
James maintains a more loyal attitude to the Durkheimian tradition in British anthropology, Hastrup’s holism grows out of a phenomenological attention to experience,
and therefore begins with the experience of the individual rather than with social
forms (while still insisting on the social basis of this experience):
In approaching the world through the individual, rather than the other way around, one
needs not endorse narcissism on the personal level or cultural fragmentation at the societal; rather such an approach allows us to see how, in a comparative perspective, agency
is shaped partly by moral horizons of different scales; there would be no “selves” without
moral spaces of orientation.… In short, it allows for a new kind of holism, one that is
neither deterministic nor simply designative, but rather an expression of those life-worlds
of people that are not otherwise manifest. (Hastrup 2002: 38)
Human life worlds, so Hastrup suggests, are by necessity holistic, and while this
sense of a whole is an illusion, it is also “an efficient framework for social action”
(Hastrup 2007: 199). Because human experience and human action are real only
because they are imagined as a whole, anthropology itself needs to be holistic in order
to capture these worlds faithfully. The break with Radcliffe-Brown and the lineage to
Lévi-Strauss and Bourdieu are palpable. Holism is not an empirical characteristic of
social organization. Instead, it is an illusory but nevertheless real aspect of the way
humans make sense in the world. Anthropological holism can be retooled to trace
phenomenologically the imaginary wholes that make individual action possible (see
Chapter 10 for a discussion of the universality of such an “experiential holism”).
c14.indd 254
8/7/2010 7:18:16 PM
255
Methodological Individualism and Holism
Others within the British tradition, however, are more skeptical about lacing holism
into a theory of action. Rapport and Overing in their encyclopedia of key concepts
in social and cultural anthropology seem to suggest that holism and “methodological individualism” are opposed and cannot be combined. Advocating the latter
approach, they suggest there has been
a venerable tradition of regarding a “methodological holism” or collectivism of this kind
not merely as a virtue of social-scientific analysis but as a sine qua non of its verity.
(Rapport and Overing 2007: 250)
The holism that Rapport and Overing diagnose within anthropology appears to
be more insidious, and much more beyond redemption than holism is for Hastrup
and James. Rapport and Overing thus detect an “impersonal holism” in much
anthropology from Geertz, Peacock, and Fortes to Giddens – a holism they accuse
with reference to Anthony Cohen of being “not merely dull, unambitious, redundant and intellectually barren, but arrogant and insensitive, discreditable”
(2007: 251).
Clearly, holism is an emotional topic for those educated within British anthropology. It is arguably so, because a debate about holism opens up the question of how
one sees the relationship between lived experience and social patterns, between phenomenology and power, between agency and structure. This is a question that
became increasingly a battle over the soul of anthropology with the practice turn in
the 1980s and with the revival of interest in both phenomenology and existentialism
(Ortner 1984). It would appear that part of the controversy within this debate
relates to different perspectives on holism. Whether one thinks that anthropology
and holism have a future together depends to a large extent on what one takes
holism to be and what intellectual genealogy one ascribes to it. The result is two
very different futures for holism. Is it possible to move beyond social wholes but
maintain some form of holism? Hastrup and James would say yes; Rapport and
Overing appear to say no.
Recently, British anthropology has seen two prominent attempts to retain the notion
of holism while breaking with some of the basic tenets of the Durkheimian heritage
within social anthropology. They come from Cambridge and Oxford, respectively.
Marilyn Strathern, professor at Cambridge University,1 has taken up the issue of holism
on several occasions (Strathern 1987, 1988, 1992b, 2004a, 2004b). Arguing against
the common notion that holism and individualism are opposed – as expressed for
instance by Rapport and Overing (2007) as well as by anthropologists working in the
tradition of Louis Dumont (Chapter 10; see also Coppet 1992) – Strathern suggests
that what prevents a contemporary holistic anthropology is not the notion of the
“individual” but the notion of “society”:
The irony is that what clouds the anthropologists’ holistic enterprise in the late twentieth
century is no longer individualism. The “death of the individual” has seen to that. Rather,
the problem is the Western dismantling of the very category that once carried the concept of a holistic entity, that is[,] “society.” Society was a vehicle for a kind of Western
holism, a totalizing concept through which modern people could think the holisms of
others. (Strathern 1992b: 77)
c14.indd 255
8/7/2010 7:18:16 PM
256
Bubandt and Otto
Strathern notices that the conception of society as a whole had come under pressure
from postmodern pluralism with its emphasis on fragmentation, hybridization, and
discontinuity. Her own analysis of modern English kinship (Strathern 1992a) demonstrates that social-functional holism is not an unproblematic notion, even in countries
that have generated and embraced this perspective. She develops this point by unpacking the problems that mid-twentieth-century British anthropology had in dealing
with English kinship. From the perspective of society, Strathern argues, the individual
person was seen as incomplete. It therefore had to be completed by its integration
into the whole through socialization and convention. From the perspective of kinship
and the domestic sphere, however, the person was already a biological and psychological whole before entering society. This entailed the existence of two partly incompatible holistic perspectives. They could not be matched or exchanged, because they
only partially overlapped: the individual as part of society and the individual as part of
nature. Both perspectives were totalizing, but neither could fully encompass the other,
thus positing a contradiction at the heart of modern English holistic perceptions of
social relations.
As an alternative to the failed efforts of using Western holistic concepts such as
society (and individual and nature) to analyze other people’s life worlds, Strathern
proposes that we try to apprehend the ways in which other people employ holistic
devices themselves to understand and construct their own life worlds. The productivity of this innovative strategy has been demonstrated by Strathern’s stimulating analyses of Melanesian societies. In contrast to Western societies, in Melanesia Strathern
does not find a holistic perspective from society on the person (or vice versa). Rather,
there is only a holistic perspective from the center (ego) on the relationships that
constitute the center. Strathern connects this insight into Melanesian holisms with her
interpretation of Melanesian personhood, which, she argues, is composed of parts
that derive from other persons in ongoing exchange relationships. Such a vision of
“dividual” or “composite” personhood allows a notion of sociality (not society) that
does not make a principal difference between social organization and the constitution
of the person. External relations have the same effect as internal ones; they are in
principle of the same nature. Therefore, the perspectives from different centers (persons or egos) are exchangeable, and this is what happens in the mutual constitution
of personhood and sociality in Melanesian exchange relationships.
The holistic practices in Melanesia and the West are thus qualitatively different and
can explain well-known ethnographic contrasts that have long puzzled anthropologists. For example, when a Melanesian person dies, the surviving relatives engage in
extensive ceremonies in which the relations centered in the dead person are severed
and rearranged. This undoing and reconstituting of relationships are necessary because
the dead person is no longer there to embody these relationships, and thus a hole in
the fabric of sociality has opened up and needs to be mended. On the other hand,
seen from an English perspective, society will endure with its formal positions and
institutions, also when an individual is taken away (Strathern 1992b). Strathern’s
approach to holism avoids the pitfalls of the classic structural-functionalist holism that
uses Western holistic concepts like “society” in a totalizing way by applying them to
other life worlds. Instead of using preconceived holistic frames, Strathern experiments
with the reconstruction of indigenous holisms to let them provide alternative perspectives on human sociality and personhood. As Tsing (Chapter 4) has argued, Strathern
thus creates an excess of contexts that avoids the stabilization and totalization of holistic frames that have haunted British social anthropology.
c14.indd 256
8/7/2010 7:18:16 PM
257
Two Chapters
The contribution to this section by Pedersen and Willerslev (Chapter 15) takes up the
suggestion by Strathern to find local forms of holism to unsettle our own preconceptions. Focusing on a classical topic of structural-functionalism, namely, that of joking
relationships, Pedersen and Willerslev show that among Darhad pastoralists and
Yukaghir hunters of Northern Asia, joking is an activity that takes place between humans
and spirits as much as it does between humans. In line with Radcliffe-Brown’s analysis
of joking relationships, they suggest that joking is a way of organizing relationships and
apportioning social respect and avoidance. But Pedersen and Willerslev suggest that
Radcliffe-Brown’s central insight was not just that joking relationships make for ordered
social ties, but also that they keep order at bay. Both the Darhad and the Yukaghir
inhabit what Pedersen and Willerslev call a “half-empty” cosmos, a cosmos that by its
very nature as well as because of Soviet political history cannot be fully brought into
discourse. The cosmos is therefore in need of “constant maintenance work.” What
needs to be maintained is precisely the half-empty (as opposed to full) nature of the
cosmos. The social maintenance of the relations to the spirits in other words produces
simultaneous order and renewal of disorder and distance. Joking about and with the
spirits performs this work of keeping humans related to the spirits while also keeping
them apart. In this practice, Pedersen and Willerslev detect amongst both groups a kind
of proportional holism, a form of ontological labor to ensure that the cosmos takes the
right shape. The “right shape” in this context entails maintaining relations to the spirits
that are neither too close nor too impersonal and distant. Joking is a holistic practice,
so Pedersen and Willerslev suggest, because it is concerned with the organization of the
world as a whole, but this world has multiple orders – the order of humans and that of
spirits – and joking allows people to walk the fragile balance between these orders by
keeping both in suspension. This ethnographic case not only provides an insight into an
animist ontology (see also Chapter 12) but also suggests that the proportional holism
practiced by Darhad and Yukaghir people when they joke is a way to avoid the threatening collapse of their multiple wholes into one whole. This may thus be indicative of a
more general way that holistic practices seek to solve the problem of cosmological
wholes: to retain and hold in balanced tension multiple wholes that are only partially
compatible. The chapter by Pedersen and Willerslev suggests, as do the chapters by
Holbraad and Tsing, that the multiple and competing wholes that people construct for
themselves in North Asia, in Cuban gambling sites, or in scientific labs in Japan and the
United States, respectively, are fragile and constantly in tension. Anthropological holism
should seek to highlight these tensions and the holistic practices that sustain them.
Chapter 16 by Hirsch and Moretti also takes up the Strathernian analytical strategy of
opposing various kinds of holism as a theoretical experiment in contrasts. Hirsch and
Moretti do so by contrasting conventional, Western notions of history evidenced in government and mission narratives from colonial Papua New Guinea as well as in more
recent scholarly histories of Melanesia and beyond with the visions of the past among
Melanesian peoples like the Fuyuge and Hamtai-Anga of Papua New Guinea. “Western”
ideas of the past are dominated, so Hirsch and Moretti argue, by a conception of history
as a coherent and meaningful unity of individuals, events, and epochs where the past is a
“single determinate realm of unchanging actuality” (Mink 1987 [1978]: 194) that
exists universally. Following Mink (1987 [1978]), Jay (1986), and others, they call
c14.indd 257
8/7/2010 7:18:16 PM
258
Bubandt and Otto
this presupposition about the past “Universal History” and suggest that, while not
unchallenged, it informs a dominant way of constructing the past in the West. “Universal
History,” they suggest, is a whole that entails a particular conception of persons and
events. This conception has an additive quality – imagined as a summation of individual
perspectives and unique events – that accumulates into one totalizing narrative. In contrast, among the Fuyuge, the Hamtai-Anga, and other groups in Melanesia, the past is
seen as an original source of unity out of which persons and performances are recurrently
differentiated. The individuality of persons or the specificity of a performance, therefore,
is in a Melanesian understanding not an essence but an outcome. This “Universal Past”
is associated, so Hirsch and Moretti argue, with the “dividual” notion of the person
described by Strathern, and entails a fractal rather than a totalizing kind of holism.
This contrast in how the past is understood and framed within an ontology of
people and events presents for Hirsch and Moretti an anthropological challenge. For
it means that a conventional historical account of the colonial past in Melanesia is
never neutral, since it will entail a kind of totalizing holism that erases local aspirations, claims, and perspectives on persons and the past. Hirsch and Moretti proceed
therefore to allow Melanesian narratives of the past – as they are recounted in local
ancestral tales and myths – to enter into dialogue with historical accounts that are
available from the colonial archives. They suggest that both Western and Melanesian
narratives of the past are holistic approaches to time. These holisms, however, encompass very different ontological ideas of being and time, and they come with different
political and moral implications. Disturbingly for an historical anthropology is the
fact that the two types of narratives also do not carry the same political weight in the
contemporary world. Attending to the Melanesian versions of the past might therefore help us not only unsettle the taken-for-granted nature of our own relation to the
past and the particular kind of holism upon which it is founded, but also highlight
how political power clings to some forms of holism more than to others.
Strathern thus marks one recent attempt to rethink holism, an attempt that has
clearly inspired the chapters by Pedersen and Willerslev and by Hirsch and Moretti.
Both chapters suggest the value of tracing holisms ethnographically and allowing
them to enter into theoretical thought experiments with the holisms we have come to
take for granted, whether those associated with human (dis)order or those associated
with the human past.
Holism, Transdisciplinarity, and One Last Chapter
The second recent attempt to rethink holism comes from Oxford. In an anthology from
2007, David Parkin and Stanley Ulijaszek have gathered a group of mainly Oxford-based
anthropologists to reflect on the future of a new “holistic anthropology” made possible by
recent advances in neuroscience and evolutionary biology (Parkin and Ulijaszek 2007). The
“sense of occasion” (Parkin 2007: 9) that the editors feel is created by this transdisciplinary
possibility is primarily used to explore questions of human and cultural evolution, and
the main “sparring” partners are found in biology, ecology, psychology, neuroscience, as
well as archaeology and material culture studies. In the British context, the volume constitutes a sharp departure from the prevailing distrust within social anthropology of
biological explanations and their assumed inherent reductionism. As such, the anthology is to be applauded as a novel contribution with a powerful alternative kind of holism,
c14.indd 258
8/7/2010 7:18:16 PM
259
a holism that has clear parallels to the ideals of the American four-field approach about
holism as the “cross-disciplinary” study of humankind (see Chapter 6). Much of the
power of this new holism comes from the cultural and scientific capital of the cognitive
and evolutionary sciences, and this not infrequently leads the contributions to more millenarian expectation than critical engagement. In his contribution to the volume, Harvey
Whitehouse for instance calls upon anthropologists to join “the revolution” (2007:
231). There are obviously many interesting areas of overlapping concerns in cognitive
and social theory, but Whitehouse expresses his belief in the prospect of providing cognitive explanations for issues of sociality and religion with a conviction and scientism that
are reminiscent of the first “holistic” revolution in anthropology that revolved around
Malinowski (see Chapter 1). History, however, has taught us to not necessarily take at
face value revolutions that assert a new scientific holism. Given its checkered history, a
more cautious and less normative approach to holism appears to us the prudent path. As
the contributions to the present volume suggest, anthropology is currently engaged in
a series of experiments in holism that are both more variegated and more guarded than
claims that anthropology is on the cusp of a paradigmatic neurobiological revolution
allow for. While Parkin and Ulijaszek’s volume is a welcome exploration of one of the
many kinds of contemporary holism (and therefore a fine complement to this volume
that might point to a future of holistic dissension), the faith in one particular kind of
holism demonstrated by most of its contributors is less ambivalent than history warrants.
The new knowledge claims deriving from neurosciences and the kind of holism it promises therefore need to be critically interrogated (like all other promises of holism) even
as we engage it collaboratively.
Advocating an “obviation” approach that breaks down the conventional boundaries between psychology, evolutionary biology, and anthropology rather than a complementary approach (Ingold 1998), Tim Ingold follows a different path from Parkin
and Ulijaszek to design a holistic approach. This kind of interdisciplinary holism, as
Ingold emphasizes, easily falls prey to a totalizing form of holism.2 In his contribution
to this section (Chapter 17), Tim Ingold therefore continues the path of critical reflection and inquiry that he has pursued for over two decades, namely how it is possible
to restore “life” to the center of anthropology and avoid falling into the trap of totalization. Anthropology, so Ingold asserts, should aim to represent life as it is lived
rather than adopt an approach that mistakes the products of life for life itself. In previous writings Ingold has called the latter approach “the building perspective,” a theoretical perspective that assumes that “worlds are made before they are lived” (Ingold
2000: 179). Ingold suggested that this notion has been the guiding premise of much
anthropology and informs the conventional anthropological concepts of “culture,”
“social organisation,” and “intentionality” as well as much of the cognitive sciences:
human action is the execution of a prior plan, and the task of science is to grasp and
give a full picture of this plan. Ingold opposed this to “the dwelling perspective,”
which insisted that the worlds that people make, “whether in the imagination or on
the ground, arise within the current of their involved activity, in the specific relational
context of their practical engagement with their surrounding” (Ingold 2000: 186).
Inspired by semiotic biologist Jakob von Uexküll – one of the leading Germanspeaking holistic scientists of the early twentieth century, and a man held in high regard
by Martin Heidegger (Harrington 1996: 34) – Ingold leveled a phenomenological
critique against sociocultural anthropology (as well as cognitive and evolutionary
biology), suggesting that anthropology should study the way people make the world
their own rather than the structures they erect as part of living3.
c14.indd 259
8/7/2010 7:18:16 PM
260
Bubandt and Otto
Ingold’s chapter to this volume continues to reflect on the difference between these
analytical perspectives by seeing them as ways of painting and ways of drawing, respectively. While the painter attempts to anticipate the totality of the complete picture when
using a brush to fill in details, the person who draws seeks to follow the continuation of
the line in its immediate spatial and temporal surroundings. While the first perspective
is holistic in the sense of conceiving a whole picture independently of or in anticipation
of action, the second follows the continuity of drawing lines as a holism of continuous
movement. According to Ingold this contrast corresponds with two paradigmatic positions in anthropology. On the one hand, there is the default position in much of social
and cultural anthropology in the twentieth century, which puts the totality of society
first and looks at the ways this shapes the person and the acting self. Ingold confesses to
the second position, which he finds “liberating,” as it tries to follow the essential continuity of the life process, in which the self is not shaped on the rebound of the social
whole, but undergoes continual generation along a line of growth. In this visionary
piece Ingold sketches the outline of an anthropology that follows the entanglements of
human life paths, that acknowledges human improvisation rather than systemic determination and totalization, and that produces studies with people rather than of them.
Notes
1 Of course, it should be kept in mind that many of Marilyn Strathern’s ideas on the topic,
including most of her key publications, originated before her appointment to Cambridge.
2 It is interesting here to note that Ingold also has a contribution in the volume edited by Parkin
and Ulijaszek (2007). This chapter is a precursor to the present chapter and appears also to
pose a counterpoint to the otherwise dominant gist of Parkin and Ulijaszek’s volume.
3 Compare Mark Mosko’s discussion of Sahlins’ and Munn’s contrasting perspectives on the
relation between human agency and cultural structures (Chapter 9: 173–4, 193), which
draws a different conclusion.
References
Barfield, Thomas (1997). The Dictionary of Anthropology. Oxford: Blackwell.
Barnard, Alan (2000). History and Theory in Anthropology. Cambridge: Cambridge University
Press.
Coppet, Daniel de (1992). Comparison, a Universal for Anthropology: From “Re-Presentation”
to the Comparison of Hierarchies of Values. In Conceptualizing Society, ed. A. Kuper.
London: Routledge, 59–74.
Hastrup, Kirsten (1995). A Passage to Anthropology: Between Experience and Theory. London:
Routledge.
Hastrup, Kirsten (2002). Anthropology’s Comparative Consciousness: The Case of Human
Rights. In Anthropology, by Comparison, ed. A. Gingrich. London: Routledge, 27–43.
Hastrup, Kirsten (2007). Performing the Worlds: Agency, Anticipation and Creativity. In Creativity
and Cultural Improvisation, ed. E. Hallam and T. Ingold. Oxford: Berg, 193–206.
Hastrup, Kirsten, and Peter Hervik (1994). Introduction. In Social Experience and
Anthropological Knowledge, ed. K. Hastrup and P. Hervik. London: Routledge, 1–12.
Ingold, Tim (1998). From Complementarity to Obviation: On Dissolving the Boundaries
Between Social and Biological Anthropology, Archaeology, and Psychology. Zeitschrift für
Ethnologie 123(1): 21–52.
c14.indd 260
8/7/2010 7:18:16 PM
261
Ingold, Tim (2000). Building, Dwelling, Living: How Animals and People Make Themselves
at Home in the World. In The Perception of the Environment: Essays in Livelihood, Dwelling
and Skill, ed. T. Ingold. London: Routledge, 172ü89.
