Dialectical Societies: The Ge and Bororo of Central Brazil (Harvard Studies in Cultural Anthropology)

Dialectical Societies The Ge and Bororo of Central Brazil

Edited by David Maybury-Lewis Contributors Joan Bamberger J. Christopher Crocker Roberto da Matta Jean Lave David Maybury-Lewis Julio Cezar Melatti Terence S. Turner

H a r v a r d U n i v e r s i t y Press C a m b r i d g e , Massachusetts and London, England 1979

HARVARD STUDIES IN CULTURAL ANTHROPOLOGY, 1 General Editors David Maybury-Lewis Stanley J. T a m b i a h Evon Z. Vogt, J r . N u r Yalman

T h e H a r v a r d Studies in Cultural Anthropology is founded in the belief that answers to general questions about the h u m a n condition may be discovered through the intensive study of other cultures. T h e series will publish books which elucidate a n d interpret cultural systems in order to contribute to comparative understanding.

Dialectical Societies The Ge and Bororo of Central Brazil

Edited by David Maybury-Lewis Contributors Joan Bamberger J. Christopher Crocker Roberto da Matta Jean Lave David Maybury-Lewis Julio Cezar Melatti Terence S. Turner

H a r v a r d U n i v e r s i t y Press C a m b r i d g e , Massachusetts and London, England 1979

Copyright © 1979 by the President and Fellows of Harvard College All rights reserved Printed in the United States of America Library of Congress Cataloging in Publication Data Main entry under title: Dialectical societies. (Harvard studies in cultural anthropology ; 1) 1. Tapuya Indians—Addresses, essays, lectures. 2. Bororo Indians—Addresses, essays, lectures. I. Maybury-Lewis, David. II. Bamberger, Joan. III. Series. F2520.LT3D5 98K004'98 79-10689 ISBN 0-674-20285-6

T h i s book is dedicated to C u r t N i m u e n d a j u in memory of his pioneering studies in ethnology a n d his deep affection for the Indians of Brazil

Acknowledgments

This volume brings together some of the results of research carried out by seven anthropologists over more than a decade. It would be difficult to list all those people and institutions who have helped each one of us during that time and impossible to acknowledge their help in any adequate fashion. Fieldwork in Central Brazil posed logistic and personal problems which could never have been surmounted with­ out the generous assistance we all received. We would, therefore, like to acknowledge our dependence on and express, however inadequately, our gratitude to our friends and sponsors in Brazil. Above all, we are grateful to the Museu Nacional of the Federal University of Rio de Janeiro, which sponsored ail of our fieldwork, and to Roberto Cardoso de Oliveira who, as director of its Department of Anthro­ pology, acted for many years as godfather to the Harvard-Central Brazil research project. O u r work has been financed from various sources over the years, but we are particularly grateful to the National Institute of Mental Health, whose grant during the period 1962-1967 provided the major funding for the field research carried out under the project. We would also like to thank the Southwest Museum in Los Angeles for permission to reproduce figure 1 in chapter 7 from Nimuendaju, The Serente, and the Clarendon Press of Oxford University for per­ mission to reproduce figure 3 in chapter 7 from David Maybury-Lewis, Akwe-Skavante Society. Finally the editor wishes to thank Judy Bevis for her work in preparing the manuscript for publication and Joan Bamberger, Jason Clay, Pia Maybury-Lewis, and Biorn Maybury-Lewis for cheerfuUy carrying out the tedious task of reading the proofs.

Contents

Foreword

IX

CHARLES WAGLEV

Introduction

1

DAVID MAYBURY-LEWIS

1.

Cycles a n d T r e n d s i n K r i k a t i N a m i n g Practices

16

JEAN LAVE

2.

T h e R e l a t i o n s h i p System of t h e K r a h 6

46

JULIO CEZAR MELATTI

3.

T h e A p i n a y e R e l a t i o n s h i p System: T e r m i n o l o g y a n d Ideology

83

ROBERTO DA MATTA

4.

E x i t a n d Voice i n C e n t r a l Brazil: T h e Politics of F l i g h t i n K a y a p 6 Society

130

JOAN BAMBERGER

5.

T h e G e a n d B o r o r o Societies as Dialectical Systems: A General Model

147

TERENCE S. TURNER

6.

Kinship, Household, and C o m m u n i t y Structure a m o n g the Kayapo

179

TERENCE S. TURNER

7.

C u l t u r a l C a t e g o r i e s of t h e C e n t r a l G e

218

DAVID MAYBURY-LEWIS

8.

