Jackson Jean E. - The Fish People Linguistic Exogamy and Tukanoan Identity in Northwest Amazonia

Cambridge Studies in Social Anthropology
General Editor: Jack Goody
39
THE FISH PEOPLE
Downloaded from https://www.cambridge.org/core. University of Birmingham, on 13 Apr 2020 at 17:30:09, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511621901
For other titles in this series turn to page 285
The Fish People
Linguistic Exogamy and Tukanoan
Identity in Northwest Amazonia
JEAN E. JACKSON
Associate Professor of Anthropology
Massachussetts Institute of Technology
CAMBRIDGE UNIVERSITY PRESS

Cambridge
London New York New Rochelle
Melbourne Sydney
CAMBRIDGE u n i v e r s i t y p r e s s
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore,
São Paulo, Delhi, Dubai, Tokyo, Mexico City
Cambridge University Press

The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the United States of America by

Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521278225
© Cambridge University Press 1983
This publication is in copyright. Subject to statutory exception

and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 1983
A catalogue recordfor this publication is available from the British Library
Library of Congress Cataloguing in Publication Data
Jackson, Jean E. (Jean Elizabeth), 1943—

The fish people.
(Cambridge studies in social anthropology; no. 39)
Includes bibliographical references and index.
1. Tucano Indians — Social life and customs.
2. Barasana Indians — Social life and customs. 3. Indians
of South America — Colombia — Social life and customs.
I. Title. II. Series.
F2520.1.T9J3 1983 306'.08998 82-23564
ISBN 0 521 23921 4 hardcovers
ISBN 0 '521 27822 8 paperback
ISBN 978-0-521-23921-9 Hardback

ISBN 978-0-521-27822-5 Paperback
Cambridge University Press has no responsibility for the persistence or

accuracy of URLs for external or third-party internet websites referred to in
this publication, and does not guarantee that any content on such websites is,
or will remain, accurate or appropriate. Information regarding prices, travel
timetables, and other factual information given in this work is correct at
the time of first printing but Cambridge University Press does not guarantee
the accuracy of such information thereafter.
For my father and in memory of my mother
Contents
List of figures, maps, and tables x

Preface xi
Acknowledgments xvi
Note on orthography xviii
1 Purpose and organization of the book 1

Social identity 2
Regional perspective 5
Fluidity 8
Ideal and real 9
Final remarks 11
2 Introduction to the Central Northwest Amazon 13

Ecological setting 13
The population 17
Language and linguistics 19
Ethnic history 21
Early explorer and missionary efforts 22
Early and recent ethnographic descriptions 23
The Maku 24
A note on acculturation 24
3 The longhouse 26
The setting 26
The people of Pumanaka buro 26
The longhouse structure 30
Outside the longhouse 31
Inside the longhouse 33
Significance of the longhouse 36
vii
Downloaded from https://www.cambridge.org/core. University of Exeter, on 03 May 2020 at 07:12:29, subject to the Cambridge Core terms of
use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511621901
Contents
4 Economic and political life 39

Daily patterns 39
The river 42
The forest 46
Cultivated foods 50
Exchange in general 59
Property 62
Leadership 65
5 Vaupes social structure 69

The settlement 69
The sib 71
The language group 77
The phratry 86
Regional integration and
interaction between settlements 96
6 Kinship 105
Kinship terminology 106
Expectations and behavior 108
Specific kinship roles 117
7 Marriage 124
Principles of marriage 125
Marriage behavior 138
Conclusions 147
8 Tukanoans and Maku 148

Background to the Maku 148
Tukanoan attitudes toward the Maku 151
Interaction between Tukanoans and Maku 154
Maku as symbol to Tukanoans 158
Conclusions 161
9 The role of language and speech in Tukanoan identity 164

Vaupes language and speech as badges of identity 165
How Vaupes languages assume features
of the nonlinguistic environment 171
The importance of language in
Tukanoan culture 177
viii
Contents
10 Male and female identity 179

Relations between men and women 181
Conclusions 192
11 Tukanoans' place in the cosmos 195

Shamanism 195
Festivals 202
The Tukanoan world 204
Conclusions 208
12 Tukanoans and the outside world 211

Extractive industries 215
Homesteaders 217
The Colombian government 217
Missions 218
MM 223
Conclusions 224
13 Conclusions: themes in Tukanoan social identity 227

Types of comparisons 227
Themes associated with social identity 231
A note on types of evidence 239
Notes 243
Glossary 256
References 259
Index 273
IX
Figures, maps, and tables
Figures
1 The longhouse setting 27
2 The inhabitants of Pumanaka buro 29
3 Ground plan of longhouse interior 34
4 Traditional Vaupes social structure 73
5 Marriage distance between language groups 95
6 Marriages during three generations at Pumanaka buro 139
7 Two conflicting models of language distance 172
Maps
1 The Eastern Colombian Vaupes 14
2 Membership of settlements by language group 80
3 Locations of settlements intermarrying with Pumanaka buro 140
Tables
1 Language group affiliation of Tukanoans in sample 82
2 Names of language groups 85
3 Estimates of populations of language groups 87
4 Zero generation terminology at three levels of inclusion 90
5 Phratries listed by three informants 91
6 Marriage between selected language groups 94
7 Bara kinship terminology 109
8 Simplified Bara kinship terminology 112
9 Language distance as measured by cognates 173
Downloaded from https://www.cambridge.org/core. Universiteitsbibliotheek Utrecht, on 06 Mar 2020 at 01:30:26, subject to the Cambridge
Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511621901
Preface
In November 1968 I arrived in the Vaupe"s territory of Colombia planning to

study beliefs and practices related to disease and curing in a Northwest Amazon
tribe. I originally intended to work with the Tikuna, south of the Vaupe"s, but
conversations with anthropologists Gerardo Reichel-Dolmatoff and Alicia Dussan
de Reichel after my arrival in Colombia convinced me that the Tikuna were too
acculturated for the study I really wanted to do. I turned to the Vaupes region,
which had the additional advantage of being the focus of a number of ongoing
research projects, including Reichel-Dolmatoff s own work with the Barasana
and Desana.
After arriving in Mitu, the airstrip town on the Vaupes River that is the
administrative seat of the Vaupes territory, I spent about three weeks making
canoe trips and flights in small airplanes to various settlements. Realizing I had to
locate myself quite far away from Mitu and the Vaupes River, I then took a
missionary plane south to Monfort on the Papuri River and soon after went on a
ten-day visit with some Desana Indians to a small, nucleated village on Cano
Virari, which, however, was still too acculturated for my purposes. It became
clear during this frustrating but valuable period of orientation that I really did not
want to compromise in terms of acculturation level, and I reconciled myself to the
inevitability of settling in an extremely isolated community reachable only by a
long canoe trip.
During my stay at Monfort I met Samuel, a Desana who spoke Spanish and
understood why I wanted to stay in the most traditional settlement possible.
Samuel agreed that a longhouse community was the only answer to my require-
ments and even listed a number of characteristics still applicable to longhouse
communities but absent in nucleated villages. He said that later I could live in a
mission town, once I understood where Tukanoans had started from in their
journey toward the Colombian-Brazilian and Christian worlds. He suggested I
live with his Bara in-laws on the Inambu River, assuring me they would be
delighted to have me, and I decided to trust this indirect invitation. First,
however, I returned to Bogotd to buy such necessary equipment as an outboard
motor and gasoline. I then returned to the Vaup6s, meeting Samuel in Mitu and
xi
Downloaded from https://www.cambridge.org/core. University of Glasgow Library, on 02 Dec 2019 at 05:11:10, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511621901.001
Preface
traveling with him to his wife's parents' longhouse, Pumanaka buro ("hill of
many leaves"), where I was to spend the next eighteen months.
The longhouse was the home of about twenty Tukanoans: Bara men, their
non-Bard wives, and their children.1 These people had heard via the Vaup6s
grapevine that I was coming, and undoubtedly were as anxious about my liking
them as I was about their liking me. (They later told me some of the worries they
had had during the beginning of our acquaintanceship.) For my part, I desperately
hoped things would work out because it seemed as though I had already wasted
an inordinate amount of time.
My first two days in the longhouse were a confused jumble of talking to people
via Samuel and making attempts to begin learning Bara; a puppy I had acquired
in Mitu provided a conversation topic of sorts. Disoriented and anxious, I was
also extremely excited, reactions shared by virtually all anthropologists upon
entering the field but perhaps heightened by several factors in my situation.2 One,
I was completely alone. Two, by then I had spent almost three months and was
extremely impatient to begin "real" fieldwork. And three, living in a Tukanoan
longhouse involves an intensity and exclusivity of contact with one's fellow
residents unmatched in the vast majority of residence arrangements found in the
world. For periods of a week or more a settlement's residents see only one
another. Pumanaka buro, like all longhouses (at least at present), is remote
from other settlements; a two-hour canoe trip separates it from its nearest
neighbors downstream, a Tuyuka longhouse community. Still, although I felt
terribly cut off from virtually everything familiar to me, I adjusted, and over time
a very deep attachment formed, both on my part and, I believe, on theirs. The
intensity of feeling has been only partly diminished with time.
Over the months I would come to realize just how dependent I was on these
people, a dependency that made me euphoric at times and depressed at others. At
the beginning, I only dimly realized some of the psychological adjustments I
would have to make; these demanded more of me than any other aspect of
fieldwork, far more than physical discomfort. It is one thing to learn to cope with
new routines, to eat strange foods, to survive wasp massacres and fungal
invasions. Much of it, in fact, I enjoyed, and the challenge posed by the rest
suited my romantic side very nicely. But my feelings of loneliness and incompe-
tence were hard to deal with, heightened as they were by my belief that
anthropologists should not become "too involved." I tried to remain objective,
not to take sides in arguments, and in general to monitor my behavior so as to
avoid permanently alienating people. Straddling the fence of emotional involve-
ment and expression (saying and feeling neither too little nor too much) was
perhaps the most difficult of requirements in a setting like the longhouse, because
these people, how they felt about me and I about them, affected my life and work
in the most profound way. Of course I failed to keep my supposed objectivity. I
did take sides in disputes, and I have come to accept the inevitability of this
xii
Preface
involvement far more than I did then. Certainly at times my role as anthropologist
conflicted with that of coresident, but I believe that in general whatever value my
research has derives to a considerable extent from the relationships I shared with
my fellow members of Pumanaka buro. I am not making the obvious observation
that the longer one lives and interacts with a group of people, the better one
understands them. True as that is, I mean to suggest here that a process of
emotional attachment must occur in order really to understand a culture but that
at times it works against one's desire to be detached, objective, scientific. The
best emotional stance to take when carrying out fieldwork must, of course, be an
individual decision, but to believe in one's total objectivity and detachment is
naive and self-deceptive: A great deal can be gained in crossing emotional
boundaries along with the cultural ones.
These rather lengthy remarks are intended not only to express my deep feelings
of appreciation toward the people of Pumanaka buro but also to call attention to
the fact that I did spend much of my time with one particular group of people in a
rather isolated location, a fact that bears directly on the avowedly regional
perspective taken by this book in portraying Tukanoan social identity. How I came
to reorient my research plan from a local to a regional perspective is important
information for a number of reasons, and thus a discussion of it follows.
Soon after arriving at Pumanaka buro I began a study of the Bara language,
in preparation for carrying out the cognitive ethnosemantic part of my research. It
quickly became apparent that the proposed research could not be done, because it
depended on an implicit research assumption that the members of the subject
group all spoke the same language. I found myself in a region with more than
sixteen languages, several of them represented in each longhouse. Although I had
been aware of this fact before, its implications did not strike me until I was in the
field. The rule of exogamy required that all inmarrying women at a settlement be
from other language groups. Every Tukanoan I talked to was at least trilingual,
and children began acquiring two languages almost from the beginning of
language learning. Most conversation was in Bara, but the women with whom I
spent much of my time also spoke in Tuyuka and occasionally in Tukano ("just
for a change").
Furthermore, the same problem arose when I contemplated research on
decision making about disease and curing and the degree of fit between normative
pronouncements and actual behavior. Other settlements of the region were at
least half an hour by canoe from each other. Worse, there were no other Bard
longhouses on the river or in the immediate vicinity of Pumanaka buro, all its
neighbors being Tuyuka, Tukano, and Desana. How could I, then, by working
with one language, claim to say anything about how these people think about and
classify disease terms and behaviors, or anything else, for that matter? This
situation, which fortunately confronts few fieldworkers, seemed to discourage
every possible avenue of research until I decided to capitalize on it and begin an
xiii
Preface
investigation of the social structure of Tukanoans, in particular those of the

Papuri drainage region, concentrating on the relation between multilingualism,
kinship, and marriage.
If I had begun with a well-thought-out research design, complete with
hypotheses to be tested and so forth, my field work would have conformed to
current standards more closely than it did. My progress toward understanding the
various linguistic and marital mysteries of the Vaupes has in fact been much
more serendipitous, full of dead ends and interesting side paths that, unfortunate-
ly, incomplete data have not allowed me to explore thoroughly. I offer these
remarks to explain some of the gaps in the data; I simply was not aware of the
complexity and reach of the system while researching it. The conclusions given
here have evolved during a ten-year period and are the result of many struggles
with the data and many conversations with fellow Vaupes specialists and other
anthropologists.
Although I spent much of the field period at a single site, I traveled frequently;
sometimes I accompanied the residents of Pumanaka buro when they went to
rituals at other settlements but more often I traveled in order to get in and out of
the region. The canoe trip from Pumanaka buro to Mitu never took less than
six days, and one memorable trip during the dry season took ten days. As I came
and went I changed boat crews frequently and slept at a different settlement each
night. I gathered as much information as I could at each stop, especially on
demography and settlement locations; by the end of my fieldwork, when I had
grasped at least in part the extent of the marriage system, I came to see this
information as extremely valuable. In every settlement some of the inmarried
affines came from distant places, making it possible to find out about communi-
ties well off my route. Furthermore, Tukanoans displayed a great interest in the
physical and social geography of the Vaupes, and they seemed to see it as a
single region. Their interest in answering my questions or finding others to
answer them gradually convinced me that I was investigating a true cultural
focus, equivalent to the interest of highland Chiapas Indians in corn or Dobuans
in sorcery.
Thus, I eventually broadened my study as far as possible, because neither I nor
Tukanoans could see any clear-cut boundaries between Tukanoans and "other
people." Even my goal of a complete census of marriages of living individuals of
the upper and mid-Papuri and its tributaries, which I eventually achieved, did not
mark off a socially bounded unit. During my trips I drew sketch maps, which,
crude though they were, demonstrated in combination with the census data the
extent to which marriage unified subsections of and, less so, the entire Vaupes
region. Thus those canoe trips that I often dreaded, greatly envying other
anthropologists who were able to fly into and out of their sites, proved to be the
source of much essential data.
It must be borne in mind that my research strategy evolved during fieldwork
and that I spent most of my time in a single longhouse community, circumstances
xiv
Preface
that have led to my taking the Bara language group and, more specifically, the
residents of Pflmanaka buro as representative of Tukanoan society. Although I
have been careful throughout the book to point out all the discrepancies and
exceptions of which I am aware, it should be remembered that in general we shall
be looking at Tukanoan society through the spectacles of the Bard living in the
Papuri drainage region. My approach has also necessitated striking a balance
between ethnographic specificity and systemic overview. The balance here seems
to me a good one, but because the work deals with both local and regional data I
have had to make many decisions regarding the amount of material to include
from either perspective. I obviously cannot, when describing an open-ended
system like the Vaupes, describe and analyze how every person, every local
group, every language-affiliated group differs from and resembles every other
one. A serious attempt to combine both a regional overview with a comprehen-
sive ethnography of every group encompassed within the overview would require
more years of research and more volumes than even the most forgiving granting
agency or publisher could conceivably tolerate.
Thus, this book represents a point - 1 hope a high but not a final one - in a long
process of investigation and analysis of Tukanoan social identity. I hope to return
to the Vaupes and continue research; indeed, this monograph was completed
before another sojourn only when it became evident that one planned for 1976
would be impossible.3 Many of my intentions for future research will become
apparent in the following pages. I am confident in the essential accuracy of my
portrayal of the form of the Vaupes social system; by adding content, admittedly
largely from Bara/Papuri sources, I believe I have sketched a reasonable portrait.
It is hoped that this book will inspire others to comment on the picture I have
painted or attempt their own interpretation of this fascinating system and the
people who have created it.
xv
Acknowledgments
Many, many people have helped me in the various stages of preparing this book.
Of the numerous people at Stanford who offered their advice and support before,
during, and after fieldwork, special thanks go to Benjamin Paul, George Collier,
and members of my dissertation committee, Charles Frake, Bernard Siegel, and
Renato Rosaldo.
Many Colombians assisted in the research, and I would like to thank the
Colombian government and its representatives in Bogota and Mitu for permit-
ting and facilitating my investigations along the way. The many individuals
attached to the Prefecture Apostolica del Vaupes gave generously of their time
and hospitality, in particular Monsenor Belarmino Correa and the missionaries
at Monfort and Acaricuara. Betty Welch, Birdie West, and Janet Barnes of the
Summer Institute of Linguistics, Colombia, also offered encouragement during
my stay. The warmth and laughter of Don Tito Vargas and his wife, Dona Alix,
of Mitu made my stops there far more pleasant than they otherwise would have
been.
My thanks also go to the many people of the Universidad de los Andes and the
Instituto Colombiano de Antropologfa in Bogota who aided the research, in
particular Alicia Dussan and Gerardo Reichel-Dolmatoff.
During the years spent writing this manuscript, various sections of it were sent
to people or used in lectures. Many people responded generously with comments
and criticism. I will simply list them alphabetically, because to do otherwise
would be very cumbersome: Kaj Arhem, Joan Bamberger, Ellen Basso, Patrice
Bidou, Roy D'Andrade, Gertrude Dole, Steven Fjellman, James Fox, Nina S. de
Friedemann, Hugh Gladwin, Irving Goldman, Joseph Greenberg, Thomas
Gregor, John Gumperz, Robert Hahn, Raymond Hames, Dell Hymes, Judith
Irvine, Pierre-Yves Jacopin, Theodore Johnson, Joanna Kaplan, Kenneth Ken-
singer, David Kronenfeld, David Maybury-Lewis, Wick Miller, Jerry Moles,
Naomi Quinn, Howard Reid, Peter Riviere, Kimball Romney, Benson Saler,
Gillian Sankoff, Judith Shapiro, Joel Sherzer, Peter Silverwood-Cope, Janet
Siskind, Carol Smith, Arthur Sorensen, David Thomas, Katherine Verdery, and
Norman Whitten. I also acknowledge the many other people who gave assistance.
xvi
Acknowledgments
Research in Colombia during 1968-70 was supported by the Danforth Founda-

tion and the Stanford Committee for Research in International Studies. An Old
Dominion fellowship and a grant from the Dean of Humanities and Social
Sciences, MIT, supported work on the manuscript.
For encouragement and, at times, forbearance, I thank my colleagues at MIT,
especially Martin Diskin and James Howe. Jim's contribution, in the form of
painstaking editing of parts of the manuscript, I and all future readers of this
book have reason to be grateful for.
As is evident from the many references to their written work and various
communications to me in the following pages, my debt to Stephen and Christine
Hugh-Jones is immeasurable.
I also thank the people at Cambridge University Press, in particular Jack
Goody, Walter Lippincott, and Susan Allen-Mills.
Finally, the people at Pumanaka buro, in particular Juanico Escobar and his
wife, Maria Tamano, and family, took me in and helped me in countless ways.
Juanico, Maria, Lois Paul, Michelle Rosaldo, and my mother, Mary Elizabeth
Gaines Jackson, have all died since the research began. I regret they cannot see
its final form, because they were all instrumental - in very different ways - in
bringing it about.
I gratefully acknowledge permission from Academic Press to reproduce Tables
1 and 2, Figures 2, 3, 5, and Map 3 of my chapter in C. Smith, ed., Regional
Analysis, Volume 2, Social Systems. I also thank Cambridge University Press for
permission to use material from my chapter in Richard Bauman and Joel Sherzer,
eds., Explorations in the Ethnography of Speaking. Mouton Publishers kindly
permitted reproduction of Chart 1, p. 121, of Nathan Waltz and Alva Wheeler's
article in E. Matteson, ed., Comparative Studies in Amerindian Languages,
which appears as Table 9 in Chapter 9. And Seminar fur Ethnologie has
permitted reproduction of part of the material on "Ethno-linguistic groups of
Colombia" in W Dostal, ed., The Situation of the Indian in South America,
which appears here as Table 3.
xvu
Note on orthography
The orthography used in this book is a phonetic transcription of Bara, simplified

to make it accessible to and convenient for English-speaking readers. The kinds
of simplifications chosen do not confuse the particular lexical items dealt with in
the text, that is, they would not reduce phonetically distinct forms to homonyms.
For a more comprehensive treatment of Bara (as spoken in the Pira-parana, a
separate dialect from Inambu Bara), see Stolte and Stolte, 1971. For example,
tone (Bard has two) is not indicated in my transcription, and stress has been
simplified (high tone and stress co-occur): Stress falls on the penultimate syllable
unless shown elsewhere. Aspiration is not phonemic; I have indicated preaspiration
where it occurs with the voiceless stops /p/, Itl, and /k/ as an aid to pronuncia-
tion (e.g., mehkd, "father's sister")-
Vowels
Unnasalized Nasalized
a as in father a
e as in eight e
i as in ped/atrics I
o as in oval 6
u as in food u
n
ii similar to German
u u
Consonants
P as input ft as in onion
b as in but m as in man
t as in fub r alveolar flap, as in Spanish pero
(between r and / in English)
d as in a"ub w as in water
k as in it it y as in y awn
g as in got h as in h at
n as in not ng as in sing
XVlll
Downloaded from https://www.cambridge.org/core. Cambridge University Main, on 28 Dec 2019 at 16:34:55, subject to the Cambridge Core
Note on orthography
In order to distinguish the linguistic stocks from which cited forms derive, I
have used italics at the first occurrence of Bara and other Eastern Tukanoan
words (as well as for scientific names, emphasis, and metalinguistic references)
and bold face for the first occurrence of Spanish, Portuguese, and Tupian loan
words.
xix
1.
Purpose and organization of the book
The social system of the Tukanoans of the Central Northwest Amazon has
intrigued virtually every scholar who has come into contact with it. Each of the
more than sixteen languages spoken there is identified with a named descent
group. Although these groups have sometimes been called tribes, they are rather
strange tribes. For one thing, with a few exceptions, they are exogamous. This
book is intended as a general introduction to the Vaup6s, the Colombian sector
of the Central Northwest Amazon, and more specifically to Tukanoan social
identity.
To a considerable extent, the book's conception and organization reflect the
way Tukanoans organize their social world. Most of the chapter topics derive
from categories and distinctions Tukanoans themselves make among kinds of
people and among other kinds of beings in their universe. By providing clues
about the essence of being Tukanoan, these categories offer a logical starting
point for discovering the content and organization of Tukanoan social identity, as
Tukanoans conceptualize it and as they reveal it in their behavior. The general
progression of chapters can be understood if we imagine ourselves to be looking
at Tukanoan society through a series of lenses, each successive lens of lower
power than the preceding one and thus encompassing units and categories of
increasing scale and magnitude. The geographical scope of an image increases
chapter by chapter, beginning with the traditional local unit, the longhouse, and
ending with the entire Tukanoan universe. The first chapters serve to distinguish
kinds of Tukanoans by such characteristics as kinship or language affiliation,
whereas later chapters take up contrasts of wider scope, such as those between
Tukanoans and Maku (the other indigenous inhabitants of the Vaupe"s), between
Tukanoans and nonhuman spirits, and between Tukanoans and whites. The field
of view does not increase evenly in every chapter: For example, discussion of the
male-female polarity comes toward the end because it is a distinction of great
scope, even though our starting point, the longhouse, includes both men and
women. In addition, other considerations have in places supplanted the progres-
sion from small to large scale: The book ends, for example, with the contrast
between Tukanoans and whites, because it is white society that is destroying
1
The fish people
traditional Tukanoan culture. Among other things, this final chapter deals with
some of the unhappy transformations Tukanoan identity will undoubtedly undergo
in the future.
The time reference of this book, unless otherwise noted, is the ethnographic
present of 1968-70. This means that some of the discussion, particularly con-
cerning acculturative influences, is now somewhat outdated. Rather than alter the
picture to adjust it to the changes I know have taken place (for example, those
caused by the penetration of the cocaine trade into the area) I have decided to
maintain consistency and hold to the time period of my fieldwork. Not having
returned to the Vaupes since then, I would otherwise skew my descriptions
toward aspects of recent change that have reached my attention, neglecting others
that have not.
A discussion of some of the theoretical and methodological issues addressed in
these pages follows.
Social identity
This book takes as its focus and orientation those components of identity that
derive from a Tukanoan's membership in various social groups, categories, and
positions. It analyzes the various social roles Tukanoans play throughout life and
how these affect the way the people are conceptualized and categorized by
themselves and others. The stage on which these roles are performed - the
Tukanoan world - is described as well. As already indicated, emphasis falls here
on social rather than strictly personal identity, that is, on those components of a
person's identity that are acquired through relations with other people. These
components include relatively abstract analytical relations, for instance those of
similarity and contrast as well as others embodied in relatively concrete relations
of genealogy, location, or language. Gender is by this definition an aspect of
social identity, whereas sex, if defined strictly in genetic, anatomical, or physio-
logical terms, can be seen as a feature of personal identity. (Transsexuals, whose
gender and sex diverge, exemplify this distinction. Even for transsexuals per-
haps, the two types of identity are not unrelated, but they are analytically
distinct.)
A number of authors have stressed the social constitution of identity. For Peirce
(see Singer, 1980), because the self is both a product and an agent of semiotic
communication, it is both social and public. Individual identity, in his theory, is
"also a social and cultural identity and is not confined to the individual organ-
ism" (Singer, 1980, p. 485). Similarly, Hallowell, while acknowledging the
universality of the concept of self, notes that the cultural form this concept takes
is highly variable: "the individual's self-image and his interpretation of his own
experience cannot be divorced from the concept of the self that is characteristic
of his society" (1955, p. 76). G. H. Mead has also stressed the social nature of
concepts of self, and Kaplan goes so far as to say: "action is not to be understood
as some by-product of the real characteristics of the person but as being the only
means of constituting reality itself. The person, in effect, creates himself by what
he does" (1961, p. 310). Given that human beings do not spend their lives in a
closet (or the Tukanoan equivalent, alone in the forest), such an existentialist
position implies that the action that constitutes social identity is interaction with
others.
Although I basically agree with these statements, my use of "social identity"
is narrower. This book concentrates on the crucial features Tukanoan society
highlights when assigning its members to different groups. Kinship, marriage,
age, sex, humanness - identity components of this sort are in large part created
and refined by the actual relations between individuals and groups. The Tukanoan
version of this essential interconnection between social identity and social struc-
ture, in particular between identity and regional and linguistic organization, will
emerge as the book progresses. Of course, identity components not comprehen-
sively treated here, such as ideas of conception and growth, anatomy and
physiology, are extremely important and do influence social identity. Space does
not allow a thorough treatment of these and other topics such as the effects of
Tukanoan socialization patterns on personality or the feelings Tukanoans have
about their individual selves. Still, in my opinion, the areas of life and culture
stressed here - the assignment, symbolism, and performance of social roles -
although they do not provide a complete picture of Tukanoan identity, do fill in a
large part of the canvas.
Of course, my information on Tukanoan social identity comes from individuals
with distinct personalities living in unique slices of space and time. My goal is to
understand what Tukanoans have in common without hiding or ignoring those
things that make them different from each other, to transcend individual variabil-
ity so as to see the logic of Tukanoan identity as a system. This goal demands that
in a number of places throughout the book I wrestle with the slippery issue of
how one distinguishes individual idiosyncrasies from other sorts of variation,
such as regional differences.
As a concept, "identity" defies easy definition. Erikson simply considers it to
be a person's "name . . . and what station he occupies in his community" (1968,
p. 61). If we broaden this a bit to include all the names and labels by which
Tukanoans are known and all the statuses and roles they play in life, we have a
fairly adequate initial definition. Given the quintessentially Western nature of the
preoccupation with identity as linked to individual identity and individualism (for
recent examples, see Burridge, 1979; Macfarlane, 1978), one runs quite a few
intellectual risks in extending the concept to small-scale non-Western societies.
To begin such an undertaking, therefore, perhaps what should be discussed first is
how the Tukanoan self differs from its more individualistic Western counterpart.
It should be noted that the Western notions of self, other, and identity as
The fish people
presented here may seem simplified and almost stereotypical without the nuances
and contradictions we know to be there. But as a foil for Tukanoan identity,
stereotypes are sufficient.
Basically, Tukanoan notions of self and other are more relational, contextual,
and evanescent than those of many Western societies, in that the more individual-
istic Western notions place greater stress on the permanence of differentiation
between self and other. Bara or other Tukanoans in contrast, although they
certainly could speak of an individual in terms of a unique intersection of
characteristics that occurs in no one else (i.e., in terms of space, time, genealogi-
cal position, gender, language, etc.), seldom seem motivated to do so. When they
distinguish X from Y or X group from Y group, they usually make the contrast in
terms of only one or two characteristics relevant to the situation at hand, leaving
other features of X and Y in abeyance. In Western society, we too make
context-specific distinctions that lump people into broad categories at the same
time they distinguish them, but as a general tendency, we are more ready than
Tukanoans to conceptualize people as distinct individuals and to make much of
the complete bundle of characteristics that makes each person unique.
Further discussion of the Tukanoan-Western contrast is saved for Chapter 13,
after the necessary ethnographic foundation has been laid. In the meantime,
however, the themes of permanence versus transience and relationality should be
kept in mind, along with the notion that the Tukanoan self, like the Bororo self,
"is created, defined, and systematically transformed by other selves; the person
does not exist except as it is reflected by these" (Crocker, 1977, p. 144).
In this book I try to show that although we can analyze Tukanoan social
identity in terms of features and dimensions, identity is both structure and action.
The abstract qualities and dimensions I have listed take on meaning only within a
contextualized action that is simultaneously moral (e.g., I am a human Tukanoan
rather than an animal if and because I continue to behave in a moral manner),
instrumental (e.g., I am able to do this because I am Bara, and would not if I
were Maku), and relational (e.g., I am Bara, which is in part defined by what I
do and am vis-a-vis Tuyuka) action. At times, contradiction results, a paradox
best seen in Tukanoan terms as a necessary outcome of sets of principles that,
when acting concurrently, overlap in incongruent fashion. Many examples of this
incongruity are given later. Of course, Western notions of self also embody
paradox and contradiction but in a somewhat different fashion, which follows
from differences in the roles played in the two systems by morality, instrumentali-
ty, and relationality (a theme also discussed further in Chapter 13). Tukanoan
society, although full of oppositions, does not oppose the individual qua individ-
ual against others nearly as much as we might expect. Instead, it transforms
oppositions into continua and it allows individuals to slide back and forth between
different positions on these continua according to the context at hand. With the
emphasis on self and other as a process, an identity to be maintained rather than
something absolute and eternal, self and other become even more flexible
4
concepts and are often literally and figuratively linked together in conceptualiza-
tion and action. Although all of this occurs in the West, I argue that it does so to a
lesser degree; we tend to see ourselves more in terms of being effectively and
permanently separated. Many markers of distinct identity exist in both systems
(among Tukanoans, the system of language identification is a prime example),
but in the West the markers tend to distinguish us as separate individuals in a
unique and more permanent manner.
The Bara (or other Tukanoan) self is indeed a distinct self, but to a greater
extent this self is constantly merged with other selves through participation -
cognitively and behaviorally - in various categories. Oppositions can involve the
self as an individual, but these are transient, and more often oppositions involve
collectivities of people. In sum, although without doubt a Bara has at his or her
disposal any number of clearly specified reference points with which to contrast
himself or herself with others as a distinct individual, I argue that the opportunity
and motive for doing so occur far less among Tukanoans than in many other
societies.
Regional perspective
Whenever possible, this book stays with the regional perspective introduced in
the Preface. As noted there, in certain respects a single Vaupes settlement is
anything but a microcosm of the larger social system. A settlement, although
often equivalent to other like units, sometimes stands in complementary, antago-
nistic, or other kinds of nonidentical relationship with other settlements. Fur-
thermore, settlements are also internally divisible along a number of structural
dimensions; they do not always stand as homogeneous, whole units. As settle-
ments are nodes in a regional network, so are their individual members and
subdivisions. Regional interaction involves, for example, marriage, residence
and visiting patterns, ceremonial and trade relationships, and, in the past,
warfare. In addition, it soon becomes obvious that Tukanoans themselves see the
Vaupes as a single system. Linguistic evidence also supports this view. Although
structurally differentiated along several dimensions, Tukanoan groups display a
remarkable degree of cultural homogeneity, using many of the same rules for
conceptualizing and participating in a single system, although at times occupying
different positions in it.
The unity of the Vaupe's social system, moreover, has been at least conceded
or implied in previous works on the area, even when they take a more particularis-
tic approach. Goldman, although confining his monographic study (1963) to a
single group, the Cubeo, notes the cosmopolitanism of the region, and in his
article in the Handbook of South American Indians (1948), he treats the Vaupe's
as a unified culture area. Briizzi Alves da Silva (1962, 1966) concerns himself
with the integration of social units in the region, emphasizing their underlying
similarity; the same is true of Fulop (1955) to a lesser extent. Reichel-Dolmatoff
5
The fish people
(1971), although more impressed by differences than similarities,also examines

the relations between language-affiliated units (see the discussions in Jackson,
1972, pp. 16-17, and S. Hugh-Jones, 1979, p. 22). Sorensen's well-known
article on Vaup6s multilingualism (1967) emphasizes how apparent diversity can
sustain integration and regional unity. Arhem (1981) also demonstrates the
existence of a complex territorial organization linked to descent and marriage
systems for Makuna in the Pira-parana.
In the last few years two new books on Tukanoan culture, by Stephen and
Christine Hugh-Jones (both 1979), discuss the importance of focusing on a
region, in this case the Pira-parand, in addition to looking at a single exoga-
mous group (the Barasana). Their books complement this one because they pay
more attention to symbolism than to the sociological concerns emphasized here.
Nonetheless, the Hugh-Joneses and I find ourselves in substantial agreement on
the form of regional integration in the Vaupes, if not on all details of its content.
In general, almost all the ethnographers of Tukanoan peoples have acknowledged
the unity of the Vaupes and grappled with the implications of that unity. I do feel
that mine is the most full-fledged regional approach; I am not castigating other
investigators for any neglect on their part but rather extending an already estab-
lished tendency (see also Goodenough, 1981, pp. 1-3).
Often it seems difficult to communicate to nonspecialists what characterizes
the Vaupes system and the lessons, both methodological and theoretical, that can
be learned from it. The kind of approach I am espousing here is discouraged by
the size of the area in question, roughly that of New England. Indeed, as pointed
out in the Preface, extremely difficult issues relating simply to size of the unit of
study must be dealt with constantly, at times in the form of rather unsatisfactory
compromises. Regions characterized by thick forest cover, no roads, frequent
rapids, and unreliable communication and transport do put obstacles in the way
of regional studies. Still, we must not allow these admittedly important consider-
ations to lead us into ignoring the reality of extralocal, and often long-range,
interaction and its symbolic significance; Tukanoans overcome such obstacles
and so must we.
Indeed, such interaction among widely dispersed local groups characterizes
virtually all low-density populations around the world, especially hunter-gatherers
such as the Shoshone, Montagnais-Naskapi, San Bushmen, or Australian Abo-
rigines, who ipso facto form regional systems because local groups are not
self-sufficient in every respect and because they exploit extensive territories.1
Dispersed hunter-gatherers, with their local group interdependence, fluidity in
territorial boundaries, and fluctuations in local group membership, offer a model,
I would suggest, for understanding the Vaupes. Although Tukanoans affiliate
with groups and categories in less flexible ways than many hunter-gatherers, and
Tukanoan institutions appear to be less amorphous and malleable, they still
preserve considerable freedom of choice within the framework of these more
rigid institutions and their rules of membership. This flexibility, operating out of
a rigidity at a more abstract and idealized level of Tukanoan social structure, is a
central concern throughout the book.
When one takes into account the Tukanoan view of the Vaupes as an
unbounded system (that is, a large geographical area in which people are
basically similar and distance and differentiation are seen in terms of degrees
rather than absolutes), one can understand why the ubiquitous term tribe does not
fit anywhere in the region. This label has been applied most often there to the
language-affiliated units encompassing a number of local settlements, but this
usage leads to so much confusion that I have substituted language group in its
place. The substitution brings its own difficulties, as we shall see, but they pale in
comparison with the confusions of tribe.
Anthropologists have not been able to agree on a single definition for the word
tribe, although many of the same definitional criteria turn up repeatedly (see
Fried, 1975; Godelier, 1977; Helm, 1968). These include (1) a shared territory
with discrete boundaries, (2) statuses and organizations for the tribe as a whole,
(3) more interaction within the tribe than without, (4) more marriage within than
without, (5) significant cultural differences with neighboring units, (6) a shared
tribal language. Singly or in combination, these criteria do not work for Tukanoan
language groups.
By definition, the members of a language group share a patrilineally inherited
affiliation with a language, but the problems encountered when trying to use
language as a criterion of tribal membership are manifold (Hymes, 1968). Very
infrequently do all members of a language group occupy a discrete territory that
is considered theirs and to which they have exclusive or almost exclusive rights.
This state of affairs is most closely approached in the Pira-parand region (see
Arhem, 1981; Bidou, 1976), but in general we may speak only of segments of
language groups occupying continuous stretches of rivers, and even in the
Pir£-parana\ where occupation of a continuous territory by all member settle-
ments of a language group may be said by them to be the ideal,' 'in practice there
is considerable overlap between descent-group territories" (C. Hugh-Jones,
1979, p. 25).
The criterion of statuses or organizations whose point of reference is the tribe
as a whole (with the implicit assumption that the tribe, at least to that extent, is a
corporate body), does not work any better in the Vaup6s. Activities involving
everyone in a language group, or an individual or group selected to represent all
the others, are conspicuous in their absence. Language groups may perhaps have
acted as political units in the past, but they certainly do not today. Although roles
in ritual often incorporate aspects of language group identity, this identification
does not make them "tribal" roles.
Because language groups are exogamous, the criterion of tribal endogamy, of
course, works not at all. And, given language group exogamy and settlement
The fish people
exogamy, people deal almost as frequently with members of other language

groups as with members of their own. Indeed, the role of language group
exogamy, more than anything else, dooms the concept of tribe from the start.
The last distinction, which assumes relative cultural uniformity within tribes
and sharp cultural differences between them, runs up against the homogeneity of
Tukanoan culture pointed out earlier. Settlements and language groups do differ
culturally, and I will spend a good deal of time describing and accounting for
such differences, but they do not produce tribes. Many of the most obvious and
frequently mentioned differences are emblematic and superficial; they fulfill the
need for social markers differentiating the units in the regional system, and as
such they point toward social and cultural integration and underlying homogene-
ity rather than heterogeneity. This is not to say that true cultural differences
cannot be found in so vast an area, variations that can be ascribed to disparate
origins, to cultural drift in different directions, to the effects of diffusion from
neighbors outside the system, and to different degrees of acculturation. None of
this variation, however, coincides in any simple fashion with the language-
affiliated units traditionally called tribes.2
Fluidity
One of the central theoretical concerns in this book is the interplay between
fluidity and apparent rigidity found in many Tukanoan social institutions. The
effects of choice and manipulation show themselves in household composition,
political organization, and almost every situation in which seemingly inelastic
and unchanging principles assign people to groups and categories. I argue that
this paradoxical juxtaposition is inherent in many small-scale societies. Variabil-
ity and invariance, operating at different levels, support and play off each other,
and in the long run probably allow societies to adapt to fluctuations in their
ecological and demographic bases.
General conclusions on this interplay must wait until the concluding chapter,
after the material showing its importance has emerged. At this juncture I need
only point out that the issue of fluidity and rigidity cannot be separated from
others already introduced: the nature of Tukanoan social identity and the utility
of a regional perspective on the Vaupes.
Fluidity also applies to the issue of the boundaries of a system like the Vaupes.
I have stated that the Vaupes is regionally organized into an open-ended system
that cannot properly be called a society in the sense of a bounded social unit (C.
Hugh-Jones, 1979, pp. xv, xvi). I am not merely saying that at places the
boundaries are fuzzy or even nonexistent but that the system presupposes no
boundaries except single-dimensional, and therefore quite arbitrary, ones (for
example, at times Tukanoans use the criterion of language exogamy to exclude
the Cubeo but clearly include them - and properly so - in the system most of the
time).
8
A big challenge I faced in writing this book was to describe the fluidity of the
system comprehensively and yet also to account for why the Vaupes, as a
system, works. And work it does: Its social structure "does" things such as
assign roles, get people married properly (or show why they have married
improperly), and point out the way to each generation on its journey to full-
fledged adult status. This is accomplished with varying combinations of the
opposites of fluidity and rigidity and difference and equivalence, but this is very
difficult to describe. Tukanoans share many cultural and social traits: related
languages and speaking patterns, intermarrying settlements, basic subsistence
patterns, similar kinship terminologies (as well as virtually all other semantic
domains), a common mythology, and so forth. Some features, moreover, nor-
mally associated with cultural and social cleavages are not to be so viewed in the
Vaupes case; examples are territorial dispersal and language differences. Find-
ing out what these differences, equivalences, and identities "really" are is not
only a problem, it is the problem. Confusion seems to reign at times, not all of
which is trivial and easily cleared up by acquiring more information or by
increasing one's awareness of one's anthropological or ethnocentric assumptions
or by some other similar scholarly operation. Confusion, when explained, is of
course no longer confusion, but I do not believe that the Vaupes system -
capable though it is of being described, explained, and formulated into models -
can ever be reduced to a coherent, logical, unitary system. Confusion gives way
to paradox, and this paradox, along with the dynamism existing between levels in
the model (i.e., fluidity and manipulability) remain. The process of discovering
the Tukanoan system, incidentally, teaches us much about our firmly built-in
epistemological constraints, in both language and modes of conceptualization,
which will become more apparent when examples of such fluidity in action are
given.
Ideal and real

The polarity between rigidity and fluidity resembles and sometimes overlaps
another dichotomy, one between ideal models of society and culture on the one
hand and so-called on-the-ground reality on the other. Although ethnographers
and informants deal in both commodities, and almost every interpretation or
piece of data inevitably mixes the two together, cultural accounts tend to empha-
size one or another. According to an anecdote narrated by Napoleon Chagnon,3
L6vi-Strauss likened society to a chambered nautilus. This mollusk combines a
beautiful structure - its shell - with a singularly ugly and slimy creature inside.
Some choose to study the shell, whereas others prefer to examine its inhabitant.
The trick is to keep oneself aware of one's preferences and of the reality of both
parts of the animal, without deluding oneself into thinking that only one part is
real or important. Seeing the two together, so interconnected and yet so contradic-
tory, is not always easy. The ugly little creature inside is responsible for the
9
The fish people
existence of the sublime shell, yet the shell far outlasts its creator and has a logic
and structure - an identity - of its own, separate from its occupant's.
As little as we may like seeing ourselves as mollusks, the societies and cultures
created by human beings embody the same duality. Structures, abstracted at one
level or another of analysis from the behavior that generates them, deserve the
absorbed attention given them by many anthropologists (including, of course,
L6vi-Strauss himself). Yet social structures do not create themselves nor myths
think themselves, any more than shells generate spontaneously; it is individuals
and individuals in collectivities, by engaging in discrete bits of behavior that
often are so seemingly lacking in logical structure, who do. When formalized
into descriptive or explanatory models, some of the empirical validity of these
behaviors is lost. That the sublime structures created are separable from and
outlive their creators, be they cephalopods, Tukanoans, or anthropologists, may
be our pleasure or our sorrow, but it is always our frustration to some extent,
because some of the richness and detail of the reality that produced the structures
is lost.
Chambered nautiluses, however, cannot, so far as I know, have false con-
sciousness; they create their shells, each one distinct from all others in some
minor respects but all of them basically to type, because that is what they are
genetically programmed to make. Human beings, on the other hand, create
various kinds of structures that can be hidden by conscious models, models that
tell us little about the structures because they are intended to perpetuate certain
phenomena rather than explain them (Levi-Strauss, 1962, p. 324). The Vaupes
illustrates the problems encountered in this regard par excellence. We might
prefer to study the equivalent of the shell or of its occupant; but the two are far
more interconnected than those of the chambered nautilus, and we perforce must
study both when encountering Tukanoans. For example, the fact that structures
outlive the individuals creating them, be they mollusks or human beings, is
extremely important: If the word dialectic, surely the most overused word at
present, applies at all to this problem, it does so in terms of the way in which
individuals and structures influence and even create behavior and idealization
(both native and ethnological). Paradoxically, whatever exists and can be found
on the ground is impermanent, whereas what is unreal in this sense - what is
structure or idealization and derived from the real-is permanent, or at least more
enduring. Here we run the risk of falling into a positivist (perhaps the second
most overused word) trap, because we.are not looking at shells or organisms but
at behavior that has no meaning without explanation. The trap is a trap because it
assumes that somewhere there is reality without idealization. Still, idealized
structure continually influences the behavioral system (which is why we cannot
talk about "reality" versus idealized expressions of that reality even when we use
the most precise of measuring instruments or statistical tests), a system that is
always both being and becoming.
In addition to the relative permanence of structures and behavior, the ideal-
10
"real" dichotomy implies another temporal question - how one deals with
variation over time. Change can be cyclical or directional, and a short-term
observer often cannot tell whether a particular phenomenon or range of phenom-
ena represents variation around some relatively stable standard or is a harbinger
of the new state a system is moving toward. (Nautiluses take much longer in
changing the shape of their shells than societies do.) The analysis presented here
depicts Tukanoan society at one particular moment. My model is, I believe, an
understandable, accurate, and tolerably sophisticated depiction of the system,
and I do not think that huge amounts of historical research (particularly given the
state of documentary sources available) would change the basic parameters given
here. But to the extent that it takes account of change, the analysis treats it as a
disruption or modification of a synchronic system rather than as part of a
diachronic process. We know a good deal already about such forces of change as
missionization and the rubber boom and about their highly variable impact on
different populations in the region. Although I try to keep such change in mind
throughout the book, and in some places speculate about earlier developments in
the evolution of the regional system, my concern with depicting a traditional
culture in terms applicable across the region makes it impossible to encompass
change and local variation as thoroughly as a more diachronic or a more localized
model would certainly do.
Final remarks
In addition to these broad concerns, the following pages bear on a number of
familiar issues of slightly narrower scope, such as descent and alliance, geneal-
ogy and category, the position of women in horticulturalist societies, and similar
topics. Although I sometimes push hard for a certain interpretation or stance on
an issue, the book does not in any comprehensive way represent a particular
school or tradition in social anthropology. My general assumptions and approaches
obviously derive from the schools of thought current in recent years, but when I
come out on one side of an issue, I am not necessarily aligning myself with the
school identified with that interpretation. More often I have latched on to it in an
eclectic manner because it makes sense of the data at hand.
This book tends less toward a particular interpretative stance than toward a
generally sociological approach. As a consequence, although I use myths to
illustrate some of my points about marriage or genealogy, I pay relatively little
attention to symbol, cosmology, and ritual. In recent years the ethnology of
lowland South America has come to the fore in precisely these areas. Structuralist
and symbolic approaches do not, however, fall in my areas of specialization, and
the appearance of several works of this sort on the Vaupes by Arhem, Bidou,
Goldman, Langdon, Reichel-Dolmatoff, and others, and especially the excellent
recent works by Stephen and Christine Hugh-Jones, convinces me that I can
safely leave this domain to my fellow ethnographers of the region.
11
The fish people
To keep publishing costs down, I have included only the quantitative material
absolutely necessary to the arguments presented in this book. Readers interested
in fuller discussion of, or quantitative data about, a particular topic (for example,
names and locations of sibs - the localized patrilineal descent groups in the
Vaupe"s), should consult my other writings, listed in the References.
12
2.
Introduction to the Central Northwest
Amazon
Ecological setting
The Vaupes region, roughly the size of New England, lies approximately
between the equator and 1° north latitude, and between 69° and 71° west
longitude (see Map I). 1 It is bisected by the Colombia-Brazil border, and in
terms of political boundaries its territory forms part of the Brazilian Territorio
Nacional do Amazonas and the Colombian Comisaria del Vaupes. The latter
comprises 90,625 square kilometers (Instituto Geografico "Agustin Codazzi,"
1969). Humid, tropical rainforest covers the entire region. This vegetation is
typical of the whole western extension of the interfluvial Guiana Highlands
(Moser and Tayler, 1963, p. 440), of which the Vaupes forms part, being more
densely wooded and shorter than the canopy forests typical of the Amazon basin
itself. The cerros, or flat-topped and domed hills that dot the landscape, date
from the Guiana Shield epoch (Reichel-Dolmatoff, 1975, p. 64). Visible from far
away, these cerros stand out in an otherwise largely undifferentiated extent of
forest, river, and smaller hills. Their prominence makes them important land-
marks for Tukanoans, as are the swamps dominated by the miriti palm (Mauritia
flexuosa). The land generally slopes eastward, and high open lands and exposed
rock and caves are typical of the western sections. Savanna is found both to the
south and north but not in the region itself.
The major rainy season lasts from April to August, the lesser for a few weeks in
October or November (Instituto Geografico "Agustin Codazzi," 1969, p. 67).
Because the dry and rainy seasons differ largely in terms of volume of rain - great
versus enormous — river height most obviously distinguishes dry from rainy
periods. The rivers, however, also fluctuate a great deal from day to day, and river
height correlates only roughly with time of year. The effects of this fluctuation
are extremely important for Tukanoan travelers, especially when canoeing with
cargo. When rivers are at their lowest, crossing from one to another demands
extra skill and endurance. Not only must travelers portage canoe and cargo
overland, but at each end of their portage they may find the river reduced to a
succession of shallow puddles. Under such conditions, they can proceed only by
raising the water level with a series of earth dams, difficult and tedious to do.
13
Cuduyari
River
0°
Map 1. The Eastern Colombian Vaupes. The sources for this map,include discussions
with Tukanoans during fieldwork in 1968-70 and the Mapa Fisico-Politico de la
Reptiblica de Colombia (Bogot£: Instituto Geografico "Augustin Codazzi," 1968).
14
*»•• w International boundary (Colombia and Brazil)
Mitu (Capital of the Comisarfa del Vaupes)
Catholic mission (in 1970)
Important trails
A Ptfmanaka buro (f ieldwork site)
0^^ . W
Scale in mile
SmSm
The section of this map depicting the Pira-parana' region has been derived mainly from
C. Hugh-Jones 1979, maps 1 and 2, and Arhem 1981, map 1. Because these maps are not
in agreement, representation of the area is approximate.
15
The fish people
Other seasonal markers are position of the constellations, availability of vari-

ous wild fruits and other forest products, harvest time for certain crops such as
corn (manioc itself is not seasonal, but the preparation of fields is certainly
geared to the yearly calendar), and animals and animal behavior, in particular the
availability offish and game (S. Hugh-Jones, 1980).
The minimum temperature varies between 10° and 20°C and the maximum
between 34° and 40° (Reichel-Dolmatoff, 1971, p. 4). A rather strange period of
weather called the aru or friagem (S. Hugh-Jones, 1979, p. 19), occurring for
approximately four days at the height of the rainy season, consists of cold
temperatures, a drizzling and distinctly untropical rain, and in some cases a
strong wind (Moser and Tayler, 1963, p. 440). This phenomenon, caused by
winds forming in the Atlantic and traveling westward, occurs every year through-
out the Amazon.
The Vaupes is one of the headwater regions of the Northwest Amazon River
drainage; in the north and northeast the rivers form part of the Orinoco system.
Some Vaupes rivers join the Amazon directly to the south via the Caqueta and
Japura rivers, but most flow east into the Rio Negro and do not meet the Amazon
itself until Manaus. The crystalline beds over which rivers of the region flow give
them their well-known inky black color and relatively acidic water (M. Bates,
1965, p. 178); as a result they lack much of the aquatic life found in slower-
moving rivers with deep beds of sediment, such as the Guaviare to the north.
Surely the most impressive characteristic of Vaupes rivers is their rapids and
cataracts. Local Spanish and Ftortuguese words for these features of the landscape
(cachiveras, raudales) invariably take on powerful associations for any visitor to
the region. Travelers, rubber gatherers, missionaries, and Tukanoans alike share a
healthy respect for the hazard and obstacle to travel that these rapids present, and
those who have not experienced accidents involving injury, death, and total loss
of cargo invariably know others who have. All of the major rivers in the region
have several major and minor cataracts, although the Vaupes River is free of
them upstream of the Yurupari Rapids. As may be expected, rapids are important
landmarks for Tukanoans, who frequently choose them as longhouse and village
sites, except in the Pira-paran£ region, where settlements tend to be hidden
some distance from rivers. Also as may be expected, rapids have great significance
in Tukanoan myth and cosmology.
Petroglyphs, or rock carvings, are found at some rapids sites. Goldman
observed Tukanoans periodically freshening the etched-in figures and geometric
designs (1963, p. 8), although Moser and Tayler (1963, p. 442) discerned little
resemblance to contemporary Tukano art (cf. also Reichel-Dolmatoff, 1967).
Apart from cerros, swamps, and rapids, the natural environment tends towards
homogeneity. Clay, quartz, and a few other materials are scarce or absent in some
areas, but a trade network evens out the distribution of both raw materials and
manufactured items. Local differences in such ecological variables as the pres-
ence or absence of leaf-cutter ants or the fertility of soil affect each settlement's
16
Introduction to the Central Northwest Amazon
adaptation, as do the size of the river on which it is located and its relative
position upstream or downstream. By and large, however, the land supports
agriculture everywhere, and the overall homogeneity of the environment does not
encourage much local specialization or social differentiation.
The population
Population density for the Vaupes is quite low; one approximation is 0.2
inhabitants per square kilometer (Instituto Geogrdfico "Agustin Codazzi,"
1969, p. xiii). Recent population estimates vary between 7,000 (Reichel-
Dolmatoff, 1975, p. 64) and 13,403 (Instituto Geografico "Agustin Codazzi,"
1969, p. xi). The latter figure includes non-Indians, approximately 1,000, mostly
in Mitu and its environs.
Because most Tukanoans live on the rivers, the region appears to be more
densely populated than it is. Missionary efforts have increased nucleation by
encouraging both the merger of longhouse settlements and movement to mission
towns. This intervention has disrupted the traditional social organization of the
region and degraded the environment near all mission towns.
The rubber boom earlier in the century devastated the indigenous populations
of the region. Rubber gatherers "recruited" labor through the most viciously
coercive techniques, scattering the settlements of those who managed to avoid
them. When given advance warning, Tukanoans sometimes chose to flee into the
forest during the recruiting season as certain rubber gatherers came into the
neighborhood. Often hiding for months at a time, refugees experienced great
privation camping in the forest, with their gardens inadequately tended. The
police and armies of both Brazil and Colombia frequently helped the rubber
gatherers recruit Tukanoans and apprehend runaways; however, the authorities of
one country could not cross into the other, and Tukanoans became adept at
border-hopping. Quite a few of the family and sib (localized exogamous patrilin-
eal descent groups) histories I collected spoke eloquently of this disruption and
dislocation. It is also true, however, that Tukanoans sometimes sought out whites,
including rubber gatherers, as sources of trade goods.
Any settlement map of the region graphically illustrates Tukanoan awareness
of the international border, which has continued up to the present. The stretch of
the Papuri between Melo Franco and Yavarete serves as the boundary between
the two countries. Except for three mission towns, all Tukanoan settlements are
on the Brazilian side, where Colombian Catholic missionaries, Javerians, have
no jurisdiction. The closest Brazilian mission is the one at Yavaret6, run by
Salesians. Although this might be taken as evidence of a Tukanoan preference for
Salesians over Javerians, the preference in fact is probably for the mission most
removed from Tukanoans' daily affairs. Howard Reid (personal communication),
an anthropologist who in his studies of the Brazilian Makfi has had prolonged
contact with Salesians, feels they pit themselves against traditional Vaupes
17
The fish people
culture more than the Javerians. He points out as evidence the complete lack of
longhouses in the Brazilian Amazon - Salesians burned out the last one, on the
upper Tiquie", in the late 1960s.
Some Tukanoans have emigrated permanently from the Vaupes, although just
how many is hard to say. A decision to leave the region permanently usually
follows a period of work at a rubber camp, where the potential out-migrant
acquires a taste for the rural Colombian (non-Indian) lifestyle. Some of these men
move as far as San Jose" de Guaviare or even Villavicencio, towns experiencing
moderate booms, although even there the chances of finding work are slim. MM
is closer but prospects are even worse.
Goldman (1963) reports a higher population density for the Cubeo during his
period of fieldwork in 1939 than is the case now. If this decline holds for all of the
Vaup6s, it is difficult to be sure what causes are most to blame, although
epidemic disease, the depredations of the rubber industry, out-migration, and the
loss of people to mission towns have obviously all contributed.
Whether the Vaup6s environment could at present sustain the large popula-
tions reported for the floodplains by such early writers as Orellana (Goldman,
1963, p. 2) is an open question. Certainly land is not a limiting factor for
Tukanoans at present. Nor was it for the Cubeo of 1939-40, although Goldman
does state that available land for manioc plantations was a decisive factor in
choosing a longhouse site, because sibs had to retain their traditional boundaries
(1981, p. 3). Still, he notes that "from terrain alone there is no doubt that the
Vaupes could have supported far greater Indian populations than it did at the time
of first contact" (1963, p. 36). On the other hand, the concentration of popula-
tion in mission towns has created real land shortages there; some women have to
canoe and walk a substantial distance to reach their fields.
Epidemic diseases introduced by whites, which include influenza, measles,
and whooping cough, have undoubtedly contributed to the depopulation of the
region. Tukanoans have a horror of all three, and with reason. A single epidemic
can ravage the population of a large area if it lacks immunity. For example, boa,
the Bard name of the river on which the longhouse I stayed at was located
("rotten," "putrid"), refers to an epidemic early in this century that killed or
seriously sickened everyone living on the river. Those who could do so fled the
area. Even today only five local descent groups survive on the river, half the
number present before the epidemic.
Reid, reporting on the health and nutritional status in the eastern Vaup6s in
1974, reports malaria and tuberculosis as the only major chronic endemic
diseases (1976, p. 20); he notes, however (personal communication), that
onchocerciasis (African river blindness) was confirmed in the Brazilian Vaupes
in 1975. The endemic eye, skin, and internal conditions caused by fungal,
helminth, nematode, and insect parasites, although not classified as major ill-
nesses, can last for years and cause intense and disabling pain. Any sore heals
slowly in so humid an area, increasing the likelihood of infection and spread.
18
Conjunctivitis can cause excruciating pain and near-blindness for weeks at a time,
and fungus infections affecting the feet are especially hard to cure, because it is
impossible to keep the feet dry or clean. Everyone plays host to a variety of
intestinal worms.
Speaking in the most general terms, one can say that population size has
declined from its levels during earlier periods; virtually all early reports lead one
to this conclusion. Beyond this, the paucity of hard data makes it impossible even
to speculate about specifics such as rate of decline. It is also clear that although
overall population and settlement size have decreased, cultivation has relatively
increased. Further discussion of some of the issues related to the matter of
population size is found in the treatments of subsistence patterns in Chapter 3 and
acculturation in Chapter 12.
Language and linguistics

Language families represented in the Vaupes include Eastern Tukanoan, Ara-
wak, Carib, Maku, and, in a limited way, Tupian. Lingua Geral (called lingoa
Geral in Portuguese and nheengatti in 1\ipian) is a Tupian trade language used
throughout the region in the past. It has been almost entirely replaced by Tukano,
the traditional lingua franca of the region. Lingua Geral was introduced by Jesuits
and other missionaries coming into the region from Brazil. Some Tukanoans
spoke it, though few fluently. No one speaks it in the region any more, although
some older men remember words and phrases. Its only continuing importance is
as the source of names of places, plants, and animals in the local Spanish and
Portuguese dialects. Undoubtedly the presence of two national languages rather
than one increases the importance of Tukano as the unifying language of the
region (Sorensen, 1967, p. 680).
The Eastern Tukanoan language family, which is identified with the Central
Northwest Amazon, appears to be the oldest in the region. Apart from Lingua
Geral itself, Tupian cultural influences go back many years (Goldman, 1963, p.
14) Arawakan languages and cultural influences, on the other hand, have arrived
more recently, as have Carib languages.
Specific named linguistic varieties presently spoken in the Vaupes include:
Carib (Carihona [Umaua]);2 Arawakan (Baniwa, Cabiyeri, Curripaco, Tariano);
Eastern Tukanoan (Bara, Barasana [Paneroa], Carapana, Cubeo, Desana,
Uanano, Makuna, Piratapuya, Siriano, Tatuyo, Tukano, Tuyuka, and YurutQ.
Waltz and Wheeler (1970) also mention twenty speakers of Papiwa (possibly
known in Bara as Wahiind). Named linguistic varieties spoken directly to the
south of the Vaupes include Yukuna and Matapi (Arawakan), and Letuama and
Tanimuca (Tukanoan).3 Other linguistic varieties named in the literature are
almost certainly dialects understood by speakers of "languages" on the list;
Taiwano (or Taibano) for instance, is mutually intelligible with Barasana (S.
Hugh-Jones, personal communication; see also Waltz and Wheeler [1970, p.
19
The fish people
120], who note that the Barasana intermarry with Taiwanos "whose language is
almost identical with their own"). Sources disagree on the status of certain
varieties. Linguists (see note 3) of the Summer Institute of Linguistics (SIL)
consider the languages in the primary list just given to be mutually unintelligible,
as does Sorensen, who states that the most closely related pair in his list of
thirteen Tukanoan languages is "considerably more distant . . . than Jutish is
from Standard Danish" (1967, p. 674). He adds that Tukanoan languages in
general seem to be less closely related than Central Algonquian or Romance
languages. Areas of differentiation include grammar and lexicon and, to a lesser
extent, phonology. Close relationships obviously link several pairs on the list,
such as Barasana and Makuna (S. Hugh-Jones, personal communication), Desana
and Siriano, which share 98.9 percent cognates in sound correspondence tests
carried out by Waltz and Wheeler (1970), and Piratapuya and Uanano, with 99.2
percent cognates. Obviously the multilingualism of all Tukanoans except some
Cubeo complicates efforts to determine mutual intelligibility.
To sum up, the linguistic situation in the Vaupds is enormously complex;
linguists disagree about the number and identity of languages in the region and
the criteria to be used in identifying them. Some of the causes of this complexity
are already apparent: the presence of multilingualism, the cultural-semantic
equivalences shared by all the languages, and linguistic exogamy. Others will be
discussed in Chapter 9. It might seem presumptuous to call into question
conclusions drawn from comparative linguistic research without having done
such work myself, except that the experts also doubt one another's conclusions.
Problems are bound to arise when attempting to enumerate linguistic varieties in
a situation of 100 percent multilingualism, in which, moreover, the native
speakers have their own nonlinguistic reasons for treating all the named varieties
as mutually unintelligible. In this respect the concepts "language" and "tribe"
present similar difficulties. There is a sense in which closely related linguistic
varieties should be considered discrete units (regardless of whether an omniscient
observer would call them languages or dialects), so long as native members of the
system consider them so. It is also true that the languages of the Vaupes form a
regional system in terms of phonology, grammar, and lexicon, with regular sound
shifts (the subject of the Waltz and Wheeler article), inversions, and transforma-
tions that are very like those of the social system. Thus, we also have to look at
everyone's (again, Tukanoans, ethnologists, linguists) assumptions about just
what a language is and what purposes or motives lie behind a concern with such
questions in the first place. Tukanoans are concerned because they marry people
identified with different linguistic varieties. SIL needs to know in order to fulfill
its goal of translating the Bible into all the languages of the world, as the title of
an SIL publication by Wallis and Bennett (1959), Two Thousand Tongues to Go,
indicates.4
20
Ethnic history
It should be apparent by now that one could hardly ask for a more demanding and
frustrating task than to attempt to piece together the linguistic and ethnic history
of the Vaupes. The apparent ease with which traits and languages are picked up
and dropped only makes the problem more difficult. This cosmopolitanism is
discussed by Goldman (1963, p. 15), and Koch-Grunberg (1909-10, vol. 2, pp.
65, 66, 81) provides evidence for exchanges of cultural traits and languages. The
multilingualism and linguistic exogamy among all groups except the Cubeo also
illustrate this.
Information on the cultural and linguistic history of the region is extremely
limited, however, and although a plurality of discrete cultural traditions and
languages characterizes the region, no assumptions can be made concerning
isomorphisms between linguistic or cultural units and the actual populations that
have maintained these traditions over time. Cultural and linguistic traditions can
be easily learned or just as easily forgotten (in favor of what may be radically
different ones) over a generation. Around the world, cultures vary greatly with
respect to their conservatism or openness in their willingness to accept intrusive
ideas and customs. This is undoubtedly an important area of research in the
Vaupe's, given what Goldman, Koch-Griinberg, Reichel-Dolmatoff, and others
say regarding the cosmopolitanism of the region, and it is obvious that any
attempts to piece together the histories of ethnic groups in the area will never
reconstruct the complete picture.
Reichel-Dolmatoff (1975, p. 66) finds a "kernel of historical truth" in the
various Tukanoan origin myths, which invariably trace a particular group's
history as one of westward, upstream migration from downriver sites in Brazil.
On the basis of these histories, he suggests that Maku, "a substratum of
band-based hunters," were the earliest inhabitants and were followed later by
Arawakans, who were subsequently invaded by Tukanoan speakers. This
Arawak-Tukanoan sequence is the reverse of the scheme proposed by Goldman
(1963, p. 14), yet such differences of opinion must inevitably occur when
working with oral histories. I collected quasihistorical accounts, concerned with
both the original peopling of the region and more recent migrations; these too
undoubtedly contain "kernels of truth." The migrations described, although
telescoped in the narratives, were almost certainly spread out over a number of
years and even generations. The often epic accounts of war, treachery, and
invasion also undoubtedly depict to some degree what actually happened.
The few general historical trends that we can be certain of are the following: (1)
Population density and settlement size have declined markedly over the past
hundred and fifty years, and at one time, some settlements may have consisted of
two or more rather than single longhouses (S. Hugh-Jones, 1979, p. 25); (2) the
ancestors of some present-day inhabitants migrated upstream from the east into
21
The fish people
the region; (3) raiding and feuding, a general though intermittent fact of life in the
past, had largely died out by the time the oldest Tukanoans alive today were born.
The cessation of raiding and the population decline undoubtedly came about in
part from white influence in the form of introduced trade goods (among which
machetes and axes very probably had a substantial impact on subsistence produc-
tivity), disease, and the appearance of an external enemy. The rubber boom, in
particular, severely disrupted every aspect of Tukanoan life. All of these devel-
opments will be discussed further.
The diffusionist or acculturational model of change, characterized by a pattern
of regularized and gradual accretion of foreign traits, will not work for the
Vaupe"s. Indeed, the inadequacy of such passive models as explanations of either
stability or change in language and culture is one very clear lesson of this book.
No archaeology has been carried out in the Vaupes, and travelers in the
nineteenth and early twentieth centuries who might have been far more informa-
tive about the earlier situation (albeit post-contact even then) did not concern
themselves with such issues. Consequently, the available information is scant,
imprecise, and highly conjectural.
Early explorer and missionary efforts

Doubtless the paucity of information on the Vaup6s from early explorers can be
attributed to its inaccessibility and hostile landscape. The territories to the north,
in contrast, which consist of open llanos (savanna) and unobstructed slow-
moving rivers, were penetrated by explorers much earlier, as were the forested
lands to the east, where the rivers, although subject to flooding, could be
navigated more easily than those farther upstream. Along with the climate and
terrain, national boundaries contributed to the region's isolation. Much of the
Vaup6s falls within the borders of Colombia, but access from the west over the
cordillera is nearly impossible. The Brazilian Vaup6s, although more accessible,
lies at the extreme periphery of the country, hundreds of miles from the nearest
major town. Thus until the introduction of the airplane in about 1935, exploration
of the area was full of risk and demanded large, well-organized expeditions
originating far downstream. Examples of the difficulties encountered can be
found in accounts written by H. W. Bates (1864), Coudreau (1887), Spruce
(1908), A. R. Wallace (1889/1972), and Whiffen (1915), some of the better-
known explorers and naturalists who traveled to the region. In general, the
Vaupes is and always has been peripheral to Colombia and Brazil, both geo-
graphically and (with a few short-lived exceptions) economically.
The earliest explorers were the conquistadors looking for the famed "Dorado
de Los Omagua" in the first half of the sixteenth century. Their expeditions left
little in the way of documentation, apart from passing mention of "Gaupe"
Indians."5
Although the Dominicans and Franciscans founded missions in neighboring
22
areas, the first mission (set up by the Carmelites) did not appear in the Vaupes
until 1852. Neither this nor subsequent attempts took hold until 1914, when the
Dutch Catholic Monfortian Congregation built two missions on the Papuri River
at Monfort and Teresita. Later, in 1929, the Salesians established a mission at
Iauarete' (Yavarete). The political unit encompassing the region, the Comisaria
del Vaup6s, was created in 1910. In 1949 the Order of St. Javier founded the
Prefectura Apost61ica de MM, which has since administered all Catholic
missions in the Colombian section of the region. Almost all the clergy of this
order are Colombian, and many come from a single region, Antioquia. This
domestic domination of mission staffing contrasts starkly with the foreign flavor
of Catholic missionization throughout most of South America and even with
early missionization of the Vaupes. In most South American countries a lack of
native-born priests and nuns forces the Catholic orders to import missionaries
from other, mostly European, countries. (Italian Salesians, for instance, oversee
the missionary effort in the Brazilian Vaupes.) Colombia, however, even exports
missionaries.
Protestantism has had a great impact on the Vaupes since its appearance in the
late 1940s. One source estimates that as many as a third of the inhabitants of the
combined comisarias of the Vaup6s and Guainia are at least nominally Protes-
tant,6 a result attributable to the enthusiastic efforts of the Summer Institute of
Linguistics, and the years of work of the near-legendary New Tribes Mission
evangelist, Sophia Muller.7
Early and recent ethnographic descriptions

One of the earliest commentators on the Vaup6s was Humboldt (1822). He was
followed later in the nineteenth century by the naturalists H. W Bates (1864), Cou-
dreau (1887), Spruce (1908), Stradelli (1890), and A. R. Wallace (1889/1972).
The first systematic work by an ethnologist came from Koch-Grunberg (1909-
10), who visited the Cubeo on the Cuduyari. as well as various groups on the Vaup6s
and Tiqui6 rivers. (He did not enter the Papuri region.) His work is voluminous
and impressive, particularly in its comprehensive surveys of language, material
culture, and some of the more visible customs and technologies. He is an
especially valuable source on customs and manufactures now abandoned. Other
explorers and ethnographers who were in the Vaup6s during the first half of this
century include Nimuendaju (1950), Rice (1910, 1914), and Whiffen (1915).
McGovern (1927) and MacCreagh (1926), primarily explorers, provide many
valuable and interesting pages of reading.
Modern ethnography in the Vaupe"s begins with Irving Goldman, who carried
out fieldwork among the Bahukiwa sib of the Cubeo on the Cuduyari River in
1939 (1948, 1963, 1964). He returned in 1970 to study the Cubeo of a much
higher-ranking sib (1976, 1977, 1981). Gerardo Reichel-Dolmatoff, among his
numerous ethnographic and archaeological publications on different peoples and
23
The fish people
regions of Colombia, has written two monographs (1971, 1975) and a number of
articles on the Vaup6s (1967, 1976). His first book, based on information
acquired from Antonio Guzma"n. a remarkable Desana informant, deals with
Desana cosmology and symbolism. The second analyzes Tukanoan cosmology
and symbolism in terms of the use and meaning of hallucinogens, primarily
banisteriopsis (Banisteriopsis inebrians; Banisteriopsis rusbyana). Two Colom-
bian anthropologists, Marcos Fulop (1955, 1956) and Alvaro Soto Holguin
(1972), have published on the Tukano and Cubeo, respectively. Rodriguez Lamus
has written on Tukano architecture (1958). Biocca (1965) and Moser and Tayler
(1963, 1965) have also published in recent years on the Vaup6s. Missionaries
who have published studies include Briizzi Alves da Silva (1962, 1966), Giacone
(1949), and Kok (1925-26). Recent studies based on long-term field research in
the area include Arhem, 1976, 1980, 1981; Bidou, 1972, 1976, 1977; C.
Hugh-Jones, 1977, 1978, 1979; S. Hugh-Jones, 1977, 1979; Langdon, 1975;
Silverwood-Cope, 1972; Sorensen, 1967, 1970; and Torres Laborde, 1969.
The Maku
The peoples I call Tukanoan share the Vaupes region with another native
population, the Maku. In contrast with the riverine Tukanoans, Maku generally
live in the forest away from the rivers and lack the elaborate longhouses, canoe
and fishing technologies, and ceremonial patterns characteristic of Tukanoans.
They do not follow a rule of linguistic exogamy, and they tolerate settlement
endogamy far more than Tukanoans do (Silverwood-Cope, 1972).
Once classified as an isolated language, Maku was placed by Greenberg
(1960) in his Macro-Tukanoan family. Since Greenberg, it has become clear that
there are actually at least two Maku languages (Silverwood-Cope, 1972; Cathcart,
1973).
The Maku, who are considered inferiors by their riverine neighbors, fre-
quently form temporary servant-master relationships with Desana, Cubeo,
Uanano, or other Tukanoan communities. In such arrangements, Makii provide
game, manufacture items, and perform odd jobs in exchange for cultivated
foodstuffs and white trade goods. Maku, however, do grow their own crops, and
it is unlikely that they were ever completely nomadic or nonhorticultural.8 Makti
are further discussed in Chapter 8.
A note on acculturation
Visitors to the Vaup6s as well as Tukanoans cannot help noticing the highly
divergent goals of outsiders resident there. Individuals and organizations not only
compete with each other, they often slip into outright hostility. Because the
primary discussion of the effects of outside intrusions does not come up until
Chapter 12, I offer a brief sketch of this subject here.
24
Mini, an airstrip town of approximately 1,000 inhabitants in 1970, forms the

administrative center for the Colombian Vaupes. Catholic mission towns never
have more than a resident priest and a few nuns and brothers, and thus Mini is
the only settlement in the eastern Colombian part of the region with a substantial
nonindigenous population. Agencies in Mitii affecting indigenous life include
(1) the clinic, with a dispensary and about ten beds; (2) the comisariato, the
government-run store; (3) the various representatives of the Colombian govern-
ment, such as the comisario, or governor, who play an increasing role in
Tukanoan affairs; (4) the Prefectura, the administrative seat for all Catholic
mission activities in the Vaupes; (5) the Caja Agraria, the government-run
agricultural cooperative that buys all rubber and makes loans (occasionally to
Tukanoans) for setting up a rubber-camp operation; (6) the police garrison, which
has only minimal effect on Tukanoans but indirectly serves as a deterrent to the
more violent and blatant abuses of Tukanoan labor and schemes to capitalize on
their ignorance; (7) the general stores (almacenes), which although having
higher prices than the comisariato, extend credit to Tukanoans and buy artifacts
from them, such as baskets.
By 1970 a number of Tukanoans had permanently migrated to Miku and its
environs (see Map 1). These individuals are fairly well acculturated. They speak
reasonably fluent Spanish, wear ironed clothing, and display such status symbols
as shoes, watches, and transistor radios. Household composition among them
follows affinal lines to a greater degree than elsewhere in the region.
I did not travel much in Brazil and thus cannot report firsthand in depth about
how the effects of white influence differ from those in Colombia. There are no
centers of non-Tukanoans comparable to Mitii in the Brazilian Vaupes, but in
general Indians are more acculturated. The Brazilian government was in greater
evidence in the Brazilian areas I visited than any comparable Colombian govern-
ment Indian agency, first through the Servic,o do Protec,ao ao Indio (SPI) and
after 1967 through SPI's replacement, the Fundagao Nacional do Indio
(FUNAI).
25
3.
The longhouse
The setting
Pumanaka buro sits on a hill overlooking an oxbow turn in the Inambu River.
Like most settlements, it takes its name from a particular feature of the local
landscape, in this case the "hill of many leaves." Settlements are also known by
the headman's Spanish or Indian name. The Inambu (tinamou bird - Tinamus
and Crypturellus spp.) is a small river on which five local descent groups - one
Bard, four Tuyuka - live at present.
It is difficult to convey how radically the character of a river in this region
changes over the course of a year. At low water no English speaker would call the
Inambu anything but a creek or stream, and huge logs, debris, and sand
dominate the landscape far more than water. Yet at high water the stream grows to
a raging torrent. The Inambu's height and volume alter its every aspect: its
appearance, its aquatic life, and the amount of time necessary to travel its length.
As is the case with most settlements, on arriving at Pflmanaka buro one
barely sees the longhouse.1 The canoe landing, however, always shows various
signs of human activity. One or more canoes, usually small one-man fishing
craft, will be tied to a tree or beached, depending on the shoreline and season.
Baskets of soaking manioc tubers may be visible in the water. A woman may be
washing clothes, cleaning an animal carcass, or scaling and gutting fish. Women
and children, or perhaps a lone man may be bathing. (Adult men usually bathe
alone, although young men bathe together at dawn.) The landing is often the
scene of social activities: Children play there several times during the day, and
women come in groups to fetch water; wash clothes, dishes, or manioc; and bathe
with their babies. The canoe landing is kept clean; poor maintenance is a sign of
low morale (Goldman, 1963, p. 38). Stairs and sometimes a railing are installed
when needed. See Figure 1.
The people of Pumanaka buro

The people and units composing the settlement I lived in are briefly introduced at
this point because they so clearly illustrate a number of important features of
26
Figure 1. The longhouse setting.
27
The fish people
residence patterns and longhouse life. Figure 2 shows how the residents of
Pfimanaka buro are related to one another (although not all consanguineal links
are shown); any discrepancies between it and the following list of who was
actually present when I arrived are due to marriages and births occurring after my
arrival.
In January 1969, five family groups resided at PQmanaka buro; I call these
hearth families, after Silverwood-Cope's (1972) "hearth group." Although not a
translation of a Bara term, hearth families labels a tacit native category (A. F. C.
Wallace 1961; see also Berlin, Breedlove, and Raven 1968), because Tukanoans
treat the people who share a hearth as a unit. A hearth family shares a fireplace,
eats together, and hangs its hammocks on the same four posts making up a
compartment (which may or may not be screened off with mats), with the
exception of unmarried initiated men, who sling their hammocks at the front of
the longhouse. The hearth family always contains an adult man and woman, most
often a husband and wife with their unmarried children. In many cases it contains
other unmarried kinsmen as well. Because it suggests that somehow such rela-
tives belong elsewhere, the term nuclear family is inappropriate for the hearth
family unit. If, for example, a brother and sister are both unmarried and have lost
their parents, they might constitute the core couple in a hearth family at that stage
in the domestic cycle. Thus, a hearth family requires one adult couple, closely
related consanguineally or affinally, and may have any number of other relatives
attached. Two married couples, however, never share a hearth. The variety of
possible arrangements is illustrated by the following list (names are pseudonyms;
except where necessary to place other relatives, dead members are not men-
tioned; members absent in January 1969 are listed in brackets):
I Mario, the headman

Juana, his Tuyuka wife
Sons: Estribino, [Lina], Berto, Pedro
Daughters: Maximiliana, Francisca
II Pedrina, Tuyuka widow of Firbino
Josefina, Bard mother of Pedrina (a very old widow)
Sons: Juanico, Angelino, Marto
Daughters: Armanda, Horacia
Others: Luisa, daughter of Emilia (dead daughter of Pedrina and
Firbino), who had been married to Emiliano, a Tuyuka who later
remarried.
Casimira, born three weeks after my arrival, to Armanda whose Tuyuka
husband, Mario, was away in a rubber camp.
III Anastasio, married to Joaquina, a Desana
Sons: [Candido], Bonifacio
Daughters: [Lina, married to a Desana], Luciana
IV Manuel, married to Esmeralda, a Tuyuka
28
1 = O TUyuka
t t
A= O
Cubec
rrrmTliyuka
r
•
i
A A •
' • = A l A r A = »
JUANICO
t Died
A * Bara
A O Not
Figure 2. The inhabitants of Pflmanaka buro. Not all consanguineal links are shown; only the name of the oldest living male is given for each hearth
family. Individuals within circles are not residents of Pumanaka buro.
The fish people
Sons: Mariquino, Nazario, Paulino

Daughters: Magdalena, Candida
V Inocencio, married to Bibiana, a Cubeo
Sons: Amelio, Lijio, Eugenio
Daughters: Micaela, Sabina
These families illustrate several stages in the life cycle of a local descent group.
At the time of my arrival the members of the younger generation in three families
(I, IV, V) had yet to have any marriages or births. This changed during the next
two years with three marriages (one unsuccessful) and two births (one the product
of a mating between a daughter in IV and a son in II). On the other hand
Anastasio and Joaquina (in unit III) had lost their two oldest children: Candido
was far away in San Jose de Guaviare, and Lina had married and moved to the
village of her husband (a Desana). Finally, family II shows some of the arrange-
ments that can result from prolonged absences and deaths. Both Pedrina and
Josefina were widows and kept each other company, although Josefina spent part
of her time at her dead husband's Tuyuka longhouse downriver with her grown
son (Pedrina's brother). Armanda, properly speaking, should have been living at
her husband's longhouse (although she was due to give birth, and some women
choose to visit their mother for the birth of their first child), but because her
husband was in a rubber camp she was living with her agnates. Similarly Luisa,
also a Tuyuka, was at PGmanaka buro being raised by her grandmother and
mother's sisters rather than with her father and stepmother at his settlement. A
number of the children present when I arrived were on vacation from the mission
school they attended in Acaricuara; they returned to school the next month.
The longhouse structure

Any Tukanoan longhouse is an imposing sight. The one at Pumanaka buro is
almost square (41 ft. x 47 ft.) but appears far more rectangular than these
dimensions would suggest owing to the height of the roof and the slope of the
eaves, which almost touch the ground at the sides. At the time of my arrival it
stood alone in the clearing except for a flimsy kitchen structure at the back. A
year after my arrival a hearth family moved out of the longhouse and built a
mudwalled house approximately twenty yards from the main structure. All
longhouses I saw were rectangular, although I was told that some on the Pira-
parana' River have rounded ends at the back to accommodate extra people.
Longhouse roofs, thatched with leaves of parana or other palm, are supported
by a framework of large posts, expertly lashed together with vines. The center
posts are enormous, and finding and bringing back tree trunks long and straight
enough demands herculean efforts. Walls are of bark, and the spaces between the
eaves and walls in the front and rear of the house are most often covered with
30
The longhouse
matting. The practice of painting the outside front walls still continues in the
Pira-parand region but has been abandoned in the Papurf region.
The headman plays the greatest part in selecting a site and organizing long-
house construction, and his role in this process identifies him as the "owner" of
the house. Construction of a longhouse testifies to his success as a leader,
because it would not have been finished without the willingness of his housemates
to follow his direction. In particular, because it requires unusual amounts of labor
and communal spirit, the transitional phase in the life of a community (between
the decision to rebuild and the completion of the structure) is an important and
demanding stage in a headman's career.
The decision to move a longhouse and the choice of a new site are influenced
by a complex mix of factors. Considerations related to agriculture, notably the
presence of suitable soils and terrain, most influence site location and the extent
of initial field clearing. A number of places may fulfill these key requirements,
however, leaving room for the influence of other considerations. These include
preferences for (1) proximity to a river; (2) higher ground, higher even than
needed to avoid flooding; (3) nearby arable land served by branch streams; (4)
distance from the nearest miriti swamp. Other seemingly idiosyncratic factors -
"There are a lot of such-and-such songbirds there," "They liked it there" - also
seem to influence decisions. On the other hand, a factor mentioned by Goldman
(1963, p. 36) in reference to the Cubeo, the pressure to maintain discrete sib
territories along the river, does not hold much importance along the Inambu,
because population densities are lower than among the Cubeo.
Outside the longhouse

The size of the clearing around a longhouse varies from one settlement to the
next. From the examination of drawings and photos from earlier periods, I have
the impression that as a general rule Tukanoans clear more now than they did
formerly. If so, this increase in clearing size possibly follows the lead of mission
towns, which are separated from the forest by extensive open ground. This style
of landscaping exposes mission towns to the torrential rains, leading to serious
erosion. Longhouses, moreover, can afford to clear more widely than they did in
the era of active warfare, when encircling vegetation offered some concealment
from enemies which sometimes included rubber gatherers. Before pacification,
longhouses also tended to be situated some distance from the river, as they still
are in some parts of the Pira-parana region.
Some settlements clear fields almost abutting the longhouse clearing, but the
normal pattern is to put even the nearest ones out of sight. In the past, I was told,
fields were always kept at some distance.2 Here again, missionary influence may
account for this change.
The plaza in front of the longhouse, kept scrupulously clean, is in most
31
The fish people
respects a male area. Men congregate there in late afternoon, sitting on stools,
smoking, talking, and working on various crafts such as basketry. Curing cere-
monies are carried out there, and during festivals the plaza turns into a dance
ground.
The area behind the longhouse, in contrast, is predominantly female, although
men appear there more often than women do in the front plaza, perhaps because
the rear area is more family oriented and informal. Ostensibly out of the sight of
visitors, it tends to be much messier than the plaza. Unlike Goldman (1963, p.
233), I did not see female guests being received at the rear during a festival.
Children and dogs congregate behind the longhouse, and women process food,
either outside or in separate kitchen structures sometimes found there.
Although no one voluntarily sits outside in the midday heat, late in the
afternoon activities needing light take place in front of or behind the longhouse.
Women gather in the rear, sitting and talking while mending clothes, polishing
newly made pottery, or feeding pet birds. Occasionally a lone woman will be
found in the front area, chatting with the men or working on some project -
particularly if it is one involving the labor of both sexes. Activities that occur
outside include basketry, plaiting, mending clothes, pounding tobacco, plucking
coca leaves, and various kinds of food processing.
Before a festival, outside activity increases greatly. The front plaza is meticu-
lously swept. The cane crusher is put in operation so that a more potent beer can
be offered to guests. Women put the final touches on the new dresses they have
made from trade cloth. Huge amounts of powdered coca are made, many
cigarettes of cured tobacco and banana leaves are assembled, and the large trough
used for manioc beer is washed and put to dry in the sun.
Forming a border to the semicircular plaza in front of the longhouse are
various fruit trees such as pupunha (Gulielma gasipaes), lime, papaya, mango,
and breadfruit. Some plants are cultivated near the longhouse, including various
medications and preventives, as well as plants for smearing on blowguns and
other weapons to attract game. Plants difficult to cultivate, such as tobacco and
pepper, are also grown near the longhouse. If a kitchen or side building is no
longer in use, these crops are sometimes planted there after the roof has rotted
away or been dismantled, because the soil is rich in ash and nutrients.
The front door of the longhouse is usually oriented to the east. River location
is irrelevant to longhouse siting, except that the men's door will be facing more
toward the river than the women's.
In the earlier times most longhouses had palisades and moats (see Goldman,
1963, p. 32), and "escape routes," consisting of a tree trunk crossing the moat,
which could be removed when danger threatened. Paths leading to the longhouse
were sometimes boobytrapped with fire-sharpened sticks.
32
The longhouse
Inside the longhouse