James, Wendy (2003). The Ceremonial Animal: A New Portrait of Anthropology. Oxford:
Oxford University Press.
Jay, Martin, ed. (1986). Marxism and Totality: The Adventures of a Concept from Lukács to
Kuper, Adam (1983). Anthropology and Anthropologists: The Modern British School. London:
Routledge.
Layton, Robert (1997). An Introduction to Theory in Anthropology. Cambridge: Cambridge
University Press.
Leach, Edmund (1984). Glimpses of the Unmentionable in the History of British Anthropology.
Annual Review of Anthropology 13: 1–23.
Marcus, George E., and Michael M. J. Fischer (1999). Anthropology as Cultural Critique. An
Experimental Moment in the Human Sciences, 2nd ed. Chicago: University of Chicago
Press.
Mink, Louis (1987). Narrative Form as a Cognitive Instrument [1978]. In Historical
Understanding, ed. B. Fay, E. Golob, and R. Vann. Ithaca: Cornell University Press,
182–203.
Ortner, Sherry B. (1984). Theory in Anthropology Since the Sixties. Comparative Studies in
Society and History 26:126–66.
Parkin, David (2007). Introduction: Emergence and Convergence. In Holistic Anthropology:
Emergence and Convergence, ed. D. Parkin and S. Ulijaszek. New York, Berghahn, 1–20.
Radcliffe-Brown, A. R. (1935). On the Concept of Function in Social Science. American
Anthropologist 37(3): 394–402.
Rapport, Nigel, and Joanna Overing (2007). Social and Cultural Anthropology: The Key
Concepts. London: Routledge.
Shiell, Timothy (1987). On Marx’s Holism. History of Philosophy Quarterly 4: 235–46.
Smuts, J. C. (1926). Holism and Evolution. New York: MacMillan.
Strathern, Marilyn (1987). Out of Context: The Persuasive Fictions of Anthropology. Current
Anthropology 28(3): 251–81.
Strathern, Marilyn (1992a). After Nature: English Kinship in the Late Twentieth Century.
Strathern, Marilyn (1992b). Part and Wholes: Refiguring Relationships in a Post-Plural World.
In Conceptualizing Society, ed. A. Kuper. London: Routledge, 75–104.
Strathern, Marilyn (2004a). Partial Connections, updated ed. Walnut Creek, CA: Altamira.
Strathern, Marilyn (2004b). The Whole Person and Its Artifacts. Annual Review of Anthropology
33: 1–19.
Whitehouse, Harvey (2007). The Evolution and History of Religion. In Holistic Anthropology:
Emergence and Convergence, ed. D. Parkin and S. Ulijaszek. New York: Berghahn, 212–33.
c14.indd 261
8/7/2010 7:18:16 PM
15
Proportional Holism
Joking the Cosmos Into
the Right Shape in North Asia
Rane Willerslev and Morten Axel Pedersen*
Holism has had a pretty bad press in anthropology for some time now. To be sure,
every now and then, there are isolated attempts at reintroducing the concept, most
recently in an edited volume on Holistic Anthropology, which aims at integrating the
“human sciences” into a “new holism of evolving life forms” (Parkin and Ulijaszek
2007). Yet, such attempts tend to be curiously apologetic – certainly, one does not
detect much enthusiasm about the concept of holism in Parkin and Ulijaszek’s book,
where it is reduced to a purely heuristic tool with no theoretical purchase. As they put
it, as long it is merely taken to be “no more than an odd-job word, [holism] can enjoy
a short moment of interrogation and revelation before it is returned to the banality of
meaning too many things” (2007: xiii).
But perhaps it is possible to convey a more upbeat message about the future of holism
as an analytical concept. Based on fieldwork among two indigenous peoples in northern
Asia, this chapter introduces the concept of proportional holism, understood as the
ongoing apportioning of different assemblages of humans and nonhumans into the
right shape. Rather than being a heuristic tool anthropologists use to make sense of their
data, holism emerges as a particular social practice that can be identified in different settings around the world, and that can be made the object of ethnographic description,
analysis, and comparison like all other practices. To study holism anthropologically is to
explore the culturally specific work actors do for their cosmos to retain its proportions.
Studying holism as an empirical phenomenon requires sensitivity to the different
conceptualizations of wholes held by different people, as well as to the sometimes
surprising practices by which such wholes are perceived to hold together. As we show
in what follows, joking is one such holistic practice in the context of North Asian
indigenous societies. More precisely, we argue, joking relationships with the spirits of
dead ancestors and wild animals serve to actualize these entities in ritual and everyday
situations, while at the same time providing them with an irreducible virtual dimension. In that sense, the work of joking as a holistic practice is to balance between
otherwise singular wholes. Acts of rude joking during Mongolian shamanic rituals
and Siberian elk hunts keep necessary cosmological distances intact by holding the
parallel realities of visible human and nonhuman bodies and of invisible human and
nonhuman souls in simultaneous view.
* The authors contributed equally to this work.
9781444333237_4_015.indd 262
8/7/2010 7:10:34 PM
Proportional Holism
263
From Totalitarian Holism to Proportional Holism
If by holism one understands the perfect integration of discrete elements into a single
all- encompassing whole, which was how the concept was used at the heyday of
anthropological modernism (Parkin 2007: 2–3), then the term is clearly of limited
merit. The problem with this traditional anthropological concept of holism is twofold.1 Not only are the external relations between different wholes left unaccounted
for, but also there is a problem with the conceptualization of the internal relationships
within each whole between its parts. In the first case, the problem is one of too little
connection (“every culture is unique”); in the second, it is one of too much connection (“cultures are homogeneous”). Against this backdrop, we can only agree with Tim
Ingold that, for the concept of holism to have any theoretical purchase, it is necessary
“to dissociate [it] from a concern with wholes. Holism is one thing, totalisation quite
another, and [the] argument for holism [must be] … an argument against totalisation”
(2007: 209). But precisely how does one go about being holistic while also avoiding
the pitfalls of totalitarian holism, which leads one to “bolt together components of
being … that should never have been carved out in the first place” (2007: 209)?
One possible answer may be found in the writings of Roy Wagner and Marilyn
Strathern, who, along with other influential Melanesianists, have demonstrated the
“fractal” properties of different social entities (bodies, persons, clans, landscapes) in the
Papua New Guinean Highlands and elsewhere (Wagner 1991; Strathern 2004; Gell
1998; Mosko and Damon 2005). Now, like Ingold, it does not seem to be Strathern’s
intention to do away with the concept of holism as such. Rather, her more restricted
agenda is to criticize the common Euro-Western assumption that social forms exist as
single wholes (the singular one) or multiple parts (the plural many). Thus “Melanesian”
wholes are not totalities, in the sense of explicate orderings of discrete, autonomous
parts, which are integrated into singular and unified totalities. Rather, holism here
amounts to an implicate order in Bohm’s sense (1983), which is made visible through
different “aesthetic forms” which highlight certain relations at the expense of others
(Strathern 1988). Here, holism is not about the integration of finite units into a single
whole; rather, we may speak of self-scaling wholes, where analogous relationships are
replicated across different scales following a “holographic” logic (Wagner 1991). As
Wagner suggests in his seminal article on Daribi kinship, all relationships
are basically analogous because all incorporate the essence of human solicitude that we
call “relating.” Every particular “kind” of relationship exemplifies in some particular way,
and comprises a (“metonymic”) part of a potential whole, a totality of which the aggregate of all the kinds of relationship represents a homologue. Each particular kind of
relationship, since it incorporates the underlying context of relational solicitude, can be
seen as an (“metaphorical”) analogue of each kind of relationship. (1977: 624)
The concepts of holism entertained by our North Asian informants come much
closer to these fractal “Melanesian” wholes than to the “Euro-American” totalities
discussed. For the Darhad pastoralists of Northern Mongolia studied by Pedersen, as
much as for the Siberian Yukaghir hunters studied by Willerslev, the cosmos is not a
perfectly integrated whole consisting of smooth and well-oiled parts, which, in principle, add up.2 On the contrary, these indigenous ontologies are perforated by gaps,
9781444333237_4_015.indd 263
8/7/2010 7:10:34 PM
264
Willerslev and Pedersen
voids, and holes as much as they are glued together by connections, relations, and
wholes (Pedersen 2007b). This is also where these North Asian (w)holes differ from
their Melanesian holographic counterparts. Due, perhaps, to their half-empty nature
of the cosmos, the aesthetic forms of Darhad pastoralists and Yukaghir hunters lack
the self-evident proportionality widely associated with Melanesian aesthetic forms,
where questions of discontinuity and instability sometimes seem to be sidelined at the
expense of questions about continuity and stability (cf. Robbins 2007). Because of
their characteristic lack of ontological fullness, both Darhad and Yukaghir conceptualizations of the cosmos come across as more flexible and dynamic, while also comprising more friction and heterogeneity, than the neat cosmologies of, say, Wagner’s
Daribi informants. This might explain why these North Asian cosmologies are felt to
be in need of constant maintenance work. Not only are numerous rituals performed
for the world to remain in place, but also there is an unspoken rule that new techniques constantly need to be invented to keep humans and nonhumans sufficiently
detached from one another.
All this begs the question of whether the “Melanesian” concept of fractal holism
represents a sufficiently radical alternative to the traditional anthropological concept
of totalitarian holism. Alberto Corsín-Jiménez has recently observed that “the trouble
with the Melanesian model is that it is … assumed that people always relate in the
same fashion.… A full model of proportional sociality … is one that takes into account
the different ways in which people inflect and qualify their relationships. Proportional
sociality tells us the factors by which the stretching out of the social takes place”
(2007: 193–4). No matter whether this depiction of the “Melanesian model” is sufficiently nuanced or not (see Holbraad and Pedersen 2009), Corsín-Jiménez’s work
opens up for the cross-cultural study of practices of apportionment, which “unlike
relations, which only tell you how to disaggregate … tell you what to disaggregate
into” (2007: 187).
Our notion of proportional holism is strongly inspired by Corsín-Jiménez’s notion
of proportional sociality. Both terms refer to the work people need to do to ensure that
“things do not get out of proportion” (2007: 186) – a sort of labor, which involves
not only how relations are made as well as unmade (and how the same processes also
make and unmake people), but also the question of what relations (and what kinds of
people) thus come into being. In our two case studies below, we discuss in some
detail how people in Mongolia and Siberia go about doing, and maintaining, their
wholes. More precisely, we examine two concrete ethnographic situations – a Darhad
shamanic ritual and a Yukaghir bear hunt – where the right balance of proximity and
distance in the relationship between human and spirits was not given beforehand, but
only became established through acts of joking by which the cosmos was apportioned
in a certain way.
But why speak of joking as a distinct holistic practice in the first place? At first
glance, joking and holism seem to be unlikely bedfellows, at least if one subscribes to
the prevalent “romantic notion that jokes are inherently disordering and disorganising phenomena” (Seizer 1997: 62). However, while this link between joking and
disorder remains popular among anthropologists and philosophers (Douglas 1968;
Koestler 1964; Morreall 1983), it has always competed with an alternative interpretation
of joking, which posits it to be central for the reproduction of social order. The bestknown example of this is Radcliffe-Brown’s work on “joking relationships” (1952a,
1952b), which he saw as institutions designated for “establishing and maintaining
social equilibrium” (1952a: 108).
9781444333237_4_015.indd 264
8/7/2010 7:10:34 PM
Proportional Holism
265
As is well known, the concept of holism held by Radcliffe-Brown and many of his
structural-functionalist peers was hair-raisingly totalizing: no attempt was made to
identify potentially conflicting interests in the societies under study, let alone to
account for the possibility of social or cultural change, for the founding gesture of
functionalism was exactly to bracket off all questions of this kind. However, we believe
that the time is ripe for offering a new reading of Radcliffe-Brown’s theory of joking
relationships, one which retains the original focus on joking as a proportionally holistic practice, while doing away with its overly functionalist bias. After all, it is worth
remembering that Radcliffe-Brown emphasized that the “continual expression of
social disjunction” constitutes an “essential part” of life, which is precisely why he saw
joking to be a means by which people are able to keep an appropriate distance. As
Radcliffe-Brown elaborated (in characteristically terse prose),
In any fixed relations between the members of … groups the separateness of the groups
must be recognized. It is precisely this separateness which is not merely recognised but
emphasised when a joking relationship is established. The show of hostility, the perpetual
disrespect, is a continual expression of the social disjunction which is an essential part of
the whole structural situation, but over which, without destroying or even weakening it,
there is provided the social conjunction of friendliness and mutual aid. The theory that
is here put forward, therefore, is that [joking relationships] are modes of organising a
definite and stable system of social behaviour in which conjunctive and disjunctive components … are maintained and combined. (1952a: 95)
As we see it, Radcliffe-Brown’s key insight was not the well-known idea that joking
relationships maintain order, but the equally central proposition that such relations at
the same time keep too much order at bay. According to this interpretation of joking
relationships, then, the acts of “perpetual disrespect” between African kinsmen
described by Radcliffe-Brown and others are apportionings – holistic gestures by which
the world is kept in a dynamic state of measured disorder as different actors try to carve
out fragile balances of proximities and distances from an evolving totality of relations.
According to Corsín-Jiménez, “It is the confusion of not knowing how and which
orders of knowledge have to be made commensurable … that makes social life continuously re-dimension itself” (2007: 192). In his analysis, the goal of this ongoing
redimensioning is human well-being. As we intend to demonstrate in the two following case studies, similar questions of “social mathematics” – the size and scale of the
social, its ratios, and its proportions – also underwrite animist practices in Northern
Asia. Joking relationships between humans and spirits are holistic practices, which
Darhads and Yukaghirs perform for the cosmos to take its right shape. Crucially, the
worlds which people are working so hard to retain are not unified and stable wholes.
Instead, they are “proportional wholes”: fragile and fleeting assemblages of humans
and nonhumans in a state of measured disorder, where heterogeneous relations and
entities continually must be reapportioned into fragile balance, which hovers dangerously, or playfully, between too much integration (as in totalitarian holism) and too
much disintegration (as in postmodernist fragmentation). Peoples’ joking with the
spirits keeps the cosmos bifurcated into incongruous realms, and enables them to
partake simultaneously in these parallel domains of life. As such, these practices
reveal the need to combine worship with disrespect, sharing with theft – along with
other proportional holistic aesthetics of conjunction and disjunction – in the human
engagement with nonhuman others, so as to continuously establish the right degree
of coexistence and friction between disparate worlds.
9781444333237_4_015.indd 265
8/7/2010 7:10:34 PM
266
Humor in Darhad Mongolian Shamanic Ritual
Our first case study concerns the Darhads, a Mongolian-speaking people of pastoralists and hunters, whose heartland is the remote Shishged Depression of northwestern
Mongolia. Although they have a long Buddhist tradition (for more than two centuries
until the 1921 revolution, the Shishged valley and its inhabitants comprised the largest ecclesiastical estate in Mongolia’s Buddhist church), the Darhads are famous for
two things: their supposedly strong shamanic abilities and associated propensities for
cursing, and their supposedly superior sense of humor. Both stereotypes have been
widely internalized by the Darhads themselves, who today use them to identity and
differentiate between new types of persons and collectivities in the context of postsocialist “transition” (Pedersen in press).
There is a lot of joking in contemporary Darhad shamanism. Not only do certain
shamanic spirits (ongod) joke with the shaman and the audience during possession
rituals, but jokes are also sometimes used to invoke the spirits. In both cases, crude
sarcasm and racy humor intersect with esoteric intricacies of shamanic cosmology, and
a pristine space of holistic analysis is laid bare. For it is clear that joking does a certain
job, has certain effects, on the shamanic “whole.” Joking is imbued with occult efficacy, and understanding its nature takes us to the heart of Darhad shamanism. Thus
laughter is the very substance of certain spirits, which are as ephemeral yet persistent
as the flow of gossip in the community. These are the so-called gossip spirits, which,
according to the late female shaman Nadmid Udgan, “always act in a funny way” and
“just love revealing peoples’ most intimate secrets in curing ceremonies.” Indeed, the
intimate link between shamanism and humor in the contemporary Darhad context is
not only played out when certain spirits make fun of humans. As we shall see in the
following example, joking is also sometimes used by humans to invoke the spirits
(ongod) from their transcendental abode, the so-called skies (tenger).
Once, when Pedersen participated in a curing ritual (zasal) conducted by Nadmid
Udgan, humor and joking were called upon in what looked like a rather desperate
attempt to attract the spirits’ attention. The ritual had at this point already lasted for
more than an hour; yet, the first spirit was still to make its appearance. This was the
famous ongon known as the Father of Harmai, which is considered among the most
powerful (hünd, literally “heavy”) of the Darhad shamanic spirits. Patiently, the shaman had gone through her extensive repertoire of praises, prayers, and invocations,
but to no avail. Meanwhile, the attention of the audience had slowly drifted away;
some had even fallen asleep. But, suddenly, the ritual séance took a new turn as Nadmid
Udgan’s “interpreter” (helmerch) – an old woman from a neighboring household, who
was a distant relative of the shaman – grabbed a bottle of vodka from the altar in the
sacred north side of the yurt (hoimor), and, whilst overgenerously splashing its contents over her possessed master, bellowed, “Why are you not arriving? We are offering
you vodka to drink, cigarettes to smoke! What is the matter with you?”
The audience was laughing. A moment later, the Father of Harmai entered (orgoh)
Nadmid Udgan’s body, as witnessed by her animalistic grunting and eerie laughter
(both well-known signs of possession). It was time for the first client to be cured
(zasah), and the level of excitement went up one further notch. New logs were added
to the fire, slumbering relatives were awoken from their dreams, and everyone stared
intently at the drumming shaman in their midst. Communicated through the fragile,
9781444333237_4_015.indd 266
8/7/2010 7:10:34 PM
Proportional Holism
267
yet strangely insisting voice of Nadmid Udgan, the room now became filled up with
the occult message, which everyone in the audience had been waiting for – the actual
“words uttered” (heldeg üg) by the spirit itself. At this point in the curing ritual, it is
usually the interpreter’s responsibility to determine who in the audience a given spirit
is calling for, as well as what is needed in order to cure this person, for it is deemed
impossible for ordinary people to decipher the garbled and archaic words of the ongod.
On this particular occasion, however, everyone seemed to understand that this person
was the interpreter herself. Yet, she vehemently refused to comply with the Father of
Harmai’s request, and this breach of conduct set off a new sequence of funny events.
“What is this? What are you calling me for? I won’t do it!” the interpreter protested
to the ongon in a mock-serious voice. At last, after a series of abuses and sarcastic
remarks, she reluctantly accepted the ongon’s call by kneeling down in front of Nadmid
Udgan. Following a probing with the shaman’s drumstick, her body was found to
contain “poison” (horlol). Yet, she did not – as clients normally would – bow her head
in silent anticipation of the shaman’s healing blows with her whip (which are understood to transport people’s “bad stuff” to the drum). Instead she retorted angrily,
“Ouch! That hurts? Hey! – Stop it, enough of this!” which in turn made the shaman
hide behind the drum, as if shy. This sparked another roar of laughter. Then, an
abrupt change in Nadmid Udgan’s composure signaled the spirit’s departure (gargah), and the interpreter returned to the crowd.
What, then, was the point of this debasement of the shamanic spirit, if there was a
point at all? Notice, first of all, the striking difference between the present case from
Northern Asia and the comparable South Asian case, as discussed by Bruce Kapferer
in his celebrated study of demon exorcism in Sri Lanka (1991) According to Kapferer,
comedy plays a crucial role in the “disambiguation process” by which the “cosmic
order” is returned to “its essential unity” as an outcome of the exorcism ritual (1991:
177). Exorcism is all about bringing back the demons to their proper, inferior
position within the cosmic hierarchy by facilitating their re-encompassment by the
deities. This happens because the ritual use of comedy makes the demons appear
monstrous – and, therefore, hilarious – in the eyes of the audience and the patient
being healed. People are liberated from the yoke of the demons as their laughter dispels evil to an unreal state of “make-believe” (1991: 315–25).