Selves a n d A l t e r s a m o n g t h e E a s t e r n B o r o r o

249

J . CHRISTOPHER CROCKER

C o n c l u s i o n : K i n s h i p , Ideology, a n d C u l t u r e

301

DAVID MAYBURY-LEWIS

References Notes Index

315 325 335

Foreword Charles Wagley

T H E E T H N O G R A P H I C P I C T U R E of t h e vast low­ lands of South America is just n o w c o m i n g i n t o focus. I n t h e late 1930s, w h e n I began to study South A m e r i c a n ethnology, Clark Wissler's The American Indian (New York, 1917) was still the s t a n d a r d survey work on the I n d i a n s of both hemispheres. I n that book, Wissler i n c l u d e d all of the tribes of tropical l o w l a n d South America in o n e vague area which he called t h e " M a n i o c Area." H e did distinguish between the so-called T a p u y a of t h e i n l a n d steppes a n d t h e Arawak, C a r i b , a n d T u p i a n g r o u p s of the tropical forest. T h e basic e t h n o g r a p h i c data on lowland South American I n d i a n s were w r i t t e n i n G e r m a n , Swedish, Spanish, Italian, a n d Portuguese. Mainly, they consisted of reports by early chroniclers, naturalists, explorers, o r anthropologists w h o u n d e r t o o k expedi­ tions that yielded fragmentary d a t a on several tribes. Most often there was m o r e information on bows a n d arrows, pottery, basketry, or body p a i n t t h a n on the social organization or ideology of any specific tribe. M o n o g r a p h s on a single t r i b e p r o v i d i n g analysis of its society a n d c u l t u r e were few in n u m b e r . T o study South American ethnology was like trying to p u t together a jigsaw puz­ zle with m a n y of t h e pieces missing. T h e major justification for u n d e r t a k i n g the m o n u m e n t a l Handbook of South American In­ dians (Smithsonian I n s t i t u t i o n , B u r e a u of A m e r i c a n Ethnology, Bulletin 143, 7 vols.) , which began p u b l i c a t i o n in 1946, was to at­ t e m p t to d r a w together and to organize i n t o some sort of a co­ h e r e n t framework t h e very diverse sources of d a t a o n this ethnographically "least k n o w n c o n t i n e n t / ' Since 1950, however, a n increasing n u m b e r of skilled a n d welltrair\ed anthropologists have c o n d u c t e d research in the South A m e r i c a n lowlands. Several of t h e m are E u r o p e a n s or N o r t h ix

A m e r i c a n s , b u t p e r h a p s for t h e first t i m e m u c h of t h e research is n o w b e i n g d o n e by S o u t h A m e r i c a n a n t h r o p o l o g i s t s , s o m e t r a i n e d abroad but others trained in universities in their own countries. T h e r e is a n i n e v i t a b l e g a p b e t w e e n a c t u a l field research a n d the a p p e a r a n c e of research r e p o r t s , b u t in t h e last d e c a d e t h e scientific "payoff" has b e g u n t o a p p e a r . Since a b o u t 1960, a series of excel­ l e n t b o o k - l e n g t h studies a n d n u m e r o u s specialized articles o n low­ land South American cultures have been published. In recent years, n u m e r o u s p a p e r s a n d symposia h a v e b e e n p r e s e n t e d at t h e a n n u a l m e e t i n g s of t h e A m e r i c a n A n t h r o p o l o g i c a l Association a n d at t h e b i a n n u a l m e e t i n g s of t h e I n t e r n a t i o n a l Congress of A m e r i ­ canists. If this g r o w i n g b o d y of research ( m u c h of it still u n p u b ­ lished) is a n y i n d i c a t i o n of t h e f u t u r e , this o n c e n e g l e c t e d a n d poorly u n d e r s t o o d e t h n o g r a p h i c a r e a will b y t h e e n d of this d e c a d e b e o n e of t h e b e t t e r k n o w n r e g i o n s of t h e e t h n o g r a p h i c w o r l d . T h e p u b l i c a t i o n of this v o l u m e b y D a v i d M a y b u r y - L e w i s a n d his associates is a n i m p o r t a n t e v e n t i n t h e c o n t i n u e d d e v e l o p m e n t of S o u t h A m e r i c a n a n t h r o p o l o g y . It is p a r t i c u l a r l y significant be­ cause it h e l p s to resolve a p u z z l i n g a n o m a l y in t h e S o u t h A m e r i ­ c a n e t h n o g r a p h i c scene, n a m e l y , t h e r a t h e r specialized a d a p t a t i o n a n d t h e h i g h l y c o m p l e x social s t r u c t u r e of t h e G e - s p e a k i n g peoples of c e n t r a l Brazil. T h i s b o o k demystifies G e social s t r u c t u r e a n d at t h e s a m e t i m e modifies a n d r e i n t e r p r e t s s o m e of o u r t r a d i t i o n a l ideas, a b o u t k i n s h i p , affiliation, a n d descent. T h e a p p a r e n t a n o m ­ aly s u r r o u n d i n g t h e G e - s p e a k i n g tribes first c a m e to l i g h t w ith t h e p u b l i c a t i o n of t w o articles i n 1937 a n d 1938 by R o b e r t H . L o w i e a n d C u r t N i m u e n d a j u , b o t h of w h i c h a r e c i t e d f r e q u e n t l y in this v o l u m e . T h e s e articles, like N i m u e n d a j i i ' s d e t a i l e d m o n o ­ g r a p h s In t h e A p i n a y ^ ( 1 9 3 9 ) , t h e S h e r e n t e ( 1 9 4 2 ) , a n d E a s t e r n T i m b i r a ( 1 9 4 6 ) , w e r e based u p o n l o n g - t e r m , d e t a i l e d , a n d con­ c e n t r a t e d field research a m o n g these t r i b a l societies. It m i g h t al­ m o s t b e said t h a t C u r t N i m u e n d a j u i n i t i a t e d a n e w era in low­ l a n d S o u t h A m e r i c a n social a n d c u l t u r a l a n t h r o p o l o g y a n d t h a t his research " t u r n e d o n e n d " s o m e of t h e p r e v i o u s l y h e l d views re­ g a r d i n g t h e e t h n o g r a p h y of this p a r t of t h e w o r l d . W h o was this m a n w h o , a l m o s t s i n g l e h a n d e d l y , c h a n g e d t h e face of l o w l a n d S o u t h A m e r i c a n e t h n o g r a p h y a n d t o w h o m this b o o k is d e d i c a t e d ? H e was b o r n i n J e n a , G e r m a n y , in 1883, as C u r t U n k e l . L i t t l e is k n o w n a b o u t his early e d u c a t i o n a n d h o w a n d w h e n h e a c q u i r e d his early i n t e r e s t in S o u t h A m e r i c a n I n d i a n s . At t w e n t y years of age, h e m o v e d t o Brazil, b e c o m i n g a Brazilian citi­ zen only i n 1922, w h e n h e was t h i r t y - n i n e years o l d . I n Brazil h e lived for several years a m o n g t h e G u a r a n i I n d i a n s of s o u t h e r n M a t o Grosso, Sao P a u l o , a n d P a r a n a . H e seems t o h a v e l e a r n e d t o f