The division of space into male and female, public and private, formal and
relaxed, and visitor-oriented and domestic areas is even more pronounced inside
the longhouse than it is outside.3 The area inside along the sides of the building,
which is divided into sections by those vertical roof supports closest to the walls,
is intended for the use of individual families. Each of these sections forms the
sleeping and eating quarters of a single hearth family. Communal activities use
the center aisle away from the sides of the longhouse and the compartments
positioned off for families. This lateral division cross-cuts the front-to-back
distinction already mentioned in reference to the space outside the longhouse.
Visitors are entertained and exclusively male activities take place toward the
front, whereas casual and spontaneous activities, especially those in which
women and children predominate, cluster toward the rear.
The massive posts supporting the roof strike the eye immediately when one
enters a longhouse. Reaching up into the darkness, they themselves have been
darkened by years of fires. Once, such posts were painted with designs, but now,
at least in the Papuri region, they are more likely to be decorated, if at all, with a
saint's picture or a calendar. The hammocks strung between the posts are
detached at one end, rolled up, and tied to the posts during the day except for one
hammock in each compartment left open for lounging.
When visitors arrive, after being greeted they are offered the traditional
quinapira (Lingua Geral term for pepperpot sauce), cazabe (manioc bread),
and a beverage of either manicuera (boiled manioc juice) or farina (toasted
manioc granules) mixed with water. Male visitors usually stay near the front of
the longhouse and sit on the finely made and decorated Tukano stools reserved for
men. Some, however, especially younger men, may be invited to lounge in a
hammock after a while or may stroll to the rear of the longhouse and join in the
activities there. Female visitors either sit on the ground or on much lower and
more poorly made stools.
The headman assigns guests a space near the front door for their hammocks.
No longhouse is ever so crowded that permanent residents fill all of its eight or
ten compartments. I was told that longhouses used to be built with sixteen
compartments, but I never saw one with more than ten. The longhouse with the
largest number of residents in my census had seven hearths occupied. Figure 3
shows in schematic form the use of space inside a longhouse.
Some hearth family compartments are screened off with mats, but they are not
so enclosed as to keep all light from entering, nor of course do the mats
appreciably deaden sound. A longhouse is noisy at almost any hour, and everyone
can hear what everyone else is saying and doing. Private conversations, most
sexual activity, and excretion take place outdoors. The longhouse falls quiet only
during the late morning and early afternoon and then only on days when everyone
is away at the same time. Tukanoans sleep easily in this noisy environment, and
33
The fish people
Headman's
family
I Griddle H
H
H
\\\\voX\\\\N
H e p
Optional partitions
H Hearth family compartments
Figure 3. Ground plan of longhouse interior.
most people wake up at least once or twice during the night. At times the
residents present a well-orchestrated choral piece, featuring dog barks, baby
cries, coughs, farts, laughter, bad dreams, conversation, goings and comings
(which involve lifting the heavy door each time), replenishing the fire, and songs.
Anyone who feels the urge to sing will do so at any hour, and I never heard a
request for quiet. Sometimes women will begin some part of the day's manioc
processing at 2:30 or 3:00 in the morning.
The larger the longhouse is, the more comfortable: it is cooler, freer of insects,
34
The longhouse
less smoky, and generally more orderly. Guests, moreover, can be accommodated
much more easily. It is ironic that in mission towns the houses that most closely
follow missionary architectural specifications also have the most difficulty han-
dling visitors — ironic because missionaries and mission Indians encourage visits,
particularly for the fiestas given on saints' days. Houses in such nucleated villages
inevitably appear much more disorderly than longhouses: they are smaller, have
lower roofs, and their mud walls and solid room divisions cut off a great deal of
light and air. Some of the smaller houses in the villages either lack one or more
walls, or their walls are flimsily constructed of branches and matting. These
houses are much more tolerable in the heat of the day but can literally be a
sometime thing in a tropical thunderstorm.
Villages without longhouses cannot hold proper festivals. Two such villages,
Yapii on the upper Papuri River and Trinidad on the Tiquie, have built long-
houses specifically for ceremonial occasions, and in 1970 a ceremonial long-
house was under construction at Melo Franco on the Papuri. A longhouse has
also been built across the river at Mitu, but this one is intended mainly for the
tourist trade, to demonstrate indigenous ceremonies and culture to non-
Tukanoans. Inasmuch as these nucleated villages are returning to traditional
ceremonies and associated activities, the new longhouses advance this renais-
sance and stand as symbols of it.
Inside the longhouse, individual hearth family compartments contain an
assortment of pots and gourds for cooking, baskets of various shapes and
function, and three pottery fire cylinders to support cooking pots. Stored food
may include some meat smoking on a rack over the fire, plantains, peppers, or
corn drying in its husks suspended on the lateral poles over the compartment. The
rubber pouch a man takes hunting to hold shot, primer, powder, and tobacco
hangs on a nail, and harpoons and similar tools are stuck in the thatch of the roof.
Any young men in the family are likely to keep a cheap suitcase holding shirts and
trousers on a shelf over the compartment. (Young men seem to care more than
young women about neat clothing; some have irons.)
Immediately outside the entrance to the family compartment sits the daily
supply of manioc bread (cazabe). kept in a basket on a palm wood pedestal. In
houses invaded by leaf-eating ants at nighttime, everything must be suspended
from the floor. In others, only rats and cockroaches (the latter impressive in both
size and number) get at the food! Fires burn all night, not only to warm the
longhouse but also to keep away roaches and bats. The latter, particularly the
vampires, which are small enough to enter even a tightly sealed longhouse, are
real pests. Sometimes they attack children, particularly those sleeping in low-
slung hammocks, from which a dangling foot often rests on the floor. Vampire
bats do not swoop down on their victims, as is commonly supposed, but hop
along the floor, and are too timid to attack except in total or near-total darkness.
Tukanoans, knowing this, sometimes place baskets over sleeping dogs and other
animals to protect them.
35
The fish people
An inventory of the many artifacts found in and around the longhouse would
require more space than is available, as would a comprehensive description of the
technology involved in making them. One impression all visitors take away with
them is the skill with which both sexes manufacture any object. The overall
picture of Tukanoan artifacts is one of a deceptive simplicity, a restrained
elaboration with some high points of flamboyant decoration, and a beauty of
function and design remarkable in its ingenuity and understatement. A wide
variety of forest products are utilized, their natural qualities well understood and
put to the best advantage.
Significance of the longhouse

The longhouse unit is traditionally the most important social grouping in the
Vaup6s. With very few exceptions the maximal unit of food consumption and
production, it (or the more recent nucleated village) is quite isolated and, as a
consequence, highly autonomous. The tightly knit group of people forming its
membership see each other constantly and others only once in a while, and the
children reared in this setting develop far closer ties with one another than with
outsiders. Members, however, always include inmarried outsiders, thus represent-
ing at least two language groups other than the longhouse's father language.
Children spend most of their time with mothers and inmarried aunts, women who
speak their own father languages among themselves. This intersection of linguis-
tic and descent group ties found within all longhouses (and all villages) gives
considerable saliency to these divisions even at the most local level of regional
organization.
The longhouse, as the setting in which children learn the nuances and intrica-
cies of proper behavior, radically shapes childhood socialization. Living within a
large, semipartitioned barn with minimal privacy, and observing the large and
small dramas of life played out on its stage, children at an early age learn tact and
the ability to interpret subtle maneuvers and semihidden meanings. And although
Tukanoan parents tend to be permissive, and Tukanoan children occasionally
indulge in temper tantrums and provocative behavior, the children, when very
young, also learn forms of politeness that North American children often resist
adopting until well into their teens.
Longhouse activities and the use of space within the building express in a
variety of ways distinctions between the individual, the hearth family, and the
community. People eat some meals by themselves (although these are really
snacks), some within the hearth family's cubicle, and others publicly in a group.
Each type of meal has its own set of rules and takes place in a different part of the
longhouse. All these forms for routine meals differ from occasions when guests
are being fed.
The longhouse provides multiple metaphors for Tukanoan reality. It symboliz-
es the human body, proper social interaction, and even the entire cosmos. Many
36
The longhouse
dimensions of existence intersect here, sometimes in opposition to each other -

public and private, individual and communal, spontaneous and formal, male and
female, secular and sacred, human and nonhuman, and cultural and natural.
Later chapters will develop this theme further.
In addition to its role as a symbolic locus of value, the longhouse is the
principal unit of social and biological reproduction, the mechanism through
which Tukanoan society largely perpetuates itself. Ideally, its male membership
is coterminous with the sib, a localized patrilineal exogamous unit, and in many
contexts Tukanoans talk of larger units as if they were longhouses. The long-
house, moreover, plays the same role in the natural and supernatural spheres:
Each species of animal has its own longhouse and its owner, who is the
equivalent of a human headman, and dead Tukanoans go to the longhouse of their
ancestors, which have their headmen.
Human longhouses, as opposed to those of animals and the dead, stand as the
epitome of life and humanity. Babies, who are not quite human, are born outside
in the fields, whereas the dead, who were fully human while alive, are buried
within its walls. In one dramatic incident I witnessed, a seven-year-old boy bitten
on his foot by a fer-de-lance was kept for several days in a hastily constructed hut
in the manioc fields, even through a torrential thunderstorm, until he was cured.
His isolation, as someone in a dangerous, not-quite-human state, confirms the
association between the longhouse and humanness. In order to keep the long-
house safe for its members, marginal beings such as newborns and snakebite
victims should not enter or even come near until achieving (or reachieving) fully
human status.
In contrast with many other societies, the longhouse does not share its role as
the locus of humanity with any other structures or spaces. The men's houses,
village centers, and dance grounds found elsewhere in Amazonia disappear in the
Vaupes. Although Tukanoans recognize sacred or supernaturally charged places
in the large world, such as rapids and the river, their spiritual significance pales
before that of the longhouse. During ritual episodes, the essential nature of the
building is transformed: It exists in a different time spectrum, and the behavior
taking place within it has cosmic meaning. In this respect, Tukanoans differ from
many cultures (such as those of Highland Guatemala, Japan, or Ancient Greece)
in which sacred places are located away from the home and profane activity,
while resembling others (e.g., the Antoni [Cunningham, 1964] or the Tetum
[Hicks, 1976] of Indonesia) in which the house itself has extraordinary supernat-
ural value.
Rituals in the longhouse transform the inhabitants as well as the structure. The
ceremonies clearly elaborate the theme of endurance through the variety of ways
they test the endurance of participants. The entire pace and plan of such
festivals seem to have as their purpose the exhaustion of all reserves of physical
and mental energy, an outcome that both symbolizes the transformation being
sought and provides the direct experience of its reality. The normal demands on
37
The fish people
stamina typical of everyday life (as on canoe trips or long jungle treks) are
exaggerated in the ritual through consumption of large amounts of chemical
stimulants - beer, cigars, snuff, coca, and the hallucinogen banisteriopsis - a
corresponding absence of food and sleep, and all manner of activities and staged
special effects. All of these charge the atmosphere with emotion and a sense of
the sacred, helping participants achieve an altered state of consciousness. The
unwritten schedule of a festival calls for a long, steady increase in tension,
encouraged and regulated by dancing and chanting, until the most sacred time is
reached, at midnight.
The longhouse and the people who are successfully transformed by the totality
of the ritual are the same building and individuals that continue existing on the
everyday plane of secular life. Most of the secular yet essentially human activi-
ties, such as transforming crops into food by cooking, share the longhouse with
the most sacred rituals (which transform and transport human beings to another
plane of existence). In addition, activities such as burning incense when danger
threatens and performing curing ceremonies sacralize the longhouse at least
partially, without achieving the radical transformation of the festivals. In the
Vaupes, the sacred and secular interpenetrate and co-occur in varying propor-
tions rather than exclude each other, and the longhouse provides the most
important site for both.
38
4.
Economic and political life
Daily patterns
With the exception of slothful anthropologists, everyone in the longhouse gets up
at least an hour before dawn, blowing fires into life and bathing in the incredibly
cold river water. Young men bathe together at this time, pounding the water in
unison (which makes a sound similar to kettle drums), on some mornings playing
the Yurupari horns and taking plant purgatives as well. Breakfast, eaten about
daybreak, consists of cazabe and quinapira and occasionally includes leftovers
from the previous day's meal. The beverage is mingao, a drink made with hot
water and tapioca granules. At other times of the day farina mixed with cold
water will do, but a proper breakfast must include a hot drink. In one Bara myth,
troubles begin with an angry wife refusing to serve her husband a hot drink in the
morning.
Women leave for their fields at about eight o'clock in the morning, taking their
small children with them. Older children play near the longhouse or by the river
or look for adventure elsewhere. Some men go off to hunt, fish, or clear forest for
a new manioc field while others occupy themselves at home with such domestic
tasks as longhouse repairs, basketry, or coca preparation.
On a typical day, manioc processing takes up much of the afternoon for most
women. The men who left to hunt or fish return in the late afternoon, bringing
with them trussed game or gutted fish glistening on a canoe paddle. Hunters
sometimes return empty-handed, although fishermen seldom do. A man's wife
cooks his catch for the afternoon meal, feeding their hearth family or a larger
group, depending on his luck. Later in the afternoon when the heat of the day has
diminished, women go out again for firewood, in groups if possible. As the day
wears on, the pace of work for both sexes diminishes gradually, and by dusk,
clusters of adults are scattered here and there, chatting and playing with children.
At sundown everyone goes inside the longhouse, the headman lowers the doors at
either end, and until morning no one leaves except to urinate, engage in sexual
intercourse, or go hunting. No one eats after dark. The men's circle convenes in
the front area of the longhouse, even if only two men are present; it will last
39
The fish people
anywhere from two to five hours. Beginning and ending their assembly with
ritual chanting, they spend the hours in between chewing coca, smoking, and
talking. The women hold a similar but much less formal gathering in their area
toward the rear of the longhouse. A woman who cannot sit in the public space
because of some work that keeps her in her family compartment will nonetheless
join in the conversation and laughter.
Tukanoans have a remarkable capacity for talking on and on, despite their
isolation from outside people and events and their near-perfect knowledge of each
other's lives. Favorite anecdotes return to conversation repeatedly, each time
provoking almost as much laughter, surprise, or anger as they did the first time
they were told. The minutiae of the day's activities are also picked over in detail.
After sitting through such talk for a great many nights, one can only conclude that
social ends far outweigh the need for information and that choice of topic has no
great importance.
In addition to whatever firelight is available, the longhouse is illuminated at
night with a burning lump of resin or a piece of slow-burning wood. Radios may
be playing or not, depending on the owner's whim. Several noisy activities often
take place simultaneously, an inevitable outcome when many individuals and
families live in one building. I witnessed a striking example of this tolerance for
noise when attending a festival on the Tiquie" River. At the village where it was
being held, our longhouse group had been crowded into a single small house with
two other groups of visitors, and during the evening one woman suddenly fell
seriously ill. She started screaming with the pain (the aspirin I gave her had no
visible effect) and continued to scream all night. Throughout, her companions
played two different radios at full volume, and all of the men stayed up processing
coca, smoking, drinking manioc beer, and laughing. Although at times I felt I
had been dropped into an especially nightmarish scene in a Fellini film, their
behavior, cacophonous as it was, was probably the best possible response to the
situation. The woman was better off screaming than trying to stifle her moans for
the benefit of the others. Had the men kept quiet and deprived themselves of their
long-anticipated revelry, they would probably not have helped her at all. On the
contrary, their resentment at having to be sympathetic and quiet would have been
felt by everyone, including the sick woman.
As a rule, people work and relax at their own speed. They seldom seem to have
to do anything truly disagreeable, and when they do, they rarely openly complain
about it. Both sexes enjoy considerable freedom of choice in planning the day's
activities, although inevitably, some tasks demand careful scheduling, hard
work, and concentration. To be sure, the effects of years of socialization patterns
that stress the value of being good-natured hide behind the apparent ease< with
which everyone approaches his or her daily tasks. And freedom in planning the
day's schedule does not include the freedom to do nothing day after day. Gossip
and ostracism are always standing ready in the wings to appear when needed - for
example, when, in the opinion of the women, a young mother plays with her new
40
baby to the neglect of her other tasks. Nonetheless, I acquired a strong impres-
sion that Tukanoans find work a largely integrated and rewarding activity. They
appear self-motivated, beginning and ending each task without direction or
coercion by others. They begin tasks and stay with them until completed with a
spontaneity and naturalness of manner both admirable and enviable. Goldman
also comments on this self-reliant enthusiasm, noting that "the Cubeo believe
that a dispirited fisherman will catch nothing anyway" (1963, p. 54).
Of course, one reason why people are not as conscious of planning and
timetables as we is because their life simply is not as hectic as ours and because
all tasks are learned from childhood and done by everyone. Another factor is that
most work is social. Deeply ingrained expectations of cheerfulness and respon-
siveness to others present govern all social situations, regardless of purpose or
participants. Almost without exception, no one considers a task so difficult as to
permit him or her to ignore or remain in a bad mood toward people in the
immediate vicinity. Prima donnas and artistic temperaments are not characteristic
of Tukanoan communal activities, and it is difficult to imagine how they could be
under the conditions of longhouse life.
Productive tasks do not monopolize the whole of every day. Many Tukanoans
keep domestic animals and any number of wild birds, such as macaws, parrots,
and toucans, as pets. (Some birds are also raised for their feathers.) Fledglings
found in the forest are saved and laboriously fed by hand. Although Tukanoans
never eat the chickens and ducks found at virtually every settlement (nor the pigs
found in a few upper Paca River settlements), they sell them occasionally to
missionaries and rubber gatherers. To protect fowl from vampire bats and wild
cats, their owners put them in coops at night or train them to sleep on roofs.
Domestic cats, though rare, are valued because they feed themselves, often on
rats and bats. Dogs are also valued as pets, and some owners lavish attention on
their dogs, extracting niguas ("chiggers") and killing fleas. Although some
dogs look underfed, I never saw any seriously mistreated. I have little sense of
how effective they are in hunting, but I did witness two occasions in which a dog
trapped a paca, which was subsequently killed by women with machetes.
Personal appearance is even more interesting and time-consuming than pets.
Women bathe and comb their hair several times a day, and they often put on
caraiuru powder (made from the leaves of Bigonia chica), especially before
leaving for the fields, where it protects them from sunburn. Babies' faces are
painted with special "jaguar" dots, and infants are given wrist and ankle bands
and beads immediately after birth, practices that confer human status as well as
decorate. Tukanoans say this is one way in which they differ from Maku, who
have no beads for their babies.
Men are often quite fastidious about their appearance, and some of them,
especially young men, will occasionally spend a half hour or more preening and
painting their faces. Traditional dress for men consists of beads, earplugs, and
loincloth. Boys proudly begin wearing a loincloth at about four years of age,
41
The fish people
although for a while they must put up with teasing from the men, who try to pull
it off at any opportunity. In 1970, the young men who had been to mission school
wore swimming trunks or trousers.
Personal appearance is most important on ceremonial occasions, when every-
one takes exceptionally great pains with adornment and dress (see Chapter 11).
From time to time, however, individuals (men or women) will treat an ordinary
day as a special occasion, for no discernible reason taking the time to paint
themselves elaborately.
Music also forms part of life's daily round. In the late afternoon, or less often
at another time of day, men sometimes play musical instruments such as panpipes
(which come in sets of seven) and various small flutes made from bone, reed,
snails, pods, and deer heads. Women do not play instruments but do sing
impromptu songs to standardized melodies. The late afternoon is also a favorite
time for manufacturing household items.
The river
The fluvial environment dominates Tukanoan life, and the Bara in particular see
themselves as a riverine people: higher-ranked Bara sibs refer to themselves as
Waf maha, "fish people."
Riverine resources vary from one place to another, as does the social value of
different longhouse sites, especially in terms of distance from headwaters and
river mouth. At the same time, other factors, especially arrangements with
Maku to trade cultivated foods and white-manufactured trade goods for game,
influence the nature and intensity of a longhouse's exploitation of the river.
Tukanoans fish with weirs, hook and line, bow and harpoon, basket traps, and
poison. Dip nets and harpoons are used when fishing with poison. Fish poisoning
expeditions, which take place in the dry season, are festive occasions, initially
proposed by the headman and including everyone in the longhouse - the only
time the women involve themselves in fishing. The men leave early in the
morning to dam up the stream and poison the water. Timbo and barbasco are
generic terms in the ethnological literature for several species of fish poison made
from both cultivated and wild plants. The choice of which one to use depends on
how toxic an effect is desired and how much river water must be poisoned. The
poison, in the form of leaves, woody stems, or vines, is pounded and washed in
the water. Although harmless for humans or dogs, it kills or stuns all aquatic life
passing through the dammed-up portion of the river, and therefore cannot be used
very often.
By late afternoon many baskets and canoe bottoms are filled with fish, and the
poison has become much more diluted. Most fish are initially stupefied by the
poison rather than killed outright, and catching them in the muddy water when
they temporarily rise to the surface resembles an Easter-egg hunt with mobile and
greased eggs. Although pregnant women and their husbands cannot enter the
42
water, they help out on the bank, gutting and packing the fish. When the party
returns to the longhouse, all the fish are cleaned at the canoe landing, everyone
helping to prepare the basketfuls for smoking. Although the expedition involves
the whole longhouse, hearth family groups catch and give away fish. There is no
anonymity in fish-poison parties, nor does the headman oversee any kind of
redistribution. Each family prepares smoking racks, on which it skewers and
smokes huge amounts of fish. Another part of the catch is boiled up in large pots
for immediate consumption, and a veritable orgy of fish-eating ensues. Neigh-
bors and relatives from other longhouses come and receive gifts of smoked fish,
truly one of the delicacies of Tukanoan cuisine. Even after a week of gluttony,
their appetite and mine for fish merely declined rather than disappeared.
Fishing technology is well developed. Worms, insects, berries, and small fish
are the principal baits. Lines are sometimes tied to branches overhanging the
river - one passes a great many on canoe trips - to be pulled in late in the day,
although by then animals have gotten at some of the hooked fish. Apart from the
flashlights occasionally used in night fishing, the only fishing equipment acquired
from whites are fish hooks, nylon line, and the metal used to tip harpoons.
The technology included detailed knowledge of the habits of various species,
of daily, seasonal, and microecological variations, and especially of the fishing
prospects at different locations. Two sites on the same river can differ substan-
tially in yield. Each man has his favorite fishing spots, and these become familiar
to his sons as they grow up. Unlike Goldman (1963, p. 55), I was not told that
individuals could lay claim to fishing rights in particular streams or stretches of
river. They can, however, install and make exclusive use of weirs, although weir
owners are expected to let through enough aquatic life to supply other individuals
and longhouses.
Once in a while a group from a longhouse will camp out in the forest for
several days to fish, hunt, and smoke their catch. The expeditions of this sort I
witnessed all consisted of single hearth families, although I was told that some-
times bigger parties go out.
The most frequently eaten fish are the waracti (Leporinus copelandi Steind.),
tarira (Erythrinus sp.), and tucunare (Chichla ocellaris). The rivers provide
other types of food, such as shrimp and caymans. Men often hunt at rivers, in
search of edible water fowl and water rodents such as paca and agouti.1
Canoes and river travel are a vital part of Tukanoan life. All men are excellent
boatmen, and although women are less accomplished at negotiating rapids, they
know all of the paddling techniques. Everyone knows a great deal about the rivers
of the region, even those far from home. This knowledge involves, it must be
remembered, familiarity with each river in several drastically different states at
various seasons of the year. A number of men have also become expert at
handling outboard motors (although unfortunately not all of them - twice I was
involved in serious capsizings while traveling by motorized canoe).
Travel on any Vaupe's river is dangerous and requires concentration and
43
The fish people
familiarity with the territory. Rapids, of course, offer the greatest challenge. The
easiest of rapids takes considerable care and skill to negotiate, and the worst ones
demand almost superhuman efforts and teamwork. Travelers also expend consid-
erable time and energy carrying cargo overland, portaging canoes around those
rapids that are completely impassable, and portaging between rivers. Travel with
cargo requires three or more boatmen, and during some parts of my trips, notably
at the three-quarter-mile portage between the Vaupe"s and Papurf drainages, we
needed as many as eight.
Between rapids, canoe travel is often pleasant. An outboard motor and a
reasonably large canoe increase the pleasure, because paddling soon becomes
wearisome, especially when going upstream, and a small canoe requires constant
trimming and bailing. Travelers do not stop for bad weather, and ten hours of rain
a day can definitely get a bit tedious. Night travel, however, is impossible except
on the upper reaches of the Vaup6s River itself, because of the hazards of unseen
snags and rapids.
Tukanoans enjoy travel. It provides an opportunity to talk and joke with others
in the travel party for long stretches of time, to meet people in canoes encoun-
tered along the way, and to visit longhouses or villages at each overnight stop. At
times irresistible hunting opportunities present themselves en route, and during
high water if a canoe party detects signs of monkeys or other game, it will paddle
off through the flooded forest in hot pursuit.
As noted in the Preface, I was fortunate to be able to travel often by canoe
because it showed me the range of geographical knowledge held by Tukanoans -
for most of them it encompasses the entire region - and their feeling that all of it
was their territory. In addition to information acquired from spending so much
time in canoes, Tukanoans add to their knowledge by showing a keen interest in
geographical and ecological topics in their conversations with people at stopover
settlements.
Canoe travel provides adventures of all sorts. Some are unpleasant in the
extreme, as when a wasp's or hornet's nest is disturbed, or a branch filled with
fire ants is unwittingly pushed aside. But even the worst incidents make interest-
ing anecdotes for recounting later on. One such adventure occurred on one of my
trips when the only shotgun in the party became jammed, a disappointment
because we were paddling and would otherwise have stood a chance of bagging
some kind of shore-dwelling game. (The noise from an outboard motor greatly
reduces these opportunities.) A large Muscovy duck (dia katd) was spotted,
flying along ahead of the boat, frustrating the men, who were helpless to do
anything. Half an hour went by in which the bird repeatedly settled on a branch
ahead, only to fly on as we approached it. As this scene was repeated over and
over, the bird eventually became so used to us that it did not fly ahead as we
approached. Finally, as the canoe passed under the branch the bird was perched
on, the men raised their arms and - as Tukanoans frequently do - yelled ' 'Bam!''
giving themselves the pleasure of at least pretending to kill it. The bird, under-
44
standably startled, reacted by covering us all with a foul-smelling, warm white

liquid. As it flew off screeching, we threw ourselves in the river, howling with
laughter. The bird's well-aimed comments on our intentions toward it were
described many times, with accompanying shrieks of laughter from everyone, in
the days that followed.
This incident illustrates how Tukanoans handle the discomforts and dangers of
river travel and, in fact, most of the discomforts of life. One must laugh if
possible and savor the details for future storytelling. Tukanoans have an astonish-
ing ability to turn the most unpleasant happenings (including many far worse than
the above incident) into something funny, at times even while it is actually
happening.
The river does offer serious danger, particularly at rapids, and drownings do
occur, especially of children, although with what frequency I could not even
guess. Water animals, however, are generally more frightening than truly danger-
ous. The anaconda is perhaps the most important figure in Tukanoan mythology
and certainly commands a healthy respect from Tukanoans. I never heard of
anyone's being killed by either an anaconda or a boa. Piranha are found in most of
the rivers, but they are neither so numerous nor nearly so dangerous as they are in
other areas of Amazonia, as is illustrated by a tragic accident I witnessed on the
Papurf, when a Javerian priest drowned in the rapids. Attempts to raise his body
were unsuccessful, and it took forty hours to surface. Yet piranha had barely
touched it. The caymans (Paleosuchus palpebrosus; Caiman schlerops) in the
region are too small to be dangerous.
Tukanoan success in their riverine adaptation depends to a considerable extent
on their canoe-building skills. A man and his sons rough out the hull of a dugout
canoe in the forest. When it is light enough, they haul it to the river, where
everyone in the longhouse joins in the hot and difficult work of finishing it by
burning it out and widening it. Some Tukanoans told me that the Bara are
specialists in canoe building, but those Bara I spoke with said that expertise in
this skill varies from one individual to another and does not follow language
divisions.
In general, rivers pervade all aspects of Tukanoan life, social and cosmological
as well as technological. As boundary markers, the connecting links between
settlements, and the most obvious landmarks in the environment, they are the
principal means by which people orient themselves, give directions, and describe
themselves. Settlements are usually located near rapids, and when they are not,
they often take their name from a characteristic feature of the river (for example,
"lagoon"). And when Tukanoans distinguish themselves from Maku, their own
association with the river and the Maku separation from it looms large.
Rivers are important to growth in a number of ways. One's humanness, one's
self-image as a full-fledged human being and a Tukanoan, involves riverine ritual
and symbolism. A newborn infant's first rites take place at the river edge. Sacred
trumpets (Yurupari) are kept in the river, " to maintain their brilliant glow."
45
The fish people
Young men bathe together before dawn in the river, taking in some of the growth
power the river and the ancestors confer to those who maintain their spiritual
purity. Sexual intercourse often takes place at the shore, the couple bathing in the
river afterward.
Rivers are fundamental in Tukanoan cosmology. The first human beings
ascended into the Vaup6s by river in an anaconda canoe. The first ancestors
emerged at the various rapids sites in the region. The main feature of the
underworld (the universe is, strictly speaking, composed of five levels, but the
lower two are often spoken of as one) is a river: Opeko dia ("milk river"). It
flows in the opposite direction from rivers of this level; that is, it goes from east to
west. Symbolism concerned with rivers is found throughout Tukanoan mythology.
The forest
The forest is conceived of as quite separate and distinct from the world of
humans, symbolized by the longhouse and to some degree by the river. Although
an extremely important domain in Tukanoan cosmology, in general the forest is
more foreign, less connected to human life, and less benign than the river.
Nevertheless, although quite separate from the habitat of humans, the forest and
its denizens intersect with the human world in a number of ways, of which
hunting is the most important.
Tukanoans use a variety of traps and weapons in hunting, including snares,
nets, deadfalls, blowguns with poison darts, and bows and arrows. The last are
used mainly on birds, often with special arrows tipped with a hard bulb that stuns
but does not kill. Tapir and peccary used to be killed with metal-tipped spears.
Breech-loading 12- or 16-bore shotguns, however, have supplanted traditional
technology to become the dominant hunting weapon. Blowguns seem to be a
boy's weapon, used on slow-moving birds like doves and other small game,
which are roasted and eaten on the spot. I never saw an adult use a blowgun.
It is difficult to assess to what degree shotguns have affected the overall
availability of game, but it is being depleted. A general consensus among
informants is that men go out hunting less than they did before the introduction of
shotguns. This is at least in part because shot is expensive and difficult to obtain.
No one, consequently, will risk a shot unless he has a good chance of bagging the
prey. Yet neither does anyone want to return empty-handed after a day of hunting.
Therefore, a man will usually stay home or go fishing unless he sees a sign of
game in the area.
Men hunt alone, except when chasing white-lipped peccaries, which travel in
herds. When they hunt with a companion, imitations of bird calls are used for
communication.
Women contribute to hunting by relaying information about game noticed en
route to their fields or on trips to abandoned fields and longhouse sites to collect
46
fruit. As noted earlier, women also occasionally kill small rodents such as pacas
and tintins {Dasyprocta sp.), with the help of dogs.
A few Tukanoans hunt for pelts such as jaguar or otter, but Colombian law has
recently and successfully curtailed the pelt trade. The enforcement was rigorous
while I was in the country, and on a couple of occasions all luggage leaving Mini
by plane was thoroughly searched.
Game is said to be attracted by certain kinds of face and body paint. The leaves
of some of the plants grown near the longhouse are applied to shotguns and
blowguns to attract game, especially deer. Dogs are also rubbed with various
plants to help them locate game.
Animals most commonly hunted are paca (Cuniculus paca virgata), agouti
(Dasyprocta aguti), tapir (Tapirus terrestris), collared peccary (Tayassu tajacu),
white-lipped peccary (Tayassu pecari), deer (Mazama youzaoubira murelia),
tintin (Dasyprocta sp.), and less frequently, armadillo (Dasypus novemcinctus).
Some animals are killed for purposes other than food. Birds prized for their
plumage are currasow (Crax sp.), toucan (Ramphastos sulfuratus), parrot
(Amazona spp.), trumpeter, egret, and macaw. Jaguar are not eaten, but their
teeth are used for belts and necklaces.
Most small game is trussed after killing and brought back to the settlement.
Larger animals may be singed and cut up before returning home, but given how
rapidly meat spoils in the tropics, any foreseen delay between killing and cooking
necessitates a preliminary smoking in the forest. A man carries his kill to the
back door of the longhouse and gives it to the appropriate woman, usually his
wife, who takes it to the port for cleaning. Except for the liver, and sometimes the
kidney and heart, no viscera are saved.
If there is enough for the entire longhouse, a communal meal of boiled meat
ensues. Exchanges between hearth families of game or other food that are not
communal meals are neither random nor based on cut-and-dried prescription but
rather follow friendship and kinship ties. Thus, normally the closest agnates
(including close inmarried women), will share more exchanges, but of course if,
for instance, two brothers are not getting along, exchanges will be curtailed.
Although the totemic features found in Vaupes social structure and cosmology
do not include a prohibition on eating one's totem (e.g., toucans, armadillos),
many animals are tabooed as food. No carnivores are eaten (including the cats,
anteaters, snakes, etc.), nor are the herbivorous sloths and many smaller animals
such as fruit-eating bats. 2 Tukanoans say that Maku will eat anything, although
this is not, in fact, true. The reasons given are either that the Maku cannot feed
themselves properly and therefore eat anything because they are starving, or that
because Maku are not quite human they cannot be expected to know what is
correct human behavior.
A vast amount of symbolism surrounds hunting, in part because hunting is
concerned with food and thus shares in the intricate symbolic system relating
47
The fish people
food and eating to many areas of physical and spiritual growth and well-being.
Metaphors, inversions, taboos, and all manner of prescriptions create a link
between a particular food and another domain. Another reason is that hunting is
tied into the equally large symbolic system concerned with (1) male sexuality and
potency, (2) fertility of humans and game animals, (3) warfare and killing in
general, and (4) male adolescent growth and transformation into adult men. A
third reason is that hunting is related to shamanism, animal-human transforma-
tions (i.e., animal spirit familiars), power, and access to the supernatural dimen-
sion in order to exchange different kinds of energy and resources. For example, if
a certain type of game is scarce, a shaman will go into a trance and visit the home
of a ' 'master'' of a given species of game animal. This home is in a cave in one of
the dome-shaped sandstone hills in the region. There the shaman will bargain
with the master, offering to exchange some human souls from a settlement far
away if more animals are released from the game owner's pen. A fourth, and final
reason is that hunting takes place in the forest, which, although quite separate and
distinct from the world of humans, is an extremely important domain in
Tukanoan cosmology.
Forest-related (although not exclusively so) dangers include poisonous snakes
(especially fer-de-lances [Bothrops atrox atrox], and bushmasters [Lachesis muta
muta]), black widows, vampire bats, scorpions, hornets, and tarantulas. Most of
these animals come into direct contact with humans infrequently (with the
possible exception of vampire bats), but over time one collects a substantial
repertoire of anecdotes about such encounters. Snakebite is a serious threat to
life. Everyone has a relative who has been bitten, sometimes fatally or with
resulting loss of a limb. The young boy bitten by a fer-de-lance almost certainly
would have died without the serum I administered, because snakebite is extremely
dangerous to children (due to the higher proportion of venom to blood). The
incident was one of the few times I felt able partially to repay the people I lived
with for all the kindness and help they showed me. It also demonstrated how
helpless Tukanoans are in the face of snakebite. In a certain sense he was already
defined as dying, and thus the treatment basically consisted of making him
comfortable, although some requirements in fact resulted in increased discom-
fort, both physical and psychological. For example, he was not allowed to talk
with or get close to his mother, and as I noted earlier he had to be kept far from
the longhouse, in a hastily built shelter in a manioc field.
Without doubt the true lords of the forest are its insects. Their variety and
numbers can make life miserable; at meal times, for instance, sometimes each
mouthful a person takes seems to be half food and half gnats. The Tucandira ant
has an unbelievable sting, guaranteeing hours of excruciating pain. Tiny fire ants
(magin&) deposit formic acid on the skin; if an ant falls in the eye, three days of
desperation and agony result. Almost all of the dogs in the region have milky,
opaque corneas trom repeated contact with magifia. Sand flies can be extremely
irritating, and female sand fleas or chiggers (Pulex penetrans) lay eggs under the
48
skin that hatch into formless white grubs. If these are not extracted, they form
painful cysts and eventually fall off, leaving a large pit in the flesh. The one
saving grace of Vaup6s insects is that bees are stingless.
In Bara\ game animals are called wai buhkiira ("old" or "mature fish").
This category is contrasted with nanard ("bad" or "useless beings"), all
nonedible animals. All life forms are part of a universe having a finite quantity of
energy (Reichel-Dolmatoff, 1976), which is literally as well as figuratively
recycled in various kinds of transformations. This idea is common in many
Amazonian societies. Each species of game animal has its own master, who can
at times be encountered in the forest. Other spirits are also found in the forests.
They are either neutral toward humans (unless provoked into an angry act by
broken taboos) or unrelievedly mean and nasty. Any encounters between humans
and forest spirits, or any indirect contact with them via the spirits' paraphernalia,
symbols, or spirit familiars, are always of great significance. Thus, animal items
from the forest brought into the longhouse have power associated with them,
regardless of how they are used. Some are worn (e.g., jaguar teeth, monkey hair),
and some are used for special occasions (e.g., skin-covered drums, or bird-bone
snuff takers).
Also important are areas in the forest known to be especially favorite haunts of
a particular spirit. Clearings are examples of these, as are miriti swamps. The
origin of all rivers is one such swamp, Ewiira tar6, which is relatively near
Pflmanaka buro. These swamps are important in Tukanoan cosmology, perhaps
in part because they are ambiguous: a combination of forest and river rather than
one or the other (Douglas, 1966). They are water but not the moving water typical
of rivers; although consisting of water, they are filled with trees and other plants,
like the forest.
A hunting accident that happened to the headman of a Bara longhouse on the
Macucii River illustrates some of these concepts. What follows is a much
abbreviated version of an account given me.
About twenty years ago, Jose of Anuyuhti longhouse went to hunt white-lipped pec-
caries. He shot one, which was especially fierce because its mate was about to give birth.
The peccary gored Jose" in the knee, but he killed it with his dogs. He limped back to the
longhouse, carrying the dead peccary. The wound got very bad. Old Arturo, the most
knowledgeable of Bara shamans, started a curing chant, but Jos6 was ready to die. His
feet were already cold and only his heart moved. Old Arturo went to Ewiira wi, a
longhouse near Ewura tar6. 3 He saw Jos6 already there, standing outside the door. He
was putting on black body paint, festive ornaments, and feathers. His mother and his
father were there, and many people awaited him. Jose had almost finished painting
himself and was about to enter the longhouse. If he had entered, he would have died in this
world. The people at Anuyuhti were already crying. Old Arturo came and pushed Jose
away from the doorway. Many peccaries were outside the house waiting, and Old Arturo
began to scare them with his power. First he summoned a big wind. The peccaries said that
wind had been around since the beginning of the world. Then Old Arturo summoned a
tremendous fire, which did scare them. They lined up and began to enter the longhouse.
He made them shut all the doors, which he closed with big logs. Old Arturo took away
49
The fish people
their aiiawiiri (long flutes), taking away their dances, taking away their hearts. When Old
Arturo returned to this world, he carried these flutes in his throat, and Jos6 recovered. But
this is why there are no peccaries in these parts today.
The forest also provides nonhunted foods of many kinds, including grubs,
termites, ants, wild fruits and berries, nuts, and palm hearts. The most important
plants are the umari fruit {Poraqueiba sericea), calaloo (Phytolacca iscon-
sandra), similar to spinach, wild Inga (Inga sp.), and assai-palm fruit (Euterpe
oleracea). Some kinds of fungus in rotting logs are edible. Collecting, except
where there are large harvests of wild fruit, is organized along hearth family
lines. It is spontaneous, and done mostly by women and children. Honey is a
delectable foraged food but is not nearly as abundant as in other areas of lowland
South America (e.g., the Ache, see Clastres, 1972; see also Kloos, 1977, on the
Akuriyo).4
Cultivated foods
Recent debates concerned with geological and ecological variation in the Ama-
zon Basin have focused on issues such as carrying capacity and prehistoric
demography (Denevan, 1976; Dobyns, 1966) and the upper limits of horticultural
potential, given the region's ecology and the staple crops that are grown.5 An
important outcome of these debates has been the increasing awareness and
recognition of extensive microvariation within the basin as well as an accumula-
tion of evidence that South American tropical ecosystems are complex and finely
honed systems impossible to understand in a cursory manner (Fosberg, 1973).
They most certainly cannot withstand many of the kinds of disturbances currently
being undergone as a result of agricultural and other kinds of development
programs. These current disturbances, often of a serious and irreversible nature
(see Denevan, 1973; Fosberg, 1973; Goodland and Irwin, 1975; Ruddle, 1974;
Sternberg, 1969), contrast with the effects of traditional swidden (slash-and-
burn) horticulture over the centuries up to the present. The genius of the various
types of swidden systems is that they admirably imitate (although only partially;
see Hames, 1980a) the natural tropical forest's cycles. This is true only when the
ratio of human beings to arable land is sufficiently low (the actual ratio varies,
depending on region and horticultural system). Where this is still the case,
indigenous utilization of natural tropical resources, including the land used for
horticulture and exploitation of wild products from the forests and rivers them-
selves, generally illustrates ecologically sound practices.
The present ecology of the Amazon Basin is extremely old (in comparison to,
for example, the age of the temperate forests of Europe) and is a delicately
balanced system. This is belied by the lush appearance of its canopy forests and
teeming animal life. We do not know all of the answers to why such good
vegetation can exist in such an environment with such a fragile base, but one
factor is certainly that it is a closed system within which 70 percent of the total
50
mineral nutrient supply is within the living biomass of the forest (Fosberg, 1973)
- a far higher percentage than in Iowa, for example. Decomposition of organic
matter is rapid and complete, particularly in the steady high temperatures of the
tropics, and the released nutrients are immediately absorbed by surface yet very
efficient root structures. One reason why this ecosystem is so easily upset is that
almost any shift in the process has far-reaching consequences. For example,
when the very thin humus layer is directly exposed to sunlight, several kinds of
permanent damage can occur, such as rapid leaching and percolation of nutrients
into the lower subsoil, where they are permanently lost. The top layer of soil can
literally evaporate. Tropical soils are thin, delicate, and of low fertility. They lack
weatherable compounds, such as silicates, and have a high iron content, espe-
cially in the older strata of the Guiana and Brazilian shields. The highly destruc-
tive effects of tropical rain, wind, and sun can quickly turn these areas into
eroded saw-grass savanna with no possibility of reforestation. Or, more drastical-
ly, the resulting erosion can produce a landscape looking remarkably like
parking-lot pavement and having about the same agricultural potential.
Although these points are somewhat technical, an understanding of the
Tukanoan subsistence base or that of any lowland South American society
requires a preliminary understanding of rainforest ecology and the biosphere of
the Amazon Basin. Furthermore, much of the acculturation undergone by these
societies at present ultimately comes from land-development schemes involving
agriculture, animal husbandry, or both. The future of these societies is inextrica-
bly linked to the outcome of these schemes, and both are in turn linked to the
conditions and limitations of their environment.
The ongoing debate over soil fertility and its exhaustion relates to a number of
diverse issues. One of these is the cause (or causes) of the periodic shifts in
settlement site that are so characteristic of all swidden-based societies in lowland
South America. Various authors offer evidence that periodic changes in settle-
ment site cannot be explained in terms of single determinants such as loss of soil
fertility.6 It is obvious that one must take into account such factors as overall diet,
the organization of economic activities (e.g., the sexual division of labor), and
the integration of these activities with other areas of the society, such as settle-
ment size and warfare. Several authors have suggested that the real limiting factor
for many groups is game, both because game and fish resources are exhausted
more quickly than cultivable soil7 and because manioc is poor in several basic
nutrients, making it necessary to have reliable sources of protein. Current studies
in the nutritional and health status of Amazonian populations will undoubtedly
shed further light on the complexity of the situation.
Cultivation in the Vaupes.

Any person who has visited the Vaup6s becomes nostalgic when remembering
activities relating to the manioc process, for it is one of the primary symbols of
51
The fish people
the region. Its salience is in part due to the high visibility of the activities in and
around the longhouse and because at least several stages, and usually the entire
process, occur daily.
A woman leaves the longhouse early in the morning with an old basket, an old
machete, a smoldering piece of wood, her very young children, and her dog. In
the fields she cuts the manioc plants, burns the tops (helping to return some of the
nutrients to the soil), and with the machete digs down to the tubers themselves.
She may peel some of the tubers in the field or wait until she is out of the sun.
Flimsily built shelters are found in many fields, which are used for relaxing and
eating a snack, peeling tubers in the shade, or escaping a sudden thunderstorm.
While in the field the woman will tend other crops as well, including plantains
(Musa paradisiaca), bananas (Musa sp.), name (no English equivalent; Xan-
thosoma sp.), sweet potatoes (Jpomoea batatas), sugar cane {Saccharum
officinarum), pineapple (Ananas comosus), melons, yams (Dioscorea trifida),
and such nonfood items as Bigonia chica (a bush producing red pigment), bottle
gourds (Lagenaria sp.), achiote (Bixa orellana; a red pigment used by females),
and a black dye-producing bush (Rubiaceae sp.). Plants tended by men include
coca (Erythroxylon coca), calabash trees (Crescentia cujete), yage (at least two
species: Banisteriopsis inebrians and Banisteriopsis rusbyana), and tobacco
(Nicotiana tabacum). Men also plant fruit trees, both in the fields and close to the
longhouse, such as jungle grape or cucura (Pourouma cecropiaefolia; its leaves
are used to wrap farina or manioc bread and are the preferred ash for mixing with
coca), Ingd (Inga dulcis — some Inga also grows wild), pupunha or peach palm
(Guiliema gasipaes), and more recently introduced trees such as mango (Man-
gifera indica), papaya (Caricapapaya), lime (Citrus aurantifolia), and caimito or
star apple (Chrysophyllum caimito). The manioc fields show some intercropping
(e.g., bananas, pineapple), but corn is usually grown separately because it needs
more care, particularly weeding; corn is often planted and tended by men.
Manioc is an admirable crop in many ways, even though deficient in certain
important nutrients. It is dependable, is not finicky about soil type or in need of
specific combinations of dryness and rain at certain crucial stages of growth (as is
true of corn, for example), and needs a relatively small amount of weeding and
protection. Compared with a number of other crops high in carbohydrates, it
provides an abundant harvest in proportion to the amount of labor required for
planting and tending. It is storable while in the ground, and when processed into
farina can keep several months. Several other manioc products will keep for
relatively long periods, a very valuable feature because storage of any foodstuff is
no mean feat in the tropics. Bitter manioc8 is used for making manioc beer and
cazabe. Thus, bitter manioc is a superior crop to sweet manioc because the
variety of products obtained from it and its storability make it a uniquely suitable
staple crop. That many Amazonian societies prefer it over sweet varieties is
understandable, despite its arduous processing requirements.
52
After several days of harvesting manioc the woman replants the manioc stems
she has accumulated, putting them in the ground at an angle and often mounding
the soil around two or three stems to delay erosion caused by heavy rain on the
relatively unprotected soil. Coca is also planted by men after a first harvest of
manioc (as well as in newly cleared fields). Men plant coca in a very symmetrical
fashion, and ordinarily tend it and harvest it by themselves. Still, I have often
seen women plucking coca leaves off the stems, although except for older
women, women generally chew the final product only at festivals.
Manioc is harvestable after six months, but the tubers are usually dug up after
twelve to twenty-four months. Whitten (1976, p. 74) states that toxicity is highest
right after flowering, decreasing in the tuber after five or six months.
Goldman (1963, p. 61) is quite correct in describing women as agronomists
rather than mere gardeners. Any woman can rattle off a long list of varieties of
manioc, describing which ones she prefers to plant and why. Each woman's
cazabe and farina are distinctive in flavor and texture because of her own special
blend of manioc varieties.
Later in the day women return from the fields with their tubers, washing them
in a small stream or in the river itself and perhaps letting them soak there in an
old basket or canoe, which makes peeling easier. Peeling, even with knives, and
the subsequent grating are arduous work. Grater boards are obtained from the
Curripaco and Baniwa, Arawak speakers living to the north of the Vaupes River.
These boards are set with sharp quartz stones. A woman sits on the ground,
places a board on her lap, and grates, holding tubers in both hands. She will grate
enough mash for at least a day's supply, more if she wants enough for two days or
plans on making farina or beer. The mash is then washed in a tripod-supported
basket and squeezed in the tipiti, a tubular basket woven like a Chinese toy finger
trap such that pulling it constricts it. At this point the manioc mash can be sifted
and toasted, producing farina or cazabe, depending on the method of toasting.
The liquid from the washing is boiled to make the drink manicuera (extreme heat
breaks down the prussic acid in manioc).9
Women also make juice from pupunha in the tripod basket strainer; the juice is
frequently allowed to ferment. Other fruit drinks are made in this manner as well.
In general, manioc products are not what a Westerner considers gourmet fare,
although some are better than others. The blue ribbon for tastelessness goes to a
very hard cake made entirely from starch. It resembles cardboard, but is good for
trips because it keeps so well. Cazabe is at its best hot off the griddle, but
Tukanoans do not eat any manioc product so soon after cooking. Little thin cakes
with no fiber content are made from tapioca; these are considered to be an
especially pure kind of food.
Although men occasionally help with cooking, they are regularly expected to
help only with making manioc beer, which is made with corn, manioc products,
and other root crops such as Xanthosoma and sweet potatoes. Fermentation is
53
The fish people
begun by women chewing on baked corn cakes and spitting pieces into a pot.
Sugar cane and fruit juices are added to beer made for festivals, resulting in a
more potent drink.
All Tukanoans are interested in gardening. Although this is mainly the domain
of women, men are also knowledgeable about it. I once brought in packets of
seeds: Both men and women became involved, examined their plots daily, and
rejoiced when a few plants actually appeared, including melons, tomatoes, and
green bell peppers. The bell peppers promised to be a big success but turned out
to be the biggest failure, even though they flourished. Everyone had assumed that
peppers four times as large as chili peppers would also be four times as hot. Of
cultivated foods, only farina and occasionally fruit are sold to missionaries.
New fields are prepared yearly, at the end of the dry season. Knowing what day
to fire and doing it properly is a tricky business. A blazing field is a spectacular
sight, and people sometimes gather to watch.
A woman's husband fells and fires a field for her after they have decided where
to clear. The field belongs to the woman and is spoken of as hers but only for as
long as she uses it. Concepts of ownership are more elaborated with respect to
horticultural land than hunting or fishing territories. If someone takes food from a
plot without telling the owner, it is considered to be theft.
Paca, agoutis, and deer are the most bothersome pests in the manioc fields.
(These animals eat manioc before it is exposed to oxygen and therefore not
toxic.) Leaf-cutter ants can also accomplish a great deal of damage when their
routes take them through a cultivated area.
The importance of food.

It is difficult to overstate the importance of attitudes and beliefs about food among
Tukanoans. Although it is an exaggeration to characterize Tukanoans as "horti-
culturalists . . . for whom the total food quest constitutes the cultural focus of
society," as Ruddle (1974, p. 4) has described incipient tropical forest horti-
culturalists in general, food is extremely important. It is furthermore symboli-
cally connected to all other important cultural foci in Vaupes society. Food has
importance as sustenance, as a form of ceremonial communication, as an expres-
sion of emotion (i.e., of affection when given, of anger when withheld), and as a
metaphor for such other important symbol domains as the body, society, and the
universe.
Foods are categorized into an elaborate system of taboos. This system is so
pervasive that it is rare that one or more people in a longhouse are not observing
some degree of restriction. When people are in a state of crisis or some sort of
precarious condition (either physical or mental), they will be prohibited those
foods corresponding to the level of seriousness and danger of their condition.
When a person falls sick, certain categories of food are forbidden. Similarly,
entering a life crisis such as puberty or pregnancy puts one in jeopardy, and a
54
certain level of food restriction is prescribed. Food restrictions are causally
related to illness and curing in a number of ways; for example, people can
become sick because they failed to observe a taboo. Many states (e.g., menstrua-
tion) and activities (e.g., hunting) require observing a particular level of food
restrictions. In cases of illness the shaman (or person substituting for a shaman if
one is not present) makes the diagnosis and prescribes the level of food restric-
tions the patient must observe in order to recover. When the danger is sufficiently
past, the shaman will also chant over each type of foodstuff being reintroduced
into the diet of the sick person. After this initial blessing, which must be done for
each kind of food, the person can safely resume eating it.
The sequence of prohibited foods is an ordinal scale - that is, if a particular
level is prohibited, all preceding levels will automatically be prohibited as well.
This scheme is a condensed version of the sequencing of foodstuffs that are
permitted to children as they grow up. When a child is ready to eat a new item, a
sample of that item will be shamanized before the child eats it for the first time. A
very abbreviated form of this sequence (C. Hugh-Jones, 1979; Langdon, 1975) is
as follows: (1) milk, after being shamanized, along with kand fruits (Sabicea
amazonensis), in the case of infants (C. Hugh-Jones, 1979, p. 118); (2) "pure"
foods (all of which should be cold): pure manioc starch with no fiber, termites,
maniwara (flying ants) - these animals have "no blood"; (3) most vegetable
products, including chili pepper and hot manioc juice; (4) small fish and all
manioc products; (5) small game; (6) large herbivorous game. Carnivores are
never eaten because not only do they have a lot of blood, but they eat other
animals that have blood. Their flesh, consequently, would be far too hot and
dangerous. There is a geheral progression from low to high (the animal's or
plant's habitat), from water to land, and from small to large (S. Hugh-Jones,
1979, p. 93).
Different methods of capture and cooking also fit into this scheme of relative
amounts of danger. The longer a foodstuff is cooked, the less dangerous it
becomes, because cooking removes some of an animal's heat. Thus, boiling is
better than roasting.
The correct approach to food and eating is one of moderation. Excess - lavish
and ostentatious serving, overly enthusiastic eating behavior, or gluttony - is very
much disapproved of. The general apprehensiveness about succulent or hot-off-
the-griddle foods underscores this general theme. Self-control and moderation
are clearly a part of the food restriction system also.
The emphasis on moderation in eating is revealed most strongly in formal
settings. For example, a guest eats a very small amount of food just after arriving
at a longhouse. Everyone is supposed to be restrained at communal meals as
well. Children, however, are allowed to eat until full and whenever they wish, and
under less formal circumstances (for instance, when women snack on bananas in
the manioc fields) people are more relaxed and boisterous.
At times people inadvertently reveal more of an interest in food than that
55
The fish people
prescribed by etiquette. It is not difficult for one who is familiar with eating rules
to notice the poorly suppressed impatience right before a communal meal,
particularly if meat has recently been scarce. Sometimes people's efforts to
maintain appearances while feeling acutely hungry are comic. I have seen
women, after a pot of meat has been put in the middle of the floor and the
invitation to eat extended to all, "walk" incredibly quickly to their baskets of
cazabe. Grabbing a piece, they "walk" back to the pot, feigning nonchalance. (It
is very bad manners to go for a piece of cazabe before one is formally invited to a
meal, and eating meat without cazabe is unthinkable.)
The requirement of moderation regarding food also means that one should hide
one's disappointment or anger when expectations about food giving are not met.
Any talking about stinginess or laziness should be strictly in private and certainly
not leave the confines of the hearth family group. Still, I heard a great deal of
gossip among the women about various kinds of nonsharing of food. It is my
impression that women gossip more than men, but men notice slights too and will
talk about them with their close kinswomen. Clearly, gossip is the main mecha-
nism of social control regarding food distribution beyond the hearth family.
The amount of formality and moderation in eating behavior directly correlates
with the number of people present at a meal and the amount of social distance
among them. This is why visitors to a longhouse are never supposed to eat to
repletion; the offerings of cazabe and quinapira that each woman makes to the
visitors are token offerings. Tukanoans feel that people should eat a genuine meal
only among intimates. A man returning from a long day's hunting is served alone
by his wife in his own compartment or immediately outside of it and will eat until
he is full. Formal communal meals involving the entire longhouse fall somewhere
in between these two extremes. People should never be greedy, but a communal
meal is not organized unless there is enough meat to give everyone more than a
taste.
Excessive consumption occurs only at festivals, and the overindulgence dis-
played then involves only those items that C. Hugh-Jones has termed "soul
food" (1979, p. 213). These include tobacco, coca, banisteriopsis, and manioc
beer. "Real" food is not consumed during a festival; this stricture applies
particularly to men.
Visitors to a longhouse should never rely on food offered them by their hosts,
regardless of how intimate the relationship is, but should always bring provisions
with them. The "Jewish mother" model found in many parts of the world of
ostentatious offerings of food and disappointment if it is not eaten would be
totally incomprehensible to lukanoans. Guests share food, having brought enough
for themselves. If they stay longer than intended, both women and men help
produce and process food. If they stay for a month or more, a section of the
manioc fields will be turned over to the women.10
A man brings any game he has killed to his wife, mother, or sister, depending
on with whom he shares a hearth. Many conversations among women focus on
56
men as providers, on their generosity or lack of it. Such judgments and invidious
comparisons about a man's laziness, lack of skill, or stinginess should never take
place in the ideal Tukanoan society, but one hears them in the real one.
Although most exchange is carried out in as spontaneous and casual a manner
as possible, a lot of rather unspontaneous and calculated account-keeping takes
place in everyone's head. The differences between this system and ones reported
for peasant patterns of reciprocity, for example, is that among Tukanoans the
ledger books are memorized and not a proper subject for public discussion. At
times a fairly pronounced egotism and spitefulness can come to the fore, particu-
larly when a given situation is somewhat outside of the traditional system of
expectations about sharing. I was caught in a number of upsetting squabbles
related to my giving gifts and paying for food, in part because these gifts and
payments were outside the traditional system, and people felt less compunction to
keep silent.
Doubtless some of the lack of fit between the ideal and practice with respect to
generosity has to do with limited resources, particularly regarding certain highly
desired foodstuffs. This is especially true during lean times in the yearly cycle. A
second factor is that whereas the hearth family is the main unit of consumption,
everyone nevertheless lives in the same large room, quite able to observe what
other families have to eat (although some attempts to conceal take place).
Longhouse-wide sharing would be the ideal under ideal conditions. Who actually
gives what to whom and when is governed by elastic conditions and allows for a
fair amount of personal inclination. Such power doubtless is an incentive to
produce and gather food, since one is rewarded not only by being able to eat well
and feed one's family, but to give to others - a supreme pleasure for Tukanoans
unless they are temporarily at odds with everyone in the longhouse. Still, along
with a flexibility allowing for personal inclination in giving food comes the
possibility of having one's feelings hurt. Women in particular give one another
such gifts as a pineapple or haunch of meat. These decisions are made entirely by
the giver, and a strong component in them is the message they carry about the
affect existing between giver and receiver. Thus, if a woman does not receive a
bunch of bananas she was hoping for, all she can do is reconsider her assessment
of the state of the relationship between the (non-)giver and herself.
The degree to which an individual is considered to be the "owner" of a
particular foodstuff depends on the amount of effort he or she has put into
acquiring it or converting it into food. This is true regardless of the nature of the
item or the age or sex of the individual. A piece of uncleared land is potentially
available to anyone in the longhouse. When a woman announces her intention to
plant it, it is considered hers as are, particularly, the crops she grows in it. Theft
is most often mentioned in connection with stealing food from fields; it is even
mentioned in a Bar£ myth. Similarly, anyone can go out and gather edible
insects. If a little boy brings back some flying ants in a pot, they are his. He can
eat them, share them, or let them shrivel up in the pot if he so wishes. The way in
57
The fish people
which food is served and received acknowledges who produced and prepared it.
Among Tukanoans, serving is not a sign of inferiority but is the privilege of
someone with something to serve - something he or she has made. Men will
serve other men manioc beer because they are the hosts of the longhouse that is
providing it (and also because they have participated in making it). They serve the
hallucinogenic banisteriopsis to their guests for the same reason. The one excep-
tion to this general rule is the role of a Maku servant, who fetches and serves
beer and lights cigars.
The question of just how limited the food resources in the Vaupes are in terms
of nutrition must remain unanswered until further studies are carried out. I was
impressed with the health and vigor of the people at Pflmanaka buro and did not
ever observe a general lack of food. Meat and, less often, fish were sometimes
scarce, but Bara told me that if things became really desperate, people would
concern themselves more with snares and blowguns to kill various small birds.
Goldman (1963, p. 85) concurs that real hunger among the Cubeo is not a
problem. Langdon (1975) describes a somewhat different situation, of times of
real deprivation, for the Barasana of Cano Colorado in the Pira-parana.
The Bara of Pumanaka buro certainly talked about hunger, especially for
meat. Juana, the headman's wife, told of a horrible period in her childhood when
a windstorm blew off the roof and two of the walls of the longhouse. Hail
destroyed the manioc fields. For a while, people were able to harvest the mature
tubers, but after this a long period of severe hunger set in until the new fields had
tubers mature enough to harvest.
Ability to feed oneself and family and, to a lesser extent, knowing that the
entire longhouse community is adequately fed is an extremely important matter
to Tukanoans. They see themselves as usually meeting this responsibility, offer-
ing the Maku as a contrast. Maku, they say, actually do starve at times, and
some Bara have rather lurid tales to support this assertion.
It would be premature to attempt to draw a tight correlation between general
characteristics of the ecosystem, the structure of food production and distribu-
tion, and values and attitudes surrounding food. Nevertheless, some speculations
might be worthwhile if they can help stimulate further thinking and research on
this topic.
The economy, as Goldman (1963, p. 85) has suggested, is relatively inelastic,
subsistence oriented, and egalitarian. This is connected to a relative lack of
abundance of foraged or cultivated foods - in comparison, for instance, to the
northwest coast of North America, some other areas of Amazonia, or the yam
cultivating societies of Melanesia. It also has to do with the fact that few
comestibles, particularly those of greatest value and scarcity, keep for any length
of time. Thus, as is true in many societies with similar levels of technology and
subsistence bases, people's resources lie in other people rather than in wealth in
the form of stored food or items of value that can be converted to food through
exchange. This is also the situation in many African horticultural societies,
58
although control over the productive potential of others is far more evident and
hierarchically organized than in lowland South America (O'Laughlin, 1974).
The difficulty of storing valued food is demonstrated when Tukanoans try to
amass enough smoked meat or fish to make a decent showing at a dabucuri, a
ceremony involving an exchange of meat or fish (brought by the guests) for beer
(supplied by the hosts). Other foods, such as miriti pulp, smoked ants, grubs, and
so on, are also given. The scurrying around to acquire enough game or fish, the
discussions, the oblique references to someone's not doing his share, are,
although by-products of the preparations, reminders to everyone of the reality of
their dependence on one another. The ceremony itself beautifully expresses the
fact that people are fed and taken care of by their own longhouse community and
by relations maintained beyond it, rather than by each person's ability to take care
of himself or herself.
The nature of the staple crop, manioc, also produces a certain amount of
interdependence. The limiting factors on producing surpluses of manioc are not
land or other forms of scarce resources (e.g., water or fertilizer) but other people.
The bottlenecks in amassing a surplus are the labor and goodwill of the women
who process it. This fact - which is very clear to Tukanoans and expressed in
many myths in one form or another — is another illustration of the necessity of
investing in other people and their good intentions toward oneself rather than
attempting to be independent and rely solely on one's own abilities. Such social
factors play a larger role in the inelasticity of the economy than technological
factors per se.
Exchange in general
The patterns I have described for exchange of food are applicable to the exchange
of nonfood items as well. A generalized type of reciprocity exists within the
hearth family. It is important to note that both generalized and balanced reciproc-
ity occurring between hearth families in the same longhouse and between closely
associated families in different longhouses are thoroughly embedded in the wider
sphere of social relations. It is fruitless to attempt a categorical separation
between economically motivated activities and other kinds. For example, a man
or woman who becomes lazy is far more likely to be expressing bad feelings of
some kind than to be wanting to get away with less work. Similarly, intentional
theft (of nonfood objects) is more likely intended as an insult or at least a sign of
displeasure at the owner than motivated simply by a desire for the object. The
kind of alienation from labor and its products that leads to desiring objects
strictly as objects and thus being able secretly to take them is only beginning to
happen in the Vaupds.
When an item is spontaneously given, very little comment is made. Certainly
no expression of gratitude is offered; the debt will be settled with a return gift at
some point in the future. In fact, the need to express any kind of thanks, whether
59
The fish people
in response to gifts from humans or from supernatural beings is a rather foreign
concept. When the more formal types of exchange are accompanied by ritual, the
rituals are not concerned with expressions of gratitude.
An example of a more formal exchange ceremony is the dabucuri festival,
which involves smoked meat and fish and beer. Another example is the basketry
that a prospective bridegroom makes and gives to the women of his intended
bride's settlement. A third is the he-tenu relationship reported by S. Hugh-
Jones (1974) and C. Hugh-Jones (1979) for the Pira-parana region. This
consists of an agreement made by two men standing in a classificatory cross-
cousin relationship to each other (and thus automatically from different settle-
ments) who are about the same age. They exchange ceremonial objects from time
to time. Whole language groups can be spoken of as standing in he-tenu
relationship to one another (C. Hugh-Jones, personal communication). Exchange
in general is an integral part of marriage alliances and indeed of all ongoing
relationships.
The mental account-keeping discussed in the section on food also applies to
exchanges of nonfood objects or services outside the hearth family. A valued
object's previous owners will be scrupulously remembered, but normally this
type of ledger-keeping is not openly discussed. An exception to this occurred
when a young man at Pumanaka buro, Mariquino, was jilted by a young
woman. He talked endlessly about getting back the gifts he had given her,
enumerating them and their value at the slightest opportunity. He also talked
about his demands to her mother that she return everything and supply farina to
compensate for those gifts that were irreplaceable, such as soap. By publicly
tallying up the amount of gifts made, Mariquino was explicitly severing the
relationship between himself and the girl and her family. One cannot openly
speak of debts in this manner without also implying criticism.
Trade and exchange occur at all levels of Vaup6s society and between people
sharing various degrees of intimacy. Many items are in constant circulation, and
ideally one must give away anything asked for. It is in this area that the
introduction of valuable goods to which people have differential access has
disrupted traditional patterns of reciprocity. Far more than was true previously,
objects have come to take on value in and of themselves. Traditionally the social
relations standing behind any exchange - which the exchange both acknowledged
and maintained - were of paramount importance. The act of exchange was
important both because of the intrinsic value in all social interaction and because
the maintenance of proper social relations was the means by which one ensured
one's continued economic well-being. This is thoroughly illustrated in myths,
which present myriad situations in which for various reasons the proper social
relations are disrupted or do not exist. Their lesson is that all types of exchange
are unsuccessful until the proper social relationships and expectations are rees-
tablished.
Thus, as is true in all small-scale societies, economic relations in the Vaupes
60
are embedded in the broader social web of relations of kinship and friendship. At
least in part, one's resources lie with other people, who are either themselves the
valued resource or the mediators necessary for access to other resources. Wealth
per se, totally divorced from other people, is a rare entity indeed.
Acculturation has already altered this picture and continues to do so at an
ever-increasing rate. The rubber camps alone have made a substantial impact, as
have the mission towns and Mitu. Obviously one area of radical transformation
is the introduction of money, although it is far from being a universal medium of
exchange in the region. Actual cash is in fact rarely seen in the area, by either
whites or TUkanoans. Perhaps if it were more abundant, white methods of
exchange would hold less mystification for Tukanoans. Traditionally, of course,
the value of an item or service could never be calculated in terms of an absolute
value in a universal currency. Specificity of value and of spheres of exchange
characterizes Tukanoan society, because in societies where most production is
oriented toward immediate subsistence, specific articles or activities can be
exchanged only for other specific articles or activities (O'Laughlin, 1975, p.
353). To sell baskets or labor in the Colombian market economy is utterly foreign
to traditional Tukanoan culture, because, first, all traditional economic activity is
embedded in the fabric of social relations. Second, many "economic" activities
concerned with production are sacred, which greatly differentiates them from
more secular ones. As a consequence, it is impossible to see the results of this
labor as equivalent in value to other objects or services undertaken in secular
contexts (unless one uses an arbitrary, non-Tukanoan standard of exchange). For
example, in order to make sacred artifacts such as feather headdresses, one must
be ritually pure, having previously observed food and other types of restrictions.
How can one calculate the value of these observances into the worth of the
resulting products? Tukanoans are dumbfounded when they discover that such
objects can be bought like any other.
The compadre (coparenthood) system has also been introduced into the
region. In the Vaupes only the asymmetrical variety is found, that is, a white
becoming the padrino (baptismal godfather) of a Tukanoan infant. Even this
form is infrequent (when compared to its development in other parts of Amazonia
[cf. Murphy and Murphy, 1974, p. 37, with respect to compadrazgo among the
Mundurucu]) and at an extremely underdeveloped stage. One reason is that at
least at present many whites are desperately poor themselves and cannot afford to
purchase the gifts required by the asymmetrical type of campadrazgo. Another is
that in general white-Indian relationships are neither formalized nor well-
established enough to permit such relationships to become customary.
It will be interesting to see the effects of increasing commercialism on native
crafts. I would predict that, first, standards of quality will decrease because they
will increasingly be applied by outsiders. This means that fine craftmanship will
disappear because it will not raise the price. Second, the inevitable homogeniza-
tion of style and pattern typical of all processes of folklorization and commercial-
61
The fish people
ization will result in the production of a few standard, recognizable types of
baskets or pots. These styles, with size of item taken into consideration, will
fetch standardized prices. To some degree this had already happened in 1970 with
respect to baskets, pottery, Cubeo bark cloth, and a number of items that are not
from traditional Tukanoan material culture - such as miniature canoe paddles,
tipiti's, and cigar holders.
Traditionally, all Tukanoans had more equitable access to scarce resources than
they do today. This is not to say there was ever complete equality; for example,
upstream and downstream longhouse locations vary in terms of the resources
available. But, in general, in the past if a Tukanoan wanted a particular object, he
either made it himself or asked someone to make it for him with an expectation of
eventual repayment. Inevitably, disagreements occurred and open quarrels erupt-
ed. Some people were doubtless seen as stingy or lazy, and others were envied. If
a wife was unhappy, she might quit working in her fields, bringing her husband
economic hardship and indebtedness. Difficulties arising from ruptured social
relations caused economic problems, whereas today increasingly it is the other
way around. Tukanoans report that in the past, quarrels were usually over women
rather than possessions, whereas today they are just as likely to be over posses-
sions. Many of the disputes I witnessed arose out of disagreements about objects.
The people in a Tuyuka longhouse on the Inambu River were irremediably
divided in half over who was the rightful owner of a shotgun.12 At present,
factions and long-term feuds seem to be produced mainly by disagreements over
white trade goods. Such goods - machetes, shotguns, clothing - are greatly
desired and are now necessities for Tukanoans, yet they are difficult to obtain.
Sadly, one is more likely to obtain them by ignoring rather than maintaining
established patterns of obligations to kinsmen.
Decisions to take care of one's own needs rather than one's obligations to
kinsmen not only can result in open quarrels but, more important, can be the
source of ongoing and unresolvable negative feelings toward oneself and others -
confusion, frustration, and envy. This is perhaps the saddest part of living in the
Vaup6s, regardless of whether one is a Tukanoan or an anthropologist feeling
angry and guilty at representing a society that is destroying a viable economic
system yet offering so little in return.
Property
As in all small-scale societies, it is difficult to speak of property as a single
concept, because, like exchange and labor, "property" is thoroughly embedded
in various social and ceremonial contexts whose meanings vary tremendously.
With respect to ownership of land, a given local descent group is associated
with a given territory. The periodic changes of settlement site occur within these
borders. Tukanoans have implicit usufruct rights to resources in these areas, but
the concepts of "tribal lands," "inalienable rights," or "title" are foreign ones.
62
In previous times, when indigenous invaders threatened, they threatened life,
limb, and above all women, but not land. Regardless of whether raiding and
warfare are in fact the result of a need for land, protein, or a similar basic material
necessity, with women being only the proximal cause (M. Harris, 1975; Ross,
1978; Siskind, 1973), this idea would not be acceptable to Tukanoans.
The presence of a government-appointed corregidor (magistrate), in 1970 a
Tukano, in Acaricuara has introduced Tukanoans to some ideas about group and
individual rights to land. But at least in 1970 this individual was very confused
himself and furthermore lacked the power he needed to put his policies into
action. His pronouncements were little more than food for thought for Tukanoans
of the Papuri region. Moreover, much of the dominant Colombian ideology is
concerned with individual rights rather than corporate rights. The latter, if
properly described, would make much more sense to Tukanoans - as they are
currently making sense to many Native American groups in the United States and
Canada in disputes over treaties and communally held lands.
Individual property seen as inalienably owned by the individual is confined to
items of almost daily use. These are the objects that tend to be buried with the
person at death because of the strong identification between object and owner.
Items used by several people and sacred items are normally communal property.
Examples of individual private property are a man's gun or fishing canoe and a
woman's old machete and baskets. All forms of adornment (in particular the
ubiquitous trade beads) are personal property, in part because body decoration
signals personhood. A naked two-week-old baby will be given beads and straps
for ankles and wrists. One of the few nonfood gifts spontaneously given to me
was a bead necklace, to make me "a Bara woman."
Examples of secular communal property are the cazabe griddle, the vats for
collecting and storing starch, the pot for smoking coca leaves, and the sugar cane
press. Examples of sacred communal property are the feather headdresses and the
Yurupari instruments. At times, boundaries can become blurred. When I left
Pumanaka buro for the final time, Mario, the headman, gave me a beautiful
black wood staff and a cigar holder, which had been made by his paternal
grandfather. (The gift was actually made to my father, because women are not
supposed to own such things.) It is rare to have an old item that can be given away
by an individual. Mario explained that normally such items are buried when their
owner dies, but his father had wanted it so much it was bequeathed to him. And
Mario in turn, had saved it when his father died.
The most sacred kinds of ritual property are always communally owned.
Examples are the sib's set of personal names or the Yurupari horns (which at
times are shared by two sibs; see S. Hugh-Jones, 1979, p. 143); similarly,
members of a given language group own their language and mythological tradi-
tion (C. Hugh-Jones, n.d., pp. 30-31). It is impossible to divorce the strictly
material from the nonmaterial, for this type of intangible property is property,
just as it is in other small-scale societies (e.g., the Northwest coast groups or
63
The fish people
Eskimo). Power is contained in ritual chants and Yuruparf music, and only people
with a right to do so should engage in activities involving them.
Another type of property right, although not an exclusive one, associated with
the language group is the manufacture of certain artifacts, often ceremonial in
nature. An example is the bone snuff taker associated with Tatuyos. No associa-
tion is made between item manufactured and name of group (for instance,
Tukanos making something from toucan birds or Tatuyos from armadillos).
Although perhaps not so in the past, any adult male Tukanoan today may
manufacture or own one of these items; but there is still a clearly symbolic
association between specific items and language groups. Another example is the
finely crafted Tukano stool. Some informants, as noted earlier, mentioned Bara
in connection with canoes. At present, canoes are not ceremonial objects,
although they are sometimes used to hold manioc beer during festivals and are
used as coffins. In the past, however, the larger canoes owned by the headmen
were ceremonial, for they brought people to festivals. Unlike the smaller individ-
ually owned fishing canoes, these canoes were carefully finished and painted on
the inside.
These different kinds of property rights are often symbolically associated with
each language group's founding ancestors and history and are connected to the
distinct roles played by each language group in Tukanoan mythology. Thus, they
are important in Tukanoan cosmology and also have a more strictly economic
role to play - which may have been even greater in the past.
The earlier role of specialization in manufacture of individually owned objects
is not certain, but it may have resembled the situation described by Chagnon for
the Yanomamo. He describes a division of labor and craft specialization by
village, adding that it is "artificial" and actually functions as "the social
catalyst, the 'starting mechanism' through which mutually suspicious allies are
repeatedly brought together in direct confrontation" (1968, p. 100).
As already indicated, Vaupe's settlements are not mutually hostile, suspicious,
or hesitant to visit one another, particularly if ongoing marriage alliances exist
between them. But a more strained and mutually suspicious situation may have
existed in a past involving active raiding and feuding. In much the same way as
basketball games between Cuba and the United States in 1976 smoothed the way
to diplomatic and economic interactions of a more delicate and potentially
volatile nature, preliminary exchanges of "needed" artifacts may have facilitated
marriage making and other sensitive negotiations in the Vaupes. Chagnon states
that the Yanomamo are so suspicious of settlements other than their own that
they say they would not visit even those with potential wives if they did not
desperately need the artifacts these villages specialize in. Yanomamo settle-
ments find it easier to accept another settlement's interest in their clay pots, dogs,
hallucinogenic drugs, cotton hammocks, and baskets than an interest in their
women. Continued intervillage contact is ensured through the obligation to repay
with a different kind of item. The exchanges are always between individuals, and
64
are reminiscent of the "ceremonial brother-in-law" relationship found in the
Pira-parana region.
According to Chagnon, such specialization among the Yanomamo cannot be
explained in terms of the distribution either of natural resources or of the
technical knowledge and skills required to make the item. As evidence for this, he
cites a case in which a settlement's alliance with a pot-making settlement grew so
cool that they could not ask for any more pots. The villagers responded to this by
suddenly "remembering" how pots were made and "discovering" that the clay
in their neighborhood was suitable after all (1968, p. 100).
Private property is always lent if borrower and lender are on good terms.
Thievery is usually an expression of disrupted social relations more than anything
else, because it is difficult to conceal the fact that one has stolen something.
Nevertheless, Tukanoans are becoming familiar with the idea of stealing and
"getting away with it," and they talk about earlier times when one's cargo was
safe in a canoe overnight, which is not always the case at present. A quasihuman
or subhuman character is frequently imputed to cases of theft. Maku are said to
steal, for example. When a danger exists of theft of cargo left overnight in
canoes, the thieves are spoken of in somewhat supernatural terms - they are
sometimes said to be able to fly, for example. The same is true for whites, who
comprise another not-quite-human category. The Bard described to me a kind of
forest demon who comes to a longhouse when everyone is gone and steals
valuable objects, including any small children who have stayed behind. He always
signs his name on pieces of paper and scatters them about so that people will
know who has been there, and he dresses in shirt, trousers, and rubber boots.
Leadership
The Vaupe"s share with other Amazon Basin societies certain characteristics in
leadership and political organization. For example, in general, leaders have
relatively little power and authority, and although headmen in many societies,
including Tukanoan society, inherit their office, they cannot become lazy or
despotic with impunity. Most often, leaders must give more than the others in
order both to attain and to maintain their positions even though the position of
headman is usually inherited (but see Kracke, 1978). This contrasts with some
other parts of the world, for example, Melanesia, where the opposite holds, and
the ability to accumulate is the path to power and prestige.
Leadership in the Vaupes and other Amazonian societies has an ad hoc
quality. Positions are often task-specific and temporary. A contributing factor in
the relative lack of importance and permanence attached to leadership positions
is the ease with which a disgruntled individual, family, or faction can leave a
settlement and live elsewhere. Because one of the main tasks of a headman (and
an important diagnostic of his success and source of his higher status) is to hold a
community together, the possibility that some of his constituency might leave is a
65
The fish people
continual and unpleasant consideration. Thus the relationship between leaders
and followers is a strongly voluntary one, a fact that undoubtedly plays a role in
determining the type of person who can become a successful leader.
In general, the political structures characterizing the Vaupes and other low-
land South American societies are relatively acephalous, with legal and juridical
procedures informal and largely implicit. Much more research needs to be carried
out in this area, and the meaning of such concepts as power, authority, or
influence in societies of this type needs to be better understood (Arvelo-
Jimenez, 1971; Riviere, 1971). These generalizations regarding personal power
do not apply as much to Tukanoan war leaders, although like the others these
leadership roles were characterized by task specificity and temporality. During
hostilities, war chiefs apparently had a fair amount of personal power, at times
over a much larger group of people than normally. This seems to be true for the
Vaup6s (Markham, 1910). It is generally true in the Amazon Basin that any
manifestation of personal power by an individual, whether over humans, super-
natural beings, or game, is met with ambivalence and suspicion on the part of
other members of the society. This undoubtedly lessens the desirability of these
kinds of positions and decreases the number of aspirants. Leaders are often little
more than older men with certain not-too-remarkable qualities - better judge-
ment, diplomatic skills, ability to organize and entertain - or simply well-
respected adult men. Many of the traits we tend to associate with political
leaders, such as arrogance, cunning, and egotism, are much more likely to be seen
as characteristics of misfits or overly ambitious malcontents, who are very seldom
chosen as leaders. (See Murphy [1961] for a well-presented discussion of the dif-
ferences in leadership qualities found between Brazilians and the Mundurucu.)
Although political structures in these societies are in general acephalous and
decentralized, social control mechanisms such as ordeals, duels, feuds, and
"voluntary" emigration do maintain a certain degree of order (Dole, 1966).
Serious cases of antisocial behavior are most often seen to be spiritual as well as
secular crimes - indeed, this dichotomy is meaningless in most instances.
Executions and other serious punishments meted out by headmen or shamans are
often seen as having been decided and sentenced by higher authorities. For
example, in some groups an execution takes place only after a divination has
demonstrated supernatural approval of it. The participation of spirits or ancestors
is necessary, because all such "criminal" cases involve a disruption of harmonic
relations between human and spiritual worlds (Butt, 1965-66; Dole, 1964).
I did not have the opportunity to witness a truly strong headman on a
day-to-day basis in the Vaupe's. When I arrived at Pumanaka buro, it consisted
of five families, and when I left, only two families were still living in the
longhouse under the headman, Mario. A satellite house built alongside the
longhouse during my stay started with one hearth family but grew into three
following two marriages. One of the three hearth families who left during my stay
66
had a history of moving in and out. They had a small house upstream. The head
of this family, Anastasio, was older than Mario, and a certain amount of rivalry
existed between them. The two other hearth families moved out over a single
issue. The precipitating reason was a young woman, Magdalena, in one family
who became pregnant by a young man, Juanico, in the other. As is always the
case, these departures were also the result of a number of long-standing frictions.
The headman, Mario, played a complex role in all of this, both as the head of one
hearth family and as headman.
It is difficult to generalize about the position of headman when only one
specific person playing the role is seen on a day-to-day basis. It is my impression
that Mario was a typical headman in many respects, despite his failure to resolve
the crisis just mentioned. Other information I have about the position comes from
myths and anecdotes and hearing news about other settlements and their headmen
from time to time.
The headman of a longhouse has some authority and almost no real power
(i.e., being able to back up requests made to others with a threat of sanction).
Vaupes headmen undoubtedly have less power and authority than those in many
other places where villages are larger or where the possibility of warfare still
exists. A Tukanoan headman rarely directly tells another individual what to do.
He leads by initiating activity himself or by suggesting what should be done to
the group as a whole. Tukanoans have learned that whites attribute more power to
headmen than they actually have, and when the people talk with whites (particu-
larly if in Spanish) they will often exaggerate the power of the office and of a
given headman. If the traditional headman is the same person as the capitan (the
position of headman recognized by missionaries) - which is not always the case -
his power is thereby increased. Sometimes an older man is the traditional
headman, and his son, more missionized and able to speak Spanish, is called
capitan by the priests, who will refer to that settlement by the younger man's
name. Some priests are unaware of the traditional headmen in the settlements
they visit.
The term for longhouse headman can be roughly glossed as "owner of the
house." A specific longhouse will be referred to by its site name (most also have a
saint's name given by the priests) or as "so-and-so's house." Sometimes a
headman's Tukanoan name is used (e.g., Pai ya wi; Pai is the traditional personal
name for a man in the headman line for the Bard Yoara sib at Pumanaka
buro), but his Spanish name is also used: Mario ya wi ("Mario's house").
Although he usually inherits the position, an effective headman must also have
certain qualities. These include a level head, good judgement, tact, an ability to
contribute more than others, and an ability to initiate and direct activities without
being overbearing. An asymmetrical exchange occurs between the other residents
of the longhouse and the headman and his wife, for they give more energy to the
group as a whole in exchange for higher status. A headman's wife works harder
67
The fish people
than the other women throughout her life. She invariably helps with communal
labor (e.g., sweeping), most often initiating it. She must also produce more
manioc beer than the other women.
Any individual who does not want to live under a particular headman will
move out. However, when a family moves out it is for complex reasons. Dis-
agreements with the headman can concern him as head of another faction in
addition to his role as headman.
Ideally, everyone who "belongs" to a longhouse should live in it (i.e., people
should live with their closest agnatic kinsmen). It is the headman's job to see that
this happens as much as possible. He should also try to see that the affairs of the
longhouse are decided and carried out in as democratic a manner as possible.
Perhaps when the office of headman did involve more decisions and power,
refusal to live under the authority of a particular headman qua headman was more
frequently the reason for a longhouse fissioning. The disputes I heard about all
concerned other types of problems. But in the past more hearth families lived
under a single roof, and thus a greater need existed for organization and leader-
ship, which when not forthcoming was more visible and intolerable. Headmen
who were successful war chiefs undoubtedly had much more power, at least
during times of crisis.
At present, no political organization in a formal sense exists in the Vaupes at a
level above that of the settlement. There is evidence that in the past, Indians
formed federations in the various uprisings against rubber gatherers and other
whites (Markham, 1910). And in the future, if the federation plans of some of the
corregidores (Tukanoans appointed by the government in each of the mission
towns; their tasks are unclear, but they are to represent the town and their
language group) are successful, a corporate decision-making body may emerge
that involves a substantial number of settlements. At present, political activity
that includes Tukanoans from more than one settlement is not organized along
any formalized lines and does not involve anything more than the most temporary
ad hoc kinds of groups.
68
5.
Vaupes social structure
The settlement
At 0.2 inhabitants per square kilometer (Instituto Geogrifico "Agustin
Codazzi," 1969, p. xii), the population density of the Vaupes is comparable to
that of many hunter-gatherer groups as well as to the more sparsely populated
horticulturalist societies in New Guinea (Glasse and Meggitt, 1965). Thus,
although Tukanoan settlements are concentrated on or near rivers and streams,
they are still extremely isolated, an isolation compounded by the difficulty of
traveling either by river or by land. Longhouses currently hold approximately 20
to 40 people (some of these are children who attend mission school for up to
eleven months of the year). Nucleated villages have anywhere from 12 to
approximately 60 inhabitants, and some mission towns have more than 100
people.
The census I took in 1968-70 shows approximately 165 settlements for the
region south of the Vaupes River, including the Papurf River and its tributaries,
the Tiqui6 above Buena Vista, and the Pira-parana above Caruni Rapids. In
1970 this figure included 24 longhouses. The remaining settlements are a variety
of arrangements, from one- or two-family groups to large mission towns, so
making generalizations about the Vaupes settlements is somewhat tricky.
Although the majority of settlements are nucleated villages, these do not
necessarily represent a simple changeover from longhouse to several small
one-family houses containing the same inhabitants. As has been pointed out,
several longhouse groups, or segments of them, may make up one of the larger
villages. And household composition in settlements on the Vaupes near MM is
sometimes based on affinal rather than agnatic ties, usually involving a couple,
their daughter and son-in-law, and at times various other relatives.1
Tukanoans located on downstream sites, on major rivers, and near mission
towns are more acculturated than their upriver counterparts. To some extent this
allows generalizations about degree of acculturation for entire language groups.
For example, the Desana, Tukano, and Uanano are as a whole more acculturated
than the Tuyuka, Bard, and Tatuyo because traditionally the former have been
69
The fish people
located in the more accessible parts of the region. As has been indicated, the
more acculturated Tukanoans feel superior to their "backwater" compatriots.
The overall recent trend has been to abandon longhouses,2 in part in response
to pressure from missionaries, and to build smaller mud-walled houses in nucle-
ated villages or mission towns. The patrilocal rule of residence and settlement
exogamy are also breaking down, both because missionaries settle members of
different language groups in the same settlement and because Tukanoans now are
starting to form households based on affinal ties between males.
In traditional Tukanoan life, people living in a settlement are linked to one
another in terms of two profoundly important organizational principles: coresi-
dence and local descent group membership. Obviously all the people living at a
settlement are its coresidents, although at times it is difficult to distinguish
between those who are residing there and those who are merely visiting. The
local descent group is (ideally) composed only of those people who are born
there and who are one another's closest agnatic kin. Inmarried women are not
adopted into this group but remain members in absentia in their natal local
descent groups. Thus these women are in a pivotal position, living in and
identifying with their new settlement as residents, creating ties to it through their
husband and children in addition to any they may have had prior to marriage. But
their ties to their own natal local descent group remain strong, and conflict
occurring between the two can produce psychic turmoil in women caught in the
middle. This theme is found in a number of myths, which picture these conflicts
in the starkest of terms, resulting in betrayal and death of either husband and
children on the one hand or of parents and brothers on the other. Where a
woman's sympathies should lie is a moot point, one of the paradoxes of this type
of social structure, and the portrayal of this in myths supports the Levi-
Straussian view that myths fundamentally deal with the unresolvable and unwel-
come contradictions of life (Leach, 1970, p. 80).
It is important to note that coresidence is in itself a crucial factor, regardless of
how the members of a settlement are otherwise related. Virtually all settlements
have people living in them who "do not belong" in terms of the patrilocal
residence rule. These people participate fully in the life of the community and,
over time, form close bonds with their fellow residents. Still, it would undoubt-
edly take at least two generations for them and their children to be assimilated
into the local descent group as bona fide members, and it might take much longer
(I have no actual cases of such assimilation). Further evidence of the strength of
the coresidence factor is that settlement exogamy is the norm even when the
actual kinship relations between two people would allow them to marry; I
recorded no such marriages, with the exception of a few made by Tukanoans
living in the same mission town. I asked informants about hypothetical matches,
for example, one involving a Bara boy and a Tuyuka girl (whose father was away
at a rubber camp) growing up at Pumanaka buro, and was told that they would
not want to marry each other.
70
The levirate (i.e., a widow's being expected to marry her dead husband's
brother if at all possible) is further evidence of the strength of ties built up over
time by coresidence and of the desirability of having the children born into a
community reared in it.
Thus the local descent group is a corporate group - the only descent-based
group that invariably is, because both descent and residence are requirements for
membership. Although some residents at a particular settlement are not members
of its local descent group, for many purposes we can consider them as if they
were. Thus, regardless of whatever the actual kinship relations are, the coresidents
of a settlement live, work, and communicate with one another as a unit, and form
strong bonds on these bases alone. Special ties are forged between pairs or larger
groupings of individuals because of their daily interaction and the special emo-
tional flavor and intensity produced by this closeness. The strength of these ties is
revealed when a chronic dispute becomes so intolerable that a family or other
segment decides to leave. The major rift of this sort that occurred at PQmanaka
buro during my stay was precipitated by a case of incest, but many other types of
quarrels can leave wounds that do not heal. When a family leaves, the entire
settlement is critically wounded, not only because of the blows to its pride and
sense of unity but also because its members must sever many personal ties
created over a long period of intimate interaction and sharing.
The sib
Vaupes sibs (clans) are named, ranked, exogamous, localized patrilineal descent
groups. Sib structure and unilineal descent in general were once thought to be far
more common in the Amazon that has turned out to be the case: Many lowland
South American societies have been reclassified from unilineal to cognatic, but
this is not true of those of the Northwest Amazon. Sibs were correctly identified
for this region by Whiffen (1915) and Nimuendajii (1950), with further
clarification appearing in accounts by Goldman (1948) and Fulop (1955).3
In principle, the local descent group is coterminous with the sib. That in reality
it most often is not is the result of many factors, one of which is the inroads of
acculturation on traditional social structure. But the lack of fit between idealized
structure and reality also derives from the fluidity found in Vaupes social groups
even in their most traditional state. Although the sib is a very important social
unit for Tukanoans, a concern with sib boundaries as such (or for language group
or phratry boundaries), particularly with respect to genealogical reckoning, does
not characterize Tukanoan thinking about sibs. An indicator of this is the lack of
tightly defined labels for sibs and other social structural units. The expressions
used are imprecise (e.g., "children of one parent," referring to the original
sibling group ancestral to a present-day sib; or "children of such-and-such
anaconda") and are sometimes applied to different levels of inclusion. This lack
of precision does not appear only when applied to specific on-the-ground group-
71
The fish people
ings over time but characterizes some of the idealized statements about sibs as
well. At the highest level of abstraction, illustrated by Figure 4, everything is
tightly bounded and precise. Of course, assimilation and fission due to demo-
graphic fluctuations and acculturation create ambiguity and contradiction; today
the system is changing in a very speeded-up fashion. But the fluidity appears at
more than an empirical level and is due not only to some sort of "breakdown"
but to built-in features of the system that allow it; this point is made by other
Vaupe's specialists as well (for instance, C. Hugh-Jones, 1979, pp. 22-23).
Paradoxically, the clearest picture to be had of the underlying principles of
Vaupe's social organization appears when a Tukanoan is asked about an unfamil-
iar group; although concrete information is more scanty, what the individual
knows will be described in regularized terms that conform to these principles - a
much more difficult task when describing more familiar groups whose aberra-
tions and exceptions to such principles cannot be ignored.
At present, very few local descent groups are exactly coterminous with the sibs
to which they belong.4 A local descent group may be a lineage forming part of a
sib, parts of two sibs, or two entire sibs (Jackson, 1972, app. 1). Usually this lack
of fit does not matter, and sibs and local groups are spoken of as if they were
always the same thing, and in many respects they are. Both are local groups.
Membership in both is reckoned primarily through agnatic links - although
Tukanoans do not care about lineage formation per se, and genealogical memo-
ries are shallow.
Sibs are named, and these names often refer to plants or animals. Sib names
can also refer to sib ancestors and their immediate descendants; this is also true
for the personal names owned by each sib. These personal names are given to
infants in a prescribed order. The eldest son of the headman ideally is the
first-born male of his generation and receives the first name on the list. This
sib-supplied name fosters growth, for it associates the newborn child with a
nurturing group of agnatic kinsmen. The infant becomes more human upon
receiving a name, for it is an explicit affirmation of membership in the sib,
entitling it to the power and nurturance available from the ancestors.
The sib name is an important component of its definition: C. Hugh-Jones has
noted that the "identity of a sib is so intimately bound up with the name that, in a
sense, the name is the sib" (1979, p. 26). She also notes that dispute over sib
membership revolves around which named sib a person properly belongs to, as
opposed to disagreements about genealogical links.
The sibs in a given language group are ranked. The order of ranking is
explained as corresponding to the order in which a group of brothers, the
ancestors of the various sibs, emerged from the rocks at a particular rapids site.
According to Bara informants, the "first people" came from the east in an
anaconda canoe5 piloted by Ko§-mahku, a culture hero. The Bard were the
last to leave the canoe, getting off at Yapu Rapids on the upper Papuri. Beyond
this point, Dir6 koa pond (literally, "piece-of-meat-bone-offspring"), two
72
I 3
n Bara TUkano Wuhana etc. Tuyuka etc. etc.
m Bard Yoara Wainakoroa etc. (approx. 30 sibs) etc. etc.
, A
(Pflmanaka
rv buro,1979)
B C etc. M N O P Y Z etc. etc.
co-agnates
Figure 4. Traditional Vaupes social structure. Level I represents the phratry, an unnamed unit composed of various
language groups; at level II is the language group, which is commonly referred to as "tribe"; at level III is the sib, a
named group that occupies one or more settlements along a stretch of river; and at level IV is the local descent group,
which can be eoterminous with the sib and is composed of coagnates who are one another's closest agnatic kin and
who share the same settlement (traditionally a longhouse). The capital letters in level IV represent current locations
of local descent groups, who are known by their settlement name, which, unlike the sib name, changes when the
local descent group moves its longhouse site.
The fish people
culture heroes who are brothers, closed off the land, making it impossible for the
canoe to travel any farther. They did this by creating Ewtira tar6, a miriti swamp
from which all rivers are said to originate.
The ranking of sibs is continued today with the use of elder-younger sibling
terms between members of different sibs. However, in some language groups the
difference in rank between certain pairs of sibs is so great that generational
divisions are brought into play. This results in an unusual and initially surprising
usage of coagnatic terminology. A person who belongs to a considerably higher
ranked sib than another will address the other as "uncle" or "grandfather." This
seemingly incongruous state of affairs is explained by Tukanoans as follows: The
first ancestors of all the sibs of one language group were brothers to one another.
The eldest brother emerged from the rocks at the rapids first, and the youngest
last. However, there were many brothers in the beginning, and obviously there
were many years between the birth of the eldest and the youngest brother. By the
time the youngest brother emerged at the rapids, the eldest was very old, and had
great-grandchildren. Thus, although the eldest and youngest brothers called each
other "brother," because many years had passed between their births the younger
brothers were addressing as "grandchild" those individuals in the eldest broth-
er's sib who were close to them in age. This is why, today, when people of about
the same age are heard using grandparent and grandchild terms to each other, it is
the one who says "grandfather" and who is called "grandson" who is of higher
rank.
Sib rank is signaled in other ways as well. One method of indicating a sib's
very low rank is to impugn its origins with the claim that it is a "new" member
of the language group. It will be implied that this sib was originally a Maku-like
group, "who were our servants, who had to be taught how to build houses and
speak our language. Then, taking pity on them, we adopted them as our
youngest-brother sib." (This is not a verbatim quotation but a composite of a
number of conversations.) The Wamutanara (named after a fernlike plant,
Selagenella sp.), a Bara" sib, is ascribed this origin by members of higher-ranked
sibs.
Another way of indicating a sib's low rank is to imply, during a recital of the
origin myths, that the sib's ancestors did not come from the underworld river
(Opeko dia, sometimes identified with the Amazon River), which joins the
Vaupes far to the east, but rather were picked up much farther upstream. The
Wamutanara sib is spoken of as boarding the canoe at Yapu Rapids on the
upper Papurf, rather than farther downstream. The ascent of the ancestors from
OpekS dia concerns the origin of both sibs and language groups. But the origin
of sibs is usually depicted as the emergence of each sib's ancestor - siblings to
one another - out of the rocks at a specific rapids site, their "birth" order
determining subsequent sib rank. Speaking of an ancestor of a Bara sib as
ascending in the anaconda canoe, in this case the Wamutanara, hints that this
sib is seen as rather distant from other, higher ranking Bara sibs. The distinction
74
between the ancestors as parents of sibs and parents of language groups is often
murky in regard to specific groups and categories; this will be discussed later.
Another mythical reference to sib rank occurs in the myth of Wai Pino
Mahk6, "Fish Anaconda Daughter." Anacondas are considered to be the pro-
genitors of all fish. The various species of fish are also ranked like sibs, with Fish
Anaconda at the head. Fish Anaconda Daughter marries a human, Yeba Mahu
who travels with her to her longhouse - where all the fish people (Wai maha)
live. Yebd Mahu sees fish all around him and wants to eat some. His father-in-
law, Fish Anaconda, seeing this tells him to eat Tarira fish (doe, Erythrinus sp.)
because, "although D06 were Fish Anaconda's younger brothers, they were like
Wamutanara are to us" (i.e., they were the lowest-ranking species and
therefore the most expendable).
A different way of pointing out low sib rank is to imply that such-and-such sib,
in the recent past or now, behaves inappropriately. One Bara man of (according
to him) the highest-ranking sib, Bara Ydara, assured me that in the past his sib
always wore bark-cloth loincloths (wedira), whereas the men of the lower-ranking
sibs went totally unclothed.
Linguistic cues can also signal low rank, at least according to members of
high-ranked sibs. Members of the Wamiitanard sib were described by other
Bara as speaking ungrammatically and with incorrect pronunciation (C. Hugh-
Jones, 1979, p. 18; also see Chapter 9).
It may be that in the past the most appropriate marriages were seen to be
between people from equivalently ranked sibs of affinally related exogamous
groups (i.e., language groups). This would follow the ideal Cubeo pattern,
although the Cubeo case does not involve different language-affiliated exoga-
mous units. I did not find any evidence of this, however, and S. Hugh-Jones notes
that contemporary Barasana deny that this would happen (1977, p. 206).
With so many available indicators of sib rank, one might expect that investigat-
ing and drawing conclusions about the names of present-day sibs, their relative
ranking, and the meaning and function of sib structure would be an easy task.
Such is unfortunately not the case. Although the lists I obtained concerning Bard
sibs were in general agreement as to their locations and rank, they never com-
pletely tallied with one another. And Bara elsewhere (for instance, in the
Pira-parand) have rather different classification systems; for example C. Hugh-
Jones's data indicate that Inambu Bara belong to the Hamoa sib; I have no
mention of such a Bard sib in my notes. In general, therefore, Tukanoans can be
seen as notoriously slippery in their evaluation of relative sib rank and under-
standing of who lives where. All Bard I spoke to did, however, agree that
Wamutanard were the lowest ranked.
The meaning of sib structure, including the question of hierarchy versus
equality and its functions in present-day Tukanoan society, is also problematic.
We can certainly conclude that sib hierarchy today has almost an exclusively
symbolic meaning. One would expect, of course, symbolic statements ultimately
75
The fish people
to reflect social processes in some fashion; in the Vaupes this association, if