Now, while this analysis seems to make a lot sense in the Singhalese context, it does
not work very well when it comes to North Asian shamanism. It would, indeed, be
rather strange if the joking in Darhad shamanic rituals were all about interpolating
doubt into an overarching ritual and performative framework of “let-believe” (pace
Kapferer 1991). In fact, it seems to be exactly the other way around. Thus the effect
of comedy in the Darhad shamanic ritual described above was to make the “demons”
(shamanic spirits) come to life, as opposed to ritually “killing them,” as is the case in
Sri Lanka (1991). From the point of view of the interpreter and the rest of the audience, adding a frame of “make-believe” to the ritual did something that the previous
frame of “let-believe” (prayers, incantations) had not been able to, namely, making
the spirits appear in the shaman’s body so people could interact with them. Far from
reducing the influence of the spirits, as in Sinhalese exorcism, joking in Darhad
shamanism thus seems to render the shamanic spirits more powerful. In that sense,
the Nadmid Udgan’s “interpreter” acted as more than a mere translator in the above
ritual.3 In fact, she emerges as a conjurer in her own right, a sort of substitute shaman,
who deployed humor and joking in order to stretch out or apportion the shamanic
cosmos in a certain way.
9781444333237_4_015.indd 267
8/7/2010 7:10:34 PM
268
Joking as Spiritual Apportioning
Our point is not that the Darhad shamanic spirits are somehow more real (let alone less
real) until they are joked with. Rather, we contend, the reality of spirits is the premise
(as opposed to the outcome) of the joking relationships at hand. The problem, then, is
not whether the spirits exist, but how they exist. This is supported by the fact that different spirits are not equally “ready-to-hand” (Willerslev 2004b), just as different kinds
of persons have different mastery of them. In fact, Mongolian shamans and laypeople
seem to entertain quite different ideas of what an ongon is (Humphrey 1996). For the
shamans, the spirits exist in an ontological continuum: they are always potentially
present inasmuch as they may be accessed through prayers at any given time. Conversely,
the laymen’s relationship with the spirits is one of ontological rupture: either the spirits
are here in the form of an invasive possession of peoples’ bodies (biye) or souls (süns),
or they are not here. This difference showed in the above ritual, which, from the point
of view of the audience, began with a boring sequence of invocations, during which the
shamanic spirits were not perceived to be present (they were “up there” in the skies).
This was followed by a much more exciting period of possession, when the spirit was
understood to be “down here” (insofar as he had taken abode in the shaman’s body).
In addition to this differentiation between shamans and laymen, there is an equally
clear pecking order between the shamanic spirits. Thus “gossip spirits” are “light”
(hüngün), whereas ancestral spirits (like Father of Harmai) are “heavy” (hünd). Unlike
the former, which both like to joke and be joked with, the latter must be treated with
the utmost respect; indeed, it is highly dangerous to laugh at “heavy spirits,” just as
they never joke with humans. Moreover, whereas “light” spirits are generally easy to
invoke, the others can be much more difficult. As a Darhad shamaness once told
Caroline Pegg, it may be necessary to invoke the spirits “more than one hundred
times before they come” (2001: 133). Indeed, according to Bumochir Dulam,
Mongolian shamans “change the meaning of song invocations when spirits do not
come. They slowly change into coaxing.” (2001: 17).
Is that what the “interpreter’s” (judging from the audience’s hysterical laughter)
shockingly rude debasement of the Father of Harmai in the aforementioned ritual was
all about? To answer this question, it is useful to briefly return to Bruce Kapferer’s study
of Sinhalese exorcism. According to Kapferer (1991), the “transformation of the
demonic” from a fearful state of unity to a laughable one of fragmentation is accomplished by means of different comical forms (dance, song, masks). Thus, as the exorcism
rite proceeds, “[T]he demon’s shape in appearance is [made] homologous with the
form of the comedy which preceded its entry. [The demon] represents in himself the
inconsistency of the comic and its revelation of contradiction” (Kapferer 1991: 303).
Now, if the first part of the Darhad shamanic ritual discussed above established a bridge
between two contradictory ideas of what a spirit is – a distal transcendental entity
residing up in “the skies,” and a much more immanent soul moving in and out of the
shaman’s body – was this not precisely because the “interpreter’s” jokes made the Father
of Harmai “homologous with the form of the comedy which preceded its entry”?
We suggest that joking in Darhad possession rituals involves the appropriation of
one aesthetic form to elicit another, homologous aesthetic form, namely, in the specific
case at hand, the invocation of the shamanic spirit Father of Harmai in its most immanent manifestation through the equally worldly phenomenon of humor. If joking thus
9781444333237_4_015.indd 268
8/7/2010 7:10:34 PM
Proportional Holism
269
emerges as a distinct shamanic modality – an occult technique with different properties
than other occult techniques like prayer and sacrifice – then this is because it is imbued
with the capacity to mold the cosmos in a particular way. After all, that is precisely what
seems to have happened in the above ritual: the “interpreter’s” jokes played the role of
a sort of occult leveling mechanism by means of which the Father of Harmai was
momentarily put down from his elevated position in “the skies.” However, this should
not be understood as a conventional carnivalesque event, where humor and laughter
serve to undermine dominant power structures (e.g., Mbembe 1992). The interpreter’s joking with the spirit put the latter down, yes; but the particular debasement at
hand did not simply involve a sociological deflation of the Father of Harmai’s rank
within the cosmic pecking order; it also gave rise to an ontological actualization as this
spirit was suddenly forced to materialize from his transcendental abode in “the skies.”
Darhad shamanic joking, then, is about making relations that differentiate – it is
about continually making nonhumans on par with humans, without, for that reason,
pretending that the two belong to one realm, as if the world added up to a single
whole.4 In that sense, the play with contrasting social ontologies that we have explored
in this first case study – between respecting and upholding hierarchies and yet simultaneously mocking and collapsing them, and between taking spirits seriously and at
the same time teasing them – allows for an ongoing escape from the always lurking
dangers of holistic totalization. More precisely, these and possibly other joking practices emerge as proportional practices in Corsín-Jiménez’s sense (2007) – as technologies for apportioning a given totality of relations into its proper shapes and sizes, in
ways that call to mind to the practices of “permitted disrespect” described by RadcliffeBrown and other structural-functionalist students of joking relationships. In both
African kinship systems and North Asian animist cosmologies, the “conjunctive and
disjunctive components” of relations are continually recombined so that a fragile balance between proximity and distance is kept intact. Thus, among the Darhads of
Mongolia and possibly elsewhere in North Asia, joking is not a matter of relativizing
the cosmos by creating reflexive distance to it. Rather, it is about securing the world’s
distance to itself through ongoing acts of spiritual apportioning.
Ridiculing the Spirits: The Role of Humor
in Yukaghir Hunting Animism
Let us move to the boreal forest and tundra of northeastern Siberia, which is the
home of the Yukaghirs, a tiny group of Paleo-Siberian hunters living along the tributaries of the Upper Kolyma River. Besides having survived centuries of demographic
decline, the people are remarkable in having maintained an economy based largely on
hunting. This condition has taken a further step after the collapse of the state farm in
1991, when people for the most part returned to a pure subsistence-based lifestyle.
Due to the paramount importance of hunting, and despite all the vicissitudes of
Sovietization, the Yukaghirs have maintained animistic beliefs in “spirit owners,” who
are said to control the prey animals and release them in a certain number as food to
hunters if they follow the proper rules of conduct. The Yukaghirs, then, do not see the
supernatural as a level of reality separate from nature. To them, the spirits are inhabitants of the same world as humans and animals, and they are experienced, at least in
certain situations, as being just as real (Willerslev 2007).
9781444333237_4_015.indd 269
8/7/2010 7:10:34 PM
270
In this respect, the brown bear is of particular significance. Not because its meat is
especially important in the subsistence economy of the Yukaghirs, who live mainly
from hunting the elk, but because the bear is believed to be loaded with an unsurpassed spiritual potency. As Ingold has stated with regard to the attitude of circumpolar peoples toward the bear, and which also holds true for the Yukaghirs, “Every
individual bear ranks in his own right as being on a par with the animal masters,
indeed he may … be [equivalent with] a master” (1986: 257; emphasis in original).
The fact that the bear, of all the animals, is singled out as being especially powerful is
perhaps most clearly reflected in the elaborate ritual treatment of its carcass after it has
been killed. Hunters generally try to make out the whole affair as an unfortunate
accident, for which they are not to be blamed. They will bow their heads in humility
before the dead animal and say, “Grandfather, who did this to you? A Russian killed
you.” Before removing its skin, they will blindfold it or poke its eyes out while croaking like a raven. This will persuade the bear that it was a bird that blinded it. Moreover,
while skinning the bear, they will say, “Grandfather, you must feel warm. Let us take
off your coat.” Having removed its flesh, the hunters then deposit its cleared bones
on a raised platform as the Yukaghirs used to do with an honored deceased relative. If
the ritual is violated, all sorts of terrible misfortunes are said to be sparked off. Thus,
Yukaghir myths and other narratives are crammed with stories about hunters who fail
to obey the ritual prescriptions and as a result lose their hunting prowess, so that the
entire camp staves to death. Likewise, other narratives describe how some disobedient
hunter is violently killed by a relative of the dead bear that seeks blood revenge for its
“murder” (Zukova et al. 1989; Spiridonov 1996 [1930]).
It is exactly because of these strict rules of etiquette governing the bear hunt that
the following observation made by Willerslev comes as a rather shocking surprise. He
was out hunting together with two Yukaghirs, an elderly and a younger hunter, and
they had succeeded in killing a big bear. While the elderly hunter was poking out its
eyes with his knife and croaking like a raven as custom prescribes, the younger one,
who was standing a few meters away, shouted to the bear, “Grandfather don’t be
fooled, it is a man, Vasili Afanasivich, who killed you and he is now blinding you!” At
first the elderly hunter doing the butchering stood stock-still as if he was in shock, but
then he looked at his younger partner and they both began laughing ecstatically as if
the whole ritual endeavor was one big joke. Then the elderly hunter said to the
younger one, “Stop fooling around and go and make a platform for the grandfather’s
bones.” However, he sounded by no means disturbed. Quite the opposite, in fact: he
was still laughing while giving the order. The only really disturbed person was
Willerslev, who saw the episode as posing a serious threat to his whole research agenda,
which was to “take animism seriously” (see Willerslev 2007). The hunter’s joke and
their ridiculing of the ritual event suggested that underlying the Yukaghir animistic
cosmology was a force of laughter, of ironic distance, of making fun of the spirits.
How could Willerslev as an anthropologist take the spirits seriously, when not even
the animists themselves did so?
Willerslev experienced several incidents of this kind of ridiculing the spirits. At one
time, for example, an old hunting leader was making an offering to his helping-spirit,
which is customary before an upcoming hunt. However, while throwing tobacco, tea,
and vodka into the fire, he shouted, “Give me prey, your bitch!” Everyone present
doubled up with laugher. Likewise, a group of hunters once took a small plastic doll,
bought in the local village shop, and started feeding it with fat and blood. While bowing their heads before the doll, which to everyone’s mind was obviously a false idol
9781444333237_4_015.indd 270
8/7/2010 7:10:34 PM
Proportional Holism
271
with no spiritual dispositions whatsoever, they exclaimed sarcastically, “Khoziain [Rus.
‘spirit-master’] needs feeding.” Now, direct questioning about such apparent breaches
of etiquette often proved fruitless. One hunter simply replied, “We are just having
fun,” while another one came up with a slightly more elaborate answer: “We make
jokes about Khoziain because we are his friends. Without laughter, there will be no
luck. Laughing is compulsory to the game of hunting.”
So what conclusion should we draw from this? Should we say that the Yukaghirs
have lost faith in their ancient animist ideology as a result of the longstanding Russian
and Soviet impact on their modes of thinking – implying that their joking about the
spirits reflects an increasing lack of belief in them? Or should we adopt a more
structural(ist) perspective, and consider such joking to be a way of maintaining a grand
cosmological alliance between the human and the nonhuman realms, in the manner of
Roberta Hamayon (1990)? We think neither is the case. Instead, we turn to the
Marxist-inspired scholar of postmodernist (and late socialist) discourse Slavoj Zizek
(1989) for inspiration. Ideology in its conventional Marxist sense, Zizek asserts, “consists in the very fact that the people ‘do not know what they are doing,’ that they have
a false representation of the social reality to which they belong” (1989: 31). Clearly,
this does not apply to the Yukaghirs, as they maintain an ironic distance toward their
official animist cosmology along with its requirements of treating the spirits with
extreme respect. Indeed, it is exactly the discordance between this prescribed ceremonial rhetoric of marked respect and the hunters’ practices of deception and manipulation that the jokes expose and that make them funny. Even so, after a good laugh, the
hunters always insist upon keeping to the official rhetoric and they continue to behave
according to the prescribed rules of ritual conduct. Thus, the formula proposed by
Zizek for the workings of ideology in the cynical and hyper-self-reflexive milieu of
postmodernism seems to fit the Yukaghirs as well: “They know very well what they are
doing, but still, they are doing it” (1989: 29).5 The Yukaghirs, therefore, are not really
naïve animists in the sense of blindly believing in the authority of the spirits. Rather,
they know very well that in conducting their ritual activities, they are following an illusion. Still they do not renounce it, but continue to do it even so.
But if the Yukaghirs are no hapless victims of false consciousness, but are fully aware
of the disparity between the rhetoric of spiritual authority and their actual practices
toward these entities, then we must ask what the importance of this gap is. In addressing this question, we need to discuss the key principle governing the Yukaghir hunting economy, which is the principle of “sharing.”
The Impossibility of Sharing
The Yukaghir distribution of resources follows in many respects a traditional huntergatherer economic model of sharing in that they run what Nicolas Peterson (1993) has
described as a “demand sharing” principle. This implies that people are expected to
make claims on other people’s possessions and those who possess more than they can
immediately consume or use are expected to give it up without expectation of repayment. This principle of sharing affects virtually everything from trade goods, such as
cigarettes and fuel, to knowledge about how to hunt, but it applies most forcefully to
the distribution of meat: “I eat, you eat. I have nothing, you have nothing, we all share
of one pot,” the Yukaghirs say (Willerslev 2007: 39). The important point for our
9781444333237_4_015.indd 271
8/7/2010 7:10:34 PM
272
argument, however, is that Yukaghir hunters engage with the nonhuman world of
animal spirits in much the same way as they engage with other humans, namely, through
the principle of demand sharing. Thus, in the forest, hunters will ask – even demand – the
spiritual owners to share their stock of prey with them in much the same way as they do
with fellow humans who possess resources beyond their immediate needs. They will, for
example, address the spirits of the rivers and places where they hunt by saying,
“Grandfather, your children are hungry and poor. Feed us as you have fed us before!”
In this sense, their animist cosmology could be interpreted as an integrated system,
an all-embracing cosmic principle based in sharing. This is in fact the argument proposed by Nurit Bird-David (1990, 1992) in her study of how hunter-gatherers relate
to their natural environments. Thus, she proposes that for the Nayaka of southern
India, the Batek of Malaysia, and the Mbuti of Zaire, the forest is regarded as a “parent,” who gives them food in overabundance without expecting anything in return –
what she labels the “giving environment” (Bird-David 1992: 28). The trouble is that
Bird-David, in proposing this argument, assumes that the hunting rhetoric of these
hunter-gatherers faithfully “matches” their activity of hunting. As such, she assumes
that she can look through their words to the actuality they point to, as if words and
deeds could be mapped onto one another in a simple one-to-one correspondence.
But they can’t – for if they could, we would have aligned these hunter-gatherers with
something akin to a cultural “death wish,” for surely a society that hunts by simply
waiting for the forest to “feed” them, without making any preparations or taking any
effort, would not survive long. What this points to, then, is that the Yukaghirs’ rhetoric about the forest being a “generous parent” is not meant to be taken too literally.
Rather, it is a sophisticated means of spirit manipulation, which is an inherent, and
even necessary, part of the Yukaghir hunting animism.
This becomes evident when we realize that a paradox is built into the moral economy of sharing, which makes it risky – lethal, in fact – to take the principle of unconditional giving at face value. We have already seen that in a sharing economy, people
have the right to demand that those who possess goods beyond their immediate needs
to give them up. With regard to the hunter–spirit relationship, this means that, as long
as an animal spirit possesses prey in plenty, the hunter is entitled to demand the spirit
to share its animal resources with him, and the spirit for its part is obliged to comply
with the hunter’s demands. However, if the wealth divide between the two agencies
becomes displaced, their respective roles as donor and recipient will be inverted and
the spirit will now be entitled to demand the hunter to share his resources with it, and
it will assert its claim by striking him with sickness and death.6 What this points to,
then, is that the condition of truly radical sharing with the animal spirits is ultimately
unsustainable and indeed self-destructive as it sooner or later ends with the roles of
donor and recipient being reversed, so that the human hunters fall prey to the spirits
of their animal prey.7
The hunters’ response is to transform the sharing relationship with the spirits into a
“play of dirty tricks” (Rus. pákostit’), which effectively means turning the hunt into a
game of “sexual seduction” by inducing in the animal spirit an illusion of a lustful play
(Willerslev 2007: 89–118).8 The feelings of sexual lust evoked in the spirit are then
extended to its physical counterpart, the animal, which the next morning, when the
hunter goes out in search for prey, will run toward him and “give itself up” to him in
the expectation of experiencing a climax of sexual excitement (Willerslev 2007: 89–118).
However, after the killing of the animal, its spirit will realize that what it took to be
lustful play was in fact a brutal murder, and it will seek revenge accordingly. The hunter,
9781444333237_4_015.indd 272
8/7/2010 7:10:34 PM
Proportional Holism
273
therefore, must cover up the fact that he was the one responsible for the animal’s death.
We have already described this procedure in relation to the bear ritual, where hunters
by means of various tactics of displacement and substitution will seek to direct the anger
of the animal spirit against non-Yukaghirs, humans and nonhumans alike. As a result,
the hunter himself will not appear to have taken anything from the spirit, at least not
formally, and no sharing relation was therefore ever established between the two. This
in turn rules out the spirit’s right to demand the hunter’s soul. In this way, hunters seek
to maximize benefit at the spirit’s expense, while avoiding the risk of falling into the
position of potential donor. This corresponds in effect to what Marshall Sahlins (1972:
195) has called “theft,” which he characterizes as “the attempt to get something for
nothing,” and which he argues to be “the most impersonal sort of exchange [that]
ranges through various degrees of cunning, guile, stealth, and violence.”
Creating Discrepancy Through Laughter
Radcliffe-Brown saw joking relationships as a particular way of ordering relationships,
which combine attachment, or “conjunction,” with separation, or “disjunction,”
between social agencies (1952a: 98). Thus, joking relationships represent the very “contrary” or “polar opposite” of “extreme respect” and “avoidance” (1952b: 106).
Nonetheless, he asserted that the two are often co-present, as the Chinese Yin and Yang
symbolism in which two opposites are required to make a unity or harmony. Relations
of joking and respect do, in this sense, add up to a whole, yet it is a whole defined by
irreducible differences – what Radcliffe-Brown called “unity in duality” (1952b: 115).
In line with this argument, we propose that Yukaghir hunting animism is characterized by a similar “unity in duality” as the morally negative reality of “theft” is
co-present with the morally positive reality of “sharing.” The latter reality is “official” and emphasizes the ethos not only of unconditional giving, but also of marked
respect and interpersonal dependency between the human and spiritual realms. In
the other, parallel reality, the hunter is engaged in thievery, with its own moral codex
of seduction, trickery, and even murder. While the two realities are essentially
opposed, they nevertheless occur in one and the same social space, which is why it
makes sense to say that both realities are equally animist (for animism is not restricted
to an official sphere, but encompasses all aspects of Yukaghir life). Contrary to the
functionalist underpinnings of Radcliffe-Brown’s theory, the two realities – the
dimension of sharing and the dimension of theft – are not and should never be collapsed into a unified whole. An essential element of noncoincidence between the
two must always be kept in place, and is kept in place, because it is exactly due to this
disparity between the official rhetoric and the practical actions of hunters that they
know the illusion of their own cosmological economy of sharing. Indeed, it is this
very “fission” within Yukaghir hunting animism that allows hunters to be the bearers of firm cosmological beliefs in the demand-sharing principle without absolutely
coinciding with these beliefs.