speak G u a r a n i a n d h e lived a m o n g t h e G u a r a n i as if h e w e r e a m e m b e r of t h e i r society. I t was from t h e G u a r a n i t h a t h e a c q u i r e d his I n d i a n s u r n a m e , N i m u e n d a j u , w h i c h h e a d o p t e d formally. H i s first two p u b l i c a t i o n s i n t h e Zeitschrift fur Ethnologie in 1914 a n d 1915 w e r e signed C u r t N i m u e n d a j u U n k e l , b u t a f t e r w a r d s h e d r o p p e d his G e r m a n s u r n a m e a n d u s e d his I n d i a n n a m e legally a n d o n all his s u b s e q u e n t p u b l i c a t i o n s . I n 1910, h e b e c a m e a n e m p l o y e e of t h e newly c r e a t e d I n d i a n P r o t e c t i o n Service ( S P I ) , w h i c h m a d e use of his t h o r o u g h k n o w l e d g e of t h e I n d i a n s of s o u t h e r n Brazil a n d subsidized his research for several years. I n fact, w h e n I m e t h i m i n 1941 in B e l e m (state of P a r i ) , h e told m e t h a t h e was a b l e t o w r i t e his first i m p o r t a n t m o n o g r a p h , Die Sagen von der Erschaffung und Vernichtiing der Welt als Grandlage der Religion der Apapocuva-Guarani (Zeitschrift fur Ethnol­ ogie, vol. 46, p p . 2 8 4 - 4 0 3 , 1 9 1 4 ) , as a n e g l e c t e d a n d u n n o t i c e d em­ p l o y e e of t h e S P I i n R i o d e J a n e i r o . I n 1913, h e m o v e d t o Belem, w h e r e h e was r a t h e r vaguely c o n n e c t e d w i t h t h e M u s e u P a r a e n s e E m i l i o G o e l d i for m a n y years. D u r i n g those years h e t r a v e l e d a n d lived a m o n g t h e I n d i a n s of t h e Brazilian A m a z o n , s u p p o r t i n g h i m ­ self by m a k i n g e t h n o g r a p h i c collections for E u r o p e a n m u s e u m s a n d by u n d e r t a k i n g occasional activities for t h e I n d i a n P r o t e c t i o n Service, such as t h e e x p e d i t i o n of 1 9 2 1 - 1 9 2 3 , w h e r e h e led t h e g r o u p which pacified t h e hostile P a r i n t i n t i n o n t h e M a d e i r a R i v e r . By t h e early 1930s, h e h a d e n t e r e d i n t o c o r r e s p o n d e n c e w i t h R o b e r t H . L o w i e , a n d t h e i r long-distance c o l l a b o r a t i o n h a d b e g u n . L o w i e b e c a m e his scientific adviser, t r a n s l a t o r (from Ger­ m a n i n t o E n g l i s h ) , a n d his e d i t o r . L o w i e seems to h a v e b e e n a b l e to find some l i m i t e d financial s u p p o r t for N i m u e n d a j i i ' s fieldwork from t h e C a r n e g i e I n s t i t u t e of W a s h i n g t o n a n d from t h e U n i ­ versity of C a l i f o r n i a at Berkeley. By t h e n , N i m u e n d a j u h a d settled u p o n t h e G e tribes of c e n t r a l Brazil as his p r i n c i p a l focus of field research. F r o m 1930 u n t i l 1940, t h e r e was h a r d l y a year t h a t h e d i d n o t w o r k w i t h o n e or a n o t h e r of these g r o u p s — A p i n a y 6 , S h e r e n t e , K r a h o , K a y a p o , a n d C a n e l a ( t h e local B r a z i l i a n t e r m for t h e R a m k o k a m e k r a a n d A p a n i e k r a of t h e E a s t e r n T i m b i r a ) . It was t h u s d u r i n g t h e late 1930s t h a t his m a j o r m o n o g r a p h s a p p e a r e d o n t h e A p i n a y 6 a n d t h e S h e r e n t e , a n d n o t u n t i l 1946 t h a t his m o s t d e t a i l e d study, The Eastern Timbira, was p u b l i s h e d . N i m u e n d a j u was n o t a n a c a d e m i c a l l y t r a i n e d a n t h r o p o l o g i s t ; in fact, h e was a l m o s t self t a u g h t i n b o t h e t h n o g r a p h y a n d a n t h r o ­ pological t h e o r y . T o m y k n o w l e d g e , h e n e v e r t r a v e l e d a b r o a d after a r r i v i n g i n Brazil a n d n e v e r a t t e n d e d scientific m e e t i n g s . If t h e c h r o n o l o g i c a l list of his fieldwork p u b l i s h e d b y H e r b e r t B a l d u s in