existing at all, did so only in the past, and the nonritualistic behavioral correlates
have been lost owing to declining population numbers and cultural loss as the
result of acculturation. The Hugh-Joneses discovered an extremely interesting
system of five-tiered sib ranking differentiated by specialist roles. These are (in
descending order): headman, dancer, warrior, shaman, and "cigar lighter." This
last category is obviously close to Maku status. (The ranking is in fact more
complex; see C. Hugh-Jones, 1977, p. 186; 1979, pp. 27, 65-67.) They also
discuss the possibility of this being a pan-Vaupes system. I cannot supply any
additional data with respect to this set of specialist roles (which play no significant
role in daily life today, and perhaps did not in the past) except to point out bits of
confirming evidence in myths, some associations made between Maku and
low-ranking Bara sibs (see Chapter 8), and some connections in present-day
Bara sib names. For example, Bara Yoara, the sib represented at Pflmanaka
buro, almost certainly corresponds to the Hugh-Joneses' dancer specialization.
Awareness of this as a system seems to have virtually disappeared for the Papurf
region (it could also be, of course, that I did not ask enough of the right
questions).
Sibs are exogamous, but this is not their function; nor can we find evidence for
any economic, political, or other social regulation that might fall under control of
the system of ranked sibs. The only current economic difference, an indirect and
very slight one, is the tendency for higher-ranked sibs to be located more
downstream than the lower-ranked ones (see Chapter 4). Thus, at least today, sib
rank does not imply any difference in power, wealth, or style of life (S. Hugh-
Jones, 1979, p. 2).
Sib organization today seems mainly to make statements about group member-
ship and relative closeness among various units. These statements are almost
exclusively symbolic. Ritual property seen as owned by the sib includes a set of
Yuruparf instruments and the right to avail oneself of the benefits from their use.
Speech differences, as mentioned earlier, also are seen to correlate with sib
boundaries (although not nearly so extensively as those that ought to separate
exogamous units); minute differences in speech behavior, therefore, are certainly
markers and perhaps can be seen as intangible property belonging to a sib (this is
discussed further in Chapter 9). Other kinds of customs, for instance, those
related to artifact manufacture, sometimes are described in similar fashion,
although expectably this is hard to pin down in black-and-white terms (the minute
one attempts to do so, one is contradicted by a Tukanoan quite clearly uncomfort-
able with someone else's making the kinds of invidious comparisons he or she
has just finished making).
In general we can say that the closer a given pair of unilineal descent-based
units is (closeness loosely conceived of in genealogical and geographical terms)
the more similar will be their ritual property and rights to its use. Because such
property and rights ultimately relate to ancestral power, "it is quite appropriate
76
that variation in rights to ritual items should mirror descent ties between groups"
(C. Hugh-Jones, 1979, p. 30).
Before leaving localized groups for larger units, it might be helpful to summa-
rize their important characteristics, disregarding for the moment the differences
existing between types of unit (e.g., local descent group, sib, settlement). The
group of people living at a single place - the settlement - is the most important
unit in the Vaupes social system. Residence and kinship, both crucial for
building cohesive, long-lasting social groups, operate at this, and only this, level.
The settlement's importance is further underscored when we consider its isolation
from other settlements and the degree of autonomy it has in many interactional,
economic, political, and ritual areas of life. Undoubtedly the more inclusive units
of Vaupes social organization are not as cognitively and affectively important to
Tukanoans.
The important features of traditional local group organization in the Vaupes
include (1) communal multifamily longhouses, one per settlement; (2) named
patrilineal, exogamous, localized sibs; (3) patrilocal rule of residence; (4) settle-
ment exogamy; (5) a tendency to classify affines as "not like us," or "outsid-
ers"; (6) a list of personal names possessed by each sib given at birth in a
specified order; (7) sib names that sometimes have a "totemic" flavor, that is, an
association with a plant or animal.6 Sometimes this association is made more
explicit by attributing similarities in qualities between the members of a sib and
its eponym.
This constellation of features is found only in the Central Northwest Amazon
area and is the key to understanding many aspects of its social organization.
Probably most important are the patrilineal localized sibs, and settlement exoga-
my.
The language group

Tukanoan settlements exhibit an autonomy in many important respects and an
equally fundamental dependence on other settlements in others. Settlements have
been structurally interdependent at least since the establishment of the rule of
local exogamy; it is perhaps this feature of Tukanoan social organization that has
been most influential in producing the degree of regional network organization
found at present in the Vaup6s. I do not know when the rule of local exogamy
came to be observed completely (as it is today, with the exception of some
marriages made in mission towns). Earlier documentary evidence (e.g., Koch-
Grunberg, 1909-10; Wallace, 1889/1972, p. 346) indicates its existence during
their visits. In some respects settlements were more autonomous in the past,
simply because longhouses were much larger than now (MacCreagh, 1926;
McGovern, 1927) and some settlements may have contained more than one
longhouse.
As is true for relationships within the settlement, most intersettlement interac-
77
The fish people
tion is structured along kinship lines. Every Tukanoan is in some respect a
kinsman of every other Tukanoan, or at least potentially so. One important way in
which kinship ties, agnatic or affinal, are structured is through the system of
more than sixteen exogamous language groups.
In Tukanoan mythology, language groups originated when the anaconda canoe
ascended Opek6 dia and the Vaup6s River, bringing the ancestors into the
region. As they traveled upstream, they became increasingly more human,
acquiring skills and knowledge about useful plants and animals. As the canoe
passed various rapids and other landmarks, the ancestors of specific groups left
the canoe and began settling the surrounding area. According to the Bara they
were the last to leave the canoe, and the ranked sibs emerged at Yapu Rapids. In
another version of the myth, the "children of one parent" (hikd pond, full
siblings), who spoke the same language {maniya waderd, "our language"),
emerged from Yurupari Rapids on the Vaupes River. This group included the
Wai maha ("fish people"), Dahea (Tukano), Kumara1 pona ("children of the
rainy season," now a Tukano sib), and Wahuna (a group living on the Vaup6s
River and probably the same as Koch-Grunberg's Uasona, "Pisa-Tapuyo").
Underneath the Yurupari Rapids is what can be glossed as the "first-people
longhouse" (Pamuri Wi), which is also known as the longhouse of the Wai
maha. Thus, whereas Yapu Rapids and in general the upper Papuri refers to the
Bard when conceived of as a group of ranked sibs, Yurupari Rapids refers to the
emergence of an exogamous group. As is typical of myths, basic "facts," or
truths, are approached from different vantage points and described in different
and at times contradictory ways. It should also be noted that the list of four groups
emerging at Yurupari Rapids is undoubtedly one of several versions of this
section of the origin myth.
Thus, the basic features of Vaupes social structure dealt with in the Bara
origin myth are (1) the arrival of separate groups of Tukanoans and the process of
becoming fully human through the acquisition of artifacts and materials, knowl-
edge, and skills; (2) the origin of the system of ranked sibs in the birth order of
the original sibling set; (3) the origin of the "children of one parent" who must
seek marriage partners from other groups; and (4) the assertion that in the
beginning, such "children of one parent" spoke the same language. The emerg-
ing of four groups at Yurupari Rapids implies that they all spoke the same
language, even though now two of these groups, Dahea and Kumara pona, are
identified with Tukano (this is discussed later in the section on phratries).
Keeping in mind that the actual situation is more complex, we can initially
define the language group as it exists today as a named patrilineal descent unit
composed of from six to more than thirty sibs, ideally identified with a distinct
language. Sixteen or more are found in the region. Their members observe a rule
of exogamy and most often terminologically distinguish agnates from other kins-
men at this level. Members of a given language group identify with co-members
as "brother people." Distinguishing features that serve as significant symbols of
78
Vaupe's social structure
identity are (1) the language and name themselves; (2) separate founding ances-
tors and distinct roles in the basic Tukanoan origin myth cycle; (3) the right to
ancestral power through the use of certain linguistic property such as sacred
chants; (4) the right to manufacture and use certain kinds of ritual property such
as the Yurupari instruments (which may actually belong to the sib); and (5) a
traditional association with certain ceremonial or near-ceremonial objects. The
difference between items 4 and 5 is that the latter type of object may be
manufactured or owned by Tukanoans not of the language group associated with
it. The exceptions to this definition of Vaupes language groups are discussed
later.
Language groups do not occupy an exclusive and continuous territory. Map 2
shows the distribution of some Vaupes settlements according to language group
affiliation. The degree of language group interspersion by settlement varies
subregionally; for example, they are more territorially circumscribed in the
Pira'-parana' region. All language groups are confined to specific areas within
the region as a whole, but it is equally true that all language groups have some of
their member settlements interspersed with settlements belonging to other lan-
guage groups.
I have chosen to emphasize the feature of language affiliation in my choice of
terminology for these units (as opposed to tribe or the Hugh-Joneses' exogamous
group) for several reasons. The first and most important one is that this is the way
the Tukanoans talked about it. Of course, what people say they do and what they
actually do may not be entirely the same. But the notion of the coincidence of
linguistic and agnatic boundaries is an extremely strong one, and my research has
shown that this ideal coincidence is almost perfectly matched in practice among
Papurf groups. C. Hugh-Jones notes that the number of exceptions to this
"among Pira'-parana' groups probably reflects their marginal position at the
southwestern extreme of the Tukanoan culture area in the same way that the
exceptional language situation among the Cubeo probably reflects their extreme
northerly position" (1979, p. 19). Doubtless in part because the Hugh-Joneses
worked in the Pira-parand and therefore the exceptions made a strong impres-
sion on them (I was unaware of any exceptions until conferring with them), they
have opted for the term exogamous group. For the same reasons, although
reversed, I stick to the term language group. I do want to make clear that I can
incorporate exceptions into the general model, and in those instances language
obviously will not be an invariant feature; the difference has to do with what we
wish to highlight with our choice of label. And, just as the existence of
exceptions raises problems with my terminology, so does the existence of a
phratric organization at a level more inclusive than "simple exogamous groups"
create difficulties with that label (C. Hugh-Jones, 1979, pp. 15-20). In sum, the
situation is complex and it is not easy to find a term both clear and yet not
oversimplified. Language group is a perfectly adequate term if the exceptions are
borne in mind. In general, the language-exogamous unit equation is extremely
79
Cano
Map 2. Membership of settlements, either longhouses or small villages, by language group of a section of the Vaupes.
BR
Language groups
Piratapuya
Carapana
Tukano
Tuyoka
Siriano
Desana
Tatuyo
YR Yuruti
Bar a
CR
TlT
DS
BR
PR
TY
TK
SR
Behuya
Cano
pervasive, being found even among the Yukuna, Tanimuka, Letuama, and Matapi
groups in the Apaporis region to the south (Jacopin, 1981).
Table 1 lists the language group affiliation of Tukanoans in the marriage
sample I collected during fieldwork in 1968-70. Although it is not by any means
a complete census of language group membership for the multilingual area, it is a
nearly complete census of the living married Tukanoans in the Papurf-upper
Tiquie drainage region in 1968-70, including the inmarried women from natal
settlements elsewhere and the women born in the area who have married into
settlements elsewhere.
Table 1 gives a list of marriages and language group affiliations collected from
Tukanoans mainly of the Papurf-upper Tiquie drainage. Thus, the names of the
language groups are not to be seen as the list of social groups identifying
themselves by language and marrying exogamously, a list accepted by all
Tukanoans in the entire Vaupes. Although the Tukanoans I talked to agreed
completely about people in nearby territory, vague and disputed language group
designations appeared when a Tukanoan was discussing groups farther away. It is
apparent that the way members of those groups describe their social universe is at
times quite different from the classifications offered by Papuri Tukanoans. C.
Hugh-Jones discusses this issue in general terms by noting that "the problem
inherent in all attempts to define social structural units in the Vaupes is that
Indians are more concerned with the types of relationship (hierarchy, etc.)
outlined in the model above than with the precise definition of social boundaries"
(1979, pp. 22-23). In this same paragraph she notes that named groups overlap
in membership at times and sometimes the categories are "sliding" - covering a
more or less extensive population, depending on context.
These complications stem from several sources, in addition to the one noted of
Tukanoans' relative lack of concern about details of boundary maintenance. First
of all, a number of Tukanoans (particularly ones in the Papun-Tiquie region, my
principal informants) do not understand, or at least do not admit, that a genuine
regional variability exists with respect to the definition of language group; these
definitional differences are apparently limited to the Pira-parana but are impor-
tant, nonetheless, although there the ideal of a coincidence of language unit and
exogamous group (C. Hugh-Jones, 1979, p. 18) does exist. Second, Cubeo do
not have to marry out of the Cubeo-speaking unit. Third, all Tukanoans tend to
make situations appear neater than they are in fact, with the result that some
individuals even in the PM-parana will simplify the situation when talking with
an outsider, as was the case with a Carapana informant from Cano Ti from whom
I collected some material on social structure. Fourth, some investigators have not
fully understood the basic outlines of Vaup6s social classification, which has led
to contradictoty lists and descriptions in the available literature. Fifth, Tukanoans
do not have words for "sib," "exogamous group," or "phratry." Sometimes
language groups are designated by using the name of one of their high-ranking
sibs as well as the language name, but at times this sib name covers some but not
81
The fish people
Table 1. Language group affiliation of Tukanoans in sample
Language Men Women Total
Bard 104 86 190

Tuyuka 160 117 277
Tukano 153 162 315
Desana 65 77 142
Carapana 40 20 60
Tatuyo 28 40 68
Siriano 73 50 123
Yurutf 16 25 41
Piratapuya 8 13 21
Uanano 4 8 12
Cubeo 8 25 33
Barasana 5 25 30
Taiwano 1 6 7
Makuna 3 9 12
Tariano 0 11 11
Curripaco 0 2 2
Carihona 1 1 2
Maku 0 1 1
Wahiina (Pisa-Tapuyo?) 0 2 2
Metuno 2 0 2
Arapaso 2 2 4
Non-Indians 11 2 13
Total 684 684 1,368
Note: The figures in this table are not representative of relative population sizes because
this sample was mainly gathered in the Papuri region.
all of the sibs in a given language group. Sixth, the boundaries of the multilingual
Central Northwest Amazon are still not clearly defined, and many of the groups
living near the boundaries are poorly studied. Seventh, all Tukanoans are under-
going rapid cultural, social, and linguistic change at different rates depending on
their location. And, eighth and finally, as noted in Chapter 1, much work remains
to be done on the languages of the region themselves. Linguistic surveys have
been made (Sorensen, 1967; Waltz and Wheeler, 1970), providing lists of
languages with their relative distance from one another. Still, all Tukanoans are
multilingual, and establishing the correlation between language differences and
cultural differences is problematic in the Vaupes, as pointed out in the discussion
of the term tribe. In addition, some groups are known to have changed their
language (Goldman, 1963, p. 26; C. Hugh-Jones, 1979, p. 19). Thus, the
question of dialect versus language is a crucial one, with informants using names
somehow associated with speech varieties to designate social units and having
strong opinions about language distance as properly correlated with affinal
82
distance. At times, as we shall see, Tukanoan cognitive models of society and
language do not coincide with those of the linguists who have worked there, but
much more research needs to be carried out on this topic.
The social structure of the Vaupes, at its most abstract so seemingly clear-cut
and unambiguous (e.g., as shown in Figure 4), is a dynamic system incorporating
a lot of slippage, fluidity, and imprecision. We are not simply dealing with a
disparity between an idealized model and on-the-ground reality. Any Tukanoan
can, if so inclined, present a clear picture of the social categories in the Vaupes
and their articulation with one another. A clear picture of sibs and language
groups will emerge, a picture involving common ancestry, geographical location,
linguistic indicators, and exogamy; variability emerges, however, in the answers
from different individuals (or even from the same individual over a period of
time), a variability built into the basic structure of the system. It is foolish to
assume that one version is right and one wrong. As we have seen, this variability
occurs in names, in definitional criteria, and in the emphasis a group of Tukanoans
will place on a given criterion over another. For example, in the Papuri region,
language identification is by far, at least at present, the most significant feature of
membership in the language group. The most important trait the Bara say they
have in common is that they share the Bara language, in a sense "owning" it,
and it is the trait that most clearly distinguishes them from members of other
language groups. In the Pir6-paran£, although language is an extremely impor-
tant criterion, Tukanoans are aware and more tolerant of deviations from the ideal
of the language-identified unit being identical with the exogamous unit. There
the exogamous unit is more clearly distinguishable from the linguistic unit, both
empirically and in PM-parana Tukanoans' normative statements about their
social structure. Insofar as the Pira-parana is less acculturated than the Papuri,
it is probably a better representation of an earlier situation existing throughout the
Vaupes, although this is extremely hard to document. It is my opinion that the
present system of linguistic exogamy in the Vaupes is unstable, the clear-cut
association between language and exogamy so strongly expressed in the Papuri
region being in part due to the loss resulting from acculturation of certain
previously far more important nonlinguistic features that distinguished exoga-
mous groups.7
Thus, we must keep in mind that whereas some of the confusion in any
discussion of nonlocalized Tukanoan social units will be clarified by further
research, a rigid association between actual individuals or groups and the more
inclusive social categories that every Tukanoan will accept will never be discov-
ered, because the Vaup6s is a dynamic system whose fluid social structure
reflects structural and regional variability and past as well as present social
processes. Any description from the point of view of more than one local group
or more than one point in time will have built-in contradictions.
All observers of Tukanoan life have pointed out the association, in all groups
except the Cubeo, between exogamy and language. Exogamy is an important
83
The fish people
distinguishing feature separating the language groups and very probably a prime
factor in bringing the present system into being (this is further discussed later).8
The sample I gathered in 1968-70 of more than 1,000 marriages (predominantly
from the Papurf-Tiquie region), showed that with one exception all Tukanoans
had observed the rule of language exogamy. Furthermore, this "exception" was
never admitted as being a real marriage by any Tukanoan I discussed it with (this
case is discussed in Chapter 7).
Any general discussion of exogamy carried out with Tukanoans invariably
involves the topic of language. A Bara is not likely to say, "We are Bara
because we do not marry one another" but to say, "We are Bara because we
speak the same language; people who speak the same language do not marry one
another."9
Table 2 shows the most common designation of Vaupes language groups used
by Spanish or Portuguese speakers (most of these terms are in Lingua Geral, the
Tupian trade language used earlier in the region), followed by their Bara name,
and then by an English gloss, when known, of the Bara term. It should be noted
that the Bara terms are only those I collected in connection with recording one or
more marriages and specified by informants as a language group. Other terms I
collected that are used for language groups included Kaiyi'ara, Mipia (Arawakan
speakers to the north of the Vaupes [S. Hugh-Jones, personal communication]),
Barea ("food-people," the Bare of Briizzi Alves da Silva, 1966, p. 89),
Kawiria, Wuhan£ (generally referring to "Makuna" [S. Hugh-Jones, unpub-
lished field notes]), Yawi piria, and Guayavero (Guayabero of San Jose" de
Guaviare).10
It should be stressed again that this list was obtained from Tukanoans from the
Papuri-Tiquie drainage. There are obviously some questions regarding the
status of groups on the boundaries of the multilingual area and whether they
ought to be included in the linguistic exogamy system. It is also clear that Papuri
Tukanoans are not aware of some of the complications in the Pira-parana.'' The
basic tendency to associate language with exogamy is clearly found in the
Pir£-parana\ Tukanoans of that region are reported as criticizing Makuna who
marry other Makuna speakers because "people should not speak like their
cross-cousins" (C. Hugh-Jones, n.d.; see also 1979, p. 17). There is also an
emphasis on language as an important marker of a social group even among the
Cubeo, who do not have an ideal of language exogamy (although many Cubeo do
marry outside of the Cubeo unit): Goldman (personal communication) reports
that a possible translation of the Cubeo self-name Pamiwa is ' 'people of the
language." The Cubeo are clearly a part of the Central Northwest Amazon
culture area in the majority of respects. In addition, to the south of the Pira-
parand, in the Miriti-Parana region, the traditionally intermarrying Yukuna,
Tanimuca, Matapi, and Letuama groups have a rule of linguistic exogamy,
although this is no longer the case with the Matapi, because owing to a measles
84
Table 2. Names of language groups
"Spanish" (usually
Lingua Geral) Bara English gloss of Bara term
Bara Bara A sweet-smelling plant placed in a

"Northern Barasano"" man's G-string over the buttocks
during festivals. Sometimes said to
be an aphrodisiac. (Goldman trans-
lates as "medicine")
TUyuka Dohka puara Mud people or clay people
Tukano Dahea Toucan bird
Desana Wina Wind
Carapana Mutea Mosquito (Goldman translates as
"gnat"); Briizzi Alves da Silva, 1966,
p. 99: genera Stegoniya, Culex
Tatuyo Pamoi, Hiina Armadillo
Siriano Hiitia Possibly clothing or masks (Siriano
purportedly use masks in mourning
rites [S. Hugh-Jones, personal com-
munication])
Yurutf Waiyiara Fish people
Piratapuya Wai mahkara Fish people
Uanano Ohkoti mahkara Water people, medicine people
Cubeo Kobeua Bard word similar to Tiikano word
kebewa meaning "the people who
are not" (Goldman, 1963, p. 25)
Barasana Panefa* Unknown
"Southern Barasano" 0
Taiwano Eduria, Edulia* Unknown
Makuna Aiiira* Bara gloss: "the people who say
'yes' in this manner"
Wuhana* Unknown
Id6 maha Water people
Tariano PavarS Unknown
Curripaco Behkara Unknown
Pisa-Tapuyo (?)c WahiinS Unknown
Metuno Metuno Unknown
Arapaso cd Konea Woodpecker
""Northern Barasano" is the designation of language given by Summer Institute of

Linguistics (see Waltz and Wheeler, 1970, which also gives Bard). "Southern Barasano"
is the SIL term for Barasana.
*Pira~parana grouping in which language and exogamous unit are not coterminous (see
Arhem, 1981).
Information from Goldman, 1948, pp. 766-67.
Înformation from Briizzi Alves da Silva, 1966.
85
The fish people
epidemic they have given up their language and now speak Yukuna (Jacopin,
1975, p. 41).
The language groups vary greatly in size. The larger ones (e.g., Tukano,
Desana) have more than thirty sibs, while others (e.g., Bara, Yuruti) have as few
as six. Such unevenness in size of structurally equivalent groups is typical of
small-scale societies with no clear mechanism for adjusting group size with
demographic fluctuations. Sibs and entire language groups have become extinct,
owing to demographic imbalances, migration, and acculturation. Some entire
sections on the periphery of the Colombian Vaup6s (and probably to a greater
extent in Brazilian territory) have experienced so much contact with non-Indians
that their indigenous inhabitants have been absorbed into it. This has happened in
the Apaporis to the south and to the west in the upper Vaup6s River area. Table 3
is a list of population estimates for Vaupes ethnolinguistic groups. I have given
all the estimates to show just how approximate they are: The table is useful only
as an indication of relative numbers.
Although in myths language groups are spoken of as a single group living in
one longhouse, it is doubtful that this was ever actually the case, even with
longhouse units far larger than they are now. Local groups will behave as if they
are entire language groups during ceremonial occasions, for they are in fact
representing the larger category, but there is no expectation that the language
group should be seen as a corporate entity that should periodically convene for
group activities. Bara informants were comfortable with the possibility of there
being Bara to the south whom they did not know at all.
The phratry
I am defining the phratry (see Figure 4) as an unnamed unit composed of various
language groups whose members are not supposed to intermarry, because an
agnatic relationship exists among them.12 The phratry thus is the most inclusive
exogamous unit recognized by Tukanoans. Qualification as a phratry includes a
consciously stated principle that a given pair or group of language groups are not
supposed to intermarry; an observation on our part that they do not in fact do so
does not suffice. Certain combinations of language groups probably do not
intermarry for reasons having nothing to do with social structure. For example,
sometimes a given pair of language groups has its constituent settlements so far
from each other that intermarriage is unlikely. Predictably, reasons are not always
so easily ascertained in the field, for certain language groups will state that some
language groups far away are their "younger brothers" and for this reason they do
not intermarry. This is the case for the Bara and Carihona - from the Bara point
of view; I do not know if any of the surviving Carihona in the upper Vaupes
region have even heard of the Bard. In this case, phratric membership is
probably epiphenomenal - that is, this pair does not intermarry because of
distance factors alone, but people observing that these groups do not intermarry
86
Table 3. Estimates of populations of language groups
Language group H S1L T DAI R S
Bani (Nor-
thern Barasano) 300 300 500 — — —
Barasana (South-
ern Barasano) 300 — 500 — — —
Carapana (Mii-
tea) 200 150 250-300 500 700 —
Carijona (incl.
Hianacoto-
Umaua) 100 — 100 60 50 —
Cubeo (Cobewa) 2,000 2,000 750-2,000 1,000 1,000 —
Curripaco (incl.
Yavarete-
Tapuya) — — 1,000 400 2,500 —
Desana 1,000 1,500 750 500 — —
Guanano
(Uanano) 600 800 400 800 1,000 —
Maku — — — — 1,000 —
Ubde (Tukano
Maku) — 150 175 — — 200-50
Cacua (Kak-
wa, Cubeo
Maku) — 100 250 — — —
Desana Maku — — 200 — — —
Makuna (incl.
Yauna) 500 400 — — — —
Piratapuya
(Uaicama) 500 500 — 600 300 —
Siriano 200 — — 200 — —
Taiwano 100 — -— 150 — —
Tariano 50 — — 30 50 —
Tatuyo (Siina) 300 300 250 600 — —
Tiikano 2,000 2,000 1,500 1,250 1,500 —
Tuyuka 200 200 200 500 — —
Yuruti (Yuruti-
Tapuyo) 200 — 150 675 — —
Note.H = Stephen Hugh-Jones; SIL = Summer Institute of Linguistics (figures compiled
in 1970); T = Loren Turnage, "Evangelical Work among the Indians of Colombia"
(Bogota), 1969; DAI = government of Colombia, Native Affairs Department (figures
compiled 1962-68); R = Reichel-Dolmatoff, 1967; S = Peter Silverwood-Cope.
Source: Dostal, 1972, pp. 393-96.
87
The fish people
group them together in the same set of siblings in an after-the-fact manner. Still,
other nonintermarrying pairs quite distant from each other are not spoken of as
"younger brothers," whereas pairs of language groups who might easily inter-
marry in terms of distance do not because they are the "same people," or
"brothers." Thus, this highest level of organization must be explained by taking
both empirical (i.e., records of actual marriages) and normative data into account.
Two points must be clarified at this juncture. It is fruitless to try to construct
exclusive groupings based on phratric principles derived either from known
marriage practices or statements by informants. As we shall see, data on actual
marriages do show striking patterns, but these do not sort into the categorical
groups we might call phratries. Phratric organizing can be done for pairs, and in
some cases for larger units, especially in the Pira-pirana. Very soon, however,
other considerations seem to enter the picture, and both the marriage statistics
and normative evidence contradict themselves. This is also the conclusion of C.
Hugh-Jones, who states that "this is a weak and variable form of organization"
that we cannot expect to become stabilized into recognizable groupings (1979, p.
22).
The second point concerns the situation in the Pira-parana\ where people one
should not marry can be classified in terms of links based on an agnatic sibling
link, original uterine sibling link (i.e., established in mythical time), or de facto
uterine sibling link. Although I obtained many statements about how individuals
may not marry people they call by the uterine sibling term, no one ever discussed
this relationship in terms of an entire language group. Hence, respecting the data
I acquired, I have chosen the definition of phratry offered at the beginning of this
section.
Some of the difficulties in understanding phratries should be clear by now. A
clear-cut congruity does not exist throughout the Vaupes between language
affiliation, exogamy, common ancestors, and use of agnatic terminology.
Although the language group is often stated to be the maximal exogamic unit,
this is by no means always the case, and the elements of Vaupes social structure
that play roles in the organization of sibs and language groups - patrilineal
descent, language, territory, and exogamy - also play a role, variable at times, in
phratric organization. For example, use of agnatic terminology characterizes
some pairs of language groups, such as Tukano and Bara, but not others, who
nonetheless are said to stand in an agnatic sibling relationship to each other.
It is not clear, historically or functionally, why such a relationship is said to
exist in many instances, in particular, why some nonintermarrying pairs are said
to be members in the same phratry (e.g., Bara-Carihona) while other pairs are
not (e.g., Bara-Siriano). It is evidence for the pan-Vaupes organization concep-
tualized by Tukanoans. At times Tukanoans take pains to present the regional
integration of the system in as neat and broadly based terms as possible, and this
may account for the tendency to group pairs of language groups from far-flung
areas into phratries. Such pairs are also spoken of as "younger-brother" military
allies. Several war stories collected from Bara men describe the Wahiina as the
younger brothers of the Bara even though the Wahiina are located on the
Vaup6s River. The Tukano are said to have allies from the Behkara (Curripaco)
and Karihona (Carihona - mentioned previously as younger brothers of the
Bara), similarly distant groups.
Other pairs of language groups that do not intermarry are said to stand in a
"mother's children" (uterine sibling) relationship to each other. This category,
which includes all matrilateral parallel cousins who are not agnates, is obviously
not equivalent in meaning to an agnatic sibling relationship. (In Bara, "mother's
children" are called pahko-mahkara or pahko-pond.) People who address each
other by these terms also cannot marry. Although Bara informants never explic-
itly named a whole language group as standing in a mother's children relation-
ship to another, this is the case in a number of myths. Furthermore, such usage is
reported for the Pird-parand region.13
C. Hugh-Jones (1979, pp. 36-37) describes a complex, tripartite division of
intermarrying groups that are mythologically described in terms of people who
are associated with either the earth, the sky, or the water, and either the jaguar,
the eagle, or the anaconda, respectively. There is some corroboration in the
myths I collected, although these are often remnants and were mostly transcribed
in Spanish. Many Bara myths deal with "eagle," or "jaguar," or "fish"
people. It is clear that Bara belong to the category of fish people and are
associated with water symbolism. An old Bara man once described "other"
people as "different, not fish people, but umurikori maha" ("day people").
These people were "born with the sun." This classification probably applies to
the Desana, who are also reported to be day people (Reichel-Dolmatoff, 1971).
Desana are called Wina ("wind") in Bard and several other languages, and the
sun and moon are very important elements in their origin myths, whereas the sun
and moon, and celestial phenomena in general, are not mentioned at all in
connection with Bari origins. The Hiina (one of the Bara names for Tatuyo
language group) are also associated with the sky, birds, and the sun and moon
(Bidou, 1976). Thus, the origin myths of the region, in addition to setting out the
principles of exogamy with respect to single "one parent" language-affiliated
groups, also offer a much broader picture of the larger universe with respect to
the various equivalent but not identical descent groups inhabiting this level of the
universe, revealing differentiation and marriage prescription as well as prohibi-
tion.14 The theme of three (only three are considered at a time) exogamous
groups that are clearly differentiated from one another in a number of symbolic
ways and are either dependent on or in competition with each other vis a vis the
exchange of women is encountered in many areas of Tukanoan culture.15 The
symbolism of differentiation extends into other areas of Vaupe"s ideation and
cosmology. Table 4 shows how these three groups are conceived of at three levels
of inclusion.
Despite the closeness of settlements to each other, several pairs of language
89
The fish people
Table 4. Zero generation terminology at three levels of inclusion
Longhouse to longhouse;
local descent group to Language group
Ego to alter local descent group to language group
First group "my siblings," "our brothers," "our brothers,"

(e.g., Inambu agnates "our real brothers," "brother people,"
Bara") agnates "We speak one lan-
guage," agnates
Second group "my cross-cousins," "our cross-cousins," "our cross-cous-
(e.g., Inambu "my mother's people," "our affines,""our ins," "our meh-
Tuyuka) father's sisters' children brothers-in-law," pairs ko-mahkara,"
of longhouses that have "father's sister
exchanged women people," "where
we exchange wom-
en," potential af-
fines
Third group "my mother's children,' ' "longhouses far away," "our mothers are
(e.g., Pirata- uterine (half-) sibling, "distant kinsmen," sisters to each
puya of waio- matrilateral parallel "our mothers are sisters; other," "mother's
peri) cousin our fathers are not children from the
[Piracuara] brothers'' waking-up times,""
' 'pahko-mahkara:
["mother's chil-
dren people'"]"
affines of affines
Note: " " around statements indicate how Tukanoans talk about relationships.
"Information from C. Hugh-Jones.
groups do not intermarry; this is the case for the Bari-Tukano pair, and the
Desana-Tuyuka pair have a much lower incidence than would be expected in
terms of numbers of potential spouses and proximity.16 The reasons given were
that these pairs are brothers to each other and should not marry their sisters.
Sometimes the explanation involves the notion that we used to call each other
"brother," and although we do not do this now, we are still too close to marry.
Table 5 gives lists of phratry members obtained from three informants.
Living with Bara facilitates the recognition of these unnamed phratries, for
not only do Bara and Tukano not intermarry, they also use agnatic terminology
between themselves.17 I do not know whether this is still the case for some of the
other pairs. The individuals I talked to did agree that previously all members of a
given phratry addressed one another as agnates and spoke the same language. In
the section of the Bard origin myth describing Bara and Tukano as "one
people," they are "of our [i.e., Bara] language" (maniya wadera). No
Bard-Tukano marriages occur in a sample of 190 Bara and 315 Tiikano
90
Table 5. Phratries listed by three informants
Language groups included in phratry

Language
group of
informant Phratry Spanish Bara
Bara 1. mani baiira Bara Wai maha ("fish people")

("our brothers")
Kawiria"
Pisd-Tapuyo (?) Wahuna
Yuruti Waiyiara
Kumara (a Tukano sib)
Tukano Dahea
2. mani teniia Barasana Panera"
("our brothers- Taiwano Edulia"
in-law," "our Makuna Yeba maha
affines")
Tariano Pavara
Piratapuyo Wai mahkara
Tuyuka Dohka puara
Desana Wina
Bohtea pona (a Desana sib)
Tatuyo Hiina
Carapana Miitea
3. ahpera mahkara
("other people") Siriano Hutid
Tuyuka 1. (no name given) Tuyuka Dohka puara
Yuruti Waiyiara
Desana Wina
Koroba ("they now speak
Cubeo")
Behkara (Arawak speakers to
the north)
2. (no name given) Bara Bara
Tukano Dahea
Barasana" Panera
Pisa-Tapuyo? Wahuna
Tukano 1. hika pona Tukano Dahea
("children of Bara Bara
one parent") Wagiiyara
Yuruti Waiyiara
Carihona Karihona
Metuno
Behkara (Arawak speakers to
the north)
91
The fish people
marriages. Bard-Tukano exogamy is corroborated by Sorensen: "no Tukano will
marry a Barasana [Bara], or a Bare, or a Baniva" (1970, p. x). Although I have
no marriages recorded for either Bare or Baniva, I do have instances in which
Behkard, an Arawak-speaking group, are spoken of as the younger-brother allies
of the Tukano.
Some informants will list language groups as being in a "brother group,"
"brother-in-law group," and "other-people group." When questioned more
closely about the way in which these lists related to the rule of exogamy and
language, Bara handled the contradiction thus: "My brothers are those who
speak my own language. I call Tukanos 'brothers' because we used to speak the
same language. They started to speak differently, and now they speak another
language entirely. But we are still close, and I still call them 'brothers.' " l 8
The younger Tukanoans I spoke with did not seem to have a clear concept of
phratric groupings as indicated by the lists given by older informants, but when
asked specifically about pairs of language groups, such as Tukano-Bard or
Desana-Tuyuka, they confirmed that such pairs do not and did not intermarry.
They did not accept the suggestion that lack of intermarrying between these pairs
was accidental or due to distance factors alone.
An obligatory elder-younger distinction is an important feature of all Tukanoan
sibling terminologies. This of course means that in a given phratry, one language
group will be of lower rank than another. Bara address Tukanoans as their "elder
siblings." Still, Bara say this is not the way it should be, that long ago the
ancestors of both groups spoke Bard but that at one point the lower-ranking sibs
of this original group began to speak differently. These same sibs also refused to
use the elder sibling terms to the others. Finally the people in the higher-ranking
sibs gave in and, continuing to speak Bara, consented to call the others by the
elder sibling terms. According to the Bard 1 spoke to, every Bard and Tukano
today knows that this is so, but only one Tukano sib, the Boho (Dasyprocta sp.)
people, near Paricachivera Mission on the Tiquie", addresses the Bard as elder
siblings.
Data collected on actual marriages demonstrate that to some degree these
phratric principles operate in marriage making. A quick scanning of Table 6,
based on a sample of 534 marriages occurring among eight language groups,
illustrates this kind of patterning. Because it is concerned only with eight
language groups and marrriage partners predominantly from the Papuri area, the
table does not represent the entire Vaupes. Moreover, insofar as these eight
language groups have settlements elsewhere in the Vaup6s, the picture is not a
Table 5. (cont.)
Note: These are lists from single informants and are not to be seen as the only cor-
rect list of phratries; glosses of names in Bard are approximate in certain cases.
"Cf. C. Hugh-Jones, 1979, Appendix 1. Rapuri River informants simplify Pira'-parana'
groupings.
92
complete account of marriages made among these particular eight language

groups during the last forty years for the Papurf-upper Tiquie drainage.
Figure 5, based on the data in Table 6, shows the pattern of marriage making in
terms of Euclidian space through the use of three-dimensional scaling tech-
niques.19 The resulting illustration demonstrates quite clearly that the eight
groups form four pairs, each of which has a high frequency of intermarriage. The
next highest frequency of intermarriage occurs between groups that are adjacent
in adjoining pairs.
In sum, it is clear that norms concerning marital preference with respect to
language group membership can be elicited from informants. In addition, pat-
terns of preference between certain pairs of language groups can be observed
from an examination of actual marriages. Still, much more research and analysis
must be done before we have a clear picture of the entire phratry system. In
particular, two directions of research suggest themselves. The first is to continue
to find out why certain language groups have no or very little intermarriage. It is
clear that Tukanoans, or at least the more traditionally oriented ones, have a
conceptualization of exogamous phratric groupings of "brother people" and
"mother's children people" that are more inclusive than any given language
group. More research on Tukanoans throughout the region and their conceptual-
izations of marriage preferences is obviously indicated. For example, both
Sorensen (1967) and Fulop (1955) report a system of five named phratries
obtained from conversations with Tukanoan informants. My informants did not
talk about a set of five exogamous units, and recent marriage patterns themselves
do not support the existence of such a tight organization. Still, various bits of
tantalizing evidence continue to emerge (for example, the number five appears in
Pird-parana Tukanoans' system of ranked sibs [C. Hugh-Jones, 1977]).
The second direction of indicated research is precisely in the area of territory
and propinquity, with the goal of a better understanding of the degree to which
these considerations do affect ultimate marriage choice. The matter is complicat-
ed; given the dispersed nature of the settlements of many language groups and the
great variation in size, one must take as the unit of study sections of language
groups in circumscribed territories. Doing so will make the measures of propin-
quity far more meaningful and will make far more noticeable the lack of
intermarriage where propinquity would predict its presence. For example, each
language group in the following list of pairs has some settlements quite close to
settlements belonging to the language group in the pair; yet marriages are either
entirely lacking or quite infrequent: Bara-Tukano, Desana-Tuyuka, Desana-
Tatuyo, Tatuyo-Siriano, and Tuyuka-Carapana (see Figure 5). Is this in reality a
reflection of proximity but proximity of an earlier era, continued through mar-
riage alliance maintenance, before the effects of acculturation brought about the
greater territorial dispersion found at present, at least in the Papuri region? The
Tiquie and the Pird-parand, areas of least acculturation, show a greater associa-
tion between territory and specific language group. It is also true that the entire
93
Table 6. Marriage between selected language groups
Husband's Wife's language group

language
group Bara Tuyuka Tukano Desana Carapana Tatuyo Siriano Yurutf Total
Bara
N 0 55 0 7 2 12 1 0 77
% 0.0 71.4 0.0 9.1 2.6 15.6 1.3 0.0 100.0 (14.4)
Tuyuka
N 58 0 71 3 0 6 7 0 145
% 40.0 0.0 49.0 2.1 0.0 4.1 4.8 0.0 100.0 (27.2)
Tukano
N 0 47 1 45 4 5 20 5 127
% 0.0 37.0 0.8 35.4 3.1 3.9 15.7 3.9 100.0 (27.2)
Desana
N 5 2 36 0 2 0 10 0 55
% 9.1 3.6 65.5 0.0 3.6 0.0 18.2 0.0 100.0 (10.3)
Carapana
N 3 0 6 4 0 10 2 2 27
% 11.1 0.0 22.2 14.8 0.0 37.0 7.4 7.4 100.0 (5.1)
Tatuyo
N 10 0 5 1 2 0 0 2 20
% 50.0 0.0 25.0 5.0 10.0 0.0 0.0 10.0 100.0 (3.7)
Siriano
N 2 8 27 14 5 1 0 14 71
% 2.8 11.3 38.0 19.7 7.0 1.4 0.0 19.7 100.0 (13.3)
Yurti
N 2 0 2 0 1 1 6 0 12
% 16.7 0.0 16.7 0.0 8.3 8.3 50.0 0.0 100.0 (2.2)
Total
N 80 112 148 74 16 35 46 23 534
% 15.0 21.0 27.7 13.9 3.0 6.6 8.6 4.3 100.0 (100.0)
Tatuyo • Carapana
Bara||1Uyuka
Figure 5. A three-dimensional spatial representation of marriage distance between

language groups. Data from Table 6; after Jackson and Romney 1973.
language group is obviously too large a unit of study for some purposes.
Although Tukanoans speak of, for example, whom Bara marry, this is most
often spoken of relative to a particular Bara settlement. Thus, Bara of one river
tend to marry Barasana and Tatuyo, whereas those on another tend to marry
Tuyuka and Desana.
Another difficulty with normative statements made by Tukanoans is that the
description being made of the "best" type of marriage may in fact operate only
under certain conditions (this is discussed further in Chapter 7). For example, a
Tukanoan might very well state that people always try to make marriages between
bilateral cross-cousins, first cousins if possible, who live in neighboring long-
houses. Yet this probably does not apply to all the marriages made by a given
generation in a settlement; it may be that once this type of marriage has been
made in two or three cases, other individuals are allowed or even encouraged to
marry into other settlements, because the new generation has reestablished its
ongoing alliances and should seek to set up new ones. This would tend to spread
out the available sources of aid and exchange relationships made by affinal ties
rather than limiting them to one settlement. Such an arrangement is explicitly
stated in the marriage rules among the Yukuna groups (Jacopin, 1975) and the
Witoto (Gasche, 1977). The rule states that no man can marry into the same
settlement in which his (classificatory) brother obtained a wife, a strategy clearly
intended to disperse marriage ties.
Space does not permit a thorough discussion of the literature with respect to
phratric organization in the Vaupes. It is scanty and confusing, particularly with
respect to earlier patterns and preferences. Sorensen, for example (1967, pp.
95
The fish people
671-72), states that each language-affiliated tribe consists of several sibs and is
aligned with one of five phratries. According to him, phratries are named and
exogamous, and all marriages conform to a rule of phratry exogamy. He does not
list the names of the phratries, however, nor does he give data on actual
marriages.
Reichel-Dolmatoff's (1971, p. 4) use of "phratry" is coterminous with my use
of language group. (In his 1975 book he refers to exogamous groups [1975, p.
66].) He does mention that the Piratapuya are considered practically a "second
Desana phratry" (1971, p. 7). Waltz and Wheeler, reporting on work done by
SIL linguists in the Vaup6s, state: "The Barasano . . . intermarry mainly with
Taiwanos whose language is almost identical with their own . . . the Guanano are
forbidden to marry the Piratapuyo 'because they are brothers'" (1970, pp.
120-25). Fulop (1955) states that the tribe Yepa" Majsa consists of five phra-
tries, each of which is aligned with a distinct language: Tukano, Carapana,
Barasana,20 Maniva, and Cubeo. Brtizzi Alves da Silva (1966, pp. 81-123) also
lists prescribed and proscribed marriages for his list of twenty-six "tribes."
The Cubeo differ from other Vaupes language-affiliated groups in being able
to marry within the Cubeo unit which contains three exogamous phratries
(Goldman, 1963, pp. 100-05). Still, the underlying pattern is identical: Several
ranked, named sibs make up an exogamous unit intermarrying with equivalent
units. Goldman (personal communication) has noted that a Desana once remarked
to him that at one time Cubeo and Desana did not intermarry because they were
too close. Finally, as has been noted, Appendix I of C. Hugh-Jones, 1979, gives
a comprehensive treatment of the Pira-parana situation.
Regional integration and interaction between settlements

We have seen some of the ways in which interaction between settlements is vital
to Tukanoan life and has brought about a regional orientation in Tukanoans'
conceptualization of their social universe. Although much of daily life revolves
around intimate relationships within the local group, marriage, visiting, and
ceremonies are also essential, for they help in constituting the local group and
expanding relationships beyond it. The amount of systemic social and cultural
integration beyond the local group is complex and varies in a nontrivial way
within the region. This is one reason why the term tribe is not powerful enough,
for it oversimplifies and obscures the nature of intersettlement relations.
Thus, in some respects, settlements in the Vaup6s are not nearly as autono-
mous as in some other lowland South American societies, despite their very real
isolation and autonomy in many other areas of life.21 For one thing, local groups
in the Vaup6s are exogamous. Marital interaction, furthermore, is not limited to
exchanges between two or three neighboring settlements. Many types of ceremo-
nial occasions require the participation of more than one local group (with the
exception of some mission towns, where traditional ceremonial life is very
96
truncated anyway), and the majority of ceremonies includes people from more
than one language group. In the past, shamanism also involved interaction of
settlements.
In earlier periods an important aspect of such interaction concerned feuding,
raiding, and the maintenance of military alliances. Although warfare22 has
ceased, its legacy is still apparent in settlement pattern, choice of settlement site,
longhouse architecture, and almost certainly marriage patterns. Raiding and the
accompanying military considerations such as defense and inculcation of bravery
and other necessary qualities in male children are important elements in
present-day ritual, myth, and various communicative events such as greeting
behavior. C. Hugh-Jones (1979, pp. 63-64) offers a nice discussion of the
interrelationships of warfare, communal rituals, shamanism, and marriage in the
past, pulling out common themes of competition and a gradation of hostilities.
What little remains of the earlier Vaupes military complex resembles the
Yanomamd of Venezuela and Brazil (Chagnon, 1968).
The dependence of a settlement on other settlements for marriage partners, for
ceremonial and military allies, and for economic exchanges has undoubtedly
influenced Tukanoan attitudes about the Vaupes as a whole. As already noted,
the extensive knowledge many Tukanoans have of the entire region is impressive.
It is evident that Tukanoans feel themselves to be much more a part of a
pan-Vaup6s system than do the peoples of the numerous lowland South Ameri-
can societies, which are characterized by a kind of xenophobia, even toward
people living quite nearby. The Tukanoan conceptualization involves an ever-
increasing geographical area even beyond the Vaupes region, with no generally
recognized natural or artificial boundaries beyond which live people who are
categorically enemies or strangers. This is not to say that "enemies" or
"other-than-us" categories do not exist but that the social and cultural differenti-
ation is gradual rather than abrupt.
Differentiation exists, but although the criteria specifying it are explicit they
are not based on territorial considerations so much as social ones. The Cubeo on
the northern boundary of the region are at times excluded in discussions of "real
people," but most of the time the implication is that they are excluded because of
their behavior rather than because of innate differences correlated with geograph-
ical location. Other times, Cubeo are very much included in categories corres-
ponding to "us." Goldman (personal communication) states that the Cubeo,
although conceiving of themselves as a tribal entity, also see themselves as part
of a larger entity.
Tukanoans visit other settlements for trade, ceremonies, or courtship. Because
the nearest longhouses are separated by anywhere from two hours' to a day's
canoe travel, a trip involves much preparation and lasts at least a few days. The
Northwest Amazon is famous for its rules of hospitality to overnight guests: One
is almost never refused lodging - a place to hang a hammock, a warm fire, and
some food - and when possible Tukanoans almost always elect to spend the night
97
The fish people
in a house rather than make camp in the forest. Furthermore, as indicated in

Chapter 4, a visiting family can participate in feeding itself; the man will hunt
and fish and the woman process manioc from her hosts' gardens.
Visiting has probably increased in recent times as a result both of more travel
to and from mission towns and the end of raiding. The frequency and importance
of visiting was forcefully demonstrated to me when I undertook the task of
censusing households. Although the patrilocal residence rule states unequivo-
cally that one should live with one's closest agnates, interpreting this in terms of
locating people in "their" settlement sometimes involved what were rather
arbitrary decisions. Tukanoans travel a great deal and can stay in other settle-
ments for prolonged periods of time, up to several months. Quarrels can result in
the permanent dissolution of a residence group, some members of which will go
to "visit" other kinsmen but in fact never return. Thus, at any point in time a
significant proportion of Tukanoans are not living where they are "supposed" to.
Such mobility and fluidity are probably an integral aspect of a system like that of
the Vaupes, a point made by Goldman (1963), among others.
A guest's status as a visitor is clearly marked in a number of ways. Greeting
and departure rituals mark both the formality of the occasion and the degree of
intimacy between hosts and guest. At important ceremonies, greeting rituals are
governed by strict protocol and are quite long. Significantly, Maku are never
formally greeted. Further evidence of this distinction between guest and host is
that members of a longhouse, even though returning from extensive trips, are not
formally greeted in this way, because one should not have to go through a formal
ritual for close family. Formal behavior is basically a recognition of social
distance and/or seriousness of the occasion, neither of which should apply to a
fellow-resident's return.
During my stay at Pumanaka buro, the headman's son, Lino, returned with
me from Mini, where he had been for seven years. During this time, only those
who had traveled to Mitu (three young men) had seen him. No one approached
when we disembarked, despite the sound of the outboard motor. Not only was
Lino not formally greeted, he was scarcely greeted at all, so unusual (and tense)
was the situation. Everyone tried to appear casual and unexcited; when his
mother was overcome by emotion (including, I am sure, anger), she ran to the
back field to finish crying.
This example illustrates that greeting forms are based on implications of
intimacy contained in social roles rather than the actual amount of intimacy. Lino
was a virtual stranger in appearance to everyone and a total stranger to many of
the younger children. Yet everyone tried to appear as though he had returned from
a few hours' fishing trip, for a member of a settlement does not go off by himself
- ideally - for longer periods of time.
As previously indicated, a settlement will normally have the most interaction
and the closest affective ties with its neighbors, particularly those on the same
stretch of river, regardless of their positions in Vaupe's social structure. For
98
example, Pumanaka buro has much more contact with the three downstream
Tuyuka settlements than with the other Bara settlements in the Macucu-Tiquie
region. Goldman (1963) is quite correct in emphasizing the importance of rivers
in Tukanoan social organization. They provide natural highways for communica-
tion and transportation, and as has already been indicated, are extremely impor-
tant in Tukanoan conceptualizations of the cosmos.23
Another kind of intersettlement dependence, although at present quite attenu-
ated, is the result of a regionally based division of labor in the manufacture of
various artifacts of a ceremonial nature. Trade and exchange of these occur in
connection with ceremonies and other visits. Some of this division of labor is
necessary because of the uneven distribution of certain raw materials (e.g., clay).
Other kinds of items are exchanged because of what appears to have been a more
artificially created specialization of manufacturing; thus, Tukanos traditionally
made a carefully carved stool (cf. A. R. Wallace, 1889/1972, p. 342, who notes
stool making by the Tariana as well), Tatuyos made a snuff taker out of the bones
of a species of bird, and so forth. The Bard are associated with canoe making as
are the Tuyuka, although their name ("mud people") implies that they might
have been the first cazabe griddle-making specialists (Briizzi Alves da Silva,
1966, p. 115). The Desana specialize in making basketsieves, and the Baniva on
the Ic,ana make graters (Briizzi Alves da Silva, 1966, p. 77). As mentioned
earlier, at present this is an association between language group and object rather
than an exclusive right to manufacture.
This kind of specialization is suggestive of a similar artificially created inter-
dependence described by Chagnon for the Yanomamo (1968, p. 100), although
it is much more elaborate and important among them. Trade between Tukanoans
of Maku-manufactured items demonstrates a similar interdependence; only
Makii make certain objects (including all manner of highly valued twined
baskets), and only some Tukanoans have direct access to Maku groups (this is
discussed further in Chapter 8).
It is evident that culturally homogeneous regional systems occur in many areas
of lowland South America, involving networks of trade and other forms of
exchange such as marriage, feasting, and in the past, military alliances.24 Still,
the Vaup6s appears to be culturally integrated into a regionally based network
system to a very marked degree. The question of why this is so in this particular
area of the Amazon-Orinoco Basin cannot be completely answered, at least at
present, but some reasonable hypotheses can be offered.
Population density is undoubtedly a crucial factor in the evolution of the
Vaup6s system to its present form. It is almost certain that the area has had, in
the past, a higher density, more territorially based control over scarce resources,
and more stratification among the groups exploiting those resources. Goldman's
(1963) description of the Cubeo in 1939-40 certainly indicates this, as do
writings from earlier travelers. At that time and earlier, groups with higher rank
and more allies - and greater population - occupied the larger rivers, particularly
99
The fish people
sites on river mouths. Such locations had military advantages and conferred
higher status. Control over groups such as Maku and lower-ranking sibs seems to
have been a major consideration. The present unimportance of this control
strategy is probably due to decline in population and to the disappearance of
active raiding and feuding.
The higher population density reported by Goldman for the Cubeo in 1939
probably also lies behind their greater use of the landscape to reflect social
arrangements. Cubeo phratric organization and sib rank are expressed quite
clearly by the location of settlements. Whether the present situation among other
Tukanoans represents a simple breakdown in the system, after falling below a
critical minimum population density, or a more creative process of adaptation to
the changing demographic situation is an open question. Some of the possible
factors producing decline in population density have been mentioned earlier.
It seems evident that changing population density and migration were also
important factors in the evolution of the present Tukanoan marriage system. Two
kinds of explanation have been offered (Sorensen, 1967) to account for the way in
which demographic and spatial factors might give rise to a system involving
linguistic exogamy. The first, a fusion model, suggests that a cul-de-sac situation
arose owing to pressure from missions, rubber gatherers, and other agents of the
national economies of either Brazil or Colombia. The resulting squeeze of
territory necessitated more interaction of distinct cultural groups, a necessity
increased by declines in population caused by disease. Various mechanisms arose
that facilitated interaction of the previously separated or hostile groups. One of
these mechanisms was intermarriage; the heretofore truly distinct tribal-lite
groups assimilated to the point of sharing a common culture, and a rule of
exogamy came to be applied to what originally were endogamous units. Lan-
guage came to be the main marker distinguishing these exogamous units, whereas
originally it was but one of many cultural differences separating them. The
peoples of the Papuri drainage area seem to have progressed the most in this
direction, and those of other areas, especially the Pira-parana, probably repre-
sent an earlier stage, with more territorially confined language groups and more
cultural distinctions separating them. In the more acculturated Papuri, some of
the complexity of the traditional system of classification of social units has been
lost.
The second type of explanation, a fission model, postulates an original situa-
tion characterized by endogamous (again, probably much more tribal-like than at
present) units with exogamous moieties within each one. Of the various markers
distinguishing one moiety from the other, speech differences came to be the most
crucial, until ultimately what was once a single protolanguage spoken by the
entire endogamous unit divided into two languages along the lines of the moiety
division.25 The rule of marriage came to be expressed as ' 'We marry people who
speak a different language."
These two models perhaps show how different segments of the Vaup6s
100
population simultaneously evolved into the present Tukanoan marriage system.
While outsider groups were increasingly being incorporated through intermar-
riage, the idea concerning marriage with a person affiliated with a different
language was becoming the most important marriage rule. It diffused into the
moiety-based groups, increasing the stress of speech differences in these groups
as the most significant marker distinguishing each exogamous unit.
Most of the ethnographers who have written recently on the Vaupes have
commented on the degree of homogeneity found throughout the region. They
note this to be so despite no or sporadic contact between many of its inhabitants
(see Goldman, 1948, pp. 763-64; Reichel-Dolmatoff, 1971, p. 16; Sorensen,
1967, pp. 673-74). The cultural homogeneity is readily apparent even to a
nonanthropologist, regardless of how one defines "culture." The similarity of
observable phenomena throughout the region is indisputable, and similarities in
the cognitive orientation of Vaupes natives can be demonstrated as well. As
mentioned earlier, Tukanoans assume that their rules of marriage apply through-
out the region, and people who do not observe these rules - whites, Maku, and
Cubeo — are criticized and ridiculed for their inappropriate behavior. The origin
myths of the various language groups, although occurring in more than sixteen
languages, share many basic similarities. They all use the entire region as their
setting, and their plots involve representatives of many of the present-day lan-
guage groups.
Linguistic evidence also shows that the discrete languages of the Vaupe"s area
are by no means indications of similarly discrete cultures. Semantic categories in
many of the languages appear to be similar if not identical. This is not surprising,
given that Tukanoans everywhere behave very similarly and must deal with the
same landscape, physical and social, regardless of formal identification with
specific father languages.
The degree to which the differences that are found in the Vaupes - particularly
linguistic ones - are cultural differences rather than significant distinctions within
a single culture is an important question. Much of the heterogeneity that exists
should perhaps be considered an aid in the organization of interaction of the
various social units within the region in much the same way as highly visible
differences in uniforms aid in the organization and playing of a football game.
The fact that it is language that serves as the emblem of distinct units, which must
remain distinct for the system to work, has undoubtedly obscured the underlying
similarities. Few anthropologists have encountered such widespread multilin-
gualism in either their own society or the anthropological literature. Multi-
lingualism is assumed to be characteristic only of complex societies. Thus,
differences separating the language groups of the Vaupes tend to be over-
emphasized (exacerbated by calling them tribes), despite the fact that differences
in language do not, a priori, indicate deep cultural divisions. The essentially
homogeneous and regionally integrated characteristics of the Vaup6s have not,
in my opinion, been given enough consideration in the ethnographic literature,
101
The fish people
although recent work has begun to remedy this. As indicated in Chapter 1, this is
also one of the major goals of this book.
A fundamental contrast inheres in the situation I have been describing. On the
one hand we have a native cognitive model that is theoretically limitless in terms
of geography, society, and language. It is a model encompassing a vast area
within which all Tukanoans potentially or actually interact but one that permits
lack of interaction to the point of total ignorance of a given settlement's exis-
tence. On the other hand we must understand the implications of such a system
from the point of view of its actors: "Part of the conceptual difficulty in
describing Vaupe"s social structure . . . is the overlap of a comprehensive system
based on descent and exogamy with a practical local organisation into longhouse
communities, each of which perceives other communities as more-or-less distant
outsiders" (C. Hugh-Jones, 1979, p. 32).
This leads us once again to the issue of rigidity versus flexibility. Tukanoans
are united by an underlying model of their social organization that involves
several dimensions of classification. These dimensions, as we have seen, are
concerned with kinship (both present-day terminological classification and ideol-
ogy about descent group ancestry), mythology, language, geography, and so on.
We have seen that these dimensions cannot be reduced to a simple model that
incorporates both the structure and content of each dimension and still represent
Vaupes reality. Even if we are speaking of a unit of analysis much smaller than
the Vaupes region, it is impossible to construct a model of Vaupes social
structure that provides both structure and a fixed content. A good example of this
is the difference in Tatuyo origins as told by adjacent sibs; these differences are
profound and cannot be explained away by searching for the one that is "more
correct" in terms of some criterion or other, for example, the one that is more
traditional (Bidou, 1976; C. Hugh-Jones, 1979, p. 32).
A crucial feature of the model, therefore, is its ability to adopt different
content, depending on the context. Flexibility at one level permits the insertion at
another of sometimes rigid classificatory principles. Our model must allow
argument about and even subversion of surface aspects of the fixed structure, for
the Tukanoan model certainly does. In some ways we are reminded of Kurosawa's
film Rashomon or Durrell's Alexandria Quartet. In the Vaupes, as in these
examples, there is no final "correct" version of action and intent. What is real
and correct is that different points of view produce different content, and yet they
nonetheless share an underlying set of common propositions about the nature and
meaning of the world. With respect to the raison d'etre behind the Vaupes
arrangement, as I have indicated, I feel that such flexibility is adaptive. Godelier's
(1975) discussion of the way in which a kinship system allows for flexibility in
relationships between people and between people and their resources is a persua-
sive example of the sort of thing I mean. Another way of putting it, if we expand
the reference to kinship to include all areas of social structure, is: "Rather than
providing a set of rules people must obey, therefore, kinship provides an idiom by
102
Vaupis social structure
which people seek to maintain or transform their relationships to others as the
situation demands" (Bledsoe, 1980, p. 29). Thus, the structure has the capacity
to take on different meaning as we move across the Vaupes landscape - seen in
physical (geographical) or social terms (e.g., the changes in perspective from one
language group to another with respect to roles assumed in myth). As we move
vertically the same thing happens, as illustrated by the example of Tatuyo sibs'
versions of the origin myth. The issue of hierarchy versus equality is intimately
connected to the one of rigidity versus flexibility. Most of the dimensions I have
mentioned incorporate some notions of hierarchical order; that is, the distinguish-
ing features of the positions in a given dimension contrast in terms of varying
amounts of a particular quality, a quality either positively or negatively evaluated.
I have given many examples of what we might call "differentiated identity" in
the preceding pages, and many more will follow. Hierarchical ordering is the
basis for such differentiated identity in many cases, for what is at issue is the
possession of greater or lesser amounts of rank, moral superiority, ceremonial
purity, and so on. C. Hugh-Jones's point about each exogamous group standing
at the center of its own social world - and this is true for each longhouse
community as well - is an example of such differentiation, relative though it be:
"From the point of view of each Exogamous Group, the others are not equal -
they range from close groups with whom women are exchanged to very distant
groups whose members are totally unknown. Only when seen from the point of
view of comparable internal structure, or of an artificial 'Vaupe's system' seen
from the outside, are Exogamous Groups equivalent to one another" (1979, p.
276). This is a different kind of nonequivalence from, for instance, sib rank,
because it is completely relative. Dyadic relations between language groups or
other equivalent units exchanging women are based on the principle of equality,
as we would expect from a system of direct exchange marriage. Still, a kind of
rank ordering occurs here, even though it is completely dependent on the actor's
point of view. The ordering is hierarchical in the sense that groups with whom
one interacts a lot are differentiated from far-away groups with respect to a
number of value-loaded dimensions. This type of "concentric" model of social
differentiation is discussed further in Chapter 13.
Other types of relationships, such as between sibs in a given language group,
between Tukanoans and Maku, or between a headman and his coresidents, are
also asymmetrical. In addition we see here the interaction of the underlying
system of fixed statuses or ranks and the built-in flexibility in their application to
specific people or social units. For example, sib rank relates to the original
ancestors' birth order, presumably quite fixed. Still, Tukanoans are notoriously
contentious about present-day sib rank, and I have argued that this is a basic
component of the system of ranked sibs. In similar fashion, although Tukanoans
and Maku are generally conceived of as qualitatively different, sometimes
Maku are spoken of as fallen angels who can be reinstated if and when they
repent and change their behavior (this is discussed further in Chapter 8). And
103
The fish people
although a headman's firstborn son ideally inherits his father's position, in reality
succession can involve a great deal of argument, schism, and denials of such a
right. A headman must initially and continually prove himself worthy of the job.
Such hierarchy does not do much of anything in terms of influencing differen-
tial access to scarce resources or even conferring symbolic status of some form or
another that is clearly marked in everyday life. Hierarchy is most apparent in
classification schemes such as the system of ranked sibs and in ritual.
We might speculate that hierarchy is one more organizing principle to show
people where they fit into the system and that it, like many other such principles
in the Vaup6s, at times seems to be available mainly for the purpose of being
manipulated, denied, or inverted. The obligatory hierarchical marking of agnatic
sibling terms, for example, tells siblings exactly where they fit, but also affords
many opportunities for challenging the implications of the hierarchy. This is
illustrated in numerous myths involving elder and younger brothers. On the other
hand, people calling each other by the uterine half-sibling term are equivalent in
rank. With less of a clear-cut specification of relative position comes more
possibility of ambiguity. And, as we shall see, the relationship occurring between
"mother's children" is ambiguous, a characteristic also illustrated in myth.
The notion of built-in flexibility can take two forms. First, it can operate via
what we might call denial mechanisms that challenge a given, clearly specified
rigid classification, often hierarchically ordered. And, second, it can emerge in
situations that incorporate some form of ambiguity in their very definition. These
themes are explored further in the following two chapters.
104
6.
Kinship
Treating kinship as a separate topic is difficult, for it pervades every aspect of

life.1 Tukanoans spend most of their lives with their family - their hearth family
and the extended family making up the longhouse unit - and when this group is
temporarily enlarged, it almost always consists of kinsmen. To a much greater
extent than in more complex societies, in which people can occupy any number
of nonkinship positions (e.g., "working class"; "Brahman caste") and assume
any number of nonkinship roles (e.g., doctor-patient; teacher-student), the
kinship-based roles a Tukanoan plays throughout life are the primary source of
social identity.
Tukanoans address and refer to one another with kin terms most of the time.
The name given to each infant at the naming ceremony soon after birth is known
by everyone but seldom heard. Naming illustrates temporality and relationality,
two themes mentioned in Chapter 1. A Bara receives a name from a very finite
list owned by the sib (see C. Hugh-Jones, 1979; Arhem, 1980), and the name
ideally comes from a FF (father's father) or FFZ (father's father's sister) who has
recently died. The deceased's losing his or her name is another stage in the
process of dying and joining the relatively undifferentiated mass of ancestor
people; the taboo on mentioning the names of the dead accelerates this process.
As noted in Chapter 1, both the naming system (which involves a cycle of only
two generations; see S. Hugh-Jones, 1979, p. 106) and the kinship terminology
system are less concerned with distinguishing each Tukanoan as a unique indi-
vidual than their counterparts in the West. Both names and kin terms can be seen
to show that a given individual's identity is in part formed by a long line of
ancestors (some of whom are namesakes as well as kin) and that the individual as
distinct personality, occupying a unique intersection of temporal, spatial, and
relational coordinates in his or her social universe, is ephemeral. Specific kin
terms are used by any number of people standing as egos and alters, and names
are not unique designators of individual Tukanoans in the way they are in the
West: thus the kin terms and names are eternal, whereas the individual is not.
These features of relationality, nondistinctiveness, and transience are further
discussed in Chapter 13.
105
The fish people
Nicknames, although more distinct, are more temporary and, as in our own
society, purposely less serious than personal names. Spanish and Portuguese
names are given more or less spontaneously and have no necessary connection to
anyone else (except to a Catholic saint, a connection very imperfectly under-
stood). Although this is changing, in 1970 patronyms were not used systematical-
ly. A Tukanoan's use of a patronym is no guarantee that his siblings or father will
use the same one, or that it will not be changed on a whim.
In Western society, that John Jacob Jingleheimer Schmidt was at his birth, is,
and will continue to be a distinct, individual self is a reasonable and meaningful
proposition. Whether or not we believe in life after death, he will survive in
numerous records; if he has composed important music or was otherwise famous,
he will have a permanent niche in history as an individual. The idea that it is
important to survive as a unique personality is not nearly so meaningful to
Tukanoans. This is not to deny that ambiguities exist in Western conceptualiza-
tions of the ways in which individual personalities maintain themselves as distinct
and with recognizable continuity throughout life and after death. But as an issue
it is far more salient in Western philosophy and religion, and, I argue, certain
features of Tukanoan society and culture can be seen as working against and
playing down similar sorts of concerns. The nonuniqueness of Tukanoan names
and kin terms and the idea of reincarnation of a person's soul (but not personali-
ty) via name transmission is an example of such downplaying.
Thus, whereas in America names are thought of as unique to each individual
(and the tag "Jr." serves this purpose when the name does not), they are much
less so in Tukanoan society. Furthermore, in the West names are only one of a
number of unique tags identifying individuals in terms of their relationship to
positions or objects (e.g., social security numbers, driver's licenses, credit card
accounts) and, less often, to other people (e.g., marriage licenses). In the
Tukanoan system, relationality eclipses such functions performed by Western
naming and tagging systems, in part because less need for such identification
occurs in Tukanoan society (where everyone knows or knows of everyone else),
and in part because our emphasis on individualism greatly exceeds Tukanoan
concern with individualism and its analogues.
Kinship terminology
Bara kinship terminology is basically a variant of the Dravidian type.2 The
following distinctions are crucial and pervade the entire Bara system:
1 Generation. Five generation levels are distinguished, no skewing is present.
2 Sex of relative. Marked in all generations.
3 Cross versus parallel (Sex of intervening relative). A distinction made in the
0, + 1, and - 1 generations. This split can also be referred to as the agnatic
106
Kinship
versus affine distinction, or the "own group" versus "other group" distinc-
tion (Buchler and Selby [1968, p. 233]).
The Bard system is different from Iroquois in that it, like all Dravidian systems,
"consistently and explicitly distinguishes 'prohibited' women (nonmarriageable
categories) from 'lawful' women (genealogically or categorically defined affines).
Such systems have positive marriage rules in the sense that Dumont ['The
Dravidian Kinship Terminology as an Expression of Marriage,' 1953] . . . uses
the term" (Buchler and Selby, 1968, p. 233).
Another important feature mentioned in Chapter 5 is the obligatory marking of
elder and younger siblings. Patrilateral parallel cousins are called by the elder or
younger term depending on the birth order of their fathers (this is simplified in
Table 7).
Sex of speaker is not a general distinction required by the terminology but is
marked in the -1 (first descending) generation, and some of the affinal terms of
the 0 generation. The basic reason for the differentiation in the -1 generation has
to do with the fact that a female's offspring are never her own agnates, whereas a
male's offspring are. Thus, although both men and women use the terms
mahku-6 for their own children, the meaning is different with respect to
agnatic ties, and this is clarified by an inspection of other kin types (e.g., a
female ego says mahku to her own son, her sister's son, and her husband's
brother's son, whereas a male ego says it to his own son, his brothers' sons, and
in general all Bard males of the first descending generation).
Patrilateral parallel cousins are separated terminologically from matrilateral
parallel cousins. The latter set of terms glosses as "mother's sister's children,"
and "mother's children" (Table 4). The implication in the latter gloss is that such
kinsmen are not "father's children" as well. The genealogically closest kinsman
called by this term is the uterine half-sibling. Since patrilateral parallel cousins
are called by sibling terms, this set of terms designates those same-generation
individuals whose mothers are of ego's mother's group but whose fathers are not
Bara. This additional set of cousin terms (mother's children) is also found in
Tukano and Barasana (C. Hugh-Jones, 1979) and Tatuyo (Bidou, 1976, pp.
117-25, 157). My own incomplete data suggest that Tuyuka can be added to the
list. I do not know whether it exists in all Eastern Tukanoan or Arawakan
languages in the area, but the fact that it is found in Maku (Silverwood-Cope,
1972), a much more distantly related language, argues for a more widespread
distribution than the five languages mentioned above. Goldman (1963, p. 134)
gives a term for matrilateral parallel cousins, male and female, among the Cubeo.
He notes, however, that these terms are "only those mother's sisters' sons
(daughters) who are of the phratry." He does not list another term for those
kinsmen outside the phratry who are not classified as cross-cousins.
Two types of cross-cousins are distinguished in Bara: those with Bara
107
The fish people
mothers (FZS [father's sister's son], FZD [father's sister's daughter]), and those
with non-Bara mothers (MBS [mother's brother's son], MBD [mother's broth-
er's daughter]).3
Table 7, accompanied by Table 8, outlines the structure of the terminology
system. The reader is also referred to Jackson, 1977, for a fuller discussion of the
zero-generation terms because virtually none of it is repeated here.
Expectations and behavior

The sociological meaning of a set of kin terms concerns the rights, duties, and
privileges existing between kinsmen. Of course one must also look at instances in
which these same rights and duties are ignored and challenged (Goodenough,
1965). The relationship of kinship behavior and expectations to the semantic
meaning of kin terms is the focus of at times very acrimonious debate.4 Some
authorities, for instance, Wittgenstein (1922; see also Tyler, 1978, pp. 170-72)
hold that such "usage" of kin terms is the way the terms ought to be defined.
Any generalized description of behavior among kinsmen will tend to be
normative in nature, closer to the official, rule-book version of kinship-related
expectations than occurs in reality. In addition, a thorough study of kinship
expectations and behavior will reveal some rules that are more hidden than others
yet probably at least as important in determining behavior as the more accessible
ones. Thus, even when limiting the notion of "rules" of kinship behavior to
cognitive rules (as opposed to statistical rules derived from observation), many
different kinds of rules operate at different levels of specificity.
As a prologue to discussing the roles kinsmen perform in Tukanoan society, I
offer some mythical illustrations of certain basic themes in Tukanoan kinship.
The synopses of myths that follow are not intended to address any issues
concerning the analysis of these myths and although used as illustrations of
kinship themes, no implication is intended that this is the "function" of Tukanoan
mythology.5
Some myths presented in this section - which emphasizes consanguineal
kinship - could just as easily have been included in the following chapter on
marriage. The choice is somewhat arbitrary because to some degree all myths
deal with agnatic and affinal relationships. The names of the myth synopses are
simply for purposes of identification.
Boraro's skin. A man found the skin of a boraro (a type of forest spirit) among some
logs while the boraro was fishing for shrimps. (The skin was like clothing; the boraro
always removed it to swim.) The man put on the skin and it took control, making him eat
the boraro, visit the boraro's house, and sleep with his wife. After a while, the man took
off the skin and returned to his own house, where two years had passed. His relatives had
108
Table 7. Bard kinship terminology
Address or
Term reference Kin types Definition
nihkii A,R FF, MF, FFB, MF1 all second ascending genera-
tion males
nihko A,R FM, MM, MMZ, all second ascending genera-
FMZ, etc. tion females
pahkii R F own father
pahko R M own mother
kakii A F own father
kako A M own mother
biigii A,R FB, MZH, FFBS, all first ascending generation
MFZS, etc. male agnates (except father)
biigo A,R MZ, FBW, FFZD, all first ascending generation
MMZD, etc. female affines (except mother)
mehkii A,R MB, FZH.MMZS, all first ascending generation
MFBS, etc. male affines
mehko A,R FZ, MBW, FFBD, all first ascending generation
MMBD, etc. female agnates
hwu A,R eB, FeBS, etc. elder brother; all patrilateral
male parallel cousins classi-
fied as elder"
ho A,R eZ, FeBD, etc. elder sister; all patrilateral fe-
male parallel cousins classi-
fied as elder"
baii A,R yB, FyBS, etc. younger brother; all patrilat-
eral male parallel cousins clas-
sified as younger"
bayo A,R yZ, FyBD, etc. younger sister; all patrilateral
female parallel cousins clas-
sified as younger"
mehk6-mahkii A,R FZS, FFBDS, etc. all same generation affinal
males with agnate (Bara)
mother*
mehk6-mahk6 A,R FZD, FFBDD, etc. all same generation affinal fe-
males with agnate (Bara)
mother*
mehku-mahku A,R MBS, MFBSS, etc. all same generation affinal
males with nonagnate (Bara)
mother*
mehku-mahkd A,R MBD, MFBSD, etc. all same generation affinal fe-
males with nonagnate (Bara)
mother*
pahko-mahku A,R MZS, MMZSS, etc. matrilateral parallel male cous-
ins (who are not Bara); uter-
ine half-brother
109
Table 7. (cont.)
Address or
pahk6-mahko A,R MZD, MMZSD, etc. matrilateral parallel female

cousins (who are not Bara);
uterine half-sister
mahku A,R S, BS,C all first descending gener-
ation male agnates, male
speaking
all first descending gener-
ation male affines, female
speaking
mahko A,R D, BD,C ZD d all first descending gener-
ation female agnates, male
speaking
all first descending genera-
tion female affines, female
speaking
ho-mahku A,R eZS" elder sister's son
ho-mahk6 A,R eZDrf elder sister's daughter
bayo-mahkii A,R yZSrf younger sister's son
bay6-mahko A,R yZiy younger sister's daughter
hwu-mahku A,R eBSrf elder brother's son
hwu-mahkp A,R eBD d elder brother's daughter
baii-mahkii A,R yBSrf younger brother's son
bau-mahk6 A,R yBDrf younger brother's daughter
parami A,R SS, DS, SSS, DSS, etc. second descending generation
male
parameo A,R SD, DD, SSD, DDD, etc. second descending generation
female
tend* A/R WB, ZH, BWB, ZHB, etc male cross-cousin; own gen-
eration male affine
tend'' R WZ, BW, BWZ, ZHZ, etc female cross-cousin; own
generation female affine
num6c R W own wife
mamie R H own husband
buibahkiie R X.f"l f ,T~\ 1*1 same generation male affine,
female speaking
buibahkoe R ZHZC same generation female af-
fine, male speaking
Ohio" R BW,d HZ, d BWZrf sister-in-law, female speak-
ing
pehu" R HW, HBW co-wife, wife of husband's
true brother, female speak-
ing
mohokuc R ZS,rf BDH,rf WBSd all first descending gener-
ation male affines, male
speaking
110
Kinship
Table 7. (cont.)
Address or
mohok6e R Z D / BSW/ WBDd all first descending gener-

ation female affines, male
speaking
pamokii6 A,R D H / BS, d etc. all first descending gener-
ation male agnates, female
speaking
d
pamoko" A,R sw/ BD, etc. all first descending genera-
tion female agnates, female
speaking
buf R DHC own son-in-law, male speaking
kamf R DHrf own son-in-law, female speak-
ing
Note: All terms are used by both male and female speakers unless otherwise indicated.
Plural forms are not given. See text in this chapter for discussion of sex of speaker
distinction in the first descending generation. The endings -6 and -ii are sometimes
nasalized, as in mehku and mehko. For reasons of diacritic simplicity, none of the
terms that have nasalized endings shows this.
Abbreviations and symbols
A = address; R = reference; F = father; M = mother; B = brother; Z = sister;
H = husband; W = wife; S = son; D = daughter; e = elder; y = younger.
"See Chapter 5 for discussion of use of sibling terminology to denote sib and language
group (within the same phratry) rank.
See Jackson, 1977, for discussion of mehk6-mahkii/-6 and mehku-mahku/-6
terms.
c
Male speaking.
''Female speaking.
e
Affinality being stressed.
•tJsed only in address between men.
given him up for dead. He told them what had happened, and one of his relatives wanted to
see the boraro's house and fishing territory for himself. They both went back and one of
them put on the boraro's skin, and the other put on the boraro's wife's skin while she was
fishing. The man wearing her skin then ate her. Finally the two men returned to this world
and told their stories, but their relatives did not believe them. They wanted to go and see
for themselves, and finally, over the protests of the two men, everyone but one went to the
other world. When they arrived, the two original men put on the two boraro skins and ate
everyone else. They returned to their settlement, told their story, and all the wives wept.
Three men remained, no more.
The most important theme in this sketch is the danger of getting too close to
beings who are not "like us." In this case, the man got too close first by venturing
111
The fish people
Table 8. Simplified Bard kinship terminology
Agnates Affines Affines of affines
<J FF/MF nihku