Let there be no misunderstanding. We do not mean to suggest that through joking,
hunters question the reality of the existence of spirits as do the Singhalese, whose
“laughter dispels [the spirits] to an unreal state of ‘make-believe’ ” (Kapferer 1991:
315–25). Rather, their joking reveals that they do not take the authority of the spirits
as seriously as they usually say they do or their mythology tells them to do. Joking and
9781444333237_4_015.indd 273
8/7/2010 7:10:34 PM
274
other types of ridiculing discourses about spirits play a prominent role in the everyday
life of hunters, not because they entail resistance to or subversion of the dominant
cosmological values of the sharing economy. To celebrate hunters’ ridiculing of spirits
as some deliberate last-ditch resistance to ideological domination would be to miss
the point entirely. Virtually all Yukaghirs ascribe to the demand-sharing principle and
regard it as both immutable and morally just. However, they are well aware that this
system or indeed whole must never become a pervasive totality of force. For the
Yukaghirs such a condition of totalizing holism would ultimately stand for “death” as
it would give the spirits the moral right to “consume” them in a series of divine predatory attacks. To avoid this, hunters must constantly steer a difficult course between
parallel realities by transcending the animist rhetoric through equally animistic forms
of seduction and deception, while also maintaining morally positive relations of sharing with the animal spirits. In this, the ongoing ridiculing of the spirits plays a key
role, for it reminds hunters not to “read” the complex of myths, beliefs, and rituals
too literally, but instead to carve out a parallel space of their own at the right angle, so
to speak, from the moral discourse of respect.
Conclusion
Numerous books on North Asia’s indigenous peoples have presented them as victims
of dominant political, economical, and cultural forces that have uprooted their livelihoods and cosmologies. Indeed, the prevailing sensation that the animist cosmos is at
the verge of losing its shape, and that certain apportioning acts must be performed to
prevent this from happening, could be seen as a result of the fact that these groups were
for 50 years subjects of state socialist polities that ridiculed, banned, and often penalized
the “superstitious” beliefs and practices we discussed in the two case studies.
However, the trouble with this line of thinking – at least when pursued with too
vigorous a sense of conviction in terms of its own explanatory superiority – is that it
risks falling into the very same totalizing trap as the structural-functionalists once did.
While the “large” societal transformations of the postsocialist transition evidently have
an impact on the cosmological ideas and practices of “small” indigenous peoples like
the Darhads and the Yukaghir, our concern in this chapter has not been to describe the
interplay between these two scales by “holistically” collapsing them one onto each
other. Instead of addressing whether the animist cosmos should represent a “whole” or
a “part” in our construction of an analytical context, our concern has been what these
people are doing for the world to assume the shape and size they find fit. Instead of
trying to answer the moot question of whether our analytical models are (and should
be) holistic or not, we have focused on the question of whether, and if so how, different
ethnographic subjects perform holism as a distinct social practice. In this sense, what we
have done in this chapter has itself been an act of (analytical) apportioning – a comparative account of two proportional wholes and how people do these, as opposed to the
wholes (or parts) envisioned by anthropology and the holism performed by us.
As we have tried to demonstrate in this chapter, joking constitutes an important
(albeit surely not the only) proportional holistic practice in the Northern Asian context. When Darhads and Yukaghirs make fun with the spirits, this should not be
understood as acts of enlightened resistance, whereby they seek to liberate themselves
from hegemonic yokes of domination. Quite the opposite, in fact: both the Darhads
9781444333237_4_015.indd 274
8/7/2010 7:10:34 PM
Proportional Holism
275
and the Yukaghirs consciously inscribe themselves at the center of their animist worlds,
without any wish of breaking out from them. Their aim is not to resist, but to actualize the spirit, which cannot function properly (elicit effects) through forms of totalization, but only through forms of separation. As our two case studies have showed,
joking relationships thus keep the animist cosmos bifurcated into parallel domains,
enabling people to partake simultaneously in incongruous economies. Peoples’ playful relationships toward their cosmological rules allow for an escape from the latent
dangers of totalization by invoking alternative scenarios that maintain the balance
between too much and too little integration, which is what animism is all about.
Acknowledgments
We thank participants in the Beyond the Whole workshop for their comments on the first version of this chapter. We would also like to thank Rebecca Empson and, in particular, the two
editors for useful suggestions as to how to improve it.
Notes
1 By the traditional concept of “holism,” we refer to how the term was deployed during the
heyday of anthropological modernism between World War I and the mid-1970s (Parkin
2007: 2–3), namely, as the perfect integration of discrete parts into one encompassing
whole. As has been pointed out by its many critics (e.g., Gupta and Ferguson 1997; Olwig
and Hastrup 1997), this concept of holism conveys the image of discrete units marked by
stable and well-defined boundaries, which are qualitatively different and mutually independent, just like different “cultures” were once said to be.
2 See Pedersen (2001, 2007a, in press), Willerslev (2004, 2007), Holbraad and Willerslev (2007),
Corsín-Jiménez and Willerslev (2007), and Pedersen and Willerslev (forthcoming).
3 Incidentally, the term helmerch, in addition to meaning “interpreter,” is also used to denote
the matchmakers who, supposedly, played a vital role at the traditional Mongolian wedding.
Apparently, each side was here represented by a helmerch, who would perform a “dialogue
… not carried on in normal conversation, but in a highly developed art of spontaneous
poetry. The bride’s speaker trie[d] to make it difficult for the groom’s spokesman by changing rhyme and meter. This, in turn, [was] matched and followed at every poetic turn by the
spokesman of the groom” (Jagchid and Hyer 1979: 84). According to Wasilewski (1983),
who identifies the role of the helmerch as that of a typical trickster, “[T]he illusion [which
the matchmaker] establishes by means of quick words – whereby he appears similar to the
prestidigitator or magician – will be recognised as deception. The Mongolian matchmakers
are cheat[ing] during the wedding and the result of their actions – leading the bride toward
her married status – is thus understood as betrayal” (51).
4 According to Hamayon, “Shamanism supposes being on equal footing with the spirits.… [S]
uch attitudes are contrary to the aloofness and respectful submission to God required in world
religions. Transcendence precludes direct contact, identification and imitation, all of which
amount to denying it” (1992: 19–20). This, however, is not to say that spirits are like humans.
As Urgunge Onon recalled from his childhood in Manchuria, the shamanic spirits (barkan)
“were not ‘higher’ than human beings, just different.” For, as he said, “if you ask me whether
barkans were higher than human beings that’s almost like asking if a horse is higher than a cow,
or a cow higher than a horse; they are just entirely different” (Humphrey 1996: 191). In the
same way, one could rhetorically state, Darhad shamanic spirits must also be “entirely different
from” people, for if they were not, why then have an “interpreter” at all?
5 Zizek dismisses the classical Marxist notion of ideology as “false consciousness.” To him,
the very objects of study are those everyday ironic narratives, which are formed largely
9781444333237_4_015.indd 275
8/7/2010 7:10:35 PM
276
outside the control of a society’s dominant ideology, and which embody significant beliefs
and practices at odds with it. This is not to say that Zizek denies that dominant ideologies
exist; but he questions that they are an important means for lending cohesion and stability
to a society, because such ruling ideologies are often internally fissured and contrary and
usually quite unsuccessful in fully shaping the consciousness of their subordinates.
6 Indeed, we find this fearful reversal of roles described again and again in Yukaghir narratives,
where it is told that a spirit out of “love” for a particular hunter makes him take all the prey
offered (Willerslev 2007: 44–7). As a result, he comes to stand out as accumulating a surplus
of animal souls, which in turn gives the spirit the right to go and kill him and drag his soul
back to its dwelling place as its “spouse.”
7 To the critical reader, this might appear to be more like an act of “reciprocity” than “sharing.”
However, as has been argued elsewhere, sharing and reciprocity are not necessarily manifestly
different forms of exchange, as many writers of hunter-gatherers would have it (see, e.g.,
Bird-David 1990, 1992; Woodburn 1998: 50; Gell 1992: 152), but are rather “reversibles”
that contain or eclipse the other within (Corsín-Jiménez and Willerslev 2007: 533).
8 Thus, before an upcoming hunt, hunters will undergo a long process of “dehumanization”
in which their bodies are transformed in the image of prey. This involves going to the sauna
the evening before the hunt, where they will wipe themselves with dry whisks from birch
trees. Hunters say that the prey recognizes the attractive smell of birch and does not flee, but
comes closer to them. Likewise, a hunter must abstain from sex altogether before undertaking a hunting trip. This is partly to avoid a human stench on his body, but also because his
sexual attention should be directed toward the animal’s spirit, which is visited by his soul
(Yuk. ayibii) during his nightly dreams in order for the two to have sexual intercourse.
References
Bird-David, N. (1990). The Giving Environment: Another Perspective on the Economic
System of Hunter-Gatherers. Current Anthropology 31: 183–96.
Bird-David, N. (1992). Beyond the Original Affluent Society: A Culturalist Reformulation.
Current Anthropology 33: 25–47.
Bohm, D. (1983). Wholeness and the Implicate Order. London: Ark.
Corsín-Jiménez, A. (2007). Well-Being in Anthropological Balance: Remarks on Proportionality
as Political Imagination. In Culture and Well-Being: Anthropological Approaches to Freedom
and Political Ethics, ed. A. Corsín-Jiménez. London: Pluto, 180–200.
Corsín-Jiménez, A., and R. Willerslev (2007). An Anthropological Concept of the Concept:
Reversibility Among the Siberian Yukaghirs. Journal of the Royal Anthropological Institute
(N.S) 13:527–44.
Douglas. M. (1968). The Social Control of Cognition: Some Factors in Joke Perception. Man
3:361–76.
Dulam, B. (2001). Degrees of Ritualisation: Language Use in Mongolian Shamanic Ritual.
MPhil thesis, Department of Social Anthropology, Cambridge University.
Gell, A. 1992. Inter-Tribal Commodity Barter and Reproductive Gift Exchange in Old
Melanesia. In Barter, Exchange and Value: An Anthropological Approach, ed. C. Humphrey
and S. Hugh-Jones. Cambridge: Cambridge University Press, 142–68.
Gell, A. (1998). Art and Agency: An Anthropological Theory. Oxford: Clarendon.
Gupta, A., and J. Ferguson, eds. (1997). Culture, Power, Place: Explorations in Critical
Anthropology. Durham: Duke University Press.
Hamayon, R. (1990). La Chasse à L’âme: Esquisse d’une Théorie du Chamanisme Sibérien.
Nanterre: Société d’Ethnologie.
Hamayon, R. (1992). Game and Games, Fortune and Dualism in Siberian Shamanism. In
Northern Religions and Shamanism, ed. M. Hoppal and J. Pentikainen. Budapest:
Akademiai Kiado, 134–7.
9781444333237_4_015.indd 276
8/7/2010 7:10:35 PM
Proportional Holism
277
Holbraad, M., and M. A. Pedersen (2009). Planet M: The Intense Abstraction of Marilyn
Strathern. Anthropological Theory 9(4): 271–394.
Holbraad, M., and R. Willerslev (2007). Afterword: Transcendental Perspectivism: Anonymous
Viewpoints from Inner Asia. Inner Asia 9(2): xx.
Humphrey, C., with U. Onon (1996). Shamans and Elders: Experience, Knowledge, and Power
Among the Daur Mongols. Oxford: Clarendon.
Ingold, T. (1986). Hunting, Sacrifice and the Domestication of Animals. In The Appropriation
of Nature: Essays on Human Ecology and Social Relations. Manchester: University of
Manchester, 243–77.
Ingold, T. (2007). Movement, Knowledge and Description. In Holistic Anthropology: Emergence
and Convergence, ed. D. Parkin and S. Ulijaszek. Oxford: Berghahn, 194–211.
Jagchid, Sechin, and Paul Hyer (1979). Mongolia’s Culture and Society. Boulder, CO:
Westview.
Kapferer, B. (1991). A Celebration of Demons: Exorcism and the Aesthetics of Healing in Sri
Lanka. Oxford: Berg.
Koestler, A. (1964). The Act of Creation. London: Arkana.
Mbembe, A. (1992). Provisional Notes on the Postcolony. Africa 62(1).
Morreall, J. (1983). Taking Laughter Seriously. Albany: State University of New York Press.
Mosko, M. S., and F. H. Damon (2005). On the Order of Chaos: Social Anthropology and the
Science of Chaos. Oxford: Berghahn.
Olwig, F., and K. Hastrup, eds. (1997). Siting Culture: The Shifting Anthropological Object.
London: Routledge.
Parkin, D. (2007). Introduction: Emergence and Convergence. In Holistic Anthropology.
Emergence and Convergence, ed. D. Parker and S. Ulijaszek. Oxford: Berghahn, 1–20.
Parkin, D., and S. Ulijaszek, eds. (2007). Holistic Anthropology: Emergence and Convergence.
Oxford: Berghahn.
Pedersen, M. A. (2001). Totemism, Animism and North Asian Indigenous Ontologies. Journal
of the Royal Anthropological Institute 7(3): 411–27.
Pedersen, M. A. (2007a). Talismans of Thought: Shamanist Ontology and Extended Cognition
in Northern Mongolia. In Thinking Through Things: Theorizing Artefacts Ethnographically,
ed. A. Henare, M. Holbraad, and S. Wastell, 141–66. London: University College London
Press.
Pedersen, M. A. (2007b). Multitude Minus Myth: Theorizing Darhad Mongolian Perspectivism.
Inner Asia 10 (1, special issue on Inner Asian Perspectivism, ed. M. A. Pedersen, R.
Empson, and C. Humphrey).
Pedersen, M.A. (in press). Not quite Shamans. Spirit Worlds and Political Lives in Northern
Mongolia after Socialism. Ithaca: Cornell University Press.
Pedersen, M.A. and R. Willerslev (Forthcoming). “The Soul of the Soul is the Body”: Rethinking
the Soul Through North Asian Ethnographies. Submitted to Journal of the Royal
Anthropological Institute.
Pegg, C. (2001). Mongolian Music, Dance, and Oral Narrative. Seattle: University of
Washington Press.
Peterson, N. (1993). Demand Sharing: Reciprocity and the Pressure for Generosity Among
Foragers. American Anthropologist 95(4): 860–74.
Radcliffe-Brown, A. R. (1952a). A Further Note on Joking Relationships. In Structure and
Function in Primitive Society. London: Cohen & West.
Radcliffe-Brown, A. R. (1952b). On Joking Relationships. In Structure and Function in
Primitive Society. London: Cohen & West.
Robbins, Joel. (2007). Continuity Thinking and the Problem of Christian Culture: Belief,
Time, and the Anthropology of Christianity. Current Anthropology 48(1): 5–38.
Sahlins, M. (1972). Stone Age Economics. New York: Aldine de Gruyter.
Seizer, Susan. (1997). Jokes, Gender, and Discursive Distance on the Tamil Popular Stage.
American Ethnologist 24(1): 62–90.
9781444333237_4_015.indd 277
8/7/2010 7:10:35 PM
278
Spiridonov, N. I. (Teki Odulok) (1996). Oduly (Yukagiry) Kolymskogo Okruga [1930]
[The Odul [Yukagirs] of the Kolyma Region]. Yakutsk.
Strathern, M. (1988). The Gender of the Gift: Problems With Women and Problems With Society
in Melanesia. Berkeley: University of California Press.
Strathern, M. (2004). Partial Connections, updated ed. Walnut Creek, CA: Altamira.
Wagner, R. (1977). Analogic Kinship: A Daribi Example. American Ethnologist 4(4): 623–42.
Wagner, R. (1991). The Fractal Person. In Big Men and Great Men. Personifications of Power
159–73.
Wasilewski, Jerzy S. (1983). The Trickster, the Initiate and Paradoxes of Cognition. Preliminary
Remarks of Selected Symbols of Initiation. Ethnologia Polana 9: 45–54.
Willerslev, R. (2004a). Not Animal Not Not-Animal: Hunting, Imitation and Empathetic
Knowledge among the Siberian Yukaghirs. Journal of the Royal Anthropological Institute
(N.S.) 10: 629–52.
Willerslev, R. (2004b). Spirits as Ready to Hand: A Phenomenological Study of Yukaghir
Spiritual Knowledge and Dreaming. Anthropological Theory 4(4): 395–418.
Willerslev, R. (2007). Souls Hunters: Hunting, Animism, and Personhood Among the Siberian
Yukaghirs. Berkeley: University of California Press.
Woodburn, J. (1998). Sharing is Not a Form of Exchange: An Analysis of Property-Sharing in
Immediate-Return Hunter-Gatherer Societies. In Property Relations: Renewing the Anthropological Tradition, ed. C. M. Hann. Cambridge: Cambridge University Press, 48–63.
Zizek, S. (1989). The Sublime Object of Ideology. London: Verso.
Zukova, L. N., I. A. Nikolaeva, and L. N. Dëmina (1989). Folklor Yukagirov Verhnei Kolymy
[Folklore of the Verkhne Kolyma Yukaghirs], pts. 1–2. Yakutsk.
9781444333237_4_015.indd 278
8/7/2010 7:10:35 PM
16
One Past and Many Pasts
Varieties of Historical Holism
in Melanesia and the West
Eric Hirsch and Daniele Moretti
In 1909 Father Alphonse Clauser of the Mission du Sacré Coeur (MSC) traveled for
the first time from the Mafulu area in the lower Auga Valley into the upper Udabe
Valley.1 These are two of the five river valleys that make up the lands of the Fuyugespeaking Papuan highlanders. Clauser did not travel alone but was accompanied by
dozens of Fuyuge men and women. Although this was the first time a missionary
had ventured into the Udabe, that certainly was not the case for the Fuyuge men
and women with him. Many of them had affinal and ritual connections in the neighboring valley, connections that were now being activated. Clauser’s journey into the
Udabe was associated with the intended mission expansion into the valley. He visited a number of the adjacent dialect groups after he entered at Evese. He reported
later that after briefly visiting Evese, he moved on to the neighboring village of
Ononuku:
[Nemb u Koge] the [Ononuku] chief approaches very timidly at first: he has never seen
a white man before. But as soon as he hears me speaking Fu[y]uge, the Mafulu language
which is also his language, he is reassured and clicks with his tongue, as a sign of surprise
and pleasure, and he reassures his fellow tribe members. (Annales de Notre Dame du
Sacré Coeur 1910: 213)
Clauser later returned to the Auga Valley and the Popole mission station at Mafulu.
The Popole station had been opened four years previously, in 1905, by Father Paul
Fastré, the first missionary established among the Fuyuge (see Fastré n.d.).
During his fieldwork in the upper Udabe Valley beginning in the mid-1980s, Hirsch
was told that Clauser was the first missionary to enter the valley. His Fuyuge hosts told
him the names of some of the people Clauser had met – such as Nemb u Koge, mentioned by Clauser – and where he traveled during his time in the Udabe. They did not
tell him much else about him, although he did return to the valley again several years
later when a mission station was established at Ononge. However, Hirsch learned
much about Clauser’s journey from the journal articles he penned for the mission
publication, Annales de Notre Dame du Sacré Coeur (hereafter, ANDSC), published
in Issoudun, France, since 1866.
9781444333237_4_016.indd 279
8/7/2010 7:10:56 PM
280
Hirsch and Moretti
By consulting the MSC Annales, Hirsch was able to read about Clauser’s historic
journey and the circumstances of his visit. He was also able to read other articles by
his fellow missionaries working in Papua, or other parts of the world, if he chose.