xii • Charles Wagley

his obituary is correct, there was not a year from 1905 u n t i l 1942 that he did not undertake fieldwork of some k i n d with Brazilian Indians (see American Anthropologist, vol. 48, no. 2, pp. 2 3 8 243). Yet he carried on a lively correspondence with E u r o p e a n and North American anthropologists, particularly with R o b e r t H . Lowie. H e m u s t have been r a t h e r widely read, at least in N o r t h and South American ethnology. W h e n I visited h i m in 1941 in his simple home in Bel^m, one large room was filled with books in Portuguese, Spanish, G e r m a n , French, and English. Although he always corresponded with Lowie in G e r m a n , he showed me books in English which Lowie h a d sent him. Yet he was a h u m b l e man, professing to know little in his field, and he plied me with ques­ tions about Franz Boas, R o b e r t Lowie, A. L. Kroeber, Paul Radin, R u t h Benedict, R u t h Bunzel, a n d other American anthropologists whose work he seemed to have read. H e seemed pleased that a young anthropologist from C o l u m b i a University would make a special trip to Belem to visit h i m . Since I was a b o u t to begin re­ search a m o n g the Guajajara-Tenetehara, he made for m e a copy of the kinship terms of this tribe, which he had collected a m o n g them many years before. H e was a superb cartographer, and for several hours he showed m e ethnographic maps of Brazil which h e had prepared (some of them appeared in the Handbook of South American Indians) . Basically, however, Nimuendajii was an ethnographer of the older G e r m a n tradition—closer to the nine­ teenth-century naturalists than to the twentieth-century social scientists. H e was always a passionate Indigenista, fighting as he knew best for the rights of the Brazilian I n d i a n against the "NeoBrazilians," as he called the m o d e r n Brazilians in his writings. It was in keeping with this life long interest that he died, probably of a heart attack, while d o i n g fieldwork a m o n g the T u k u n a In­ dians on the Solimoes River in the u p p e r Amazon. Nimuendajii's field reports opened u p a series of questions for students of lowland South American cultures. H o w could these G e tribes living on the Brazilian steppe or savannah (better desig­ nated by the Portuguese term, cerrado [bush land]) support vil­ lages of eight h u n d r e d , a thousand a n d even more people, as the historical data seem to indicate? A l t h o u g h by Nimuendajii's time they were already reduced in n u m b e r s , these villages were larger than those of the tropical forest societies. T h e s e Ge-speaking tribes with an exceedingly simple technology, who lacked pottery, boats, hammocks, perhaps any elaborate domestic house forms, who depended marginally u p o n horticulture in the narrow gal­ lery forests, and wandered part of each year almost like h u n t i n g and gathering nomads, were suddenly revealed as societies with

Foreword • xiii

complex social systems a n d a very elaborate ceremonial life. T h e y were an anomaly in lowland South American ethnography. T h e y were similar in material traits to the marginal tribes of the southern pampas and T i e r r a del Fuego, b u t more complex in so­ cial structure a n d ceremonial life than the peoples of the Ama­ zonian tropical forest. T h i s contradiction was hardly resolved by the editor of the Handbook of South American Indians, who con­ signed t h e m with certain caveats to volume 2 in company with the simple marginal tribes of the extreme southern portion of the hemisphere. Others attempted to explain their apparent tech­ nological simplicity combined with sociological and ideological complexity by interpreting them as "degenerate remnants of higher South American civilizations" or as peoples driven out of the tropical forest by more aggressive groups (see the Introduc­ tion to this b o o k ) . Neither explanation seemed tenable his­ torically, ethnographically, or functionally. T h u s , when welltrained sociocultural anthropologists were able to d o field research in Brazil, many t u r n e d to further work with the Ge-speaking tribes. O n e of the earlier restudies of the Ge tribes is the longitudinal study of the Eastern T i m b i r a (Ramkokamekra a n d Apaniekra) being carried out by William H . Crocker of the Smithsonian In­ stitution, which began in 1957 a n d still continues. A long series of articles, many of which are cited in this book, have been pub­ lished and two monographs are near completion. Crocker has cor­ rected many of N i m u e n d a j u ' s earlier observations a n d has pro­ vided new data and new interpretations on the c u l t u r e of these important Ge-speaking tribes. T h e second research program, after Nimuendaju's time, on these Ge groups and the Bororo (a tribe with only a distantly re­ lated language b u t many Ge-like features in its social structure) was the H a r v a r d Central Brazil Project, which was carried o u t be­ tween 1962 and 1907 u n d e r the direction of David MayburyLewis. T h i s program was a cooperative endeavor in which meth­ odology, concepts, and theoretical interpretations were shared and a series of independent field research projects were u n d e r t a k e n by the individual participants. T h i s was the first time in the history of lowland South American ethnology that a planned comparative research program had been carried out a m o n g the lowland peoples of South America. (Almost simultaneously, a program involving some of the same anthropologists on Areas of Interethnic Friction was being carried o u t u n d e r the leadership of R o b e r t o Cardoso de Oliveira, then of the Museu Nacional of R i o de J a n e i r o , now of the University of Brasilia). T h e participants in the Harvard pro-