+2
9 FM/MM nihk6
6 FB biigu MB mehku
+1
9 FZ mehk6 MZ biig6
<J eB hwu yB baii FZS mehko-mahkii MZS pahk6-mahkii

0
9 eZ ho yZ bay6 FZD mehk6-mahko MZD pahko-mahko
6 S mahkii ZS mohokQ
-l
9 D mahk6 ZD mohok6
6 SS/ZSS p&ami
-2
9 SD/ZSD parameo
Note: Terms for male ego only. No individualizing terms (e.g., F, M) are included. Bara
terminology fits a basic Dravidian pattern except for separate "affine of affine" terms and
optional MBS/MBD terms when alter is not the child of a female agnate. See Jackson,
1977, for fuller discussion. Cf. C. Hugh-Jones, 1979 and Arhem, 1981, for discussion of
Barasana and Makuna terminologies.
too near and then by assuming the other's appearance, eating his food, and
sleeping with his wife. Tragedy follows, with the protagonist eventually eating
virtually all of his own relatives.
Kamaweni. One day the older of two brothers, Kamaweni, was making ceremonial
headdresses and other objects. Despite the warning of his younger brother, Kamaweni ate
some fish. He then grew very fat and soon was repugnant to his brother. Kamaweni ate
everything in sight, and when climbing a log over a river, he vomited up all the fish he had
eaten (which turned into a number of present-day fish species). After this, being thin and
little, he climbed up on his brother's shoulder and stuck his legs into his brother's chest
behind the collarbone (which is why we have holes there today). After some more
adventures, his brother became bored with this, and once when a thirsty Kamaweni (after
eating too much of a certain fruit) demanded to be taken to some water, his younger
brother tricked him into following a kind of toad (which has a voice just like humans) into
the forest. When Kamaweni returned, flying, his brother had closed all the doors of the
longhouse where they lived. Kamaweni sat on the top of the longhouse, saying, "We are
one family, of one father; why don't you love me?" Later, because Kamaweni did not
112
Kinship
believe what his younger brother said about some jaguars he had summoned, Kamaweni
was eaten by them.
A number of points are illustrated in this myth. An older brother is entitled to
respect, but only if he behaves properly. Kamaweni does not and begins a process
of moral and physical decline that ultimately ends in chaos. He breaks an
important food prohibition, eats excessively, becomes ugly, then vomits exces-
sively. Weak and small, he becomes a literal burden on his younger brother,
which leads to more excessive eating and drinking, until finally his brother
defaults on his obligations and lies and plays tricks. Kamaweni's brother refuses
to be his brother's keeper when such excessive demands are made and, instead,
becomes his slayer. Thus is Kamaweni's brother sucked into Kamaweni's decline.
The myth illustrates the need for moderation in one's obligations to kinsmen and
the necessity for implicit, unquestioning trust, at least between full brothers.
Kahpea orero. This myth deals with a family's trip to Ewiira taro (a miriti swamp
important in Tukanoan cosmology) to collect thatch. The oldest brother, a shaman,
recognized a little creature of the forest, Kahpe"a orero ("eye-extractor"), who was
eating eyes and making tiny tipiti's (manioc-juice extractors) and saying, "these are for
taking out the eyes of those people over there." The shaman returned and warned his
relatives to hang their hammocks high that night so they would not run the risk of having
their eyes stolen. But they did not believe him. That night Kahp6a orero did come, and
he took out the eyes of everyone with the little tipiti. The shaman woke up and shouted, but
his family did not wake up, because they had already lost their eyes. He tried to save the
eyes, but it was too late; they had already been cooked. He killed Kahpea orero. After
this, the shaman had to take care of his brothers, leading them with a vine. He finally gave
up and deserted them, saying it was their own fault. But then he regretted this action and
returned, saying, "These people are my family, how can I throw them away like this?",
but it was too late. They had turned into spider monkeys, mahokd ahke, living in trees
and making noises like these monkeys. The shaman returned to his house alone.
This myth also deals with the theme of relatives becoming overly burdensome,
here also because they lacked trust and did not take their brother seriously. The
protagonist ends up having literally to lead them because they refused his initial
guidance. When he finally gives up on them, they lose their humanness, and he is
totally bereft of kinsmen. People must fulfill their kinship obligations if they wish
to remain human and continue to have kinsmen.
Live Woman. This myth is about the Tuyuka, Yiikiiro, a man whose father, who knew
many chants and spells, died before having an opportunity to teach them to his son.
Yukuro's father's ghost started to harvest coca in Yukuro's manioc field; they met and
Yukuro was taken to theTuyuka dead-spirit house on Behuya (weapon) stream. Yukuro's
father taught him the proper chants and spells, and Yukuro danced with the dead spirits
till dawn and then returned to his house. Although he had been gone only a night, his wife,
Live Woman, had become very angry and had burned all the dance costumes and thrown
out all his coca and banisteriopsis. Yukuro became angry and then became sick from so
much anger and sadness. He realized his death was imminent and told his wife, "When I
113
The fish people
die, go to such-and-such a path." After he was buried, his spirit departed, following the
path. He met his wife and told her to follow, keeping her eyes closed, but she saw a little -
the dead-spirit house and the stream. His relatives received him, saying "now our elder
brother has arrived." He was put into a large vat and covered with a cottonlike substance
and came out a newborn infant. Live Woman slept outside the house, getting very cold
because of the exposure and lack of fire. A number of unpleasant events occurred to her,
all because she was a human being in the dead-spirit house: She mistook penises for fire,
was disobedient, and, after giving birth to a son, died along with her child.
This myth introduces the idea of the importance of a father teaching his son all
the important knowledge; if the father dies before this is finished, difficulties will
set in. In Yiikuro's case, he deserts his wife (even though only for a night) and
she, in retaliation, destroys his ceremonial paraphernalia - symbols of his
humanness and membership in Tukanoan society. The story continues with the
constant themes of separation, anger between husband and wife, and the wife's
jealousy over her husband's loyalty to his own sib-mates. Finally, everything is
wrong: Yiikiiro has died too soon and his wife is in a dead-spirit longhouse
without having died. Husband and wife are truly alienated at this point, one dead
and one living. They cannot continue as a couple and his descent line cannot
continue.
Namaku.ru. A man died, leaving a wife, two sons, and a little girl. His wife married a
wahti (forest spirit), but her children did not like him because he was not human. The
mother had tried to keep her second husband a secret: She would let him in only when the
children were asleep, but sometimes they stayed awake and consequently knew everything.
The wahtf would bring food, but sometimes it was not human food, such as snakes. The
mother would serve the wahti good things to eat. When the mother became pregnant and
gave birth, it was to a deer child, Namakuru. She kept it suspended in a sack high from the
ridgepole.
Various versions of the myth differ from here on, but the thrust of the story concerns the
mother's deception of her children regarding her newborn child. She makes them spend
long hours bathing, even during the night, while she spends time with their stepfather and
nurses the deer child. Her children, in turn, deceive her, spying on her and killing their
wahti stepfather by poisoning him with manioc that has not been fully processed.
Although in one version the children, particularly the little girl, play with the deer child
Namakuru {hamd, "deer") for a while, Namakuru becomes increasingly animal-like,
eating only leaves, and finally runs off into the forest. The children fill up the sack he was
kept in with ashes, which, when discovered by their mother, enrages her so much she
punishes them by smearing hot peppers all over their bodies. They have to "sleep" in the
river two nights because of the pain. The myth continues with more and more aberrant
behavior, more and more of their world going awry, and after a series of symbolically
significant transformations, the myth ends in chaos and disaster for the mother and
children.
The story of Namakuru illustrates the jealousy and rivalry that develop between
siblings, especially half-siblings. It also demonstrates the inevitable tensions
between children and stepparents. Food and sex are the foci of struggle as well as
the symbols of estranged affection.
114
Kinship
A number of other myths illustrate the theme of competition between "moth-
er's children" as well. In some myths the characters are uterine half-siblings, as
in the Namakuru myth, and in one myth collected by S. Hugh-Jones the plot
involves cohusbands of a woman (who would also address one another by the
pahk6-mahku ["mother's son"] term). Groups of people (settlements, or
entire language groups) who marry women of a third group but do not exchange
women between themselves also address one another by this term. The myth
about cohusbands of the same wife illustrates this type of relationship between
entire groups. The association between wahti and "mother's children" (pahko-
mahku-6) appears several times in Tukanoan myths. For example, in several
stories the protagonist is on his way to visit his pahk6-mahkara and ends up in a
wahti's house. The ambiguity is clearly intentional.
This myth also deals with good and bad mothers and good and bad children.
The mother of Namakuru has made an improper second marriage and lies to her
children for her own selfish purposes rather than for their own good. She makes
her children do strange, potentially harmful things (such as bathing for long
periods of time late at night). When she becomes angry she punishes them far too
severely. She demonstrates her "badness" by giving birth to an animal-like
infant. Because of her association with her wahti husband and her rejection of her
children, she herself becomes progressively more animal-like, eventually turning
into a bird.
For their part, her children also become progressively "bad," unchildlike, and
eventually unhuman, turning into birds themselves. They fly into a little hole in a
rock, their new home. Their mother, also a bird by now, cannot follow them: She
becomes stuck because of her large breast (the one she fed Namakuru with).6
Koa-mahkU. At the beginning of this myth, the culture hero Koa-mahkU is
seducing both daughters of Wahobiro, who gets angry at this sexual greediness and also
because Koa-mahkU eats excessively and is lazy. Wahobiro succeeds in abandoning
Koa-mahkU in a tall tree (a trick that worked partly because Koa-mahkU was so
greedy), telling him, "You can stay there till you die." KoS-mahkU is saved by some
birds, eyoa (wood ibis) because he helps them get past a huge pillar of fire that reaches
from a mountain to the heavens. They take him to the house of their sister, who is the
daughter of Nim£ Pin6 ("Curare Anaconda"). During their visit they keep Koa-
mahkU hidden, but she discovers him, again because of his greediness. She thinks he has
been left by her relatives the Eyoa to be her Maku, that is, her servant. She puts him to
work, sending him for water and firewood. He becomes sexually interested in her, and she
says, "Make a comb for my hair." Her pubic hair is filled with poisonous snakes,
tarantulas, scorpions, and similar stinging creatures. He makes the comb and they begin
sleeping together, but he is so greedy that she becomes fed up with his demands. She
decides to take him to her father's house, which was exactly what Koa-mahkU had
hoped would happen. They arrive at her father's house, which is all rock, like a cave, and
made entirely of curare. She tells Nima Pin6, her father, that she has a servant for him,
and NimS Pin6 puts Koa-mahkU to work. But soon Koa-mahku turns himself into
a flea, enters Nima Pin6's nose and from his heart takes curare, which is the very best
kind. Koa-mahkU then escapes through a hole he has previously drilled in the rock and
115
The fish people
uses the poison to kill the hawk Name\ which had been terrorizing and eating all the
people.
The myth continues with Koa-mahku pretending to be the pahko-mahkii of Ana
(poisonous snakes) in order to be able to kill them, and he succeeds except for one who
escapes to the lower world. He returns to his grandmother's house.
Three kinship-related themes are stressed in this story. The first concerns the
inappropriateness of greed for women and sexual intercourse. Koa-mahku is
pictured as being generally overly lustful, both for the daughters of Wahobiro
and for the daughter of Nima Pino. The theme of a man wanting one or, worse,
two women, with no indication of a sense of responsibility or reciprocity on his
part, occurs in several myths, involving other protagonists as well as Koa-
mahku. In the areas of sex and marriage, greed and selfishness will lead to
unintended and undesirable consequences.
The second theme concerns proper exchange behavior between relatives.
Nim£ Pind's daughter is given a gift from her classificatory siblings, the Eyoa:
Koa-mahku disguised as a Maku-like servant. In turn, she gives the servant to
her father.
The third theme is another example of the ambiguity and potential danger
inherent in the "mother's children" (pahko-pona) relationship. Koa-mahku
pretends that the Ana are his pahko-pona, which enables him treacherously to
kill them.
Some of the general themes expressed in these myth fragments are the
following:
1 Relatives are supposed to love one another and look out for one another; when
expectations are not met, everything can go haywire, with people eating their
own kinsmen, causing their kinsmen to be eaten by jaguars, and punishing
their kinsmen too severely or deserting them. But normally one should trust
and respect one's kin, for they have one's best interests at heart.
2 Although authority and respect can be initially ascribed by birth order, it must
be maintained by behaving appropriately.
3 Ambiguity, potential tension, and even danger inhere in certain relationships,
such as between parents-in-law and children-in-law and between pahko-
mahkara ("mother's children").
4 Conflicts of interest frequently occur between the husband and wife and the
parent and offspring.
5 Kinsmen who are greedy and irresponsible will upset things in many dire and
unpredictable ways. Irresponsible relatives can literally become unbearable
burdens, and finally even very close kin will betray and desert the offender.
Even the most devoted kin have limits of energy and forgiveness for an erring
kinsman.
6 People should associate with their own kind. Because things are seldom what
they seem, consorting with beings too different is never successful and
116
Kinship
inevitably brings many unpleasant surprises. Furthermore, such association is

contaminating, and leads to one's becoming like these others, creating disloy-
alty toward and betrayal of one's other kin.7
Specific kinship roles

Before discussing specific kinship dyads, I wish to point out certain factors that
affect all Tukanoan kinship-related behavior. One of the most crucial of these is
the social structure of the longhouse. The physical structure of the longhouse
encourages very close contact among all children and adults in it, although the
hearth family is the basic unit of socialization. This creates a primary cleavage
between coresident and noncoresident types of kinship relations.
Another consideration is the amount of time men spend at the longhouse: Not
only do children constantly interact with many people besides those in their
hearth family, but they do so with their uncles and older male cousins to a much
greater degree than in most other societies. While growing up, children of both
sexes have intimate and long-lasting contact with infants and small children; the
amount of preparation for parenting they receive is far greater than occurs -
especially for boys - in most other types of domestic arrangements. The contrast
in this regard between traditional Vaupes society and our own is particularly
striking.8 The dimensions of age and sex obviously pervade every kind of kinship
relationship. The firstborn son or daughter has more responsibility and does not
have the experience of being parented by an older sibling. The last child a couple
has is said to be the favorite and thus the one most likely to be spoiled. Still, birth
order does not matter nearly as much in Tukanoan society as it does in many other
traditional societies. As mentioned earlier, the birth of the first child of the new
generation in a local descent group is an important event. Ideally it is the son (or
grandson) of the headman, because this position implies certain differences in
status and expectations.
Parents and offspring.

Both parents nurture very young children, and the sex of the child is irrelevant to
the amount of parenting from each. But as they grow up, children spend
increasingly more time with their same-sex parent.
From a Western viewpoint, parents are very permissive with young children.
Access to parents is almost never denied. For example, no work periods in and
around the longhouse require parents' separation from their children. Except for
the few genuine dangers, such as fire, very few areas or objects in the longhouse
are forbidden. (It is surprising to return to the United States and realize just how
lethal a world confronts a toddler in his own home.) If it is impossible to grant
117
The fish people
very young children what they want, they will be distracted, amused, or simply
bought off with a treat; the denial is smooth, soft, and although at times blatantly
manipulative, very seldom an outright rejection.
Children should always be in the company of others, but this is true for
everyone else as well. The myth of di wdhako (described in the following
chapter) demonstrates the tragedy that can befall a family if children are punished
by being shut out of the longhouse and left alone. Other myths recount a variety
of gruesome ends met by children alone in the longhouse or who have strayed into
the forest.
Children learn how to behave both by imitation and by gentle suggestions from
parents. Past a certain age, they seem far less likely than children in our own
society to act out or deliberately to test the limits of patience of their elders. I
noticed very little intentional misbehavior in children over three, although this is
not to say that children, especially younger ones, do not at times yell with rage or
go through phases of seeming to burst into tears at the slightest provocation.
When children disobey or otherwise cause displeasure, they are very rarely
directly punished, and most often disapproval is not even apparent. The two times
I saw a parent hit a child were the opposite of a "rational" technique of
socialization, and were obviously unusual behavior on the mothers' parts.
At some point in children's lives (when depends on development, personali-
ties, and birth order position), their parents begin to toughen up and expect them
to be more responsible for their actions. None of this is explicitly acknowledged;
I heard very little parental philosophy per se discussed. The methods used are not
explicitly justified or rationalized and occur rather randomly and unevenly -
which iii itself is probably an important lesson to be learned about life. Parents
apply techniques of social control found in other areas of Tukanoan life: ignoring
the behavior, collective ostracism, gentle laughter (at times escalating to ridi-
cule), and similar methods.
Little girls begin to work at an earlier age and for longer periods of time than
little boys. In large part this obviously has to do with the nature of women's as
opposed to men's work. Nevertheless, little boys take care of younger siblings and
run errands for their mothers and aunts. This is partly because the sexes help each
other with tasks, even though the tasks themselves are almost invariably desig-
nated as either men's or women's. Most often, children are very willing to run
errands, and at times they are very proud to be asked. Still, little girls, who
sometimes carry water, wash clothes, and gather firewood in quite grown-up
fashion, cannot be forced to work, for no one can. This points out a general fact
of the parent-offspring relationship: It is founded on an almost natural and
spontaneous line of authority based on skill and knowledge, and both parties
must voluntarily accept such authority. Parents nurture and give to children far
more than the reverse and so have more authority. But the overriding characteris-
tic of this relationship is respect, on both parents' and children's parts; it is also
generally true of relations between husband and wife. To a certain degree, parents
118
Kinship
treat their children like adults, or rather like potential adults, in that they have
their own autonomy in many respects. The enthusiasm children at times display
in helping out is not abused, because no one can coerce another into doing his or
her job. Help may be offered and gratefully accepted, but this never releases a
person from working. Even Maku servants do not relieve their masters and
mistresses of daily tasks.
Admittedly the above statements are somewhat subjective. Still, it is clear that
Tukanoans believe that respect and trust should flow between parents and chil-
dren. It is a theme in a number of myths (which usually teach by demonstrating
the negative consequences when such respect and trust are lacking). Further-
more, a lack of respect (of recognizing that everyone has a right to be one's own
person) is an oft-heard complaint when difficulties erupt between parents and
their older children. At present, this problem is exacerbated by the younger
generation's sometimes being more able to elicit prestige and respect according
to the new, intrusive value system as a consequence of their knowledge of Spanish
and of mission town culture; a personal acquaintance with missionary personnel;
an appointment as the catequista of a settlement; and the possession of shirts,
shotguns, and other forms of wealth.
The worst fight I witnessed at Pumanaka buro involved a man and his eldest
son. Many accusations of lack of respect on everyone's part were hurled. Eventu-
ally the entire longhouse became involved when the son picked up a coca-
pounding club made of extremely hard wood, and opened a four-inch gash in his
father's head.9
Uncles, aunts, nephews, and nieces.

Relations between these kinds of relatives depend, first, on whether they are
agnatic or not, and second, on whether the individuals are coresidents or not. A
child lives with some of his classificatory father's brothers, who are called
bugti; their wives are called btigo (classificatory mother's sister). Both terms,
especially the bugo term, extend to many more kinsmen; thus the terms can
refer to the very close relationship an ego has with coresident father's brothers
and their wives (who may be actual mother's sisters, although this seldom
happens), whereas in other cases, the terms refer to + 1 generation male agnates,
or + 1 generation female nonagnates, respectively.
A child calls his MBs (any + 1 generation male belonging to his mother's
language group) mehku and their wives mehko. Ideally, only + 1 generation
Bara women are supposed to be called mehko, but the wife of a mehku is
addressed with this term, although a mehku's non-Bara wife may be called
mehkU-numd ("mehku's wife"), which usually indicates that she is not
Bard. Ego calls all FZs ( + 1 generation Bara females) by the mehko term. As
with the biigu/biig6 set, expectations of behavior cannot be generalized: The
amount of interaction determines these expectations to a greater extent than with
119
The fish people
the preceding set because these individuals always live in other settlements.
Certainly a close mehku or mehko, if visited frequently (especially if a
potential marriage with a daughter is in the offing) is a highly significant
individual.
Mdniku and mdniko are terms that refer specifically to parents-in-law and
are never used in address; when used in reference, the affinal relationship is being
emphasized. Otherwise mehku/-6 is used.
Siblings.
All older children take care of younger ones to some extent, although girls, if
available, will do more. The traditional Tukanoan settlement's isolation undoubt-
edly plays an important part in producing the close bonds between many sibling
pairs. Closeness in age and compatible personalities obviously matter as well in
creating particularly strong affective bonds both between real siblings and between
coresident parallel cousins.10 Both brother-sister and same sex sibling pairs can
be very close. If an unmarried girl is without a classificatory sister of her age at
the settlement, she will probably feel quite deprived. A solution is to try to
establish intimacy and confidence with a sister-in-law, but this is more difficult,
especially if that woman is struggling with her mother-in-law (who is sometimes
the unmarried girl's mother). Close sisters (or a brother-sister pair) face a painful
period when one of them marries and moves away.
As siblings grow up, tensions often develop between them. For several reasons
this is particularly likely in the case of brothers. First, cultural tradition requires
that the bond between brothers be paramount, affectively strong, and permanent.
The local descent group's core, ideally if not always in fact, is the set of full male
siblings. In fact, a lot does depend on brothers' (particularly full brothers) getting
along with one another, and so when difficulties occur, the implications are
always serious. Furthermore, brothers often find themselves in situations of real
competition and rivalry with one another (e.g., for women or prestige). In
addition, because the role of the male adult, more than that of the female,
emphasizes face-saving and the need to maintain respect and self-esteem, there is
a greater possibility of real or imagined threats to this self-esteem, most often
from other men. Finally, whereas sisters eventually leave the sibling group,
brothers do not, and thus more difficulties occur between brothers simply because
they interact more.
Grandparents and grandchildren.

Relations between grandparents {nikul-6) and their grandchildren (pdrami) are
secure, warm, affectionate, and generally lack threat or anxiety. Few tensions
are built into these reciprocal roles in the first place, and the great difference in
age limits the manipulation, ambiguity, and accommodation possible. It is also
120
Kinship
true that this is primarily a relationship between older people and young children,
although some grandparents live to see their grandchildren well into their teens:
one four-generation family (Pedrina's) resided at Pumanaka buro.
Mehko-mahkul-6: cross-cousins.
Although in part genealogically denned in terms of consanguinity, these relation-
ships are also affinal and are more thoroughly discussed in the chapter on
marriage, which follows. Unless a marriage has taken place, because mehko-
mahkara are always living in different settlements, interaction is infrequent and
matters less. Moreover, this relationship, both as a potential husband and wife or
as one's sibling's husband or wife, is infused with affinal meaning. Of course,
while growing up, a Tukanoan child's interactions with mehko-mahkara are not
heavily flavored with marriageability issues. Nonetheless, such interaction ide-
ally sets the stage for later marital negotiations. Thus, while necessary analytical-
ly, it is hard to separate the relationship traced through parents from one traced
through potential affines.
Pahko-mahkul-6: matrilateral parallel cousins.

I have already briefly discussed this relationship in the form it takes between
uterine half-siblings. Otherwise, no specific prescriptions or expectations obtain
between pahko-pond.u People who are pahko-mahkii/-6 to each other are
matrilateral parallel cousins whose fathers are from different language groups.
They cannot intermarry, but neither are they classificatory siblings. The implica-
tion is that they compete for women from the language group from which they
both obtain women - their respective mothers' language group. Thus the relation-
ship has many affinal aspects implicit in it and is discussed more fully in the
following chapter.
Affines.
Kin terms that are affinal as well as consanguineal (but, obviously, never agnatic)
are bug6 (FBW/FZH), mehkii, mehk6-mahku/-6 (cross-cousins, mother
Bard), mehku-mahku/-6 (cross-cousins, mother not Bara), and pahko-mah-
ktiy-6 (refers to affines of affines; see the reference to the myth about cohusbands
of the woman earlier in this chapter). Other terms (see Table 7) refer to specific
affinal relations and are not generalizable either from ego or alter to a category of
people.12 These terms have a number of special qualities, connotations, and
constraints on usage. In general, they are not used if a more polite consan-
guineal-affinal term can be used (e.g., mehkii [MB] rather than mlnikii
[WF], or mehk6-mahk6 [FZD] rather than tend: the tend term is so loaded
with strictly affinal meaning, especially in cross-sex address, that it is almost
121
The fish people
never heard). This results in these strictly affinal terms often being used as a
default category, having somewhat bare, almost derogatory qualities. (This is
because they refer only to affines, which is an awkward and potentially hostile
category when no other relationships exist between ego and alter.) These terms
are used in discussing affinal relationships, when marriage and its problems and
interactions are being dealt with as separate from other kinship relations.
Several terms show, at least in part, some of the strains placed on a basically
Dravidian structure when several rather than only two intermarrying groups are
actually exchanging women. I have briefly discussed the pahko-mahku/-6 set
of terms in this regard. This set acknowledges the presence of three intermarrying
units in the + 1 generation: ego's group, ego's mother's group, and the group that
ego's mother's sister married into - which is obviously not ego's own, because
the sibling term is not used for these parallel cousins. The tend term gives a
strong emphasis to the marital relationship itself, because, when1 possible,
cross-cousin terms are normally used between sisters-in-law and (in particular)
between brother and sister-in-law. Tenti, although it also stresses the affinal as
opposed to the consanguineal (cross-cousin) relationship, is used more fre-
quently between brothers-in-law, because this relationship, even though strictly
affinal in meaning, is important in its own right. Another term, pehu, when
applied to females literally means "co-sister-in-law," "co-wife," and again is
used (almost never in practice) when no other term can be substituted. The
following myth of women who share the same husband shows the ambivalence
and potential conflict of interest between co-sisters-in-law who are bound by no
other relationship.13
Pehu. A widower takes a second wife. Thefirstwife, now a wahti, appears to the second
in her maniocfield,invites her to collect edible ants with her, offers her cazabe and offers
to take care of the second wife's baby. Little by little the second wife gets suspicious and
discovers the truth when thefirstwife, while taking a nap, snores and says:' 'pehutorono•,''
acknowledging that she is a wahti co-wife, the second wife's pehu. After a series of
adventures involving some forest creatures described as the pahko-mahkara of the wahti,
the second wife is saved at daybreak by her husband. They leave thefirstwife in the forest
and are never bothered again.
Virtually all myths deal with affinal relationships. Many of them stress the
same themes, one of which is that affines are necessary, can be the source of gifts
and benefits, and will make many sacrifices. Still, affines are always different
from one's own kind and hence ultimately unknowable. Furthermore, certain
built-in risks, conflicts of interest, and competition arise in affinal relationships,
for example, the competition with one's same-generation agnates and with one's
pahk6-mahkara when acquiring a wife.
The myth of Wai Pino Mahko, Fish Anaconda Daughter, portrays affinal
relationships in several ways. This woman, who marries Yeba Mahu (a human)
is responsible for introducing tobacco and banisteriopsis to humans and illus-
trates the help and benefits that can be obtained for establishing good affinal
122
Kinship
relationships. Still, acquiring these two plants predictably comes about through a
series of mistakes and ignored instructions. The next section of the myth is
concerned with a trip Yeb£ Mahu makes to visit his in-laws. This episode was
mentioned previously in the discussion of sib rank. Yeb£ Mahu visits the
longhouse of his Wai maha ("fish people") in-laws, the headman of which is
Wai Pin6 (Fish Anaconda), the father of Wai Pino Mahko, Fish Anaconda
Daughter. The Wai maha eat only "pure" food, such as termites. However, Wai
Pin6, seeing that his son-in-law is hungry, offers him his (Wai Pino's) own
relatives to eat, illustrating the sacrifice one must make of one's agnates in favor
of one's affines. Wai Pin6 offers Yeba Mahu the Doe (Tarira) fish to eat
because they are the lowest-ranking species of fish there and are described in the
myth as "like the Maku . . . like the Wamutanara sib is to the Bara."
Another theme in the myth of Wai Pin6 Mahko is the risk and danger inherent
in visiting in-laws, because they are by definition dissimilar. At one point, Yeba
MahU is almost suffocated by his father-in-law's anaconda skin, in part because
Yebd Mahu disobeyed instructions. Wai Pin6 returns in time and strikes the
skin, which of course obeys him. This is a graphic demonstration of the very real
risk of being suffocated or squeezed to death by one's affines. It also illustrates
the fact that all affines are potentially two-faced: generous and friendly when it
suits them and when they are obeyed but dangerous when things go contrary to
their liking. This last point is also demonstrated when Yebd Mahu is greeted by
his affines: They greet him first as anacondas by crawling all over him (their
customary greeting!), and he, although full of fear, must withstand it and
maintain his equanimity. Following this, he is greeted the way people greet one
another.
Fictive and ceremonial kinship.

Some Tukanoans form ritual coparenthood (baptismal godparent) relationships
with rubber gatherers during sojourns at rubber camps. My impression is that
these are extremely attenuated; for one thing, as in all areas in Latin America
where compadrazgo (ritual coparenthood) is found, it requires upkeep and if
allowed to lapse is effectively terminated.
S. Hugh-Jones (1979) and C. Hugh-Jones (1979) observed a "ceremonial
brother-in-law" (he tehiia) relationship in the Pira-parana. This involves an
agreement between two actual or classificatory brothers-in-law to be friends and
to exchange ceremonial artifacts. Pira-parand Tukanoans also speak of groups
as each others' he teniia. This is, in other words, an elaboration of and a
heightening of the affect implicit in an already existing kinship relationship.
Unlike the special bonds that can develop between brother and sister, however,
these are necessarily always between individuals from different settlements,
which makes a difference in the amount of intimacy that can develop.14
123
7.
Marriage
Marriage in the Vaupes is a kind of movement: People, goods, and intangible

commodities such as prestige follow marital paths linking families and settle-
ments. It is thus not surprising that Tukanoans spend a great deal of time and
energy negotiating marriages and observing and discussing how similar negotia-
tions progress in other households.
Probably low population density, dispersed settlement pattern, egalitarianism,
and a swidden horticultural base with a strong overlay of hunting, fishing, and
gathering are the most important reasons why so much extralocal economic
interaction is linked to affinal relationships. A strong component of the economic
system is that people meet the demand for nonlocal goods and services by
depending on their network of ties to other people and settlements through
organized marriage exchanges rather than through direct commodity exchanges.
This economic system suits an environment and level of technology requiring
both extensive lands that must lie fallow for longer than they are cultivated and
controls on overexploitation of hunting, fishing, and gathering resources. An
evenhanded and flexible economic system involving a strong element of general-
ized reciprocity and little capital or stored surplus is necessary. A marriage
system that allows for continual adjustments to the ecosystem and fosters some
economic dependence on social relationships maintained through affinal links in
a number of other settlements fits in well with this setting, for it places the most
emphasis on exchanges of personnel rather than on exchanges of goods. It thus
helps limit the potential for exclusive control of wealth in either goods or scarce
sources of energy by any one group. The flexibility of a system preventing such
exclusive control permits the most judicious long-term utilization of the ecosys-
tem.
Even the beginning of a marriage (I hesitate to use the term wedding) fits this
description. Significant contractual agreement, transferral of wealth, or major
ritual contact with supernatural powers does not characterize Vaupes marriage
making as it does in many societies. Unlike some Ge cultures, Tukanoan
initiation ritual is not at all concerned with marriage (Maybury-Lewis, 1979). As
C. Hugh-Jones points out, "marriage is solely a redistribution of female repro-
124
Marriage
ductive powers which makes a new generation socially possible" (1979, p. 160),
noting that the physiological ability to reproduce is established during male and
female initiation. Birth and death in Tukanoan culture are also marked by ritual
and contact with the ancestors, but marriage is distinctly secular. Sometimes (as
when, for instance, a woman is already living at her husband's settlement),
nothing pinpoints the beginning of a marriage. Certainly economic exchanges
follow the establishment of a marriage, but they depend on availability of
resources and how well the social relationships between affinally related families
and settlements are going. To the degree that an alliance is established or
continued and strengthened, ongoing movement between local descent groups
resulting from a successful marriage is a permanent process that outlives all the
participants.
Ceremonial bride capture and subsequent negotiations are in a sense wedding
rituals, but the ideal marriage lacks such evidence of unfamiliarity and mistrust
and is initiated long before a woman moves to her husband's settlement.
I analyzed a marriage sample of 684 marriages, made by 636 men and 664
women. I actually gathered data on many more marriages (more than 1,000),
most of them between the parents and grandparents of Tukanoans included in the
final sample. The marriages eliminated from the final sample were discarded
because of unreliable sources of information or too many missing variables.' The
marriages accepted for the sample were scored in terms of reliability and were
coded for twenty-six variables concerned with information about each partner and
about the marriage itself.2 As mentioned earlier, this sample is not a total census
of marriages for the Vaupes, nor is it a systematically selected random sample. It
is complete with respect to all marriages made by living individuals in the
Papuri-upper Tiqui6 drainage region: men and their inmarried wives, and
women born in this subregion who have married into settlements elsewhere.
Principles of marriage
The following discussion examines some of the underlying principles of marriage
in the Vaup6s. I am defining "principle" loosely to indicate that in Vaupes
marital behavior certain patterns are observable to some extent and that Tukanoans
verbalize them as norms when discussing marriage making. I do not (and cannot,
at this point) imply that marital behavior is determined by these principles. By no
means do they predict actual marital behavior all of the time, nor is it clear
whether they in fact operate even as norms in all marriages. A future research
objective is to discover the degree to which such principles do in fact influence
marriage making. Anthropologists tend to be overly juridical in their interpreta-
tions of how ' 'rules'' operate in small-scale societies and too quick to assume that
what is stated to be the preferred form of marriage (or any other type of behavior)
is the preference under all possible conditions. It was pointed out in Chapter 1
125
The fish people
that there is a seeming overabundance of rules and categories in Vaupes social

structure, particularly as concerns marriage. We cannot rule out the possibility
that some of these rules are epiphenomenal: Though stated to be determinants by
Tukanoans, they are in fact only after-the-fact rationalizations. Perhaps in actual-
ity such "rules" play much more of a role in ordering the Tukanoan social
universe as a conceptual system. Still, it is also possible that some rules are
constructed in such a way as to override certain other rules under given condi-
tions. In this case, what we lack is knowledge of the criteria that activate given
sets of rules and given them precedence. Silverwood-Cope (1972, p. 176) gives
an example of this type of ordering in his discussion of Maku marriage rules:
Although Maku are quite aware of the Tukanoan rules that prescribe marriage
with a person in the "potential affine" category, Maku have other rules
prescribing subregional endogamy that have priority. Tukanoans do not have such
stringent territorial rules, and therefore they condemn the higher frequency of
improper Maku marriages much more than do the Maku themselves. Undoubt-
edly, Tukanoans do have rules regulating regional endogamy, but these involve a
far vaster territory and thus are not as constraining as those of the Maku. In any
case, it is evident that we must avoid overly simplistic conclusions about
Tukanoans' obeying or disobeying their marriage rules, whether in thought or
action.3
Direct exchange and polygamy.

In the ideal Tukanoan marriage, a man obtains a wife by exchanging his real
sister for another man's real sister. "Exchanging our daughters for daughters-in-
law" also expresses this principle. This phrasing illustrates how, in addition to
the one or two pairs of prospective spouses, the two local descent groups perceive
their vested interest in any marriage making.
Although an exchange of actual sisters is the ideal, it seldom occurs in practice
because of demographic and other factors. In 18.7 percent (79) of a sample of
423 marriages, one or both wives were the actual sisters of each other's husband;
in 14.4 percent (61), the wives were the classificatory sisters and of the same
settlement as each other's husband; in 24.1 percent (102), exchanges involved
two generations (e.g., an FZD for the FZ given in the preceding generation); in
5.7 percent (24), the classificatory sisters were from different settlements from
each other's husband; and in 37.1 percent (157), no exchange took place. This
last category includes only marriages known to have had no exchange (informa-
tion on this variable was not available for 249 marriages in the total sample of
684). Several one-couple marriages were described to me as future exchange
marriages, but I recorded them as nonexchange marriages. Most involved a sister
too young to exchange at the time of the first marriage. Thus, although the ideal
type of exchange marriage is rarely realized in actuality, it is evident that
126
Marriage
marriages where two men exchange classificatory sisters or complete the exchange
in the next generation are still very much a fact of Vaupes life.
The Bard terms for marriage describe the exchange. The term hikaniya (hikd,
"one") glosses as "one-couple marriage," and puaniya (piid, "two") glosses
as "two-couple marriage." A married woman will be a hikaya mahko or a
pudya mahko, depending on whether or not she was part of an exchange
marriage.
That the ideal consists of an exchange of real sisters can be understood by
examining the negotiational aspects of marriage and the risk that the second half
of an exchange marriage may never come about, despite protestations of good
faith (and, in fact, genuinely good intentions) on the part of the local descent
group that owes a woman to another group. For example, although an exchange
can be completed in the next generation, it is risky to expect people to live up to
obligations incurred at a much earlier time, and thus sisters are preferred. It also
means that for a considerable amount of time one local group will be indebted to
another, which considers itself deprived of a woman. In addition, a man has the
most influence over his actual sisters. Classificatory sisters are certainly as
acceptable in exchange, but they are likely to be most influenced by their father
and brothers. Furthermore, a man's real sister will be more likely to consider
marrying a particular man if by so doing she can help her brother get a wife. She
will be less inclined to do such favors for a less closely related agnate, particu-
larly if he is not of her settlement. In the latter type of marriage, it always seems
that the woman would have married the man anyway; the exchange was conve-
nient to all and did not involve any persuasive techniques directed at her. Women
always have the final say, at least at present. This is one area in which the Vaup6s
system differs from certain other direct exchange systems in lowland South
America, most notoriously the Yanomamo.4 Furthermore, a man can try to
persuade his actual sister to return to her natal settlement if his wife leaves him.
Because marital quarrels threaten two marriages, more pressure is brought on the
couple to patch things up. Thus, Tukanoans say, an exchange of two women
guarantees a greater amount of marital stability. An exchange of actual sisters,
when possible, has the greatest potential for achieving two stable marriages.
Only women can be exchanged for women. According to alliance theory,
"marriage forms may be regarded as partial and incomplete expressions of
certain underlying principles of reciprocity" (Buchler and Selby, 1968, p. 103).
Marriage in the Vaupes is indeed part of a larger relationship - between families
and between pairs of local descent groups. Still, no goods can be given in lieu of
a woman, and only another woman given at that time or promised for the future
compensates this loss. Goldman says that a headman's prestige and affluence can
help him obtain a wife through bride price alone (1963, p. 145), but it is my
impression that to some degree he will still be seen as owing a woman to the
group that gave him his second wife. If a man, even a headman, receives a
127
The fish people
woman from a group, he and his own group stand in a position of indebtedness
that cannot be erased with bride price.
Tukanoans offer the fact that only a woman given compensates for a woman
received as the reason behind a general disapproval of polygyny. A man (almost
always a headman) with a second wife is indebted to the local group that she came
from - even if she was a widow, divorcee, or without close kinsmen. The
unavoidable implications of polygyny, then, are that such men have used two of
their close female agnates in exchange for their wives, or that one or two are owed
to another local group. Although a headman's wife works very hard and is seen as
more deserving of the help of a co-wife (although she herself may not see things
this way), his position as a headman does not formally entitle him to make two
marriages. A man is entitled to one wife, for general opinion has it that otherwise
there would not be enough women to go around. Tukanoans generally feel that all
men would get married if they could, and all women could get married if they
(only) would.5
I did not collect systematic data on age at marriage (an extremely difficult
task), but usually women do not marry before or immediately after puberty. This
contrasts with some other lowland South American societies (e.g., the Cuiva,
Yanomamo, Guayaki [Ache"], Siriono), in which men are often much older
than their wives. In general, in a first marriage husband and wife do not differ
markedly in age. Some Tukanoan women do not marry until well into their
twenties. That some women live with one or two men prior to settling down with
the man who turns out to be their permanent mate complicates obtaining informa-
tion on age at marriage, for informants giving marriage histories tend not to
remember such information easily.
Tukanoans thus see monogamy as the most equitable arrangement for all
concerned. In a sample of 672 marriages, only 6 were polygynous, and 4 of these
were made by one rather remarkable Tukano man. When a priest discovered the
situation, he persuaded this Lothario (my informants giggled at the story) to send
three of his wives back to their natal longhouses. It seems reasonable to assume
that polygyny was practiced to a greater degree in the past and declined because
of Catholic missionary disapproval.
A marriage at Pumanaka buro demonstrates how the principle of exchange
operates even in the absence of brother-sister pairs. Estribino, the headman's
eldest son, took a Tuyuka wife, Isabel. She came from a family living down-
stream that consisted of three sisters and no brothers. A week later, Isabel's
father, Armando, arrived, saying he wanted Estribino's younger sister, Maxi-
miliana, in exchange. This precipitated a huge quarrel, mainly between Armando
and Estribino's father, Mario. Armando maintained that he was owed a woman
and, moreover, that he had a right to a second wife because he had no sons and
therefore no daughters-in-law to look after him and his wife when they were old.
Mario retorted that Armando was too old, that he already had a wife, who could
still bear him sons, and that Maximiliana did not want to marry him anyway.
128
Marriage
Hearing this, Armando took his daughter back home with him, although she did
not want to go. Armando was counting on Estribino to pressure his sister
Maximiliana into marrying him. His plan was not successful, and Isabel did not
return to Pumanaka buro. When I left, Estribino was still unmarried.
When Armando and Mario had discussed the upcoming marriage, they had not
mentioned an exchange of women: The demographic imbalance (Armando had
all daughters) and the fact that he lived alone with his family (a quarrel had
resulted in his leaving his natal longhouse farther downstream) meant that the
potential bride, Isabel, had no classificatory brothers wanting to use her in their
exchange marriages. Armando waited until Isabel was firmly established at
Pumanaka buro, when his bargaining position was strongest, before coming
forward with his demands. Using his daughter as leverage in this fashion with this
particular family was perhaps the best possible strategy, for it was common
knowledge that Estribino had been trying to get married for several years (he was
at least twenty-six). The pressure on him to marry was particularly strong,
because a headman's eldest son should be the first to get married.
It is difficult to speculate about the incidence of piianiya in earlier times,
although "one-couple" marriages might perhaps occur more frequently today,
owing to acculaturation, depopulation, and out-migration. Nevertheless, the
principle of sister exchange is still quite strong, and Tukanoans who are quite
aware of the changes in marriage patterns due to acculturative influences maintain
that piianiya is still the best arrangement. The Catholic missionaries consider it a
form of selling women and have prevented exchange marriages from taking place.
I do not have much data on divorce per se. Information on its incidence is
difficult to come by because early marriages are often unstable, and a divorce
before children are born is obviously far less significant than one that takes place
after one or more children. Still, such breakups do happen, and the children tend
to move with the mother. Presumably it is difficult for them if she marries a
second husband belonging to a language group different from her first husband's,
but I have information on only one such case - a Carapana woman who left her
Tukano husband for a Tatuyo. That this does not happen often illustrates how
much current Vaupes marital patterns differ from those in societies with greater
tendencies toward polygyny, brittle marriages, and strong male dominance. In
such societies, for example, wives are often "stolen" by other men. Although the
wife is often blamed, that men want to take away other men's wives is seen as the
natural state of affairs. This is not the case in the Vaupes, although adultery,
always a serious matter, does occur - at least accusations are often made. A
woman who chooses to leave her husband after having several children by him to
run off with another man is seen as unusual. The case mentioned above provoked
comments about the first husband's inability to keep his wife, but no coercion
was used to bring her back.
I did not record any unquestionable cases of polyandry, although I observed a
curious situation involving a woman and two brothers in a downstream long-
129
The fish people
house. She was said to be married only to the older one, but the three had been
together a long time, she fed them both, and she had given birth to several
children. They traveled as a unit. I am suspicious of this case mainly because of
informants' surprising reluctance to discuss the younger brother's failure to
obtain a wife. I could only conclude that this was a case of polyandry or that he
was not interested in women at all.
I did not witness any accusations of adultery at Pumanaka buro itself, but I
certainly heard of them at other settlements. Their absence at Pumanaka buro
was probably due to the fact that all married people had been married for a long
while; in fact each couple had children who were ready to get married. At the
time of my arrival, no members of the next generation had married, with the
exception of a Bard woman who was staying at Pumanaka buro because her
husband was away at a rubber camp. Thus, the most potentially adulterous
situations (i.e., involving young couples) did not typify this group. It is likely
that sometimes young wives dissatisfied with their situation provoke a quarrel to
make their presence felt. Because they have little power, one way of doing this is
to flirt with their husband's real or classificatory brother. In any case, women
tend to be seen as the instigators of adulterous liaisons. Wives are seen as more
likely than husbands to experience difficulties early in a marriage. Women
receive more blame and punishment for sexual misbehavior than men, although
men are subject to more prohibitions in sexual activities (in connection with
hunting or ritual, for instance). Although it was my impression that the young
man was equally guilty in the case of incest at Pumanaka buro (see Chapter 5),
the young woman was blamed more.
Regardless of who in fact is responsible for beginning a flirtation, the least
disruptive view to take is that the outsider women are the troublemakers (see
Goldman, 1963). This characterizes a number of patrilineal and patrilocal socie-
ties (Collier, 1974; Collier and Rosaldo, 1981). Male solidarity must be main-
tained, and a rift between brothers is a serious matter, particularly when seen as a
result of active seduction rather than the milder crime of passive surrender to
overwhelming temptation. Themes of women being sexually provocative,
demanding, or even voracious appear and reappear in myths. Other societies'
stress on female purity, innocence, and maternal longings does not appear in the
Vaup6s. An emphasis on chastity, in fact, applies more to men, particularly
during special ritual periods. To be seductive, a woman must make an overture;
she cannot merely "look" sexually inviting. A picture of a seductively supine
woman in an advertisement in my American magazine was only puzzling to
Tukanoans, who asked if she were sick.
Kinship relations between spouses.

When specifically asked, Tukanoans will state a preference for marriage to a
genealogically close kinsman within the category of cross-cousin (mehko—
130
Marriage
mahku-6). The advantages of such a marriage are several. It is better from the
woman's point of view not to move into a totally strange longhouse, and a
longhouse where a woman's closest cross-cousins reside will also usually be
relatively nearby. Furthermore, not only is this longhouse likely to be inhabited
by familiar people and in familiar territory, but most of these people will be
related through several well-established connections. This means her link to them
will not be merely an affinal one. Should trouble arise, such an inmarried woman
will be seen not only as daughter-in-law but also as the close niece of several of
the residents. Living with the real or close classificatory siblings of her parents,
whether on her father's, mother's, or both sides, probably helps to resolve
marital difficulties more easily. She also speaks the language of the longhouse
where a real or close classificatory mother's brother is living and is very likely to
speak the longhouse language where a real father's sister is living. (She will, of
course, be familiar with the language of any longhouse belonging to her mother's
language group, and only some of these will contain genealogically close kins-
men.)
In 21.1 percent (n = 360) of marriages in which the husband's relationship to
the wife's mother ("FZD marriage") was known, the wife's mother belonged at
least to the husband's sib or was even more closely related. In 42.8 percent of this
sample, the wife's mother belonged to the same language group as the husband.
In a sample of 360 marriages in which the husband's mother's relationship to
the wife was known ("MBD marriages"), 20.0 percent involved a woman from
the husband's mother's sib or an even closer relationship. Of this sample, 40.8
percent of the marriages involved a woman of the same language group as the
husband's mother.
Although a woman remains a member of her local descent group throughout
her life, her membership in a residential group shifts at marriage from her natal
longhouse to her husband's. As pointed out earlier, this at times places women in
a position of conflicting loyalties. A woman's children belong to the local descent
group of her husband, and when quarrels arise between the two groups her
loyalties to them sometimes conflict with her continuing allegiance to her own
local descent group. A pair of local descent groups experiences more pressure to
resolve disagreements equitably when they have a long-standing exchange rela-
tionship involving more than one married couple, even though it is often these
same affinal relationships that sow the seeds of many of the disagreements that
occur. A pair of allied local descent groups gains much from continued contact
and loses much in a permanent rupture, because so many individual relationships
are involved. Therefore, both groups are more likely to keep their promises and
work at resolving conflicts when more than one inmarried woman connects them.
In many societies with prescriptive marriage rules, marrying too close a
relative - a double cross-cousin or a first cross-cousin - is prohibited. In the
Vaupes, not only are these close cousins possible spouses, they are stated to be
the preferred type. In societies that prohibit such marriages, a moiety system (as
131
The fish people
opposed to the Vaupes's open-ended system) sometimes operates, or a greater

chance exists of local groups forming too exclusive an alliance with only one
other local group - because this would most likely always be the preference of the
individual parties involved in marriage making. Forming too consolidated an
alliance with one other group seems less likely in a situation such as that in the
Vaupes, where demographic constraints and the presence of so many exogamous
groups mitigate against the likelihood of ending up with all one's affinal eggs in a
single basket. In the past, other factors may have served adequately to prevent the
formation of too exclusive an alliance between pairs of local descent groups.
In neighboring groups, such as the Yukuna (Jacopin, 1981) and Witoto
(Gasche, 1977), a man cannot marry into any longhouse in which his brothers
have already married. Tukanoan marriage patterns also show dispersal of group
members, although proscriptions of this sort are not explicitly stated.
At present I cannot discern a tendency from the actual marriage data to marry
more closely through one of the two patrilines connecting the spouses. Speaking
from the perspective of a male ego, Bara informants stated a preference for
marriage with an FZD rather than an MBD. Their explanation of this preference
concerns a general structural principle rather than presumed advantages to the
individuals involved. FZD marriage is preferred to MBD marriage because it is
"getting back a woman (FZD) for the one we gave out in the previous generation
(FZ)." Marriage with an MBD, on the other hand, creates a double debt to the
affinal group because an MBD's local descent group has now given both an M
and an MBD.
We cannot combine these two stated preferences into a single composite
preference (e.g., "the ideal marriage is with a true FZD"), because informants
argue them from different grounds. A woman probably prefers to be both an FZD
and a genealogically close kinswoman of her husband, for she would then very
probably move into her mother's natal longhouse. This preference for FZD
conflicts with ethnographic reports from some societies having prescriptive
bilateral marriage in which natives state a preference for MBD as opposed to FZD
(Elkin, 1953; Radcliffe-Brown, 1953). Furthermore, some ethnographers work-
ing in the Vaup6s also report a preference for marriage with MBDs (e.g.,
Goldman, 1963, p. 138).
I must stress that although Tukanoans may state that marriage to a specific type
of relative is preferred, what they actually do when arranging marriages is
influenced by many other considerations as well.6 As noted, I tried to see whether
any pattern emerged in actual marriages in terms of a greater genealogical
closeness between spouses through the wife's father's versus mother's line. But
the tests I ran, using a sample of 180 marriages in which the relationship through
both lines was known, were statistically inconclusive (see Jackson, n.d. b, for
further discussion of this and similar points). A possible consideration is the
relative ease with which a father can negotiate for a son's wife through his WB
(wife's brother) versus his ZH (sister's husband) (and it certainly is easier for a
132
Marriage
mother to influence her B than her HZ and HZH). The longhouse in which a WB
lives tends to be more of a known and secure place than one in which a man's Z
has married (assuming these are different). Although a man's bond with his sister
may be strong, he cannot as easily control her affinal relatives, because she,
being an inmarried woman, is on less sure footing herself. Still, if a strong
marital bond exists between husband and wife, perhaps they can exert consider-
able pressure on the wife's brother. A man's relationship with his kinsmen at his
mother's natal longhouse (where a close MBD would come from) is usually
stronger than with people whom an FZ lives with, and therefore we might
speculate that whereas from the structural point of view just summarized
Tukanoans prefer FZD marriages, from a bargaining point of view they know
MBD marriages are generally easier to negotiate. (But I am not saying that we
can collapse all the parameters of actual marriage making into this single factor.)
Genealogical closeness does not guarantee familiarity between prospective
spouses. For example, during my stay, a marriage was arranged between Nazalio
and his first cousin, Fernanda, a Tuyuka girl from the Tiquie. The people at
Pumanaka buro knew very little about this girl and her family, because contact
had not been maintained after Nazalio's mother, Esmeralda, had moved to her
husband's home.
Although Tukanoans agree that marrying a woman from an unknown and
far-away longhouse is not the best type of marriage, such arrangements are not
automatically "default" marriages. Making new alliances can be advantageous
for any number of reasons, many of which are illustrated in myths. Moreover, any
thorough analysis of marriage should examine both individual marriages and the
total pattern of marriages made during a generation at a given settlement. Where
one or two men have obtained wives perhaps has bearing on where their brothers
and patrilateral cousins marry. Marriages between distant relatives or between
those in settlements previously unknown to either partner increase a settlement's
pool of affines, allowing a network with a greater geographical range (discussed
later, and in Jackson, 1976). Although such marriages are less likely to be seen as
desirable in the abstract, they occur in the real world. Tukanoan men visit, court,
and fall in love with little-known women despite the knowledge that the "ideal"
marriage is with a neighbor MBD or FZD.
Apparently, mock, or ceremonial, marriage abduction still occurs from time to
time, although I did not witness one and was not told of any happening during my
stay. Real raids for women no longer occur, and they never characterized mar-
riages gradually arranged by close kin. Wife-capturing raids are still a part of
every Tukanoan's heritage, however, in that they figure prominently in stories and
myths.
When an "abduction" is planned today, the groom and his close kinsmen
arrive during the night in one or two canoes and ' 'kidnap'' the bride by sneaking
up to the longhouse and entering and escaping through side doors used only for
these elopements and by the women during Yuruparf rites (S. Hugh-Jones,
133
The fish people
1979). The degree to which this is a real elopement, that is, secret from the
bride's family, undoubtedly varies from case to case. The rest of the longhouse
inhabitants wake up in time to fight a mock battle at the canoe landing, involving
a lot of yelling but no real damage.
Although not a real abduction, this ritual emphasizes the seriousness of the
occasion and the latent antagonism and suspicion on both sides. At times,
apparently, the bride's family shows displeasure with the match (see Goldman,
1963, p. 142). This feigned or actual reluctance gives the bride's family leverage
in the ensuing negotiations a few days later when they visit the groom's settle-
ment to see if the bride is happy and to arrange for an exchange marriage as soon
as possible. The bride must have consented to some degree from the beginning,
because at present no woman can be taken from her home completely against her
will. She should not, however, give the appearance of happily deserting her home
and agnates to live with affines and make their descent line strong, and this
ceremony demonstrates, although at times hypocritically, her proper ambivalence
and hesitation.7
The range of wife-getting procedures, from peaceable exchange to elopement
to, in the past, actual capture, illustrates the complexity of the marriage system as
a whole. We can assemble an impressive amount of evidence for this range of
"proper" marriage and thus eliminate much of the contradiction that first strikes
our attention. If we see the range as depending, in part, on the closeness of the
potential affines - genealogical, geographical, linguistic, and perhaps other
forms of closeness - then we need to know the parameters of each type of marital
option rather than assume that one type of marriage is the best and the others are
by default or due to a breakdown in the system or similar exogenous factor.
Despite idealized statements about the benefits of marrying genealogically close,
advantages and disadvantages to all types of marital strategies exist, whether one
is speaking from a structural or transactional perspective. Contradiction and
conflict are inherent both among people with vested interests in specific mar-
riages and among the various goals individual Tukanoans hope to achieve by
getting married. The marriage data I collected do not allow for a fine enough
breakdown of all the determinant variables to demonstrate which strategies
emerge under which conditions, but the various other kinds of evidence we have
are compelling.
At one extreme, what we might call a "bilateral first cross-cousin model" says
it is preferable to marry this kind of woman, exchanged for another just like her.
Each woman will be moving to her mother's natal longhouse, probably nearby,
and will also be an FZD. The two marriages will be relatively stable and will join
a group of marriages that cement an ongoing alliance between a pair of
hyperaffinal settlements, an alliance spanning at least a generation.
What makes these marriages unlikely, in both behavioral and formal terms, is
that they depend on ideal conditions - and not only demographic ones. When
Tukanoans state that such marriages are the ideal, they are speaking of the way
134
Marriage
things ought to be: the way marriages ought to be and the way social relations
ought to be in general. But reality is not ideal, and Tukanoans are quite pragmatic
- in their myths, their anecdotes, and certainly their actual marital negotiations -
when operating under less than ideal conditions. Thus, they marry (and talk
about marrying) MBDs who are not FZDs (and vice versa), cousins who are only
distantly related or who become cousins only after marriage becomes a possibili-
ty, or women from a "third" group who are not the traditional affines, either at
the settlement or language group level. This "more distant" kind of marital
strategy includes a wide range of marriage types, which we can lump together
and term a "portfolio strategy model" because it contains the notion that it is
best not to have all one's eggs, or assets of any kind, in one basket - at least under
less than ideal conditions, which, unfortunately, are the ones under which both
we and Tukanoans usually find ourselves.
The fuzziness and ambiguity that emerge - for example, when discussing the
meaning of the "mother's children" terms, when discussing affines as generous
and yet treacherous, and certainly when trying to understand actual marriages -
are intrinsic to the system. Although we cannot conclude that, even when
demographic conditions permit, a single type of marriage is the ideal (except
under Utopian conditions), we can begin to comprehend Tukanoan marriage as a
consistent and logical system - albeit a more complex one than conversations
with informants about "how we marry" might initially lead us to believe.
Residential exogamy.
Residential exogamy, an automatic result of the rules of language group exogamy
and patrilocal residence, can claim to be a principle in its own right, one followed
even when nonagnates happen to live together. Feelings about this are especially
strong when cross-cousins are raised together (see Goldman, 1963, p. 43).
Residential exogamy is a principle found throughout the Northwest Amazon,8 in
contrast to many societies elsewhere in the Amazon-Orinoco Basin in which
marriages can be residentially endogamous (for example, in the Xingu Basin, in
most of the Panoan and Ge groups, and among the Piaroa and Yanomamo).
The few exceptions to this rule at present in the Vaupes occur in the mission
towns, where members of several language groups have been encouraged to live
together - a revolutionary change in Vaupes social and spatial arrangements. Out
of a sample of 635 marriages, only 19, all of them in mission towns, were
residentially endogamous. In the layout of these towns each language group and
sib occupy distinct territories.
Language group exogamy.

The principle of language group exogamy is the one most perplexing to non-
Tukanoans - missionaries, local Colombians, and anthropologists alike. Despite
135
The fish people
the sweeping changes brought about by acculturation, language exogamy is still

very much the rule among Tukanoans to the south of Cubeo territory (and many
Cubeo do marry into other language groups). Only one exception (and this one
questionable) occurred in my sample of more than 1,000 marriages.9 And, with
some exceptions in the Pira-parana (see Chapter 5), all Tukanoans affirm that
the rule of language group exogamy must be observed.
A two-year union between a Tukano man of the mid-Papuri and a Tukano
woman of Yavarete is the one observed exception in my sample. Informants were
reluctant to discuss this case, at times saying it was not really a marriage at all
and at other times saying that the woman's sib, the low-ranking ahpukeria
("crab"), was not Tukano and spoke a different language.
Tukanoans see the rule of language group exogamy extending far beyond the
Papurf-Vaupes-Tiqui6-Pira-parana region. They are critical of Cubeo inter-
marriages. Bara war stories generally end with the victors going north and
killing off all Cubeos except a brother-sister pair, who then had to marry each
other in order to start the Cubeo people again. This is why Cubeos "marry their
sisters" today.
How long the rule of language group exogamy has been in effect is open to
speculation; Koch-Griinberg reports it (1909-10), but we do not know whether
it was as strongly enforced as it is today. A. R. Wallace (1889/1972) reports
marriage with blood relatives, presumably cross-cousins, and Kirchoff states that
tribal exogamy is practiced by Tukano groups and blood-relative marriage by
Arawak groups. Goldman (1948, p. 780) correctly points out that the two
statements are not necessarily contradictory.
As we have seen, certain same-language groups in the Pira-parana region
intermarry. Other Pird-parana Tukanoans state that this should not happen
because "people should not speak like their cross-cousins." Some Pira-parana
Indians also state that these same-language affinal units spoke different languages
in the past (C. Hugh-Jones, n.d.). The Yukuna-Tanimuka-Letuama-Matapf
groups to the south also have a rule of language exogamy.
The strength of the rule of language group exogamy is more impressive when
we consider the degree of acculturation in the Vaupes, particularly in the Papuri
drainage. Change agents, especially Catholic missionaries, actively discourage
cross-cousin marriage and say the language group exogamy rule is absurd. This
part of the marriage system has remained viable, however, despite extensive
inroads by such agents into many areas of Vaupes culture.
Sorensen (1967, p. 672) notes that Tukanoans assume that whites also practice
.language group exogamy. When talking with me, Tukanoans always assumed that
my mother and father spoke different languages, and other anthropologists who
have worked in the Vaupes report similar experiences.
During my stay, a Tukano man and woman from the mission town of Acaricuara
announced that they intended to get married, thereby provoking a huge scandal.
Both individuals were highly missionized (for example, both were catequistas),
136
Marriage
and the priest actively encouraged the marriage. He argued that they had a right
to marry each other, but he also supported it because it provided a chance to drive
a wedge into the system of language group exogamy. The proposed match would
have violated the generally observed rule of residential exogamy as well, but it
was the fact that two Tukanos were contemplating getting married that was the
most reprehensible to Tukanoans.
Ironically, both the young man and woman had the same Spanish surname; this
was pure happenstance, for they came from different sibs. Still, Tukanoans,
especially those in Acaricuara, seized on this point to bolster their argument,
saying that the marriage would violate Catholic incest prohibitions as well.
I did not find out how the situation was resolved before I left the area. The
families of both were horrified, and rumors circulated of violent quarrels and
threats made with shotguns.
If, after I departed, the couple was in fact married in the church, then the
missionaries had indeed made a significant step in dismantling the traditional
marriage system. Such a marriage would have meant that the Tukanoans' many
efforts to thwart it were unsuccessful and that the church had triumphed over
traditional values. Furthermore, because the marriage involved a man and a
woman from respected mission-town families belonging to high-ranking sibs and
the young man was the corregidor of Acaricuara, the flaunting of traditional
mores perhaps demonstrated to Tukanoans the encroaching power of the Colom-
bian government as well as the church.
Alliance.
"Alliance" simply means that two affinally related local descent groups consider
it advantageous to continue to exchange women over time. Alliance is a long-
term strategy and as such differs from the more limited, direct exchange strategy,
although the two are obviously related. The latter is concerned more with
ensuring the replacement of women of the local descent group who are lost
through marriage.
Lounsbury (1962, p. 1307) discusses two general types of alliance strategies.
The first is concerned with the continuation and reaffirmation of an already
established alliance that continues to be seen as advantageous to both sides. As
noted previously, some of these advantages in the Vaupes might be geographical
proximity, close kinship ties, facilitation of the ceremonial and economic
exchanges characterizing most pairs of neighboring longhouses, and increased
benefits to the women who have already married into that particular longhouse.
The second type of marital strategy is establishing new alliances that promise
economic, political, military, or other advantages. In the Vaup6s, alliance with
more distant longhouses might be made for such reasons as being assured of
hospitality on various river routes or in certain mission towns or having one's
close affines dispersed in order to have a wider range of individuals in one's
137
The fish people
affinal and neighbor relationships. Furthermore, although raiding and feuding

have ceased, in the past, military considerations were very much an aspect of
marital strategies. Distant marriages being made today may be continuations of
alliances formed originally for the purpose of having dispersed military allies.
The influence of geographical distance is a crucial factor in comprehending the
way in which principles of alliance actually influence decision making about
marriage. For example, if a settlement considers itself owed a woman, its
demands will not have much impact if it is located far away from the debtor
settlement. This distance is also probably an important contibuting factor in the
debtor settlement's continued disregard of the other's claims.
The distance factor also attests to the importance of analyzing the total sample
of marriages made by a local descent group over a period of time, such as a
generation. For Tukanoans the meaning of distance between marriage partners'
settlements is not simply one of a direct correlation between increasing distance
and increasing undesirability of a prospective marriage alliance. For example,
given that a specific settlement has married several of its women into one or two
nearby settlements, it may be less "costly" from its point of view to make the
next marriages with settlements relatively far away. This is, however, still proba-
bly a less desirable type of marriage from the point of view of the marrying
individuals.
It is evident that settlements in the Vaupes do not marry within a tightly
constricted geographical field. Looking at the marriages made by any given
settlement over two or three generations, we see that some of the longhouse
groups exchanging women are quite far apart. Furthermore, many of the "poten-
tial affine" settlements (not counting the settlements that cannot be sources of
spouses due to prohibitions imposed by the social structure) are closer than some
of those that actually have supplied spouses (Figure 6 and Map 3 give an example
of this). Yet it is also the case that some marriages are between local descent
groups so far away from each other that they are statistically very improbable and
are seen as such by Tukanoans.
Measuring geographical distance between marriage partners' settlements at the
time of marriage for a sample of 635 marriages gives a mean linear distance of
(very roughly) 22 miles, with a standard deviation of 19.8 miles. Some marriages
occur between settlements as distant as 90 linear miles. The actual distances are
greater, for the river and trail routes used by Tukanoans when traveling between
the settlements must be taken into consideration.l0
Marriage behavior
All Tukanoans possess a remarkable knowledge of the settlements of the region
and their inhabitants, even those they do not personally know. Most know a lot
about many quite distant settlements, because a favorite topic of conversation is
fellow Tukanoans.
138
Wafoperi (Desana)
^ _ _ _ _ - ^ ^ ^ - — Siriria (Desana)
Hchpirotudu
(TUyuka)
Yehpupuna • EreyS (Tuyuka)

Ahpuratudu 1 & 2
(Tliyuka)
(T\iyuka)
Figure 6. Marriages during three generations at Pumanaka buro. Total marriages: 27 (13
women, 14 men); total number of settlements involved: 8. The solid lines represent men
bringing women to Pumanaka buro; broken lines represent women marrying out.
The preferred method of obtaining a spouse is slowly to develop an understand-

ing with a cross-cousin from a familiar settlement. Another courtship pattern is
for a group of young men to go traveling together to visit relatives in settlements
where there are marriageable women. One such group came to Pumanaka buro
during my stay. They remained a week, obviously enjoying themselves, and
provided excitement and a change of pace. The group consisted of two Tuyuka
and two Siriano men from fairly distant settlements. No one ever mentioned the
ostensible reason for their visit. This is understandable, however, because to refer
explicitly to a visitor's interest in marrying one of the settlement's women is to
impugn his good intentions. These men made baskets for their hostesses, as is the
custom in this type of visit. No marriages came of this particular expedition, at
least during my stay.
On another occasion a woman, Fernanda, was brought to Pumanaka buro as
a prospective wife for the unfortunate-in-love Estribino, the headman's eldest
son. She did not take to Estribino but did end up marrying Nazalio, the second
son of Manuel. This not only upset Estribino personally but increased the already
well-established rift between the two families. It also upset Nazalio's older
brother, Mariquino, who ideally should have married before Nazalio but like
Estribino had had problems getting a wife. The woman whom Mariquino was
most interested in was Gabriela, a Tuyuka from a downstream longhouse. She
had gone to mission school and spoke Spanish. For a while she seemed disin-
clined to live with such backwoods Indians and took up with a cabuco near
Mini." But by the time I left, she had finally settled down with Mariquino. The
139
Vaupes River
o 4,
Siriria
Map 3. Locations of settlements intermarrying with Pdmanaka buro. The Bara settlement of Pumanaka buro is located at 1, and Tuyuka
settlements of Yehpupuna, Ahpuratiidu and HehpSrotiidu are located at 2. 3. and 4. respectively (Jvlot to scaled
Marriage
first baby born to the succeeding generation was, in fact, Nazalio and Fer-
nanda's.
Couples in mission towns are expected to marry in the church before the
woman changes residence, but this seldom happens. In the past, priests periodi-
cally made trips to settlements to baptize and perform marriage ceremonies, but
at present they encourage the couples to have the religious rites performed during
a trip to a mission town. A substantial number have done so, but the idea of being
married in the church before starting to live together is still incomprehensible to
Tukanoans. They are not concerned about the virginity of either spouse and know
how brittle many marriages are during the first years.
Those marriages at Pumanaka buro that had lasted a number of years seemed
very solid. I never saw a serious open quarrel between a husband and wife;
quarrels certainly occurred, but, like all other truly private interaction, they took
place well away from the longhouse. Couples joke about their quarrels. By the
time I left I was aware of one or two long-standing points of friction in each
marriage.
One of the closest and most affectionate marriages was between a Bara man
and a Cubeo woman. They claimed they had not understood each other's
languages at all when they married. This is one of the rare instances when two
multilingual Tukanoans do not have a language in common. Being Cubeo, she
did not, like almost all other Tukanoans, speak the lingua franca, Tukano
(although many Cubeo do speak it). She said it took her two years to learn Bara.
In public, spouses almost never physically show affection, not even touching
one another. Physical contact occurs only when joking or during the few public
gestures of intimacy, either grooming or administering medication of some sort.
A wife will apply body paint to her husband (although he does not reciprocate);
she will also delouse him. Men do not delouse each other but women delouse one
another and their children. (Both men and women take extraordinarily good care
of their dogs, keeping them fairly free of fleas and niguas, "chiggers.") Other
intimate public gestures are related to health: picking scabs, removing thorns and
niguas, and applying stinging nettle to muscular pain. Otherwise, in public both
men and women touch children and members of their own sex far more than
members of the opposite sex. The two couples who married during my stay (not
counting the week-long union between Estribino and Isabel) were more openly
amorous but in a very lighthearted, teasing manner.
The lack of public physical contact between spouses is in keeping with the
general pattern of sexual segregation. There is also a general tendency to play
down any exclusive or highly charged emotional relationship that might conflict
with the interests of the community as a whole. All married partners must stay in
good humor and be polite to each other and the rest of the longhouse, regardless
of any strains the marriage is temporarily experiencing. A general muting of all
displays of sexual attraction when things are going well makes it easier to
maintain equilibrium and friendliness when they are not. An indication of the
141
The fish people
soft-pedaling of the emotional closeness between spouses is the lack of publicly
used terms of address. One can refer to yti manti ("my husband") or yii numo
("my wife"), but direct forms of address consist only of a spouse's Spanish
name, or, after the birth of a child, teknonymy. Frequently, the child's Spanish
name is used.
This general stricture also helps newlyweds overcome the initial difficulties of
adjusting to each other. The couple can ease into deepening love and intimacy
rather than having to be demonstrative immediately after marriage. It is difficult
for the new wife to adjust to the change of surroundings and associations, the
absence of familiar loved ones, and the new longhouse language. Although the
husband's life changes very little, it is an awkward time for him as well. Should
his wife and his mother disagree over something, and they almost inevitably will,
the husband is equally inevitably caught squarely in the middle.
Thus, even when husband and wife have known each other a long time before
starting to live together, the period right after marriage is stressful. The wife must
not only adjust to her husband and her new surroundings but to her new
mother-in-law and sisters-in-law as well, for she will be spending most of her
time with them. When young women are contemplating getting married, they
consider their future relationships with these women very carefully. Competing
claims on loyalty, time, services, and other resources create tension between
coresident in-laws as does the fact that affines are always outsiders and agnates
are always one's "own group."
Still, some of the rivalry, jealousy, and complications found elsewhere - for
example, at times in American families - are not so marked in the Vaupes. For
one thing, the segregation between the sexes is extensive, which reduces the
claim either mother or wife has on a son's/husband's time, attention, or displays
of loyalty and affection. It is difficult unwittingly to show favoritism in the
Vaup6s, whereas in some other societies a man can be more easily manipulated
in behind-the-scenes power struggles between his mother and wife or wives.
Furthermore, female solidarity is strong in Tukanoan life. Women enjoy one
another's company, and members of the same sex enjoy greater ease and familiar-
ity among themselves than with members of the opposite sex. This helps a young
wife adjust to her female in-laws.
When problems do surface, the new bride finds herself more powerless than
she will ever be again, but not as powerless as she would be in some other
strongly patrilineal societies or in some peasant communities. For example, a
Vaupds wife can return home knowing she will not be turned away, and this is a
threat she can use to her advantage in her new home. She also can make her
presence felt by threatening to commit adultery or by engaging in other disruptive
behavior. A new bride's mother-in-law does not have as much authority over her
as in many other societies (e.g., Northern India), and although a husband is
supposed to listen to his mother's advice, everyone is expected to compromise a
bit. If a mother has difficulty liking her son's wife even after a substantial amount
142
Marriage
of time, the general courtesy and good humor demanded of all who live together
will prevail in daily life, at least on the surface.
Mythological references to marriage behavior and expectations.