Clauser’s written narratives were also potentially connected to other narratives about
Papua, such as those being written by colonial government officials, or other narratives about the mission, or about travel in the Pacific at the time – the list is seemingly
endless. We are not suggesting that this actually happened, but that Clauser’s writings
about his progress into the Upper Udabe Valley could combine with other narratives
covering other landscapes, persons, or things. In other words, here we have the idea
of an unfolding series of stories that cover the entire world, each possibly added to
others. More specifically, Clauser’s journey figures in several of the narratives written
about the mission, appearing briefly, for example, in the account composed by Delbos
(1985) to mark the hundredth anniversary of the MSCs in Papua that took place in
1985. In another sense, though, Clauser’s journey and the texts he produced about it
are informed by particular presuppositions that derive from the early Latin Church
and the requirement of the Church to expand and reproduce itself through the replication of bishoprics – a process whose telos of salvation calls for the progressive conversion and incorporation of hitherto “heathen” peoples into the fold of the Christian
Church. But these Roman Catholic presuppositions coexist with a more modern,
progress-orientated historical sensibility – an expansive “civilizing process.” This
expansive enterprise was most clearly manifested in the projects of the colonial government, then in its early years in Papua.
The manner in which these different but related expansive conventions were understood and narrated stemmed from a common idea. This idea is what has been called
Universal History or what Jay (1984: 105) refers to as “history as a coherent and
meaningful unity … a progressive longitudinal totality.” According to the philosopher
of history Louis Mink (1987: 190), “Universal History was not the idea that there is
a particular plot in the movement of history but the assumption that underlay all proposals to display one sort of plot or another.” This idea was already found in the works
of ancient historians like Polybius but was later developed in a Christian manner
derived from the Bible by Augustine in his City of God, and in its modern form in
Vico’s The New Science (Burrow 2007: 68–9, 342, 468). As further argued by Mink
(1987: 193; cf. Megill 1995), the notion of Universal History did not disappear with
the rise of Romanticism and expanding historical and national consciousness but survived in the doctrine of progress, in orthodox historical materialism. and in more
recent works by scholars like Fukuyama (1992) or Diamond (2005 [1997]) (see
below). Nevertheless, Mink suggests that whereas this idea once explicitly served as a
“regulative principle of thought about the past” (1987: 188–9), now it continues to
serve in this fashion, but more implicitly.
Mink’s thesis is that this idea is complex, but that it is informed by several common
features (1987: 190). The first is the assertion “that the ensemble of human events
belongs to a single story,” whether the work of divine Providence or later, in its “secularized” form – that of no one – just a story that is there (1987: 190). This later form
is the story of the discipline of history itself (see Megill 1995). The second idea is
“that there is a single central subject or theme in the unfolding of the plot of history”
(Mink 1987: 190). Hegel, and more recently Fukuyama (1992), provided a unified
and complex statement of this notion. The third aspect is that “it is implied that the
events of the historical process are unintelligible when seen only in relation to their
immediate circumstances” (Mink 1987: 191; emphasis in original). Thus, many
9781444333237_4_016.indd 280
8/7/2010 7:10:56 PM
Historical Holism in Melanesia and the West
281
Westerners by virtue of their education and schooling carry around certain outlines or
grids of historical development over lengthy durations into which details about the
past can be entered (cf. Lowenthal 1985; Samuel 1994). Lastly, according to Mink
(1987: 191–2), “Universal History did not deny the great diversity of human events,
customs, and institutions; but it did regard this variety as the permutations of a single
and unchanging set of human capacities and possibilities, differentiated only by the
effects of geography, climate, race, and other natural contingencies.”
Along with other scientific disciplines, anthropology rejected the idea that history
is a divinely ordained “longitudinal totality.” At the same time, it both proposed and
critiqued alternative, more “secular” readings of human history as a linear progression
between, say, “simpler” and “more complex” modes of thought or of social, political,
and economic organization. As it documented more and more thoroughly the variability of human social capacities, conventions, and institutions, modern anthropology
also developed a “relativistic challenge” to the aforementioned notion of an unchanging and universal underlying “human nature.” This does not mean that such religious
or secular versions of history as a universal narrative with a clear plot and telos have
been completely displaced from the West or, indeed, from anthropology itself. The
idea that history moves to an end determined by the will of God and that all past,
present, and future events can be tied back to the events and prophecies narrated in
the Bible or other religious texts is still alive in certain quarters. At least up to the
recent weakening of neoconservativism and the global economic crisis, many also
continued to believe in and ostensibly promote a vision of history as a universal progression toward liberal democracy and capitalism. And, as the editors note in the
introduction to this volume, ideas about the fundamental “universalities of human
existence” are still evident in anthropology and other sciences, particularly in fields
such as biological anthropology or studies of cognition.
As argued by Mink (1987), few believe anymore that a single, unified narrative of
the past could ever be actually written or read. But this does not mean, he notes, that
Westerners have truly abandoned “the concept of universal history,” only that they
have put aside “the concept of universal historiography” (1987: 94; emphasis in original). In other words, even when they think that past events are not really guided by
a unitary plot or are simply too many to ever be fully recorded and read as a single
account, Westerners still hold on to the fundamental notion that “everything that has
happened belongs to a single determinate realm of unchanging actuality (‘What’s
done is done. You can’t change the past.’)” (1987: 194). Further, they view this
“past actuality” as a story that may never be told in full but that is nevertheless out
there and always knowable at least in part(s) (1987: 189). The world is full of potential narratives of the past, a never-ending “place for stories,” as the historian William
Cronon (1992) has phrased this.
There is an important associated element to Mink’s argument which has been highlighted by others, who we refer to below. He suggests that the above assumptions
informing the notion of Universal History also contain incompatibilities about the
idea of event and narrative. The orthodox view is that narratives are constructed out
of events, res gestae as one damned thing after another (see Lowenthal 1985: 218).
Mink contests this assumption:
[T]he concept of event is primarily linked to the conceptual structure of science (and to
that part of common sense that has adopted the language and methods of science); but
in that conceptual structure it is purged of all narrative connections, and refers to
9781444333237_4_016.indd 281
8/7/2010 7:10:56 PM
282
Hirsch and Moretti
something that can be identified and described without any necessary reference to its
location in some process of development – a process which only narrative form can
represent. Therefore, to speak of a “narrative of events” is nearly a contradiction in
terms. That it is not perceived as such, I suggest, again reflects the extent to which the
idea of Universal History survives as a presupposition. To the extent that historical
actuality is regarded as an untold story, then in that untold story our conceptual problems disappear.… “Events” (or more precisely, descriptions of events) are not the raw
material out of which narratives are constructed; rather an event is an abstraction from
a narrative. (Mink 1987: 200–1)
Thus, Mink (1987: 199), in line with Hayden White (1987) and as expressed in different ways by Lévi-Strauss (1966 [1962]), argues that narrative form in history, as
much as in myth or fiction, “is an artifice, the product of individual imagination.”
Nonetheless, these writers explore the social conventions and constraints instructing
what the imagination is able to conjure.
With these ideas in mind, as well as the brief opening example about mission
expansion in Papua, we want to propose that there is a kind of figure–ground association at work in the way Westerners perceive particular pasts or histories and the
unstated assumption of Universal History. To narrate a particular representation of
the past – as in the form of written history – presupposes that the world is filled with
all sorts of untold stories that can be told, a universal unchanging past actuality; and,
furthermore, that the world is there to create new landscapes inhabited by these
stories, as in Clauser’s account of his journey. At the same time, the unspoken
assumption of Universal History is sustained by the expanding number of particular
accounts of the past or histories that are continually produced. Universal History is
a “whole” that exists nowhere in particular, except through the specific histories and
accounts of the past that exist, potentially, everywhere.
It is also important at this point to highlight another element in Western understandings of the past: the difference between what have been called, following
Oakeshott (1983), “the practical past” and “the historical past.” The practical past is
the past that is carried around and evoked for specific purposes, such as the way politicians summon images of the past to legitimate their policies or rhetoric. This is supposed to differ from the historical past that is studied by disinterested historians of
various persuasions and that implies a separation between past and present. Certainly,
both forms of past overlap distinctly in the Western context – Lowenthal (1985: 237)
even suggests that the distinction is ultimately illusory because “the practical man’s
past is seldom exclusively operational; the historian too is unavoidably presentminded.” Nevertheless, the existence of this conceptual distinction is further evidence
of the Western assumption that “the past” can be apprehended “objectively” as well
as evoked “instrumentally” because it exists “out there” as a universal, unchanging
totality distinct from “the present” and “future.”
In the remainder of this chapter, we draw on existing literature, the aforementioned
narrative about Clauser’s mission’s expansion in the Udabe, and other examples of historical and mythical narratives originating from or linked to Papua New Guinea (PNG)
to contrast these Western views about the past with those which have been documented
in Melanesia. We suggest that a different kind of figure–ground relation is at work in
Melanesian conventions about particular pasts and what we wish to refer to as a “Universal
Past.” In choosing the term “Universal Past,” we are clearly drawing a parallel with the
Western idea of Universal History. As far as we are aware, Melanesians do not have such
a term, but the notion of Universal Past conveys what we think is a prevalent but unstated
assumption in Melanesian imaginings and perceptions. Here one finds a similar
9781444333237_4_016.indd 282
8/7/2010 7:10:57 PM
283
proliferation of stories about the past but there is the presupposition that they are made
possible or derive directly from a single creative source that accounts for the origins and
divisions of the whole world. Thus, all the stories about the past are made possible by a
single source from which the past derives, while evidence of this single source is provided
by the proliferation of accounts to do with the past. In contrast to Universal History, we
suggest, the Universal Past is a whole that exists, potentially, everywhere. Evidence of its
existence is confirmed and sustained by particular stories of the past (and narratives in
the present). Although the Universal Past exists conceivably everywhere, this ever-presence derives ultimately from one source, from one past unity.
A further significant distinction between Universal History and Universal Past is
that the latter is inherently “practical.” By this we do not mean so much that in
Melanesia to tell a myth or other kind of narrative is very often (though of course, not
exclusively) to elicit a response from, and thus to have an effect on, those who listen,
whether they be other Melanesians, mining companies, or white people more generally.2 More than this, we wish to suggest that in the narrative conventions of the Universal
Past, the past “is not a foreign country” or a realm of actuality objectively separate
from present and future. Rather, it is something immanent in, inseparable from, and
always highly consequential for them. In saying this, we mirror some of the insights
developed in Battaglia’s (1995) and Strathern’s (1995) reflections on Western and
Melanesian modes of evoking the past and their perceptive distinction between what
they define as “synthetic” versus “practical” or “substantive” forms of nostalgia.
The aim of this chapter is thus to draw on Melanesian models of the past to bring
into relief the unspoken Western assumptions of Universal History. This comparative
“thought experiment” will not only highlight what we feel to be the dominant social
conventions and constraints instructing “what the imagination is able to conjure” in
Melanesian and Western narratives about the past, but also reveal the asymmetry
between them. If “Universal History” and “Universal Past” are, as we argue here, two
different holistic approaches to the past, our analysis will attempt to show that they
are also “partial wholes” whose narrative conventions embody significantly “partial”
(in the sense of “interested”; see Busse 2005: 457–8, drawing on Strathern 1991)
claims and counterclaims regarding past and present power inequalities, connections,
and obligations (or lack thereof) between Westerners and Melanesians.
Papuan “Hero Tales,” Tidibe, and Missionization
Among the Fuyuge
A good initial example of what we are referring to as narratives about a “Universal
Past” are the Papuan “Hero Tales” described by Wagner (1972: 17–37). These linked
myths cover a wide area of Western Papua and the total effect, from the local
perspective, is the “continued adventures of a single wandering hero” traveling from
what the anthropologist perceives as one society to another (1972: 20). As Wagner
(1972: 20) describes, these stories often commence with some shaming interaction
that precipitates the hero’s journey: “he travels across the sea seeking women and
bringing vegetable food, or journeys to the land of the dead, or flees from a pursuing
woman with whom he has shamed himself. Landmarks and curious features along his
route are often linked to his passage.” It is in the course of this journey that “the
landscape is created or brought into its present state, so that the initial incidents are
universalized and extended through the world” (1972: 35; emphasis added).
9781444333237_4_016.indd 283
8/7/2010 7:10:57 PM
284
Hirsch and Moretti
More recently, Busse (2005) has considered these tales in the context of contemporary economic, political, and religious affiliations affecting this large geographical
region. In the Lake Murray–Middle Fly area, the wandering hero is called Nggiwe, or
often Jesus Nggiwe. These stories are closely connected with Christian mythology,
and the men who told them to Busse said that they were “the same” as the stories in
the Bible (2005: 451). Busse later describes how this wandering hero was the source
of both white people and local Papuans (and, by extension, black people everywhere),
all deriving from a common source. He was told,
[W]hen Nggiwe first came to the southern lowlands, his skin was white. He went part
way up the Maro River and then headed east to Suki Lagoon. From there, he travelled
up the Fly River to Kiunga, went to the Digul River, then down the Digul to Hong
Kong, Singapore, Australia, and the United States, where he gave “the good life” and
“all the things” to “white people” in those places. Later, Nggiwe returned to the southern lowland, and it was during his second visit that he formed the ancestors of the people
who now live there. (2005: 453)
In local imaginings, the wandering hero is the source of both white and black people, both deriving from this unity. His wanderings also led to all the valued items and
conventions that are now possessed by white people, but that, the narrative suggests,
stem from a single landscape and set of places. In this sense, the modern Papuan hero
tales described by Busse (2005) resemble many so-called cargo myths from other
parts of PNG and Melanesia, which also connect distant global landscapes and black
and white peoples, goods, religious narratives, and forms of knowledge to a single
source usually located in or near the places where the myths originated (e.g., Bashkow
2006; Burridge 1995; Jorgensen 2006; Kirsch 2006; Lawrence 1971; Trompf 1990,
1994; Worsley 1970).
We now want to introduce a correlative way of understanding the distinction
between Universal History and Universal Past that we have proposed. These differing apprehensions of the past – Western and Melanesian – are also associated with
distinct conceptions of the person; conceptions that intimate understandings of the
past that we have just described. On the one hand, modern, Western historical consciousness has a distinct preoccupation with individuals and individuality, whether of
“epochs, regions or persons” (Burke 2002: 22; cf. Ouroussoff 1993). The proliferation of histories and versions of the past about an individual, Napoleon for instance,
does not negate alternative perspectives on that person. Rather, there is the presupposition evident in historical representation as much as in orthodox views more generally held that each individual possesses her or his own unique point of view.
Everyone, in short, “is the originator of a singular view, and knowledge of this fact
allows one to be added to another without expectation of closure or summation”
(Strathern 2005: 158). In other words, an account of the past that builds on the
work of others, or which is jointly authored, can be said to have multiple origins, but
the “multiplicity comes from the way persons are added to one another’s enterprises”
(Strathern 2005: 161). The perceived singularity of “events” has a similar additive
quality in the way, it is imagined, they are tallied together to produce a particular
narrative form; events in a progression, one following – being added to – the next.
The situation in Melanesia as we have described it is different. Melanesian ideas of
the past do not tend to take separation and difference as a starting point and do not
suggest an emphasis on individuality. Rather, they relate differentiation to an original
source of unity that provides the basis and moral compulsion for complementary
9781444333237_4_016.indd 284
8/7/2010 7:10:57 PM
285
engagement or for the obviation of present inequalities (see below). Multiplicity
arises in a different manner: it “comes from the way people divide themselves from
one another. Singularity (individuality) is an outcome not an origin” (Strathern
2005: 161). The Papua Hero Tales portray an originary figure and his journey from
which all else in the world derives through unique movements and acts of detachment. This corresponds to the person that is neither simply singular nor plural
(Strathern 1991).
The issue we want to raise, then, is that from a Western perspective informed by the
presuppositions of Universal History, the key unit of understandings of the past,
including that of history, is the event. Narratives told, or yet to be told, are formed
by what are taken to be discrete events of varied durations. The delineation of events
is what is seen to enable history to be written, even if, as it is argued, an event is an
abstraction from an explicit or implicit narrative or code (Mink 1987: 200; White
1987: 33–34; 1999: 66–67). As Lévi-Strauss (1966 [1962]: 257) notes,
What makes history possible is that a sub-set of events is found, for a given period, to
have approximately the same significance for a contingent of individuals who have not
necessarily experienced the events and may even consider them at an interval of several
centuries.
From a Melanesian perspective, as we describe it here, the key unit of the past is not
the individual, segregated event but the “fractal” person – the person as an image
composed of, and enchained with, other persons, all ultimately deriving from one
source from which separation ensued (Wagner 1991). This is exemplified by the
Melanesian emphasis on naming that is used indigenously to “speak of, order and
conceptualise existence as identity” (Wagner 1991: 163). Whether used for personal
or collective designations, names are just names. However, as a Daribi friend told
Wagner (1991), “When you see a man, he is small; when you say his name he is big.”
Narrative and person are not conceived as separate and distinct in the Melanesian
region. Rather, persons are formed by talk – myths, names, stories – just as narratives
are formed by persons through their constitutive acts. This is what Wagner (1991:
166) refers to as the “fractality of the Melanesian person: the talk formed through the
person that is the person formed through the talk.”
With these contrasts in mind, we can return to our opening example of Clauser’s
movements into the Udabe Valley. The Fuyuge who encountered Clauser are recalled
by later Fuyuge generations, and it is these latter generations who narrated to Hirsch
the names of the people Clauser met and the places to which he ventured. This is
knowledge that derives from the remembered talk (av) of the aked inoge – the great
ancestors of the living. The mundane Fuyuge recollections of Clauser’s movements in
the upper Udabe Valley were not the only way he is recalled. There is also a narrative
form – known locally as tidibe – in which Clauser figures centrally. Tidibe has the
characteristics of what is commonly called myth; it is simultaneously a creative force
that not only is of the past, but also inheres in the present. Tidibe is presupposed in
everything that Fuyuge know, the world they inhabit and that beyond their perceptual
boundaries. Tidibe is associated with a particular local landscape and set of places as
well as all possible landscapes, persons, and places. Black people and white people and
all the knowledge and things they possess derive from this source. There are stories
and figures in these stories that account for how these separations led to current divisions in the world.
9781444333237_4_016.indd 285
8/7/2010 7:10:57 PM
286
Hirsch and Moretti
Tidibe is an original unity or whole from which all is differentiated. It is concurrently
one and many: one past and many pasts – one person and many persons. Tidibe is comparable to the “fractal person” mentioned above in being “whole and part at once.”
For the Fuyuge to know about the people and places that Clauser met during his
time in the Udabe is to explicitly or implicitly know the particular tidibe that contains
him – to which we now turn – as much as to know tidibe in general as a unity, as one.
The particular tidibe containing Clauser accounts for how the missionaries (and
government) came into Fuyuge lands, and especially the upper Udabe Valley. A key
agent in various versions is the tidibe figure known as Solusia.3 Different versions
begin in the following manner:
Solusia caught a “cuscus” [marsupial] but he found that no matter how he tried to cook
it, the “[c]uscus” remained uncooked;4 finally, in despair, Solusia threw the [c]uscus in a
river. The animal floated in the river until it came to the coast.
In different versions the “cuscus” is found by another tidibe figure (but with different
names, depending on the versions), and this figure comes into the mountains in search
of Solusia. After asking repeatedly for Solusia (again, portrayed differently depending
on the version), he is finally found.
In the version Hirsch was told, Solusia5 and the figure named Ulgio come to
Ononge and then Visi in the upper Udabe Valley (where Hirsch was based during
fieldwork). At Visi they find the “base” man of Visi, the man from whom others imagine they derive, and his two sons. Solusia and Ulgio then go to the coast and tell
Father Clauser6 to go to Visi and meet the men they saw there. Clauser travels to Visi
and subsequently “puts the law into the heads” of these men by breaking a spear over
the head of one of them. The version Hirsch was told in Visi ends as follows:
Later with other Fathers he built houses of reeds all down the valley. Today, [the missionaries] live in permanent buildings [i.e., they have returned and settled here forever].