g r a m w e r e b o t h B r a z i l i a n a n d N o r t h A m e r i c a n , as is n o t e d i n t h e f o l l o w i n g I n t r o d u c t i o n . As a r e s u l t of this r e s e a r c h , a l o n g series of articles (citations m a y b e f o u n d i n t h e references for t h e in­ d i v i d u a l p a p e r s ) , a l a r g e n u m b e r of d o c t o r a l theses (most of t h e m still u n p u b l i s h e d ) , a n d b o o k s b y D a v i d M a y b u r y - L e w i s o n t h e A k w e - S h a v a n t e , by J u l i o C e z a r M e l a t t i o n t h e K r a h o , a n d by Roberto da Matta on the Apinay£ have appeared. T h e p r e s e n t b o o k , Dialectical Societies, is a n i m p o r t a n t r e s u l t of this s e m i n a l research p r o g r a m . A l t h o u g h t h e e d i t o r d e n i e s t h a t t h e p r i m a r y a i m of this b o o k is " t o set t h e e t h n o g r a p h i c r e c o r d s t r a i g h t , " it d i d j u s t t h a t for t h i s r e a d e r . A n d it also t a u g h t m e m u c h m o r e a b o u t B r a z i l i a n social a n t h r o p o l o g y a n d t h e g e n e r a l t h e o r y of social s t r u c t u r e . T h i s b o o k c o m b i n e s e x c i t i n g e t h n o ­ g r a p h i c d a t a w i t h n e w t h e o r e t i c a l i n t e r p r e t a t i o n s . I t shows clearly t h a t t h e social systems of t h e G e ( a n d B o r o r o ) tribes a r e b a s e d u p o n u x o r i l o c a l d o m i c i l e s w h i c h a r e cross-cut a n d s e g m e n t e d b y c e r e m o n i a l m o i e t i e s , age-sets, a n d o t h e r social categories o r classifi­ c a t i o n s . I t shows affiliation in several g r o u p s to b e i n t i m a t e l y i n t e r ­ t w i n e d w i t h a c o m p l e x system of t r a n s m i s s i o n of n a m e s a n d asso­ c i a t e d r i g h t s . A n d , a b o v ^ a l l , t h e a u t h o r s of t h i s b o o k e x p l a i n h o w ideology ( t h e n a t i v e s ' t h e o r i e s of t h e i r o w n c u l t u r e ) h e l p e d t h e m to u n d e r s t a n d t h e logic of t h e s e i n t r i c a t e social systems a n d t h e m e a n i n g a n d use of k i n s h i p w i t h i n these systems. B u t let u s a l l o w t h e b o o k to s p e a k for itself. T h e r e is m u c h t h a t is n e w a n d fresh b o t h for s t u d e n t s of l o w l a n d S o u t h A m e r i c a n social a n t h r o p o l o g y a n d anthropological theory in general. U n i v e r s i t y of F l o r i d a , G a i n e s v i l l e O c t o b e r 1978

Dialectical Societies

W e have, t h r o u g h o u t this volume, followed the convention of desig­ n a t i n g kinship terms by their first letter (M for m o t h e r ) , except Z for sister to distinguish it fronjf S for son.

Introduction David Maybury-Lewis

T H E INDIANS of Central Brazil have always been some­ thing of a mystery. In colonial times they were little known, feared and despised alike by the peoples of the coast and by the Portu­ guese, who were quick to explore but slow to settle the interior. Reports dating back to the seventeenth century (Barlaeus 1647, Marcgrav 1648, Baro 1651) speak of the hinterland as being peopled with wild Tapuya, nomads who lived by h u n t i n g and gathering and enjoyed running races with tree trunks on their shoulders. These tribes were probably related to but not the direct ancestors of the Ge and the Bororo who are described in this vol­ ume. It was not until the mid-nineteenth century that von Martius (1867) drew u p a preliminary classification of the languages and peoples of Central Brazil and named the Ge family. T h e first reasonably full accounts of the Ge-speaking peoples and the Bo­ roro came much later. It was Curt Nimuendaju, writing on the Apinay£ (1939), the Sherente (1942), and the Eastern Timbira (1946), as well as the Salesian fathers Colbacchini and Albisetti with their monograph on the Bororo (1942), who established the existence of highly complex social systems among peoples who had hitherto been classified as "marginal." T h e marginal peoples of South America were a residual cate­ gory defined in terms of cultural traits which they were reported to lack, such as agriculture, tobacco, alcohol, boats, pottery, ham­ mocks, and so on (Steward 1946). T h e writings of Nimuendaju and the Salesians, therefore, merely deepened the mystery. How could peoples whose technical ability appeared to be so rudimen­ tary have developed such cultural complexities? Had we under­ estimated their technology, or perhaps been overimpressed by their social institutions? Perhaps the anomaly could be explained 1

1

2 • David May bury-Lewis

by claiming that the Ge and Bororo were degenerate remnants of a higher South American civilization (Levi-Strauss 1944) or at the very least by assuming that they were peripheral barbarians who had acquired, let us say, the cultural panoply of the Incas, without the economic and social substratum necessary to re-create the whole civilization (Haeckel 1939). Such learned speculations seem, to the writers of this volume, somewhat strained. W e have been engaged on research i n t o the c u l t u r e of the Ge and the Bororo for the past decade and this book is, in effect, a t r i b u t e to the people we studied. T h e i r view of the world a n d of their place in it, their social theory and social or­ ganization, are of a complexity and sophistication which have proved d a u n t i n g . T o us, w ho elected to try a n d understand them and to communicate this understanding, it has seemed as if each analytical advance was like scaling a foothill. It presented us with views of more important peaks which w ere yet to be climbed. W e do not claim to have arrived even now at the Everest of o u r research. Yet we believe that \ h e essays in this book provide an adequate preliminary analysis of these i n t r i g u i n g systems a n d at least point to some of the comparative implications of o u r inquiry. T h e present series of investigations started in the mid-1950s, when Professor H e r b e r t Baldus, w h o was my teacher at the Escola de Sociologia e Polftica of the University of Sao Paulo, urged me to take u p the study of the Ge-speaking peoples once again. H e pointed out that the rich materials already provided by N i m u e n dajii a n d the Salesians raised a host of interesting analytical ques­ tions which could only be resolved by further fieldwork. I decided to u n d e r t a k e the task. Like my predecessors, I was intrigued by the social systems of the Central Brazilian peoples and, above all, by their dual organ­ ization. T h e i r interlocking systems of exogamous moieties and agamous moieties, with all their overlapping social fields and a t t e n d a n t complications, were a model maker's delight; b u t how did they actually work on a day-to-day basis? W h a t did they mean to the people who ordered their lives by them? I felt that if I could answer such questions, then we w ould be in a better position to understand the meaning and function of dual organization the world over. But in order to answer these questions, I had to find a dualistic society, going about its business with a m i n i m u m of in­ terference from the outside world, and to study its workings. T h i s was not going to be too easy. N i m u e n d a j u had stressed that dual organization was the key to the u n d e r s t a n d i n g of the Gespeaking peoples; yet there were disquieting hints in his mono­ graphs that the system was breaking down when he had studied it, T