As is apparent from the myths already recounted, affinal relations are a vital part
of Tukanoan mythology. As indicated in Chapter 6, a basic theme is the uncer-
tainty and risk inhering in any dealings with affines. But sometimes the message
is that one only truly becomes human upon establishing affinal ties. At times the
advantages and risks are dealt with at almost the same point in the narrative,
producing a counterpoint effect. Frequently given advantages include being able
to perpetuate one's own people, reciprocal exchange of services and goods, and
sexual activity. For example, Yeba mahti receives three quintessentially human
gifts from Wai Pino Mahko (Fish Anaconda Daughter), his wife: manioc, coca,
and banisteriopsis. Other benefits are obtained from interactions with in-laws; for
example, Wai Pino offers Yeba Mahu his hospitality, feeds him Tarira fish
(Wai Pino's own agnates), and protects Yeba Mahu when he disobeys instruc-
tions and is about to be strangled by the anaconda skin.
This myth is a "charter" for the ceremonial food exchange (dabucuri) found
in the Vaupe's. In a part of this myth, recorded by S. Hugh-Jones (personal
communication), the prototypical exchange relation between father-in-law and
son-in-law is laid out by Wai Pin6 giving (via his daughter) manioc and all
cultivated crops to Yeba Mahu in exchange for meat. A present-day dabucuri
consists of the hosts offering beer and guests offering fish or meat (see Chapter
4). These exchanges imply (and symbolize) exchanges between men (father-in-law
and sons-in-law) for women and exchange between men and women.
It is important to note that although affines are "other people" and potentially
treacherous, one receives many benefits from them unavailable from one's "own
people," and this dilemma is expressed over and over in myths. The primary
benefit, of course, is daughters-in-law in exchange for daughters. Marital
exchanges and the mutual ongoing benefits of ceremonial interactions such as
dabucuri that accompany them either cannot or in fact generally do not occur
between agnates. Another example is the gifts of baskets of various sorts (between
brothers-in-law and from a visiting potential husband to his prospective mother-
in-law), paralleling the gifts of baskets from husband to wife in exchange for
manioc products.
Expectations about proper relationships of exchange between affines in fact
are found in most myths. For example, in a myth about "jaguar people"
(yaiwa mahd), the grandmother of Diro koa pona (literally, "piece-of-meat-bone-
offspring") is ashamed and angry when she becomes afraid that her grandsons
have not provided enough fish as a gift to her relatives, the jaguar people. But the
Diro koa pona actually have acted in good faith, and it is the jaguar people
who are treacherous (by nature, is the message) and who are destroyed in the end.
143
The fish people
In the myth of di wdnako (a type of long horn), a wahti shows his gratitude to
his human wife's people by supplying them with a huge amount of food, by
warning them of an upcoming raid from their affinal enemies and ultimately by
sacrificing himself in the war effort. His body is burnt and the plants growing
from the ashes are the source of a protective unguent and of powerful horns (di
wanako) that when played render the enemy helpless.
Ambiguity and unpredictability appear often in myths dealing with affinal
relations. I have already mentioned the built-in ambiguity of people who stand in
an "affine of my affines" relationship with each other - who call each other
pahko-mahku/-6 ("mother's children"). In the Barasana myth recorded by S.
Hugh-Jones mentioned previously, two men who are cohusbands of the same
woman call each other pahko-mahku. The potential for competition, jealousy,
and suspicion is obvious in the myth. The children in the Namakuru myth are also
jealous of their pahk6-mahku but for different reasons; however, the situation is
an identical one of competition for affection and favors. The crucial lesson is
concerned with the potential jealousy inherent in the relationship, even if it is not
always present.
Of course another source of ambiguity in the pahko-pona relationship arises
when we recall the phrase found in many Dravidian systems, that "affines of my
affines are (like) my brothers." The pahko-pona relationship is truly ambiguous;
denoting neither agnate nor affine, it is a semantic lightning rod for all sorts of
interactions and emotions containing ambiguous elements. We can, thus, inter-
pret negative statements about pahko-pona at times as referring to the tensions
and antagonisms that crop up between affines (such as cohusbands) but at other-
times as indirectly characterizing the conflicts that occur between brothers, who
at times compete for women, prestige, and other scarce resources but who are not
supposed to ever let these tensions become publicly known. Obviously such
tensions sometimes do get out of hand and this indeed is expressed in many
myths. But pahko-pona lack the built-in affection, mutual dependence, and
solidarity that usually exist between agnates, especially coresident ones, and thus
when serving as metonym for brothers the term allows for a stark presentation of
competition and jealousy.
A Tukanoan will classify any Indians who live far away as pahko-mahkara
("mother's children people"), provided they fall within the category of maha
("people"). This merely indicates that they are "people like us" who live far
away and with whom there is no ongoing interaction. In this context the set of
terms expresses relative affective distance with respect to trustworthiness, loyalty,
and commitment (or the lack of these). It also expresses relative cultural distance
or the notion that the farther away people live, the more likely they are to be
different - to do things differently, to speak languages extremely different from
the ones one knows, and so on. The terms also express relative social distance,
indicating the degree to which one can expect regularized reciprocity, military
aid, absence of conflicts of interest, and so forth. Thus, if in a given context a
144
Marriage
Tukanoan wants to indicate an ambiguous, if not suspicious, socially distant type

of relationship, the terms pahko-mahku/-6 are selected.
As indicated, many myths illustrate the relatively great social and psychologi-
cal distance and the ambiguity implied by these terms. An individual's pahk6-
mahku is a person rather than a wahtf. Still, almost always in myths one's
pahko-pona play tricks. A man's pahko-pona are not necessarily his antago-
nists but rather people who cheat him or do him harm if there are benefits to be
gained from such behavior. Reciprocal pahko-pona are either indifferent to each
other's well-being, or when convenient, dangerous. Myths dealing with the
relationship between pahko-pon£ do not illustrate failure to live up to expecta-
tions so much as potential trickery and treachery brought about by opportunism
and a lack of obligations. As we have seen in the Namakuru myth and the one
about cohusbands of the same wife, however, at times the lack of intimacy and
trust between pahko-pona is not caused merely by indifference and ambivalence
but by conflict of interest and open competition.
These themes of jealousy and competition also appear in the pehu story: the
two co-wives of the same husband are structurally identical to the two men who
are cohusbands and call each other pahk6-mahku. Another illustration of this
motif of conflict between pahk6-pond is provided in the myth about Koa-
mahkii, when he pretends to be the pahko-mahkii of Ana (poisonous snakes)
in order to slay them more easily.
Although great benefits are gained when expectations are met, ambiguity and
risk inhere in all affinal relations. Some of the danger and unpredictability
deriving from such relationships are due to the fact that affines are always
different, "not-us," "other." The story of Wai Pin6 Mahko, Fish Anaconda
Daughter, illustrates this; she is not a human and her father is an anaconda. One
danger of associating with people not "like us" is, of course, becoming too
much like them. The mother in the Namakuru story is depicted as becoming
progressively more like her wahti husband: One breast becomes excessively
large, she learns to speak his language, and she gives birth to an animal. In the
myth of di wdnako, which involves a little girl locked out of the longhouse by
angry parents, a wahtf carries the girl off, adopts and rears her, and eventually
takes her as his wife. When she returns to her parents, she is described as having
become less human in certain ways, a crucial one is having an enlarged breast.
A similar risk is the possibility of starting to like one's affines more than one's
agnates. This is expressed in the Namakuru story when the mother excessively
punishes her children and sends them away in order to conceal her attachment to
her new husband and baby.
In a sense, women are always suspected of loving the group they marry into too
much. The grandmother in the myth about jaguar people is torn between concern
for her grandchildren, Dir6 koa pona, and her agnates, the jaguar people; the
idea is that it is she who has brought them together, and it is her loyalty that is
compromised.
145
The fish people
Women in the myths are shown at times as not only being pivotal and having
divided loyalties but also as betraying either agnates or affines out of selfish
reasons. In two myths, women are punished for this by being made unknowingly
to eat their lovers' genitals.
In some myths, particularly war stories, women are portrayed as more loyal to
their agnatic kin than to their affines. In one, not only are the women loyal but
they arm themselves and fight alongside their male agnates. In another war story
an inmarried Tuyuka woman overhears her Carihona affines planning to hold a
feast and murder their Tuyuka enemies by poisoning the manioc beer, and she
warns her agnates of their treachery.
Thus, danger inheres in affinal relations not only because affines are "different
from us," but also because real conflicts of interest arise. In this case, it is not so
much a question of differences between " u s " and "them" as one of similarities:
similarities in wanting the same scarce resources, in trying to make a good
bargain, and in mutual distrust and suspicion.
The themes of greed and excess frequently appear in myths about affinal
relations. In several myths Koa-mahkii is depicted as wanting two women at
the same time, with no offer of an exchange for even one of them. In one, this
leads to difficulties with Wahobiro, the women's father, who leaves Koa-
mahku abandoned in a tree.
A frequently encountered progression in a myth's plot begins with a rather
innocuous misunderstanding or misdemeanor that sets up a reaction that finally
plunges everything into chaos. One war story begins with a man getting angry
and beating his wife, who returns to her natal longhouse. Eventually, the two
local groups try to kill each other.
Affines are also dangerous because of the risks of sexuality itself. At times this
is connected to the dangers of excess and at times to the theme of inherent
differences between own group and affines and the consequent danger of all
sexual congress with them. Koa-mahkii has to comb out all of the stinging and
biting vermin from the pubic hair of Nima Pino's (Curare Anaconda) daughter
before having intercourse with her, an obvious vagina dentata motif.
Excesses, inherent differences, conflicts of interest, and other types of dangers
encountered in interactions with affines will lead to total disaster if not corrected.
For when affines are killed, chaos and animal-like behavior threaten, because one
clearly needs affines to continue being human. Animals have sexual intercourse
with anybody, as do Maku or any people who "marry their sisters." Thus, myths
demonstrate the necessity of maintaining proper affinal relationships if one wants
to continue one's agnatic line. When things go awry and get progressively out of
kilter, tragedy is the result. The theme of ever-increasing chaos also occurs in
narratives about other kinds of relatives. The protagonists in the story of boraro
skin ate their own relatives. Kamaweni is killed by his own brother after a series
of excesses. The relatives of the protagonist in the story of kahp6a orero turn
into monkeys after they lose their eyes and become burdens. Several myths end
146
Marriage
with such total annihilation that brother must marry sister in order to continue
"the people."
Conclusions
Kinship and marriage roles are extremely important components of Tukanoan
social categories and groups. Local descent groups and the individuals compris-
ing them care about their reputations as honorable and trustworthy people, and
their relations with agnates and affines form a major segment of this reputation.
Unlike other simple marriage systems, such as those found among the Ge or
Australian Aborigines, group membership in the Vaupes does not determine
affinal relations to the extent of specifying the only group or category one may
marry into. This is true despite the basically Dravidian kinship terminology found
in many (perhaps all) Tukanoan languages. Adjustments have been made in at
least several Tukanoan languages that fit the terminology more closely to the
reality - a regionally based system in which more than sixteen exogamous units
exchange women. Thus such structural principles as the kinship terminology and
the patrilineal language groups and phratries only set constraints on Vaupes
marriage making. An acceptable model of marriage for the Vaupes, therefore, is
one based on network formation as the outcome of at least three sets of factors:
(1) environmental constraints, (2) social structural principles, and (3) shared
decision-making rules for choosing among the alternatives offered by the first two
sets.12
147
8.
Tukanoans and Maku
It should be apparent by now that the kinds of interaction occurring between

Tukanoans and Maku and the categorizations each group makes of the other are
very important components of identity. In this chapter, I will deal far more with
Tukanoan perceptions of Maku than vice versa, however, because I had very
little direct contact with Maku.'
Tukanoans stand in a superordinate position to Maku and use the differences
between their respective subsistence modes, principles of social structure, and
other areas of culture to explain and justify these claims of superiority. Tukanoans
are more sedentary than Maku, practice more horticulture, and are more ori-
ented toward the rivers and fishing. These are observable differences; when asked
to make comparisons, Tukanoans will also list others, which are more open to
question. It should be noted at the outset that although Tukanoans will usually
describe differences between themselves and Maku in terms of absolutes, in
most of the areas of differentiation (for example, horticultural practices, mobility,
property owning, or use of hunted and gathered food), the contrasts are ones of
degree rather than category.
The image each group holds of the other embodies key elements of its general
conceptualization of the world and humanity. These characteristics at times vary
from the flesh-and-blood reality of the other group, and we shall see that they are
systematic and meaningful discrepancies.
My own contacts with Maku were infrequent and superficial. Thus, as noted,
this chapter is biased: Most of the data on Tukanoans are from my own fieldwork,
whereas most of the data on Maku are from published reports and communica-
tions with investigators who have had firsthand experience.
Background to the Maku

Simply put, Maku differ from Tukanoans in that they are more sylvan than
riverine, they speak non-Tukanoan languages, they can marry within the linguis-
tic unit and have a stronger constraint toward regional endogamy, and they enter
into various kinds of "symbiotic" relationships with Tukanoans in which they
148
Tukanoans and Maku
play a subordinate role. The latter can be relatively temporary, consisting of
one- or two-day events in which meat, labor, or various forest products are
exchanged for cultivated goods and white trade items. Or they can be long-term
servant-master types of relationships occurring between a Maku family or larger
group and a specific Tukanoan settlement.
The term Maku is generic (Metraux, 1948; Silverwood-Cope, 1972), and like
several other terms (e.g., "Guaharibo") can at times designate nothing more
precise than the "wild" Indians of a given region - those least contacted, least
clothed, and who are alleged to practice no farming, have no houses, and lead a
totally nomadic existence. This is the meaning in the Vaupes behind most local
Colombians' use of the term and is at times the meaning intended when Tukanoans
speak of Maku.
The Maku of the Central Northwest Amazon are divided into three linguistic
groups: (1) Cacua, also known as Bara Maku (Silverwood-Cope, 1972); (2)
Jupda, also known as Ubde (Giacone, 1955) or Hupdti (Reid, personal commu-
nication); and (3) Yohop (Silverwood-Cope, 1972), also known as Yiihup (Reid,
personal communication).2
The origin of the Maku, like the origins of Tukanoans, is problematical.
Goldman (1963, p. 14) makes a case for a relatively early Tupian influence in the
area with later Carib and Arawak contacts, with no specific mention of Maku
origins in either his 1948 or 1963 publications. In the Handbook of South
American Indians, Metraux (1948, p. 865) gives a scheme in which the Maku
and similar groups are "generally considered to be the last representatives of an
ancient people who occupied vast areas of the Amazon Basin before they were
exterminated or assimilated by the Carib, Arawak, and Tucano, the carriers of a
more advanced culture based on farming. " 3
Still, more recent ethnographic and ethnological work suggests it is a futile
exercise to use an extremely rigid and genetic-based classification of lowland
South American linguistic and tribal groups for either historical reconstruction or
conclusions about present-day interbreeding populations.4 Many groups that at
present substantially depend on foraging may have devolved from earlier, more
complex systems. Evidence exists that segments of foraging groups like the
Maku are periodically assimilated into a local group of one of their horti-
culturalist neighbors. In the case of the Maku, they, as the assimilated segment,
take on the status of the lowest-ranked clan of the horticulturalist group (Goldman,
1963, p. 100; Koch-Griinberg, 1906; Silverwood-Cope, 1972). In sum, where
the Maku came from is not known, the degree to which they represent a
distinguishable breeding population both now and in the past is uncertain
(although some ethnographic accounts describe physical differences between
Maku and neighboring groups),5 and it has not been conclusively established
whether they have devolved from a previously more complex state or have always
been more marginal than their neighbors.
Indeed, a crucial question is to what degree do groups such as Maku maintain
149
The fish people
their identity and distinctiveness - in language, subsistence mode and other

economic activities, or the display of certain stereotypical characteristics that set
them apart from their neighbors - in response to and for the purpose of maintain-
ing this symbiotic and unequal interaction? It is possible that some of these less
complex groups would have been assimilated long ago, were it not for ongoing
pressures to maintain distinctiveness. Thus, their current status may not be due to
their historically different origins so much as due to the structure of their
interaction with other groups.
In contrast to the riverine Tukanoans, Makii inhabit the interfluvial areas of
the region. Maku lack the highly developed Tukanoan technology surrounding
fishing and river transportation. For example, they make no canoes, whereas
Tukanoans excel at this craft. Maku do fish, however, and they practice a fishing
technique - for catching eels - unknown to the neighboring Desana (Silverwood-
Cope, 1972, p. 86).
Makti practice horticulture at present, and it is not certain whether they ever
lived entirely by hunting and gathering. It is questionable whether a strict hunting
and gathering life is possible in many areas of lowland South America.6 Further-
more, no good evidence exists that the Maku were ever totally nomadic.
Nevertheless, their houses are far more makeshift than the elaborate traditional
Tukanoan longhouse, and the foods that Maku do cultivate apparently do not
supply them with all they need and want. In some cases specific crops are not
grown at all but are obtained entirely from Tukanoans. In other cases not enough
of an item is planted to last through the year. Makii make extended treks into the
jungle on hunting expeditions more frequently than Tukanoans. According to
Silverwood-Cope (1972, p. 103), Maku actually gather less than Tukanoans.
Silverwood-Cope's description of the Maku as "professional hunters" who
specialize in acquiring game to be used for exchanges with Tukanoans is very
much to the point (1972, p. 103).
Maku can marry within the linguistic unit and generally prefer to marry within
a circumscribed region, a preference that at times takes precedence over marrying
a person in the proper kin category; according to Silverwood-Cope, Maku
explained this by saying that Maku in other regions were dangerous sorcerers
and jealous with their women (1972, p. 176). These "improper" marriages are
usually between individuals in lineages that consider each other agnates.7 H.
Reid (personal communication) notes that marriages do occur between members
of different regional groups and that no specific rule proscribes regional exoga-
my. Tukanoans also occasionally make improper marriages, but the rule of
language exogamy is much more stringently observed by them; exceptions are
exceedingly rare. Thus, being ignorant of Maku lineage or clan exogamic rules,
they are shocked that Maku "marry their sisters" so often. Silverwood-Cope
quotes a Uanano who specifically compares Makii marriage to animal marriage,
in that both are randomly incestuous (1972, p. 176).
Other elements of Maku social organization include patrilineal exogamous
150
Tukanoans and Maku
clans, a Dravidian kinship terminology, a preference for classificatory sister
exchange, and marriage into a class of relatives including one's bilateral cross-
cousin (Silverwood-Cope, 1972, p. 175).
In the areas of ideology and belief system, Maku have an elaborate conceptu-
alization of the cosmos and human beings' place in it. In the area of food
restrictions and taboos, Maku appear to have a far less stringent set of require-
ments, if one compares the information available in Silverwood-Cope (1972)
with information on Tukanoans.8
Tukanoan attitudes toward the Maku

The Tukanoans who supplied most of the primary information on the Maku (i.e.,
the Bara and Tuyuka of the Inambu River) considered the Maku to be not
entirely "people." Classifications of this sort are common among ethnic groups
that do not share the same level of economic and political complexity yet that
regularly interact with each other (see Barth, 1969; Lee and DeVore, 1968). The
origin of the Maku is not a part of any Bara origin myths I recorded during
fieldwork, nor is it in any of the Barasana myths recorded by S. and C.
Hugh-Jones (personal communication). Giacone reports that the Tukano, Tariana,
and Desana consider the Maku not as people but as "jaguar children" (1949, p.
88). This general subhuman classification frequently appears in reports from
travelers as well (Koch-Griinberg, 1906; MacCreagh, 1926; McGovern, 1927;
and A. R. Wallace, 1889/1972). Reichel-Dolmatoff (1971), however, states that
the Desana do include Maku in their origin myths. Maku definitely appear in
Bard and other Tukanoan myths; their origin, however, is not that of "true
people," that is, emergence from a rapids site or upstream travel in an anaconda
canoe. Evidence for Maku inferiority is also indicated in the Tukanoan state-
ments that the Maku do not farm, do not fish or travel by river, and build no
houses.
Maku were once described to me as having no duhptia kai (a Bara word
glossing as "head-spirit" or perhaps "consciousness"). In this context it proba-
bly refers to Maku lack of shame about "improper" behavior. Maku are said to
observe no food taboos, eating snakes, sloths, rats, and vultures. Maku are also
"shameless" because they "marry their sisters." A disparaging way for Bara to
call attention to Maku orientation toward the forest rather than the river is by
saying Maku urinate in the river. According to Goldman (1963, p. 91), a Cubeo
origin myth that describes a certain sib's arrival at a river via an overland route
stigmatizes this sib, because this mode of travel is considered typical of Maku.
In fact, in the Vaupes any comparative statement about the greater forest
orientation of a given group with respect to another is most often a statement
about relative inferiority of the former to the latter. In this sense, Maku are to be
seen as at the end of a continuum. For example, McGovern (1927, p. 227) makes
the clearly deprecatory statement that the Desana of the Papurf are more forest-
151
The fish people
oriented than are neighboring Tukano and Waikana (Piratapuya?). He describes

all of the Bara of the upper Tiquie in a similar fashion, contrasting them with
their Tukano and Tuyuka neighbors farther downstream. Such a distinction is also
made between high- and low-ranking sibs within a particular language group.
Low-ranking sibs will occupy headwater locations, resulting in a relatively
greater dependence on game than on fish (Goldman, 1963, p. 49). The degree to
which a group classifies itself as river-oriented and is disdainful of the forest is
probably in part a function of whether it has other groups supplying it with game.
For example, Silverwood-Cope describes the Desana as afraid of the forest,
particularly of hunting in it at night (1972, p. 44). For the Desana, forest spirits
are far more malignant than those of the rivers. In fear of them, Desana lock up
their houses at dusk. Silverwood-Cope also notes that at the time of his research,
Desana did virtually no hunting, told him they were fishermen, and that hunting
in the forest was for Maku (1972, p. 44). 9
The Bara I lived with certainly hunted, often at night. They also certainly
leave the trails when hunting, something the Desana that Silverwood-Cope
observed are reluctant to do. It may be said that the Bara, because they are
located on upstream tributaries more than Desana, and because they have no
Maku to bring them game, stress hunting more. Still, Bara, like Desana, do
associate many dangers with hunting, do observe taboos in connection with
hunting expeditions and weapons, and certainly know of many dangerous spirits
inhabiting the forests.
At times the low-ranking sibs of a given Tukanoan group will be described as
being servants to the higher-ranking sibs - hunting for game and carrying out
other tasks in a manner parallel to Maku activities. For example, McGovern
(1927, pp. 208-09) states that the Waikano consider Maku to be too inferior, and
had turned their own lowest-ranked sib into Maku-like servants.
Tukanoans will also describe Maku as wearing no clothes. Clothing, body
paint, and jewelry all confer human identity; thus, it is to be expected that
derogatory descriptions of Maku will include statements that they wear no
clothing or body adornment. In general, Maku do decorate their bodies less than
Tukanoans (Goldman, 1963, p. 153; Silverwood-Cope, 1972).
The association of cannibalism with Maku is mentioned in the literature
(Goldman, 1963; Koch-Griinberg, 1909-10; Reichel-Dolmatoff, 1971, p. 260).
This theme, especially one concerned with an organism consuming itself, is
recurrent in Maku mythology (Silverwood-Cope, 1972), but otherwise no spe-
cial cannibalistic practices have actually been reported for them. A Bara infor-
mant told me that Maku were people who hunted and ate human beings simply
for food. This remark was derogatory not only in its reference to cannibalism but
also in its implication that Maku were not able to secure enough food by other
means.
In Tukanoan opinion, in addition to their unwillingness to observe basic rules
of behavior, Maku are stupid. A country bumpkin flavor comes out in such
152
Tukanoans and Maku
descriptions. Most frequently the inability of Maku to feed themselves is a
theme of a joke. I was told a probably apocryphal tale about some Maku who
arrived at a Tukanoan longhouse at night so crazed with hunger that they wolfed
down raw manioc mash and subsequently died from the prussic acid it contained.
A picture in a magazine I had of starving Biafrans never failed to produce howls
of laughter from the Bara" I lived with. I was asked if they were "our" (i.e.,
Americans') Maku, because like Maku, the Biafrans obviously could not
adequately feed themselves. That Maku dogs starve to death was given as
another piece of evidence that Maku are too stupid to take care of themselves
without periodic assistance from Tukanoans.
Occasionally I observed Tukanoans poking fun at Maku who happened to be
present, but this was done behind their back. One Maku man who wore an old
pair of plastic shoes occasioned much giggling, as did the asymmetry of an old
woman's breasts. (In several myths, having only one breast or asymmetrical
breasts is a clue that an otherwise normal-appearing woman is really a wahti.)
A final point made by Bara and other Tukanoans about Maku inferiority is
their lack of knowledge about ceremonial lore and behavior. Bara say that Maku
adolescents are not initiated (this is not supported by Silverwood-Cope's observa-
tions). Nor are they said to have the other ceremonies common to Tukanoan
groups of the region. Makii are said neither to sing nor to dance (again, not
supported by Silverwood-Cope; however, during festivals held by Tukanoans, any
Maku present remain in the background). Maku are also said to have special
knowledge about sorcery (Goldman, 1963, p. 107; Reichel-Dolmatoff, 1971, p.
260), which, some authors say, gives Tukanoans an incentive to deal fairly when
trading with them.10 Still, Silverwood-Cope states that Makii are afraid of
sorcery practiced by experienced Desana shamans (1972, p. 98). He believes that
any current Tukanoan reliance on Maku shamans is probably the result of the
relatively less acculturation of Makti than some local groups of Tukanoans,
who, as a consequence, have lost their shamans. Koch-Griinberg states that
Maku were used as scapegoats; when anything went wrong, Maku sorcery was
blamed (1909-10, p. 270). Whiffen concurs, calling them the "proverbial cat"
in always taking the blame (1915, p. 70).
Bara informants told me that Maku could make powerful poisons (a Maku
specialization is the manufacture of curare) and that some of the older Maku had
special knowledge in making trance-producing substances. I was told that the
snuff taken by shamans of the various Tukanoan language groups was available
only from Maku and was a trade item. This snuff, Virola sp. (Schultes, 1972),
was said to be more powerful than banisteriopsis, the hallucinogen taken by all
initiated men during festivals, and hence only shamans took it.
153
The fish people
Interaction between Tukanoans and Maku
Maku and Tukanoan interaction is most often described in the literature (and was
described to me by Bar£ informants) as a servant-master relationship. What this
in fact consists of - what it means in terms of expectations and obligations - is
complex and variable. At one extreme is a short-term exchange of goods by
barter. At the other extreme are long-term (at times described as occurring over
several generations) relationships between specific Tukanoans and "their" Maku.
An example of the first type of relationship is given by Silverwood-Cope
(1972, p. 97), who reports that some Maku in the Macucu-parana region, in
addition to being the traditional servants of specific Desana, will exchange meat
obtained during an extended hunting expedition with Uananos living on tributar-
ies of the Vaupe"s River. Expectations in terms of either an extended period of
contact or "servant" obligations are minimal in these situations; the interaction is
fairly straightforwardly economic in nature. Trade objects manufactured by Maku
for exchange, in addition to smoked or fresh game, include the highly prized
twined baskets of various shapes and sizes, Virola snuff, curare for blowguns,
and occasionally pelts destined ultimately for trade with whites. At present, in
addition to food (e.g., manioc, chili peppers) and other cultigens such as coca
and tobacco, it is my impression that Tukanoan items of trade are almost entirely
white-manufactured articles such as machetes, beads, and cloth. I observed
bartering taking place in longhouses where Maku were visiting. This bartering
mostly involved Tukanoan women placing orders for custom-made baskets, and
paying with beads, cloth, and similar items.
The question of specialization in manufactured objects in the Vaupes -
between Maku and Tukanoans and among Tukanoans - is a complex one. Allen
(1947, p. 574), for example, lists blowguns as Maku trade items, stating that
Tukanoans had forgotten how to make them. Bara and Tuyuka men of the Papuri
drainage still make blowguns, and probably other Tukanoans do so as well, and it
is interesting that these people do not have regular contact with Maku, in
contrast with some groups of Desana, Uanano, and Cubeo. Silverwood-Cope
states that Maku do supply those Tukanoans with blowpipes and hunting poison
(1972, p. 97). As we have seen, "forgetting" how to make certain necessary
objects that must then be acquired by trade with groups specializing in their
manufacture has been reported for the Yanomamo (Chagnon, 1968). S. Hugh-
Jones (personal communication) states that Barasana of the Pira-parana say that
animal products such as jaguar teeth, bones, and peccary teeth for making ritual
gear used to come from Maku. Silverwood-Cope lists additional forest articles
for exchange: wild fruits, palm leaves for roofing, tree resins, and reeds for
panpipes (1972, p. 97). H. Reid (personal communication) notes that Hupdii
Maku supply many objects for use in Tukanoan rituals, such as feathers,
nutshells, monkey fur, and teeth.
154
Tukanoans and Maku
Straightforward barter often occurs when a group of Makii stays near a
Tukanoan settlement. During their visit the Maku exchange their labor for food
and the opportunity to participate in a festival. On one occasion I observed a
Maku group of about eight adults and several children who were helping with
communal work at a Desana settlement on the Papuri River. These Maku helped
rethatch a roof, clear away jungle growth near the house and nearby fields, and
secure the stairway to the canoe landing. They were paid with a two-day festival
during which large amounts of chicha, coca, and tobacco were consumed. Maku
were fed during this time, and Maku women helped with the food preparation.
Tukanoans from nearby settlements placed orders for Maku baskets.
Another pattern that occurs is the semipermanent attachment of a Maku group
to one longhouse. In this situation, the Makii seem to be less independent than
those who more regularly maintain their own settlements and cultivations. I
observed this type of arrangement when a Maku group stayed for approximately
six months at a Tuyuka longhouse downriver from Pumanaka buro. It was
difficult to believe that this particular group could have lived entirely by itself for
any appreciable length of time. Their personal possessions were pitifully scant,
consisting of a broken and almost useless machete, two hammocks, and some
articles of clothing. They had nothing in the way of cooking utensils or hunting
equipment, yet they bartered their baskets for nonutilitarian white-manufactured
goods such as beads, cloth, and earrings.'' The one dog with them was obviously
starving to death and could barely get to its feet (which it tried to do very
infrequently). This group slept in a very flimsy shelter some distance away from
the longhouse, although most of their time was spent at the longhouse (with the
exception of the dog, who, not being allowed near it, consequently received very
little food). After six months, the group departed as a result of a fight between
one of the Maku men and a young Tuyuka man. I was told they had gone to the
Tiquie in search of another longhouse to which they could attach themselves.
The group consisted of two men, three women, a teenage girl, and four small
children. One of the men was absent most of the time, and the other was the
personal servant of the longhouse headman. It was this Maku's job to accom-
pany the headman whenever he was wanted, to light his cigars,12 to light and
maintain the resin torch at night, to fetch chicha for the guests during a party, and
in general to be at the headman's beck and call. In exchange for these services,
there was food for him and his family and occasional drinks of chicha, puffs on
the cigar, and spoonfuls of coca.
The women helped in the food-processing chores and in the fields, and the
young girl occasionally looked after the children.13 The women also chopped
firewood, tended the fires, fetched water, and deloused other women. I never
observed a Maku woman performing the final stages of cooking or serving other
people. The Maku waited to be asked to eat and in general were reserved and
kept in the background. No Maku woman will perform all or even a substantial
155
The fish people
amount of the work of another woman. It is my impression that no Tukanoan
woman would dream of sitting back idly while another woman, even a Maku,
did her work. The nature of Maku servanthood is to help out.
Except when drunk, Tukanoans are not directly cruel to Maku, nor do they
openly taunt and jeer at them. Maku status is manifested in more subtle ways.
For example, Maku are never formally greeted. Maku are also visibly distin-
guished by their relatively poorer quality of clothing and relative lack of jewelry.
They will eat when asked to and talk when invited to join a conversation, but
otherwise they are silent and generally inconspicuous.
During the period when this Maku group stayed at the downriver Tuyuka
longhouse, a large party of Tuyuka and Bara, including the Bara I lived with,
traveled to the Tiquie River for a large festival. The young Maku girl went
along, as did the man who was the headman's personal servant. The girl had been
given a new dress for the occasion, but her Maku status was quite apparent
during the trip. She was included little by little in conversations held by her
Tuyuka and Bara age-mates, but throughout the trip she took care of the
headman's youngest child and fetched water and firewood when asked to.
Making generalizations about this type of extended servant-master relation-
ship from observations of a single Maku group is, of course, somewhat risky.
However, I was assured by the Tukanoans I lived with that the situation was a
typical one. 14 1 also received confirmation from informants when they discussed
in the abstract norms governing interaction between Tukanoans and Maku. The
patterns I have described are also reflected in myths dealing with Maku and
Makti-like relationships.
Several authors mentioned Maku women being used as sexual objects by
Tukanoans (Goldman, 1963, p. 107; Koch-Griinberg, 1906; McGovern, 1927,
p. 259; Reichel-Dolmatoff, 1971, p. 19; Whiffen, 1915, p. 262). l5 Tukanoans
denied that such behavior was ever engaged in and condemned the idea, but this
does not automatically mean that it does not happen. Ideally, Maku are not
considered potential marriage partners.16 However, some liaisons turn into per-
manent unions. I knew of one Maku woman married to a Desana in the mission
town of Monfort. The information about her tribal identity was very difficult to
obtain and cross-check. Goldman reports that according to other more high-
ranking clans, the Bahkukiwa sib he studied was originally a Maku group. The
elevation of Maku groups is probably a periodically occurring process: Koch-
Griinberg states that three Cubeo clans were probably originally Maku (1909-10,
p. 136). Silverwood-Cope thinks that intermarriage between Maku and Tukanoans
is probably the crucial factor in the assimilation process. Once intermarriage
takes place, steps will be taken by both sides to erase the taint of Maku ancestry
(1972, p. 104).
A somewhat different picture emerges from earlier accounts of Maku-Tuka-
noan interaction. Several reports state that Maku were "slaves," although what
was meant by this rather vague term is difficult to ascertain with any precision. At
156
Tukanoans and Maku
issue is the degree of control a slave's owner had over life and limb, what
resources were open to a master should the slave escape, and whether slaves
could be lent or sold to other masters.
Some evidence exists of Indians' enslaving other Indians - both Maku and
Tukanoans - in previous periods. Such enslavement was connected with the
presence of warfare, much stronger headmen, and rapacious white traders and
rubber gatherers, all of which are either absent or greatly diminished at present.
Both Wallace (1889/1972, pp. 206-07) and Spruce (1908, p. 294) report
Tukanoans' preying on each other, sometimes because a headman has been
required to turn over a specific number of Indians to a government agent or
upriver trader. It should be borne in mind that such reports involve disruptions
caused by whites and do not necessarily tell us anything about the aboriginal
situation. Furthermore, the two individuals who have firsthand knowledge of
Maku seriously question such reports of enslavement. Silverwood-Cope con-
cludes that Maku were less vulnerable to whatever dangers actually existed than
were Tukanoans, because at the first opportunity of escape they could more easily
disappear into the trackless jungle and survive for a long period, moving con-
stantly. And Reid (personal communication) notes the general tendency of
nineteenth-century explorers and travelers to see much more hierarchy than
actually existed. Therefore, some of the earlier reports that will be discussed
should be taken with a grain of salt. No Tukanoan ever described to me any form
of Tukanoan-Maku interaction that could be called "slavery"; and at least at
present no Maku can be coerced to stay in a place or work when not willing to.
Several earlier reports state that Maku attached themselves to Tukanoan
settlements out of a need for protection against marauding bands of Indians.
Koch-Griinberg gives this as the reason why Maku stayed at Paricachivera
Mission on the Tiquie" River despite their "slave" status (reported by Metraux,
1948, p, 866). He reports that Maku worked for Tukanoans, who treated them
like pet animals. According to him, Maku women were at the sexual disposition
of the men. Briizzi Alves da Silva (1962, p. 463) reports that a letter of
appointment written by the Franciscan director of Paricachivera Mission in 1885
explicitly prohibits the sale of Maku, and Van Ernst (1966b, p. 172) reports the
same of a similar document for 1882. Koch-Grunberg states that a Tukano chief
on the Tiquie claimed to own several hundred Maku slaves (Silverwood-Cope,
1972, p. 317).
Several authors conclude that greater contact with whites increased the en-
slavement of Indians. Spruce (1908, p. 294), for example, is quite critical of
this practice. A. R. Wallace (1889/1972, p. 236) remarks that he was given a
Maku boy for the duration of his travels on the Vaupes River in payment of a
debt. Goldman discusses the symbiotic nature of this "slavery" and states that
the fact that Maku were traded for ceremonial objects (one Maku male suppos-
edly being equal to one ceremonial staff) does not indicate the absolute power
over a person it seems to imply. Maku were fairly easily traded because they had
157
The fish people
no great attachment to a particular place. Goldman (1963, p. 106) also states that
although Maku were valued for their labor, they were mainly wanted to increase
the number in a household.
At present, Maku are definitely not "slaves," but otherwise their inferior
status has not changed. Van Ernst (1966b, p. 191) affirms this for Maku living in
mission towns, noting that they live in poorly constructed huts on the periphery
of the village and are very subservient to the Tukanoans there. For example,
according to Van Ernst, if a priest gives a Maku an item, a Tuyuka or other
Tukanoan can take it away from him or her then and there, and no one will find it
unusual.
In conclusion, although Bara informants denied the potentiality for sexual
relations between Tukanoans and Maku and had not heard of a Maku band
attaching itself to a longhouse for the purpose of protection from potential
enslavers, in general what is reported in the literature on Maku was confirmed
during conversations with Bara and other Tukanoans. Informants also mentioned
Maku as having been military allies in earlier times. Still, it should be noted that
a general tradition exists in which lower-ranking sibs are automatically spoken of
as allies, and Maku may have been spoken of in this manner for similar reasons.
(When describing a raid that occurred in the past, the excitement is heightened if
the narrator comments on the large numbers of people massed on both sides.)
Bara informants never showed any guilt about how Maku were treated and
were never reluctant to discuss any aspect of Maku-Tukanoan interaction. They
would readily list the services that Maku perform for Tukanoans and would just
as readily discuss why they see Maku as inferior.
Maku as symbol to Tukanoans

The range of possible relationships occurring between Maku and Tukanoans
clearly shows that interaction between the two groups is a significant part of life
for all Maku and for many Tukanoans. Additionally, each group provides the
other with illustrations of certain basic premises it holds about the world, human
nature, and proper behavior.
Because my encounters with Maku were so infrequent, this section is limited
to how Tukanoans use Maku as symbols. For an account of Maku perceptions
of Tukanoans, see Silverwood-Cope, 1972, and Reid, n.d.
For Tukanoans, Maku exist not only in everyday life but also in the realities
described in and explained by their cosmology. Maku appear time and again in
myths and in the Tukanoan variety of just-so stories. When Maku are used as
symbols the resulting picture often diverges from what Maku are like in actuali-
ty. It is obviously important always to distinguish between the two. In my
opinion, Bara and other Tukanoans also distinguish between Maku as flesh-and-
blood beings and Maku as symbols. This is not to say that at times Tukanoans
158
Tukanoans and Maku
are not simply mistaken about real-life Maku, nor do they always clearly make
the distinction between symbolism and reality.
Statements from Bara and neighboring Tukanoans about Maku that illustrate
the Tukanoan organization of the universe express at least three themes: (1)
Maku as examples of what "true people" do not do; (2) Maku as prototypes of
the lowest-ranking category in any hierarchy, most typically the ordering of sibs
within a given language group; and (3) Maku as examples of the servant-master
relationship.
We have already examined the first theme. Tukanoans say Maku are not "true
people" because they are animal-like, eat tabooed food, marry their sisters,
cannot provide enough food for themselves, live wild in the jungle, and so forth.
They thus serve as stern reminders of what can happen when the rules governing
proper behavior are ignored. Highly visible negatively evaluated subgroups often
play this type of symbiotic role within the dominant culture: Gypsies and lepers
are good examples.
To understand the second theme requires examining the similarities between
Maku and the lowest-ranked sib in any given language group. A low-ranking
sib's origins can always be impugned by suggesting that it was originally a Maku
band. Unlike all true Tukanoan sibs, it did not emerge at a rapids site but attached
itself to a longhouse and slowly learned the language of its adopted Tukanoan
settlement and finally was assimilated. The motives behind the settlement's
taking in the Maku group are said to be kindness and pity because of their state
of near starvation. As mentioned earlier, just how much this reflects an actual
process that occurs periodically is uncertain.
It is important to note, however, that a Tukanoan who is describing a sib origin
in this way is not primarily concerned with historical reality. The very next day
the same Tukanoan might describe the sib in question as a full-fledged Bara or
Tuyuka sib in good standing. Imputed Maku origins are a means by which a
sib's present status can be examined and criticized. Such a statement may simply
indicate a Tukanoan's quarrel with an individual or settlement belonging to that
sib, or a group of Tukanoans may be punishing an individual or group that has
transgressed one of the rules by which all "true people" live.
Thus, although members of any low-ranking sibs of any Tukanoan language
group are in many ways distinct from Maku, the differences are not as extensive
as it at times seems, and Tukanoan evaluations of these differences can vary over
time. Some criticisms of Maku - not speaking properly, marrying people who
speak the same language, not observing taboos - will sometimes be made about
lower-ranked sibs as well. The Bara with whom I lived described the lowest-
ranked Bara sibs as having originally been Maku. This was mentioned in
conjunction with statements about the inferior dialect of Bara these groups
spoke, their sloppiness with regard to proper ritual attitudes, and their supposed
previous state of total nakedness. Similar statements are made about Maku.
159
The fish people
Silverwood-Cope (1972, pp. 195-96), discussing Tukanoan attitudes toward
Maku, says that Tukanoans stress those differences that distinguish humanity
from the natural world, showing Maku to be far closer to the latter. Human
beings do not live in the forest but make large clearings and build large elaborate
houses. Human beings do not marry their own kind, and social structure is
reflected in the structure of language affiliation and in territorial arrangements
(especially in residential exogamy). Animals do not make gardens, plant, make
cazabe and beer, nor do they hold rituals. Animals do not wear ornaments. Thus,
Maku insofar as they fail to be human in these respects, are much more
animal-like.
The third theme concerns the role of Maku as servant. When one is a servant,
one is often thought of as qualitatively different from the master. In this sense,
the servant-master relationship between Tukanoans and Maku differs from most
other relationships existing among Tukanoans - relationships involving trade,
services, differential status, or a combination of these. For example, according to
C. Hugh-Jones (1979, p. 81), two men can stand in a "ceremonial brother-in-
law" relationship with each other. They will belong to different language groups,
but the relationship is completely egalitarian: It is between "men who exchange
sisters." Asymmetrical relationships involving exchange of goods or services are
also found. For example, the headman is entitled to his position partly through
inheritance, but he maintains it mainly through providing more of certain ser-
vices and goods than others, receiving prestige in return. He is not seen as an
inherently different human being from his fellows because he is headman. This is
also the case for the specialized positions of headman's wife, head dancer, or
shaman (although certain shamans become so powerful they are transformed into
a different kind of being). Except for the sexual division of labor, none of the
occupational specializations is full-time.17 This is significant, for Maku are
always seen as being in the position of servants. (This is, of course, the Tukanoan
view of things.)
In short, always to act as a servant is good evidence of one's inherent and
permanent lower status, and this is the essence of being Maku. To some degree,
other characteristics seen as typifying the Maku (such as a lack of ceremonial
knowledge) are thought to be a consequence of the inherent Maku state of
servitude. This point can be overstated, however, for at times Tukanoans, when
feeling expansive, will give the impression that Maku are not categorically and
eternally excluded from the status of true people: When Maku start behaving as
they ought to, their penalized status will assuredly also cease. This recalls other
ethnographic situations where members of a dominant group claim that when a
discriminated-against minority's behavior changes, their status will also change.
In all of these cases as well, ambivalence and contradictory statements are found.
The ambiguity of the minority's status - one day described as nonhuman and one
day described as potentially human "if only they would . . ." - is perhaps a
general characteristic of dominant-subordinate interaction.
160
Tukanoans and Maku
The theme of a class of beings who are permanently consigned to a position of
being servants to others is revealed in several Bara and other Tukanoan myths.
Certain animals are often described as being the "Maku" of other animals.
Frequently a symbiotic relationship actually occurs between the two species
described (cff Reichel-Dolmatoff, 1971, p. 211). Many times, for example, a
myth's protagonist has as his Makii little birds who keep him company and
occasionally take messages or perform other chores. At times these species are
spoken of as military allies.
In the myth of Koa-mahkii and Nima Pino ("Curare Anaconda") summa-
rized in Chapter 5, Koa-mahkii initially gains entrance into both Nima Pino's
daughter's house and Nima Pino's cave through the ruse of presenting himself as a
Maku-like servant. Koa-mahkii's previous stay with Eyoa (wood ibis birds)
had also been as a servant, but they finally decided to encourage him to leave,
because he had practically eaten them out of house and home. This is a
frequently heard complaint about Maku. The daughter of Nima Pino also accepts
Koa-mahkii as a servant, but he becomes not only a voracious eater but
sexually demanding as well; this may express the sexual tensions between Maku
and Tukanoans who live in very close contact.
In the myth of Wai Pino Mahko ("Fish Anaconda Daughter"), summarized in
the preceding chapter, Yeba Mahii visits his wife's father and relatives in their
longhouse under the river. All the species of fish are ranked in a fashion similar to
the set of sibs of a Tukanoan language group. Fish Anaconda, noticing that Yeba
is hungry, tells him to eat tarira fish, the lowest-ranked species. Fish Anaconda
says that although all fish are of one family, tarira are "our Maku," and
presumably their sacrifice is not horribly serious.
Having Maku or Maku-like servants can definitely indicate prestige and
popularity. Goldman states that Maku are most wanted among the Cubeo
because they swell the household. An example of this in a Bara myth is in the
story of Oa ("oppossum"), a trickster who outsmarts his visitors by claiming to
have Maku and therefore to be an important person - which is in fact not the
case. At one point Oa wonders out loud where his Maku could be and then
"discovers" a termite nest that he says his servants have apparently just brought
him (termites are a delicacy).
Conclusions
Relationships between Maku and Tukanoans can be analyzed from economic,
political, and symbolic frames of reference. From the economic point of view,
these relationships are exchanges of goods for goods or of goods for services. In
this they are similar to the servant-master interaction occurring between Pygmies
and Bantu Negroes of the Congo (Turnbull, 1965, pp. 293-94). Similar exchange
relationships have been reported for several nomadic or seminomadic groups.
This type of symbiosis is probably an old pattern of economic specialization,
161
The fish people
occurring between horticulturalists or pastoralists and hunter-gatherers since the

neolithic revolution itself. In many cases it undoubtedly has been affected by a
greater perceived need for white-manufactured articles more easily obtained by
one of the parties.18
Some of the political aspects of Maku-Tukanoan interaction are probably less
important than in the past. Maku certainly are not "slaves" at present, but it may
be that in earlier times, when Tukanoans were less acculturated and less domi-
nated by missionaries, and when feuding was a much more normal state of affairs
(and thus raiding for captives a real danger), groups of Maku were more like the
"field slaves" described by Goldman (1963, p. 106). Still, that one investigator
with long-term experience with Maku seriously doubts that they were ever slaves
(Reid, 1979). Additionally, the right to use Maku women sexually might have
been more widespread although never fully sanctioned (McGovern, 1927, p.
249). Finally, Maku might very well have served as military allies, giving their
services when needed as warriors in return for protection while attached to a
particular settlement.
As a symbol, Maku serve at least two purposes, and in a certain sense this
necessarily leads to contradictory statements from Tukanoans about why Maku
are what they are. On the one hand, Maku serve as boundary markers and
maintainers, reminding Tukanoans of correct behavior. To do this, Maku must
be seen as like Tukanoans to some degree - people, or at least potential people,
who break the rules and whose fate will be shared by anyone who behaves in
similar fashion. On the other hand, Maku are exploited, and it is easier to
maintain this type of relationship if Maku are seen as qualitatively and unchange-
ably different. It is this tension between the purposes to which Maku-as-symbol
are put that in part produces the ambivalence and ambiguity of statements elicited
from Tukanoans when discussing the Maku.
The literature on ethnic groups and other minorities frequently refers to ways in
which members of a stigmatized group are a boundary-maintaining "message"
for the dominant group (Barth 1969; Goffman 1963). The outside group's
existence affirms each insider's membership and specifies by illustration what the
criteria are for membership. Examples abound: the Jews in Nazi Germany and
elsewhere, the Ainu in Japan, the Irish in Boston at the turn of the century.
Outsiders at times serve as negative examples of what can happen if cultural
norms are not adhered to. Although to some degree the outside groups buy the
dominant group's view of things,19 it is also true that frequently each group has a
remarkably disparate view of itself and the other group. Both characteristics fit
the Maku.
When Tukanoans are discussing human nature in general, the tendency is to
include all people - whites and Maku - in the discussion. In this framework
Maku are not seen as categorically and eternally excluded from the status of
"true people" but potentially as members of the human race in good standing if
and when they start to comply with certain rules and regulations. This is apparent
162
Tukanoans and Maku
when a Tukanoan discusses certain behaviors as right and natural to everyone and
disapproves of groups that do not subscribe to the same rules. If Maku were
really beyond the pale with regard to their humanity, it would seem that other
rules would apply for judging their behavior. In this regard, however, the ambigu-
ity and inconsistency appear to be built in; at times even animals are discussed in
terms of human-like characteristics and evaluated accordingly. Thus, when it is
useful to use the Maku in a discussion of proper behavior, applicable to all, they
are so used.
Still, Tukanoans are also interested in justifying and continuing their exploita-
tion of Maku in the servant-master relationship. This is accomplished in part by
emphasizing the differences between Maku and Tukanoans, saying Maku are
inferior in these respects and claiming that these are inherent differences. These
qualitative and unchangeable differences are what necessitates and justifies
Tukanoans' treating Maku as second-class citizens. This type of rationalization
is almost universally found in situations of social stratification.
In the Vaupes, tendencies toward stratification are not nearly as pronounced as
in more complex and densely populated societies. Fixed, hereditary ranking is
looser than that found in many other societies because of the ecological setting,
demography, and economic organization of Tukanoan society. This is reflected in
the actual freedom that Maku have, even though in the past Maku in some
instances seem to have been less free than now. The periodic assimilation of
Maku by Tukanoan local groups reported by various investigators would not
occur so easily in more rigidly stratified societies in which genealogical reckon-
ing is of more importance. Nevertheless, a hierarchy definitely exists in the
Vaup6s, and expectably, an ideology exists to support and explain it and its
correctness. This is what occurs when a dominant group refers to differences
between it and a subordinate group as permanent, inborn qualities that make
unequal treatment just and inevitable. As we have seen, when it is convenient to
describe Maku in these terms, they are so described.
163
9.
The role of language and speech in
Tukanoan identity
It is by now obvious that language and speech are very important areas of Vaupes
culture, among other things signaling identity in several social groups and
categories. My choice of terminology for the exogamous language groups is
based on the overwhelming importance of this factor in Tukanoan conceptualiza-
tions of these units. Although everyone is multilingual, individuals identify with
and are loyal to only one language, their father language. Linguistic criteria are
used as markers for differentiation among other kinds of social units as well - for
example, sibs and phratries. In this chapter, I will examine in greater detail the
ways in which language and speech are important to Tukanoans and will show
how the Tukanoan model of language makes connections between linguistic
phenomena and certain features of the nonlinguistic environment in which they
occur.'
In the Vaupes (as elsewhere), language and speech are used to communicate
information in at least two ways. First, languages (and the various genres and
registers of any given language) are codes that are used for exchanging referential
information. Speakers intentionally use these codes to send messages to each
other. Second, language and speech are codes for communicating - sometimes
intentionally and sometimes totally unconsciously, social information about the
identities of the speaker, interlocutor (the individual to whom the message is
directed), and audience.
Tukanoans see language and speech as correlated with a number of features of
Vaupes social organization and culture. This process can be one of making links
between actual speech forms and nonlinguistic social forms; or the connection
between linguistic phenomena and the environment in which they occur can be a
more diffuse one not attached to specific speech patterns. For example, Tukanoans
in general assume that Vaupes languages are equal to one another - in contrast to
many multilingual situations in which the interacting codes are ranked along
dimensions such as beauty, logicality, and so on. This linguistic egalitarianism
doubtless derives from the social egalitarianism Tukanoans see as existing
between the language groups.
The relative emphasis placed on language and speech in a society in general
164
Role of language and speech in Tukanoan identity
can indicate much about the social environment of the society. The abundant
evidence of the importance language has for Tukanoans suggests that a thorough
understanding of Tukanoan "ethnolinguistics" (Moerman, 1965) is crucial for a
thorough understanding of most areas of Vaupes society and culture.
Vaupes language and speech as badges of identity

People tend to wear badges in situations in which there is a high degree of
interaction and in which the differences being signaled by the badges are
important for the successful outcome of the interaction. Badges can signal many
kinds of social and cultural roles: Examples are sheriff's badges, the crosses,
worn on the necks of Maronite Christian soldiers in Lebanon, and the name tags
people wear at annual conventions. As such, badges are a kind of message, a
highly specialized kind of message that must meet specific requirements if
communication is to be successful.
Badges can signal many kinds of limited or temporary social positions and
roles such as "doctor" or "Hell's Angels member." Thus, although one may be
a medical doctor all one's life, this part of one's identity is not always signaled -
and if it were, it would at times occasion humor. Although being a member of the
Hell's Angels may be a total experience, it probably will not last an entire
lifetime. In contrast, being a member of a cultural, racial, or ethnic group usually
is a permanent identification and frequently an extremely pervasive one.
A number of authors have discussed such badges of identity. Blom discusses
multicultural or ethnic situations in which cultural differences separating the
interacting groups become codified (1969, p. 84). Barth (1964, 1969) refers to
these codified differences as "diacritica." In situations characterized by frequent
interaction between different categories of people, some of the differences
separating these categories become standardized and eventually stereotyped. The
more such units interact and the longer the period of interaction, the more these
categories of people become structurally similar and differentiated only by a few,
clear diacritica. The total inventory of cultural differences is reduced, but the
differences that remain, because of their new role as badges or emblems of
identification with distinct groups, become more important. Such codified
differences function as markers so that interaction can run more smoothly. They
are thus quite unlike real cultural differences that are genuine impediments to
interaction and communication.2
If badges and emblems of identity are a kind of message that must be
transmitted, received, and decoded for interaction to take place, they must be
reasonably parsimonious and successful most of the time. The reduction of
differences satisfies the requirement of parsimony, and those differences that
remain undergo transformations to ensure a maximum amount of effectiveness.3
The particular differences that remain after this reduction have usually changed in
directions that increase their visibility, unambiguity, and discreteness, because
165
The fish people
these characteristics facilitate the successful transmission of the message that
badges and emblems are intended to send. Often the process includes opposition
and polarization of the remaining differences, which increase the impact of the
message.
Many badges that signal cultural or social identity are literally visible: Hoklo
silver hair ornaments worn as "self-conscious marks of differentiation and local
pride" are a good example (Ward, 1965, p. 133). Yet badges need not be literally
visible to be effective; what is essential is that they be able to be displayed in
public, either continually or when needed. For example, Moerman notes that
"slight peculiarities of speech may serve as emblems (Fortes, 1945, p. 136; cf.
[E.] Sapir, 1949) of a community which has a tribal name . . . " (1965, p. 1218).
These slight peculiarities are probably far more important as markers of distinct
communities than as linguistic differences per se. 4 Evidence abounds that
language is a frequent and highly successful marker of cultural and social
differentiation. The Canaanites' inability to pronounce Shibboleth is one of
numerous cases. Gumperz, one of the scholars who has carried out significant
research on language and speech as indicators of social identity, has studied a
situation of code-switching between Hindi and Punjabi speakers in which a few
linguistic items with a high text frequency "suffice to preserve the necessary
minimum of symbols of role specificity" (1964a, p. 1123).
If the message transmitted by a badge of social identity is to be successfully
communicated, the sender and receiver must see its content as important enough
to warrant their time and energy. It generally seems to be the case that when
differences evolve into badges and emblems of identity, they in fact do become
very significant in the minds of the individuals involved and become quite
charged with emotion. The fact that any outsiders observing this frequently judge
the differences separating the groups as quite trivial, superficial, and overemphasized
is not accidental. For the participants, the features that render each badge distinct
from others in the set become highly charged with meaning. Such meaning may
have a negative or positive value, for both the badge wearers and the badge
observers, but will never produce mere indifference. Similarly, the dimension
within which all of these distinctive features are contained can also become
highly significant to insiders and exaggerated to outsiders. In the Vaup6s, the
dimension is language or linguistics, and the features are those linguistic
elements that are seen by Tukanoans as making Vaupes languages mutually
unintelligible. Vocabulary and grammatical differences and the co-occurrence
rules that serve to keep Tukanoan languages as discrete category systems are
vigorously maintained by Tukanoans. And the dimension of language is of
extreme importance to Tukanoans, one that is woven into much of the fabric of
their social life.
Although the linguistic picture in the Vaup6s is unusual in a number of ways
(and unique in a few), language and speech serve as markers of identity
throughout the world. Furthermore, the process by which great import becomes
166
attributed to the specific linguistic features marking social differences in class,
sex, race, or ethnicity - is the same wherever it occurs. It seems always to be the
case that perception becomes selective when specific linguistic features take on
this role. The study of the Kannada and Marathi languages made by Gumperz
(1964b) is an excellent illustration. In this case, despite high contact frequency,
speakers believe linguistic diversity is far greater than it actually is, because of the
importance of maintaining the distinction between the social groups identified
with each code (Ervin-Tripp, 1969, p. 145). Because these rather trivial differ-
ences have a high text frequency, however, they are quite "visible" to the
speakers.5
In addition to being parsimonious, visible, and significant, badges are also
likely to be somewhat arbitrary. The more arbitrary a badge is - the less it is an
inherent characteristic of an individual - the more easily it can be manipulated
always to apply to just the right group of people even when changing conditions
alter the membership of the group. Thus, the form of a badge is most suited to
this function when the association between it and the category of people it
represents is completely learned as a symbol. A good example of such arbitrari-
ness is the Star of David as an emblem of Jewish identity, literally worn as a
badge in Nazi Germany. Yet it is a potent enough association to have led to many
consequences - for example, Egypt prohibited the display of any six-pointed stars
on imported products after the Six-Day War, regardless of the product's lack of
connection to Israel.
Obviously, many badges and emblems of identity are not entirely arbitrary;
skin color is a good example. Nevertheless, racism is effective in keeping many
people oppressed because of the learned association between various phenotypi-
cal traits and socially stigmatized categories of people, and because it serves this
purpose so well, it is a well-established institution, despite the sometimes
ludicrous expressions that arise.6
Language and speech are fairly well adapted to fulfill this requirement of
arbitrariness, although not completely so. Shaw's Pygmalion dramatizes this
theme. Much humor is devoted to chronicling ways in which upwardly mobile
individuals betray their origins with their speech, either through slip-ups or
hyperforms (Labov, 1966). In the Vaupes one cannot cross boundaries by
changing one's speech patterns or identification with a particular father language.
This is because one inherits one's father language and cannot change it in any
legitimate way, and it is doubtful that any Tukanoan has tried to do it illegitimate-
ly. Aside from the fact that there are virtually no motives for wanting to switch
father-language identity, it would be impossible because everyone knows or
knows about everyone else. This is not the case when a Tukanoan is trying to
"pass" into the category of "Colombian," particularly if he or she lives in Mitu
or San Jos6 de Guaviare. An ability to speak Spanish with very little Indian
accent is an important factor influencing the degree to which it is possible to carry
this off.
167
The fish people
Languages are also arbitrary in the sense that they can take on the social values
associated with the group of people speaking the language. A fair amount of
research on this topic has been carried out, showing how languages come to be
regarded as superior (or inferior), musical, logical, and so on, mainly by serving
as a symbol of the community identified with them, to which these traits are also
seen to apply. Speech itself can at times be said to be "high," "thick," or
"heavy," because of an association with certain groups (J. D. Sapir, 1975).7
Thus, badges and emblems serve their purpose of sending messages best when
they are clear-cut, unambiguous, and discrete. This is why the emblematic
differences found among interacting social units are standardized, often to the
point of caricature. In situations in which dual membership, marginal member-
ship, boundary crossing ("passing"), or secret membership can occur, the
orderliness of the interaction and people's assumptions about predictability can
become eroded.8 Although such flexibility may be beneficial to individuals
attempting to pass into other social categories (hence the frequency of this theme
in fairy tales, Shakespeare, Gilbert and Sullivan, etc.), they do so at the expense
of the social system - another reason why this theme is so popular in fiction.
As indicated earlier, identification with a specific language is clear-cut and
permanent in the Vaupes. In situations where languages or speech varieties are
badges of identity, the criterion of discreteness may be met by a relatively few
differences, as is the case in an example of Kannada and Marathi, discussed by
Gumperz and Wilson (1971), in which two unrelated languages have greatly
converged yet are seen as distinct by their respective speakers who identify with
only one of the two. Tukanoans see their languages as quite distinct, with
clear-cut boundaries. Some Vaupes languages are more different from each other
than in some of the situations discussed by Gumperz (1964b, 1969) and Moerman
(1965, p. 1218). Yet interesting questions arise in connection with the Vaupes
material regarding the nature of dialect versus language and the influence speakers'
attitudes and loyalty to only one code has on language change. For example,
Sorensen remarks: "It has occurred to me that the exogamic and other cultural
institutions . . . may be exerting a force that makes a speaker want to render
closely-related languages further apart, even to an artificial extent, but so far I
have detected no linguistic innovations to this end" (1967, p. 676).
Much of the sociolinguistic literature deals with the ways in which individuals
send and receive information about social identity through the use of language.
Most of this work is concerned with speech behavior itself, and some studies have
concentrated on the ways speakers themselves understand how speech can
indicate social status. Labov's discussions (1966) of stereotype and hypercorrection
are good examples.
In some situations in the Vaupes, the public display of language group identity
through actual speech occurs. Sorensen gives an example: "Each individual
initially speaks in his own father-language during such a conversation in order to
168
assert his tribal affiliation and identification" (1967, p. 678). A Tukanoan who is
publicly acknowledging his language group membership is usually reaffirming
this aspect of his social identity rather than announcing a previously unknown
fact about himself. Most of the time, Tukanoans interact with other Tukanoans
who are well known to them, and hence they know one another's language group
identity. Thus, the fact that very few speech events allow one to signal one's
language group identity does not indicate that language group identity is not an
extremely important aspect of Vaupe's life. On the contrary, it is so important
that there is little need for Tukanoans continually to remind one another of this
aspect of their social identity. In most situations, and in all informal speech
events, various other sociolinguistic rules determine which languages will be
used in speaking.
That Tukanoans consciously try to maintain linguistic boundaries when speak-
ing is further indication of the role language plays as a badge of identity. Sorensen
(1967, p. 675) states that languages appear to be kept fastidiously apart and that
when two languages are closely related, Tukanoans will "carefully and even
consciously keep them apart." He also states that Tukanoans do not attempt to
speak a language being learned until they feel competent to speak it correctly.
This suggests that interference in speech from one's father language or another
language already in one's verbal repertoire is disapproved of socially. I observed
instances of this type of social disapproval when women were scolded for
allowing words from other languages to creep into conversations in Bara. Other
Tukanoans would comment that such women were not setting a good example for
their children, who should learn to speak their father's and mother's languages
correctly. Occasionally it was remarked to me that I would shame the longhouse
if I learned Bari with Tuyuka words mixed in. I did not try to measure the overall
amount of interference from other languages that was allowed to pass or try to
estimate the frequency of criticism of such mistakes. It is my impression that
attempting to measure such interference objectively would be very hard, because
to a large extent what interference is in the Vaupes is what is seen as such by the
participants at a given point in time. It seems clear that Tukanoans disapprove of
using a word from another language in speaking and see it as a mistake. Thus,
while convergence may be taking place (and quite probably is, in fact) among
Tukanoan languages, at any given point in time strict co-occurrence rules operate
to keep the languages separate in a specific individual's or group's repertoire.
The presence of these rules is evidence for the emblematic nature of Vaupe's
languages. That Tukanoans are aware of speech "mistakes," that they place a
high value on correct speech, and that they see Vaupes languages as mutually
unintelligible support the hypothesis that the separate languages serve as em-
blems of structurally similar interacting groups. Otherwise, languages might
converge or several languages be dropped, which would certainly be more
efficient linguistically. When I directly asked a Tukanoan why they spoke so many
169
The fish people
languages rather than, for example, relying on Tukano exclusively, because they
all knew it anyway, he responded, "If we all were Tukano speakers, where would
we get our women?"
Thus, the cognitive model Tukanoans have of linguistic discreteness, espe-
cially with regard to mutual unintelligibility. is of great importance in under-
standing the linguistic situation. For example, I was surprised several times to
find that a word that was frequently used in Bara was considered by Bara
speakers to be a Tuyuka word. When asked for a translation, they always
produced a Bara word and would give a reason for preferring the Tuyuka word.
Most important, I would be reassured that "everyone knows it is a Tuyuka word."
This suggests that Tukanoans are aware of intrusive words in a language's lexicon
but that this is accepted as long as the co-occurrence rules separating languages
are not seen to be in danger of breaking down.
Messages about social identity can be signaled by language in two ways. The
first is through speech itself. Although identifying one's social identity through
speech does not constantly occur in the Vaupes, this is due to the fact that
Tukanoan language group affiliations are known by almost everyone almost all of
the time, reducing the need continually to display this emblem of membership. In
other multilingual situations, speech serves this function to a greater degree. For
example, in the Xingu Basin, where Indians from various linguistically discrete
groups come together for ceremonies, each group continues to speak its own
language, even though others present may not understand it (Basso, 1973). The
same is reported by Rivero for the Chiricoa-Guahibo and the Achaguas in the
Colombian llanos (in Morey and Morey, 1975, p. 9). Such a situation is also
found among the Yurok, Karok, and Hupa groups in Northern California. Each
community is identified with one tribal language, and within the community only
the tribal language is spoken. Separate names exist for many of the same physical
landmarks (Bright and Bright, 1965).
The other way in which language can be an emblem of social identity occurs
when people have a formal affiliation with one language and limit their loyalty to
it, despite any knowledge they might have of other languages. To the degree to
which this is public knowledge, it will be a factor in their interactions with
others, regardless of their speech behavior itself. Obviously, this second way of
having language signal social identity requires that the interacting parties be aware
of each other's formal identification with given codes. In situations in which
individuals are perfect bilinguals or multilinguals, their formal identification with
one language must be known by everyone if it is not overtly displayed in speech.
Of course, how one evaluates one's own linguistic affiliation may differ from how
others evaluate it: Most Germans probably differ from most Frenchmen regarding
the privilege of identifying with German rather than French as one's mother
tongue.
Linguistic data on Vaupes languages show the close association between
statements about language and statements about language group membership,
170
supporting the assertion that languages serve as badges of such membership. For
example, the question he wadegti niti mii in Bara, which glosses roughly as
"What [male] language-speaker are you?" invariably elicits a response about the
interlocutor's language group membership. This is unambiguous, and a Bara
male will answer, yii ni barayti, "I am Bara," or iyti-) ye waderd ni bard,
"My father language is Bard." These inquiries about language group member-
ship are grammatically distinct from inquiries and responses about speech itself,
such as he wadegati mii, "What do you say?" or nohkoro waderd mahiti
mii, "How many languages do you know [how to speak]?"
Other evidence available from conversations with Bara informants indicates
their awareness of the emblematic nature of Tukanoan languages. An example of
this is the quotation given earlier regarding the relationship between sibling
terminology and marriage rules. The relationship between common ancestry and
common language is quite clearly drawn.9
How Vaupes languages assume features of the nonlinguistic

environment
Phratries: language distance.

As we saw in Chapter 5, the degree to which phratries exist in the Vaupes is
problematical, whether one is speaking of Tukanoan conceptualizations of
phratries or actual marriage patterns. The Tukanoan model of a phratry probably
does not require conceptualizing it as a distinct entity; at least this seems likely in
the case of the younger Tukanoans who served as informants. Older Tukanoans,
however, give lists of language groups whose members are not supposed to
intermarry because of a sibling relationship between them. This may or may not
be accompanied by actual use of agnatic terminology. Members of the same
phratry are also spoken of as having spoken a single language at some time in the
past. This is in keeping with a general tendency to associate an exogamous
marriage class with a single language and, conversely, an assumption that
intermarrying people are always associated with distinct languages. Thus, agnatic
relatedness and linguistic relatedness are indicators of social closeness.
In other words, when confronted with the fact that certain pairs of language
groups do not intermarry but nonetheless speak different languages, Tukanoans
state that although these languages are mutually unintelligible, they are nonethe-
less relatively close, because the two groups used to speak the same language.
Language distance is, thus, still mirrored in social distance, both being based on
a (relative) genetic model. Bard say that Tukano is easier to learn than other
languages, particularly those languages of the groups with which Bara intermar-
ry. Figure 7 illustrates the lack of fit between Bara statements about relative
genetic distance among Vaupes languages and Sorensen's reconstruction of
171
The fish people
Language groups:
BR: Bara TK: Tukano
BS: Barasana TT: Tatuyo
CB: Cubeo TW: Taiwano
CR: Carapana TY: Tuyuka
DS: Desana UN: Uanano
PR: Piratapuya YU: Yurutf
SR: Siriano
Figure 7. Two conflicting models of language distance. The solid lines represent the
Eastern Tukanoan family as reconstructed by Sorensen (1967); the dashed lines represent
the Bard model. Groups that intermarry in at least one region of the Vaupes are
connected by dotted lines. The geographical locations of the intermarrying language
groups shown here are Cano Vina (DS and SR); lower Papuri (PR and UN); middle
Papuri (TK and TY); upper Papuri (CR, TT, and BR); Pira-parana (BS and TW).
Eastern Tukanoan languages (1967, p. 682), using the comparative method.

Sorensen states that "the resulting subgroupings definitely do not correlate with
phratric groupings" (1967, p. 674).
Sorensen's reconstruction is generally corroborated by the work on a recon-
struction of Proto-Tukanoan carried out by Waltz and Wheeler (1970). Table 9,
based on their work, shows the percentage of shared vocabulary for each pair of
most closely related languages.
As can be seen in Figure 7, many of the groups most closely related
linguistically also intermarry. This is also supported by Sorensen (1967, p. 674)
and, with one exception (ignoring the Cubeo, who marry within the tribal unit),
by Waltz and Wheeler's information on marriage patterns. Waltz and Wheeler
state that Bara marry with Tuyuka, Uanano with Tukano, and Carapana with
Tatuyo (they do not give marriage data on all language groups). The exception is
the Uanano, who are "forbidden to marry the Piratapuyo 'because they are
brothers.' " 1 0 The Piratapuya and Uanano, then, are the only linguistically valid
examples of the Tukanoan model, in which social and linguistic relationships
have a close connection.
Bard informants state that there is a close genetic relationship between
sibling-related languages and a distant genetic relationship between affinally
related languages. Sorensen's and Waltz and Wheeler's work indicate that this is
172
Table 9. Language distance as measured by cognates
Languages % cognates
Piratapuya and Uanano 99.2

Uanano and Tukano 91
Tukano and Papiwa" 93
Papiwa and Tatuyo 90.2
Tatuyo and Carapana 96.3
Carapana and Bara 91
Bara and Tuyuka 97.2
luyuka and Barasana6 93
Barasana and Makuna 98.3
Makuna and Siriano 94.5
Siriano and Desana 98.9
Desana and Cubeo 79
"Possibly Wahiina; if the same as Sorensen's Yuruti, the two reconstructions shown in
Figure 7 differ on which languages are most closely related to Yurutf-Papiwa.
Îdentical to Sorensen's Paneroa.
Source: Waltz and Wheeler, 1970; Chart 1.
Note: Tukanoans consider Taiwano and Barasana to be separate languages, and they are
listed as distinct in Sorensen, 1967. Waltz and Wheeler (1970) and S. Hugh-Jones
(personal communication) consider them the same language.
not true in the majority of cases. Frequently those languages that are genetically
the most closely related are those of language groups that do intermarry in one or
more subregions of the Vaupes. This suggests that an origin by fission from a
protolanguage might be the explanation for such presently separate languages as
Siriano and Desana, Carapana and Tatuyo, Bara and Tuyuka, and Uanano and
Tukano. This might have involved a previous situation where two moieties
speaking the same protolanguage with some differentiation in speech gradually
came to be identified with distinct languages. Evidence from the Barasana-Taiwano
split (which Tukanoans consider to be separate languages but which Waltz and
Wheeler and S. Hugh-Jones [personal communication] do not) supports this
assumption. Perhaps, if allowed to evolve, the two intermarrying Makuna groups
might eventually separate linguistically as much as some other intermarrying
pairs (e.g., Bard and Tuyuka, who share 97.2 percent cognates).11 Sorensen
(1967, p. 670) suggests that the Tuyuka and Yuruti languages (and language
groups) are perhaps derived from a common source.
Other evidence exists supporting the case of this type of origin for a number of
intermarrying pairs of language groups in the Vaupes. For example, the
Piratapuya retain a nonfunctioning moiety system. Goldman (personal communi-
cation) has information on an earlier Cubeo marriage system based on moieties
with fixed intermarrying ranks. Furthermore, we have seen that a number of
173
The fish people
Tbkanoan languages have a basically Dravidian type of kinship terminology,
which organizes kinsmen into dual structures of agnate-affine and is quite
compatible with a moiety structure.
With the exception of Waltz and Wheeler's information about nonmarriage
between the Piratapuya and Uanano, Tukanoan assumptions of relatively greater
linguistic distance existing between certain other pairs who call themselves
"brothers" is not supported by the comparative linguistic data.
Other well-documented instances exist in which social attitudes toward lan-
guage have led to the disregarding of purely linguistic evidence about relative
linguistic distance. Gumperz cites the case of speakers of certain German dialects
in Alsace-Lorraine who pay language loyalty to French rather than to German. He
also discusses situations where "speakers' claims that they do not understand
each other reflect primarily social attitudes rather than linguistic fact" (1968, p.
124). It is evident that attitudes toward language based on social criteria influence
popular notions about relative distance between languages, speech boundaries,
and the amount of energy invested in keeping two code systems visibly distinct.
The Vaupe"s obviously offers many opportunities for investigating how social
attitudes can, over time, affect the "purely linguistic criteria" upon which
measures of linguistic diversity are based.
Language groups: linguistic discreteness.

Tukanoan attitudes toward language in the context of the language group per se
might be characterized as separate but equal. That they are maintained as
separate enough for Tukanoans to state they are mutually unintelligible (which is
also the position of the Summer Institute of Linguistics) is doubtless at least in
part due to their role as badges of the language groups.
In addition to the feature of mutual unintelligibility is the one of equality.
Tukanoans do not agree with suggestions that some Tukanoan languages are
superior to others. This corresponds to a similar resistance to suggestions that
certain language groups are superior to others. This is expectable in a situation
characterized by marriage classes participating in a symmetrical direct exchange
system. In contrast to an asymmetrical system (Levi-Strauss, 1969; Needham,
1962), this type of system does not permit intermarrying classes to have different
status. It would therefore be surprising if the emblems symbolizing these classes
were seen as having greater or less value vis a vis each other. Status differentia-
tion does exist in the Vaup6s, but apart from the use of elder and younger sibling
terms among language groups in the same phratry, it is not an aspect of the
language group system. (See Zuidema, 1969, on this matter.)
Sorensen (1967, p. 679) states that Tukanos claim a mild prestige as the senior
clans of the area. I have heard Tukanos claim (in Spanish) that they were los
primeros ("foremost") among Vaupes Indians. It is also true that Bard address
Tukanos with elder sibling terminology, which generally is an indicator of respect
174
and higher status. Thus, Tukanos may be an exception to the rule of egalitarian-
ism among language groups.
Still, Tlikanos are in a somewhat special category, which may account for the
mild prestige they claim. They are apparently relative newcomers to the Vaupes,
and the distribution of Tukano settlements suggests that they displaced other
groups when they entered the region, forcing these groups into headwater areas.
This may indicate that in the past Tukanos were accorded respect because of their
military strength. Tukanos are also the largest language group and the most
acculturated. The fact that there are more Tukanos and that more Tukanos wear
Western clothing, are practicing Catholics, and speak Spanish than do members
of most other language groups may be sufficiently important to account for their
special status. Moreover, the Tukano language is the lingua franca of the region.
In any case, Tukanos do not have more power or authority than other
Tukanoans. It is my impression that Tukanos are more likely to claim such
prestige in their interactions with whites; open assertion of superiority in the
presence of other Tukanoans would surely be met with hostility. In addition,
although Bari freely admit that they address Tukanos with elder sibling terms,
they also assert that this is not the way it should be, because in fact Bara are the
elder siblings of the Tukano.
Vaupe"s languages are in no way ranked. If Tukanos do in fact have a higher
prestige, such prestige is not extended to their language. Despite the fact that
Tukano is a lingua franca, it is not considered preferable because of its greater
clarity or more extensive vocabulary or similar qualities. In fact, Tukanoans
resent the fact that change agents such as Catholic priests consider Tukano the
only language worth learning. I was told that the Summer Institute of Linguistics
has a much better approach than the Catholic missionaries, because its personnel
try to learn all the languages. The reluctance of Tukanoans to make invidious
comparisons involving language groups or language contrasts with their willing-
ness to do so with respect to different sibs' ability to speak a single language; this
is discussed in the following section.
This characteristic of egalitarianism among languages in the Vaupes is an
important one because in so many multilingual situations ranking is an integral
part of the system (Ferguson, 1959; Lambert, 1967; Labov, 1966). The situation
in the Vaupes clearly supports Gumperz's statement that "the common view that
multilingualism wherever it occurs also reflects deep social cleavages is clearly in
need of revision" (1969, p. 447).
Although unusual, the concept of partial or total language exogamy is not
unique to the Vaup6s. More research should be done regarding native attitudes to
codes, be they dialect differences (Sankoff, 1970, 1974) or mutually unintelligi-
ble languages, when these are connected to exogamous marriage classes in some
manner. Partial linguistic exogamy apparently occurs in small-scale societies
with greater frequency than is usually supposed (Owen, 1965; W R. Miller,
1970). Total linguistic exogamy has been reported for some Australian groups in
175
The fish people
which intermarrying moieties speak separate languages: "The members of each

moiety are supposed to speak different languages . . . Each class is said to have a
language" (Warner, discussing the Murngin, 1937, p. 30; also see Tindale,
1953).
Sibs: linguistic proficiency.

Sibs are named patrilineal descent groups, each one tracing descent from a
mythical ancestor. Each language group is composed of a set of ranked sibs, each
of which should, and usually does, occupy a continuous territory, although as we
have seen, this does not always fit the ideal pattern of one sib per settlement. Sib
rank is seen as derived from and equivalent to sibling birth order.
Tukanoans will account for differences in rank in the sibs composing a given
language group in a number of ways. One of these is by referring to the relative
proficiency with which members of a given sib speak Bara, Tukano, or whatever
father language the sib speaks. This criterion was the one most frequently offered
by Bara informants when asked why they addressed the members of lower-
ranking sibs as "younger brothers." These judgments are most often concerned
with lexical criteria, although occasionally pronunciation will be mentioned.
Whenever I asked why the Bara Yoara sib addressed the Wainakoroa or
Wamutanara sibs as younger siblings, they responded with statements such as,
"They do not speak Bara properly . . . listen to the [inmarried] Bard women of
the longhouses downstream . . . they use the wrong words." It is interesting to
note that at times a possible hyperform usage enters the picture (Labov, 1966). I
was told to use certain words and grammatical constructions I had never heard
used by the Bara of Pumanaka buro themselves. When I pointed this out, they
would respond, "This is the language of the Bara Yoara . . . this is the correct
language, the language of Mario's [the headman's] father and grandfather." It is
interesting that my speech was corrected for certain speech forms even though in
general there was a hesitance on the part of my informants to correct my speech.
For example, I was told to say bdrihke ("food") rather than bdrihe, the form I
always heard. I was told always to use the /hk/ in words having that suffix, which
is a frequently occurring one. Whether this was the common form at an earlier
period, whether the /hk/ is used in another, more prestigious dialect of Bara
(quite possible, despite the claims of Pumanaka buro to belonging to the
highest-ranked sib; such claims should always be taken with a grain of salt), or
whether it is an example of the way in which a sound change begins, I cannot say.
When I protested, saying that I wanted to speak like the rest, they would say that I
had to learn the proper form in order to demonstrate to other Tukanoans that I was
being taught the correct speech, the Bara of the Bara Yoara sib. These Bara
said that the Summer Institute of Linguistics missionary living with the
Wamutanara Bara (the lowest-ranking sib) on Cafio Colorado in the Pira-
parana" region would never learn to speak proper Bara, because he had chosen to
176
live with Bara who could not speak it properly themselves. During my stay I
collected numerous examples of improper forms spoken by lower-ranked sibs and
of "more proper" forms, now "lost," which had been in the language of the
ancestors of the higher-ranking sibs.
The importance of language in l\ikanoan culture

The discussion of Tukanoan attitudes toward language distance, discreteness, and
correct speech gives an idea of the overall importance of language and speech in
Tukanoan culture and society. Aspects of language and speech are markers of
identity in several kinds of social groups and categories in the Vaupes. In
addition, language - as a concept and in the sense of linguistic performance - is
extremely important in the Tukanoan model of society and, indeed, the universe,
and linguistic criteria symbolize many crucial facts of Vaupes life.
Sorensen points out that "Indians are quite unself-conscious about their
multilingualism" (1967, p. 679), and this is true in the sense that a polyglot in a
society in which all members are multilingual is not the remarkable individual he
or she is in American society. Nonetheless, Tukanoans possess distinct father
languages, and the ability to speak, to speak well, and to be able to use power
available from linguistic performances are considerable assets for those individu-
als and groups that have such abilities and can claim such prerogatives.
Tukanoan mythology frequently refers to a connection between power, humani-
ty, or life itself, and the ability to speak (and, by analogy, to make sounds).12
Shamans speak an esoteric language in ritual, which nonspecialists cannot
understand (S. Hugh-Jones, 1979, pp. 58-59). They validate their claims to their
special status by their ability to communicate with nonhuman beings in this world
and in other levels of the universe. Much of the respect they acquire in their
career is due to these special linguistic abilities.
In other areas of Tukanoan life, speech is one of the means of distinguishing
among classes of people in terms of status and prestige. A command of ritual lore
and knowledge and an ability to speak effectively and beautifully are ways of
distinguishing high-ranking sibs from low-ranking ones, men from women, and
old men from younger men (Reichel-Dolmatoff, 1971, pp. 249-52). Thus,
formal occasions provide opportunities to demonstrate one's grasp of esoteric and
ritual knowledge and of etiquette in general. In fact, speech itself plays many
important performative roles (Austin, 1962), and in this capacity is crucial in
Tukanoan ritual and symbolism (Bidou, 1976; S. Hugh-Jones, 1974; Jacopin,
1972). The ability to perform chants and myth narrations well is important in
tapping the power of the ancients.
Language also distinguishes humans from nonhumans, although in some
instances this is not signaled simply by the presence or absence of a language.
Each species of animal is seen as having its own language, as do the ubi-
quitous wahtia, forest spirits. All the other denizens of the universe, insofar as
177
The fish people
they have any humanoid characteristics, have speech. For example, during my
fieldwork, at times an uncanny noise would be heard in the sky right before a
thunderstorm. I imagined this to be bird cries that were greatly distorted by the
unusual acoustics of the very low clouds. My Bara informants told me that this
was "sky-people language," which becomes loud enough to be heard when the
sky people become excited about something and start yelling and cleaning out
their longhouse.
Many other examples can be given. One way Tukanoans distinguish them-
selves from Maku is by saying the Tukanoans speak a superior, truly human
language. Goldman (personal communication) states that a translation of the
Cubeo self-name Pamiwa is "people of the language."
This chapter has been limited to a discussion of certain aspects of social
structure and the way in which these are incorporated into the Tukanoan model of
linguistics. It is evident that much more remains to be studied and analyzed if we
are to acquire a comprehensive picture of the importance and function of
language and speech in Tukanoan culture.
178
10.
Male and female identity
It is difficult to discuss Tukanoan sex roles and identity comprehensively in a single

chapter. First, an entire book would be required to do justice to the necessary
ethnographic material, particularly with respect to the extraordinarily rich ritual
and mythological expressions of sex differences. Fortunately, other investigators
have discussed some of the symbolism of Tukanoan sex roles.1 (Sex roles refers
to all of the ways in which Tukanoan males and females behave differently, see
themselves as different, and symbolically express these differences.)
Second, the nature of the subject matter itself requires more extensive treat-
ment than is possible within one chapter. The fledgling state of the field of sex
roles as a topic in its own right creates certain difficulties, which have been
discussed in many articles and books. Both our own society and lowland South
American societies see men and women as fundamentally different (see Jackson,
n.d.-a.). "Vet what does this mean? Seemingly insoluble problems arise in
connection with objectivity and agreement over basic assumptions and correct
research design and goals. How do we evaluate the differences existing between
the sexes in all Amazonian societies in terms of crucial questions concerning
status, power, solidarity, egalitarianism versus hierarchy, among others? What do
these differences mean, comparatively speaking, in a society so unlike our own,
a society where age and sex are the main criteria of social differentiation? Yet we
can hardly accept uncritically the "inside" view - the native interpretation - of
sex differences, even supposing there were a single such view. It is a truism that
people are often unaware of significant structures and dynamics in their own
society and, when convenient, ignore and distort others. Such systematic (i.e.,
ultimately explainable) distortions occur as much, if not more so, in the area of
sex roles as in other areas of social life.
Still, it is impossible to imagine a book on Tukanoan identity without a chapter
on at least some of the issues concerning relations between the sexes. This is
particularly important when we consider the prominent place that material from
horticulturalist societies, with South American societies prominently included,
occupies in many of the more theoretical debates on sex differences.
Lowland South America has been the focus of a number of inquiries into
179
The fish people
male-female relations, particularly with respect to questions of opposition and
antagonism between the sexes. One important debate involves whether these are
expressions of "real" male oppression or only wishful thinking on the part of
men in the face of strong contrary evidence. A related question concerns the
actual amount of female "power" in these societies and to what degree it results
from female contributions in economic production and distribution, reproduc-
tion, and other areas.2
Too many discussions of the position of women in horticultural societies are
uninformed, simplistic, and naive. Frequently such concepts of high or low
"status" are used in cross-cultural comparative studies with shockingly impre-
cise definitions. As Lamphere points out (1977, p. 613), the same can be said for
"asymmetry," "subordination," "equality," or "power." Serious problems
arise in selecting and defining equivalent units of comparison, regardless of the
type of evidence (e.g.. economic, symbolic) being marshaled. How can the
differences between image and reality found in all cultures be adequately
translated into an analytic code for comparative purposes? Of course, this is a
general problem in description and comparison - and translation - but the
difficulties seem especially thorny with respect to the study of sex roles and
identity. This is in part because of the general tendency for anthropologists to
investigate, record, and publish only the official, idealized version of cultural
patterns and mores, or at least to give undue emphasis to them. In all cultures
male and female views of reality diverge, at times extensively. Nevertheless, both
men and women in a given society tend to give any outsider the official,
male-oriented view, and research concerned with discovering differences between
the sexes will be grossly inadequate if the investigator does not probe more
deeply. The dangers of such inadequacy are increased when the researcher is
male, talks only to males, is unaware of his own ethnocentric assumptions about
sex differences (which we all have to some degree and are all somewhat unaware
of), and is satisfied with a "rule-book" version of culture. To use such
information for drawing cross-cultural generalizations about sex roles is obvi-
ously unwise; sad to say, at times such reports are the only ones available for a
given society or region.
Differences in sex roles and their causes are obviously important aspects of life
for all of us, and few fields offer such an abundance of explanatory models and
theories, each one predicting horrible yet inevitable consequences should its
implications be ignored. Of course, some of the more bizarre or dogmatic
explanations have little effect on anthropological research and analysis, but the
existence of so many theories argued from such divergent premises poses a
serious problem to scholars when conducting research, be it devoted to a single
culture or comparative in scope.
These comments are intended not only as warnings with respect to the
reliability and validity of ethnographic sources dealing with these topics and the
theoretical arguments buttressing them but also as reasons for my hesitancy to
180
draw any sweeping conclusions about sex-role differences and their determinants
among Tukanoans.
This chapter looks at some "givens" in social structure, economic production,
and reproduction in Tukanoan society and examines some of the cognitive
models that Tukanoans have evolved for expressing and making sense out of these
givens. Some tentative conclusions about Tukanoan views regarding sex-role
differences are proposed: (1) To some extent in Vaupes society, women are seen
as "outsiders," or "the other," to use de Beauvoir's (1953) terminology; (2)
women are seen to be relatively more "natural" than men (who are more
"cultural," although this is not a simple association by any means);3 (3) women
need to be ' 'controlled" in a number of ways, at times because of their superiori-
ty, at times because of the danger they pose, and at times simply to avoid an
excess of what is in itself not a negatively evaluated quality or trait.
It is important to note that although it is true that Tukanoans recognize differ-
ences existing between the sexes in all areas of life - and this recognition is ex-
pressed in spatial arrangements segregating the sexes, in economic tasks, in social
interaction, in ritual, and in myth - such opposition and, often, antagonism are
not the entire picture. The processes of synthesizing and permuting male and fe-
male symbolic elements are also very much in evidence. Above all, it is important
to bear in mind the virtual absence of absolutes that can be lifted from the struc-
tures and contexts in which they occur without great distortion. What is one thing
at one level is something quite different - at times the reverse - at another.
Relations between men and women

Probably the most important social structural factors affecting female status in the
Vaupes are (1) the organization of female economic contribution (particularly
food production); (2) patrilineality and patrilocality; (3) the traditional settlement
pattern of communal multifamily longhouses, one per settlement; and (4) the rule
of direct exchange at marriage. Other significant features are undoubtedly the
prescription for language exogamy, the kinds of relationships formed between
affinally related settlements, the low incidence of polygyny, the fact that mar-
riages are fairly stable after an initial period of a few years, and the (current)
absence of raiding and feuding.
It cannot be denied that Vaupe"s society is male-dominated. Male activities are
valued more highly than female ones. As was pointed out in Chapter 4, a
powerful mystique surrounds hunting, male sexuality, and male-controlled reli-
gion, including sacred objects and power forbidden to women. Only men are
thought of as true spiritual beings (S. Hugh-Jones, 1977, p. 13). Male-focused
ritual dominates Vaupe's religion, and men can be seen as ritually controlling
female (as well as male) areas of life, such as menarche and birth. Male rituals
are needed to make relationships with females safe, ordered, and moderated -
keeping women at a proper distance describes much of what these rituals are
181
The fish people
intended to accomplish. Rituals control women and female symbols in a variety

of ways - for instance, by neutralizing potentially destructive female elements, or
limiting the force of others. Other rituals can be seen as expropriating female
power and creativity for male-oriented uses. An example of the latter is the male
initiation ceremony, which for most of the ritual excludes noninitiates, particu-
larly women, yet which on the symbolic level utilizes many potent female
images, sources of power, and kinds of creativity. Thus, these rituals are both an
acknowledgment of female power and evidence of male dominance, for they
utilize female power to promote what are predominantly male interests, such as
the continuation of the patrilineal descent group. At times women are symboli-
cally "sent back to where they came from, outside and beyond male society" (S.
Hugh-Jones, 1977, p. 210).
Much evidence can be found in myth as well as ritual for expressions and
justifications of male supremacy. Females are variously depicted as dangerous,
selfish, treacherous, and nonhuman. They are enemies who can never really
become friends, destructive to "good" male interests, and threats to male
solidarity. Although this is not the only kind of female image contained in
Tukanoan myths, nor are negative images the only ones of women to be found in
Tukanoan ritual, neither is the picture one of equality, a picture where male and
female spheres are separate but complimentary and equally valuable. The real
situation lies somewhere in between these two poles and is far more complex.
To understand these images more fully requires a careful examination of
everyday relations between the sexes, in particular of the relative amounts of
power, autonomy, and options available to each. This allows a better understand-
ing of the symbolic plane, not only to see whether such ritual expressions of male
dominance are "accurate" reflections of daily arrangements but also better to
understand why male dominance is asserted so vehemently in certain rituals.
Native cognitive models, whatever their medium of expression, in part exist as
mechanisms of perpetuating, rather than actually explaining, social reality
(LeVi-Strauss, 1962). They also have an internal logic that to some degree
determines their nature apart from any "function." Thus, it is quite probable that
myth and ritual sometimes buttress unsteady male positions, giving expression to
anxieties and insecurities about the inevitability and justice of male superiority,
serving as just-so stories. This argues for initially separating on-the-ground
reality from symbolic reality in any analysis of male and female roles - although
at a later stage these two levels must be combined.
Food production.
The food Tukanoan women produce and process is a substantial contribution to
the Tukanoan diet, although given our state of knowledge about diet in such
horticulturally based societies, it is impossible precisely to compare male and
182
female contributions in terms of actual nutrition. Women prepare almost all
food.4 In addition to processing and cooking, women provide most cultivated
foods and many kinds of foraged foods. It is the women who plant, tend, and
process almost all the edible cultigens (i.e., excluding coca, banisteriopsis, and
tobacco), although men are responsible for some fruit trees, some corn, and
pineapple plants. It is mainly women who gather fruits, vegetables, nuts, and
small insects.
In general men and women are viewed as complementary with respect to
subsistence contributions. Fish and game, which supply protein, are the most
highly valued foodstuffs, but manioc is the staff of life and other foods cannot be
eaten without it. One can, and does at certain times of the year, go without
male-supplied foods for one or more days, but it is unthinkable to get through a
single day without manioc products. This, added to the fact that women transform
virtually all foodstuffs from a raw to a ready-to-eat stage, means that men are
constantly reminded of their dependence on women for food. If a man has no
wife, he depends on his mother or unmarried sister for his meals. Lacking these,
he must eat at his brother and sister-in-law's hearth. The strict sexual division of
labor prohibits men from engaging in those aspects of food production and
preparation assigned to women, just as women cannot clear fields, hunt, or fish.
When men discuss women, their role as food producers and processors is
mentioned time and again.
In sum, sex-specific activities related to subsistence are markedly segregated,
both in terms of who performs an activity and die symbolic meanings associated
with it.5 An observer is given the impression of interdependence and complemen-
tarity, but this complementarity, although very real in both objective and concep-
tual terms, is not the whole story.6
Tukanoan ritual and symbolism portray male and female roles in subsistence in
many ways. For example, although women are excluded from much of the male
initiation ceremony, certain women are crucial to it, such as those who prepare
the food eaten by men and initiates. Furthermore, throughout the male initiation
ceremonial cycle, many symbolic equivalences between women and female-
related subsistence activities, especially concerning manioc, are ritually ex-
pressed. Another example is found during food exchange ceremonies, when the
two communities adopt male and female symbolic stances vis a vis each other.
The host community, receiving smoked meat, is likened to a woman (who
similarly receives game from a man); the food is brought in through the men's
door and the host group provides beer, a female contribution (C. Hugh-Jones,
1977, p. 200). Rituals celebrating the harvest of jungle fruits, in particular the
miritf palm fruit, are replete with sexual symbolism. I was present at an attenu-
ated one, which involved canoes laden with miritf fruit arriving at the port
accompanied by Yurupari horns; at the arrival the women left the longhouse for
the fields, hearing the horns during the ceremony but not returning to the
183
The fish people
longhouse until the men and horns had left again. In the past these harvest
ceremonies were more developed (Goldman, 1963, p. 192), sometimes express-
ing more overt antagonism between the sexes.
Apart from symbolism contained in ritual, one finds many kinds of associa-
tions made between men and women and food or agriculture. For example, C.
Hugh-Jones discusses the agnatic core of a community as the respective recipi-
ents or "consumers" of children, just as they are of women's economic produc-
tivity: "There is no doubt that manioc is regarded as women's offspring for
women are called 'manioc-stick food mothers'. . .in ritual language" (personal
communication). She also notes a correlation made between the umbilicus giving
nourishment to the foetus and the path from the longhouse to the manioc fields
(1976, p. 14).
In subsistence as in other areas, men ritually control women and their prod-
ucts. Mother's milk and all other types of food must be shamanized before they
are safe for consumption. Cooked food - associated with female symbolism in a
number of ways - is dangerous in itself and must be understood and eaten under
the proper circumstances and never to excess. As was noted in Chapter 4, a
complicated series of food rituals exists that both restricts and makes safe the
various categories of food for people going through various kinds of crises.
Domestic roles.
As in the case of subsistence-related activities, there is a marked distinction
between the sexes in domestic activities and an interdependence, although an
absolute balance does not exist either in complementarity of tasks or in how
Tukanoans conceive of this interdependence. Women make bags of Mauritia
flexuosa fibers twisted into string, and from Bmmelia fibers they make the ritual
garters worn by men (the latter is a rare example of women manufacturing
ceremonial items). They also wash and mend all clothing, sweep, gather firewood,
and fetch water. All pottery, including items used exclusively by men, is made by
women. Men construct all the basket and wood items used by both sexes, build
the houses, and make canoes, fishing gear, and hunting weapons. They manufac-
ture nearly all ritual equipment. True, much of men's time is devoted to mascu-
line activities - such as coca-growing and processing - but because men are
supposed to be the spiritual guardians of the longhouse, such activities are seen
as necessary to the well-being of the entire community. Women's activities more
directly serve their families.
With respect to actual work effort, it was my impression that women worked
longer and harder than men. Still, I did not keep work diaries and am very aware
that such a statement is subjective, because I spent my time mostly with the
women. One must also assess the impact of acculturation on the work loads of each
sex. Women, for example, must now produce a surplus of manioc, a truly arduous
task, in order to make farina to sell to missionaries and rubber gatherers. The
184
Western-style clothing now worn by Tukanoans means a substantial increase in
time spent mending and washing, which sometimes must be done without soap.
In conclusion, regardless of the actual proportion of time spent in "work," it
is unquestionable that women's economic contributions, both with respect to
food production and other tasks, are considerable, necessary, and seen as such by
all lukanoans. Probably the fact that women produce the only cash crop, farina,
has made men's dependence on them especially apparent in recent times.
Roles of wife and husband.