Each version, in its distinctive way, commences with the nonpresence of either government or missionary. The “cuscus” that would be conventionally cooked, divided,
and shared with others (and not eaten by the hunter) is unable to be transformed
in this manner. What it does enable is the subsequent incorporation of missionary and
government into the social lives of the Fuyuge and the bringing of “law.” This is
achieved through a series of substitutions or displacements, and in the process the
limits of Fuyuge social conventions are revealed and obviated by the modes of conduct brought by these outside agents and their local appropriation. At first glance, this
may suggest that these narratives take separation and difference between Fuyuge and
white missionaries as the “original state” or “starting point” to be “historically” mediated. But while these particular Clauser tidibe tales seem to portray an original or first
connection between the Fuyuge and white missionaries and government, this is in fact
a reconnection. We say this because there are other tidibe, noted above, that describe
how white people and black people derive from an original pair of brothers who
became divided and separated in the ancestral past. The multiple versions of the past
associated with the coming of the missionaries and Clauser, in particular, cannot be
understood outside of the background of the “Universal Past” specified by tidibe. But
tidibe makes no sense unless it is understood against the background of the multiple
narratives (including those which are also known as tidibe) of which it is a part.
9781444333237_4_016.indd 286
8/7/2010 7:10:57 PM
287
An important outcome of this figure–ground association between specific local
narratives about past colonial “contacts” and the “Universal Past” of origin tidibe is
that, from the Fuyuge perspective, whites appear not as unconnected strangers but as
long-lost siblings, and the Christian “law” and associated “transformations” they
bring along represent something that in a sense had always been part of the local
landscape, from which it had originated, and to which the Fuyuge had been entitled
all along. Further, whereas the mission’s own accounts portray Clauser as the “culture
hero” whose agency brings a new law and, ultimately, salvation to the Fuyuge, local
narratives about his journeys suggest that he, like other important tidibe figures, was
enticed to travel to the area from the coast by the actions of local persons (i.e., their
visits to the coast and the throwing of the cuscus in the river).
An Australian “Hero Tale” and Its PNG Counterpart
With this in mind, we now turn to a different example from a Fuyuge neighboring
area that has been the site of nearly uninterrupted intensive gold mining by foreign
companies and independent expatriate and indigenous miners since the mid-1920s
(Halvaksz 2008; Moretti 2006, 2007). The locale is at the fringe of the Anga (formerly Kukukuku) ethnolinguistic area, a region spanning across the three PNG provinces of Gulf, Eastern Highlands, and Morobe that is inhabited by a collection of
people who speak 13 related languages (Gordon 2005).
By the 1920s–1930s, the Anga had been involved in various raids against populations already under colonial control and “protection” and in violent confrontations
with white miners operating in the Lakekamu region on the Papua side and the Morobe
Goldfields in New Guinea. Furthermore, they posed a challenge to prospectors who
wished to penetrate deeper into Angaland to find whether its core contained the same
kind of riches already discovered at its margins (see McCarthy 1963; Nelson 1976).
In 1933, the New Guinea government decided to establish a patrol post at the
heart of the Anga region for the double purpose of finding out if it was suitable for
mining development and to facilitate the “pacification” of its peoples. Patrol Officer
J. K. McCarthy was chosen for this mission and set out on a patrol from the Lower
Watut River with 12 police and 24 Waria carriers. His main task was to locate a piece
of flat land in central Angaland that could host an airstrip for the patrol post. He did
so in a place called Menyamya. Accounts of this patrol and those connected with it are
found in his popular book Patrol Into Yesterday (McCarthy 1963) and in unpublished
patrol reports he penned at the time. Another version is found in Simpson’s book
Adam With Arrows (Simpson 1954), which contains a detailed interview about this
patrol, conducted by Simpson. The following is gleaned from McCarthy (1963:
90–113, 116–26) and Simpson (1954: 40–57).
After finding Menyamya, McCarthy’s patrol began its return trip. About one week
away from Menyamya, they passed through the hamlets of Kobakini at the headwaters
of the Kapau River. According to McCarthy, people there looked nervous and seemed
very keen for the patrol to move on quickly. He also noticed that some local men carried steel axes and wore various items of European clothing. Two men also sported
stone clubs that had shotgun barrels for handles. He suspected that these items had
been stolen from Europeans, and the following day this was confirmed by the discovery of the decomposed bodies of two white men and several Papua New Guineans
9781444333237_4_016.indd 287
8/7/2010 7:10:57 PM
288
Hirsch and Moretti
along with some European belongings. McCarthy later learned that they were two
prospectors named Naylor and Clarius and their carriers. McCarthy decided that it
would be impossible to properly “investigate” the murders to find who was actually
responsible because “these primitive peoples had no conception of the white man’s
law” (McCarthy 1963: 107). Instead, he resolved to capture some local men, notably
the two sporting the shotgun barrel stone clubs, to be taken to the coastal city of
Salamaua as prisoners with the idea that “after a suitable indoctrination period they
could be returned to their homes to spread the gospel [i.e., the new colonial law]”
(McCarthy 1963: 107). He captured eight men, but later his patrol was ambushed by
a group of warriors who had followed them from Kobakini. One of the prisoners was
accidentally killed by the attackers, and McCarthy and some of his police were seriously injured. Eventually, they made it back to the Otibanda post in the Anga-inhabited
Upper Watut area of the Wau-Bulolo District of Morobe Province (i.e., the Morobe
Goldfields), from where they were taken to Salamaua for medical treatment. Four
months later, McCarthy was ordered to make another patrol into the area and eventually established a post and airstrip at Menyamya. On the way there he returned to
Kobakini, where another patrol officer had since opened a fortified post. McCarthy
was surprised to find that the people there were happy to trade with him and did not
seem totally hostile. Indeed, he even remarked that “nobody among them was very
interested when we told them the prisoners we had taken to Salamaua were alive and
well and would be returned to the villages in a short time” (McCarthy 1963: 117).
In the following decades, McCarthy was appointed M.B.E. (1943) and C.B.E.
(1965), rose to the rank of lieutenant colonel during the Second World War, and
eventually became director of native affairs and deputy speaker of the elected House
of Assembly of Papua New Guinea. His many exploits as an explorer, administrator,
soldier, and writer, including the Menyamya patrol popularized in the aforementioned
accounts, gained him a “heroic status in Australian history” as shown, for instance, by
his entry in the Australian Dictionary of Biography (Nelson 2000), a resource with
the stated mission of providing “scholarly biographies of persons who were significant
in Australian history” (see also Australian Dictionary of Biography n.d.).
As already noted by others, the written accounts of McCarthy’s Menyamya patrols,
and especially that of Simpson, are “rich in the legend-making material of the Australian
penetration of New Guinea” (Watson 1956: 1163). In fact, writing in the 1950s,
when the process of Australian colonial expansion was still ongoing in Papua New
Guinea, the anthropologist Watson (1956: 1163–4) suggested that Simpson’s “book
itself is a part of that very active cultural process, for it already is referred to for authority in many stations and outposts of the Territory.” In other words, Simpson was telling (and promoting) the untold story of colonial expansion and pacification into this
“unknown” world and describing the landscapes and peoples that were being refashioned by “civilisation and law.” Here was a particular history against the background
of Universal History.
But McCarthy’s own published accounts of the Menyamya patrols were also
informed by this same figure–ground relation. As the above citations suggest, for
instance, his book framed the Kobakini arrests, not as an act of white retribution or
subjugation but as a necessary step in a process which, not unlike Clauser’s beating of
the (Christian) law into Visi heads, would transform the Anga from “primitive peoples [who] had no conception of the white man’s law” (and therefore “knew no better”) into “more civilised” and “law conscious” ones (McCarthy 1963: 107). This
transformation was seen to necessitate the physical removal and encompassment of a
9781444333237_4_016.indd 288
8/7/2010 7:10:57 PM
289
few of them into the white man’s world, where they could be properly “indoctrinated” and then sent back to their home communities to, as McCarthy puts it, “spread
the gospel” of the new colonial law and order (1963). Far from being fortuitous, the
evangelical metaphor McCarthy used to describe the rationale for his arrests underlines the shared basis of the aforementioned “Universal Historical” expansive conventions found in mission and government narratives of material and spiritual “progress”
and salvation from colonial Papua New Guinea.
At this point, it is also useful to note that an essential element of these conventions was
the trope of colonization as “time travel.” By this we mean that Western colonial narratives about PNG portrayed specific “missions,” such as McCarthy’s Menyamya patrol, as
expressions of a wider “civilising mission” that brought colonial agents into a temporally
as well as spatially distant, “still primitive world” so that they could eventually make – to
borrow the anthropologist Beatrice Blackwood’s (1939, 1950) revealingly oxymoronic
branding of the Anga people of the Upper Watut – the “modern stone-age” peoples of
Papua New Guinea truly coeval (Fabian 2002 [1983]) with the “properly modern”
West. Indeed, this trope of time travel is evident from the very titles of the historical
narratives penned by McCarthy (Patrol Into Yesterday [1963]), Simpson (Adam With
Arrows [1954]) and others, such as the explorer and prospector Michael J. Leahy, who
in 1937 published a memoir of his PNG adventures (including a period spent in the
Morobe Goldfields) entitled The Land That Time Forgot (Leahy and Crain 1937).
From a political point of view, this “time travel trope” had the effect of justifying
colonial domination in two apparently contradictory but mutually reinforcing ways.
On the one hand, the “backwardness” of Papua New Guineans was invoked as proof
that only the “truly modern” whites could propel PNG into the present and that, as
such, the “civilising process” of colonization was nothing short of a “moral duty.”
But on the other, the associated Western notions that the past is a far-stretching and
unchangeable realm of actuality separate from the present and that history can only
progress according to a single “plot” – or, in other words, through a given series of
steps – promoted the belief that, despite the best efforts of its colonizers, Papua New
Guinea could not possibly be expected and expect to accomplish in a few years or
decades what had taken Westerners centuries or even millennia to achieve. In other
words, this second aspect justified enduring domination by portraying the “civilising
process” of colonialism as a necessarily protracted enterprise and by asserting that
current inequalities between colonizers and colonized were merely the outcome of an
indelible series of past events that could be corrected only in some more or less
distant future.
But McCarthy’s encounters in Kobakini also figure in a different narrative form
arising in the Anga area itself. What we recount here is a tumbuna stori (ancestral
story) that Moretti recorded during a recent period of fieldwork (2004–5) in the Mt.
Kaindi area of the Wau-Bulolo District of Morobe. Its narrator was Nakoeo (a pseudonym), a local gold miner whose family had come to Mt. Kaindi from a HamtaiAnga7 village in the neighboring Aseki area of Morobe. Nakoeo’s story is meant to
have a particular effect because he intended to tell it in some court one day in order
to make a claim for compensation against the Australian and/or Papua New Guinea
governments. His immediate claim is that McCarthy arrested one of his maternal
grandfathers, who never returned alive to his home, and he wants compensation for
this loss. But as we shall discuss momentarily, this incident is inextricably connected
to wider and more longstanding claims that stretch back all the way to the mythical
time of creation. The following is a condensed version of the story told.
9781444333237_4_016.indd 289
8/7/2010 7:10:57 PM
290
Hirsch and Moretti
Nakoeo claimed that two of his maternal grandfathers were from Kobakini. They were local
“chiefs,” as he called them using the English word, and they possessed great magic power
that enabled them to become invisible, change shape, quickly travel over vast distances and
so on. One day the two men saw that a local river had flooded and that the rushing water
had cut the surrounding trees very sharply. Hamtai ancestral tales told of Niyantona (meaning “the metamorphosis”) – a place at the heart of Anga country where all the peoples of
the world were said to have originated. The place had once contained a water pool (e’ä
pnga) where all black and white people had emerged from the transformed remains of a
single murdered ancestor. According to these myths, the whites had to eventually depart
the area. Various reasons were given for this, the most common being that they had stolen
a powerful “fire” that belonged to the “culture hero” Akheänqa. This “fire” was said to be
the source of all the “knowledge,” “cargo,” and “power” that whites now hold and most
versions of the myths stated that Akheänqa had originally intended it for the Anga.
According to Nakoeo, his grandfathers thought that the cut trees were signs that whites had
now returned with their “stolen/borrowed” power and had settled near the mouth of the
local river. Keeping this secret from other people in their village, the two used their magic
to transform into spirits and travel down the river and on to Salamaua, a natural harbor that
was the major port town and administrative center servicing the Morobe Goldfields in
colonial times. There the grandfathers saw the whiteman’s houses, all of which had iron
roofs. By looking through the windows, they saw white men and women sitting around
tables surrounded by many steel objects such as knives and axes. According to Nakoeo, his
grandfathers then realized that the way the trees were cut by their local river had foretold
the arrival of these sharp new tools. After summoning a third Kobakini man who had the
power to “cool” and neutralize dangerous magic, the grandfathers entered one of the
houses through the iron roof, took several steel axes and blades, and returned home.
The three men thought they had gotten away unseen but, thanks to their own “shamanlike” vision (hingo hivä’u), the whites had perceived “little men in bark capes, grass skirts
and yellow bandoliers” (the “traditional” Anga attire) walking around the city, just as if
they had been there in material form. As a result, they knew that the objects missing from
one of their homes had been taken by these visitors and departed in search of them.
Back in the village, the Kobakini people used the sharp steel axes obtained by the grandfathers to cut trees quickly and clear large tracts of land for gardening. Meanwhile, Makarti
[McCarthy] was chosen by his people to lead a patrol to find the men who had come to
Salamaua and was now coming from the coast. On approaching Kobakini he looked through
his binoculars and saw the local men cutting trees with steel tools. According to Nakoeo,
the villagers too spotted the patrol and, unaware of the grandfathers’ “theft,” started to
wonder what they could have possibly done to bring them up (see Strathern 1990).
The grandfathers were not afraid and used their magic to rob the whitemen of their
strength. They killed two of them but Markarti escaped and later returned with more
police. When he came back the villagers were so busy gardening with the new tools that
Makarti caught them by surprise. Many were captured, including one of the grandfathers, and they were taken to Salamaua.
At the end of his story, Nakoeo asserted that the name Kukukuku, which whites had
given to his people in colonial times, had actually come from the name of the grandfather captured by Makarti.8 In turn, he said, this showed that his grandfather had been
the first Anga person to contact and get the attention of the returned white people.
Moreover, it showed that whites had considered him a very powerful and important
person, which is why they never returned him to his people and why the Australian and/
or PNG governments should now pay Nakoeo “many millions” in compensation for his
arrest and failed return.
Nakoeo’s story frames the first encounter between Hamtai (represented by the grandfather, whose name identifies as a personification of “the Kukukuku” as a whole) and
white colonizers (represented by their “chosen champion,” Makarti) through narrative
9781444333237_4_016.indd 290
8/7/2010 7:10:57 PM
291
conventions traditionally associated with shamanic exploits. In particular, the powers of
the three Kobakini men and those of the whites they encounter, such as the capacity to
see far-away places and persons, perceive the ordinarily invisible, set and neutralize magic
traps (i.e., the dangerous iron roof), and quickly travel vast distances, are those commonly attributed to shamans, sorcerers, and ancestral or place spirits. The grandfathers’
journey to the coast is also depicted in terms analogous to those associated with the
disembodied journeys that shamans undertake into the invisible world of spirits to
acquire important knowledge, retrieve lost objects, or recapture the stolen souls of people made sick by sorcerers or angry spirits. For example, during such journeys shamans
were said to use “cooling magic” like that described in the story to make a safe passage
in the roofs of houses belonging to ancestral or place spirits (see Herdt 1977).
But like the aforementioned Fuyuge narratives, Nakoeo’s tale is also set against the
background of myths (or Akheänqa stori) about the common origins of Anga and white
peoples and of all the knowledge, power, and things the latter hold. As in the Fuyuge
case, there too Hamtai and whites are portrayed as “brothers” that were once united,
then divided, and now potentially reconnected. Against the background of this Universal
Past, Makarti and the grandfathers – and especially the one whose name stands for the
Anga people as a whole – are “fractal persons” who represent whites and Anga respectively whilst simultaneously being themselves part of an original personified whole: the
single ancestor whose remains transformed into all the peoples of the world.
Similarly, the ways in which Makarti and the grandfathers act toward each other are
motivated by, and simultaneously encapsulate, make apparent, and expand upon, those
of the white and Anga ancestors of the Akheänqa stori. Thus Nakoeo told Moretti that
the white people his grandfathers encountered on the coast had not come to New
Guinea to discover new lands or to look for gold but to locate and reconnect with the
Hamtai people themselves, whom they knew to have originated from the same pool
and ancestral body as themselves and thus to be the “long-lost siblings” their own
ancestors had told them about. In other words, the colonial penetration of PNG was
a reversal of the outward journey that the first white ancestors had made soon after
creation. Similarly, the grandfather’s “theft” of some steel tools was only an attempted
redress of the primordial appropriation by white people of “white things” that had
always (also) been intended for the Anga. Equally, Makarti’s arrest of Nakoeo’s never
returned grandfather, whose only “crime” had been to take back what had been due
to him and his people all along, was not just a reaction (or overreaction) to the grandfathers’ exploits but also the replication of the initial white (mis)appropriation of
Akheänqa’s powerful “fire” at creation. As such, the characters of that story and their
interactions also embody and symbolize the obligations that whites and Hamtai have,
but have hitherto failed to meet, toward each other as “kin.” And in this sense,
Nakoeo’s “stolen grandfather” is a personification not just of the narrator’s perceived
claims against the Australian or PNG governments, but also of the rights to equality
and “recognition” (Robbins 2003) that all Anga feel to have vis-à-vis whites.
“Partial Wholes”
Both Western and Melanesian narratives about Clauser and McCarthy entail a “holistic approach” to the past that rests on a figure–ground relation between those specific
accounts and what we defined respectively as “Universal History” (after Jay 1984;
Mink 1987) and “Universal Past.” As we tried to show, however, the kinds of wholes
9781444333237_4_016.indd 291
8/7/2010 7:10:57 PM
292
Hirsch and Moretti
and part–whole configurations that they delineate are different in some important
respects. And while we do not mean to argue that this is their only rationale and outcome or that they are necessarily intended to do so, we wish to suggest that these
differences reflect and justify or challenge enduring power inequalities between
Westerners and Melanesians.
A significant theme in both the Fuyuge narrative involving Clauser and the Anga
narrative about McCarthy is the connection between mountains, as the place of blackmen, and the coast and beyond, as that of whitemen. Each is united through an
ambiguous act, a killing that is not a killing in the Fuyuge case (the cuscus that cries
out and is thrown into the water), or a theft that is not a theft in that of the Anga (the
reappropriation of “white” objects that were always intended for them). In the
substitutions and displacements that are enacted in the narratives, a certain kind of
sense is made obvious, that is, that people now know things of whitemen that they
possessed in the past but that were lost or stolen, and that the conditions are potentially in place for a permanent reconnection between whites and Papua New Guineans,
who share a common origin.
Wagner (1986: 80) suggests, “What is made problematic here is time … as in the
evanescent condensation of a world of meanings into the moment of its realization.”
What kind of “now” or “presence of time” is engendered in these Melanesian and
Western narratives? Clearly they all derive from different chronological times (i.e., the
early twentieth century, late twentieth century, or current century), but even with
these differences it is still possible to discern enduring power relations and inequities.
The Melanesian instances reveal people who imagine themselves as both the source
and rightful heirs (or at least equal co-heirs) of the world and simultaneously displaced and possibly marginal from this power. The division of black and white peoples
from an original base of unity is one that requires and indeed demands obviation.
There are no untold stories, only stories that are made possible by, or derive from, the
conditions of a Universal Past.
In the Western narratives, by contrast, we glimpse the presence of time through the
tropes of “progress” or “civilisation and law.” This is an expansive, cumulative process that transforms unknown lands and people – unknown stories – into new narratives that often naturalize and legitimize long-term inequality and domination. As
recently argued by Kirsch (2006: 140–3, 155–6, 217), then, Western historical narratives about the colonization of Papua New Guinea, such as the above narratives
by and about Clauser and McCarthy, take as a starting point and emphasize separation and difference between Westerners and Melanesians and predicate their resolution on a long but inevitable process of hierarchical encompassment of the latter by
the former.