r

r

Introduction

• 3

that the Indians themselves n o longer lived by their own rules— and N i m u e n d a j u was describing it as he h a d seen it in" the 1930s. Where, then, was I to find a fully functioning system of dual or­ ganization in Central Brazil twenty years later? T h e answer seemed to be among the Shavante. T h e problem with the Shavante was that they were hostile. At­ tempts were being made to establish peaceful contact with them, but these had not so far been successful. In any case, they could be expected to b e monolingual, so that fieldwork a m o n g them would require me to learn their language somehow. I therefore went off to study the Sherente in 1955-56. T h e Sherente were r e p u t e d to have been one people with the Shavante in the distant past and therefore presumably spoke a similar or perhaps even identical language. T h e y were also settled and comparatively accessible, a n d they spoke Portuguese as well as their own language. I r e t u r n e d to the field in 1958-59. By this time it was possible to work with the recently contacted Shavante. I discovered that they and the Sherente spoke dialects of a c o m m o n language, Akw e. T h e result was that I could communicate with the Shavante when I got into the field, although they found my Sherente version of Akwe hilariously funny. My work with the Sherente had already taught me that Nimuendajii's forebodings about the i m m i n e n t collapse of their culture a n d the demise of their dualistic system twenty years previously had been p r e m a t u r e . T h e y h a d certainly adapted to changing cir­ cumstances a n d learned to live very m u c h like the poor back­ woodsmen of that part of Brazil; b u t they spoke their own lan­ guage and m a i n t a i n e d a living culture of their own, including the traditional dualism. It seemed, then, that it was not too late to undertake a comparative study of the social institutions of the G£speaking peoples. I therefore launched the H a r v a r d Central Brazil Project in an effort to solve some of the puzzles of Central Bra­ zilian ethnology to which I have already alluded. My collaborators and I resolved to try and study the societies of Central Brazil through controlled comparison. W e would carry out an all b u t simultaneous investigation of the Ge-speaking peo­ ples, focusing on the dual organization which they were all re­ puted to have. Since these peoples all spoke closely related lan­ guages, it was reasonable to suppose that they had had q u i t e close historical connections and had thus developed similar institutions. W e hoped that our investigations would b e able to elucidate the themes and variations in these institutional arrangements. T h e Bororo presented a difficulty. Linguists could not decide whether their language could be classed as belonging to the Ge r

4 • David

Maybury-Lewis

family or not. If not, it was at least a not too distant relative. It was clear that the Bororo h a d lived cheek by jowl with the Ge speakers for a long time. T h e i r culture presented fascinating ana­ logues with, and equally interesting divergences from more conven­ tional Ge patterns. If they were n o t q u i t e Ge, it seemed that they were not q u i t e non-Ge either. W o u l d it not, therefore, b e both fruitful a n d methodologically defensible to try to focus on a G£Bororo cultural complex? T h i s was the line of inquiry that we re­ solved to pursue. T h e N a m b i k w a r a case was more difficult. T h e N a m b i k w a r a were reported to have dual organization a n d a two-section system of kinship terminology. T h e y lived o n the fringes of the lands traditionally occupied by the G e a n d Bororo speakers, b u t their language was q u i t e different from that of t h e former. W e sched­ uled a study of the N a m b i k w a r a in our initial plans for the proj­ ect, feeling that the data would b e instructive, if only for purposes of contrast. As the research proceeded, however, we decided to focus on o u r controlled comparison of the Ge and the Bororo a n d to treat the N a m b i k w a r a as a separate case. T h e interrelationships between the various peoples who speak languages belonging to the Ge family are set out in T a b l e 1. At the m i n i m u m level of differentiation, the languages or dialects TABLE 1. Linguistic affinities of G£- and Bororo-speaking peoples Northern Ge Eastern Timbira Apaniekra Ramkokamekra KrTkati Krah6 Western Timbira Apinay£ Kayap6 Gorotire Xikrin Suva Central Gi Sherente Shavante Bororo Southern Gi Kaingang Shokleng