Tukanoans see marriage as difficult to accomplish from a man's point of view but
say that any woman could get married if she wanted to. It is probably true that
from the perspective of any individual Tukanoan, arriving at a happily married
state is difficult to accomplish regardless of whether one is a man or a woman.
Nonetheless, all Tukanoans recognize that women make more sacrifices upon
marrying. This fact has implications for any study of Tukanoan sex roles.
A woman's marriage benefits, besides herself, three parties: her husband, her
husband's local descent group, and her own local descent group. Aside from gain-
ing the services a wife provides (which her spouse reciprocates), a man also
achieves the status of married male and can establish a separate compartment in
the longhouse. Furthermore, he is on the way to achieving full adult status, which
comes when his wife grows manioc in the fields he has cleared for her and bears
children. A woman's status also improves upon becoming a wife and later on a
mother; however, in some senses her status as a new wife is the lowest position
she will ever have in her life - in terms of the realpolitik of a longhouse commu-
nity. Furthermore, because Tukanoans see marriage as difficult to accomplish for a
man, getting married is seen as much more of an achievement for him than for a
woman. To congratulate a woman on having caught a man would be nonsense.
Obviously statements that all men would get married if they could and all
women could get married if they only would are to some extent cover-ups, that is,
rationalizations about marriage and a glossing over of the fact that women
frequently do want to get married and clearly see benefits for themselves from
doing so. Tukanoans understandably play down the idea that women marry out of
self-interest, for this implies that they are willing to forsake their agnatic kinsmen
and contribute their companionship and labor to an affinal longhouse, increasing
its strength by providing it with the new generation. It is easier to see women as
getting married for the sake of their brothers, thus enabling them to obtain wives.
A woman's husband's local descent group benefits from the marriage because
she will contribute to it economically and help provide it with sons and daugh-
ters. As is true in all patrilineal and patrilocal societies characterized by direct
exchange or brideprice, Tukanoans regard the arrival of a daughter-in-law to a
longhouse a difficult and highly desirable accomplishment.
Finally, a woman's own local descent group benefits from her marriage because
185
The fish people
although they are losing a close agnatic kinswoman and the benefits of her eco-
nomic contributions, her brother is thereby enabled to obtain a wife. This new
daughter-in-law will replace the daughter in terms of food production and domes-
tic services. Because only women may be exchanged for women, and sisters can-
not provide sexual services nor reproduce the local group, the men must lose their
sisters. Furthermore, a woman's marriage helps either to create or to strengthen
an exchange relationship between the two local descent groups - an alliance.
In one important respect the Vaup6s situation differs from certain other
patrilineal or bilateral societies, in which women are reared by their families only
to be sent as wives to other groups, frequently involving dowry expenses met
only at considerable sacrifice (M. Wolf, 1974; J. Schneider, 1971; Michaelson
and Goldschmidt, 1971). The contrast is particularly strong in some peasant
communities where the outmarrying women of a local descent group are seen as
"rubbish." This difference in attitudes toward the sisters and daughters of a
local group is in part due to the system of direct exchange in the Vaup6s and the
importance of Vaupe"s women's economic roles.
Nevertheless, although Tukanoan women do derive benefits from marriage -
such as full adult status, sexual expression, and parenthood - the costs of this
action are far greater for them than for men. No matter how close her husband's
longhouse is to her natal longhouse, it is the woman who must leave, permanent-
ly. In the Vaup6s, the emotional stresses that accompany this move are many and
in some cases extreme. Residential groups are isolated, autonomous, and (at least
ideally) tightly knit, making it difficult for a daughter to leave or a daughter-in-law
to enter. Furthermore, not all affinally related longhouses are near one another,
with the result that those women who marry into distant settlements are able to
visit their natal longhouses only sporadically if at all. Even if a woman moves
into her mother's natal longhouse, she nonetheless moves into a house with
people whom she does not know nearly as well as those she has lived with all her
life. These new in-laws speak a different language from her father language and
are ajfines, regardless of whatever consanguineal relationships she might have
with them. The anthropological literature is replete with examples of difficulties
existing between in-laws, even those with the best intentions. Such built-in
stresses certainly exist in all Tukanoan affinal relationships, as was discussed in
Chapter 5, and they are particularly painful to the individual living mainly among
affines. A new wife must not only adjust to husband and new surroundings but
also spend a large part of her day with her mother-in-law and sisters-in-law,
whereas before she was with her own mother and sisters (and sisters-in-law). As
is true in many societies (for example, see Collier, 1974), the period immediately
following marriage is extremely difficult for the young wife, regardless of her
love for her husband, and in terms of personal influence and security is probably
the lowest point in a Tukanoan woman's life.
A man, on the other hand, undergoes few changes when he marries, and most
of these are, from his point of view, positive or neutral ones - excepting those he
186
experiences because of his wife's difficulties in adjusting. He stays in his own
longhouse with people he has known since birth. Even though he must adjust to a
new wife, this problem is alleviated by the fact that interaction between newly-
weds is kept at a minimum. His position at this time is pivotal, fluctuating
between loyalty and concern for his new wife and his continuing loyalty and
obligations to his local group, in particular his mother and sisters.
New wives' being "strangers" in their new homes in certain ways throughout
their lives is a crucial theme in Tukanoan culture. Symbolically, the idea of a
united patrilineal core and separated, female peripheral units is expressed in the
longhouse structure itself (C. Hugh-Jones, 1977, p. 197), in ritual, in ethno-
geography, in concepts of time, and in many other areas (also see Goldman,
1977). Opposition exists between the core group and the separate, structurally
equivalent units represented by the wives, an opposition between insiders and
outsiders. This is made more apparent by the fact that in the majority of
longhouses several languages are spoken and represented by the inmarried wives.
The theme of antagonism between affmally related people is ever-present, and
even though raiding for women has ceased, the ideology is still very apparent.
This cannot fail to affect a woman's attitude toward becoming and being a wife as
well as to influence Tukanoan conceptualizations of women and female elements
in general. The effect of marriage practices, therefore, is that men live together in
an agnatic core with feelings of solidarity, ownership of the community, and
dominance as men, whereas women are seen as individuals, outsiders, on the
periphery of the longhouse group, and related to each other via the links they
share through the men. (Of course, some women in a longhouse can in fact be
classificatory or real sisters, but it is the ideology that is important.)
Although women are always peripheral as wives in the longhouse group, they
are necessary for its survival. This means that marriage, and more broadly
speaking, women, must be supervised and controlled both practically and ritually
by men. There is no doubt that men are the agents arranging marriage and that
Tukanoans see things this way, although, as noted earlier, Tukanoan women are
not nearly as powerless regarding whom they will marry as the women in groups
like the Yanomamo seem to be. An association exists between continuity of
growth of the longhouse community and its destruction by women, because
marriage breaks up the sibling groups of an established generation (C. Hugh-
Jones, 1979, p. 161). Women, thus, are seen as "potentially creative as sisters
and actually creative as wives."7 The general theme of destruction leading to
creation, renewal, and growth permeates Tukanoan thought. Women are seen as
both destructive and creative; the essential process of the one evolving into the
other occurs by exchanging women, and to be successful it must be regulated by
men. These themes, derived from Tukanoan social structure and other areas, are
symbolically expressed in many areas of Tukanoan cosmology.
Finally, by leaving one group as sisters and coming into another group as
wives, women are the expression of reciprocity between groups. This is so
187
The fish people
fundamental a symbol in Tukanoan myth and ritual that one encounters its
expression constantly; many examples are found, for instance, in the synopses of
myths in Chapters 6 and 7. 8
Roles as parents.
As in all patrilineal descent systems, it is the nonagnatic, outsider women who
supply the local group with the new generation. This is why daughters-in-law are
ultimately of greater value to the local group than the daughters it gives away, an
interesting fact because of its indisputable effects on Tukanoan conceptualiza-
tions of women in general. The need for women as procreators, accompanied by
their necessarily equivocal position as outsiders, and the ensuing anxiety about
their loyalty and devotion as mothers and wives are reflected in several Vaupes
myths, most notably the ones surrounding the origin of the Yurupari instru-
ments.9 The following is a synopsis of one version collected in Bara:
At one time, after they were made, these horns fell into the hands of the women, who were
not as lazy as the men. The women would get up earlier in the morning and go down to the
river to bathe before dawn. These women obtained the horns by being clever, for the horns
succeeded in hiding themselves from the women for a while. The women kept the horns
stored in their bodies, first in the humerus (which is why women's elbows are shaped
differently from men's), and then in the vagina. The men could not get the horns back. The
women grew very strong, and refused to have sexual relations or bear children. The men,
realizing that the people would cease to exist, asked a shaman what to do. He suggested a
trick, which worked, and the men recovered the horns. Since then the men have got up and
bathed earlier than the women. And also since then women cannot see the horns, for they
would fall sick and die.
It is obvious that one theme of this myth is an expression of anxiety about

women becoming too powerful or seeing their interests as totally at odds with
men's. That it is an anxiety based on fact has already been illustrated in some of
the myth synopses given in Chapter 7. Although a woman identifies very closely
with her children's descent group, she remains a member of her own throughout
life. It is significant that Tukanoans whom I talked to did not know whether a
woman goes to her natal spirit longhouse at death or to her husband's (and
children's). The facts of social structure mean that although women naturally
share an uninterrupted physical relationship with their children from the time of
conception (which the couvade in part allows fathers to do), they are never linked
socially with their children. Thus, male control of women not only breaks up
actually existing female solidarity groups by sending out sisters in exchange for
wives, but mothers are never identified with their daughters in terms of patrilineal
group membership. This continuity in the male line is expressed ritually in many
ways: Men give their children "soul-stuff" (C. Hugh-Jones, 1977, p. 188) by
various associations made between soul and semen, bone, and names. It is the
men, rather than women, who in their role as parents, create the continuity
188
between generations seen as social units. Thus, the themes of women as outsider,
as "destructive" (in that by leaving they break up the longhouse group), and as
needing to be ritually controlled also pertain to women as mothers. It will be
recalled that all food, even mother's milk, must be made safe by male shamans
before being given to babies.10 Even when female power is seen as neutral or even
good, it still must be supervised by men. The fact that men who warm themselves
too much at a hearth fire will procreate only daughters is a good example of
Tiikanoan assumptions that an excess of female power is inimical to male and
patrilineal group interests.11 This is also an example of the overriding theme of
moderation and control in general found in Tukanoan thought, which particularly
applies to the need for a balance in male and female components and energies.
Male and female sexuality.

Socially sanctioned and openly acknowledged sexual behavior occurs only within
marriage. Of course, this requires a narrow definition of "sexual behavior." Men
will lie together in hammocks fondling each other's genitals, and women will
stroke one another's bodies (including breasts) and play with the genitals of
children of both sexes. To speak subjectively, although Tukanoans are not as
uninhibited about sex as, say, the Tahitians, in many respects (certainly with
regard to touching and joking), they are not pathologically repressed about it.12
Sexual relations that are not approved of certainly occur, as attested to by
reports from Tukanoans of cases of incest and adultery. Goldman has noted a
relationship that can develop between a Cubeo brother and sister in which many
kinds of sexual behaviors are permitted with the exception of intromission (I did
not obtain data regarding this type of relationship). It is possible that pregnancies
occasionally develop in these relationships because the prohibition on intercourse
is disregarded. C. Hugh-Jones reports that premarital affairs are common (1979,
pp. 160-61) and that these can and do occur between people who do not stand in
the relationship of potential marriage partners to each other. The settlement
pattern obviously affords little opportunity for a sexual relationship with someone
who is neither an agnate nor already married.
It is obviously very difficult to draw conclusions about sexual activity, particu-
larly with regard to frequency and participants, because most serious sexual
encounters in the Vaupes (as elsewhere) take place in private. Attempting to
describe Tukanoan attitudes toward sexuality is a bit easier than discussing the
behavior itself but is still a topic full of pitfalls. Although I am not certain
whether women are seen as wanting sexual intercourse as much as or more than
men (to say nothing of whether they do in fact), the Victorian stereotype of the
sexually unresponsive wife who obliges her husband certainly does not apply. In
fact, Tukanoans place more emphasis on the possible damage to men and
patrilineal groups in general resulting from excessive or improper sexuality rather
than on the possible damage to females.
189
The fish people
I have already noted taboos on sexual activity during sacred periods and before
hunting and raids. Male sexual activity is linked to a general loss of male potency
and purity. The themes of women as outsider and polluter are pertinent to this as-
sociation as is the overall theme of potential danger resulting from excessive con-
tact with women regardless of the activity. Women are, then, definitely not seen as
sexually passive - they are accused of being the instigators in many adultery cases
(Goldman, 1963, p. 150). It was the woman who was blamed for instigating the
case of incest occurring at Puemanaka buro during my stay. Again, the actual
behavior of women in initiating illicit sexual relations is difficult to ascertain.
On the one hand, in any patrilineal and patrilocal society, one would expect out-
sider, troublemaker women to be blamed for initiating affairs regardless of the
facts because they are the most convenient scapegoats. On the other hand, adultery
is one of the strategies available to a newly married woman (or a disgruntled
woman of any age) for making her presence felt and her needs known. This can
be an effective ploy, despite the fact that it involves reprisals.
The foregoing characterization of Tukanoan women as the opposite of the
Victorian ideal of innocent, passive women - in that they are sexually active and
will initiate sexual encounters and are therefore potentially disruptive because of
their sexual drives - is found in many other accounts of horticulturalist groups,
particularly in the Amazon Basin and Highland New Guinea.13
Male sexuality, although in some senses just as "natural" as female sexuality,
is so very much linked to ritual that it is both difficult and probably unwise to try
to separate them completely. To some extent both women and female imagery are
seen as more natural than male symbols, and this of course applies to male
sexuality as well as to other areas. In other words, men in general are seen as
more "cultural" ("spiritual," or "social"). Furthermore, many functions asso-
ciated with the sexuality of both men and women (i.e., fertility, conception,
gestation, growth, nourishment) are explicitly carried out by men in the ritual
sphere far more than by women. Thus, although it is true that, for example,
banisteriopsis vines and bone are explicit male sexual symbols,14 both phallic in
shape and inseminating (as are many other symbols, such as the Yurupari
instruments), they also represent other, more female elements, depending on the
context. For example, banisteriopsis is associated with breast milk and the
umbilical cord in some rituals. A good example of this is the overtly male ritual
of Yurupari, in which the initiates are openly compared to menstruating women,
ritually imitating the loss of menstrual blood, dying, and being reborn (S.
Hugh-Jones, 1974). Although in one sense menstruation will always be a female
symbol, it is not sufficient to state that this rite is an imitation of female
menstruation and leave it at that - it may be enough for the anthropologist, but for
Tukanoans this is a male rite involving male menstruation, which accomplishes
real tasks for male initiates. The rite does not only imitate, it expropriates female
power available through the menstrual process of losing skin (blood) and growing
190
new skin. The skin change as indicative of immortality or a longer and healthier
life is one of the most powerful idioms of growth (C. Hugh-Jones, 1977, p. 189).
Another example of this is the male sexual symbol of the tipiti (manioc
squeezer), which, according to C. Hugh-Jones, is male both because of its
phallic shape (termed - in Bara - pino-wu, "anaconda-long, hollow instru-
ment") and because it exudes "urine." In other contexts, however, it is female,
for example, in its association with manioc.
The most sacred ritual objects and concepts in Tukanoan culture share a dual
sexual imagery (see S. Hugh-Jones, 1979, with respect to werea [beeswax],
perhaps the most sacred substance of all).
One must be careful to avoid oversimplifying the emotional meaning of male
and female symbols in either their negative or their positive value for Tukanoans.
On one level, menstruating women are dangerous, polluting, and must be
controlled and at times avoided. On another level, as we have seen, menstruation
has strong positive, creative, life-giving associations. C. Hugh-Jones (1977, p.
189) notes that the destructive quality of menstruation is limited to the "loss"
phase and it is at this time that women are to be avoided. This is tied to a general
theme of a two-phase, destruction-creation symbolism. Destruction and loss,
almost always involving dispersal, is the only means to construction and creation:
Loss of sisters means acquisition of wives and mothers of the next generation;
loss of vegetation (in cutting and burning the forest) allows for planting fields;
and loss of blood must take place before a new lining of menstrual blood and its
benefits (longer life, the possibility of conception) can be realized. Obviously,
this complexity of symbolic meaning argues against a too-pat association in
cross-cultural studies between menstrual contamination beliefs and an assump-
tion that women and menstruation are only seen as polluting and dangerous (see
Kelly, 1976, with respect to sexual intercourse).
The same complexity is found in the Yuruparf rituals. These unmistakably
symbolize - to Tukanoans as well as ethnographers - male dominance and
superiority. But in addition to rejecting women (both actually and symbolically)
and demonstrating the dangers of female sexuality, they are also an expropriation
of female power - an excellent example of flattery, envy, and nonrejection.
Furthermore, the transformation in Yuruparf ritual of these female elements into
male power is beneficial not only to the initiates and the community of men
(usually, but not always, the core of agnates living in a circumscribed area, for
example on a section of a river) but also to the entire community and society as a
whole. The same can be said for male sexuality in general. On one plane it is
opposed and antagonistic to female sexuality. Some female symbols in the He wi
rituals (e.g., the beeswax gourd) are dangerous to real women (S. Hugh-Jones,
1979). Certainly the restrictions on male sexual behavior contain many state-
ments of the danger and destructiveness of female sexuality. But male-centered
rituals and restrictions are also the route to the ancestors, to "good" energy, to
191
The fish people
reciprocity with the cosmos, to growth and continuity - without which all
Tukanoan society would certainly die.
Thus, in the final analysis, male and female sexual symbols are not only
opposed and seen as mutually hostile. They, as well as all symbols that are seen at
some time as male or female, can take on a dual sexual imagery, combining
whatever particular aspects of both are appropriate to the specific symbolic task
at hand. As C. Hugh-Jones has so eloquently stated, such a coming together
"creates a unity of a higher order than the creative elements. Semen and blood
make a foetus; male-produced fish and female-produced manioc make a meal,
and men and women make a heterosexual community who eat the meal in the
center of the house" (1977, p. 203).
Conclusions
In this chapter we have looked at some of the differences between male and
female roles in Vaupes society - as they appear to an outsider, as they are
conceived of by Tukanoans, and as they are symbolically expressed in ritual and
myth. The contributions both sexes make in the subsistence, domestic, and
procreative areas show that women are seen as valuable and powerful individuals,
both in and out of marriage. Yet strong and unmistakable avowals of male
supremacy are found in everyday, ritual, and mythic life. It is clear that at times
women are seen as potentially dangerous, as definitely dangerous when contact
with them is excessive, and as in need of ritual and practical control.
The disadvantages and disruptions encountered by women when they marry,
particularly at first, are beautifully and artfully captured in myths about affinal
relations; this transition, never fully accomplished, is one of many factors
contributing to a view of women as ambivalent, unpredictable, selfish, and at
times treacherous.
Nonetheless, when compared with some other societies in lowland South
America, the status of women in the Vaupes seems significantly higher. One
suggestion as to why this might be so involves precisely the fact that women more
than men face disadvantages at marriage, with the result that marriage is seen as
"the problem of getting daughters-in-law" (as opposed to, for instance, the
problem of getting a dowry raised for one's own daughter). Yet women are not
coerced into marriage or carried off against their will as they are among, for
instance, the Yanomamo. The fact that women have the final word in the
process of marriage making, seen as extremely difficult and yet vital, undoubt-
edly contributes to the overall amount of power and status they have and are
perceived of as having. It is also suggested that the very real contributions made
by women in the economic sphere plus the marked segregation of the sexes
contribute to their relatively high status.15 Furthermore, the effects of accultura-
tion, notably the need for the cash crop farina produced by women, and the
cessation of warfare at least fifty years ago have had an impact. Another factor is
192
the relative egalitarianism found in the Vaup6s. Although not as egalitarian as
hunter-gatherer societies - for ranking is present - social differentiation and
stratification are not nearly so developed as in more complex societies.16
Tukanoans display a general dislike of invidious comparison and at present no
stratification exists of the kind that implies greatly unequal control of scarce
resources, power, and positions of authority. The area in which the differentiation
of power and scarce resources is greatest is, in fact, in relations between men and
women. Thus, although it is true that traditional Tukanoan society is marked by
strong statements and evidence of male superiority and supremacy (e.g., only
men are spiritual beings), i t is equally true that when the society is compared
with more complex societies, the actual differences are seen to be more apparent
than real. The relative egalitarianism in the Vaup6s greatly curtails the actual
power men have over women.
Another factor contributing to women's higher status in the Vaupes is the
general emphasis on cooperation and internalization of expectations concerning
work and other obligations. One should be self-motivated rather than dependent
on an authority figure to give orders. A nuclear family is not greatly stratified;
husbands do not tell their wives what to do, and children are rarely told what to
do. Women can use sanctions such as gossip and ostracism as much as men. The
sexual segregation also obviously lessens the opportunity for men to give women
orders. Tukanoan women are not bullied into working hard, nor are they seen as a
kind of domestic servant when they do menial tasks or services or serve men food
and wait for them to finish before they eat. -17
The fact that marriage making is seen as difficult and the continuing emphasis
on the desirability of sister exchange probably contribute to the low incidence of
polygyny. Even if a man is a powerful headman, the stipulation that only women
can be exchanged for women means that the man who wants a second wife will be
considered greedy. Added to this is the pressure from Catholic priests against
plural marriages. All the Tukanoan women I talked with on the subject did not
like the idea of polygyny, and its rarity is at least an expression of their autonomy
and influence, even if other factors actually determine its low incidence. Only
headmen had second wives in my sample of marriages (0.8 percent of 672
marriages). Thus, although the "harem" or "gerontocracy" may have been
typical in the past, and still is in many Australian groups and a number of South
American lowland groups (see Chagnon, 1968, and Siskind, 1973b), it is not
now found in the Vaup6s.
Insofar as I can ascertain, no discrepancy exists between the sexes in terms of
numbers in the respective pools of potential mates. I found no significant
differences in numbers of men and women in my household censuses and
genealogies. Furthermore, although approximate, the age differences recorded
for husband and wife do not seem significantly biased for one sex, although
husbands seem to be on the average three to four years older than their wives.
Still, some men have wives who are considerably older than they.
193
The fish people
The relative stability of marriages in the Vaupes also appears to be related to
the higher status of Vaupes women. Such stability is not seen in many classic
South American tropical societies, such as the Yanomamo (Chagnon, 1968;
Shapiro. 1972), Sharanahua (Siskind, 1973a), Mehinacu (Gregor, 1974), Siriono
(Holmberg, 1969), Krfkati (Lave, 1966), Shavante (Maybury-Lewis, 1967).
Although it is difficult clearly to delineate cause and effect, at present women are
not moved around as pawns in the displays of male brinkmanship typical of
societies like the Yanomamo.
The cessation of "warfare" in the Vaupes, probably more than fifty years ago,
is undoubtedly an extremely important factor in female status. Warfare per se
does not preclude the possibility of relatively high female status, measured in any
of a number of ways; the Iroquois are perhaps the best example of this (Brown,
1970a). Given certain conditions, such as matrilineality, the presence of female
groups, and strong feelings of solidarity among females, a highly developed
military complex with its concomitant emphasis on masculine warrior roles and
values can peacefully coexist with opposed female-connected values and roles.
Such coexistence was economically rational for the Iroquois and actually aided
the military effort, for it was the women who outfitted the warparties with food-
stuffs for long expeditions and maintained homes during the men's absence. Such
a situation does not seem to have occurred in the internal warfare of Amazonian
societies, however, where a frequent component of warfare ideology includes
conceptualizing women as pawns and even as treacherous betrayers in the strug-
gle rather than as contributors to the war effort - although reports of' 'Amazons''
fighting alongside their men are certainly available as well. Undoubtedly the
frequently found expression of affines as "enemies" is an important consider-
ation here.
In general, although ethnologists disagree on the causes of warfare, which can
range from proteins to prestige,' 8 it would probably be generally accepted that
polygyny and unstable marriages are likely to be correlated with the Amazonian
pattern of active raiding and feuding. Thus, at an earlier period in Vaup6s
history, it is probable that both polygyny and unstable marriages were much more
frequent and were directly connected to feuding, because many more marriages
were made by coercion. Following from this, it is likely that the position of
women was lower. This is, however, speculation, based on similar situations in
other areas of lowland South America and evidence from myths and quasi-
historical accounts of earlier periods from Tukanoans.
194
11.
Tukanoans' place in the cosmos
Shamanism
The shaman is an extremely important figure in traditional Tukanoan culture. As
is the case with other specialists in Vaupes society, this position does not relieve
him of the normal duties and obligations of an adult man. The shaman is the one
who "sees" and understands the relations between his community and the world
it inhabits. He interprets and explains this relationship to others, and thereby
warns and prescribes. He diagnoses and cures and above all protects by knowing
how to maintain harmony and balance between his community and the rest of the
universe. Part of this task of protecting the individuals under his care may involve
bringing illness or misfortune to others, but good shamans never practice sorcery
against people of their own community. Protection involves performing necessary
rituals, prescribing correct food restrictions, making divinations, and "seeing"
by going into trances. His activities help to ensure fertility of food sources and
human fertility. He officiates at all transformations. He ushers newborn babies
into the world of humans with water and food rituals and establishes their
identities as members of particular language groups by giving them names, and
through this, the "soul-stuff" which is passed along patrilineal lines (see C.
Hugh-Jones, n.d., p. 9). He officiates at initiations for young men when they are
first introduced to the Yuruparf, other ritual substances, and sacred knowledge,
and at the menarche rituals for young girls. He officiates at rituals intended to
transform the community from a disturbed state to one of equilibrium, as, for
example, when a member of the community is bitten by a snake, falls ill from
sorcery, or dies and needs to be buried properly and sent on his or her journey to
the other world. Very important to the well-being of the community are all large
ceremonies in which the shaman transforms the longhouse and its occupants in
space and time. The rituals to accomplish this, in which all adult men participate,
are organized and conducted by the shaman. At the most sacred point in this
process, the group enters ancestral time: The longhouse is the universe, and the
people in it, along with the sacred horns, are ancestral people (S. Hugh-Jones,
1979).
195
The fish people
Shamans are able to officiate and transform not because they are priests in any
sense but because they know and understand myth. Myth forms the foundation of
Tukanoan symbolic structure, and through this, Tukanoan society; it describes
and explains the world and provides the rules for the proper maintenance of
human society within this world. Shamans must understand myth well enough to
be able to translate it for the rest of the community in terms of its applicability to
current problems. (To some degree, however, everyone understands certain myth-
ical symbols and meanings.) The esoteric language spoken by shamans (S.
Hugh-Jones, 1979, pp. 58-59) is based on analogies and synonyms that only a
thorough understanding of Tukanoan mythology can render comprehensible.
Thus, because shamans not only know and see this world but understand the
other worlds of the universe and how these figure in the activities of this one, they
understand the importance of balance and moderation in both human and non-
human life. They know how to regulate, transform, and recycle energy through
the cosmos - an extremely important concept in Tukanoan cosmology.
As ushers-in and transformers, shamans are intermediaries and for this reason
are frequently seen as ambiguous. In a number of ways they straddle both sides of
important divisions. They are human, yet they have contacts with the nonhuman
world. Some shamans become evil and use their power for vengeful or other
selfish motives. When this happens, these individuals are no longer considered
men and come to be identified with the animal world in several ways - with
animals, with the forest, and with the spirits inhabiting the forest. "Normal"
shamans are humans but can perhaps be described as more than human. They
speak a number of Tukanoan languages, as does everyone, but they also have
their own. They do everything any ordinary man will do, as well as perform
many special activities. They see everything ordinary adults see, and more.
To become a shaman takes at least a year of apprenticeship and involves
learning to learn from one's dreams, one's trances. It was emphasized that during
this time, a young man only "gets advice" from his teacher. To become an
effective curer one must travel to all the houses in the other worlds to learn the
proper chants. As expected, the apprentice observes many restrictions in food and
activities. Whereas apprentice shamans must travel to all the houses in the rest of
the universe, a very powerful shaman can persuade the shamans in these houses
(who had been shamans when they were alive in this world) to return here. I
collected a long story of a very powerful Bara shaman who was able to do this;
the two other-world shamans who visited him charmed a section of territory so
that people, so long as they stayed in it, did not die.
People are always somewhat wary of a shaman, even those not suspected of
having become evil. A certain amount of ambivalence and potential suspicion
seems always to adhere to any position involving leadership and specialized
knowledge in Amazonian (and other small-scale) societies. Furthermore, in the
Vaupes such a role is dangerous because ritual power and knowledge are always
dangerous. Thus a renegade shaman can forsake all feelings of responsibility to
196
human society and become destructive; when this happens a shaman actually
turns into a predatory spirit animal. In most, if not all, Tukanoan languages the
word for shaman is synonymous with the word for a class of predatory animals
including the jaguar (yai). It is probably also the case that shamans are somewhat
mistrusted, at least in the abstract, because of their intermediary position per se,
in the same manner that extremely old people, so evidently straddling the border
between the living and the dead, are seen as not quite human and are mistrusted.'
Shamans are also the intermediaries between balance and imbalance, for they
can divine causes and restore equilibrium through ritual. It is they who protect
the local group from the fury of an angered spirit-person. It is they who bring
"unbalanced" people through their personal crises - which also always affect the
rest of the community - brought on by birth, first menstruation, illness, or a
similar condition. It is they who regulate and balance the nutritional state of the
longhouse. In many senses shamans are food intermediaries (Langdon, 1975).
They symbolically stand between food and "soul food" (C. Hugh-Jones, 1979) -
the sacred substances of coca, tobacco, banisteriopsis, and chicha.
According to S. Hugh-Jones (1977, p. 213), shamans are intermediaries in
several senses in the male initiation ceremonies: For example, they are members
of both the class of elder men and of the initiates. They also symbolically stand
between the world of women and that of men.
Because no shamans lived on the Inambu River, my information about them is
incomplete. Very few shamans are left in the Vaupes; of those remaining, most
are found in the Pira-parana region or in a few far upstream locations. During
my stay, no real shaman ever came to Pumanaka buro, although I did meet one
Bara shaman during a trip to the Tiquie. At Pumanaka buro the headman
officiated at whatever ceremonies continued to be performed, and he performed
some curing rituals.
What follows are very abbreviated accounts of the kinds of crises in human
lives that shamans oversee and restore to normalcy via ritual. These accounts are
meant to illustrate certain basic points about shaman roles and are not intended to
be comprehensive descriptions.
Birth.
After a child is born (the birth takes place in the manioc fields), the shaman
performs a number of rituals to give the child human status and make it a member
of its particular language group. In addition to defining its status, these birth
rituals serve to protect it (for example, by making safe the foods it eats) and to
protect the rest of the community during this period of heightened risk. Many
precautions and restrictions during the course of pregnancy, birth, and nursing
have this effect. These restrictions apply to certain foods, activities and rituals,
and contact with particular objects. The entire community is involved in a birth
and participates in the accompanying rituals, but the stages at which the parents
197
The fish people
and offspring pass from one stage to another - that is, are transformed in some
manner - are conducted by a shaman.
Neither babies nor mothers are washed for two days following a birth. They are
then painted and undergo a ceremony at the port, during which a gourd of
mother's milk is made safe (henceforth permitting the child to suckle). Mother
and child are then bathed, and the child is given a name. The shaman also cleans
the house out with pepper smoke.
Sorcery and disease.

Diseases never "just happen," and usually malice lies behind an individual's
illness or accident. Disease can be caused by an unknown person directing
sorcery at a particular individual, by a wahti, or by a shaman who is exchanging
human souls from a distant settlement for game animals. It is also true that people
can be in a state in which they are particularly likely to succumb to illness. If they
have seen a wahti, for example, and brood about it, or if excessive sadness or
grief is not alleviated, they will fall ill. The prayers after a funeral are in part to
ensure that the bereaved relatives do not think too much about the dead, lest they
too fall sick and die. Ordinary human beings can cause illness in any number of
ways, from actually slipping a toxic substance into someone's food to learning a
particular spell and practicing sorcery. In fact, it is interesting to note that killing
by sorcery can be described with the verb doarihe, which basically means "to
cook." A sorcerer might send a cigar made of' 'other-world tobacco" through the
air to bury itself in a distant longhouse's floor. As it travels through the air it
resembles a species of fly, but upon divination and discovery it looks like a cigar
(which, however, only a shaman can see). The cure is to return it to the sender's
longhouse, where it will cause much sickness. A counter-sorcery exists to punish
sorcerers. It is a complicated process based on contagious magic using "dirt"
from the sick person's body - hair, nails, urine - which is kept in a bottle and
prayed over for many days and finally placed near a fire. When the bottle bursts,
the sorcerer dies and the sick person usually gets well.
A spell can also be cast on an object, which will then cause harm to anyone
who comes in contact with it. I was told that only the Siriano and Desana can cast
these spells onto objects, such as machetes, axes, and canoes. Even trees can be
sorcerized so that they fall and close off the trail. Contact with these objects can
cause cramps or paralysis.
Most often everyone knows what kind of disease a person has from the
symptoms. Still, a shaman must diagnose by divination the ultimate cause of a
disease (i.e., the agent willfully causing it), and only shamans can cure a serious
illness. Shamans go into a trance to divine the evildoer's identity and to locate the
sickness-bearing object. The most frequent disease-causing objects are wahkd -
intrusive thorns, hair, pieces of cotton, and similar items that enter the patient's
body and must be extracted. Some of these never lose their power; one reason a
198
longhouse is abandoned after a shaman dies is that the particular type of object
strong enough to kill a shaman (looking like a tiny canoe paddle) will kill anyone
else who remains in the longhouse.
Sorcery can also be practiced on unmarried girls so that they will be barren or
lose their children during pregnancy or after birth. Informants gave unrequited
love as the motive. Birth defects can be caused by sorcery. A pregnant woman
who eats dirt has probably been sorcerized.
A shaman's equipment includes Virola sp. snuff, which was obtained from
Maku in the past but which, I was told, is no longer available. The few
still-practicing shamans left are thus very reluctant to use their supply unless
absolutely necessary. A shaman also owns a cylindrical quartz crystal (yai-ga);
the quartz is found only in the Tiquie region. One of its names means "thunder
stone," and the shaman uses it to "see" - to see into the body of the patient and
discover who fee culprit is. These quartzes also can be used to send illness in the
form of cramps and spasms.
Curing mainly consists of extracting the foreign substance from the patient's
body. Shamans cure with techniques involving tobacco smoke, massage, and
water throwing. Water throwing requires the most advanced training. The most
important part of curing, however, are the prayers. Any adult knows remedies for
minor illnesses such as muscle ache, fungus infections, swellings, and eye
trouble. And many forms of health maintenance and disease preventives can be
used by everyone, such as purgatives, emetics, and snakebite preventives. Many
plants used to ward off malign influences are grown. Older men and occasionally
old women as well will blow tobacco smoke onto children for this purpose. Only
shamans can cure serious illnesses, however, because only they know the proper
prayers.
Blowing tobacco smoke, praying, and the water cures occur on the patio in
front of the longhouse. The shaman massages the patient, which causes the
disease-bearing wahka to go to the extremities of the patient's body and enter the
shaman's. The wahka then migrate to his mouth and he pulls them out. I was
told that while these items look like thorns and so forth, they really are not these
things.
The difference between yai and kumu, the two words roughly glossing as
"shaman," was never made clear to me. I was told that both kumu and yai cure
by blowing smoke and throwing water, but only the yai performs sorcery.
Tukanoans state that many white-introduced diseases do not respond to tradi-
tional curing techniques but that traditional cures are needed for traditional
illnesses. Thus, it is a matter of concern that no more shamans are being trained
in the Papuri region.
199
The fish people
Death.
When someone dies, the corpse is washed and buried inside the longhouse.
Prayers are offered to ensure that the spirit leave quickly; chants also help the
bereaved. A woman is buried with her small pouch of personal belongings, but
her other personal property (e.g., baskets, pots) is broken or burned. A man is
buried with whatever ritual paraphernalia is closely identified with him. Destroy-
ing or burying such property gives the spirit less to miss and thus hastens its
speedy departure. Fires are sometimes made over a fresh grave to lessen the
possibility of seeing the spirit leaving.
I was told that "before we had canoes and only fished from the riverbank"
people used to be buried in a crouched position in urns; this is the way children
and babies are buried today. For an adult, an old canoe (but one without a split),
or a new one if an old one is not available, is lined with the person's hammock.
The corpse is washed and groomed but not painted. The canoe is sealed with a
piece of wood and resin and bound with fiber ropes. After burial, werea is burned
and relatives come to chant and mourn. It is not a proper festival, however,
because it lacks chicha and dancing. Three days later normal routines are
resumed.
After death, spirits normally go to "dead-people spirit" (mahokd diari
wahti) houses, where the ancestors live. These houses are found in specific
locations, depending on an individual's language group. For example, Tuyukas
go to a longhouse on Behuya stream in the upper Paca River, and Bard go to
Pamuri wi at Yuruparf Rapids in the Vaupes. The spirits of people who have had
certain kinds of difficulties in life either linger around the area where they died or
begin the journey but never succeed in actually entering the dead-spirit long-
house. In a sense, the final stage of dying is entry into this house (see the account
in Chapter 4 of Old Arturo's saving the headman Jose from death by preventing
his entrance into the dead-spirit longhouse). The spirits of the dead are not
intrinsically bad, but they should be kept separate from the world of the living.
When this does not happen, something is wrong and the living begin to be in
danger. One example of this is when a man prays and buries a special cigar in the
house of his enemy. Later, when the man dies, his dead spirit also goes to his
enemy's house to bother him and his housemates. This happens even if the man
has died in a way totally unconnected with his enemy. His dead spirit is heard
pounding on coca and swimming in the environs of his enemy's house. To
exorcise this spirit a shaman has to divine the location of the cigar and destroy it,
releasing the spirit.
The spirits of people who steal from manioc fields stay around their graves and
have to be banished by prayers of a shaman. This is also true of people who
commit incest. Adulterers go to the dead-spirit house but continue to be trouble-
makers there. Murderers, too, tend to remain at their graves rather than setting
off on their journey. The victims of bloody murders (e.g., killed by guns or by
200
knives) are doomed to stand forever outside their dead-spirit house, with blood
continually flowing from their wounds. Those who meet accidental, yet bloody,
deaths continue to "breathe poorly just as they did right before they died," 2
indicating that they have not quite died. Thus, a connection is made between an
improper death and an incomplete process of dying and separation from the
world of the living.
Although the dead-spirit houses are sometimes spoken of as far away, the dead
spirit, emerging from the grave, sees them as quite close. I was told that some
dying people describe the dead-spirit houses they are already seeing.
Reports about the dead-spirit houses are conflicting. One day people will say
no one hunts or fishes there, and another day people will say that people "have
everything there that we have in this world." Sometimes dead spirits are said to
eat regular food, and yet other times they are described as eating only tobacco,
chicha, coca, and banisteriopsis. I was once told that old people turn into young
adults again. As noted earlier, Tukanoans are vague regarding whether women
stay with their husband's group or return to their own patrilineal dead-spirit
house.3
Usually dead spirits are not powerful enough actually to take the living with
them (however, see the myth of Live Woman, Chapter 6, about a wife who
follows her husband to the dead-spirit house).4 The spirit of a powerful shaman
may succeed in this, though, by accompanying his widow everywhere. By thus
making her cry all the time, she eventually falls sick (although he does not
actually send the sickness), and "he can take her with him." No woman could
take her living husband with her in this manner, however (see the myth of the
co-wives, Chapter 6).
If a bayd (a head dancer) were to die without teaching the chants to the next
generation, his relatives might decide to wait seven to ten years and send two men
to revisit the abandoned longhouse site during a lunar eclipse. They would make a
fence with bamboo slats near the grave and move away, saying, "Grandfather,
we've come to learn a chant." Eventually the earth would move, and the dead
spirit would emerge and teach them everything.
In summary, the ideas associated with death are that dead spirits are potentially
harmful only if in some way they do not completely ' 'die'' - a process that lasts
until they have left the area and entered the dead-spirit longhouses of their
ancestors. This happens when something in the world of the living is not right.
Whether the dead were brutally murdered or themselves were murderers or
whether they missed their spouses too much, something made the separation and
mourning period an abnormal process. The effect of this is to throw other things
out of balance so that eventually the whole community suffers.
201
The fish people
Festivals
In general, festivals are basically concerned either with relationships within a
language group (or sections of it) or with relationships between affines. Within
these two categories are several more specific types of festivals, some of which
are no longer held in the Papuri region. Regardless of specific type, all festivals
have the following characteristics. A festival is a special, nonordinary, sacred
event. This is reflected in (and in many symbolic ways signaled by) the alterations
that occur in time, space, activity, state of consciousness, and, at certain points,
reality - at the most sacred point of the proceedings, the participants actually
become one with the ancestors. Thus, the festival is a time to transcend everyday
reality, to reveal and celebrate its underlying truth; in short, to ' 'make sense'' of
it. All normal activities stop: One neither eats nor sleeps during a festival. Those
items that are consumed are explicitly opposed to ordinary food; C. Hugh-Jones
has referred to them as "soul food" for this reason. These soul foods - chicha,
tobacco, banisteriopsis, and coca - are intoxicants and are seen as nourishment
for the soul rather than for the body. They are also sacred (although coca, chicha,
and tobacco can be ingested on nonsacred occasions). How long the participants
go without sleep varies, depending on the type of festival and how carefully
traditions are followed.
The entire festival, including the preparations prior to its official beginning,
resembles a piece of music in that every aspect of the event - the chanting, the
instruments and their music, the dancing, the intoxication - all lead to a climax
late at night. By this point the people, the paraphernalia, the music, and the
longhouse itself have all undergone a profound transformation. This is the
opportunity for participants (although we are speaking here mainly of initiated
men) partially to experience what shamans experience - to "see" beyond
everyday reality and observe the underlying meaning, in effect, the underlying
reality, of the world and life itself.
As already noted, festivals that are basically making statements about relations
between affines, dabucuri, involve an exchange of smoked meat or fish, brought
by guests, for beer, supplied by the hosts. Festivals that focus on the patrilineal
concerns of the sib and language group are the ceremonies at which the sacred
horns are played over recently harvested jungle fruit and the series of ceremonies
involved with male initiation. The food exchange rituals are concerned with
reciprocity. The fruit and initiation rituals are concerned with growth, fertility,
and continuation. In a deeper sense these two types are linked, the one ensuring
the continued production of people and food and the other ensuring the proper
exchange of these between groups of people.
Currently in the Papuri drainage, festivals are also held simply for the purpose
of drinking and dancing. They do not involve a ceremonial exchange of protein
or fruit foodstuffs nor the playing of sacred horns. Whether this practice was also
true at earlier, less acculturated periods is uncertain; however, both Langdon,
202
1975, discussing the Pira-parana region, and Goldman, 1963, discussing the
Cubeo, state that such "simple" festivals take place in these groups.
For festivals the Tukanoans energetically clean up the longhouse and surround-
ing area (which may include making substantial repairs to the roof, the port,
etc.). amass huge amounts offirewood;prepare large quantities of chicha, usually
of more than one kind; grind sugar cane; process a great deal of coca; finish work
on new dresses and other items of clothing and decoration; and prepare enough
cazabe to serve early arriving guests. Technically no cazabe is eaten after the
festivities formally begin, but some women surreptitiously eat throughout the
festival, and children are not expected to go without food. Before a dabucuri,
those who will be guests feel a great deal of pressure to kill and smoke sufficient
amounts of game or fish, because they would be embarrassed if not enough were
provided. This obligation itself nicely demonstrates the pressures and responsi-
bilities affines feel toward each other.
Early arriving visitors help prepare the wooden percussive staves (made anew
each time) that are used in the dabucuri festival, assist in making the banisteriopsis
infusion, and help to process coca. At a certain time, the basket suitcase holding
the more sacred ritual paraphernalia is lowered from its shelf and the final
dressing and painting begin.5
The main activity of festivals is dancing and drinking. One word for these
occasions is hinirike ("drink"). Dances last anywhere from fifteen to thirty
minutes and are followed by breaks for consuming chicha and, for the men,
banisteriopsis, tobacco (in the form of cigars and snuff), and coca. Men chant
while seated at stools or along benches; some of the older men and the shamans
do not dance at all. The more formal dances, consisting of rows of men who are
joined by women after the dance is begun (each woman placing herself under the
outstretched arm of a man), are led by the baya\ Several dances are performed,
each named, and each varying in step and the paraphernalia (rattles or staves,
etc.) used. It is the younger men who initiate panpipe-playing dances, starting off
in a line, playing and dancing to intricate rhythms and harmonies. Much more of
an element of challenge and competition is apparent in these dances. Women join
the men, either beside them in the row, or, for panpipe dancing, in side-by-side
couples, after a dance has begun and leave before it is finished. During some
dances, a woman (ytigd) - usually one married to a headman or a baya - stands
in the middle of a circle of dancers and periodically emits a long, piercing shriek.
At the beginning of the festival the host group and the visitors are clearly
demarcated according to which group dances and which watches. Later on,
however, the groups intermingle, expressing symbolically how settlements, par-
ticularly affinally related ones, interact.
The official beginning of a festival is always about 8:00 a.m., although panpipe
playing and coca chewing may have kept the men up quite late the previous night.
In any case, almost everyone in the longhouse is involved in making chicha the
night before a festival, and no one sleeps much. The peak of the ceremony is after
203
The fish people
midnight the following night. Festivals vary in duration, but a day and a half
of dancing is no mean feat, especially considering the time spent in preparations
and the added effects of the intoxicants. Always toward the end, tempers are
frayed and hangovers monumental. Not every festival I attended ended in quar-
reling (as was true in Goldman's experiences with the Cubeo), but I did see
a number of disagreements erupt and one serious argument. And, of course,
both the hosts and the guests griped and gossiped after parting company.
Regardless of the specific type of festival being held, the host group usually
invites its closest neighbors, irrespective of their language group affiliation. S.
Hugh-Jones (1979) reports this also to be true for festivals held to celebrate male
initiation. Sometimes young men from more than one language group are
initiated together, although according to tradition this is not supposed to happen.
The Tukanoan world

Tukanoans know that the world is much, much more than what it appears to be to
the five senses. The world was different in ancestral times,, which are recreated
during sacred ceremonies. The world as perceived space is also more; for
example, the sky is not only a sky but the underside of the level of the universe
above this one. 6 Furthermore, many of the distinguishable natural features on this
level are more than what they appear to be to the naked eye, although usually only
shamans can see beyond this visible reality (other men sometimes travel and see
the less immediately apparent reality of this level when taking banisteriopsis
during ceremonies). At the most sacred point in Tukanoan rituals, vertical,
horizontal, and temporal "space" is transformed. The longhouse itself becomes
the universe, the participants become the ancestors, and the realities behind the
outward appearances of space, time, and matter are revealed. These characteris-
tics of insubstantiality, mutability, and multiple reality are crucial ones in
Tukanoan conceptions of the universe. Any descriptions, therefore, that depict
the world only in terms of substantive, concrete, and permanent characteristics
are misleading.
Levels of the universe.

The universe consists of five levels, but not much is known about the very top and
bottom layers. Whereas the levels adjacent to ours are populated with reasonably
nice beings and have features similar to ours, the very top and bottom ones are
not very interesting, for they have only ugly, dirty, black, and bad inhabitants.
The next highest level is associated with air, this level with earth, and the level
below with water. Beneath the layer of this world is Opek6 dia ("milk river"),
a very important feature of the Tukanoan universe. Its current flows from east to
west, unlike all the rivers of our level, which flow west to east. This is one of
many examples of inversions found in the levels adjacent to ours. For example,
204
while it is nighttime here, it is daylight in the level below us, because the sun is
crossing that sky after it sets on our level. The same inversion is found in the sky
level above us. A myth about "star girl" demonstrates many of these. Star girl
was bitten by a vampire bat. The wound did not heal and soon became filled with
worms. She smelled so bad that her relatives threw her out, and she fell to this
layer. A man found her and washed and cleaned her. When her wound healed, he
took her as his wife and they returned to the sky. However, everything was
backwards for him: When his wife slept he was wide awake. When the couple
engaged in sexual intercourse everyone could see what they did, because it was
broad daylight. While he was fishing, she would sleep with other men. Finally,
disgusted, he returned to this world.
The heavens.
The Bard at Pumanaka buro were not particularly knowledgeable about celes-
tial bodies and phenomena. To what extent this lack is due to more than fifty
years of missionary influence and to what extent it is due to their considering
themselves "water people" is not known.7 For example, although various con-
stellations had names and their movements were tied to the seasons of the year, no
one knew a name for the Milky Way, which figures prominently in Desana
mythology (Reichel-Dolmatoff, 1971). Bara identify Desana, Tatuyo, and
"maybe Siriano" with the sky and say that these groups are "the sun's children"
and that they know much more about celestial phenomena. Insofar as I could
ascertain, the people at Pumanaka buro placed the most importance on the
planet Venus, which has two names as morning and evening star. When Venus is
near the moon, it is called his "bride" and this portends a marriage.
Rainbows have separate names, depending on whether they are in the sky of
the rising sun or setting sun. An association exists between rainbows, thunder and
lightning, snakebites, and rotting fingernails and teeth. A rainbow around the
sun is a sign that shamans either are creating sickness in some settlement or
teaching apprentices how to cast spells.
Thunder and lightning are closely associated with sickness; they can cause an
illness or aggravate an already existing case. The thunderstorm that occurred
after the snakebite incident at Pumanaka buro (see Chapter 2) was seen as
connected to the event: The anger of the other fer-de-lances in the region and of
the spirit snakes brought the thunder, which in turn increased the possibility of
additional snakebites and the likelihood of the little boy's dying. Pain connected
to thunder is described as similar to a bolt of lightning.
Sometimes one can hear the "sky people" running around above the clouds
just before a thunderstorm. They are afraid, and so they yell and bang on pots and
pans, which is also the proper behavior during a moon eclipse. These people are
three feet tall and have no hair.
The headman, Mario, said that in earlier times a screen of tapir hide would be
205
The fish people
attached to the pillars just inside the men's door during a festival. This would
protect the two chanting shamans sitting behind it from sicknesses sent by enemy
shamans through thunder, lightning, and winds.
In stark contrast to a general lack of interest in heavenly bodies and events is
everyone's fascination with lunar eclipses.8 In general the moon is seen as a very
sinister personage. The shadows on his face were put there by his sister when she
smeared a black dye, weh (Rublaceae), on him during the night to be able to
recognize her mysterious lover, the moon, in the daytime. During the periods
when the moon "sickens" or "dies," which include eclipses and evenings when
the moon is a dark red (this is corroborated by Goldman, 1963, p. 245), the moon
comes down to this level, hides the light he is carrying under a pot, digs into the
earth in abandoned longhouse sites, and eats the corpses. People must make
every effort to force the moon into the heavens again, for eclipses are prototypes
of periods when everything is out of balance and greatly disrupted. Wahtia of
many kinds are about, returning to their graves or dancing in the forests.
Everyone must hurry to accomplish tasks and in general create a lot of commo-
tion and activity, neither resting nor sleeping. During this period, as in other
times of disturbance, chili peppers are burned, the smoke protecting the long-
house and its inhabitants from the dangers that threaten.
Important places and features of the landscape.

Many geographical formations and specific locations have mythical references
and are seen as special, if not sacred. Types of terrain likely to be imbued with
sacred meaning include rapids, sandstone outcrops, caves, dome-topped hills,
clearings in the forest, miriti swamps and lakes, and lagoons in rivers. The
origins of most large and unusual rock formations are given in myths. Particular
sandstone hills are known as the houses of specific wahtia. Many features are
named and described in the section of the origin myth cycle that deals with the
upstream journey of the ancestors of the language groups in the anaconda canoe.
People should avoid tatd, or clearings in the jungle, all of which are shaman
houses. Everyone knows anecdotes about those tatd near their own settlement;
these anecdotes usually concern people who remained near a tatd boa (boa
refers to the "rotten" quality of the clearing - i.e., that trees do not grow there)
and were swallowed up into the ground.
Ewiira tar6 is an extremely important miriti swamp because it is said to be
the source of the rivers in the region (the Inambti, the Tiquie, and Timi'ya on the
Pird-parana'). Shamans can see a house inside the earth near this site, which is
the ancestral home of peccaries. In the center of the swamp is a huge miriti palm
(neno), which grows to the sky each day. It is the "center of the world";
shamans see it as a huge snake.
Many other localities are known to Tukanoans as the place where such-
206
and-such happened in a particular myth; furthermore, these sites are associ-
ated with various important facts of present-day life, such as the ancestral home
of a species of game animal or the original source of medicinal plants, or
first-menstruation longhouse, and so forth. Although these places are distinctive,
it is only shamans who can "see" the other reality they have.
At a deeper level, however, every part of the world is sacred. The hidden
realities of all places can be comprehensively described only by shamans or
people who have "seen" them during their travels under the influence of
banisteriopsis. Rivers, the forests, cultivated fields, the longhouse - all have great
symbolic meaning, all have a reality behind their ordinary appearance, and all are
described and explained in Tukanoan mythology.
Wahtia.
In general, wahtia are humanlike creatures but are not ordinarily seen by humans
and are not ordinarily found in areas frequented by humans. Although not all
wahtia can be seen as representatives of aggressive and asocial sexuality
(Reichel-Dolmatoff, 1971) many wahtia do resemble projections of human feel-
ings and drives. Inasmuch as they represent human yet unacceptable emotions,
wahtia are caricatures of human nature, and this is reflected in their appearance:
humanlike yet wrong in some fashion. Their appearance can be a rather mild
departure from that of normal human beings, as when wahtia are described as
looking like human beings but lacking nipples or toes or having only one breast,
or they can be truly monstrous. Not all wahtia are monstrous, because the
ancestors are a kind of wahtia, as are the dead-people spirits (mahoka diari
wahtia). Still, all wahtia do seem only to resemble human beings and, in so
doing, expressly highlight an essential nonhumanness.
With the exception of ancestors and dead-people wahtia (which are, at a deep
level, identical to and in fact are merged with the living during part of the festival
proceedings), all wahtia eat people.9 A particular kind of wahtia {weori maha)
specializes in looking exactly like someone's kinsman, in order to trick the
person into accompanying it into the forest, where he is attacked. Another kind
(hiigard) is actually two beings, attached at their buttocks back-to-back such
that when one wants to go one way, the other has to follow walking backward.
Another kind (huobari wahti) calls to victims while facing away from them in
order to trick them into thinking the danger is far away. This wahti's arms are so
long that it doubles them up and carries them on its shoulders. Many more kinds
exist, all having qualities of exaggerations or reversals of human features: Some
are huge and black but otherwise humanlike, whereas others are little and "like
whites."10 Any Tukanoan knows many stories about wahtia. Some have actually
heard them yell or have heard of nuts or pits being thrown at someone by a wahti.
Face-to-face encounters are rare; I never talked to anyone who had personally had
207
The fish people
such a meeting, but many Tukanoans talk about others who have and who were
scared out of their wits. Encounters with some kinds of wahtia automatically
result in death.
Ordinarily, humans and spirits do not encounter each other; contact is a sign
that something has disturbed the normal state of affairs. Dead-people spirits
haunt an area only during a lunar eclipse or if something was wrong about their
life or the way they died. Both wahtia and spirits in general (e.g., wai mahl,
who live below this level — "fire women," "snake women," "star people"),
including mythical culture heroes, are normally retired and removed from this
world. Only when the world itself is radically altered (i.e., during festivals) is
there a merging of human and other spirits, but at these times human beings are
not ordinary humans either. Otherwise, only shamans who go into a trance and
travel and "see" will have contact with dead people and other-level houses.
Conclusions
Despite the brevity of this presentation of Tukanoan cosmology, it is hoped that
the reader has gained some knowledge of the richness of Tukanoan ritual, myth,
and symbolism. That such richness is still apparent despite more than a half-
century of missionary influence in the Papuri region is remarkable. The effects of
missionary activities have been extensive everywhere in the Vaupes, greatly
disrupting the coherence and integrity of its culture.
Several general themes can be seen in the preceding discussion. One is that the
dichotomy between sacred and profane is relative rather than absolute: Whether
something is considered sacred or not depends on which of several levels is being
conceptualized. At a very profound level everything is sacred, because behind
what is perceived by humans as everyday reality, including the most prosaic of
events or situations, lies an extraordinary reality or realities. Certain items or
concepts are more sacred than others (e.g., men-women; initiated-uninitiated).
Some things are always sacred, for instance, beeswax or Yurupari horn music.
Obviously, shamans, and to a lesser degree initiated men who have knowledge
and who drink infusions of banisteriopsis on ritual occasions, can perceive more
of the behind-the-visible reality than women and children. Still, all Tukanoans
experience these extraordinary realities to some extent during their lives, both as
spectators at ceremonies and during their own periods of life crises.
Another general theme in Tukanoan cosmology is the conceptualization of the
world as in a state of dynamic equilibrium. In a sense, myths provide a code that
explains the system and gives instructions for maintaining or reestablishing this
equilibrium. Rituals, with shamans acting as guides and intermediaries, allow
people to pass safely through periods of danger.
The symbolism pervading all levels of Tukanoan life also explains and makes
sense of the universe and Tukanoans' place in it. For example, although
Tukanoans are concerned about proper and moral conduct, issues are frequently
208
too complex to be reduced to questions of right versus wrong behavior. Although
at times the status quo is thrown out of kilter by a consciously immoral act (e.g.,
incest or murder), at other times a community enters a period of danger that is
simply due to the process of living. One must eat and therefore run risks by
hunting. One must acquire a spouse, invariably a risky undertaking. One must go
through puberty rites and be concerned with the fertility of crops and humans,
and yet these necessarily affect the overall balance of energy in the universe. This
concern - maintaining a balance - in part accounts for the complex associations
linking hunting, sexuality, spirit owners of species of game animals, and the
strict observance of food and other restrictions. To maintain this equilibrium
properly while at the same time exploiting the energy reserves in the universe is a
tricky business. To run the risks required for living - to face the dangers of falling
sick, of dying, of being attacked (openly or by sorcery) by one's enemies - is,
thus, a necessary part of the exchange Tukanoans carry out between themselves
in their daily and sacred activities and the rest of the world, a world seen as a
closed circuit of energy. With each generation of people, a greater amount of
game animals and fish are consumed as food. This inevitably means that there is
an ever-increasing depletion of the finite and irreplaceable energy reserves in the
universe.
The idea that energy is limited and nonrenewable is a conscious aspect of
Tukanoan cosmological structure: Living organisms clearly deplete the resources
of their environment (an idea rather new to Western society). Yet these resources
can be replenished and new energy created through rituals that foster contact with
the nourishment and growth-power available from the first people, the ancestors.
S. Hugh-Jones (1979) shows how this is accomplished by rituals that produce a
collapse of time, in particular the descent model of time: Lineage time, or sib
time - history - is dissolved and, again through ritual, rebirth and renewal are
achieved. In essence, these rituals contradict the idea that the energy resources of
the world are nonrenewable, for by collapsing genealogical time the power and
energy in the universe are recycled. The theme of recycling energy is a powerful
one, expressed in many symbolic forms. Rituals to obtain power and nourishment
from the ancestors, according to S. Hugh-Jones, use metaphors of rebirth
(including skin-changing, perhaps the most powerful symbol of rebirth and
immortality), because these metaphors, being concerned with cyclical rather than
unidirectional time and space, provide a continual opportunity to tap into the
energy available at the very beginning of the world. Tukanoan concepts of
reincarnation fit into this structure: Although Tukanoans at death are spoken of as
being permanently removed to their ancestral homes located in specifiable areas
of the Vaupe"s region, dead-people spirits are also reincarnated as energy - as life
- and remain on this level. It is significant that Pamuri wi, the name of the Bara
ancestral house, refers to birth. One translation of this name is "the house where
our ancestors were born." It also refers to the birth of present-day Bard, or more
specifically, rebirth: metamorphosis and reincarnation. In time, dead-people
209
The fish people
spirits are forgotten and become further and further removed from the affairs of
humans and from any specifically human characteristics. But they remain in this
world in their transformed state, eventually turning into little birds. Certain birds,
such as the oriole, Guacamaya, and egret are closely linked to Tukanoan concep-
tualizations of life after death. At some final point in time, human spirits become
almost totally inanimate, turning into mist and foam on the rivers.
210
12.
Tukanoans and the outside world
No phrase captures so well what 1\ikanoans face in their present and future
contacts with the outside world as the subtitle of Erving Goffman's book Stigma:
The Management of Spoiled Identity (1963). Although the degree and nature of
change vary greatly from region to region in the Vaupes, it is apparent to all but
the most ethnocentric change agents (regardless of the particular type of change
being promulgated) that Tukanoans have suffered greatly from disturbances in
their traditional culture. Although the situation in the Vaupes may be better than
in most of Amazonia, as Corry (1976) maintains, in my opinion Tukanoans face a
bleak future.
In the vast majority of cases the results of regular contact between a small-
scale society and a powerful colonizing one are massive, far-reaching, and for the
most part negative. This has been well documented for hundreds of band and
tribal societies (Bodley, 1975); Tukanoan society is no exception.
Some of the changes in the society resulting from contact with outsiders have
been mentioned in previous chapters. Diseases introduced by outsiders brought
death and out-migration, reducing the size of local groups so drastically that
some customs and institutions have disappeared owing to lack of people to fill the
required roles.
The effects of an expanding, development-oriented national and international
economy on local systems like the one in the Vaupes are, with few exceptions,
devastating (Davis, 1977). All of a sudden native economies that had no universal
currency, all exchange being embedded in the network of kinship relations, have
to adjust to a cash-based market system. Systems like that of the Vaupes have
very little centralization and are predominantly subsistence-oriented. Having to
adjust, and adjust very rapidly, to radically different systems inevitably creates a
great deal of stress - in the native economic systems and in the natives them-
selves. New materials and objects are introduced, as are ideas about the monetary
value of labor. Many assumptions are introduced regarding what motivates
people economically. Also new are the sanctions surrounding economic transac-
tions with members of the dominant culture, such as being kept at a rubber camp
until a debt is paid. Most important, of course, is the fact that Tukanoans become
211
The fish people
increasingly powerless and are exploited in the vast majority of their transactions
with outsiders.
In the dominant ideology, cost effectiveness and, usually, profit underlie
most strictly economic transactions. Most change agents, whether in the Vaupes
for personal economic gain or not, see an internal colony in need of development
- "development" meaning improvement in the region's ability to supply raw
materials and native manufactures to the national economy. I am not concerned
here with denying that the Vaupes's economy can improve with development or
that Tukanoan attitudes toward money must change. I am concerned with the
problems that these processes create and with understanding the ' 'development
of underdevelopment" that too often occurs in marginal areas like the Vaup6s.
It is evident from the foregoing chapters that traditional Tukanoan social
organization has broken down in many areas. For example, Tukanoans have little
acquaintance with true social stratification in their traditional political organization
(i.e., in terms of limited or exclusive access to and control over scarce resourc-
es). Ranking clearly is a crucial concept in the Tukanoan world view, but is
rudimentary with respect to linkage with exclusive economic privileges and thus
bears little resemblance to the hierarchy introduced by representatives of the
national culture. Traditional headmen (except for war chiefs) have largely
ceremonial tasks, and although their authority is certainly greater than that of
other adult males, their ability to govern rests on the support given them by the
local community. Ability to use power not sanctioned by the local group is
minimal, because factionalization and "voting with one's feet" are always possi-
ble options. Most decisions are consensual, and most social control informal. In
general, leaders rise to their position by their ability to give more rather than to
accumulate more. Private property exists, but many checks provide effective
limits to an individual's ability to accumulate significant amounts of valuable
objects and resources. Before the need for trade goods became so well estab-
lished, private property consisted mainly of goods that were available to every-
one: One could either manufacture an item or acquire it through relatives or
friends. Otherwise, property is communal, and often it is sacred as well. Fur-
thermore, effective sanctions exist against nonsharing. Most often people who
refuse to share are not so much economically motivated as motivated to make
statements about ruptured social relations.
Equally different behaviors and concepts are being introduced in the areas of
religion, philosophy, and values. Perhaps, at least in the Papuri drainage, it is in
this area that the greatest change has occurred, because by far the most influential
change agents have been the Monfortian, Javerian, and Salesian missionaries.
These missionaries have introduced a myriad of concepts foreign to traditional
Tukanoan culture. There is no codified body of traditional Tukanoan knowledge,
law, theology, or ethics. Mythology gives meaning to the world, but it is absurd
to view myth as equivalent to Western moral, religious, legal, and philosophical
systems that are expressed in nonmythical terms. Certainly the individualism that
212
is so celebrated and so much a part of the very foundation of Western philosophy
lacks its equivalent in traditional Tukanoan philosophy. The conflict between
Tukanoan values and the Western tradition as taught and practiced by Catholic,
and more recently Protestant, missionaries is great. It is virtually impossible to
translate Western philosophy into traditional Tukanoan culture. This, however, is
not really the goal of missionaries, who wish to eliminate and replace far more
than translate.
Thus, a general process of deculturation has been occurring in all spheres of
traditional Tukanoan life. Most damaging, however, are the messages Tukanoans
receive that they are inferior and that they should relinquish control over their
lives in a number of areas. Missionaries may state, and some naively believe, that
such psychological blows are not their intention; nevertheless, it is undeniable
that they are the result of every increment of missionary presence and influence in
the region. Part of the reason the effects on Tukanoans are so devastating is that
the traditional Tukanoan culture provides very few counterexplanations or mech-
anisms for coping with their increasing sense of inferiority and powerlessness to
determine and control their affairs. Part of the reason is also that the change
agents in the region can back up their messages with so much supporting
evidence. In a number of respects the Catholic mission town is a microcosm of a
feudal fiefdom. Missionary personnel rule with a heavy hand, although not as
heavy as in the past. The governmental bureaucracy, on the other hand, even in
such an out-of-the-way place as Mini, carries the unmistakable stamp of routin-
ized, impersonal transaction. On the surface its representatives rigidly follow
policy and are organized along strict hierarchical lines. Behind the scenes, of
course, much is accomplished by the tried and true tactics of palanca (personal
influence); but for the uninformed and naive Tukanoan, such niceties are often
missed and usually unavailable.
From the very beginning, whites entering the region have shown their superior-
ity in rather convincing ways - with guns, chains, and other demonstrations of an
ability to coerce and apply punishments surpassing any Tukanoan counterparts,
technological or ideological. It is true that in the beginning missionaries failed
more often than they succeeded at mission founding, but the die was cast even in
the unsuccessful efforts, for the creation of needs that could be filled only through
contact with whites had begun. Unfortunately, Tukanoans are initially exposed to
what appear to be the positive aspects of civilization, seeing the negative ones
only later on. This is not to say that they look favorably on the ways guns are used
to intimidate them, for instance, but when they look at a missionary or govern-
ment representative they see tools, outboard motors and planes, communication
equipment, a seemingly unending supply of food mysteriously arriving in planes,
and a startling amount of luxuries - many changes of clothing, radios, combs,
and so on. And the unmistakable arrogance and self-confidence of some whites
(at times pathetically imitated by Tukanoans) also dramatically bespeaks power.
Furthermore, an economy that supports such nonfood-producing specialists as
213
The fish people
teachers, police, or priests is striking evidence of wealth. These people occupy

enviable, high-status positions; accompanying the envy is the desire to emulate,
and thus begins a process of deculturation that can produce some of the most
miserable and pitiful human beings on earth. It is impossible for Tukanoans to
understand the nature of the dominant social system that is affecting them so
totally. A rather complete confusion of cause and effect occurs (encouraged by
missionaries, rubber gatherers, and general-store owners), and Tukanoans become
confident that having the trappings of power will ease the discrimination against
them. It is painful to see the enthusiasm (and desperation) with which accultu-
rated Tukanoans seize upon the symbols of being a civilizado. Rather arbitrary
white customs such as small items of dress, bearing, and speech are meticulously
copied. Yet Tukanoans are not stupid; they know what alcohol abuse does. They
see what having a mutilated sense of self-worth does to their own and their
kinsmen's and friends' characters. They see the items bought at great personal
sacrifice rust, break, peel, or otherwise disintegrate far faster than handmade
objects, most often without possibility of repair, and their frustration increases.
An important component of this process is the intermittent reinforcement by
whites of Tukanoan attempts to mimic them. On the one hand, laughter and
ridicule are always waiting in the wings, to step onstage when a Tukanoan fails to
get something exactly right. On the other hand, praise and encouragement are
often the reward when a Tukanoan does master a specific behavior, such as asking
a priest's permission before removing a shirt when paddling a canoe. Frequently
the appropriate behavior indicates a willingness to become more dependent on
the national economy and culture in general and to become more appreciative and
loyal to the individual change agent in particular. I have on several occasions seen
priests show approval of Tukanoan men who bought plastic shoes, even though
well aware of the impracticality of such items, which split open in two or three
weeks. We can laugh at this, because we are affluent enough to have lost the
horror of being too poor to afford shoes. But North American change agents in
the Vaupes have their own lists of approved and disapproved behaviors. I believe
much of the pleasure so apparent in the faces and comments of these change
agents derives from two sources.
First, most change agents, like most people, enjoy personal power. They enjoy
being influential and imitated. They enjoy always being on top, but one remains
on top only when the value system placing one there is shared by all.
Second, attracting Tukanoans to material goods acquirable from a change
agent creates a dependency that helps achieve other goals. In this manner,
Tukanoan values slowly undergo a profound transformation without a great deal
of direct coercion. Respect for the giver declines, and respect for the receiver
increases. Group consensus evolves into a reliance upon individualistic decision
making. Arguments about social relations change into arguments about objects.
Respect and loyalty to the aged and one's kin turn into resentment of the implied
obligations, because not everyone has equal access to white trade goods. These
214
obligations are increasingly avoided whenever possible but not without bad
feeling on all sides.
Although change agents often have their own axes to grind when they encour-
age breakdowns in the traditional system, many times negative consequences
have little, if any, relation to the official agenda. One of the sadder parts of
acculturative processes in areas like the Vaupe"s is the amount of harm done by a
pervasive, yet often unconscious, ethnocentrism. Especially sad to see is the
unnecessary havoc wrought by well-intentioned change agents. Some of the
ridicule, racism, exaggerated emphasis on outward signs of inner worth, and
encouragement to engage in ultimately harmful behaviors are parts of con-
sciously planned programs, because they have a goal of creating dependency and
new felt needs. But other instances in which deep shame is induced, pride and
confidence destroyed, and dignity trampled into the ground are unintentionally
brought about by people who believe they have the target population's best
interests at heart. It is a sobering lesson, an illustration of the truth of the
proposition that a seemingly necessary requirement for a successful social system
is that its members define its values as the only ones possible and condemn out of
hand any others. Brief discussions follow of the specific types of change-
introducing programs in the recent history of the Vaupes region.
Extractive industries
The only natural resource whose exploitation has ever become firmly established in
the Vaupe's is rubber. This may be changed in the near future, if rumors about the
presence of uranium in the area have any basis in fact. Uranium has been discovered
in other areas of Northern Amazonia,1 but all attempts to locate deposits of
valuable minerals in the Vaupes have been unsuccessful up to the present.
Sporadic attempts to make lumber extraction a successful enterprise have
always met with failure of one kind or another.2 A major problem of industries
involved in the extraction of any natural resource in the Vaupes other than
precious minerals is the tremendous cost of transportation. This is a major reason
why the Vaup6s is considered to be such a backwater area - and, incidentally,
why white-Indian contacts have been relatively benign compared with many other
areas. The rivers of the Vaupes region all flow east into Brazil, far from any
Colombian centers of population. All rivers of any size are filled with rapids and
flood much of the landscape yearly, making road construction extremely difficult.
Any exports from the region must first be transported by water or overland to
MM under perilous conditions and then flown out at great cost. This in part
accounts for why there have been no national (not to speak of multinational)
corporations represented in the region until very recently.
The domestic rubber industry of Colombia has had its ups and downs since the
end of the rubber boom during the first quarter of the century, despite protective
legislation that requires rubber companies to purchase a certain amount of
215
The fish people
Colombian rubber. I was told of some rubber companies that buy their quotas of
domestic rubber from the Caja Agraria (the rubber cooperative) in Mitu and
immediately dump it in the river. This is cheaper than paying air transportation
costs for rubber that is often unusable because it has been poorly processed or has
been damaged during river transport.
The cauchero (rubber gatherer) works a section of the forest he has staked out
and bought a license for. Some Tukanoans have begun to work their own rubber
territories, which involves getting a license, acquiring a loan from the Caja
Agraria for the necessary equipment (including the extremely heavy
laminadora, the wringer-press), and transporting the equipment to their territo-
ry. White caucheros continue to travel by canoe to Tukanoan settlements before
the rubber season to sign up one or more men as laborers. Tukanoans who buy
trade cloth, machetes, fishhooks (and, more recently, shotguns), ready-made
men's clothing, and transistor radios from these men must work off their debt,
which, until recently, could involve a number of years. With very few exceptions,
caucheros are universally hated by Tukanoans. Although a few compadrazgo
relationships between caucheros and Tukanoans have occurred, in general the
situation is too oppressive and the cultural cleavages too great for there to exist
any more than the minimum understanding and trust. There seems, in fact, to be
a large amount of broken promises and abuses of even the minimal mutual
expectations that do exist. Many of the most violent incidents I heard about were
justified by caucheros as being necessary punitive measures against Tukanoans,
who were described as lazy, irresponsible liars who did not learn without such
harsh lessons. I was told that revolvers and other means of delivering on-the-spot
justice were necessary as deterrents and sometimes for protection. Caucheros
often advised me to carry a revolver.
For the most part, Tukanoan reminiscences of the rubber-gathering period's
heyday describe an unrelievedly horrible epoch in their history. Everyone has
anecdotes to recount, both about the cruelty of a particular cauchero and illustra-
tions of the general abuses of the system. Caucheros were sometimes refugees
from the period of violencia in Colombia (a period of political and general
violence lasting from 1948 to 1958); some were notorious criminals. Violence
and cruelty were inevitable with such men coming to a frontier area like the
Vaup6s, hoping to get rich as quickly as possible and leave. The rudimentary
forces of law and order tended to side with the caucheros, helping to round up
and punish runaway Tukanoans and terrorize settlements.
Nowadays, rubber gathering is very much in decline, mainly because of the
superiority of imported and synthetic rubber products. Few mourn its passing,
with the exception of the various "retired" caucheros living in Mitu, almost all
of whom are in dire poverty. These men are eager to recount their adventures and
are full of stories of betrayals, tragedies, and occasional moments of hilarity.
Most are nostalgic about the past, and a rather different picture from the
Tukanoan one emerges from their recollections. Their opinions about the
216
younger generation of Tukanoans, Catholic and Protestant missions, and the
situation in Mitu and nationally are salty and occasionally illuminating.
In 1970 the Colombian government issued a decree that Tukanoans must be
paid in cash for any labor in rubber camps; by 1974, according to Stephen Corry
(1976), debt-bondage had virtually ceased. It is not known whether or not buying
and selling the labor of Tukanoans by paying off a man's debts has completely
ended. The practice has been illegal for some time, but in Mitu several times I
was offered a Tukanoan in exchange for payment of his debts. All debts are
supposed to be cancelled yearly by government representatives in Mitu. Some-
times a priest will support a Tukanoan against a cauchero in a dispute. The
present situation is a far cry from earlier periods when police were used in raids
on settlements to help carry off Tukanoan men, who were manacled and threat-
ened with whipping and shooting. Oftentimes Tukanoans had no recourse but to
disappear into the forest for months at a time.3
Homesteaders
Very few colonos (individuals who migrate into the region with the intention of
establishing a permanent farm) were in the Vaupes during my stay, and this
continued to be the pattern in 1974 (Corry, 1976, p. 20). One has the impression
that the few men who have settled permanently in Mitu and its environs have
done so in a spirit of resignation and hopelessness; for some of them it would be
dangerous to return to areas of Colombia where they previously lived. Most of
them have Tukanoan wives. These families barely manage to eke out a living, and
many colonos become involved in affinal exchanges with their Tukanoan
brothers-in-law. Most of them drink a lot of alcohol and a few chew coca. This
picture contrasts greatly with towns farther to the west, such as San Jose" de
Guaviare, which are veritable boom towns. These towns are also reported to have
a healthier climate than Mitu, although diseases such as tuberculosis and malaria
pose serious problems.
The Colombian government

As indicated in Chapter 2, the presence of the Colombian government in the
Vaupe"s has been low-key and, with a few exceptions (i.e., the police during the
rubber boom), inoffensive. Doubtless its unobtrusiveness is in part due to the
region's being economically stagnant and with little potential. In such a setting,
an indigenous population's self-sufficiency is desirable to a government
struggling with terrible poverty in many other areas of the country. Conse-
quently, the government has played a minimal part until recently, putting the
Catholic church in charge of providing various necessary services, mainly in the
areas of education and health. This arrangment is spelled out in the Concordato,
an agreement between the church and the government. This arrangement con-
217
The fish people
trasts somewhat with the more direct influence of the Brazilian government in its
section of the Vaupes, which is even more distant from the Brazilian centers of
population and areas of development. Colombia is a very Catholic country, a fact
that undoubtedly has played a major role in its policy in the Vaupes. It is the only
Latin American country that takes care of all of its own missionary needs and
actually exports missionaries to other countries. In contrast, Brazil imports many
of its Catholic missionaries; all the priests at Yavaret6 (the Brazilian mission
town on the border where the Vaup6s and Papuri rivers converge) are Italian.
The greater number of non-missionary whites in the Brazilian Vaupes has resulted
in Brazilian Tukanoans' possessing more white-manufactured luxury articles such
as high heels, and distilling and drinking more alcohol.
In contrast, the only government representatives outside of Mitu during my
stay in 1968-70 were a DDT-spraying team and the occasional half-hour visits
made by the comisario (the equivalent of a governor) to mission towns on the
monthly plane. The promoter, a civil servant whose job was to promote indige-
nous affairs, made only occasional trips to Mitu and was never seen to travel in
the rest of the region. He was supposed to act as a liaison between the various
government agencies and private concerns dealing with Tukanoans. Unfortunate-
ly, he was universally and correctly seen as less than intelligent, and the only
thing he was seen to promote was free drinks from everyone.
In 1970 a few agronomists on the government payroll were sent into two or
three mission towns. They were to instruct Tukanoans about agriculture, particu-
larly cattle raising. I do not know the outcome of this program; it did strike me
that Tukanoans knew a good deal more about tropical horticulture than the
individual I talked to in Acaricuara.
Because most of the Colombian government's impact has been until recently
through the Catholic church or the Summer Institute of Linguistics, what remains
to be said regarding the roles of the state will be discussed in the context of
missions.
Missions
After many failures in the previous century, the Catholic church succeeded in
establishing a permanent base in the Vaupes with the founding of Monfort in
1914 by the Dutch Monfortian order. It is a task of no mean proportions to
attempt to evaluate its success - whether "objectively" or using the church's
own criteria. A distinction must be made among the various subregions in the
Vaupes, for they vary greatly not only in the general level of Tukanoan accultura-
tion but also, and even more so, in their degree of exposure to missions. For
example, at the time of my stay the Pira-parand had had SIL personnel for only
the past ten years and a permanent Catholic mission base only during the last
five. No SIL missionaries were on the Macucu or Tiqui6 rivers, and the nearest
218
Catholic mission was Paricachivera, in Brazil. Furthermore, the nature of the
Catholic missionary endeavor has changed through the years, particularly after
Vatican II had made its impact on the Colombian church. Its effect on the
Javerian order was apparently a strong one.
In addition, the very visible and well-organized Protestant missionary effort
has influenced the church's methods and strategy in many ways. The Vaupes has
been exposed to Protestantism for some forty years through the efforts of the
near-legendary Sophia Muller, a New Tribes missionary whose evangelical fervor
led her single-handedly to convert thousands of Indians to Protestantism in all
parts of Colombia (Stoll, 1981). More recently, SIL established missionary teams
in the region, leading to a remarkably high proportion of religious change agents
to Tukanoans and creating serious conflict between Catholics and Protestants in
the region. These two groups differ from each other in a number of fundamental
respects, not all having to do with religious doctrine. Each envies the other; each
believes that what it has to offer Tukanoans is superior and at times the only
possible solution to indigenous problems. Reading the literature published by
each provides glimpses into the internal organization of each group, its implicit
assumptions, and its justifications for claiming that it deserves full control of the
Vaupes region. Despite a fairly successful policy of maintaining on-the-surface
cooperation and congeniality the two organizations are in fact archenemies, and
when letting their hair down, members of each will discuss their opponents in
highly critical terms. The Catholic church points to the Concordato, its contract
with the Colombian government, which states that in exchange for the church's
agreeing to be responsible for educating the Indians, the government will expressly
forbid any proselytizing efforts by non-Catholic religions throughout Colombia.
SIL is very aware of the disadvantage it faces; none of the literature published by
it for public distribution in Colombia acknowledges the missionary aspects of the
organization.4 Instead, pamphlets and other printed materials stress the goals of
SIL in the areas of contributions to linguistics, bilingual education in Colombia,
health services, and preparing Tukanoans for their integration into Colombian
national society. Furthermore, SIL has its own agreement with the government
(via the Indian Affairs Agency), which gives the organization permission to set
up bases in specified areas of the country. The Colombian government pays for
the gasoline for airplanes and for constructing the airstrips at each site where a
team of two linguists are working. In exchange, SIL agrees to provide transporta-
tion, for a fee, to the general public (ironically, some of its most regular
customers are Catholic mission personnel). It is an expensive but invaluable
service. An important result of the arrangement is that airstrips are being built in
remote, otherwise inaccessible areas of Colombia. There are at least ten SIL-
constructed airstrips in the Vaup6s alone.
SIL's practice of relying exclusively on light aircraft for transportation has had
some unexpected effects. Many Tukanoan communities wonder if they will be
219
The fish people
selected to be the recipients of a SIL team - and the money and goods that
inevitably accompany such an event. Once when I was in Mitu I was approached
by two Cubeo Indians who requested that I radio SIL and relay the message that
their settlement had finished clearing an airstrip and was ready for "their" SIL
team. When asked why they wanted a team, they said they needed payment for the
work and other kinds of merchandise. I could not ascertain whether in fact an
airstrip had been built at that village. The SIL people I talked to about this
incident said that they were always receiving offers from Tukanoans to construct
an airstrip for a missionary team, but they had not heard of an airstrip actually
being built without prior arrangement with SIL.
In accordance with the agreement SIL made with the Colombian government,
some information on linguistics and certain ethnographic topics has been made
available to scholars (e.g., Summer Institute of Linguistics, 1975). SIL is also
interested in establishing a hegemony in bilingual education, because the gov-
ernment's interest dovetails with its own evangelical goals. Its successes in
winning important government officials and agencies over to its side has put the
Catholic church and in particular the two missionary groups concerned with the
Vaup6s (the Javerians and the Holy Sisters of the Mother Laura) on the defen-
sive. It is ironic that one of the battlefields of the war between Colombian
Catholicism and foreign-oriented Protestantism is the Vaupes, because it is in so
many respects a forgotten, backwater region of the country. Yet owing to the
region's multilingualism, SIL has organized a large campaign in the Vaup6s (the
resemblance to a military effort - including a vocabulary that speaks of "bases,"
"furloughs," etc. - is not lost on anyone). Because of the political advantages of
doing so, priests and nuns have recently - and ironically - found it worthwhile to
speak favorably of bilingual education. In fact, however, the concept that Tukanoan
children should be taught to respect their own language and the more radical idea
that they should be taught in both the indigenous language and Spanish are
anathema to most Catholic missionaries. Their three goals, very lofty and
impeccable in their eyes, and not usually kept separate, are to teach Tukanoans a
respect for their country, to teach them a respect for the Spanish language, and to
make them good Catholics. Thus, many Catholic missionaries still hold the
attitude that the sooner Spanish can be made to replace native dialects, the better,
and it is not difficult to bring it to the surface. It is my impression that although
some missionaries, both SIL and Catholic, make statements about the value of
cultural relativity and the spirit of the Declaration of Barbados on freedom of
religion and thought, most in fact are cultural imperialists to a pronounced
degree. (The Barbados declaration was promulgated by a group of anthropolo-
gists in 1971 on the situation of South American Indians.)
Mutual hostility has always existed between missionaries and caucheros. Often
the presence of Catholic priests, particularly in the Papuri, has acted as a
deterrent to some of the most excessive abuses practiced by rubber gatherers
toward Tukanoans' persons and property. This is recognized by Tukanoans in
220
conversations and in a story they tell of a jungle spirit called Kusir6 (undoubt-
edly from cauchero, although informants denied any connection, stating that this
name is derived from the call of Kusir6). Kusir6 is a cannibalistic, three-foot-
high "demon" who is fat, wears modern clothing,5 and smokes cigarettes.
Tukanoans flee into the forest or lock themselves in their longhouses in fear of
him. Barricaded in their longhouses, they quickly run out of food and water but
are too afraid when they hear his call to go to their fields or the river. This
describes how Tukanoans used to behave when they heard that caucheros were in
the vicinity. Finally, a priest banished Kusiro, causing him to flee the Papurf
region. This event is described at length in terms of a Catholic ritual using
prayers, incense, holy water, and tobacco.
Although the presence of both Catholic and Protestant missionary enterprises6
serves as a mutual check, and the competition for the approval of the relevant
Colombian governmental agencies inspires proposals that at least sound as
though they are concerned with Tukanoan welfare in a framework broader than a
strictly religious one, the Tukanoans themselves are the real losers in the ideolog-
ical warfare occurring in the Vaup6s. The Tukanoans of the mission town of
Acaricuara are a good case in point. This town of some thirty households has a
SIL team that has been there about ten years and a full-fledged Catholic mission
that includes several nuns, a brother, and a priest. Relations are superficially
cordial between the two sets of missionaries, but incidents periodically happen,
and rumors are constantly being circulated that serve to create misunderstandings
and factions among the Tukanoans there. It is quite clear from the SIL literature I
have received that SIL members perceive their main task as saving Tukanoan
souls - through long-term efforts at translating the Bible and through short-term
efforts at living a good, exemplary life and teaching any Tukanoans who show
interest about the life of Christ through songs, stories, and the like. This results in
a great deal of confusion for Tukanoans, although they are very much aware of
the competition existing between the two missions and will play one side against
the other when convenient (although not always successfully). In other areas of
the region, such as certain stretches of the Vaupe"s River, so much competition
has developed that one must declare one's religious persuasion before Tukanoans
will permit an overnight stay in a given settlement.
In 1969 a catechism program was begun by the Javerian mission to provide
religious instruction in all the settlements of the Vaupe"s rather than only in the
mission towns. The priests appointed one or two catechists per settlement who
were to conduct nightly sessions and give a little instruction in arithmetic and the
alphabet. They received a small remuneration for this. It is difficult to evaluate
the program, because it was only begun during my stay. I do know many
anecdotes about disputes involving catequistas at many settlements. This was
only to be expected in a culture in which people are suspicious of situations that
give one person authority over others. The fact that these persons are appointed
by an outsider rather than agreed upon by consensus only serves to exacerbate the
221
The fish people
problem. The catequistas tend to be younger Tukanoans rather than the traditional
holders of authority, and are the most missionized individuals in each settlement.
The particular situation at Pflmanaka buro was ludicrous. The catequista was
the eldest son of the family that was constantly at odds with the others in the
settlement (this family built a satellite house alongside the longhouse during my
stay). In addition, this particular young man was a difficult person in many ways
and had very ambivalent feelings toward priests and whites in general. Further-
more, despite three years of mission schooling, he could not speak Spanish (an
outcome not unusual in the Vaup6s). After a while, people stopped coming to
his sessions, despite dire warnings from him that they would bring punishment on
themselves from the priest in Acaricuara. In other settlements, however, the
catequistas commanded much more respect and were popular enough to weather
the inevitable disputes and bad feelings that arose from time to time.
It is completely true that Catholic missionaries' treatment of Tukanoans has
vastly improved from earlier periods (Corry, 1976, p. 19).7 Such punishments of
schoolchildren as whipping or making an entire class stand out in the noonday
sun with their arms raised shoulder-high have been discontinued. Still, great
variation exists among the present Javerians, even though as a unit they are less
strict than the earlier Monfortians. One priest proudly described to me in detail
how he personally had ordered two longhouses on the Inambu River destroyed
and had moved the inhabitants to the town of Los Angeles on the Papuri River. It
is ironic that this town, created by priests, is now the location of the SIL team
working on the Tuyuka language. A few of the clergy are liberal-minded, but
most priests are reactionary and narrow-minded. Furthermore, it is difficult for all
Colombian (and foreign, for that matter) missionaries to detect the instances in
which ethnocentrism rather than received gospel motivates them to alter Tukanoan
attitudes and behavior. Priests tell Tukanoans how to "eat properly," "walk
properly," and "sit properly." In general they are authoritarian, have an ingrained
sense of their own superiority, and at times seem to revel in the sense of sacrifice
and martyrdom they feel without acknowledging the pleasures they receive from
their personal power and status. They will not let Tukanoans bathe without
clothing on, and even on canoe trips a Tukanoan man must ask a priest's
permission before taking off his shirt. Priests must give permission before a
couple living in a mission town may get married. I was told of one case involving
a woman from Acaricuara and a man from Cano Vina to whom permission to
marry was never granted. Tukanoans said permission was not granted because the
woman, who worked in the mission kitchen, was a favorite of the mother
superior, who did not want the woman to leave.
Trouble erupts between missionaries and Tukanoans from time to time, which
sometimes is serious and which is never really satisfactorily resolved. One such
instance, described earlier, involved a priest who was openly in favor of a
marriage taking place between two Tukanos. Another case, which still rankles
Tukanoans, involved a mother superior who secretly paid a Tukanoan to obtain a
222
set of Yuruparf horns for shipment to Bogota. I also heard rumors of sexual
advances made by priests to Tukanoan women. It is, of course, difficult to assess
the validity of these rumors, because Tukanoans, especially those not living in
mission towns, are very confused about the social structure of the missionary
enterprise and often will make unwarranted assumptions about it. Incidents do
occur, however, and they create a great deal of anger and resentment in Tukanoans.
The fact that Catholic missions have been in the Papuri drainage region for
more than fifty years is readily apparent upon examining traditional Tukanoan
religious beliefs. Many intrusive elements from Catholic dogma are included in
stories people tell of the supernatural - encounters, strange occurrences, magical
formulas, and such. Less syncretism occurs in myths; the effect of Catholicism
on mythology is more one of simple extinction rather than radical transformation
of form or content. But when Tukanoans are telling quasisupernatural anecdotes
about themselves or other living people, often very strange mixtures of tradi-
tional and new themes and characterizations appear. For example, a fairly
acculturated Tukanoan explained to me that a particular kind of wahtia are red
"because they are from Russia." One list of directions for acquiring shaman
powers begins with the traditional notion of obtaining magical thorns and insert-
ing them right under the skin but also involves using a needle to open the flesh
and obtaining powerful splinters from the cross in the cemetery in Mitu. This has
to be done on a night of a lunar eclipse and requires engaging in mortal combat
with the wahti who protects the cross.
MM
At times Mini's inhabitants seem to be characters out of a Bunuel film -
surrealistic caricatures rather than real human beings. At other times they seem
smaller than life, oppressed and belittled by the same system oppressing the
Tukanoans. And sometimes they simply seem to be warm, sympathetic, coura-
geous human beings with remarkable capacities for self-satire and humor when
life is at its bleakest. Although not all of Mini's inhabitants deal with Tukanoans,
the overall impact of the town on them is extensive, both directly and indirectly.
Mitu's two or three bars sell soft drinks and beer, both exorbitantly expensive
due in part to air transport costs. Tukanoans also consume large quantities of
aguardiente (cheap cane liquor), and alcoholism is a problem in a number of
families, although thus far confined to Mitu and its environs. It is at bars that one
of the many forms of discrimination is openly practiced: Although the Tukanoans'
patronage is welcome, their presence is not, and they usually consume their
beverages outside. (On the rare occasions when a Tukanoan pays for a full meal at
one of the two residencias [lodging houses] he sits at a table like the others.) A
familiar nightly scene consists of Colombian whites gathered inside a bar,
drinking and playing pool, and a group of Tukanoans standing outside looking
through the windows and door, like moths drawn to the light. It is half-jokingly
223
The fish people
said that the only place where Tukanoans are welcome inside are the whorehouse
on the outskirts of town and the jail.
Occasionally one sees open and hostile discrimination in Mitu, particularly if
a white is angry or drunk. Although it is true that some of the abuses against
Tukanoans such as "buying" Indians (contracting for the labor of an indebted
Tukanoan) have been greatly curtailed, the existence of two classes of citizens is
apparent to all.
The terms each group uses for the other are ethnic rather than racial
classifications.8 Everyone affirms that most "whites" (blancos) in the region
have partial, if not total, Indian ancestry. No term in any Tukanoan language
classifies whites racially. Colombians of Negro descent are unquestionably
blancos to Spanish-speaking Tukanoans. The local term for half-breed, cabuco,
comes from the Portuguese caboclo, meaning "backwoodsman." It almost
always refers to a person with a bianco father and a Tukanoan mother. Apart from
this term, no labels exist that classify individuals in terms of their supposed racial
origin or by physical features. It is my impression that the criteria used to classify
people as Indian or white are (at least from the Tukanoan point of view) in the
first place his or her father language and in the second the desire to become a
bianco. This means that in one or two generations a Tukanoan can make the
switch, even if other Tukanoans in the area continue to be seen as Indian. I have
heard a Tukanoan woman living in Mitu speaking only Spanish to her children,
although she spoke her own language with friends. This is evidence that the way
to "become bianco" involves both the desire to switch and identification with
Spanish as one's first language. Of course, other factors are at play as well,
involving matters of dress, gestures, and other customs, and a repudiation of
behavior seen as specifically Indian.
It is my impression that for a Tukanoan to come to be perceived as white by
other whites, a generation is needed. A Tukanoan may be very fluent in Spanish,
but if his parents are known to be Tukanoans and their language and customs
were taught to the individual as a child, he will not be considered a bianco,
despite his volition and behavior. Still, his children will probably be considered
blancos if their first language is Spanish. Of course this process sometimes
involves denial of knowledge of any Tukanoan languages. Once when this
happened in my presence in Mitu, it was greeted with giggles from the compan-
ions of the young Bard man who was trying to pass as a bianco by speaking only
Spanish with me.
Conclusions
Many kinds of change agents are found in the Vaupe"s - their number and variety
is astonishing when one takes into account the small numbers of Tukanoans who
form their target population and the economic stagnation of the region itself. It is
difficult to predict whether the Catholic church or SIL and other Protestant
224
organizations will have increasing influence in the years to come. Although