Rather than being confined to mission or secular narratives from the colonial era,
this pattern is also evident in more recent scholarly narratives that continue to be
shaped by the holistic conventions of “Universal History.” A good illustration of this
is outlined in Errington and Gewertz’s (2004) recent critique of the bestselling,
Pulitzer Prize-winning book Guns, Germs and Steel (2005 [1997]) by Jared
Diamond.
Diamond states that he wrote the book as an answer to Yali (the “cargo cult”
leader made famous by Peter Lawrence [1971]) who, during a conversation with
Diamond back in 1972, asked him, “[W]hy is it that you white people developed so
much cargo and brought it to New Guinea, but we black people had little cargo of
our own?” (Diamond 2005 [1997]: 14, in Errington and Gewertz 2004: 7). The
9781444333237_4_016.indd 292
8/7/2010 7:10:57 PM
293
book offers an account of how, over some 13 millennia, people differentiated and
spread across the globe. Although all were equally intelligent and capable, some
found themselves in geographical conditions that allowed them “to have the power
of guns, germs, and steel on their side” (2004: 7), thus increasing their wealth and
power and projecting them on a trajectory of conquest and subjugation of which
colonialism is but the latest chapter. For Errington and Gewertz, this kind of “universal history” has the advantage of challenging the idea that some people were
inherently less able to achieve “development” than others. Under the right geographic and biogeographic conditions, Diamond stresses, Papua New Guineans
would have developed the same “cargo” as Europeans (in Errington and Gewertz
2004: 9). But Errington and Gewertz also point out that the book was so well
received by Westerners because it reflects their own ingrained assumptions about
history and because it unwittingly justifies and naturalizes current inequalities by
painting them as the inevitable outcome of a random but cumulative chain of causations that could not have been otherwise (2004: 9–10, 44–6). For Errington and
Gewertz, this historical narrative supports the idea that if anyone else had achieved
the power of “guns, germs and steel,” they would have behaved exactly like Westerners
did and also subjugated others, created a colonial empire, and continued to enforce
a system of inequalities in the postcolonial world (2004: 11–12). What is more, for
Errington and Gewertz this narrative also implies that, as it took Europeans millennia to get where they are, Papua New Guineans cannot expect to get there too
immediately, and Europeans have no obligation to wholly and promptly share what
they have with them.
In contrast with this, they suggest that Papua New Guinean narratives portray a
situation where Europeans and local people had not just encountered each other in
colonial times but also shared a prior history (what we called a common, “Universal
Past”), which, although it could vary in detail, always implied that they should treat
each other as equals in the here and now (2004: 26). Analogous conclusions were
reached by many other anthropologists whose works have shown two things. The first
is that the idea of a primordial unity between Westerner and Melanesian landscapes,
persons, material possessions, beliefs, and so on and the often concurrent characterization of Europeans as long-lost brothers, cross-cousins, parents, mother’s brothers, or
other kinds of kin are very common throughout Papua New Guinea and Melanesia.
The second is that these ideas constitute an indigenous critique of enduring power
inequalities and a model for the kind of reciprocal moral relationships that Melanesians
would like and expect to have with Westerners (e.g., Bashkow 2006; Burridge 1995;
Busse 2005; Kirsch 2006; Lawrence 1971; Trompf 1990; Worsley 1970). In this light,
the holistic narrative conventions of what we have herein defined as the “Universal
Past” seem to provide a common model for how Melanesians approach the past as a
source of critique against past and present inequalities. And as Sahlins’s chapter
(Chapter 7, this volume) shows at length through ethnographic and historical examples from Africa, East Asia, and Oceania, certain aspects of these “Melanesian” conventions about the past, and most especially their basis for claims of common descent
between what the anthropologist would consider local and outside agents, are in fact
not peculiar to Melanesia. Rather, they constitute a specific instantiation of what he
describes as a broader “politics of alterity” in which indigenous peoples “the world
over” engage to attempt the holistic appropriation and transformation of powerful and
potentially dangerous Others into local relations, allies, and protectors (Sahlins,
Chapter 7, this volume).
9781444333237_4_016.indd 293
8/7/2010 7:10:57 PM
294
Hirsch and Moretti
Conclusion
In our discussion of Western and local narratives about Clauser and McCarthy, we
have contrasted what we defined as “Western” and “Melanesian” modes of imagining
the past and their relation to present and future. In so doing, we have of course engaged
in a heuristic thought experiment that has inevitably homogenized and simplified what
are much more diverse and highly complex realities on the ground. For example, not
all Melanesian narratives about the past conform to our model of the “Universal Past.”
To name just two recent examples, Knauft (2002a, 2002b, 2002c) and Robbins (2004,
2005a, 2005b) have provided case studies of Melanesian communities whose currently
dominant approaches to the past appear to fit more closely with what we defined as
“Universal History.” In other words, these peoples appear to emphasize difference
between local and outside actors, institutions, behaviors, and values and the need to
break with the “indigenous past” and exchange it with a “Western future.” And in the
case documented by Knauft in particular, the Gebusi seem to frame this need for rupture and transformation according to Western ideas of “progress.”
What is interesting to note, however, is that Knauft and Robbins themselves interpret this as the outcome of the local internalization of the “means of cultural debasement” (Sahlins 1992: 24; Robbins 2004: 9) associated with colonization,
missionization, and postcolonial exploitation, including those stigmatizing discourses
about material, social, moral, and spiritual “backwardness” and “temporal behindness” that we have suggested to be connected to certain aspects or versions of
“Universal History.” What is more, these two cases seem to us quite extreme. In most
Melanesian contexts, one may well find narratives emphasizing difference and separation between local actors and Westerners and advocating a total rupture with a local
past to be exchanged for a new “Western-like” or global future. Nevertheless, these
are usually connected to, and thus complemented and potentially obviated by, other
narratives about a common, Universal Past that makes the two – to borrow Busse’s
(2005: 451) informants’ observations about local and biblical persons, landscapes,
and stories – “the same” and ultimately equal (see, among others, Bashkow 2006;
Burridge 1995; Kirsch 2006; Lawrence 1971; Trompf 1990).
As the editors point out in the introduction to this volume, the concepts of social,
cultural, political, and human “holism” have carried important moral and political
implications in anthropology and beyond. In Western thought, the concept of the
whole has been intrinsically connected to that of the individual. This connection has
also been foundational to much anthropology, where the whole has been primarily
associated with “the other” and deployed to criticize Western ideas about the individual self. But Strathern (1988) and others have long highlighted the dangers associated with the imposition of Western notions of and preoccupations about wholes such
as “society” and their relations to “the individual” on Melanesians. Instead, they have
drawn on Melanesian perspectives to provide alternative models that could both help
to foreground our own conceptions of the whole and of the person and offer novel
comparative critiques of them.
In this chapter, we have pursued this kind of strategy in relation to a type of “whole”
that is as central to Western thought and practice (including anthropology) as those
of “society,” “culture,” or “human nature,” but that has been subjected to much less
attention and comparative critique, at least by anthropologists. This is the notion of
9781444333237_4_016.indd 294
8/7/2010 7:10:57 PM
295
“history as a whole” entailed in “Universal History.” Drawing on a model of
Melanesian narrative conventions about the past, we sought to bring into critical relief
this largely implicit Western notion. In so doing, we have argued that “Universal
Past” and “Universal History” are both “holistic” modes of narrating the past and its
relation to present and future that nonetheless entail different modes of imagining
wholes and part–whole relations.
Just as the editors opened this volume by pointing out the importance of thoroughly grasping and engaging with the moral and political agendas and critical
projects implied in different Western conceptions of the social, political, and cultural
whole and their projection onto others (e.g., as a critique of Western liberal and neoliberal individualism or as a legitimizing force for cultural or social fundamentalism),
so we sought to highlight the asymmetry between Universal History and Universal
Past. It is an asymmetry that informs current realities in Melanesia as much as our
attempts to understand those realities through the use of ethnography and history.
Thus, if it is the case that all comparisons and perspectives are, as Busse (2005:
457–8, drawing on Strathern 1991) recently suggested, inevitably “partial” (in the
sense of “interested”), then our “partial” comparison aimed to highlight the fact that
Universal History and Universal Past are also intrinsically “partial wholes,” whose
holistic narrative conventions about the past embody and promote important “interested” (political and moral) claims and counterclaims that deserve close anthropological attention.
Acknowledgments
Hirsch and Moretti would like to thank Ton Otto and Nils Bubandt for the invitation to contribute to the Sandbjerg Megaseminar and to seminar participants for their stimulating questions and comments on a draft of this chapter. In revising it for this volume, we have benefited
from the critical input of the editors as well as that from Charles Stewart. Moretti is grateful to
the British Academy and the ESRC for funding his part of this research. The authors remain
solely responsible for all errors of fact and interpretation that may be found in this chapter.
Notes
1 Hirsch has drawn on this example in previous publications and for different comparative
purposes, most recently in Hirsch (2008).
2 These effects can be varied and overlapping. For example, a myth, ancestral story, or other
such tale may be told to promote physical growth in young initiates. It may be narrated to
make a moral point and induce others to change their behavior accordingly (Kirsch 2006:
151–4). Or it may be recounted to induce another person or a mining company to recognize
one’s rightful claims over a piece of land or property and perhaps obtain control over and/or
compensation for their exploitation (Filer and Macintyre 2006: 225).
3 When Hirsch was told this tidibe during the mid-1980s, the narrators referred to this figure
sometimes as Sol Si and other times as Solusia or Solusi, reflecting differences in the way the
name was pronounced and, perhaps, differences in the way Hirsch heard (and recorded) the
pronunciations of the name.
4 In the later version Hirsch was told that the animal cried out as if it were still alive, although
it had its intestines cut out and fur removed.
5 Elsewhere Hirsch (1999: 815n19) refers to Solusia as Sol Si, based on a version he was given.
9781444333237_4_016.indd 295
8/7/2010 7:10:57 PM
296
Hirsch and Moretti
6 His name is known as Cross or Clossi (Klossi) in the local Fuyuge dialects.
7 Hamtai is the largest language group in the Anga ethnolinguistic family and spans across the
Gulf Province of Papua and the Morobe Province of New Guinea (see Blackwood 1939,
1950; Bamford 2007; Gordon 2005).
8 We do not report the name, as Nakoeo may not want it publicized in case others should
appropriate it to “court the government” themselves.
References
Annales de Notre Dame du Sacré Coeur (ANDSC). (1866–present). Issoudun, France:
ANDSC.
Australian Dictionary of Biography (N.d.). Australian Dictionary of Biography Online. http://
adbonline.anu.edu.au/adbonline.htm (accessed May 26, 2009).
Bamford, S. (2007). Biology Unmoored: Melanesian Reflections on Life and Biotechnology.
Bashkow, I. (2006). The Meaning of Whitemen: Race and Modernity in the Orokaiva Cultural
World. Chicago: University of Chicago Press.
Battaglia, D. (1995). On Practical Nostalgia: Self-Prospecting Among Urban Trobrianders. In
Rhetorics of Self-Making, ed. D. Battaglia. Berkeley: University of California Press, 77–96.
Blackwood, B. (1939). Folk-Stories of a Stone Age People in New Guinea. Folklore 50:209–42.
Blackwood, B. (1950). The Technology of a Stone Age People in New Guinea. Oxford: Oxford
University Press.
Burke, P. (2002). Western Historical Thinking in a Global Perspective: 10 Theses. In Western
Historical Thinking: An Intercultural Debate, ed. J. Rüsen. Oxford: Berghahn, 15–30.
Burridge, K. (1995). Mambu: A Melanesian Millennium. Princeton: Princeton University Press.
Burrow, J. (2007). A History of Histories: Epics, Chronicles, Romances and Inquiries From
Herodotus and Thucydides to the Twentieth Century. London: Allen Lane.
Busse, M. (2005). Wandering Hero Stories in the Southern Lowlands of New Guinea: Culture
Areas, Comparison, and History. Cultural Anthropology 20:443–73.
Cronon, W. (1992). A Place for Stories: Nature, History, and Narrative. Journal of American
History 78:1347–76.
Delbos, G. (1985). The Mustard Seed: From French Mission to a Papuan Church. Port Moresby:
Institute of Papua New Guinea Studies.
Diamond, J. (2005). Guns, Germs and Steel: A Short History of Everybody for the Last 13,000
Years [1997]. London: Vintage.
Errington, F., and D. Gewertz, eds. (2004). Yali’s Question: Sugar, Culture and History.
Fabian, J. (2002). Time and the Other: How Anthropology Makes Its Object [1983]. New York:
Columbia University Press.
Fastré, P. (N.d.). Manners and Customs of the Fuyuges [Moeurs et Coutumes Foujougheses]
[1937–9] (trans. M. Flower, O.C.D. and E. Chariot, O.C.D.). Boroko, Papua New
Guinea: Boismenu College.
Filer, C., and M. Macintyre. (2006). Grass Roots and Deep Holes: Community Responses to
Mining in Melanesia. Contemporary Pacific 18:215–31.
Fukuyama, F. (1992). The End of History and the Last Man. London: Penguin.
Gordon, R. G., ed. (2005). Ethnologue: Languages of the World, 15th ed. Dallas: SIL
International.
Halvaksz, J. (2008). Whose Closure? Appearances, Temporality, and Mineral Extraction in
Papua New Guinea. Journal of the Royal Anthropological Institute 14:21–37.
Herdt, G. H. (1977). The Shaman’s Calling Among the Sambia of New Guinea. Journal de la
Société des Océanistes 33:153–67.
Hirsch, E. (1999). Colonial Units and Ritual Units: Historical Transformations of Persons and
Horizons in Highland Papua. Comparative Studies in Society and History 41:805–28.
9781444333237_4_016.indd 296
8/7/2010 7:10:57 PM
297
Hirsch, E. (2008). God or Tidibe? Melanesian Christianity and the Problem of Wholes. Ethnos
73:141–62.
Jay, M. (1984). Marxism and Totality: The Adventures of a Concept From Lukács to Habermas.
Jorgensen, D. (2006). Hinterland History: The Ok Tedi Mine and Its Cultural Consequences
in Telefolmin. Contemporary Pacific 18:233–63.
Kirsch, S. (2006). Reverse Anthropology: Indigenous Analysis of Social and Environmental
Relations in New Guinea. Stanford: Stanford University Press.
Knauft, B. (2002a). Critically Modern: An Introduction. In Critically Modern: Alternatives,
Alterities, Anthropologies, ed. B. Knauft. Bloomington: Indiana University Press, 1–54.
Knauft, B. (2002b). Exchanging the Past: A Rainforest World of Before and After. Chicago:
Knauft, B. (2002c). Trials of the Oxymodern: Public Practice at Nomad Station. In Critically
Modern: Alternatives, Alterities, Anthropologies, ed. B. Knauft. Bloomington: Indiana
Lawrence, P. (1971). Road Belong Cargo. Manchester: Manchester University Press.
Leahy, M. and M. Crain (1937). The Land That Time Forgot: Adventures and Discoveries in
New Guinea. London: Hurst & Blackett.
Lévi-Strauss, C. (1966). The Savage Mind [1962]. Chicago: University of Chicago Press.
Lowenthal, D. (1985). The Past is a Foreign Country. Cambridge: Cambridge University Press.
McCarthy, J. (1963). Patrol Into Yesterday: My New Guinea Years. Sydney: Cheshire.
Megill, A. (1995). “Grand Narrative” and the Discipline of History. In A New Philosophy of
History, ed. F. Ankersmit and H. Kellner. Chicago: University of Chicago Press, 151–73.
Mink, L. (1987). Narrative Form as a Cognitive Instrument [1978]. In Historical Understanding,
ed. B. Fay, E. Golob, and R. Vann. Ithaca: Cornell University Press, 182–203.
Moretti, D. (2006). The Gender of the Gold: An Ethnographic and Historical Account of
Women’s Involvement in Artisanal and Small-Scale Mining in Mount Kaindi. Papua New
Guinea. Oceania 76:133–49.
Moretti, D. (2007). Ecocosmologies in the Making: New Mining Rituals in Two Papua New
Guinea Societies. Ethnology 46:305–28.
Nelson, H. (1976). Black, White and Gold: Gold Mining in Papua New Guinea, 1878–1930.
Canberra: Australian National University Press.
Nelson, H. (2000). McCarthy, John Keith 1905–1978. In Australian Dictionary of Biography.
Melbourne: Melbourne University Press, 170–1.
Oakeshott, M. (1983). On History and Other Essays. Oxford: Basil Blackwell.
Ouroussoff, A. (1993). Illusions of Rationality: False Premises of the Rational Tradition. Man
28:281–98.
Robbins, J. (2003). Properties of Nature, Properties of Culture: Possession, Recognition, and
the Substance of Politics in a Papua New Guinea Society. Suomen Antropologi 28:9–28.
Robbins, J. (2004). Becoming Sinners: Christianity and Moral Torment in a Papua New Guinea
Society. Chicago: University of Chicago Press.
Robbins, J. (2005a). Introduction – Humiliation and Transformation: Marshall Sahlins and the
Study of Cultural Change in Melanesia. In The Making of Global and Local Modernities in
Melanesia: Humiliation, Transformation and the Nature of Cultural Change, ed. J.
Robbins and H. Wardlow. Burlington, VT: Ashgate, 3–21.
Robbins, J. (2005b). The Humiliations of Sin: Christianity and the Modernization of the
Subject Among the Urapmin. In The Making of Global and Local Modernities in Melanesia:
Humiliation, Transformation and the Nature of Cultural Change, ed. J. Robbins and H.
Wardlow. Burlington, VT: Ashgate, 43–56.
Sahlins, M. (1992). The Economics of Develop-Man. Res 21:13–25.
Samuel, R. (1994). Theatres of Memory: Past and Present in Contemporary Culture. London:
Verso.
Simpson, C. (1954). Adam With Arrows: Inside Aboriginal New Guinea. Sydney: Angus and
Robertson.
9781444333237_4_016.indd 297
8/7/2010 7:10:57 PM
298
Hirsch and Moretti
Strathern, M. (1988). The Gender of the Gift. Berkeley: University of California Press.
Strathern, M. (1990). Artefacts of History: Events and the Interpretation of Images. In Culture
and History in the Pacific, ed. J. Siikala. Helsinki: Finnish Anthropological Society,
25–44.
Strathern, M. (1991). Partial Connections. Savage, MD: Rowman and Littlefield.
Strathern, M. (1995). Nostalgia and the New Genetics. In Rhetorics of Self-Making, ed. D.
Battaglia. Berkeley: University of California Press, 97–120.
Strathern, M. (2005). Kinship, Law and the Unexpected: Relatives Are Always a Surprise.
Trompf, G. W. (1990). The Cargo and the Millennium on Both Sides of the Pacific. In Cargo
Cults and Millenarian Movements, ed. G. W. Trompf. Berlin: Mouton de Gruyter,
35–94.
Trompf, G. W. (1994). Payback: The Logic of Retribution in Melanesian Religion. Cambridge:
Wagner, R. (1972). Habu: The Innovation of Meaning in Daribi Religion. Chicago: University
of Chicago Press.
Wagner, R. (1986). Symbols That Stand for Themselves. Chicago: University of Chicago Press.
Wagner, R. (1991). The Fractal Person. In Big Men and Great Men: Personifications of Power
159–73.
Watson, J. (1956). Review of Simpson, C. 1954. American Anthropologist 58:1163–4.
White, H. (1987). The Content of the Form: Narrative Discourse and Historical Representation.
Baltimore: Johns Hopkins University Press.
White, H. (1999). Figural Realism: Studies in the Mimesis Effect. Baltimore: Johns Hopkins
University Press.
Worsley, P. (1970). The Trumpet Shall Sound, 2nd ed. London: Paladin.