Introduction

Road State boundary Towns

MAP 1. Location of Central Brazilian tribes

•

6 • David

Maybury-Lewis

are mutually intelligible. T h e Krikati have no difficulty under­ standing the Kraho, the Sherente the Shavante, the Kaingang the Shokleng, and so on. People speaking languages from different N o r t h e r n Ge groups (for example, the Eastern T i m b i r a and the Kayapo) may understand something of each other's speech, b u t with difficulty. T h e r e is n o m u t u a l intelligibility between North­ ern Ge\ Central Ge, Bororo, and Southern G£, respectively. T h e s e Ge- a n d Bororo-speaking peoples have lived on the Cen­ tral Brazilian highlands for as long as we have any information a b o u t them. T h e i r m o d e r n descendants are scattered over a huge area (see M a p I ) . From the most northeasterly of the N o r t h e r n Ge (the Apaniekra and the Ramkokamekra) to the most south­ westerly (the Suva), these peoples are found spread over a wide arc on a diameter of a b o u t eight h u n d r e d miles. F u r t h e r south, it is about six h u n d r e d miles from the Sherente in the northeast to the Bororo, who live just beyond the Shavante to the southwest. A n d it is a b o u t a thousand mVles from the Apaniekra to the Bororo. T h e Southern G e are even farther afield. T h e y live in the state of Santa Catarina a b o u t eight h u n d r e d miles south of Brasilia. T h e y are not shown on Map 1. Indeed, w e did not include them in o u r original research plan because we thought, erroneously, I am happy to say, that they had died out, or at least that their way of life was extinct. Recent research by Silvio Coelho dos Santos (1973) and Gregory U r b a n (1978) has shown, however, that the Kaingang and the Shokleng are still living in recognizably Ge style. T h e following investigations were actually carried out u n d e r the auspices of the Harvard Central Brazil Project: T

J o a n Bamberger Terence T u r n e r R o b e r t o da Matta J u l i o Cezar Melatti Jean Carter Lave Dolores Newton Christopher Crocker Cecil Cook

Kayap6

(research started in 1962)

Apinay£ Krah6 Krikati

(research started in 1962) (research started in 1962) (research started in 1963)

Bororo Nambikwara

(research started in 1964) (research started in 1966)

Since the time when these research projects were completed and the initial dissertations based on them written, the g r o u p of Cen­ tral Brazilianists has been widely scattered in various universities on two continents. It has thus been very h a r d to produce a joint

Introduction

• 7

volume. The administrative difficulties were heightened by intel­ lectual ones. It has been genuinely difficult to reach the point w here we felt we had a reasonable comparative understanding of the Ge and the Bororo. W e would meet as a g r o u p to discuss these societies, as we did at H a r v a r d in 1968, at Oxford a n d Stutt­ gart (during the Congress of Americanists) in 1968, a n d again at Harvard in 1971, a n d each time the euphoria of the meeting w ould soon give way to solitary analytical doubts about each others' suggestions and interpretations. D u r i n g this period of cogitation we have also benefited from other people's collabora­ tion. Pierre a n d Elli M a r a n d a spent a s u m m e r analyzing G6 myths and contributed their conclusions to o u r discussions. William Crocker, from the Smithsonian Institution, who had done exten­ sive fieldwork among the Apaniekra, was such a frequent con­ tributor to our discussions and meetings that we have come to re­ gard him as a classificatory m e m b e r of the project itself. T h e members of this g r o u p are now so busy with other activi­ ties, including the preparation of their own books on the Ge and Bororo, that it has taken m u c h editorial pressure to get this vol­ u m e ready for press. Synchronization of the submissions has thus been very difficult. In order to avoid an endless process of revision in the light of the latest paper submitted, I have summarily de­ cided at various moments that certain papers were ready and held them for the volume. T h u s the papers by Melatti and da Matta have been ready for some years now. Crocker, Lave, a n d May­ bury-Lewis completed their contributions next. T h e Kayapo ma­ terials, by Bamberger a n d T u r n e r , are the most recent. W i t h their submission, the volume was ready to go to press. T h e papers on the N o r t h e r n G£ peoples are presented first T h e n there is a single paper on the Central Ge, followed by a paper on the Bororo. T h e concluding paper of the volume is a summary one, which comments on the results of our research as a whole. T h e r e are brief notes at the beginning of most papers on the present situation of each people discussed. T h e s e notes are for con­ venient a n d summary reference only. T h e current problems that these societies face in their dealings with outsiders, particularly Brazilians, are being treated in other publications. It should be ex­ plained, however, that the agency of the federal government of Brazil responsible for I n d i a n affairs used to be the SPI, or Servi^o de Prote^ao aos fndios. It is referred to t h r o u g h o u t this volume as the Indian Protection Service. T h e SPI was disbanded in 1967, after a governmental inquiry into scandals within its administra­ tion, and supplanted by the F U N A I , or Funda^ao Nacional do r