denunciations of the presence of SIL in Colombia regularly appear in the national
press and other interested publications, to date SIL has not been forbidden to
work in Colombia. Rumors to the effect that "SIL has been given three years to
get out of Colombia" surface periodically, yet SIL is strong enough (counting on
important support from other U.S. interests in Colombia) to withstand this
criticism.
It is certain that if both SIL and the Catholic church maintain their toeholds in
the Vaup6s, the major battles will be fought over education. Tukanoans recog-
nize that several years of schooling for their children is absolutely necessary,
given the world they will inherit. This does not lessen the seriousness of charges
that can be made against education as it now exists in the Vaupe"s. In some
senses, what occurs in the classroom is representative of the entire religious-
cultural acculturation process occurring in the Vaupe"s up to the present: A very
sad corollary is the fact that much of the damage is unnecessary, even taking the
missionary perspective and its goals into account. Educational content and
teaching methods are archaic, and the real lessons taught to Tukanoan children
are the subverted ones mentioned at the beginning of this chapter - the lessons
that instill in each child an awareness of his or her own unworthiness. Further-
more, much of what Tukanoans learn is inefficiently taught and virtually useless
for the life they will lead. Lessons in Colombian geography dealing with terrains
of which Tukanoans can have no conception are a good case in point. From an
outsider's point of view, this is true for all religious instruction as well, except for
the advantage such knowledge provides a Tukanoan in his subsequent dealings
with missionary personnel. In addition, the Spanish that children leam in school
is minimal, considering the many hours of instruction given. The outdated,
inefficient, rigid teaching methods do teach part of the hidden curriculum rather
effectively, however, because they encourage passivity, dependence, resignation,
and self-deprecation. And, although instruction outside the classroom is more
oriented to practical matters such as how to use a sewing machine, these lessons
also have the effect of creating a greater dependence on white goods.9 These are
necessary lessons in any effort at turning autonomous, reasonably self-sufficient
groups into "good citizens" - Spanish-speaking Colombians who are at least
Christian (and, specifically, Protestant or Catholic, depending on one's point of
view) and who participate in the national economy. It is certainly true that by now
most Tukanoans want to go in this direction, and in a few years almost all will be
interested in giving up what is left of their culture and embracing the variety of
Colombian culture existing in mission towns and places like MM.
The cost of this change is exorbitant, however, and Tukanoans, although they
are the ones who will pay the bill, are unaware of most of the items on it. One
item is the decreasing control over their lives, both as individuals and collective-
ly, and decreasing pride and feelings of self-worth. Another is the great reduction
in security and predictability provided by their knowledge of their environment,
225
The fish people
including the complex way it is integrated into their systems of mythology and
other symbolic forms. Mythology explains, justifies, and elaborates the world
and Tlikanoans' place in it; the substitutes being provided are surely not as rich,
integral, or satisfying.
Tukanoans are embarked on a journey of increasing integration into the
national society and culture, and for many it is a journey toward real poverty,
alcoholism, disease, malnutrition, alienation, and despair. To some degree,
Tukanoans have unwittingly and often unwillingly sold their birthright for a mess
of pottage. It is true that this process has been going on for centuries and that even
the most traditional of cultures is never wholly autonomous and lacking contact
with other, frequently far more powerful cultures. Nevertheless, the situation in
the Vaup6s is exceptional because this particular mess of pottage is very poor
fare indeed and because it is being forced down Tukanoan throats in a very
speeded-up process. The reasons why this change is so rapid, so coercive - at
times in the extreme - and replaces so rich a traditional culture with one so
deficient are very complex. They have to do with the fact that it is really not
Colombian national culture Tukanoans are being integrated into but a position on
the lowest rung of an economic and cultural ladder whose highest rungs are
outside Colombia, insofar as Colombia itself is a client-state of the United States
and the many international corporations that fund the development efforts taking
place within its borders. The answers to why the future appears so bleak and
poverty-stricken for these people lie in analyses of these global economic and
political processes, and not in anything exclusively or intrinsically Colombian.10
226
13.
Conclusions: themes in Tukanoan
social identity
Types of comparisons
Much of the discussion in this book has followed out of postulating a series of
polarized contrasts in Tukanoan classifications of people. Some of these contrasts
constitute entire chapters (e.g., Tukanoan-Maku, male-female, Tukanoan-
white), whereas others (e.g., coresident-other, own-language group-other) are
described in several chapters. Although most contrasts are dichotomies, some are
trichotomies, most notably the agnate-affine-mother's child contrast. Accentuat-
ing such polarities is a useful tactic for explicating the major differences between
various classes of people and nonpeople important in Tukanoan social structure.
Another type of comparison consists of a class of elements ordered along a
continuum according to their possession of varying amounts of a particular
property. Examples of this are the birth order of siblings and the system of ranked
sibs within a language group.
A third type of comparison also involves the notion of continuum, but in this
case one end of the continuum is the core or nucleus and the other end is per-
ipheral, whose significance mainly derives from being the opposite of the core. An
example of this is the longhouse, the symbol of all that is human, social, and spir-
itual. Next in order on the scale are the river, the forest, and finally the wild and
anomalous features of the landscape, such as tatd boa (naturally occurring forest
clearings) and the rocky outcrops and caves sporadically looming high above the
forest floor. Another example of this type of continuum is the progression from
"us" to ' 'not-us,'' going from own group (maniya wadera,''our-language people''
or "brother people") to affines (mehko-mahkara) to mother's children (pahko-
mahkara) to other people (ahpera mahkara) and, presumably, to classes of people
such as whites and Makii. This particular set of contrasts is also a trichotomy;
whether it should be seen as three basic contrasts or as a continuum involving the
notion of increasing distance (geographical, social, marital, linguistic, emotional,
etc.) and differentiation depends on the context and what is being emphasized.
Identity components vary with respect to permanence. Those that are tem-
porary and relative rather than fixed were discussed primarily in the beginning
227
The fish people
of the book. One is normally both child and parent in the course of a life-
time. The roles of host and guest are similarly relative. These temporary roles
often contain symbolic elements derived from more permanent identities; for
example, hosts can be seen as female, guests as male. And as S. Hugh-Jones
(personal communication) notes, at Yurupari ceremonies hosts can be seen as
humans and guests as spirits.
It is by now apparent that Tukanoan classifications concerned with social
identity involve distinctive features and symbols in multiple levels of meanings.
The ideas expressed in the agnate-qffine-mother's child contrast can vary,
depending on what is being stressed. It can be a simple tripartite division,
especially when three actual language groups are being discussed and the
earth-air-water trichotomy is involved, or it can be seen as a core-periphery
ordinal scale. The merging or elimination of certain features at a higher level of
contrast figures prominently in Tukanoan conceptualizations of identity.
Some comparisons are taxonomic in structure, and some are paradigmatic.1
An example of a taxonomy is the Tukanoan contrast like-us-not-like-us made at
different levels of inclusion: One would perhaps begin with one of the divisions in
the agnatic terminology such as generation and end with the opposition
human-nonhuman. This taxonomy could involve as many as nine levels. A
paradigmatic relationship, on the other hand, is not based on levels of inclusion
but on the simultaneous occurrence of at least two dimensions, with a minimum
of two distinctive features per dimension. The set of personal pronouns in English
is a good example of a paradigm (the dimensions are person and number). An
example of a paradigmatic classification in Tukanoan identity conceptualization
occurs when the male-female and Tukanoan-Maku contrasts co-occur. Given
the association sometimes made between qualities such as spirituality and true
humanity and the features Tukanoan and male, it is not surprising to find an
association sometimes made between categories lacking these qualities: female
and Makii. We have seen that at times both women and Makii are associated
with each other and with natural symbols. But this association is appropriate only
in some contexts, and in others a connection made between women and Maku
would be wrong. It is important to resist the temptation to assume that symbolic
connections are universally applicable and permanent. Such simplistic and overly
rigid treatment ultimately leads to confusion rather than clarification.
Many more examples can be offered of multiple levels of meaning. One
already mentioned is that a sacred-profane continuum exists on one level and yet
on another level a sacred quality is imparted to everything. This is in part because
reality itself consists of many levels, extraordinary time and space lying behind
ordinary reality. This characteristic of multiple levels of meaning is clearly an
aspect of the interplay between rigidity and fluidity in Tukanoan social structure,
both in terms of on-the-ground realizations of the structure and in terms of the
structure as cognitive model. The fixed and the variable in Tukanoan social
228
Conclusions: themes in Tukanoan social identity
structure are linked, serving as foils for each other. For instance, Bar£ (fish
people) are associated with water, Tuyuka (mud people or clay people) with
earth, and Desana (wind people) with the air. In the Inambu region, the classic
affinal relationship is seen as occurring between Bara and TUyuka. Bara also
marry Desana, but Tuyuka, according to some informants, are "not supposed"
to marry Desana, and very few marriages have taken place between this pair of
language groups. Bara are not supposed to marry Tukano; it may be that Tukano
(toucan) are seen as sky people as well. On the Pir£-parana\ Barasana and
Tatuyo enter the marital picture with Bara, and Tukano, Desana, and Tuyuka
leave it. Thus, we see variability in terms of actual groups, and yet the structure -
a tripartite one with one pair seen as archetypical affines, one pair as not
intermarrying - is preserved. It is important to note that the situation is variable
not only with respect to which language groups are involved but also as regards
the location of the language groups' member settlements and these settlements'
neighbors. This is part of the reason why it is so difficult to arrive at a description
of the phratric organization of the Vaupes.
Another example involves the category Maku. Sometimes Maku is clearly
opposed to Tukanoan and associated concepts such as humanity, morality, and
knowledge. Here Maku tends to be paired off with wahtia, in the sense of "like
human but something wrong," with associations of excessiveness, imbalances,
and lacks (e.g., wahtia can lack nipples or toes; Maku are said to lack clothing,
body paint, and jewelry). Unacceptable behaviors such as improper language,
food, and sex partners, cannibalism, and an association with the forest are also
common to both categories. \et at other times Maku are not opposed to
Tukanoans but associated with the bottom position in a ranked group; Tukanoans
may speak of the lowest-ranked sib of a given language group as being of Maku
origin. Both categories share the designation "cigar lighter" and various sym-
bolic associations. The association with the forest is also shared: Lower-ranked
sibs are seen as inhabiting relatively upstream sites. Lower-ranked sibs are not,
however, associated with wahtia or powerful sorcery as are Maku.
Another example involves the contrast between Tukanoans and wahtia: Usu-
ally wahtia are the antithesis of Tukanoans, in appearance, in character, in
behavior. Yet as we have seen, sometimes in myths an ambiguous association is
made between wahtia and mother's children, pahk6-mahkara. Here what is
being stressed are the shared qualities of unpredictability, potential danger, and
absence of ongoing social, emotional, and exchange relationships, which do not
occur between Tukanoans and wahtia nor, structurally (they may occur in
individual cases), between pairs of pahk6-pon£ or groups that are pahk6-
mahkara to each other.
A final example is the contrast male-female. At times this is a simple
dichotomy, but at other times a core-periphery connotation is introduced, as
when men are spoken of as the only spiritual, cultural beings, and women are
229
The fish people
simply the category lacking these qualities. At other times there is a merging into
a kind of androgynous state: "A unity of a higher order than the elements" (C.
Hugh-Jones, 1977, p. 203) is created.
The longhouse and the shaman, both extremely complex concepts laden with
symbolic meaning, admirably illustrate what I mean by multiple and merging
meanings. The longhouse is probably the key metaphor for human identity. It is
where such quintessentially human activities as cooking, eating, dancing, and
singing are carried out. Babies, not yet human, are born outside and brought in;
the dead, recently human, are buried beneath it. Curing is done near its front
door. The boy mentioned in Chapter 2 who had been bitten by the snake had to
recuperate outside until he had sufficiently returned to human status to be allowed
back in. But at times this human-nonhuman association disappears, and all
beings are spoken of as having their longhouses: animals, wahtia, dead spirits,
and the ancestors. Sometimes, rather than being opposed to nonhuman and
nature, the longhouse merges with them, becoming the whole universe. A
taxonomic structure consisting of human body-family-sib-language group-
universe can be constructed, with the longhouse representing each level.
The shaman is a more complicated symbol. Having a similar quality of
multiple meanings, he also demonstrates characteristics of ambiguity and bound-
ary straddling not found in the symbolism of the longhouse.
In a sense shamans are both human and more than human. They are at the top
of a hierarchy of people possessing wisdom and sacred knowledge (see Reichel-
Dolmatoff, 1971, p. 249), at the bottom of which are women and children. They
interpret myth, see many levels of the universe, speak several Tukanoan lan-
guages plus their own esoteric language, and in general engage in more activities
associated with humanness than do ordinary humans. But shamans are also pre-
cariously close to the nonhuman side and at times slip over into it, as when they
turn into evil sorcerers and practice magic against their own community. This dan-
ger always potentially exists because shamans associate with nonhumans, travel to
other levels of the universe, and have supernatural powers. As we have seen, there
is always the danger of becoming like that with which one associates.
Shamans are also intermediaries: They usher in new states and statuses at
rituals and mediate at other events. Thus, again by association, they are at times
seen to be "like" women, animals, or the dead. Shamans are thus a prime
symbol of the merging that occurs throughout Tukanoan conceptualizations of
identity. By confusing some of his own identities and crossing some boundaries,
the shaman clarifies what these are and preserves them - by preventing inappro-
priate merging, divining what has been crossed or confused, and restoring
boundaries already transgressed. This is also what happens in the way Tukanoan
social identity is structured and played out in other areas of life, an illustration of
my suggestion that ambiguity and variation are features intrinsic to the system of
social identity, for they help to define and maintain what does not vary by
showing what does and the limits to this variability.
230
Themes associated with social identity
Crucial to an understanding of Tukanoan social identity are the themes of

maintenance, moderation, balance, risk, relationality, permanence versus tran-
sience, and regionality. Maintenance refers to two related but distinct ideas in
Tukanoan thought: (1) the need to maintain one's identity by behaving properly
and (2) the need to maintain Tukanoan society by means of proper individual and
collective behavior. The idea of maintenance is connected to the notion that
identity is process as much as idea (Crocker, 1977). Numerous examples of this
can be offered with respect to individual identity. Kamaweni, by not behaving like
a good elder brother, forfeited his claim to be one, with grievous consequences:
He was indirectly killed by his younger brother. One must maintain one's identity,
one's claim to being human, a kinsman, a Tukanoan, or whatever, by marrying,
speaking, and living properly. One must also continue to associate with appropri-
ate others. Many myths demonstrate how people can become like the beings they
associate with, whether wahtia or animals, as was mentioned in the case of
shamans. Shamans and other leaders, particularly headmen, must continue to
show the proper qualities of moral restraint, selflessness, wisdom, and so forth,
to continue to merit their positions. And all Tukanoans must engage in certain key
behaviors, such as speaking, dancing, and singing, to show they continue to
deserve to be considered Tukanoans - individuals and settlements - in good
standing with their neighbors and Tukanoan society as a whole.
The second idea concerning maintenance - proper individual and collective be-
havior to maintain the group and, at the broadest level, Tukanoan society itself-
has to do with a longhouse group's observation of proper restrictions on food and
other behaviors. Such activities are also for the benefit of the individual, but infrac-
tions hurt not only the individual at fault but the entire settlement. In addition are
the periodic ceremonies that, when carried out correctly, renew the local group by
maintaining continuity with the ancestors and the past in general. Food rituals,
hom playing and making music in general, myth chanting, periodically burning
beeswax, are all ways of ensuring the continuance of Tukanoan society.
Such activities concerned with maintenance and prevention contrast analyti-
cally with rites and other behaviors that can be seen as restorative in nature.
These concerns introduce the next three themes pervading Tukanoan conceptual-
izations of social identity: balance, moderation, and risk. Themes of balance and
equilibrium appear time and again in Tukanoan thinking of how to live a good
and happy life, and much of the content of restorative ritual has to do with
discovering whatever imbalances exist and restoring harmony. Death, birth,
taboo breaking, and illness are all instances of imbalance, and everyone must
work to restore harmony and balance when such threats disrupt the moral order.
The activities connected with lunar eclipses are excellent examples of this
concern, the purpose being to get the moon back into the sky and to restore him
to his proper place and activity.
231
The fish people
Risk and danger cannot be completely avoided, however, for they are inherent
in the process of life itself. Life means change: Growth, initiation, marriage,
birth, and death are all supposed to happen, and yet all are risky. They must be
properly ushered in and lived through, usually with the help of a shaman (the
exception is marriage, a completely secular event). Processes inherent to living,
in both its physical and social aspects, are potentially dangerous. Eating means
killing animals and the possibility of breaking a taboo; sexual activity is danger-
ous in itself and necessitates association with other classes of people. Mainte-
nance of the moral order is thus a kind of tightrope: To live is to risk and to
become imbalanced during certain periods such as birth and initiation. Certain
rituals reduce the risk of such periods by ensuring maximum equilibrium in the
individual and the community. When the balance is upset, certain ceremonies re-
store it. For example, naming, painting, washing the infant, and making the
mother's milk safe for drinking are ways of restoring balance and eliminating the
imbalances in the cosmological order brought about by birth.
The necessity of maintaining boundaries is relevant here, for it is tied in with
notions about balance and harmony in Tukanoan thinking about the world and its
occupants. In part, both shamans and Maku are complex symbols because they
cross boundaries. Shamans cross over into the other levels of the universe, and
Maku cross the boundary between themselves and Tukanoans. This occurs
conceptually when lower-ranked Tukanoan sibs are described as "like Maku"
and in actuality when intermarriage occurs or when a Maku group is absorbed by
a Tukanoan language group as its lowest-ranked sib. But the boundaries between
shamans and supernatural beings and between Maku and Tukanoans should be
kept distinct. The same is true of the boundary between the living and dead as the
myths of pehu, and Live Woman demonstrate. The best example of boundary
maintenance is perhaps that between the language groups. As shown in Chapters
5 and 9, the differences between Tukanoan languages are seen to be as important
as they are because of the role language plays in marking the boundaries between
these otherwise similar exogamous units in Tukanoan society.
The theme of moderation appears as an important element in many areas of Tu-
kanoan life, particularly in activities concerned with maintenance and restoration
of health and balance in the community and the cosmos. Moderation in eating,
sleeping, sexual intercourse, and many other activities goes a long way toward
keeping everything in balance, particularly for initiated men. Again, many illus-
trations in myths can be found. Women, Maku, and wahtia are often contrasted
with initiated men with reference to excess. Male restraint and moderation are
expressions of male morality, male solidarity, and male responsibility for main-
taining the community. Many kinds of ceremonies vividly display such concerns,
often through the use of inversion: For example, during rituals excessive con-
sumption is the rule but of soul food rather than food (C. Hugh-Jones, 1979).
With respect to the theme of relationality, in the United States and other
Western countries, if we choose to pay attention, we can discern many kinds of
232
identity "tags" that remind us of our uniqueness. These tags range from a myriad
of account numbers (such as social security, credit card, driver's license) to all
manner of legal conventions that highlight individual existence and responsibili-
ty, often to the exclusion of any other kind. An example is the increasing
acceptance of an individual's right to take his or her own life; this argument is
almost always couched in terms of individual rights, as contrasted to arguments
concerned with the rights of others (e.g., the stance of the Catholic church).
Arguments both for and against abortion are similarly phrased. A look at earlier
periods in Western history reveals that such was not always the case; notions of
collective identity in terms of privilege or responsibility, have been greatly
eroded. Familiar examples are collective blood guilt and other duties between
kinsmen. Such a stress on individualism is not characteristic of Tukanoan
culture, even though Tukanoans would probably agree that individual identity
derives from a unique intersection of social, spatial, and temporal variables. But,
I would argue, for them this has meaning only relationally, whereas there is a
much stronger element of permanent distinctiveness in Western notions. True, all
those classifying numbers and markers of distinct identity are related logically to
the other markers in the same category, but their primary function is to separate
each individual from all others (even if, as is the case with social security
numbers, it is a separation to allow the bureaucracy to treat us all the same).
Social security numbers are relational only in the sense of showing our relation-
ship (one of distinction) to all other workers in the United States. And they
continue to distinguish us after a change of name, address, or even death.
Thus, while it is true that certain identity components are clear-cut (no
ambiguity, no dual memberships or fence-straddling, etc.), ambiguity and con-
tradiction inhere in all Tukanoan social roles. An example is the pivotal position
Tukanoan women assume when they marry. They remain agnates (sisters, daugh-
ters, loyal supporters, and members in absentia of their natal longhouses) and
must reconcile this with their new affinal roles (wives, mothers, loyal supporters
of their husbands' and children's longhouses). These potential contradictions are
usually handled by stressing one role in a given situation and deemphasizing
another. It is only when this is impossible (e.g., in answer to a question of
whether women go to their natal descent groups' ancestor longhouses after death
or to their husbands' and children's) that the contradiction becomes apparent.
What is important to note here is the familiar theme of fluidity and variability - or,
put another way, relationality, because who people are in very large part depends
on the social context defining the roles they play. Like the Bororo of Central
Brazil, the Tukanoan self exists mainly with respect to, in relation to, other
selves (Crocker, 1977, p. 144).
Obviously vast differences are found between Tukanoan society and the West or
other complex society - in social structure, scale, political centralization, and so
forth. I cannot discuss all the implications of these differences here, but it is clear
that certain purposes, if I may speak in functionalist terms, are served by the mech-
233
The fish people
anisms that establish and maintain individual identity in the West that are not
served in the Tukanoan case. Tukanoans have no need to distinguish individuals
bureaucratically or otherwise in the fashion we are all familiar with; indeed, the re-
verse is true. For many reasons the more fluid (temporary) and collective (rela-
tional) system of marking selves and others in Tukanoan society suits their purposes.
Another example of ways in which putatively unchangeable, ascribed statuses
are affected in Tukanoan society by an individual's relationships with others is
language group identity. This is a crucial aspect of a Tukanoan's notion of self,
and I would argue that it is a central one, following Barnett (1977, p. 277) and D.
R. Miller (1961, pp. 282-83) on the notion of centrality of certain identity
components. The meaning this has, however, is highly dependent on the location
of one's settlement and its relationships with the other settlements in its sphere of
interaction (and their language group identity). Although, for instance, once a
Bard always a Bard, what this means is extensively shaped and influenced by
whether one interacts with Tuyuka, Barasana, or other language group. How one
talks about being a Bard, the normative statements one will make, the mytholog-
ical statements, are influenced by these factors.
Another example of contextuality that obscures an underlying contradiction is
the never-ending dispute about relative patrilineal sib rank. Only after careful
discussions with various Tukanoans does one realize that no agreed-upon ranking
system exists that classifies all presently existing sibs, although the top ranking
one(s) will usually be acknowledged as such. Even here, however, at times some
Tukanoans will aver that in earlier times the reverse was the situation, but "those
people refused to call us elder siblings, so we had to address them this way!'
These seemingly clear-cut features of identity, when being specified, show that
their clarity depends on contextualization and reveal an underlying ambiguity and
contradiction: A relational quality emerges.
Tukanoan public ritual expresses relationality and collectivity in many respects
and downplays individuality. For example, Tukanoans have no wedding ceremo-
ny, and the rituals that express and celebrate affinal relationships and ongoing
alliances do not say anything about the individuals as individuals presently or
previously in such relationships.
Male initiation ritual is another example - as unlike our initiation rituals (e.g.,
a Bar Mitzvah) with respect to individuality as one can imagine. Tukanoan boys
are initiated in a group with little acknowledgment of them as distinct individuals
(S. Hugh-Jones, 1979; rituals such as menarche ritual that do mark individual
change are smaller and more private). Furthermore, public rituals in general stress
collective identity and behavior, in such activities as dancing and chanting. In the
few instances in which an individual does perform alone, this person is con-
strained by a role that suppresses much of his or her individuality. Of course this
is true for many rituals in many societies, but it is pronounced in Tukanoan ritual.
It is interesting to note that the shaman who performs more as an individual
than other Tukanoan ritual roles allow is seen as more likely than others to misuse
234
his power. We can characterize this process as one of increasing individuation (in
terms of power, self-interest, etc.) accompanied by decreasing humanness (this
type of shaman eventually turns into, and is called, a jaguar).
In sum, Tukanoans obviously recognize and think of one another as distinct
personalities. The differences between the way they acknowledge the individual
and our individualism lie in the degree of overt and covert distinguishing of the
individual as such. Ambiguity and contradiction certainly inhere in Western
conceptualizations of the individual, whether they be religious, legal, or psycho-
logical, but we have many institutions that explicitly work at avoiding ambiguity
in some spheres of social action and social classification. These types of institu-
tions are lacking in Tukanoan society. The need for all the distinguishing individ-
ual identity features and markers (such as birth certificates) fosters a notion of the
individual as existing apart from the group for some purposes, and in the West a
person's existence is demonstrated, validated, legitimated, and to some degree, I
would argue, created by these various mechanisms. Many instances can be found
in Western literature of identities changed or even created or eliminated by
various documents (just look at several Gilbert and Sullivan plots). Although
Tukanoans certainly go through various ceremonies involving the establishment
of or change in a given social status, the group does the work, and to a greater
extent the individual is always submerged in a group. Tukanoans can be seen to
be selves and others through one another, in collectivized and contextualized
relationships. In the West, many relationships are mediated by some form of state
or municipal machinery that contributes to a definition of that relationship (e.g.,
marriage or adoption). This finally has a profound influence on how a citizen,
parent, worker, or even deceased person comes into being and is conceptualized.2
With respect to the theme of permanence versus transience, it would be
difficult to make the argument that Tukanoans conceive of an individual as
existing as an individual self throughout time. For us, that John Jacob Jingle-
heimer Schmidt was at his birth, is, and will continue to be a distinct, individual
self is a reasonable and meaningful proposition. This idea of survival of individ-
ual identity is not, I would argue, nearly so meaningful to Tukanoans, for many
of their identity features are not seen as permanent, and many others are played
down more than in Western society. This is not to deny that ambiguities exist in
Western conceptualizations of the ways in which individual personality maintains
itself as distinct and with recognizable continuity throughout life and after death.
But such maintenance and survival is a theme in many religious and secular
philosophies (in Christian thought, in purgatory, hell, and heaven; in countercul-
tural and occult thinking, in other worlds or planes of reality). Present in many of
these is a concern for a continuation of some identity features acquired during a
period of life on earth, especially with respect to responsibilities for one's
actions. Why this is true is the topic of another book.
Such ideas are not foreign to Tukanoan thought; for example, murderers are seen
as suffering different fates, at least immediately after death, from people who have
235
The fish people
committed no serious crimes. Westerners are hardly free of ambiguity and uncer-
tainty about survival of the personality, but I would maintain it is of greater con-
cern to many in the Judaeo-Christian tradition than it is to Tukanoans. Tukanoans
speak of what we might term "spirits" and "souls," but none of this is clearly
worked out: "The 'other world' is not a unitary concept in Indian thought: it em-
braces the distant natural world of the remote forest, the ancestral world and the
sinister world of the recently dead" (C. Hugh-Jones, 1979, p. 113). Some as-
pects of death, especially matters to do with corpses, are unpleasant matters, a-
bout which the less said the better (see Moser and Tayler, 1965, pp. 51-54).
The cyclical features of Tukanoan conceptualizations of the individual self can
be seen as a denial of anything like permanent individual identity. Naming
involves a cycle of only two generations. Thus, in the name and in the idea of
reincarnation of a person's soul via name transmission, the notion of a distinct
personality's survival is downplayed. Tukanoans say that after death the spirit
goes on a journey to ancestral homes under the rivers. But such longhouses are
not the subject of elaborate and comprehensive description (unlike other areas of
Tukanoan cosmology). There is only one myth in my collection that deals with
life after death (this is also true for the corpus collected by the Hugh-Joneses).
Another way of speaking of life after death is in connection with animals, where
an equivalence is made between game animals and people, especially children. A
third way is as a kind of reincarnation (although it should be noted, my data are
very skimpy) in which Tukanoans eventually turn into small birds, then into
yellow butterflies, and finally into river foam. A process of decreasing individua-
tion is clearly apparent.
It is important to note that "life" does not end at death: The soul lives on in its
namesake and other processes take place involving travel to ancestor longhouses
or reincarnation. But the idea that individual personality continues is very
minimal and vague. In general, the Tukanoan individual comes from an undiffer-
entiated source and returns to one (such as river foam). Certain ancestors are
indeed spoken of as distinct personalities, but these are far from humanlike (S.
Hugh-Jones, 1979). They can also be seen as inimical to individual identity,
because contact without proper precautions is fatal, for the ancestors want to
return with such individuals to ancestral time and space.
Naming is a clear example of the principles of temporality and relationality. If
one is a Bard or other Tukanoan, one receives a name from a very finite list
owned by the sib (C. Hugh-Jones, 1979; Arhem, 1980). The name ideally comes
from a FF or FFZ who has recently died. An association is thus made that works
in two ways: The deceased loses his or her name (confirming a state of affairs in
existence since death because people are very reluctant to mention the names of
the dead); and a connection is made between the infant's identity and the long line
of ancestors who have possessed that name. Relationality, nondistinctiveness,
and transience (the name is eternal, but the individual is not) are thus expressed.
Nicknames, although more distinct, are more temporary. The other ways of
236
referring to and addressing individuals, such as by personal pronouns and kinship

terms, are even more relational and suppressive of individual identity. Spanish
and Portuguese names do differ in certain ways from this characterization. A
Tukanoan's Spanish name is given more or less spontaneously and has no
necessary connection to anyone else (except to a Catholic saint, a connection
very imperfectly understood).
Thus, in this book I have tried to show that although we can analyze Tukanoan
social identity in terms of features and dimensions (i.e., in terms of gender,
genealogy, exogamy, language, territory, etc.), identity is both structure and
action. An.important difference between Tukanoans and ourselves is our greater
tendency to see identity as "true" or "false" in a rather absolute sense. This
tendency is due in part to the greater possibility of being an imposter in the West;
in Tukanoan society everyone knows or knows of everyone else. A second
difference is a stress on consistency in the presence of legal or other kinds of
formal requirements. For example, one cannot be married and single at the same
time in the United States, but the situation is often more ambiguous in Tlikanoan
society. What one is in Tukanoan society - locationally, genealogically, maritally,
and so on - is sometimes far from either categorical or necessarily true or false
but something socially negotiated and either agreed upon or disputed and, if
disputed, may or may not be resolved. My characterization of Tukanoan social
identity as action and process is applicable here. Although we do encounter
similar situations in our own society (see, for example, Dominguez's discussion
of supposedly rigid racial identity [1977]), in general everyone in Western society
would say that such problems are ultimately resolvable when all the "facts" are
known - and that they should be resolved. Of course, such "facts" can often be
negotiated. In Tukanoan society, however, such situations either do not arise
(e.g., of being an imposter or, as occurs in Shakespeare's plays, of mistaken
identity) or when they do exist they are not resolved. They simply fluctuate and
remain in contradiction or fade from consciousness as interest in them diminish-
es. An example of this is the chronically disputed question of relative sib rank.
Thus, although a person, whether American or Bara\ is obviously unique, the
focus in the respective societies is quite different. Tukanoans do not focus on the
particular individual as a unique intersection of attributes but emphasize one or a
set of attributes, often opposing them to others. This opposition is determined by
context and is thus transient; in other contexts it is irrelevant. Memberships in
groups and categories are stressed or ignored, depending on the situation. At a
Bar Mitzvah the individual is ritually passed from one category to another, as in
all initiations, and certainly some features of his identity are irrelevant. But he as
an individual is the focal point. Tukanoan initiation rites stress the individual
participants as individuals far less; at some initiations even the extremely impor-
tant exogamous group memberships of the initiates are irrelevant (S. Hugh-
Jones, 1979). Therefore, the position of an individual Bara or other Tukanoan is
not one of a lack of differentiation but is far less one of unique and permanent
237
The fish people
differentiation than in our own and many other societies. Discussions abound in
both scholarly and popular publications of the implications of this view for
Western psychological identity, and specifically mental and spiritual health.
My final theme is the importance of regional orientation and awareness in
Tukanoan social identity. The sources of this regional perspective are numerous,
as we have seen throughout this book. Tukanoan social structure is a contributing
factor, in particular its system of named patrilineal exogamous local sibs. This
system has obviously helped to create an extension and complexity of interaction
and cognitive orientation far beyond the local group. Tukanoan cosmology
mirrors this complexity with, among other things, its tripartite symbolism of
earth, sky, and river and their connections with specific language groups. Nearly
all aspects of Tukanoan society and culture, from cosmology to language and
speech to kinship terminology to marriage patterns, show the pervasiveness of
this regional orientation. I have mentioned Tukanoans' pan-Vaup6s (and beyond)
perspective, discussing their interest in and knowledge about geography, showing
their social geographical model to be one of an ever-increasing area with ever-
increasing and gradual social and cultural differentiation but one which in theory
has "no social, geographical, or linguistic limits" (C. Hugh-Jones, n.d.).
The portion of Tukanoan space we have examined is a rather large one, but I
hope that I have illustrated some of the advantages in taking a regionally oriented
approach. It is obvious that at least a rudimentary grasp of the regional
organization of the Vaupds, both in interactional and cognitive terms, is essen-
tial for any comprehension of Tukanoan social identity.
An associated factor is the underlying cultural homogeneity that unifies the
Vaup6s region. On the strength of numerous pieces of evidence presented
throughout the book, I have argued that this homogeneity is extensive, much
more so than first appears to be the case - when one is initially confronted with
the complex linguistic and "tribal" situation. I am calling the tribes by a new
term, which (appropriately) places an emphasis not on a real cultural diversity
but on a diversity that serves to organize interaction between basically similar
social units. A games analogy suggests itself: Football needs easily distinguish-
able uniforms and Monopoly needs easily distinguishable tokens. Of course, in
many respects the game played in the Vaupes is far more complex - life always
is. For one thing, a true cultural diversity does exist, as exhibited by the regional
variation described. Such diversity is sometimes due to simple geographical
variation, sometimes due to differences in local history, and sometimes due to
different degrees of acculturation.
But these factors do not completely explain the diversity occurring in lan-
guage, in associations made in myths between animal and geographical symbol-
ism and specific language groups, nor in the specialization in manufacture of
certain artifacts. These differences, unlike diversity produced by underlying
differences in cognitive models of the world, spring from underlying agreements
about it. The basic patterns that give rise to these differences are shared -
238
including rules for action, assumptions about peoples' motives for following or
not following the rules, symbolic representations of these rules and motives - in
short, much of what life is all about. This does not mean that there is total
agreement, for the participants in any social system never completely share their
perceptions and understandings. The Vaupes is no exception, as the variability is
organized into a rather large and complex system, and this is why one cannot talk
about Bard or Tukano society, even though genuine cultural differences can be
found between this pair, or any pair, of language groups. We must speak of
Tkkanoan society and culture because most of the differences we can observe
(and they are actually far fewer than the similarities) are the result not of cultural
barriers separating units but of cultural links uniting them into a single system.
A note on types of evidence

There is no question but that an epistemologically eclectic approach raises
serious problems in research and interpretation. Still, to address the topic at
hand, that of Tukanoan identity, such an approach is mandatory because many
levels of behavior and thoughts - revealed in mythological vignettes, Tukanoans'
comments about real-life cases, statistical descriptions of marriage patterns -
pertain to the questions being asked. Extremely thorny problems arise, however,
when trying to tie together information from such disparate sources, especially
when the results from one type of research are not complete and therefore are
suggestive rather than conclusive.
Usually, of course, pieces of information resting on each rung of the ladder
reaching from "real" on-the-ground behavior to the most abstract, idealized
structures somewhere in the stratosphere reinforce and clarify one another. Many
examples of this have been offered in the preceding discussion. It is when they do
not that problems of data reliability and validity and painful choices about
interpretation arise. I have shown that some of the discrepancies in the data
gathered so far are due to the extremely complex nature of the system itself and
the genuine regional variability contained within it. The final picture of Tukanoan
identity, a goal not yet attained, will incorporate and structure this variability,
showing how it fits into the overall scheme.
Another source of complication stems from the fact that ideal and "real" are
not simple polar opposites but the top and bottom rungs of a many-runged ladder.
The "ideal" statements one obtains from informants about how things ought to
be are not the only pieces of information from the idealist section of the ladder.
Some cognitive rules for behavior, like grammatical rules, are accessible and can
easily be verbalized. Others, at least equally determinative of behavior, are
deeper and less likely to surface during interviews and similar types of research.
Ample evidence of this has emerged in the discussions of Tukanoan-Maku
interaction, male-female roles, and choices about marriage. Maku is a category
in a postulated Tukanoan domain we might gloss as "beings" or perhaps
239
The fish people
"human and humanlike beings." As members of this category, Maku serve as

projections: They marry their sisters, have no shame, are "jaguar children," do
not speak properly, and so forth. But really to understand the meaning of this, we
obviously need to understand flesh-and-blood Maku. It is fortunate that two
investigators, R Silverwood-Cope and H. Reid, have provided material that takes
us a long way toward that goal. We have discovered that Maku really are
"professional hunters" and provide some Tukanoan settlements with forest-
derived products for manufacturing ritual objects. Thus they merit their associa-
tion with the forest. In other respects, however, the chasm between ideal and real
is a wide one indeed. With this information we are far better equipped to
understand the total nature of Tukanoan-Maku interaction. We are better able to
understand the degree to which Maku are different because they actively main-
tain their distinctiveness (rather than its being due merely to the residue of
different origins). It also is clear that some Maku are periodically assimilated
into Tukanoan groups via a mechanism proscribed by the rules: intermarriage.
Looking at questions of identity from all angles - seeing the view from as
many rungs of the ladder as possible - allows us to see more clearly where ideal
and real do in fact fit together. For example, in the Papuri region, the marriage
sample reveals a number of patterns that reinforce the information obtained about
Tukanoan social structure from more idealist types of evidence. There is in fact
extremely low settlement endogamy, all of it due to the effects of recent concerted
efforts on the part of Catholic missionaries. The marriage sample also reveals
virtually no language group endogamy for the Papurf-Tiquie region. Another
finding was minimal (0.8 percent) polygyny.
Statistical patterns also support evidence obtained from other sources of data.
For example, I have argued that Tukanoans have a regional orientation rather than
a tightly constricted one of three or four neighboring longhouses. There is some
evidence that Tukanoans maintain both narrow and wide perspectives; Tukanoans
do think of neighboring longhouses as much more important than those far away:
Close neighbors tend to be guests at festivals, for instance, and Tukanoans
characterize the ideal marriage as one occurring between double cross-cousins -
with the implication that this partner's natal longhouse is quite near. Still, the
marriage sample revealed an average of roughly twenty-two linear miles between
spouses' natal settlements, with an actual travel distance of many more miles.
The average marriage, then, does not occur between neighboring settlements (of
course many actual marriages do; this figure is the mean distance). It is also
interesting to see how many marriages are "two-couple" (direct exchange)
marriages, which supports the ideal Tukanoan version. Further work needs to be
done on questions of kinship relatedness between spouses (what I have termed
FZD and MBD marriage) and on phratric organization, but results obtained so far
have shown several very promising leads.
Of course the relationship between "real" and "ideal" is complex: For
example, interference from ideal conceptualizations of lukanoan society undoubt-
240
edly influences informants' reports about actual marital behavior. More marriages
with Maku undoubtedly have occurred than the single one I recorded (H. Reid,
personal communication). Difficulty in obtaining information about the Tukano-
Tukano marriage discussed in Chapter 7 is also due to interference from Tukanoan
notions of proper marriage - this marriage, in fact, was described as not a
marriage. Marriages that do not survive for long and are without children tend to
be forgotten, a difference between our legalistic concept of what a marriage is and
Tukanoans' processual one. This is the main reason I have not offered any
quantitative data on divorce: Determining a rate of divorce in such a situation is
extremely difficult to do.
The problems of incompleteness and difficulties in reconciling and synthesiz-
ing different types of data that do not of their own accord mesh together are not
inconsiderable. But neither are they insurmountable. What I have been able to
paint here is less than a complete portrait of Tukanoan society and culture, but the
basic outline and some of the detail are clear. We have looked at the ecological
setting, briefly examined the history of the area, and noted the important
demographic, settlement, and economic features of Tukanoan society. With an
understanding of these levels of the substratum, itself in part created and molded
by the social system it supports, and with an understanding of the main features
of Tukanoan social identity, we can begin to understand the entire social order in
both behavioral and ideational terms. We can begin to comprehend the role of
physical and social features of the landscape in formulating this social order: from
rivers and forests all the way to Maku, whites, and wahtfa. We understand far
better the enormously complex system of languages, seeing them in their roles as
communicators of both referential (semantic) and social information from and
about their speakers. We begin to understand the logic of Tukanoan concepts of
time and space as they pertain to identity, so that, for instance, we see the sense of
signaling high status by referring to certain agnates as "grandchild." We under-
stand far better the range and limits of this social order and are well on our way to
comprehending the totality of Tukanoan experience. This experience differs for
individual Tukanoans, depending on their location in the Vaup6s territory, on
whether they are Bar£, Tukano, Tuyuka, or other, but it is a similar experience in
most respects. More important, we can see that these differences are, at a
fundamental level, understood and agreed upon by all Tukanoans as meaningful
and essential differences that unite them all in an overarching system in a universe
with its own multiple identities.
We have looked at slices of time and space in the Vaupes region, concentrat-
ing on understanding social identity in traditional Tukanoan society. ' Tradition-
al" is not a static concept, for it is obvious that Tukanoan society has always been
evolving, during the pre-Columbian and postcontact periods. Still, as pointed out
in the preceding chapter, Tukanoans now face unprecedented change at unprece-
dented rates, with the tragic result that traditional Tukanoan culture is no longer
able to adapt and instead is becoming extinct. The final polarity in Tukanoan
241
The fish people
social identity - the one of Tukanoan-white - is, like the others, couched in terms
of absolute and categorical contrasts, but not simply because such contrasts are
meaningful in Tukanoan conceptualizations of identity. Unlike the contrasts of
Tukanoan-Maku or male-female, for example, this contrast is not so much used
by Tukanoans in their own construction of their world view but is one imposed
upon them. The contrast is immense and the cultural barriers are all too real.
Many of the differences are incomprehensible to Tukanoans, and others are all
too comprehensible, for they convey unmistakable messages about unworthiness,
ugliness, powerlessness, shame, ignorance, and the advisability of dependency.
Soon we will have one more instance of the label "traditional" referring to a past
time and an extinct culture rather than to a group of people alive and surviving
with their understanding of the world and their place in it intact. I hope that these
pages give some glimpse into the genius of Tukanoan culture and the tragedy of
this loss, both theirs and ours.
242
Notes
Preface
In general, Vaupes refers to the region and Tukanoan to the sedentary, riverine
populations living there. Except in this Preface, Tbkanoan personal names are
pseudonyms; site names are not, because many of them have changed since 1970.
See, for instance, the sensitive and perceptive accounts by Berreman (1972), Briggs
(1970), Maybury-Lewis (1965), Powdermaker (1966), and Rabinow (1977).
I was granted permission to do fieldwork by the Instituto Colombiano de Antropologia
on the condition that I supply $10,000 worth of equipment (in the form of a small
airplane) for its anthropological projects. This was unacceptable to the National
Science Foundation, which was to fund the project, even if the $10,000 were to come
from another source. Several of the restrictions on foreign anthropologists carrying out
research in Colombia were eased in 1979.
1. Purpose and organization of the book

1 For discussions of hunter-gatherer populations and regional organization, see
Leacock, 1955, pp. 31-47, and Lee and DeVore, 1968, especially pp. 150-57.
2 It is also true that tribe has been used to characterize Vaupes social units other than
the language-affiliated ones; see Jackson, 1972, pp. 8-18; S. Hugh-Jones, 1979, pp.
22-24; and Sorensen, 1967, for discussions of this.
3 Reported during a symposium on "Amazonian Marriage Practices" at the 1973 annual
meeting of the American Anthropological Association, New Orleans.
2. Introduction to the Central Northwest Amazon

1 Most of the primary data discussed here were gathered in the Colombian half of the
Central Northwest Amazon, resulting in the book's having an overall Colombian
perspective. Generalizations made about the entire region are to be seen as more
provisional when applied to the Brazilian sector, but for reasons of simplicity, Central
Northwest Amazon and Vaupes are used interchangeably.
2 See Goldman, 1963, p. 14; Koch-Griinberg, 1909-10, vol. 2, p. 65.
3 For information on specific Vaupe's languages, see Briizzi Alves da Silva, 1962;
Koch-Griinberg, 1909-10; Kok, 1921-22; West and Welch, 1967, 1972; and Waltz
243
Notes to pp. 20-49
and Wheeler, 1970, on TUkano; Waltz and Waltz, 1972, on Uanano; Smith and Smith,
1971, on Southern Barasano (Barasana); Stolte and Stolte, 1971, on Northern Barasano
(Bara); and Salser, 1971, on Cubeo; See Jacopin, 1972, on the status of Yukuna,
Matapf, Letuama, and Tanimuca. Wycliffe Bible Translators / Summer Institute of
Linguistics, Inc., dedicates itself to translating the Bible into all the languages of the
world. Founded by William Cameron Townsend in 1942, its international headquarters
is in Santa Ana, California (Wallis and Bennett, 1959; Hvalkof and Aaby, 1981).
4 I am extremely grateful to S. Hugh-Jones for help in thinking these points through.
5 See Reichel-Dolmatoff (1975, pp. 62-63) for a discussion of the history of the region.
See also Coudreau, 1887, and Humboldt, 1822, vol. 7, p. 383; vol. 8, p. 145. The
history of missions is mentioned in Goldman (1948, 1963); Koch-Griinberg (1909-10,
vol. 1); Misiones del Vaupes (1966); and Stradelli (1890, p. 433).
6 Misiones del Vaupds, 1966, p. 13, quoted in Reichel-Dolmatoff, 1971, p. 7.
7 New Tribes Mission (NTM), founded in 1942 by Paul W Fleming, has its headquarters
in Sanford, Florida, and currently over 1,500 missionary members. "According to an
NTM leaflet: 'The New Tribes Mission is a fundamental, non-denominational faith
missionary society, composed of born-again believers and dedicated to the evangeliza-
tion of unreached tribal peoples . . .'" (Wright and Swenson, 1981, pp. 3-4). For a
discussion of nativistic movements in the region, see Goldman, 1963, p. 16; Van
Ernst, 1966b; and Wright, 1981.
8 Sources on the Maku include Biocca, 1965; Cathcart, 1973; Koch-Griinberg, 1906;
MacCreagh, 1926; MacGovern, 1927; Reid, 1979; Silverwood-Cope, 1972; Tastevin,
1923; Terribilini and Terribilini, 1961; Van Ernst, 1966; A. R. Wallace, 1889/1972 ;
and Whiffen, 1915.
3. The longhouse
1 This contrasts somewhat with Goldman's description of Cubeo longhouses and
surroundings, which, he says, are oriented toward the river such that "Along a
well-populated river such as the Cuduiarf the communities resemble rough-shaped
beads widely spaced along the thin strand of the sinuously curving river" (1963, p.
28).
2 See Goldman, regarding poorer Cubeo communities, which tend to locate their fields
closer to the house. Fields are always more distant in the more "prosperous and
prideful communities" (1963, p. 29). In the Pira-parana, fields are adjacent to the
longhouse clearing (C. Hugh-Jones, 1979, p. 28; S. Hugh-Jones, 1979, pp. 43-44).
3 The longhouse and its environs and the symbolism of architecture and space are
discussed more fully in Goldman, 1963, pp. 39-42; C. Hugh-Jones, 1977, 1979;
S. Hugh-Jones, 1977, 1979; Moser and Tayler, 1963, pp. 443-45; Reichel-Dolmatoff,
1971, pp. 104-09; and Rodriguez Lamus, 1958.
4. Economic and political life

1 See Silverwood-Cope (1972, pp. 83-86) for a description of an eel-catching
technique found only among the Maku. This is noteworthy because they generally
have a reputation for being nonriverine people.
2 S. Hugh-Jones (personal communication) states that the Makuna and Taiwano eat
snakes.
3 I.e., Old Arturo went into trance. I am not certain whether this longhouse is an
animal longhouse in the sense described earlier.
244
Notes to pp. 50-77
4 Further discussion of forest- and hunting-related topics can be found in Arhem, 1976;
Dufour and Zarucchi, 1979; Goldman, 1963; C. Hugh-Jones, 1979; S. Hugh-Jones,
1979; Koch-Grunberg, 1909-10; Reichel-Dolmatoff, 1971; Silverwood-Cope, 1972;
and Torres Laborde, 1969.
5 See, for example, the debate regarding settlement pattern and cultural evolution
occuring in situ or as a result of migration (Carneiro, 1960; Lathrap, 1970; Meggers,
1973). See Johnson, 1974, on the question of carrying capacity.
6 See, for example, discussions in W L. Allen and Holshouser de Tizon, 1973;
Denevan, 1971; and Harris, 1971.
7 Population numbers vs. resource availability has been the topic of much recent debate
(cf. Beckerman, 1979; Gross, 1975; Hames, 1980b; Ross, 1978).
8 Bitter manioc is so called because of its high prussic acid content; this acid must be
washed and leached out or cooked sufficiently to render it nontoxic. All varieties of
manioc are the same species, Manihot esculenta (see Rogers, 1963, and Schwerin,
1971).
9 For an excellent and comprehensive discussion of the entire manioc process, see C.
Hugh-Jones, 1979, 174 ff.
10 C. Hugh-Jones (personal communication) comments that this does not totally
eliminate guests' being a burden on their hosts. She observed a number of cases of
visiting freeloading relatives, who were an undeniable drain on the resources of the
families putting them up.
11 It should be noted that the military reasons for staying together in a well-fortified
longhouse have disappeared, and this is doubtless also a contributing factor in the
current widespread disruption of longhouse units.
5. Vaupes social structure

1 Although I have no real data on why this trend has occurred, I find the arguments
correlating uxorilocality with periodic absence of the son-in-law (see Helms, 1970;
Casselberry and Valavanes, 1976) and with the father-in-law's availing himself
of the labor of the son-in-law and trying to have the balance of power in his favor (see
Oberg, 1955; T. Turner, 1979) persuasive and probably applicable to the Tukanoan
case. Of course, all of this must be related to the general breakdown in traditional
norms in the region as well as why agnatic groups of males are no longer as adaptive
as they apparently once were.
2 There are no longer any longhouses in Brazil (Howard Reid, personal communica-
tion).
3 See Goldman, 1963, pp. 90-113, for a discussion of Cubeo sib ancestors as distinct
personalities. Discussions of Vaupe's sib structure can also be found in Bidou, 1976,
pp. 135-80; Briizzi Alves da Silva, 1966, pp. 84-123; C. Hugh-Jones, 1979, pp.
22-31; and Reichel-Dolmatoff, 1971, pp. 189-201.
4 C. Hugh-Jones (1979, pp. 28-29) discusses evidence for several sibs being located at
a single site in separate longhouses in earlier times. She notes, however, that this may
always have been an ideal only and that then, as now, sibs were located along a stretch
of river at different settlement sites.
5 Pind kumoa, sometimes described as a submarine-like vessel, looking like an
anaconda, with the ancestors inside. See Kumu and Kenhiri (1980) for a drawing of a
Desana anaconda canoe.
6 Many of these characteristics apply to the Northwest Amazon as a whole, although
data are lacking for a number of societies. The Tukuna are exceptions to items 2 and 4
in this summary.
245
Notes to pp. 83-97
7 This is similar to the argument presented by Barth (1969) when discussing ethnic
groups and boundaries and the processes of polarization and differentiation occurring
among increasingly structurally similar social groups.
8 See Sorensen, 1970, p. v: "The exogamy among these tribes, reinforced by the
cultural identification of tribe with language, must supply the energy that keeps each
language distinct and alive."
9 C. Hugh-Jones (1979, p. 18) also notes the use of language as a way of talking about
descent.
10 Cf. C. Hugh-Jones (1979, pp. 283-86) and Arhem (1981, p. 345).
11 Space does not permit a full discussion here (cf. Arhem, 1976; Bidou, 1976; C.
Hugh-Jones, 1979; S. Hugh-Jones, 1979). Briefly, the Hugh-Joneses have classified
the Pira'-parana' into seven exogamous phratries. Although language normally corre-
lates with exogamy, there are a few exceptions: (1) some speakers of Makuna belong
to the phratry Makuna, whereas others belong to the Barasana phratry; (2) some
speakers of the "Barasano del Sur" language (Summer Institute of Linguistics
terminology) belong to the Taiwano phratry while others belong to the Barasana
phratry; (3) a Tatuyo-speaking sib belongs to the Bara phratry; and (4) an Arawak-
speaking sib belongs to the Bard phratry. The following section offers a somewhat
different treatment of the concept of exogamous phratry, which also attempts to make
sense of the lukanoan tendency to include speakers of a different language into a
single exogamous group.
12 C. Hugh-Jones (1979, p. 21) has a slightly different definition: the phratry "is an
association of Exogamous Groups united by the rule of exogamy but not occupying a
continuous area." Cf. also S. Hugh-Jones, 1979, pp. 24-25.
13 For a more comprehensive discussion of the pahk6-mahkara set of terms, see
Chapter 7 and Jackson, 1977.
14 I fully concur with C. Hugh-Jones's warning that "it is impossible to arrange
Exogamous Groups into three neat and mutually exclusive categories on this basis,
just as it is impossible to arrange Exogamous Groups into a pan-Vaupe's phratric
pattern" (1979, p. 36).
15 For a discussion of Bard kinship terminology with respect to this, see Jackson,
1977.
16 Three Tliyuka men out of 145 were married to Desana women; 2 Tuyuka women out
of 112 were married to Desana men in the marriage sample I collected. Bruzzi Alves
da Silva (1966. p. 90) also notes no Desana-Tuyuka marriages.
17 Bruzzi Alves da Silva (1966, p. 87) states that some Tukano said that the Bara" are
their paxko-ro (grandparents), and that he could not find the reason for this
designation because there is no allusion to this in their legends, but it is certain that
intermarriage is proscribed.
18 This is not a translation of an actual quotation, but a condensation of several
conversations.
19 See Jackson, 1976; Jackson and Romney, n.d. A. Kimball Romney analyzed the
marriage data and derived the model in Figure 5; I gratefully acknowledge his interest
and contributions. Figure 5 represents marriage preferences as they appear after
removing the effects of the size of the groups involved, using a technique developed
by Romney (1971, pp. 191-213).
20 His list of sibs indicates that this group is Bard.
21 For a discussion of differences in outlook and practices in a sample of lowland South
American societies illustrating just how varied their inward or outward focus can be,
see Shapiro (n.d.).
22 If such a word can be applied to hostilities as traditionally practiced; "raiding" and
246
Notes to pp. 99-115
"feuding" are perhaps more indicative of the actual situation. Cf. Service, 1968, p.
160, on Yanomamo "warfare"; also see Fried, Harris, and Murphy, 1968, p. xvii.
23 However, it is my impression that Bard and Tuyuka, at least those of the Inambu
River, are more forest-oriented than the Cubeo as described by Goldman. His
statement about the Cubeo that "the forest is undifferentiated terrain" does not apply
to the people of Pumanaka buro.
24 See, for example, the discussions of Colson (1973), Coppens (1971), and Thomas
(1972) on the Gran Sabana of Venezuela; Harner (1972), Oberem (1974), and
Whitten (1978) on the Ecuadorian Montana; Bodley (1972) on the Campa; Lathrap
(1973) on the Shipibo; Morey and Morey (1974) and Leeds (1964) on the Orinoco
llanos; and Basso (1973) on the upper Xingu.
25 Speech differences used for marking exogamous unilineal units have been reported
for New Guinea (Sankoff, 1968) and Australia (Owen, 1965). See also Hymes,
1968.
6. Kinship
1 Kinship in this chapter is to be seen as interchangeable with relationship as a label for
the set of terms that show consanguineal and affineal relationships. (See Needham,
1966, for a discussion of alliance systems of social classification, which purportedly
have nothing to do with kinship, genealogically speaking; a good case can be made
for Bard as well [see Jackson, 1977].) I use the term kinship because the naming and
pronominal systems also show kinds of relationships among Tukanoans and because I
use the term relationship in other contexts. My use of kinship does not necessarily
imply that Bari "kin terms" are ultimately derived from genealogical relations.
2 The discrepancy between Bard kinship terminology and Tukanoan society is
unavoidable in a kinship system involving marriage between patrilineal descent
groups identified with different languages. Also, classic Dravidian terminologies do
not distinguish between types of cross-cousins or between matrilateral and patrilateral
parallel cousins, as does Bar£ (Jackson, 1977), but Bara is still to be seen as a
variant of the basic Dravidian type.
3 I stress the " ± mother Barf" feature because the transcription of the Bara terms
with FZS, MBS, etc., although a good literal gloss, can be misleading. When the
MBS, MBD terms are used, they really are not referring so much to the fact that a
child is the son or daughter of an MB but to the fact that the child's mother is not
Barf; it is a default category and avoided wherever possible (Jackson, 1977).
4 For a discussion of the long history of both approaches and the debates between the
two schools of thought, see Buchler and Selby, 1968; Goodenough, 1970; Leach,
1958; Lounsbury, 1956, 1965; Needham, 1971; Scheffler, 1966, 1973; D. M.
Schneider, 1965, 1972.
5 Space does not allow a presentation of all the myth material I collected. My major
focus of research was not on mythology, and I make no claim regarding the
completeness of the myths collected or the methodological rigor used when eliciting
them. Most of the myths were gathered from non-Spanish-speaking BarS men and
were written down with the help of Lino, Mario's second son, who lived near Mitu
and spoke Spanish.
6 Both mother and children show progressive dehumanization in their speech, an
important point in this and a number of other myths. The mother talks to her wahtf
husband in his language, thereby showing her accommodation to him and the
accompanying loss of human characteristics. In the end, while her children are
247
Notes to pp. 117-27
accusing her of loving Namakuru too much, their dialogue turns into pure bird talk.
The same situation is portrayed in the end of the Kahp6a orero story: The
protagonist's relatives turn into monkeys after losing their eyes, and the myth ends
with an imitation of their monkey talk.
7 The dilemma of the necessary but risky association with afflnes, who, by definition,
are never one's "own kind," and for this reason alone somewhat dangerous,
frequently appears in myths.
8 With perhaps far-reaching consequences (see Chodorow, 1978; Dinnerstein, 1976).
9 Until this happened, the fight was mostly screaming and rather ineffectual fisticuffs
(at least compared with the experts in U.S. films and television). Use of any weapon,
however, is deadly serious. After the fight, the eldest son left the settlement for some
weeks.
10 Goldman (1963, p. 114) finds Cubeo suffixes distinguishing lineage, sib, and phratry
siblings. In Bara, true siblings can be distinguished from patrilateral parallel
cousins, but a more lengthy construction must be used to be more precise about other
degrees of sibling closeness.
11 Pond glosses as "children of one set of parents." Mahkara glosses as "people."
Thus, the term pahko-pond is usually used with reference to a single ego's set of
matrilateral parallel cousins, whereas pahko-mahkara is a more inclusive term,
referring, for instance, to a settlement's position vis a vis another settlement or at
times to an entire language group.
12 The one exception is the seldom-used tenara, "in-law-people," from tehiil-6,
"brother-/sister-in-law?'
13 The tactic of showing the essence of a relationship by picturing it in its closest terms,
even when this is impossible in actual life, occurs frequently in myths. The
Namakuru myth is an illustration of the conflicts and dissimilarities in the uterine
half-sibling relationship between totally human children and their deer-child half-
brother. In another myth, referred to earlier, the pahk6-mahkii/-6 set's affine-of-
affine meaning is portrayed by showing the relationship between two husbands of the
same wife.
14 See Goldman, 1963, on ceremonial friendship and S. Hugh-Jones, 1979, pp.
114—15, on ritual kinship in connection with male initiation.
7. Marriage
1 Several ethnographic and geographic factors place limitations on gathering such data
in the Vaupes. One is the difficulty of travel, which placed constraints on my ability
to collect and cross-check firsthand information. The multilingualism of the region is
another. A third is the absence of names of local descent groups (when not
coterminous with sibs), confusing in an area in which residence sites are temporary
and a significant proportion of people do not currently reside where they are
"supposed" to according to the patrilocal rule of residence. A fourth related factor is
the built-in slippage between named social groups, e.g., sibs, and the people who
constitute these groups. A fifth limitation is the general taboo on pronouncing
personal Tukanoan names, in particular those of the dead.
2 See Jackson, 1972, pp. 127-34. for further discussion of the sample.
3 See C. Hugh-Jones, 1979, p. 104, and Silverwood-Cope, 1972, p. 194, in this
regard.
4 I speak of exchanging sisters rather than exchanging brothers because, first, this is
the way lukanoans talk about it, and second, it is the women who change residence at
marriage. This usage in no way is meant to imply that women are to be thought of as
248
Notes to pp. 128-49
pawns or "rubbish" (Fox, 1967, p. 117) as has been suggested for other strongly
patrilineal systems. The Yanomamo (Chagnon, 1968; Shapiro, 1972) speak of
women in derogatory terms; this and the manipulation and coercion of women and
the frequency of polygyny are not nearly as characteristic of Tukanoan society.
5 Out-migration and perhaps other factors have resulted in many unmarried women in
mission towns. Nuns told me that these young women could find no men because
they did not want to leave the mission towns to marry "pagan" Tukanoans.
6 One of the goals of this book is to illustrate just how complex decision making about
marriage usually is. The above comments are not to be seen as for or against either
the "structures" or "sentiments" position in the polemics on "proximate" and
"final" causes, because both types of argument are far too simplistic when they
claim to explain marriage behavior (as opposed to marriage norms). See Homans and
Schnieder, 1955; Le"vi-Strauss, 1969; Needham, 1962.
7 See Briizzi Alves da Silva, 1962, p. 414; Giacone, 1949, p. 21; C. Hugh-Jones,
1979, pp. 64, 69, 93ff., for further discussion of mock capture of wives and actual
raiding for women.
8 With the exception of the Tikuna (Cardoso de Oliveira, 1961, 1964).
9 See C. Hugh-Jones (1979, p. 58) regarding close agnatic marriage in a Barasana
"servant" sib; also see my discussion of Maku marriage in Chapter 8.
10 For further discussion of these and other measurements, see Jackson, 1972, 1976.
11 This is the Spanish version of Portuguese caboclo, "backwoodsman," but in the
Vaupe's it means half-breed.
12 Just how important a role environmental constraints play in the marriage systems of
small-scale societies is a topic of current debate. Some authors argue that the
ecological and demographic givens cause the particular system being examined. In
this view, other factors (e.g., rules of marriage prescription or preference) are
"epiphenomena" (Gilbert and Hammel, 1966) or "idealizations of the parameters of
performance" (Lehman, 1974). No one would disagree that environmental givens set
limits on any abstract marriage system structure, as well as forming the substratum to
its design (that is, are determinative in a broadly conceived way - e.g., elementary
systems are found in societies with certain levels of techno-socio-cultural integration,
and complex systems are found in others). Some excellent studies address themselves
to correlations existing between environmental features and marriage patterns - for
example, between rainfall and section system development in Australia (Birdsell,
1953; Yengoyan, 1968). And certain demographic variables (related to birth rate, or
relative age of spouses) seem to set definite limits on the presence and operation of
particular marriage patterns (Gilbert and Hammel, 1966). It would be good if more
studies of this kind were carried out among tropical forest horticulturalist societies,
such as those found in lowland South America and highland New Guinea (see Glasse
and Meggitt, 1965, for promising beginnings).
8. Tukanoans and Maku

1 This discussion of the Maku, particularly in the areas of ecology and social
structure, has profited greatly from criticisms of an earlier draft from Stephen
Hugh-Jones and Howard Reid.
2 Cathcart (1973), a SIL linguist, uses the term Cacua ("people") for all Maku. She
states that Maku is a single language with two principle dialects (1973, p. 101). Reid
(personal communication) states that at least one, and probably as many as four,
different languages of the Maku family are spoken on the lower west tributaries of
the Rio Negro and north tributaries of the Japura-Caqueta (personal communica-
249
Notes to pp. 149-56
tion; see also his doctoral dissertation on the Hupdii Maku [1979]). Mason classifies
Maku as an "independent" (1950, p. 257), whereas Greenberg places it within
the Macro-Tbkanoan subfamily of the Andean-Ecuatorial family (1960). Word lists
are available in Biocca, 1965; Giacone, 1955; Koch-Griinberg, 1906; and Rivet and
Tastevin, 1920.
3 Summaries of research on the Makii are found in Biocca, 1965, and Silverwood-
Cope, 1972, p. 312. Investigators giving firsthand accounts of contact with Makii
include Biocca (1965), Giacone (1949, 1955), Goldman (1963, p. 70), Koch-
Griinberg (1906), MacCreagh (1926), McGovem (1927), Spruce (1908, p. 344),
and Whiffen (1915, p. 59-61). Silverwood-Cope (1972) and Reid (1979) are the only
investigators to carry out full-scale investigations of the Makii. Other sources of
information on the Maku include Briizzi Alves da Silva, 1962; Galvao, 1959;
Moser and Tayler, 1963; Nimuendaju, 1950; Reichel-Dolmatoff, 1967, 1971;
Schultz, 1959; Van Ernst, 1966a, 1966b; and Whiffen, 1915.
4 See, for example, Cardoso de Oliveira and de Castro Faria, 1971; Hymes, 1968;
Lathrap, 1968; and Martin, 1969.
5 See, for example, Koch-Griinberg, 1906; MacCreagh, 1926; McGovem, 1927, p.
357; Silverwood-Cope, 1972; and Whiffen, 1915, pp. 59-61. I never heard a
Tukanoan classify Maku in physical terms.
6 Examples of such a subsistence mode are the Ache" (Guayaki) described by Clastres
(1972), the Akuriy6 (Kloos, 1977), and the Cuiva (Arcand, 1972). Many groups
who have been classified as South American hunter-gatherers in fact either practice
horticulture, engage inregularizedand fairly frequent food exchanges with horticulturalist
neighbors, or both. Examples include the Sirion6 (Holmberg, 1969), and the Yaruro
(Leeds, 1964). See Hames, 1978, on the question of a purely hunting-gathering
existence in Amazonia.
7 McGovem (1927, p. 183) reports exogamous units among the Maku.
8 Including my information on the BarS and that in C. Hugh-Jones, 1979; S.
Hugh-Jones, 1974; Langdon, 1975; and Reichel-Dolmatoff, 1971.
9 For a different description of Desana orientation toward hunting, see Reichel-
Dolmatoff, 1971, p. 17.
10 This is a similar situation to that described by Turnbull (1965) for Pygmy-Bantu
relations.
11 H. Reid (personal communication) notes that such seeming "poverty" is a typical
trick of the Hupdii, in order to scrounge more items from their Tukanoan hosts.
12 Maku as cigar lighters is a frequently heard description in the Vaupe's (see Galvao,
1959, p. 42), and it is interesting to note a parallel classification for the lowest-ranked
sib in those Tukanoan language groups that are characterized by a hierarchy of five
ranked sibs. In this scheme, each position is labeled with an occupation: headman,
head dancer, warrior, shaman, and cigar lighter (C. Hugh-Jones, 1977; S. Hugh-
Jones, 1977).
13 See Fock, 1963, p. 153, for a similar account of a girl taking care of children among
the Wai Wai.
14 H. Reid (personal communication) is of the opinion that such a personalized
servant-master relationship is rare and represents the Tukanoan ideal rather than the
norm.
15 Reichel-Dolmatoff discusses the sexual side of Tukanoan-Makii relationships the
most thoroughly, noting that "close emotional and sexual relationships" exist
between Desana and Maku and that Maku are "not only servants but represent the
female element, a sexual object, upon which very ambivalent ideas and emotions are
projected" (1971, pp. 19-20). For Reichel-Dolmatoff, the Desana exhibit a very
250
Notes to pp. 156-68
marked sexual repression, are puritanical, and view sexual relations "with great fear
and anxiety?' Unfortunately, he does not give any concrete evidence for this, in
particular whether Desana sexual impulses toward Maku are always repressed and
sublimated or are ever acted upon.
16 This is confirmed in all the literature on the Maku, with the exception of Whiffen,
who says "the Maku will intermarry with any . . ." (1915, p. 61). H. Reid (personal
communication), however, who knows of about eight Maku women married to
Tukanoans, states that ideally Maku do not mind Maku women marrying Tukanoan
men (although at times they object in practice), but that the opposite is disapproved
of.
17 Perhaps the only division of labor among the Tukanoans that is the result of
qualitative differences between the people concerned is the sexual one.
18 See Barth, 1969; Haaland, 1969; Knutsson, 1969; and Seligmann and Seligmann,
1911, for accounts of these types of "symbiotic" relationships. Both Lee (1968, p.
31) and Woodburn (1968, p. 52) report regularized trading between hunter-gatherers
(!Kung Bushmen and Hadza, respectively) and agriculturalists. The agriculturalists
requested certain foodstuffs either because they were considered a delicacy, such as
honey, or because crop failure had brought about widespread famine.
19 A good demonstration of this happening in an unconscious way among French-
Canadians is found in Lambert et al., 1960.
9. The role of language and speech in Tukanoan identity

1 "When specific categories of people are identified with specific speech varieties,
such varieties will come to symbolize the cultural values associated with certain
features of the non-linguistic environment" (Gumperz, 1970, p. 10).
2 Blom notes that "differences in speech between various kinds of groups that are in
frequent contact are not in themselves responsible for the establishment and mainte-
nance of social boundaries. These differences rather reflect features of social
organization through a process of social codification" (1969, p. 83).
3 See G. Miller, 1963, p. 45, regarding the need for efficiency and redundance in
messages.
4 E.g., "How do we explain the fact that language conflict between competing
intergroup aspirations and interethnic stereotypes is symbolized by what to the
linguist are almost trivial linguistic differences?" (Gumperz, 1972, p. 14)
5 See also Leach (1954, p. 51), who discusses the situation in Highland Burma: "In
such conditions (gumlao), I suggest, where each petty village leader is prepared to
assert that he is as good as his neighbor, we may expect to find an obstinately
persistent linguistic factionalism even in the fact of nominally centralized political
authority'
6 For example, the use of the expression "money whitens'' or the recent declaration by
South Africa that Japanese were "honorary Caucasians" and thus could sit in the
front of the bus.
7 "Matched guise" studies conducted by W. E. Lambert and associates (1960) of
attitudes toward groups of people as "revealed" through their language and speech
are quite enlightening. Ferguson notes that Arab attitudes toward other Arabs and the
non-Islamic world can be derived from investigating their attitudes toward Arabic
itself (1959, pp. 75-82; see also Irvine, 1974).
8 Although it is also true that many systems seem to tolerate or actually to exploit such
ambiguity: See Leach's example of an individual in Highland Burma whose family
had been both Kachin and Shan for some seventy years (1954, p. 2).
251
Notes to pp. 171-89
9 For a discussion of other topics related to multilingualism research and the Vaupes
situation, see Jackson, 1974, pp. 53-59, 63-64.
10 This is also supported by my own rather skimpy data on these groups. In 21
Piratapuya marriages (8 men, 13 women) and 12 Uanano marriages (4 men, 8
women), no intermarriage occurred between Piratapuya and Uanano.
11 Mason (1950) considers Bara and Tuyuka to be the same language, as well as
Piratapuya and Uanano.
12 Yeripuna, "soul" or "life force," was described to me as the soundof the heartbeat;
when it is silent the person is dead. In a myth about Tapir, an association is made
between the loss of his voice and his ceasing to be a feared predator. Musical
instruments give power; in the story of Jose' in Chapter 4, the peccaries become weak
and without courage when the shaman Old Arturo takes their horns away. The
association between speech and power occurs in many lowland South American
societies. For example, Harner (1972, p. 139) notes that an indication that a Jivaro
has acquired an arutam soul is his changed speech, with the consequence that visits
between Jivaro men are characterized by near-shouting. If a man is lacking in
forcefulness of speech, it is a sign that his arutum soul has been taken away.
10. Male and female identity

1 See Arhem, 1981; Goldman, 1940, 1963, 1964, 1976; S. Hugh-Jones, 1979; and
Reichel-Dolmatoff, 1971.
2 See, for example, discussions by Murphy (1956); Murphy and Murphy (1974); J.
Shapiro (n.d.); and Siskind (1973a; 1973b).
3 See Ortner, 1974, for a discussion of this concept. Also see Lamphere, 1977;
MacCormack and Strathern, 1980; and Quinn, 1977, for critiques. C. Hugh-Jones
(1979) also makes this distinction, variously distinguishing between "natural" and
"social," "spiritual," or "ritual," depending on the context.
4 Men smoke fish obtained during fish poisoning expeditions and the large joints of
game. They also help brew chicha before a festival.
5 Very seldom, however, is any activity completely thought of in terms of symbols
related to only one sex. This dual sexual symbolism is especially true of the most
sacred symbols in Tukanoan culture. Concrete phenomena can represent male or
female elements, depending on context and on the relationship between them and
their surroundings.
6 "However, the economic tasks of the sexes are not balanced in such a way that they
are equal and opposite any more than men and women in Pira'-parana' are equal and
opposite" (C. Hugh-Jones, personal communication).
7 See, C. Hugh-Jones (1979) for a discussion relating this kind of "social periodicity''
brought about by the rules of exogamy to women's natural periodicity with respect to
two phases of the menstrual cycle.
8 The reciprocity is direct and does not involve certain language groups being classified
as "masculine" or "feminine" because of a greater tendency to hunt or fish (fishing
being considered more "feminine"), as Reichel-Dolmatoff states (1971, pp. 17-18).
9 See Bamberger, 1974; Bolens, 1967; S. Hugh-Jones, 1974; Reichel-Dolmatoff,
1971.
10 C. Hugh-Jones (personal communication) has heard of one or two cases of women
doing some types of shamanism for infants.
11 This example is from C. Hugh-Jones, 1979, p. 257. Domestic fire is a symbol of
women's creativity in childbearing.
12 In particular, Reichel-Dolmatoff s description of the Desana with regard to sex does
252
Notes to pp. 190-201
not fit with my impressions. Desana are characterized by "a very marked sexual
repression . . . puritan trend in which all that refers to sexual relations is viewed with
great fear and anxiety. Sex is the greatest danger in life . . . The percentage of
homosexuals, female as well as male, seems to be quite high, as well as the incidence
of sexual assaults" (1971, pp. 19-20). His statement that Tukano women use herbal
concoctions which "in varying concentrations, cause temporary sterility" (1976, p.
312), allowing women to space births, is not corroborated by my information. Sexual
abstinence is prescribed following birth, and, insofar as this is practiced, probably
accounts for most birth spacing. The existence of infanticide (see Chapter 5 and C.
Hugh-Jones, 1979, p. 128) also seems to argue against knowledge of effective
chemical contraceptives.
13 See Collier and Rosaldo (1981) on this theme; because an important aspect of their
analysis depends on uxorilocality, it does not really apply to the Vaup6s situation.
See also Gregor, 1974; Murphy and Murphy, 1974; and Siskind, 1973a, for
comparative material on the Amazon Basin, and Brown and Buchbinder, 1976;
Langness, 1967; Lindenbaum, 1972; and Meggitt, 1964, for material on the New
Guinea Highlands.
14 C. Hugh-Jones (personal communication) discusses this at length, for example:
"PM-parana culture arranges sexual symbols in a nesting system so that each unit
upon which we choose to focus has an overall sexual identity, although it contains
within it both male and female elements which may be separated out and combined
with other elements from other units."
15 Many studies have pointed out that a strict sexual segregation can result in women
having substantial power and autonomy, both because men are not around to keep
women under surveillance and because the resulting complementarity of tasks means
that men are dependent on women as well as vice versa. See, for example, Brown,
1970b; Friedl, 1975; Murphy and Murphy, 1974; Netting, 1969; Quinn, 1977; and
Rosaldo, 1974.
16 See Goldman, 1963, 1977, 1981, for discussions of incipient stratification in the
Vaup6s.
17 Still, this should not be overstated. The relationship between Maku as servant and
women as servant is not clear (see Reichel-Dolmatoff, 1971). See also S. Hugh-
Jones, 1977, p. 213, with respect to a "servant" sib (the lowest-ranked in a series of
five positions) being represented in part by women and children in the he wi (male
initiation) ritual, and the association of this rank with Maku. A special flute played
by the initiates is described as being the "wives" of the other horns. See also Briizzi
Alves da Silva, 1962, p. 307, in this regard.
18 See Chapter 2; also see Chagnon, 1968; Gross, 1975; Harris, 1975, pp. 338-41;
Ross, 1978; Siskind, 1973b; and Shapiro, 1972, for a variety of explanations.
11. Tukanoans' place in the cosmos

1 I collected no firsthand evidence of the occasional geronticide reported by C.
Hugh-Jones (1979, p. 108).
2 The relationship between the sound of breathing and the heartbeat and death - or, as
is indicated here, stages of death- is very important. In many ways, sound serves as a
symbol of life, vigor, and growth.
3 See Amorim, 1926/28, which includes a description of a woman influential enough
to have her body returned to her natal longhouse group for burial.
4 C. Hugh-Jones (1979) offers a comprehensive discussion of two quite distinct
elements of dead-spirit houses: One is connected to the physical death of the person
253
Notes to pp. 203-24
(and concerned with underground, rotten liquid, and generally unpleasant and
sinister features) and the other concerns the places where the spirit goes - the
ancestral dead-spirit houses, which might be spoken of as "under" a rapids, for
example, but do not ever have an association with yellow earth and underground.
5 Space does not permit a more thorough description of ritual paraphernalia, musical
instruments, or other objects associated with festivals. For a very comprehensive
description of He wi and He rika soria wi, the festivals associated with male
initiation, see S. Hugh-Jones, 1979.
6 Two words are used to describe these levels. Bara distinguishes between yehpd,
"ground," "earth," "flat"; and tuti, any layer in the five-tiered universe.
7 Unfortunately, some of the discussion about cosmology took place during the
summer of 1970, when the Apollo program landed a man on the moon. The
headman's second son was visiting after a long period of living in Mitu. He spoke
fairly good Spanish and listened to the radio constantly. I was asked a number of
questions about the moon flight, and this probably influenced the willingness of
people at Pumanaka buro to talk about how they conceived of the heavens.
8 No one knew anything about solar eclipses, although they had heard that such things
do occur.
9 This assertion seems contradicted by one particular type of wahti (wahti-biyo) that
rapes girls going through the menarche if they should enter the forest but, I was
explicitly told, does not devour them; however, one translation of its name is
"like-a-wahtf."
10 The UmUa.ro wahti, however, has the body of a howler monkey and the head of a
human being.
12. Tukanoans and the outside world

1 See International Work Group for Indigenous Affairs, 1976, and Jackson, 1978, n.d.-c:
the presence of uranium in the territory of Roraima, Brazil, has been known since 1954.
2 A sawmill at the Internado (Catholic boarding school) above Mitu produces milled
lumber for local use and provides a few jobs to Tukanoans.
3 For a more thorough description of the rubber-gathering industry in the Amazon
Basin, including descriptions of patron-client relationships as they occur in other
areas, see Murphy, 1960; Murphy and Murphy, 1974; and Murphy and Steward,
1956. Documentation of the atrocities committed by rubber barons and their hired
goons is found in Hardenberg, 1912, and Casement, 1913. For a discussion of the
situation in the Vaupes, see Fulop, 1953.
4 Individual members of SIL, however, make no attempt to hide the fact that they are
there to convert Indians to their form of fundamentalist Protestantism. See also Wallis
and Bennett, Two Thousand Tongues to Go, 1959, regarding the missionary policies
of SIL, and Hvalkof and Aaby, 1981.
5 Sometimes described as a black sultana - priests wear white sultanas.
6 The program of groups such as New Tribes and Unevangelized Fields (another North
American Protestant missionary group) are included with SIL for this discussion,
although obviously some important differences exist among them.
7 During my stay a number of priests and nuns discussed previous missionary efforts in
the Vaup6s with me; although they acknowledged that many measures taken by those
missionaries were too harsh, they frequently added that "it was a difficult time then;
the Indians were not as tame [manso] as they are now?'
8 The Bard term for whites is pehkd-maha, "fire people," most probably a
254
Notes to pp. 225-35
reference to firearms. Spanish words for Tukanoans are indigena, "indigene," and
the far more rarely used derogatory indio.
9 Conscious attempts by change agents to create needs in order to have more access to
target populations is an old story. See Holmberg's advocating creating an "addic-
tion" to salt among the Sirion6 for this purpose (1969, p. 85).
10 Good discussions of the kinds of international economic and political processes that
are the ultimate determinants of the future Tukanoans face can be found in Davis,
1977; Davis and Mathews, 1976; Dostal, 1972; Frank, 1967; Ribeiro, 1972; Whitten,
1976; and Wolf and Hansen, 1972.
13. Conclusions: themes in Tukanoan social identity

1 For a discussion of types of classificatory structures, see Kay, 1966.
2 Sudnow, 1967, provides an interesting discussion of the Western conceptualization of
a deceased person.
255
Glossary
This glossary contains words appearing in the text more than once with the exception of
kin terms (see Table 7) and names of social units such as sibs and language groups (see
Tables 2 and 9). Most scientific identifications of plants and animals are found in Chapter
4. Definitions of Tbkanoan social units are given in Figure 4. Italics indicate BarS and
other Eastern Ihkanoan words, and bold face indicates Spanish, Portuguese, and Tupian
loan words.
ahperd mahkara "other people."

almaceti general store.
and poisonous snakes.
ani period of weather in the rainy season consisting of misty rain and
cold temperatures.
banisteriopsis Banisteriopsis inebrians; Banisteriopsis rusbyana. A Malpighia-
ceous genus containing several hallucinogenic species. Known lo-
cally as yahi.
Bard ydara Bard sib at Pumanaka buro.
barbasco fish poison.
bayd head dancer.
Behkard Arawak-speaking language group to the north of the Vaupe's River.
boa rotten, putrid.
boho see tintfn.
borarp forest demon
biihkii old, mature.
cabuco half-breed (almost always of a Colombian or Brazilian father and a
Tukanoan mother).
cachivera rapids.
caja agraria the state-run rubber cooperative in Mitii.
capitan headman.
caraiuru a chalky red powder made from the leaves of Bigonia chica.
catequista catechist.
cauchero rubber gatherer.
cazabe manioc bread.
cerros flat-topped or domed sandstone hills.
coca Erythroxylon coca.
comisaria a Colombian political unit similar to the national territories of
Canada.
256
Glossary
comisariato government-administered general store.

comisario director of a comisaria; similar to governor.
compadrazgo co-parent relationship between the parent of a child and its
godfather.
compadre co-parent.
corregidor magistrate; among Ttikanoans, a government-appointed position in
each mission town.
dabucuri ceremony involving an exchange of meat or fish for beer.
Dahea toucan bird; the Tukano language group.
dia kat&r Muscovy duck.
DM kod pond "piece-of-meat-bone-children," mythical culture heroes.
doe See tarira.
Ewura tard a mythical miriti swamp, the source of all rivers.
eyoa wood ibis birds.
farina toasted manioc granules.
friagem see aru.
he tenti ceremonial brother-in-law (PM-paranS groups can be he teniia
to each other).
hikaniya "one-couple marriage."
hikd pond children of the same parents (full siblings).
Inambu tinamou bird (Tinamus and Crypturellus spp.).
hand Sabicea amazonensis.
Karihona Carihona, Carib speakers to the west of the Vaup6s.
Kod-mahkii "Bone son," a mythical culture hero.
kumod canoe.
kumu shaman.
Lingua Geral Tupian-based lingua franca.
llanos savanna.
magutt fire ants.
maha r people.
mahkii, mahkd son, daughter.
mahokd ahke spider monkey.
mahokd diari wahti dead-people spirit.
Maku indigenous inhabitants of the Vaupes who speak a non-Tukanoan
language and are more forest than river oriented.
maniya waderd "our-language people."
miriti Mauritiaflexuosa, a palm tree that grows in swamps and provides
an important fruit.
Namakuru a mythical character, "deer child."
nfguas chiggers, Pulex penetrans.
nimd __ curare.
Opekd dia mythical river flowing in the underworld in an east-west direction;
"milk river."
padrino baptismal godfather.
pahkd-mahkara "mother's children people"; affines of affines.
Pamiiri wi "first-people longhouse." Ancestral longhouse of the Bard under
Yurupari Rapids.
pehu co-wife; wife of husband's true brother.
pind anaconda; snake.
puaniya two-couple marriage.
257
Glossary
Pumanaka bum location of field study on the Inambu River.
pupunha peach palm; Gulielma gasipaes.
quinapira pepperpot sauce.
tarira fish, Erythrinus sp.
tatd swamp; land that floods annually.
laid boa clearings in the forest.
tenara "in-law people"; affines.
timb6 fish poison.
tintfn small rodent, Dasyprocta sp.
tipitf tubular basket for squeezing manioc mash.
umurikori maha "day people."
Virola sp. a hallucinogenic snuff, made by Maku.
wahti a category of nonhuman beings including forest demons, ghosts,
and the ancestors.
wai fish.
Wai maha fish people; most of the Bara language group.
Wamiitanard low-ranking Bara sib, (from wamiita, a fernlike plant,
Selagenella sp.).
werea beeswax.
wi longhouse; house.
wind wind; the Desana language group.
yai class of predatory animals including the jaguar; a shaman who
sorcerizes.
Yurupari sacred flutes associated with the male initiation rites.
258
References
Allen, R H. (1947) Indians of Southeastern Colombia. Geographical Review 37, pp.