9781444333237_4_016.indd 298
8/7/2010 7:10:58 PM
17
Drawing Together
Materials, Gestures, Lines
Tim Ingold
As individuals express their life, so they are. What they are, therefore, coincides with
their production, both with what they produce and with how they produce.
– Karl Marx and Friedrich Engels (1977 [1845–6]: 42)
The only thing that is given to us and that is when there is human life is the having
to make it.… Life is a task.
– José Ortega y Gasset (1941 [1935]: 200)
The manner in which we humans are on the earth is Buan, dwelling. To be a
human being means to be on the earth as a mortal. It means to dwell.
– Martin Heidegger (1971 [1954]: 147)
A Biographical Prelude
Anthropology, in my view, is a sustained and disciplined inquiry into the conditions and
potentials of human life. Yet generations of theorists, throughout the history of the discipline, have been at pains to banish life from their accounts, or to treat it as merely
consequential, the derivative and fragmentary output of patterns, codes, structures, or
systems, variously labeled genetic or cultural, natural or social, that are already assembled
and joined up. My own work, over the last quarter of a century, has been driven by an
ambition to deliver life from the exile into which it has thus been driven and to restore it
to where it belongs, at the heart of anthropological concern.
It all began with my attempts to understand the meaning of production. Comparing
the work of human beings and nonhuman animals, I argued that instead of thinking
of production in terms of a transitive relation between an initial image (such as a
design for a building) and a final object (such as the completed house), it should be
understood intransitively, as a process of life that is continually carried on through
the forms and architectures that it brings into being (Ingold 1983: 15). I found a
precedent for this view in the early work of Marx and Engels, for whom a mode
9781444333237_4_017.indd 299
8/7/2010 7:11:22 PM
300
Inglod
of production was, in itself, a mode of life (1977 [1845–6]: 42). Subsequently, I came
across the writings of the Spanish philosopher José Ortega y Gasset, much cited by
mid-twentieth-century cultural anthropologists in the belief that they lent support to
their idea that culture, and not nature, shapes human experience. In reality, Ortega
had argued nothing of the sort. His celebrated claim – that man has no nature, only
history – was in fact a declaration of the primacy of life. “Let us say,” Ortega wrote,
“not that man is, but that he lives” (Ortega y Gasset 1941 [1935]: 111). Humanity
cannot be taken as a given; it is something we have continually to work at. What we
are, or what we can be, does not come ready-made in any kind of program, genetic
or cultural. We have, perpetually and never-endingly, to be making ourselves. That is
what life is, what history is, and what it means to produce.
At around the same time, I also encountered the philosophical writings of Alfred
North Whitehead and Henri Bergson, and in my book Evolution and Social Life (Ingold
1986), I set out to link this sense of production as a life process to the ideas of creativity
and duration that I drew from their work. From Whitehead, I took the idea that the
world we inhabit is never complete but continually surpassing itself. Creativity lies in the
movement of the world’s self-surpassing, or what Whitehead called “concrescence”
(1929: 420). “For a conscious being,” as Bergson reasoned in similar vein, “to exist is
to change, to change is to mature, to mature is to go on creating oneself endlessly”
(1911: 8). Crucial to Bergson was the claim that in this movement of self-creation, of
life and growth, lies the essence of time: “Wherever anything lives, there is, open somewhere, a register in which time is being inscribed” (1911: 17). Thus production – indeed,
life itself – far from being encompassed within the finalities of any particular project,
forever overtakes them in the process of their implementation. “And it does so,” as
I wrote in my Evolution book, “by virtue of its direction by the agency of consciousness
as a movement in real time [that] … is essentially continuous” (Ingold 1986: 327). One
may set out to build a house or to cultivate a field, but conscious life goes on.
Later, I would find this same distinction between transitive and intransitive senses
of production reflected in the phenomenological reflections of Martin Heidegger on
the difference between building and dwelling (Heidegger 1971: 145–61). Building is
transitive: it ends in something built, already envisaged at the outset. But dwelling is
intransitive: like life, it carries on. Against the commonplace idea that building is but
a means to dwelling, Heidegger argued the reverse: “We do not dwell because we
have built, but we build and have built because we dwell.… Only if we are capable of
dwelling, only then can we build” (Heidegger 1971: 148, 160). In a collection of
essays on The Perception of the Environment (Ingold 2000), I took this as the founding
statement of what I called the “dwelling perspective,” by which I meant a perspective
that treats the immersion of the inhabitant in an environment or life world as an inescapable condition of existence (2000: 154, 185–7). He who would build a house or
cultivate a field must already dwell and, in dwelling, be literally embarked upon a
“movement along a way of life” (2000: 146). The producer is thus a wayfarer, and the
mode of production is itself a trail blazed or a path followed. Along such linear paths,
I concluded, lives are lived, skills developed, and understandings grown.
It is from this conclusion that I have set out in my current work, comprising a series
of explorations in the comparative anthropology of the line (Ingold 2007c). What if
the living being is the line of its own movement? Then it can no longer be imagined as
a bounded totality, surrounded by its environment. We have rather to think of it as a
line of growth or concrescence – or, more realistically, as a bundle of such lines – and
of the environment as a zone in which these lines become comprehensively entangled
with one another. Ecology, I suggested, would then be “the study of the life of lines”
9781444333237_4_017.indd 300
8/7/2010 7:11:22 PM
Drawing Together
301
(Ingold 2007c: 103). Now I am not so sure. Can such lines really be objects of study?
Literally, an object is a thing that has been thrown before the mind, in a form that can
be apprehended. Life, however, is in the throwing and in the apprehension. It is the
becoming of things perceived and of ourselves as perceivers. Perhaps, then, as an inquiry
into the conditions and potentials of life, anthropology is not so much a study of lines
as in them. If every living being is a bundle of lines, then this must apply as much to
ourselves as students of life as to those with whom we study. Our lines become enmeshed
with theirs. In this chapter I shall present a sketch of what an anthropological study in
lines might look like. And in doing so, I aim to give my provisional answer to the question raised in Chapter 15 of this volume by Morten Pedersen and Rane Willerslev: how
can one go about being holistic without having recourse to totalization?
Painting and Drawing
In my search for an answer, I have been inspired by a recent article by the art historian
Norman Bryson (2003), written to accompany a major exhibition on the theme of
drawing. In it, Bryson compares drawing with painting, or more specifically, with the
Western tradition of oil painting.1 One particular aspect of the comparison caught my
attention. It starts with the painter or draughtsman poised at that inaugural moment
when the hand is about to make its first trace on an initially blank surface. You might
think this is a moment that drawing and painting have in common. But in reality, Bryson
argues, the perceptions of blankness, and of the potentials it holds, are radically different
in each case. The painter perceives a surface that has to be filled throughout its extent,
an extent that is nevertheless bounded by the four sides of the frame. This frame exerts
a kind of pressure that rebounds inward on the composition in such a way that every
element that is added – every trace of the brush – has to anticipate the totality of the
complete picture of which it will eventually form a part. It is, in other words, subject to
what Bryson (2003: 150–1) calls “the law of the all-over.” Drawing, by contrast, is not
compelled to observe this law. Instead, although the blank surface of the paper is perceptually present, it does not have to be conceived as a surface, as an area that needs to
be filled. It becomes rather a “reserve,” a kind of insurance against finality and closure.
Thus, the drawn line can unfold in a way that responds to its immediate spatial and
temporal milieu, having regard for its own continuation rather than for the totality of
the composition. “The reserve,” Bryson writes, “introduces a principle of noncompositionality, an antitotalizing force that relieves the drawn line of the responsibility to
always put the totality first, to put the collective first, and to assume in relation to the
surface a secondary, derivative function” (2003: 151). Compared to painting’s logic
of the all-over, drawing’s logic is of “localized space and time.” Operating with these
different logics, the painter’s brush and the draughtsman’s pencil follow distinctive
trajectories. The brush, before it can touch the surface of the canvas, has to hover – to
hesitate – while it takes in the sum of marks already made and seeks out an appropriate
channel of entry that is consistent with the overall compositional aim. The pencil, by
contrast, freed from this “complex calculus of the totality,” does not hover but carries
on its way from where the hand is now positioned, responding only to the present
conditions in its vicinity rather than to any imagined future state. Indeed, the conception of the surface as a reserve ensures that no drawing is ever finished. The last line
to have been drawn is never the last that could have been drawn: even that final line
“is in itself open to a present that bars the act of closure” (Bryson 2003: 150). So as
9781444333237_4_017.indd 301
8/7/2010 7:11:22 PM
302
Inglod
painting moves to completion, drawing carries on: it is always work in progress, manifesting in its lines a history of becoming rather than an image of being.
Reading Bryson, it immediately struck me that what he says about the processes of
art could readily be transposed to those of social life. Metaphorically, the brush and the
pencil could stand, respectively, for two ways of thinking about human action, and
about the context of that action. One way is to imagine that to act as a responsible social
being is to put the totality first, to “make one’s mark on society” (as we say colloquially), by contributing to the overall picture. Society, here, covers the world, within the
limits of an institutional frame that rebounds inwardly on the constitution of the person
or the self. Another way is to imagine the social world as a tangle of threads or life paths,
ever raveling here and unraveling there, within which the task for any being is to improvise a way through, and to keep on going. Lives are bound up in the tangle, but are not
bound by it, since there is no enframing, no external boundary. Thus, the self is not
fashioned on the rebound but undergoes continual generation along a line of growth.
The comparison between these two ways of thinking is of course a heuristic one,
and I have no desire to set up a crude dichotomy, let alone to argue that one way is
right, and the other wrong. I would venture, however, that the first has been the
default position in social anthropology for much of its twentieth-century history, and
that it underwrites the discipline’s long-held profession to address the totality of social
phenomena. The second surely has its antecedents too, but it has always remained as
something of an undercurrent in a discipline that seems determined to enframe others, to paint them into the picture, and thereby to bring closure to their lives. For my
part, however, I have found it profoundly liberating (Ingold 2007c: 4). As Bryson
said of the drawing, it is premised on a principle of noncompositionality, and harbors
an antitotalizing force that enables us better to understand how lives are lived not in
closed social worlds but in the open. These lives are social not because they are framed
but because they are entwined. All life is social in this sense, for a fundamental principle of the life process is that it is inherently multistranded, an intertwining of many
lines running concurrently. The great Swedish geographer Torsten Hägerstrand
(1976: 332) calls it the “principle of togetherness”; this is not, he writes, “just resting
together. It is also movement and encounter.” There can be no life that is not social, or
that evades this principle. Like the lines of a drawing, the lines of social life manifest
histories of becoming in a world that is never complete but always work in progress.
Every life is a trajectory through time, issuing forth from its tip. “Seen from within,”
Hägerstrand observes, “one could think of the tips of trajectories as sometimes being
pushed forward by forces behind and besides and sometimes having eyes looking
around and arms reaching out, at every moment asking ‘what shall I do next?’ ”
(1976: 332). A living anthropology is one that would join with these trajectories of
pulling and pushing, and follow them wherever they may lead.
Yet it is not just human beings who, among themselves, encounter one another in
the course of resting and moving together. Nor, conversely, do assorted nonhumans
seek and enjoy only each other’s company. Writing over 30 years ago, Hägerstrand
already anticipated the fall of the great divide between Nature and Society. We can
bring these divisions under one perspective, he argues, “because what is all the time
resting, moving and encountering is … humans, plants, animals and things all at
once” (1976: 332). Togetherness binds all things, but they are not bound into a
totality, or placed within a common frame. Hägerstrand’s call is for a holism which,
like that of the drawing, is processual and open-ended and, by the same token, both
noncompositional and antitotalizing. My proposal, then, is to redraw anthropology
9781444333237_4_017.indd 302
8/7/2010 7:11:22 PM
Drawing Together
303
along these lines. The drawing is part metaphorical, but also part methodological.
Metaphorically, it is about our understanding of persons and other things as drawing
together or binding the trajectories of life. Each, we might say, is a togethering.
Methodologically, it is about the potential of drawing as a way of describing the lives
we observe and with which we participate, both in movement and at rest, in what is
sometimes called the “ethnographic encounter.”
Toward a Graphic Anthropology
I find it extraordinary that in all the debate about “writing culture,” the assumption
has always been that the graphic part of ethnography is writing and not drawing.
“What does the ethnographer do?” Clifford Geertz once asked rhetorically; “He
writes” (Geertz 1973: 19). What a limiting view this is! Given that by all accounts,
drawing is an immensely powerful tool of observation, and given also that it combines
observation and description in a single gestural movement, why has it been all but
forgotten in anthropology? I want to suggest that the answer lies in a residual commitment, within the mainstream of the discipline, to a painterly aesthetic that values
compositionality and totalization over improvisation and process. Ethnography
remains beholden to the “law of the all-over,” which it satisfies through means that
are antithetical to the waywardness of the drawn line. Rather than joining with the
togetherings of life, and carrying them forward, its tendency is to want to retroject
the fullness of the phenomenal world, caught at a particular moment, back upon the
surface of the page as if on a blank canvas or screen.
This is what Alfred Kroeber, in a paper published in 1935, meant by “an endeavour
at descriptive integration” that aims, as he put it, to grasp “the totality of phenomena”
(545–7). Subsequently endorsed by E. E. Evans-Pritchard (1950: 122) and thence
passed to the mainstream of social anthropology, Kroeber’s ideal of descriptive integration was a direct heir to the “art of describing” perfected by Dutch painters in the seventeenth century (Alpers 1983). The objective was to render a moment in the collective
life of a people with the same completeness and accuracy with which the Dutch rendered
their landscapes. As the painting covers the entire canvas, leaving no space unfilled, so in
an ethnographic account that grasps the totality there should be no gaps. Every detail
should be filled in. Indeed, the very “thickness” of the ethnographer’s description, to
use Geertz’s (1973: 6) term,2 brings to mind the density and opacity of oil paint, which –
as Bryson explains (2003: 149) – covers over and obliterates the workings of the picture.
All the revisions, alterations, and false starts that went into making it remain hidden,
buried under the surface that meets the eye, leaving the picture as a completed whole
that preserves in its compositional arrangement the totality of phenomena represented.
And so too, the perfect ethnography hides the traces of its inscription, presenting a picture of the life world as if it were arrayed, fully formed, upon a surface.
By contrast, an anthropology that takes drawing as its medium – that is, a graphic
anthropology (Afonso and Ramos 2004: 73) – would appeal to the openness of the
reserve rather than the closure of a surface that has been completely filled in. Indeed,
in its sights, the life world would have no surface. Finding a way through rather than
covering over, the drawn line contrives to make the surface disappear, or, as Bryson
puts it (2003: 151), to be “perceptually present but conceptually absent.” Though
practically inscribed as traces on a surface, the lines of the drawing appear like threads
9781444333237_4_017.indd 303
8/7/2010 7:11:22 PM
304
Inglod
in a void (Ingold 2007c: 57). If anything, they weave a surface rather than being laid
upon it. And, like threads, they cannot be erased.3 Since drawing does not cover,
neither can it be covered up. In life as in drawing, what is done cannot be undone.
You can only carry on from where you are now, leaving a trail behind you as evidence
of where you have been. Drawing, Bryson writes, is relentless: “it forces everything
into the open, into a field of exposure without shields or screens, with no hiding
places, a radically open zone that always operates in real time” (2003: 149).
A graphic anthropology, then, would aim not at a complete description of what is
already there, or has already come to pass, but at joining together with persons and
other things in the movements of their formation. The impulse of life, and thus of
graphic anthropology as a process of life, is to carry on, punctuated – but neither initiated nor terminated – by the finalities of the projects it brings into being. Like any
other life process, such an anthropology would operate in real time, coupling our own
movements of description with our observations of what is going on in the world,
which are, in turn, necessarily coupled – through the participatory act of togethering – with the trajectories of those with whom, or with which, we join. In this way,
the three corners of the ethnographic triangle – doing, observing, and describing –
can be unified in a single approach that I propose to characterize in terms of three
injunctions. These are to follow the materials, to copy the gestures, and to draw the
lines. In the following sections, I shall elaborate on each of these injunctions.
Follow the Materials
In his notebooks, the artist Paul Klee repeatedly urged that the processes of genesis
and growth that give rise to forms in the world we inhabit are more important than
the forms themselves. “Form is the end, death,” he wrote; “Form-giving is life” (Klee
1973: 269). This, in turn, lay at the heart of his celebrated “Creative Credo” of 1920:
“Art does not reproduce the visible but makes visible” (Klee 1961: 76). It does not,
in other words, seek to replicate finished forms that are already settled, whether as
images in the mind or as objects in the world. It seeks, rather, to join with those very
forces that bring form into being. Thus the line grows from a point that has been set
in motion, as the plant grows from its seed.
Taking their cue from Klee, philosophers Gilles Deleuze and Félix Guattari argue
that the essential relation, in a world of life, is not between matter and form but
between materials and forces (2004: 377). It is about the ways in which substances of
all sorts, enlivened by cosmic forces and with variable properties, mix and meld with
one another in the generation of things. Whenever we encounter matter, as Deleuze
and Guattari insist, “it is matter in movement, in flux, in variation.” And the consequence, they go on to assert, is that “this matter-flow can only be followed” (Deleuze
and Guattari 2004: 451). What Deleuze and Guattari call “matter-flow,” I would call
material. Accordingly, I recast the assertion as a simple rule of thumb: to follow the
materials. Consider, for example, the operation of splitting timber with an axe.
The practiced woodsman brings down the axe so that its blade enters the grain and
follows a line already incorporated into the timber through its previous history of
growth, when it was part of a living tree. “It is a question,” write Deleuze and Guattari,
“of surrendering to the wood, and following where it leads” (2004: 451).4
Much has been written, in recent years, on the relations between persons and things,
guided by the thought that the material world is not passively subservient to human
9781444333237_4_017.indd 304
8/7/2010 7:11:22 PM
Drawing Together
305
design. Theorists have expressed this, however, by appeal not to the vitality of materials
but to the agency of objects. If persons can act on objects in their vicinity, so, it is
argued, can objects “act back,” causing persons to do what they otherwise would not.5
Thus the log, in our example of splitting timber, is understood as an object that exercises its agency by guiding or inflecting the path of the axe, of perhaps by exerting such
a countervailing force as to cause the blade to jam. At one moment, a human subject is
acting on a material object; at the next, the object takes over, acting as a kind of subject
vis-à-vis the human practitioner who, in turn, becomes object-like. Instead of subjects
and objects, we are told, there are “quasi-objects” and “quasi-subjects,” connected in
relational networks (Latour 1993: 89). Yet paradoxically, these attempts to move beyond
the modernist polarization of subject and object remain trapped within a language of
causation that is founded on the very same grammatical categories and that can conceive
of action only as an effect set in train by an agent (e.g., Gell 1998: 16). At best, they lead
only to contradiction and confusion (Ingold 2007b: 52). At worst, they have led theoreticians to make fools of themselves in a way that we would be ill-advised to emulate.
For the world we inhabit, I maintain, is not made up of subjects and objects, or
even of quasi-subjects and quasi-objects. The problem lies not so much in the sub- or
the ob-, or in the dichotomy between them, as in the -ject. For the constituents of this
world are not already thrown or cast before they can act or be acted upon. They are
in the throwing, in the casting. The point may best be illustrated by means of a simple
experiment that I have myself carried out with my students at the University of
Aberdeen. Using fabric, matchstick bamboo, ribbon, tape, glue, and twine, and working indoors on tables, we each made a kite. It seemed that we were assembling an
object. But as soon as we carried our creations outside, they leaped into action, twirling, spinning, nose-diving, and occasionally flying. How did this happen? Had some
animating principle magically jumped into the kites, causing them to act most often
in ways we did not intend? Were we witnessing, in their unruly behavior, the consequences of interaction between – in each case – a person (the flyer) and an object
(the kite), which can only be explained by imagining that the kite had acquired an
“agency” capable of counteracting that of the flyer?
Of course not. The kites behaved in the way they did because, at the moment we went
out of doors, they were swept up – as indeed we were ourselves – in those currents of air
that we call the wind (Ingold 2007a: S30–1). The kite that had lain lifeless on the table
indoors, now immerse