568-83.
Allen, W L., and J. Holshouser de Tiz6n (1973) Land use patterns among the Campa of
the Alto Pachitea, Peru. In D. W Lathrap and J. Douglas, eds., Variation in
Anthropology: Essays in Honor of John C. McGregor, pp. 137-53. Urbana: Illinois
Archaeological Survey.
Amorim, A. B. de (1926/28) Lendas em Nheengatu e em Portuguez. Revista do Instituto
Historico Geogrdfico Brasileiro (tomo 100) 154, no. 2 (1926).
Arcand, B. (1972) The Urgent Situation of the Cuiva Indians of Colombia. International
Work Group for Indigenous Affairs, Document 7. Copenhagen.
Arhem, K. (1976) Fishing and hunting among the Makuna: Economy, ideology and
ecological adaptation in the Northwest Amazon. Goteborgs Etnografiska Museum
Arstryck, pp. 27-44. Gothenburg.
(1980) Observations on life cycle rituals among the Makuna: Birth, initiation, death.
Goteborgs Etnografiska Museum Arstryck, pp. 10-47. Gothenburg.
(1981) Makuna Social Organization: A Study in Descent, Alliance and the Formation of
Corporate Groups in the North-Western Amazon. Uppsala Studies in Cultural Anthro-
pology 4.
Arvelo-Jimenez, N. (1971) Political relations in a tribal society: A study of the Ye'cuana
Indians of Venezuela. Ph.D. diss., Cornell University.
Austin, J. L. (1962) How to Do Things with Words. Cambridge, Mass. Harvard University
Press.
Bamberger, J. (1974) The myth of matriarchy: Why men rule in primitive society. In M. Z.
Rosaldo and L. Lamphere, eds., Woman, Culture and Society, pp. 263-80. Stanford
University Press.
Bamett, S. (1977) Identity choice and caste ideology in contemporary South India. In J.
L. Dolgin, D. S. Kemnitzer, andD. M. Schneider, eds., Symbolic Anthropology: A
Reader in the Study of Symbols and Meanings, pp. 270-91. New \fork: Columbia
University Press.
Barth, E (1964) Ethnic processes on the Pathan-Baluch boundary. In G. Redard, ed.,
Indo-Iranica. Wiesbaden.
(1969) Introduction. In E Barth, ed., Ethnic Groups and Boundaries, pp. 9-38.
Boston: Little, Brown.
Basso, E. B. (1973) The use of Portuguese in Kalapalo (Xingu Carib) encounters:
Changes in a Central Brazilian communications network. Language in Society 2, pp.
1-19.
Bates, H. W (1864) The Naturalist on the River Amazons. London.
259
References
Bates, M. (1965) The Land and Wildlife of South America. Amsterdam: Time-Life
International.
Beckerman, S. (1979) The abundance of protein in Amazonia: A reply to Gross. Ameri-
can Anthropologist 81, pp. 533-60.
Berlin, B., D. E. Breedlove, and R H. Raven (1968) Covert categories and folk
taxonomies. American Anthropologist 70, pp. 290-399.
Berreman, G. D. (1972) Prologue: Behind many masks, ethnography and impression
management. In Hindus of the Himalayas, Ethnography and Change. Berkeley:
University of California Press.
Bidou, P. (1972) Representations de l'espace dans la mythologie Tatuyo (Indiens Tucano).
Journal de la Sociiti des Amiricanistes 61, pp. 45-105.
(1976) Les fils de 1'anaconda cdleste (les Tatuyo): Etude de la structure socio-
politique. Thesis for Troisieme Cycle, University of Paris.
(1977) Naitre et etre Tatuyo. Actes du XLJI' Congres International des Amiricanistes
2, pp. 105-20.
Biocca, E. (1965) Viaggi tra gli lndi Alto Rio Negro - Alto Orinoco: Appunti di un
biologo, vol. 1: Tuk&no - Taridna - Baniwa - Maku. Rome: Consiglio Nazionale
delle Ricerche.
Birdsell, J. B. (1953) Some environmental and cultural factors influencing the structuring
of Australian Aboriginal populations. American Naturalist 87, no. 834, pp. 171-207.
Bledsoe, C. (1980) The manipulation of Kpelle social fatherhood. Ethnology 19, pp.
29-46.
Blom, J. R (1969) Ethnic and cultural differentiation. In E Barth, ed., Ethnic Groups and
Boundaries, pp. 74-86. Boston: Little, Brown.
Bodley, J. (1972) Tribal Survival in the Amazon: The Campa Case. International Work
Group for Indigenous Affairs, Document 5. Copenhagen.
(1975) Victims of Progress. Menlo Park, Calif.: Cummings.
Bolens, J. (1967) Mythe de Jurupari: Introduction a une analyse. LHomme 7, no. 1, pp.
50-66.
Briggs, J. (1970) Never in Anger: Portrait of an Eskimo Family. Cambridge, Mass.:
Harvard University Press.
Bright, J. O., and W. Bright (1965) Semantic structures in Northwestern California and
the Sapir-Whorf hypothesis. American Anthropologist 67, pp. 249-58.
Brown, J. K. (1970a) Economic organization and the position of women among the
Iroquois. Ethnohistory 17, no. 3-4, pp. 151-67.
(1970b) A note on the division of labor by sex. American Anthropologist 72, pp.
1073-78.
Brown, R, and G. Buchbinder, eds. (1976) Man and Woman in the New Guinea High-
lands. Special Publication 8. Washington, D.C.: American Anthropological Associa-
tion.
Briizzi Alves da Silva, A. (1962) A civilizacao indigena do Uaupes. Sao Paulo:
Centra de Pesquisas Iauaret6.
(1966) Estrutura da tribo Tukano. Anthropos 61, pp. 191-203.
Buchler, I. R., and H. A. Selby (1968) Kinship and Social Organization: An Introduction
to Theory and Method. New York: Macmillan.
Burridge, K. (1979) Someone, No One: An Essay on Individuality. Princeton, N.J.:
Princeton University Press.
Butt, A. J. (1965-66) The shaman's legal role. Revista Museu Paulista 16, pp. 151-86.
Cardoso de Oliveira, R. (1961) Alianga interclanica na sociedade Tukuna. Revista de
Antropologia 9, no. 1, pp. 15-32.
260
References
(1964) O indio e o mundo dos brancos: A situacdo dos Tukuna do Alto Solimoes.
Corpo e Alma do Brasil, 12. Sao Paulo: Difusao Europ&a do Livro.
Cardoso de Oliveira, R., and L. de Castro Faria (1971) Interethnic contact and the study
of populations. In E Salzano, ed., The Ongoing Evolution of Latin American
Populations, pp. 41-59. Springfield, 111.: Thomas.
Carneiro, R. L. (1960) Slash-and-burn agriculture: A closer look at its implications for
settlement patterns. In A. E C. Wallace, ed., Men and Cultures: Selected Papers of
the Fifth International Congress of Anthropological and Ethnological Sciences, pp.
229-34. Philadelphia: University of Pennsylvania Press.
Casement, R. (1913) El libro rojo del Putumayo. London.
Casselberry, S., and N. Valavanes (1976) "Matrilocal" Greek peasants and a reconsidera-
tion of residence terminology. American Ethnologist 3, pp. 215-26.
Casson, R. W (1973) Paired polarity relations in the formal analysis of a Turkish kinship
terminology. Ethnology 12, pp. 275-98.
Cathcart, M. (1973) Cacua. In Aspectos de la cultura material de grupos etnicos de
Colombia. Lomalinda, Meta, Colombia: Ministerio de Gobierno, Republica de
Colombia.
Chagnon, N. A. (1968) Yqnomamd: The Fierce People. New York: Holt, Rinehart and
Winston.
Chodorow, N. (1978) The Reproduction of Mothering: Psychoanalysis and the Sociology
of Gender. Berkeley: University of California Press.
Clastres, R (1972) Chronique des Indiens Guayaki: Ce que savent les Ache, chasseurs
nomades du Paraguay. Paris: Plon.
Collier, J. (1974) Women in politics. In M. Z. Rosaldo and L. Lamphere, eds., Woman,
Culture and Society. Stanford University Press.
Collier, J. E, and M. Z. Rosaldo (1981) Politics and gender in simple societies. In S. B.
Ortner and H. Whitehead, eds., Sexual Meanings: The Cultural Construction of
Gender and Sexuality, pp. 275-329. Cambridge University Press.
Colson, A. B. (1973) Inter-tribal trade in the Guiana Highlands. Antropologica 34, pp.
5-69.
Coppens, W (1971) Las relaciones comerciales de los Yekuana del Caura-Paragua.
Antropologica 30, pp. 28-59.
Corry, S. (1976) Towards Indian Self-determination in Colombia. Survival International
Document 2. London.
Coudreau, H. A. (1887) La France equinoxiale. Vol. 2: Voyage a trovers les Guayanes et
I'Amazone. Paris.
Crocker, J. C. (1977) The mirrored self: Identity and ritual inversion among the Eastern
Bororo. Ethnology 16, pp. 129-46.
Cunningham, C. (1964) Order in the Atoni house. Bijdragen tot de Taal-land en
Volkenkunde 120, pp. 34-68.
Davis, S. H. (1977) Victims of the Miracle: Development and the Indians of Brazil.
Cambridge University Press.
Davis, S. H., and R. O. Mathews (1976) The Geological Imperative: Anthropology and
Development in the Amazon Basin of South America. Cambridge, Mass.: Anthropol-
ogy Resource Center.
De Beauvoir, S. (1953) The Second Sex. Translated by H. M. Parshley. New York: Knopf.
Denevan, W M. (1971) Campa subsistence in the Gran Pajonal, Eastern Peru. Geographi-
cal Review 61, pp. 496-518.
(1973) Development and the imminent demise of the Amazon rain forest. Professional
Geographer 25, no. 2, pp. 130-35.
261
References
ed. (1976) The Native Population of the Americas in 1492. Madison: University of
Wisconsin Press.
Dinnerstein, D. (1976) The Mermaid and the Minotaur: Sexual Arrangements and Human
Malaise. New York: Harper & Row.
Dobyns, H. F (1966) Estimating aboriginal American populations. Current Anthropology
7, no. 4, pp. 395-416.
Dole, G. (1964) Shamanism and political control among the Kuikuru. In H. Becker, ed.,
Beitrage zur Volkerkunde Sud-Amerikas: Festgabe fur Herbert Baldus, pp. 53-62.
Hannover.
(1966) Anarchy without chaos: Alternatives to political authority among the Kuikuru.
In M. J. Swartz, V W. Turner, and A. Tuden, eds., Political Anthropology, pp.
73-87. Chicago: Aldine.
Dominguez, Y (1977) Social classification in Creole Louisiana. American Ethnologist 4,
no. 4, pp. 589-603.
Dostal, W, ed. (1972) The Situation of the Indian in South America. Geneva: World
Council of Churches.
Douglas, M. (1966) Purity and Danger. Boston: Routledge & Kegan Paul.
Dufour, D. L., and J. L. Zarucchi (1979) Monopteryx angustifolia and Erisma japura:
Their use by indigenous peoples in the Northwestern Amazon. Botanical Museum
Leaflets, vol. 27, no. 3-4, pp. 69-91. Harvard University.
Elkin, A. R (1953) Murgin kinship re-examined and remarks on some generalizations.
American Anthropologist 55, pp. 412-19.
Erikson, E. (1968) Identity. In D. Sills, ed., Encyclopedia of the Social Sciences, vol. 7.
New York: Macmillan.
Ervin-Tripp, S. (1969) Sociolinguistics. In L. Berkowitz, ed., Advances in Experimental
Social Psychology, vol. 4, pp. 91-165. New York: Academic Press.
Ferguson, C. A. (1959) Myths about Arabic. Languages and Linguistic Monograph
Series 12, pp. 75-82. Georgetown University Press.
Fock, N. (1963) Waiwai: Religion and Society of an Amazon Tribe. Ethnographic Series,
vol. 8. Copenhagen: National Museum of Denmark.
Fortes, M. (1945) The Dynamics of Clanship among the Tallensi. Oxford University
Press.
Fosberg, E R. (1973) Temperate zone influence on tropical forest land use: A plea for
sanity. In B. Meggers, E. S. Ayensu, and W. D. Duckworth, eds., Tropical Forest
Ecosystems in Africa and South America: A Comparative Review. Washington, D.C.:
Smithsonian Institution Press.
Fox, R. (1967) Kinship and Marriage. London: Penguin Books.
Frank, A. G. (1967) Capitalism and Underdevelopment in Latin America. New York:
Monthly Review Press.
Fried, M., M. Harris, and R. Murphy, eds. (1968) War: The Anthropology of Armed
Conflict and Aggression. New York: Doubleday.
Fried, M. H. (1975) The Notion of Tribe. Menlo Park, Calif.: Cummings.
Friedl, E. (1975) Women and Men: An Anthropologist's View. New York: Holt, Rinehart
and Winston.
Fulop, M. (1953) El cauchero en el Vaupe's. Revista Colombiana de Folklore, 2d ser. 2,
pp. 243-55.
(1955) Notas sobre los te"rminos y el sistema de parentesco de los Tukano. Revista
Colombiana de Antropologia 4, pp. 123-64.
(1956) Aspectos de la cultura Tukana. Mitologfa. Revista Colombiana de Antropologia
5, pp. 335-373.
262
References
Galvao, E. (1959) Aculturagao indigena no Rio Negro. Boletim do Museu Paraense
Emilio Goeldi 7, pp. 1-60.
Gasche, J. (1977) Les fondements de 1'organisation sociale des Indiens Witoto et
l'illusion exogamique. Actes du XLIIe Congres International des Americanistes 2,
pp. 141-61.
Giacone, A. (1949) Os Tucanos e outras tribus do Rio Uaupes. Sao Paulo: Impresa
Oficial do Estado.
(1955) Pequena gramdtica e dicionario Portugues Ubde-Nehern ou Macu. Recife.
Gilbert, J. R, and E. A. Hammel (1966) Computer simulation and the analysis of
problems in kinship and social structure. American Anthropologist 68, pp. 71-93.
Glasse, R. M., and M. J. Meggitt (1965) Pigs, Pearlshells and Women. Englewood Cliffs,
N.J.: Prentice-Hall.
Godelier, M. (1975) Modes of production, kinship, and demographic structures. In M.
Bloch, ed., Marxist Analyses and Social Anthropology. London: Malaby.
(1977) The concept of the "tribe": A crisis involving merely a concept or the empirical
foundations of anthropology itself? In Perspectives in Marxist Anthropology. Cam-
bridge University Press.
Goffman, E. (1963) Stigma: Notes on the Management of Spoiled Identity. Englewood
Cliffs, N.J.: Prentice-Hall.
Goldman, I. (1940) Cosmological beliefs of the Cubeo Indians. Journal of American
Folklore 53, pp. 242-47.
(1948) Tribes of the Uaupes-Caquetd region. Bureau of American Ethnology Bulletin
143, no. 3, pp. 763-98.
(1963) The Cubeo: Indians of the Northwest Amazon. Champaign: University of Illinois
Press.
(1964) The structure of ritual in the Northwest Amazon. In R. A. Manners, ed.,
Process and Pattern in Culture: Essays in Honor of Julian H. Steward. Chicago:
University of Chicago Press.
(1976) Perceptions of nature and the structure of society: The question of Cubeo
descent. Dialectical Anthropology 1, pp. 287-92.
(1977) Time, space, and descent: The Cubeo example. Actes du XLIIe Congres
International des Amiricanistes 2, pp. 175-83.
(1981) Foundations of social hierarchy: A Northwest Amazon case. Paper presented at
the New York Academy of Sciences, Section of Anthropology, New York, February
23.
Goodenough, W (1965) Rethinking "status" and "role": Toward a general model of the
cultural organization of social relationships. In M. Banton, ed., The Relevance of
Models for Social Anthropology. London: Tavistock.
(1970) Description and Comparison in Cultural Anthropology. Chicago: Aldine.
(1981) Culture, language, and society. Menlo Park, Calif.: Cummings.
Goodland, R. J. A., and H. S. Irwin (1975) Amazon Jungle: Green Hell to Red Desert?
The Hague: Elsevier.
Greenberg, J. H. (1960) The general classification of Central and South American
languages. In A. E C. Wallace, ed., Selected Papers of the International Congress of
Anthropological and Ethnological Sciences. Philadelphia: University of Pennsylva-
nia Press.
Gregor, T. (1974) Publicity, privacy and Mehinacu marriage. Ethnology 13, pp. 333-49.
Gross, D. (1975) Protein capture and cultural development in the Amazon Basin. Ameri-
Gumperz, J. J. (1964a) Hindi-Punjabi code-switching in Delhi. In H. G. Lunt, ed.,
263
References
Proceedings of the Ninth International Congress of Linguists, pp. 1115-24. The

Hague: Mouton.
(1964b) Linguistic and social interaction in two communities. In J. Gumperz and D.
Hymes, eds., The ethnography of communication. Part 2. Special Publication.
American Anthropologist 66, no. 6, pp. 137-54.
(1968) The speech community. In International Encyclopedia of the Social Sciences,
vol. 9, pp. 381—86. (Reprinted in Language in Social Groups: Essays by John J.
Gumperz, pp. 114-28. Stanford University Press, 1971.)
(1969) Communication in multilingual societies. In S. Tyler, ed., Cognitive Anthropol-
ogy, pp. 435-48. New York: Holt, Rinehart and Winston.
(1970) Verbal strategies in multilingual communication (Working Paper 36).
Language-Behavior Research Laboratory, University of California, Berkeley. Mimeo.
(1972) Introduction. In J. J. Gumperz and D. Hymes, eds., Directions in Sociolinguis-
tics: The Ethnography of Communication, pp. 1-25. New York: Holt, Rinehart and
Winston.
Gumperz, J. J., and R. Wilson (1971) Convergence and creolization: A case from the
Indo-Aryan/Dravidian border. In D. Hymes, ed., Pidginization and Creolization of
Languages. Proceedings of a Conference Held at the University of the West Indies,
Mora, Jamaica, April 1968, pp. 151-68. Cambridge University Press.
Haaland, G. (1969) Economic determinants in ethnic processes. In E Barth, ed., Ethnic
Groups and Boundaries, pp. 58-73. Boston: Little, Brown.
Hallowell, A. I. (1955) The self and its behavioral environment. In Culture and Experi-
ence, pp. 75-111. Philadelphia: University of Pennsylvania Press.
Hames, R. (1978) A behavioral account of the division of labor among the Ye'kwana
Indians of Southern Venezuela. Ph.D. diss., University of California, Santa Barbara.
(1980a) Diversity within monoculture: The polyvarietal gardens of the Ye'kwana and
Yanomamo. Paper presented at 79th Annual Meeting, American Anthropological
Association, Washington, D.C., December 6.
ed. (1980b) Studies in Hunting and Fishing in the Neotropics (Working Papers on South
American Indians, no. 2). Bennington College.
Hardenburg, W S. (1912) The Putumayo: The Devil's Paradise. London.
Harner, M. J. (1972) The Jivaro: People of the Sacred Waterfalls. New York: Doubleday.
Harris, D. R. (1971) The ecology of swidden cultivation in the Upper Orinoco rainforest,
Venezuela. Geographical Review 61, no. 4, pp. 475-95.
Harris, M. (1975) Culture, People and Nature. 2d ed. New York: Crowell.
Helm, J., ed. (1968) Essays on the Problem of Tribe. Proceedings of the 1967 Annual
Spring Meeting of the American Ethnological Society. Seattle: University of Wash-
ington Press.
Helms, M. (1970) Matrilocality, social solidarity, and culture contact: Three case histo-
ries. Southwestern Journal of Anthropology 26, pp. 197-212.
Hicks, D. (1976) Tetum Ghosts and Kin: Fieldwork in an Indonesian Community. Palo
Alto, Calif.: Mayfield.
Holmberg, A. (1969) Nomads of the Long Bow: The Siriono of Eastern Bolivia. New
York: Doubleday.
Homans, G. C , and D. Schneider (1955) Marriage, Authority and Final Causes: A Study
of Unilateral Cross-cousin Marriage. Glencoe, 111.: Free Press.
Hugh-Jones, C. (1977) Skin and soul: The round and the straight. Social time and social
space in Pira-paranS society. Actes du XLIIe Congris International des Amiricanistes
2, pp. 185-204.
(1978) Food for thought - Patterns of production and consumption in Pira-parana
society. In J. LaFontaine, ed., Sex and Age as Principles of Social Differentiation,
264
References
pp. 41—66. London: Academic Press.

(1979) From the Milk River: Spatial and Temporal Processes in Northwest Amazonia.
(n.d.) Untitled manuscript on Pira-parana social structure.
Hugh-Jones, S. (1974) Male initiation and cosmology among the Barasana Indians of the
Vaupe*s area of Colombia. Ph.D. diss., Cambridge University.
(1977) Like the leaves on the forest floor: Ritual and social structure amongst the
Barasana. Actes du XLIIe Congres International des Americanistes 2, pp. 205-15.
(1979) The Palm and the Pleiades: Initiation and Cosmology in Northwest Amazonia.
(1980) Stars and seasons in Barasana cosmology. Paper presented at the 79th Annual
Meeting, American Anthropological Association, Washington, D.C., December 5.
(1982) The Pleiades and Scorpius in Barasana cosmology. In A. E Aveni and G. Urton,
eds.,Ethnoastronomy and Archaeoastronomy in the American Tropics, pp. 183-201.
New York Academy of Sciences.
Humboldt, Alexandre von (1822) Voyage aux regions equinoxiales du Nouveau Conti-
nent. Paris.
Huntingford, G. W B. (1955) The economic life of the Dorobo. Anthropos 50, pp.
602-34.
Hvalkof, S., and R Aaby, eds. (1981) Is God an American? An Anthropological Perspec-
tive on the Missionary Work of the Summer Institute of Linguistics. International
Work Group for Indigenous Affairs, Copenhagen, and Survival International, Lon-
don.
Hymes, D. (1968) Linguistic problems in defining the concept of "tribe." In J. Helm, ed.,
Essays on the Problem of Tribe, pp. 23-48. Seattle: University of Washington Press.
Institute Geogr£fico "Agustm Codazzi" (1969) Atlas de Colombia. Bogota.
International Work Group for Indigenous Affairs (1976) Newsletter (December). Copen-
hagen.
Irvine, J. T. (1974) Strategies of status manipulation in the Wolof greeting. In R. Bauman
and J. Sherzer, eds., Explorations in the Ethnography of Speaking, pp. 167-91.
Jackson, J. E. (1972) Marriage and linguistic identity among the Bara Indians of the
Vaupds, Colombia. Ph.D. diss., Stanford University.
(1974) Language identity of the Colombian Vaupe's Indians. In R. Bauman and J.
Sherzer, eds., Explorations in the Ethnography of Speaking, pp. 50-64. Cambridge
University Press.
(1976) Vaupes marriage: A network system in the Northwest Amazon. In C. A. Smith,
ed., Regional Analysis. Vol. 2: Social Systems, pp. 65-93. New York: Academic
Press.
(1977) Barii zero generation terminology and marriage. Ethnology 16, no. 1, pp.
83-104.
(1978) Instant underdevelopment for Colombia's Indians. Anthropology Resource Cen-
ter Newsletter (December), p. 4. Cambridge, Mass.
(n.d.-a) On trying to be an Amazon. In T. Whitehead, M. E. Conaway, and M. Clark,
eds., Sex and Gender in Fieldwork: Exploring Problems and Prospects for Cross-
cultural Research and Communication. Manuscript.
(n.d.-b) Vaup6s marriage practices. In K. Kensinger, ed., Marriage Practices in
Lowland South America. Champaign: University of Illinois Press. In press.
(n.d.-c) The impact of the state on small-scale societies. Studies in Comparative
International Development. In press.
Jackson, J. E., and A. K. Romney (n.d.) A note on Bara exogamy. Manuscript.
265
References
Jacopin, P (1972) Habitat et territoire Yukuna. Journal de la Societe des Americanistes

61, pp. 107-139.
(1975) Mission Pierre-Yves Jacopin (1969-71). In J. Gabus, ed., Amazonie Nord-
ouest. Musee d'Ethnographie de NeucMtel.
(1981) La parole generative: de la mythologie des Indiens Yukuna. Ph.D. diss.,
University of NeucMtel.
Johnson, A. (1974) Carrying-capacity in Amazonia: Problems in theory and method.
Paper presented at 73d Annual Meeting, American Anthropological Association,
Mexico, November 24.
Kaplan, B. (1961) Personality Study and Culture. In B. Kaplan, ed., Studying Personality
Cross-culturally, pp. 301-11. Evanston, 111.: Row, Peterson.
Kay, E (1966) Comment on "Ethnographic semantics: A preliminary survey," by B. N.
Colby. Current Anthropology 7, no. 1, pp. 20-23.
Kelly, R. (1976) Witchcraft and sexual relations: An exploration in the social and semantic
implications of the structure of belief. In E Brown and G. Buchbinder, eds., Man and
Woman in the New Guinea Highlands. Washington, D. C.: American Anthropological
Association.
Kloos, E (1977) The Akuriyo of Surinam: A case of emergence from isolation. Interna-
tional Work Group for Indigenous Affairs, Document 27, Copenhagen.
Knutsson, K. E. (1969) Dichotomization and integration. In E Barth, ed., Ethnic Groups
and Boundaries, pp. 86-100. Boston: Little, Brown.
Koch-Griinberg, T. (1906) Die Maku. Anthropos 1, pp. 877-906.
(1909-10) Zwei Jahre unter den Indianern; Reisen in Nordwest-Brasilien. 2 vols.
Berlin: Ernst Wasmuth.
Kok, E E (1921-22) Ensayo de gramatica Dagseye o Tokano. Anthropos 16-17, pp.
838-65.
(1925-26) Quelques notices ethnographiques sur les Indiens du Rio Papuri. Anthropos
20-21, pp. 624-37, 921-37.
Kracke, W (1978) Force and Persuasion: Leadership in an Amazonian Society. Chicago:
University of Chicago Press.
Kumu, U. E, and T. Kenhiri (1980) Antes o Mundo Nao Existia. Sao Paulo: Livraria
Cultura Editora.
Labov, W (1966) Hypercorrection by the lower middle class as a factor in linguistic
change. In W. Bright, ed., Sociolinguistics, pp. 84-113. The Hague: Mouton.
Lambert, W E. (1967) A psychology of bilingualism. In J. Macnamara, ed., Problems of
Bilingualism. Journal of Social Issues 23, no. 2, pp. 91-109.
Lambert, W E . , R . C. Hodgson, R. C. Gardner, and S. Fillenbaum (1960) Evaluational
reactions to spoken languages. Journal of Abnormal and Social Psychology 60, pp.
144-51.
Lamphere, L. (1977) Anthropology. Signs: Journal of Women in Culture and Society 2,
pp. 612-27.
Langdon, T. (1975) Food restrictions in the medical system of the Barasana and Taiwano
Indians of the Colombian Northwest Amazon. Ph.D. diss., Tulane University.
Langness, L. L. (1967) Sexual antagonism in the New Guinea Highlands: A Bena Bena
example. Oceania 37, pp. 161-77.
Lathrap, D. W (1968) The "hunting" economies of the tropical forest zone of South
America: An attempt at historical perspective. In R. Lee and I. DeVore, eds., Man
the Hunter, pp. 23-29. Chicago: Aldine.
(1970) The Upper Amazon. New 'York: Praeger.
(1973) The antiquity and importance of long distance trade relationships in the moist
tropics of pre-Columbian South America. World Archaeology 5, pp. 170-86.
266
References
Lave, J. C. (1966) A formal analysis of preferential rârriage with the sister's daughter.
Man 1, pp. 185-200.
Leach, E. R. (1954) Political Systems of Highland Burma: A Study of Kachin Social
Structure. London School of Economics.
(1958) Concerning Trobriand clans and the kinship category "tabu." In J. Goody, ed.,
The Developmental Cycle in Domestic Groups. Cambridge University Press.
(1970) Uvi-Strauss. London: Fontana/Collins.
Leacock, E. B. (1955) Matrilocality in a simple hunting economy (Montagnais-Naskapi).
Southwestern Journal of Anthropology 11, pp. 31-47.
Lee, R. B. (1968) What hunters do for a living, or how to make out on scarce resources. In
R. Lee and I. DeVore, eds., Man the Hunter, pp. 30-48. Chicago: Aldine.
Lee, R. B., and I. DeVore, eds. (1968) Man the Hunter. Chicago: Aldine.
Leeds, A. (1964) Some problems of Yaruro ethnohistory. Proceedings of the 35th Interna-
tional Congress of Americanists 2, pp. 157-75.
Lehman, E K. (1974) Foreword. In E A. Ballonoff, ed., Mathematical Models of Social
and Cognitive Structures: Contributions to the Mathematical Development of Anthro-
pology, pp. vii-xviii. Champaign: University of Illinois Press.
Levi-Strauss, C. (1962) Social structure. In S. lax, ed., Anthropology, pp. 321-50.
Chicago: University of Chicago Press.
(1969) The Elementary Structures of Kinship. 2d ed. London: Eyre and Spottiswoode.
Lindenbaum, S. (1972) Sorcerers, ghosts, and polluting women: An analysis of religious
belief and population control. Ethnology 11, pp. 241-53.
Lounsbury, E G. (1956) A semantic analysis of the Pawnee kinship usage. Language 32,
pp. 158-94.
(1962) Review of Structure and Sentiment, by R. Needham. American Anthropologist
64, pp. 1302-10.
(1965) Another view of Trobriand kinship categories. In E. A. Hammel, ed., Formal
Semantic Analysis. Special Publication. American Anthropologist 67, no. 5, pt. 2,
pp. 142-185.
MacCormack, C , andM. Strathera, eds. (19S0) Nature, Culture and Gender. Cambridge
University Press.
MacCreagh, G. (1926) White Waters and Black. New York: Century.
Macfarlane, Alan (1978) The Origins of English Individualism: The Family, Property, and
Social Transition. Cambridge University Press.
McGovern, M. W. (1927) Jungle Paths andlnca Ruins. London: Hutchinsqn.
Markham, C. R. (1910) A list of the tribes of the valley of the Amazons. Journal of the
Anthropological Institute 40, pp. 73-140.
Martin, M. K. (1969) South American foragers: A case study in cultural devolution.
American Anthropologist 71, pp. 243—60.
Mason, J. A. (1950) The languages of. South American Indians. Handbook of South
American Indians, vol. 6, pp. 257-58. Washington, D.C.: Bureau of American
Ethnology.
Maybury-Lewis, D. H. E (1965) The Savage and the Innocent. London: Evans Brothers.
(1967) Akwe Shavante Society. New York: Oxford University Press,
ed. (1979) Dialectical Societies: The Ge and Bororo of Central Brazil. Cambridge,
Mass.: Harvard University Press.
Meggers, B. J. (1973) Some problems of cultural adaptation in Amazonia, with emphasis
on the pre-European period. In B. Meggers, E. S. Ayensu, and W D. Duckworth,
eds., Tropical Forest Ecosystems in Africa and South America: A Comparative
Review. Washington, D.C.: Smithsonian Institution Press.
Meggitt, M. J. (1964) Male-female relationships in the Highlands of Australian New
267
References
Guinea. In J. B. Watson, ed., New Guinea: The Central Highlands. Special Publica-
tion. American Anthropologist 66, no. 4, pp. 204-24.
Metraux, A. (1948) The hunting and gathering tribes of the Rio Negro Basin. Handbook
of South American Indians, vol. 3, pp. 861-67. Washington, D.C.: Bureau of
American Ethnology.
Michaelson, E. J., and W. Goldschmidt (1971) Female roles and male dominance among
peasants. Southwestern Journal of Anthropology 27, pp. 330-52.
Miller, D. R. (1961) Personality and social interaction. In B. Kaplan, ed., Studying
Personality Cross-culturally, pp. 271-300. Evanston, 111.: Row, Peterson.
Miller, G. (1963) Language and Communication. 2d ed. New York: McGraw-Hill.
Miller, W R. (1970) Western Shoshoni dialects. In E. H. Swanson, ed., Languages and
Cultures of Western North America: Essays in Honor ofSven S. Liljeblad, pp. 17-36.
Pocatello: Idaho State University Press.
Misiones del Vaupe's (1966) 1914-1964: Los misioneros del Vaup6s a sus amigos y
colaboradores. Medellfn: Misiones del Vaupe's.
Moerman, M. (1965) Who are the Lue? Ethnic identification in a complex civilization.
Morey, R. V, and N. C. Morey (1974) The early trade system of the Orinoco llanos. Paper
presented at the 41st International Congress of Americanists, Mexico City, Sept. 3.
(1975) Relaciones comerciales en el pasado en los llanos de Colombia y Venezuela.
Universidad Catolica 'Andres Bello," Institututo de Investigaciones Hist6ricas,
Caracas.
Moser, B., and D. Tayler (1963) Tribes of the Piraparana. Geographical Journal 129, no.
4, pp. 437-49.
(1965) The Cocaine Eaters. London.
Murphy, R. E (1956) Matrilocality and patrilineality in Mundurucu society. American
Anthropologist 56, pp. 414-34.
(1960) Headhunter's Heritage: Social and Economic Change among the Mundurucu
Indians. Berkeley: University of California Press.
(1961) Deviance and social control I: What makes Ward run? Kroeber Anthropological
Society Papers 24, pp. 55-61.
Murphy, R., and J. Steward (1956) Tappers and trappers: Parallel process in acculturation.
Economic Development and Cultural Change 4, pp. 335-53.
Murphy, Y, and R. Murphy (1974) Women of the Forest. New York: Columbia University
Press.
Needham, R. (1962) Structure and Sentiment: A Test Case in Social Anthropology.
Chicago: University of Chicago Press.
(1966) Age, category and descent. Bijdragen tot de Taal-land en Volkenkunde 122, pp.
1-35.
(1971) Introduction. In R. Needham, ed., Rethinking Kinship and Marriage, pp.
xiii-cxvii. London: Tavistock.
Netting, R. M. (1969) Women's weapons: The politics of domesticity among the Kofyar.
Nimuendajo, C. (1950) Reconhecimiento dos Rios Igana, Ayarie, Uaupes. Journal de
la Sociite des Americanistes 39, pp. 125-82.
Oberem, U. (1974) Trade and trade goods in the Ecuadorian Montana. In P. Lyon, ed.,
Native South Americans: Ethnology of the Least Known Continent, pp. 346-57.
Boston: Little, Brown.
Oberg, K. (1955) Types of social structure among the lowland tribes of South and Central
America. American Anthropologist 57, pp. 472-87.
268
References
O'Laughlin, B. (1974) Mediation of contradiction: Why Mbum women do not eat

chicken. In M. Rosaldo and L. Lamphere, eds., Woman, Culture and Society, pp.
301-20. Stanford University Press.
(1975) Marxist approaches in anthropology. In B. Siegel, ed., Annual Review of
Anthropology, pp. 341-70. Palo Alto, Calif.: Annual Reviews.
Ortner, S. B. (1974) Is female to male as nature is to culture? In M. Rosaldo and L.
Lamphere, eds., Woman, Culture and Society, pp. 67-88. Stanford University Press.
Owen, Roger (1965) Patrilocal band: A linguistic and cultural heterogeneous unit. Ameri-
Powdermaker, H. (1966) Stranger and friend: The Way of an Anthropologist. New York:
Norton.
Quinn, N. (1977) Anthropological studies on women's status. In B. Siegel, ed., Annual
Review of Anthropology, pp. 181-226. Palo Alto, Calif.: Annual Reviews.
Rabinow, R (1977) Reflections on Fieldwork in Morocco. Berkeley: University of Califor-
nia Press.
Radcliffe-Brown, A. R. (1953) Dravidian kinship terminology. Man 53, p. 112.
Reichel-Dolmatoff, G. (1967) A brief field report on urgent ethnological research in the
Vaupe's area, Colombia, South America. Bulletin of the International Committee on
Urgent Anthropological and Ethnological Research, no. 9, pp. 53-62.
(1971) Amazonian Cosmos: The Sexual and Religious Symbolism of the Tukano Indi-
ans. Chicago: University of Chicago Press.
(1975) The Shaman and the Jaguar: A Study of Narcotic Drugs among the Indians of
Colombia. Philadelphia; Temple University Press.
(1976) Cosmology as ecological analysis: The view from the rainforest. Man 11, pp.
307-18.
Reid, H. (1976) Health and nutrition in the Eastern Vaupe's. In S. Corry, ed., Towards
Indian Self-determination in Colombia. Survival International Document 2, pp.
20-21. London.
(1977) Comparative discussion of Maku and Tukanoan social structure. Cambridge
University.
(1979) Some aspects of movement, growth, and change among the Hupdu Maku
Indians of Brazil. Ph.D. diss. Cambridge University.
Ribeiro, D. (1972) The Americas and Civilization. New York: Dutton.
Rice, H. (1910) The River Vaupe's. Geographical Journal 25, pt. 5, pp. 682-700.
(1914) Further explorations in the North-West Amazon Basin, Geographical Journal
44, pp. 137-68.
Rivet, E, and C. Tastevin (1920) Affinites du Maku et du Puinave. Journal de la Societe
des Amiricanistes 12, pp. 69-82.
Riviere, P. G. (1971) The political structure of the Trio Indians as manifested in a system
of ceremonial dialogue. In T. O. Beidelman, ed., The Translation of Culture: Essays
to E. E. Evans-Pritchard, pp. 293-311. London: Tavistock.
Rodriguez Lamus, L. R. (1958) La arquitectura de los Tukano. Revista Colombiana de
Antropologia 7, pp. 251-70.
Rogers, D. J. (1963) Studies of Manihot esculenta Crantz and related species. Bulletin of
the Torrey Botanical Club 90, pp. 43-54.
Romney, A. K. (1965) Kalmuk Mongol and the classification of lineal kinship terminolo-
gies. In E. Hammel, ed., Formal semantic analysis. Special Publication. American
Anthropologist 67, no. 5, p. 2, pp. 127—41.
(1971) Measuring endogamy. In E Kay, ed., Explorations in Mathematical Anthropolo-
gy, pp. 191-213. Cambridge. Mass.: MIT Press.
269
References
Romney, A. K., and R. G. D'Andrade (1964) Cognitive aspects of English kin terms.
American Anthropologist 66, no. 3, pt. 2, pp. 146-70.
Rosaldo, M. (1974) Woman, culture and society: A theoretical overview. In M. Rosaldo
and L. Lamphere, eds., Woman, Culture and Society, pp. 17-42, Stanford University
Press.
Ross, E. B. (1978) Food taboos, diet and hunting strategy: The adaptation to animals in
Amazon cultural ecology. Current Anthropology 19, pp. 1-36.
Ruddle, K. (1974) The Yukpa Cultivation System. Berkeley: University of California
Press.
Salser, J. K. (1971) Cubeo phonemics. Linguistics 75, pp. 74-79.
Sankoff, G. (1968) Social aspects of multilingualism in New Guinea. Ph.D. diss., McGill
University.
(1970) Mutual intelligibility, bilingualism, and linguistic boundaries. In International
Days of Sociolinguistics, pp. 839-48, Rome: Institute Luigi Sturzo.
(1974) A quantitative paradigm for the study of communicative competence. In R.
Bauman and J. Sherzer, eds., Explorations in the Ethnography of Speaking, pp.
18-49. Cambridge University Press.
Sapir, E. (1949) The nature of language. In D. G. Mandelbaum, ed., Selected Writings of
Edward Sapir in Language, Culture, and Personality. Berkeley: University of Cali-
fornia Press.
Sapir, J. D. (1975) Big and thin: Two Diola metalinguistic terms. Language in Society 4,
pp. 1-16.
Scheffler, H. (1966) Ancestor worship in anthropology: Or observations on descent and
descent groups. Current Anthropology 7, pp. 541-51.
(1973) Kinship, descent and alliance. In J. J. Honigman. ed., Handbook of Social and
Cultural Anthropology, pp. 747-93. Chicago: Rand McNally.
Schneider, D. M. (1965) Some muddles in the models: Or, how the system really works.
In M. Ban ton. ed., The Relevance of Models for Social Anthropology, pp. 25-86.
London: Tavistock.
(1972) American Kinship: A Cultural Account. 2d ed. Englewood Cliffs, N.J.:
Prentice-Hall.
Schneider, J. (1971) Of vigilance and virgins: Honor, shame and access to resources in
Mediterranean society. Ethnology 10, pp. 1-24.
Schultes, R. E. (1972) An overview of hallucinogens in the Western Hemisphere. In R T.
Furst, ed., Flesh of the Gods: The Ritual Use ofHallucinogens, pp. 3-54. New York:
Praeger.
Schultz, H. (1959) Ligeiras notas sobre os Makfl do Parana-Boa-Boa. Revista do Museu
Paulista, 11, pp. 110-32.
Schwerin, K. (1971) The bitter and the sweet: Some implications of techniques for
preparing manioc. Paper presented at the 70th Annual Meeting, American Anthropo-
logical Association, New York, November 30.
Seligmann, C. G., and B. Z. Seligmann (1911)The Veddas. Cambridge Univeristy Press.
Service, E. R. (1968) War and our contemporary ancestors. In M. Fried, M. Harris, and
R. Murphy, eds., War: The Anthropology of Armed Conflict and Aggression, pp.
160-67. New York: Doubleday.
Shapiro, J. R. (1972) Sex roles and social structure among the Yanomama Indians of
Northern Brazil. Ph.D. diss., Columbia University,
(n.d.) Marriage rules, marriage exchange, and the definition of marriage in lowland
South American societies. In K. Kensinger, ed., Marriage Practices in Lowland
South America. Champaign: University of Illinois Press. In press.
SIL. See Summer Institute of Linguistics.
270
References
Silverwood-Cope, E (1972) A contribution to the ethnography of the Colombian Maku.

Ph.D. diss., Cambridge University.
Singer, M. (1980) Signs of the self: An exploration in semiotic anthropology. American
Anthropologist 82, no. 3, pp. 485-507.
Siskind, J. (1973a) To Hunt in the Morning. New York: Oxford University Press.
(1973b) Tropical forest hunters and the economy of sex. In D. Gross, ed., Peoples and
Cultures of Native South America, pp. 226-41. New York: Doubleday.
Smith, R., and C. Smith (1971) Southern Barasano phonemics. Linguistics, no. 75, pp.
80-85.
Sorensen, A. E, Jr. (1967) Multilingualism in the Northwest Amazon. American Anthro-
pologist 69, pp. 670-84.
(1970) The morphology of Tukano. Ph.D. diss., Columbia University.
Soto Holgufn, A. (1972) Mitos de los Cubeo. Acts of the 39th International Congress of
Americanists 6, pp. 59-65.
Spruce, R. (1908) Notes of a Botanist on the Amazon and Andes. 2 vols. London.
Sternberg, H. O. (1969) Man and environmental change in South America. In E. J.
Fittkau, ed., Biogeography and Ecology of South America, pp. 413-41. The Hague:
Junk.
Stall, D. (1981) The adventures of Sophia Muller. ARC Bulletin 9, pp. 5-7. Boston:
Anthropology Resource Center.
Stolte, J., and N. Stolte (1971) A description of Northern Barasano phonology. Linguistics
75, pp. 86-92.
Stradelli, E. (1890) Leggenda dell'Jurupary. Bolletino della Societd Geografica Italiana,
3d. ser. 3, pp. 659-89, 798-835.
Sudnow, D. (1967) Passing On: The Social Organization of Dying. Englewood Cliffs,
N.J.: Prentice-Hall.
Summer Institute of Linguistics (SIL) (1975) Aspectos de la cultura material de grupos
etnicos de Colombia. Bogota': Summer Institute of Linguistics.
Tastevin, E C. (1923) Les Makfi du Japura. Journal de la Sociiti des Americanistes 15,
pp. 99-108.
Terriblini, M., and M. Terriblini (1961) Enquete chez des Indiens Maku (Br6sil). Bulletin
de la Society Suisse des Americanistes 21, pp. 2-10.
Thomas, D. J. (1972) The indigenous trade system of Southeast Estado Bolfvar, Venezue-
la. Antropoldgica 33, pp. 3-37.
Tindale, N. B. (1953) Tribal and inter-tribal marriage among the Australian aborigines.
Human Biology 25, pp. 169-90.
Torres Laborde, A. (1969) Mito y cultura entre los Barasana: Un grupo indlgena Tukano
del Vaupes. Bogota.
Turnbull, C. (1965) The Mbuti Pygmies of the Congo. In J. Gibbs, ed., Peoples of Africa,
pp. 281-317. New York: Holt, Rinehart and Winston.
Turner, T. (1979) Kinship, household, and community structure among the Kayap6. In D.
H. E Maybury-Lewis, ed., Dialectical Societies: The Ge and Bororo of Central
Brazil, pp. 179-217. Cambridge, Mass.: Harvard University Press.
Tyler, S. A. (1978) The Said and the Unsaid: Mind, Meaning, and Culture. New York:
Academic Press.
Van Ernst, E (1966a) Vrijwillige onvrijheid: De Maku in West-Amazonas. Bijdragen Tot
der Taal-land en Volkenkunde 122, pp. 110-28.
(1966b) Indians and missionaries on the River Tiquie, Brazil-Colombia. Institute
Archives of Ethnography 50, pt. 2, pp. 145-97.
Wallace, A. E C. (1961) On being just complicated enough. Proceedings of the National
Academy of Sciences 47, pp. 458-64.
271
References
Wallace, A. R. (1889/1972) A Narrative of Travels on the Amazon and Rio Negro. 2d. ed.
(1st ed., 1853) London. (Reprint. New York: Dover ed.)
Wallis, E. E., and Bennett, M. A. (1959) Two Thousand Tongues to Go. New York: Harper
&Row.
Waltz, N., and C. Waltz (1972) Fonologfa del Guanano. In B. Elson, ed., Sistemas
fonologicosde idiomasColombianos. Vol. 1, pp. 29-40, Lomalinda, Meta, Colombia.
Waltz, N. E., and A. Wheeler (1970) Proto Tucanoan. In E. Matteson, ed., Comparative
Studies in Amerindian Languages, pp. 120-49. The Hague: Mouton.
Ward, B. (1965) Varieties of the conscious model: The fishermen of South China. In M.
Banton, ed., The Relevance ofModels for Social Anthropology, pp. 113-37. London:
Tavistock.
Warner, W L. (1937) A Black Civilization: A Social Study of an Australian Tribe. New
•fork: Harper & Brothers.
West, B., and B. Welch (1967) Phonemic system of Tucano. In V Waterhouse, ed.,
Phonemic systems of Colombian Languages. Norman, Okla.: Summer Institute of
Linguistics Publications in Linguistics and Related Fields.
(1972) Sistema fonol6gico del Tucano. In B. Elson, ed., Sistemas fonologicos de
idiomas Colombianos. Vol. 1, pp. 13-28. Lomalinda, Meta, Colombia.
Whiffen, T. (1915) The Northwest Amazons: Notes on Some Months Spent among
Cannibal Tribes. London: Constable.
Whitten, N. E., Jr. (1976) Sacha Runa: Ethnicity and Adaptation of Ecuadorian Jungle
Quichua. Champaign: University of Illinois Press.
(1978) Ecological imagery and cultural adaptability: The Canelos Quichua of Eastern
Ecuador. American Anthropologist 80, pp. 836-59.
Wittgenstein, Ludwig (1922) Tractatus Logico-philosophicus. London: Routledge &
Kegan Paul.
Wolf, E., and E. Hansen (1972) The Human Condition in Latin America. New York:
Oxford University Press.
Wolf, M. (1974) Chinese women: Old skills in a new context. In M. Rosaldo and L.
Lamphere, eds., Woman, Culture and Society. Stanford University Press.
Woodburn, J. (1968) An introduction to Hadza ecology. In R. Lee and I. DeVore, eds.,
Man the Hunter, pp. 49-55. Chicago: Aldine.
Wright, R. (1981) The history and religion of the Baniwa peoples of the Upper Rio Negro
valley. Ph.D. diss., Stanford University.
Wright, R., and S. Swenson, eds. (1981) ARC Bulletin 9. Boston, Mass.: Anthropology
Resource Center.
"Yengoyan, A. (1968) Demographic and ecological influences on aboriginal Australian
marriage sections. In R. Lee and I. DeVore, eds., Man the Hunter, pp. 185-99.
Chicago: Aldine.
Zuidema, R. T. (1969) Hierarchy in symmetric alliance systems. Bijdragen tot de
Taal-land en "Volkenkunde 125, pp. 134-39.
272
Index
acculturation, 24—5; see also change, decul- toward work, 40-1

turation aunts, kinship role of, 119-20
and cultivated foods, 51 Austin, J. L., 177
impact on women's work of, 184-5 Australian Aborigines, 147
and language group exogamy, 136 authority
and language groups, 69-70 of headmen, 65, 67, 212
and patterns of exchange, 61-2 in parent-offspring relationship, 118-19
and regional variation, 238
and territorial dispersion, 93, 95 balance, in social identity, 231, 232
and traditional social structure, 71 banisteriopsis, 58, 153, 197
adultery, 129, 130, 189, 190, 200 at festivals, 203
affines ritual drinking of, 204, 208
in kinship system, 121 symbolism of, 190
and marital patterns, 186 Bantu Negroes (African society), 161
in mythology, 143 Bara (fish people), 229
age kinship terminology of, 109-11
at marriage, 128 language group of, 83
of spouses, 193 Barnett, S., 234
agricultural cooperatives, government-run, Barth, E, 151, 162, 165
25, 216 Basso, E. B., 170
airstrips, SIL constructed, 219-20 Bates, H. W, 22, 23
alcoholism, 223 Bates, M., 16
Allen, E H., 154 beeswax, 191, 208
alliance, marital strategy in, 137-8 behavior
Amazon, Central Northwest: kinship expectations and, 108
ecological setting of, 13-17 kinship-related, 116-17
population of, 17—19 marriage, 138-47
see also Vaupes region see also daily life
Arawakan languages, 19 belief systems; see cosmology, Tukanoan,
Arawakans, 21 mythology
Arhem, K., 6, 7, 11, 24, 105, 236 Bennett, M. A., 20
artifacts, 36 Berlin, B., 28
and gender differences, 184 beverages
in settlement interaction, 99 mingao, 39
see also craftsmanship manicuera, 53
aru, 16 manioc beer, 52, 53-54
Arvelo-Jimenez, N., 66 offered to visitors, 33
attitudes see also banisteriopsis
toward money, 212 Bidou, E, 7, 11, 24, 102, 107, 177
toward sexuality, 189 bilingual education, 220
toward sick, 37, 40 Biocca, E., 24
273
Downloaded from https://www.cambridge.org/core. University of Cambridge, on 03 Feb 2020 at 08:10:21, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511621901
Index
Bledsoe, C , 103 results of, 241

Blom, J. P, 165 see also acculturation, deculturation
blowguns, 46, 154 change agents, 212
Bodley, J., 211 and deculturation, 213-14
border, Brazilian-Colombian, 17 powers of, 214
Bororo (South American society), 4, 233 chicha, 197, 203
boundary maintenance, 232 childbirth, role of shaman in, 197-8
Breedlove, D. E., 28 children
bride grandchildren, 120-1
adaptation of, 142 in longhouse, 36
role of, 134 raising, 117-19
bride capture, ceremonial, 125 "cigar lighter," 229
bride price, 127-8 classification system, Tukanoan
Bright, J. O., 170 comparisons in, 228
Bright, W, 170 continuum in, 227-8
"brother-in-law" relationship, ceremonial, longhouse in, 230
65 Maku in, 229
brothers, in kinship system, 120 multiple meaning in, 228-9
brothers-in-law, 122 "other world" in, 236
Brown, J. K., 194 polarized contracts in, 227
Briizzi Alves da Silva, A., 5, 24, 84, 99, 157 shaman in, 230
Buchler, I. R., 107, 127 wahtia in, 229
Burridge, K., 3 climate of Vaupe's region, 16
Butt, A. J., 66 coca, 197, 203
collectivity, in Tukanoan public ritual, 234
cachiveras, 16; see also rapids Collier, J., 130, 186
Caja Agraria, 25, 216 Colombia
cannibalism, 152 as client-state of U.S., 226
Carib languages, 19 interaction with Tukanoans, 217-18
cataracts, 16; see also rivers colonialism, results of, 211
category systems, languages as, 166; see also Comisarfa del Vaupes, 23
classification system, Tukanoan comisariato, 25
Cathcart, M., 24 comisario, 25
Catholic church, and Colombian government, commercialism, effects of increasing, 61;
217-18 see also world, outside
Catholic missionaries communication
influence of, 224-5 regional perspective on, 6
and marriage, 129 while hunting, 46
and Tukanoan language, 175 see also language, language groups
see also Javerians, missionaries compadrazgo, 123
caucheros compadre (coparenthood) system, 61
in debt-bondage system, 217 comparisons, taxonomic vs. paradigmatic,
Tukanoan view of, 216 228
cazabe, 33, 35; see also manioc process competition, as mythological theme, 145
census, Tukanoan, 1968-70, 69 concordato, 217, 219
ceremonies conjunctivitis, 19
food exchange, 183-4 cooking
greeting rituals at, 98 and sexual division of labor, 53
intersettlement relations and, 96 taboo associated with, 55
naming, 105 coresidence, 70
see also festivals, ritual com, as crop, 52
cerros, 13 corregidor (magistrate), government-appointed',
Chagnon, N. A., 9, 64, 65, 97, 99, 154, 63, 137
193, 194 Corry, S., 211, 217, 222
change cosmology, Tukanoan
concept of, 11 heavens in, 205-6
costs of, 225-6 important features of landscape in, 206-7
274
Index
levels of universe in, 204-5 dialect
Makii in, 158 vs. language, 82
property in, 64 as linguistic variety, 19
rivers in, 46 diet, in horticulturally based societies,
shamanism in, 195-202 182-3
swamps in, 49 discrimination against Tukanoans, 223-4;
symbolism of differentiation in, 89, 90 see also world, outside
wahti'a in, 207 disease
cosmopolitanism, 21 diagnosis of. 198-9
Coudreau, H. A., 22, 23 epidemic, 18
courtship, patterns of, 139 impact on Tukanoan society of, 211
cousins see also curing
cross, 121 divorce, 129, 241
marriage of, 131 Dobyns, H. F, 50
matrilateral parallel, 121 Dole, G., 66
matrilateral vs. patrilateral parallel, 107 Dominguez. V, 237
couvade, 188 Douglas, M., 49
co-wife, 122 dreams, and shamanism, 196
craftsmanship drinking, at festivals, 202; see also beverages
effects of increasing commercialism on, 61 Dumont, L., 107
and sex roles, 184
Crocker, J. S., 4, 231, 233 eagle people, 89
crops, variety of, 52 economy, Tukanoan, 61
Cubeo (language group) characteristics of, 58
exogamous phratries of, 96 inelasticity of, 59
festivals of, 203 and introduction of money, 61
regional integration of, 97 see also exchange
Cuiva (South American society), 128 ecosystem
Cunningham, C , 37 and food production, 58-9
curing South American tropical, 50-1
and food taboos, 55 education
by shamans, 199 bilingual, 220
and religion, 225
dabucuri festival, 59, 60 Elkin, A. P, 132
daily life emblems of identity, 165
activities of, 39-42 energy, concept of, 209
exchange in. 59-62 environment; see Vaupes region
forest in, 46-50 epidemics, 18
impact of river on, 42-6 Erikson, E., 3
leadership in, 65-8 Ervin-Tripp, S., 167
manioc process in, 51-4 ethnic history, 21-2
property in, 62-5 ethnocentrism, 215, 222
dancing ethnographic descriptions, 23-4
at festivals, 202 ethnolinguistics, 165
symbolism of, 203 excess, as mythological theme, 146
daughter-in-law, role of, 185-6 exchange
Davis, S. H., 211 at festivals, 202
dead-spirit houses, 201 impact of acculturation on, 61-2
death, ritual associated with, 200-1 and intersettlement dependence, 99
de Beauvoir, S., 181 marriage as, 124, 129
dcculturation patterns of, 59-62
process of, 213, 214 and property rights, 64-5
role of extractive industries in, 215-17 and social structure, 60-1
see also world, outside exogamous group, 79
Denevan, D. M., 50 exogamy
Desana (wind people), 229 Bara-Tukano, 90, 92
DeVore, I., 151 language, 7-8, 83-4, 135-7, 175-6
275
Index
exogamy (com.) gardening, 54
linguistic, 20, 21 Gasche\ J., 95, 132
residential, 135 gender, in social identity, 2; see also men,
settlement, 70 sex roles, women
sib, 76 general stores, 25
expectations, kinship-related, 116-17 generation, in kinship system, 107
explorers, early, 22 Giacone, A., 24, 149
gift giving, gossip about, 57
farina, 33, 184 Glasse, R. M , 69
Ferguson, C. A., 175 Godelier, M., 7, 102
festivals Goffman, E., 162, 211
dabucuri, 59, 60 Goldman, I., 5, 11, 16, 18, 19, 21, 23, 26,
food consumption during, 56 32,43, 53, 58, 71, 82, 84, 96, 97.
main activity of, 203 98, 99, 100, 101, 107, 127, 130, 132,
preparation for, 203 134, 135, 136, 149, 151, 152, 153,
purpose of, 202 156, 157, 158, 161, 162, 173, 178, 184,
"soul food" at, 202 187, 189, 190, 203, 204, 206
feuding, and control strategies, 100 Goldschmidt, W, 186
fish, kinds of, 43 Goodenough, W, 6, 108
Fish Anaconda Daughter myth, 75, 122-3, Goodland, R. J. A., 50
143, 161 gossip, 40, 57, 193
fishing, 42-3 government, Colombian, 217-18, 226
fish people, 42, 89; see also Bara governmental bureaucracy, 213
fluidity, concept of, 8-9, 228 grandchildren, role of, 120-1
folklorization, 61 grandparents, role of, 120-1
food greed, as mythological theme, 146
cooked, 184 Greenberg, J. H., 24
cultivated, 50-9 greeting ritual. 98
distribution of, 47, 56 Gregor, T, 194
importance of, 54-9 Guayaki [Ache] (South American society), 128
nonhunted forest, 50 Gumperz, J. J., 166, 167. 168, 174, 175
ownership of, 57-8 Guzman. Antonio, 24
"real" vs. "sou!," 56
"soul," 197, 202 Hallowell. A. I.. 2
taboos, 47, 54-5 Hames, R., 50
see also beverages Harris. M., 63
food exchange ceremonies, 143, 183-4 harvesting, of manioc, 52
food production headmen
sex roles in, 182-4 qualities of. 67
and social structure, 58-9 and succession. 104
see also manioc process support for, 212
forest tasks of. 65
animals of, 47 typical, 67
dangers of, 48 wives of, 67-8
hunting in, 46 hearth family, 28, 105
insects of, 48-9 Helm, J.. 7
nonhunted foods of, 50 Hicks. D., 37
spirits of, 49 hierarchial ordering, 103-4
in TUkanoan classification, 227 Holmberg, A.. 194
Fortes, M., 166 Holy Sisters of Mother Laura. 220
Fosberg, E R., 50 hospitality, rules of. 97-8
friagem, 16 Hugh-Jones. C , 6. 7, 11, 14, 24, 55, 56, 60,
Fried, M, H., 7 63. 72. 75, 76. 77, 79. 81. 82, 84,
Fulop, Marcos, 5, 24, 71, 93 88. 89. 93. 96. 97, 102. 103. 105. 107.
Fundacao Nacional do Indio (FUNAI), 123. 124. 136. 151. 160. 183. 184,
Brazilian, 25 187, 188. 189. 191. 192. 195. 197. 202,
fungal infections, 19 230. 232. 236, 238
276
Index
Hugh-Jones, S., 6, 11, 16. 19. 20. 21, 24, lictive and ceremonial kinship, 123
60. 63. 75. 76. 79. 84. 87. 102, 105, grandparent and grandchild, 120-1
115. 123. 133. 151. 154. 173, 177. 181, nephews and nieces, 119-20
182, 190. 192, 195, 196, 197. 204, parents and offspring. 117-19
209, 228, 234. 236, 237 siblings. 120
Humboldt. Alexandre von. 23 uncles and aunts. 119-20
hunter-gatherers, regional systems among, 6 knowledge, ritual, 196
hunting. 43, 46 Koch-Griinberg, T. 21. 23. 77. 136, 149,
animals hunted. 47 151. 152. 156. 157
symbolism surrounding. 47-8. 49 Kok. R P.. 24
in Tukanoan world, 209 Kracke, W, 65
husband, role of, 185-8; see also spouses Krikati (South American society), 194
Hymes. D., 7
labor, sexual division of, 183, 184
ideal, concept of, 9-11 Labov, W, 167, 168, 175, 176
identification, with specific language, 168 Lambert, W E . , 175
identity, badges of. language and speech as, Lamphere, L., 180
165-71; see also social identity Langdon, T., 11, 24, 55, 58. 197, 202
illness, and food taboos, 55 language
incest. 130. 189, 200 as badge of identity, 165-71
individualism changing, 82
and names, 106 and exogamy, 83-4
of shaman. 234-5 families represented, 19
in Tukanoan social identity. 233 importance in Tukanoan culture, 177-8
Tukanoan vs. Western concepts of. 212, and nonlinguistic environment, 171-7
235 and passing as white, 224
initiation ritual, male, 204, 234, 237 Spanish, 167, 220, 224
insects, variety of, 48-9 language distance
intermarriage. Tukanoan-Makii, 156. 240 measured by shared vocabulary, 173
intimacy models of, 172
and exchange. 60 language groups, 7
and tictive kinship, 123 and acculturation, 69-70
and greeting forms. 98 adequacy of term, 79, 81
Iroquois (North American society), 194 boundary maintenance between, 232
Irwin. H. S., 50 definition for, 78
dyadic relations between, 103
Jackson. J. E.. 6. 72. 108. 132. 179 egalitarianism among, 164, 175
Jacopin, P. 95. 132. 177 exogamy of, 7-8, 135-7
jaguar people. 89 genetically related, 173
Javerians (missionaries), 17, 23. 219. 220, geographic distribution of, 79
222 and kinship lines, 78
jealousy, as mythological theme. 145 and linguistic discreteness, 174-5
ofMaku, 149
Kaplan, B . 3 and marriage, 81, 82, 100
Kelly. R.. 191 marriage between selected, 94, 95
kinship names of, 85
and expectations and behavior, 108-17 population estimates for, 87
and language groups, 78 and settlements, 80
mythical illustrations of. 108. I l l , 112-16 and social boundaries, 81
names and, 105-6 in social identity, 234
and settlements, 77 Tukanoan, 82
in social structure, 102-3 Lave, J. C , 194
specific roles. 117 Leach, E. R., 70
terminology. 106-8. 109-11 leadership
kinship roles headmen, 65, 67, 104, 212
aftines. 121-3 nature of, 212
cousins. 121 voluntary nature of, 66
277
Index
leadership (com.) Tukanoan interaction with, 240

war chiefs, 66 Tukanoan view of, 178
Lee, R. B., 151 malaria, 18
levirate, 71 manicuera, 33
Levi-Strauss, C , 9, 10, 70, 174, 182 manioc beer, 53
Lingua Geral, 19 manioc bread, 33, 35
linguistics manioc process, 39, 51-2
and descent group ties, 36 harvesting in, 52
in ethnographic descriptions, 23 peeling and grating in, 53
language distance, 171-4 in ritual, 183, 184
and social attitudes, 174 manioc products, 52, 53, 183
variations, 19-20 Markham, C. R., 66, 68
llanos, 22 marriage
local descent groups, 70 "abduction," 133
alliance of, 137-8 acceptable model of, 147
and kinship relations between spouses, 131 adjustment of newlyweds in, 141-2
and marriage, 185-6 age at, 128
in traditional social structure, 73 and alliance, 137-8
longhouse, 1 and divorce, 129, 241
abandonment of, 70 exchange associated with, 60, 126
ceremonial, 35 function of, 185
choosing site for, 18, 31 ideal, 240
ground plan of interior, 34 impact of missionaries on, 136-7
hearth family of, 28, 105 importance of, 124
inside, 33-6 incidence of, 139
outside of, 31—2 initiation of, 125
protection for, 32 and kinship relations between spouses,
setting for, 26, 27 130-5
significance of, 36-8 and language group exogamy, 135-7
structure of, 30-1 and language groups, 81, 82, 100
in Tukanoan classification system, 227, 230 myths concerning, 115-16
Lounsbury, E G., 137 polygamous, 128, 193
lumber extraction, 215 principles of, 125-6
lunar eclipses, 205, 206, 208, 223, 231 real vs. ideal forms, 134-5
and residential exogamy, 135
MacCreagh, G., 23, 77, 151 stability of, 194
McGovern, M. W, 23, 77, 151, 152, 156, in Tukanoan cosmology, 209
162 Marriage making, phratric principles in, 92-3,
maintenance, in social identity, 231 94
Makii (South American society), 21 Maybury-Lewis, D. H. P, 124, 194
attached to longhouse, 155-6 Mead, G. H., 2
background to, 148-51 meals
characteristics of, 149 breakfast, 39
ethnographic description of, 24 formal communal, 56
interaction between Tukanoans and, 154-8 restrictions during, 55—6
intermarriage of, 156, 240 see also food
linguistic groups of, 149 meaning, multiple levels of, 228; see also
marriage rules of, 126, 150 symbolism
origin of, 149 Meggitt, M. } . , 69
poison manufactured by, 153 Mehinacu (South American society), 194
role as servants of, 160 men
as slaves, 162 crops tended by, 52
social organization of, 150-1 and gardening, 54
as symbol to Tukanoans, 158-61 in male-female dichotomy, 229
status of, 156 personal appearance of, 41-2
Tukanoan attitudes toward, 103, 151-3 as providers, 56-7
in Tukanoan classification system, 229 supremacy of, 192
278
Index
see also sex roles in daily life, 42
menstruation, symbolism of, 190-1 at festivals, 203
metamorphosis, in Tukanoan cosmology, 209 see also Yurupari instruments
methodology mythology, 212, 226
fluidity in, 8-9 affinal relationships in, 122
ideal-real dichotomy in, 9-11 cannibalism in, 152
regional perspective in, 5-8 exchange in, 60
social identity, 2-5 language groups in, 78, 86
Mteaux, A., 149, 157 marriage behavior and expectations in,
Michaelson, E. J., 186 143-7
migration, and marriage system, 100 male supremacy in, 182
"Milk river," 204 myths
Miller, D. R., 234 "Fish Anaconda Daughter," 75, 122-3,
Miller, W R., 175 143, 161
mingao, 39 about "first people," 72
miriti palm, 13 function of, 208
miriti' swamp, in Tukanoan cosmology, 206 with kinship-related themes, 108, 111,
missionaries 112-16
arrival of, 218-19 of life after death, 236
as change agents, 212 Maku in, 158
and deculturation, 213 origin, 72, 89, 101
ethnographic descriptions of, 24 and shamanism, 196
first, 23 star girl, 205
ideological warfare of, 220-1 about women's position, 70
impact on Tukanoan social structure of, 17 Yurupari instruments in, 188
and marriage customs, 136-7
rivalry among, 219, 221 names
trouble with, 222-3 of language groups, 85
mission towns patronyms, 106
houses of, 35 sib-supplied, 72
and land shortages, 18 naming
marriages in, 141 purpose of, 105
population of, 69 in social identity, 236
Mini (town), 25, 218 Needham, R., 174
and government representation, 218 neighbors, settlement, 98-9
homesteaders in, 217 nephews, kinship role of, 119-20
inhabitants of, 223-4 newborns, 37; see also childbirth
moderation, in social identity, 231, 232 Nheengatu (language family), 19
Moerman, M., 165, 166, 168 nieces
money, introduction of, 61 kinship role of, 119-20
Monfortians, 218, 222 Nimuendaju, C., 23, 71
monogamy, view of, 128 noise, tolerance for, 40
moon eclipse, 205, 206, 208, 223, 231
moral conduct, attitude toward, 208-9 O'Laughlin, B., 59, 61
Morey, N. C , 170 onchocerciasis (African river blindness), 18
Morey, R. V, 170 Orellana (early writer), 18
Moser, B., 13, 16, 24, 236 ostracism, 40, 193
Muller, Sophia, 23, 219 out-migration, 18
multilingualism, 20, 21 Owen, R., 175
Indian attitudes toward, 177 ownership
and social cleavages, 175 concept of, 57
widespread, 101 of land, 62
see also language groups see also property
murder, 200-1
Murphy, R., 61, 66 pahko-pona relationship, ambiguity of, 144
Murphy, Y, 61 Papuri River, 83
music parasites, 18
279
Index
parents settlements intermarrying with, 140

and childbirth, 198 see also daily life
role of, 117-19, 188-9 Pygmies (African society), 161
parents-in-law, 120
patronyms, 106 quarrels, 62
Peirce, C , 2 quinapira (pepperpot sauce), 33
people, in Tukanoan classification, 227
pepperpot sauce, 33 racism, 167
permanence, in social identity, 235 Radcliffe-Brown, A. R., 132
personality, in social identity, 236 raiding, and control strategies, 100
personal appearance, attention given to, 41 rain forest, ecology of, 51; see also forest
petroglyphs, 16 rainy season, 13
pets, 41 ranking, in TUkanoan world view, 212
phratry rapids, and river transportation, 16, 44; see
definition of, 86 also rivers
and language distance, 171-4 raudales (cataracts), 16
and language groups included in, 91 Raven, P H., 28
and marriage making, 92-3, 94 reality, concept of, 9-11
organization of, 88 rebirth, in Thkanoan cosmology, 209
in traditional social structure, 73 reciprocity
Pira-parana River, 83 peasant patterns of, 57
Pira-parana region women as expression of, 187-8
map of, 14 see also exchange
settlements of, 16 regionality, in social identity, 238
poison, fish, 42-3 regional perspective, concept of, 5-8
political organization, 68 Reichel-Dolmatoff, G., 5, 11, 13, 16, 21, 23,
polyandry, 129-30 49, 87, 89, 96, 100, 151, 152, 153,
polygamy, 128 156, 161, 177, 205, 207, 230
polygyny Reid, H., 17, 18, 149, 150, 154, 157, 162,
disapproval of, 128 240, 241
incidence of, 193 reincarnation, in Tukanoan cosmology, 209
population relationality
acculturation of, 24 in social identity, 4, 232-3
decline, 19 in Tukanoan public ritual, 234
density, 17, 21, 69, 99-100 residence
impact of disease on, 18 patterns of, 28
language group estimates, 87 settlement, 77
power, ritual, 196 see also longhouse
prefectura, 25 respect, in parent-offspring relationship
property 118-19
individual, 63 Rice, H., 23
lending of, 65 rigidity, vs. fluidity, 8, 9
manufacture of artifacts, 64 risk, in social identity, 231, 232
ownership of land, 62-3 ritual
private, 212 associated with death, 200-1
ritual, 63-4, 76 associated with river, 45
secular communal, 63 function of, 208
Protestant missionaries, 23 greeting, 98
influence of, 224-5 individuality downplayed in, 234
rivalry with Catholic missionaries, 219 in longhouse, 37-8
see also Summer Institute of Linguistics male, 181-2
puberty rites, 209; see also initiation ritual male initiation, 234
Pumanaka buro property associated with, 63-4, 76
catechism program at, 222 sex roles in, 183-4
health and nutrition at, 58 Yuruparf, 191
marriages at, 139 see also symbolism
people of, 26, 28-30
280
Index
characteristics of, 16 and domestic activities. 184-5
and communication, 44-5 and food production, 182-4
dangers of, 45 and male and female sexuality, 189-92
and fishing, 42-3 of parents, 188
ritual associated with, 4 5 - 6 problems with research in, 180-1
and social organization, 99 relations between men and women, 181-2
and transportation, 43 of spouses, 185-8
in Tukanoan classification system, 227 shamanism, 195-7
Riviere, P G., 66 birth in, 197-8
Rodriguez Lamus, L. R., 24 death in, 200-1
Rosaldo, M. Z., 130 and hunting, 48
Ross, E. B., 63 sorcery and disease in, 198-9
rubber boom, 22 shamans, 66
rubber gatherers ambiguity of, 196
dealing with Tukanoans of, 211 curing by, 199
labor "recruited" by, 17 and food, 184, 189
Tukanoans as, 216 as food intermediaries, 197
see also caucheros function of, 208
rubber industry, of Colombia, 215 individualism of, 234-5
Ruddle, K., 50, 54 language of, 177
snuff taken by, 153
Salesians (missionaries), 17—18 in Tukanoan classification, 230
San Jose de Guaviare (town), 217 and wahtfa, 208
Sankoff, G., 175 Shapiro, J. R., 194
Sapir, E., 166 Sharanahua (South American society), 194
Sapir, J. D., 168 Shavante (South American society), 194
savanna, 22, 51 sibling rivalry, 114
Schneider, J., 186 siblings, in kinship system, 120
Schultes, R. E., 153 sibs (clans), 37
seasons, markers of, 16 exogamy of, 76
segregation, sexual, 141, 142, 181, 183, 193 and linguistic proficiency, 176
Selby, H. A., 107, 127 and local descent groups, 72
self meaning of, 75-6
concept of, 2 names of, 72
Tukanoan vs. Western, 3-4 ranking of, 72, 74-5
see also individualism ritual property of, 76
servant-master relationship, Maku in, 163 system of, 71-2
servants sickness
role of, 119 attitudes toward, 37, 40
role of Maku as, 160 and food taboos, 54-5
Servic.o do Protagao ao Indio (SPI), Brazilian, and thunder and lightning, 205
25 Silverwood-Cope, P, 24, 28, 87, 107, 126,
settlements 149, 150, 151, 152, 153, 154, 156,
clearing in, 31—2 157, 160, 240
importance of, 77 Singer, M., 2
interaction between, 96-104 Siriono (South American society), 128, 194
intermarrying among, 140 Siskind, J., 63, 193, 194
language groups of, 80 sisters
and local exogamy, 77 in kinship system, 120
regional perspective on, 5 marriage exchange of, 126, 127, 129
size of, 21 sisters-in-law, 122
sex skin-changing, 209
in kinship system, 107 slavery, 156-8
myths concerning, 115-16 smoking, of fish, 43
in social identity, 2 snakebite
sex roles danger of, 48
defined, 179 thunder and lightning connected with, 205
281
Index
snakebite (com.) Stradelli, E., 23
victim, 37 starvation, 58
snuff Stemberg, H. O., 50
at festivals, 203 storage, food, 58
shamans', 153 stratification, social, 163
social identity subsistence, sex roles in, 183-4
balance in, 231, 232 Summer Institute of Linguistics (SIL), 20, 23,
concept of, 2-5 87, 174, 175, 176
language as emblem of, 170 and Colombian government, 219-20
language group in, 234 influence of, 224-5
language and speech as indicators of, 166 swamps
maintenance in, 231 in Tukanoan cosmology, 49, 206
moderation in, 231, 232 in Tukanoan environment, 13, 16
naming in, 236 swidden horticultural systems, 50
permanence vs. transience in, 235 symbiosis, of servant-master interaction, 161
vs. personal identity, 2 symbolism
regional orientation in, 238 concerned with rivers, 46
relationality in, 231-3 destruction-creation, 191
risk in, 231, 232 of differentiation, 89
socialization, 118 of hunting, 47-8
social structure Maku in, 158-61
kinship in, 102-3 sex roles in, 183-4
and language distance, 171-4 of sib structure, 75-6
and language group, 77-86 see also mythology
and linguistic discreteness, 174-6
and linguistic proficiency, 176-7 taboos
phratry in, 86-96 food, 47, 54-5
regional integration, 96-104 incest, 130, 189, 200
and residence, 69-71 on mentioning names of dead, 105
role of language and speech in, 165-71 taxonomy, 228
sibs (clans) in, 71-7 Tayler, D., 13, 16, 24, 236
traditional local group organization in, 73, technology
77 canoe-building, 45
Sorensen, A. R, Jr., 6, 19, 20, 24, 82, 93, fishing, 43
95, 100, 101, 136, 168, 169, 171, hunting, 46
172, 173, 174, 177 temperature, of Vaup£s region, 16
Soto Holgui'n, Alvaro, 24 terminology
"soul food," 197, 202 kinship, 106-8, 109-11
Spanish language, 220 simplified Bara kinship, 112
ability to speak, 167 zero generation, 90
and passing as white, 224 theft
speech changing attitudes toward, 65
as badge of identity, 165 spirits and, 200
value placed on correct, 169 Tindale, N. B., 176
see also language tipiti, symbol of, 191
spirits tobacco, 197, 203
"dead-people," 200 Torres Laborde, A., 24
of forest, 49 towns; see mission towns, MM
see also wahti'a trade, and intersettlement dependence, 99
spouses trade goods
ages of, 193 disagreements over, 62
behavior of, 141 introduced by whites, 22
kinship relations between, 130-5 trance, and shamanism, 196, 208
see also marriage, sex roles transience, in social identity, 4, 235
Spruce, R., 22, 23, 157 transportation
star girl, 205 and extractive industries, 215
Stoll, D., 219 and rainy season, 13
282
Index
regional perspective, 6 and headman, 67
technology, 4 3 - 4 relationships with, 61
transsexuals, 2 Tukanoan view of, 213
tribe see also missionaries, rubber gatherers
definition of, 7 Whitten. N. E., Jr., 53
in traditional social structure, 73 wife-getting procedures, 134
tuberculosis, 18 wives
Tukanoan speakers, 21 of headman, 67-8
Tukano language, 175 role of, 185-8
Tlipian (language family), 19 see also spouses
Tumage, L., 87 Wilson, R., 168
Tumbull, C , 161 Witoto (South American society), 132
Tuyuka (mud people or clay people), 229 Wittgenstein, L., 108
Tyler, S. A., 108 Wolf, M., 186
women
uncles, kinship role of, 119-20 crops tended by, 52
universe, Tukanoan, 1, 46; see also and gardening, 54
cosmology in horticultural societies, 180
uranium, 215 and hunting, 46-7
in male-female dichotomy, 229
values, Tukanoan, and deculturation process, and marriage stability, 194
214 personal appearance of, 41
Van Ernst, P, 157, 158 pivotal position of, 145, 146, 185, 233
Vaup6s region solidarity of. 142
cultivation in, 51-4 status of, 192-3
ecological setting for, 13 value of, 192
language families of, 19 see also bride, sex roles
map of, 14-15 world, outside
population density for, 17 Colombian government, 217-18
Vaupes system, fluid boundaries of, 8 and deculturation process. 211-15
villages, population of, 69 extractive industries from, 215-17
visiting, importance of, 98 homesteaders from, 217
missionaries from, 218-23
wahtfa (spirits) Mitu, 223-4
characteristics of, 207 world, Tukanoan; see cosmology, Tukanoan
encounters with, 208 work
in Tukanoan classification, 229 attitudes toward, 40-1
Wallace, A. F C . 28 and sexual division of labor, 183, 184
Wallace, A. R., 22, 23, 77, 99, 136, 151,
157 xenophobia, 97
Wallis, E. E., 20
Waltz, N. E., 19, 20, 82, 96, 172, 173, 174 Yanomamo (South American society), 97, 99,
war chiefs, power of, 66 127, 154
Ward, B., 166 interaction with other settlements of. 64
warfare and trade goods, 128
cessation of, 194 women of. 192, 194
and settlement patterns, 97 Yukuna (language group). 132
Warner, W. L., 176 Yurupari instruments, 190, 191
weapons, for hunting, 46 origin of, 188
Wheeler, A., 19, 20, 82, 96, 172, 173, 174 sacredness of, 208
Whiffen, T., 22, 23, 71, 156 Yurupari Rapids, 16
whites, 1
discrimination by, 224 Zuidema, R. T., 174
283
CAMBRIDGE STUDIES IN SOCIAL ANTHROPOLOGY
General Editor: Jack Goody
1. The Political Organisation of Unyamwezi
R. G. ABRAHAMS
2. Buddhism and the Spirit Cults in North-East Thailand*
S. J. TAMBIAH
3. Kalahari Village Politics: An African Democracy
ADAM KUPER
4. The Rope of Moka: Big-Men and Ceremonial Exchange in Mount Hagen, New
Guinea*
ANDREW STRATHERN
5. The Majangir: Ecology and Society of a Southwest Ethiopian People
JACK STAUDER
6. Buddhist Monk, Buddhist Layman: A Study of Urban Monastic Organisation in
Central Thailand
JANE BUNNAG
7. Contexts of Kinship: An Essay in the Family Sociology of the Gonja of Northern
Ghana
ESTHER N. GOODY
8. Marriage among a Matrilineal Elite: A Family Study of Ghanaian Senior Civil
Servants
CHRISTINE OPPONG
9. Elite Politics in Rural India: Political Stratification and Political Alliances in Western
Maharashtra
ANTHONY T. CARTER
10. Women and Property in Morocco: Their Changing Relation to the Process of Social
Stratification in the Middle Atlas
VANESSA MAHER
11. Rethinking Symbolism*
DAN SPERBER, Translated by Alice L. Morton
12. Resources and Population: A Study of the Gurungs of Nepal
ALAN MACFARLANE
13. Mediterranean Family Structures
Edited by i. a. PERISTIANY
14. Spirits of Protest: Spirit Mediums and the Articulation of Consensus among the
Zezuru of Southern Rhodesia (Zimbabwe)
PETER FRY
15. World Conqueror and World Renouncer: A Study of Buddhism and Polity in
Thailand against a Historical Background*
s. J. TAMBIAH
16. Outline of a Theory of Practice*
PIERRE BOURDIEU, Translated by Richard Nice
17. Production and Reproduction: A Comparative Study of the Domestic Domain*
JACK GOODY
18. Perspectives in Marxist Anthropology*
MAURICE GODELIER, Translated by Robert Brain
285
19. The Fate of Shechem, or the Politics of Sex: Essays in the Anthropology of the
Mediterranean
JULIAN PITT-RIVERS
20. People of the Zongo: The Transformation of Ethnic Identities in Ghana
ENID SCHILDKROUT
21. Casting out Anger: Religion among the Taita of Kenya
GRACE HARRIS
22. Rituals of the Kandyan State
H. L. SENEVIRATNE
23. Australian Kin Classification
HAROLD W. SCHEEFLER
24. The Palm and the Pleiades: Initiation and Cosmology in Northwest Amazonia
STEPHEN HUGH-JONES
25. Nomads of South Siberia: The Pastoral Economies of Tuva
S. I. VAINSHTEIN
26. From the Milk River: Spatial and Temporal Processes in Northwest Amazonia
CHRISTINE HUGH-JONES
27. Day of Shining Red: An Essay on Understanding Ritual
GILBERT LEWIS
28. Hunters, Pastoralists and Ranchers: Reindeer Economies and Their Transformations
TIM INGOLD
29. The Wood-Carvers of Hong Kong: Craft Production in the World Capitalist Periphery
EUGENE COOPER
30. Minangkabau Social Formations: Indonesian Peasants and the World Economy
JOEL S. KAHN
31. Patrons and Partisans: A Study of Two Southern Italian Comuni
CAROLINE WHITE
32. Muslim Society
ERNEST GELLNER
33. Why Marry Her? Society and Symbolic Structures
LUC DE HEUSCH
34. Chinese Ritual and Politics
EMILY AHERN
35. Parenthood and Social Reproduction: Fostering and Occupational Roles in West
Africa
ESTHER N. GOODY
36. Dravidian Kinship
THOMAS R. TRAUTMANN
37. The Anthropological Circle: Symbol, Function, History
MARC AUGE
38. Rural Society in Southeast India
KATHLEEN GOUGH
39. The Fish-People: Linguistic Endogamy and Tukanoan Identity in the Northwest
Amazon
JEAN. JACKSON
40. Karl Marx Collective: Political Economy of a Siberian Collective Farm
CAROLINE HUMPHREY
286
41. Ecology and Exchange in the Andes
DAVID LEHMANN
42. Traders without Trade: Responses to Trade in Two Dyula Communities
ROBERT LAUNAY
^Published also as a paperback
287

Jackson Jean E. - The Fish People Linguistic Exogamy and Tukanoan Identity in Northwest Amazonia

Uploaded by

Copyright:

Available Formats

Jackson Jean E. - The Fish People Linguistic Exogamy and Tukanoan Identity in Northwest Amazonia

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Jackson Jean E. - The Fish People Linguistic Exogamy and Tukanoan Identity in Northwest Amazonia

Uploaded by

Copyright:

Available Formats

Cambridge Studies in Social Anthropology

General Editor: Jack Goody

THE FISH PEOPLE

CAMBRIDGE UNIVERSITY PRESS

Cambridge University Press

Published in the United States of America by

© Cambridge University Press 1983

This publication is in copyright. Subject to statutory exception

First published 1983

A catalogue recordfor this publication is available from the British Library

Library of Congress Cataloguing in Publication Data

Jackson, Jean E. (Jean Elizabeth), 1943—

ISBN 978-0-521-23921-9 Hardback

Cambridge University Press has no responsibility for the persistence or

List of figures, maps, and tables x

1 Purpose and organization of the book 1

2 Introduction to the Central Northwest Amazon 13

4 Economic and political life 39

5 Vaupes social structure 69

8 Tukanoans and Maku 148

9 The role of language and speech in Tukanoan identity 164

10 Male and female identity 179

11 Tukanoans' place in the cosmos 195

12 Tukanoans and the outside world 211

13 Conclusions: themes in Tukanoan social identity 227

In November 1968 I arrived in the Vaupe"s territory of Colombia planning to

investigation of the social structure of Tukanoans, in particular those of the

Research in Colombia during 1968-70 was supported by the Danforth Founda-

The orthography used in this book is a phonetic transcription of Bara, simplified

(1971), although more impressed by differences than similarities,also examines

exogamy, people deal almost as frequently with members of other language

Ideal and real

Other seasonal markers are position of the constellations, availability of vari-

Language and linguistics

Early explorer and missionary efforts

Early and recent ethnographic descriptions

Mini, an airstrip town of approximately 1,000 inhabitants in 1970, forms the

The people of Pumanaka buro

I Mario, the headman

Sons: Mariquino, Nazario, Paulino

The longhouse structure

Outside the longhouse

Inside the longhouse

Figure 3. Ground plan of longhouse interior.

Significance of the longhouse

dimensions of existence intersect here, sometimes in opposition to each other -

standably startled, reacted by covering us all with a foul-smelling, warm white

Cultivation in the Vaupes.

The importance of food.

n Bara TUkano Wuhana etc. Tuyuka etc. etc.

m Bard Yoara Wainakoroa etc. (approx. 30 sibs) etc. etc.

to reflect social processes in some fashion; in the Vaupes this association, if

The language group

Table 1. Language group affiliation of Tukanoans in sample

Language Men Women Total

Bard 104 86 190

Bara Bara A sweet-smelling plant placed in a

""Northern Barasano" is the designation of language given by Summer Institute of

Language group H S1L T DAI R S

Table 4. Zero generation terminology at three levels of inclusion