r

8 • David

Maybury-Lewis

fndio. T h i s is referred to in this volume as the National I n d i a n Foundation. It is important to r e m e m b e r that the F U N A I often carries on administering the old SPI posts. W h e n we write of an I n d i a n Protection Service post, (or simply of an Indian Serv­ ice p o s t ) , we refer to a post founded by the I n d i a n Protection Service and now operated by the National I n d i a n Foundation. While the names of these government agencies should present little difficulty, there is a problem as to how to distinguish verbally between Indians a n d non-Indians. In Central Brazil the difficulty of defining an I n d i a n is not serious. For the most part, Indians are those very few people who live in Indian villages. T h e r e is a neg­ ligible n u m b e r of people living close to, b u t not in, the villages whose status is ambiguous. T h e rest are not Indian. T h e non-In­ dians are usually referred to by local people, including the Indians themselves, as Cristaos (Christians) or civilizados (civilized peo­ ple) . T h i s is interesting in itself, b u t we did not wish to adopt that ethnocentric usage in this volume. T h e use of the term "whites'' to refer to non-Indians is much rarer in Central Brazil, for obvious reasons. In the population as a whole, skin color varies from light to dark tan, a n d the Indians are not always noticeably darker or redder than their non-Indian neighbors. W e also felt that Nimuendaju's scrupulous designation of non-Indians as neo-Brazilians, leaving the term "Brazilian" to be applied (if at all) to the only " t r u e " Brazilians, the Indians themselves, was a little precious. T h e whole concept of Brazil and Brazilians is, after all, an idea introduced by the settler society a n d one which was q u i t e foreign to the Indians themselves. W e have therefore chosen the terms " I n d i a n " and "Brazilian" as being the least objectionable ones through which to express the opposition between Indians and the rest. O u r research has led to a substantial revision of Central Bra­ zilian ethnography and, we believe, to a deeper understanding of the cultures of the G£- and Bororo-speaking peoples. Yet it must be stressed that it is not the primary aim of o u r book to set the ethnographic record straight. T h i s is being done elsewhere. W e have thus faced a difficult problem of exposition. W e could not assume that our readers were familiar with the data we were discussing, yet we wished to present a series of arguments, n o t a collection of m i n i a t u r e ethnographies. W e have accordingly tried to provide just the a m o u n t of ethnographic information needed to follow the arguments. T h e s e arguments have been profoundly affected by the com­ parative n a t u r e of our project. T h e interpretation of each individ­ ual society has been deepened and modified by a consideration of 2

Introduction

• 9

the data on the others. It would be difficult to reconstitute the stages in this process of understanding, and they would in any case be of little consequence. Yet something needs to be said at the outset concerning our common approach. Initially, we devoted special attention to problems of social or­ ganization, for it was here that the data were especially puzzling. T h e descent system reported for the Apinay£ was anomalous (Maybury-Lewis 1960) . T h e N o r t h e r n Ge were reportedly matrilineal, the Central Ge patrilineal, and the Bororo matrilineal again. Since these societies were so closely related to each other, we wondered why they differed in such a fundamental fashion. Nor was it clear how these descent systems worked or how they were articulated with other institutions such as age-sets, cere­ monial organizations, and moieties, both agamous a n d exogamous. Above all, how was all this related to dual organization? Was the latter merely a matter of moieties or did it go deeper than that? Was it a philosophical idea or an institution that ordered people's daily lives? O r was it both? W e therefore u n d e r t o o k a comprehensive investigation of the cultural categories of each society. W e wanted to understand what their social theories were, how these were reflected in their institu­ tional arrangements, and how all this related to their everyday behavior. W e soon discovered that the features of the Ge-Bororo complex were somewhat different from what had hitherto been supposed. All of these societies traditionally inhabited circular OT semicircular villages whose layouts reflected their view of the world a n d of themselves. T h e y make a sharp distinction between the forum, or central, ceremonial sphere (which was conceptually a male place) and the houses, the peripheral, domestic sphere (which was conceptually female). T h e i r rule of postmarital residence was thus everywhere uxorilocal. O n this c o m m o n base, each society had constructed its own peculiar set of institutional arrangements. T h e Eastern T i m b i r a and the Apinaye attached great importance to ceremonial moieties to which people were affiliated through their names. Kayapo life was focused more on age-grades and men's associations. These w ere also important a m o n g the Central Ge, who had invested them with ceremonial functions, in contrast to their politically oriented descent groups. Finally, the Bororo appeared to have the most thoroughgoing dual organization of them all, with exogamous matrilineal moieties and name-based clans engaged in a constant series of exchanges. T h e varying emphases of the papers in this volume reflect, to some extent, the differences between the societies with which they deal. T h e r e are also differences of opinion regarding the theoretiT

10 • David May bury-Lewis

cal conclusions which can be drawn from o u r exercise in compari­ son, and these will be noted as they occur. Jean Carter Lave, for example, focuses on n a m i n g a m o n g the Krikati. T h e issue of names, naming, and the affiliation to social groups through one's name r u n s through all the discussions of the Ge and Bororo, b u t it is nowhere so salient as a m o n g the Krikati. Lave discovered that it was impossible to make a satisfactory anal­ ysis of Krikati society or of the Krikati kinship system without first understanding the overriding principles of their n a m i n g sys­ tem. H e r paper therefore focuses on Krikatf n a m i n g a n d suggests why this institution has come into such prominence a m o n g them. J u l i o Melatti takes u p the issue of n a m i n g too, b u t first insists that the relationship system of the Kraho cannot be understood until we have grasped the principles of the dualism which per­ meates their society. T h e Kraho distinguish between central, ritual activities, which concern the society as a whole and periph­ eral, instrumental ones, which are held to be of domestic con­ cern. T h e former are performed by groups to which a person is affiliated through a name; the latter by groups whose members are held to be physically related to each other. T h i s opposition, Melatti argues, is expressed through their relationship termi­ nology. Roberto da Matta takes u p a similar theme in his analysis of Apinaye ideology. In his formulation, the Apinay£ distinguish be­ tween two types of kinship relations—"given' relationships of com­ m o n substance, which involve a person in domestic activities, and "constructed" relationships, in which names play a p r o m i n e n t part, and which involve a person in ceremonial activities. After showing how N i m u e n d a j u was mistaken a b o u t their descent sys­ tem a n d thus resolving the "Apinay£ anomaly," da Matta goes on to show how the Apinay^ relationship system, like that of the Kraho, articulates the various domains of Apinay6 culture. H e then considers the comparative implications of this, not only for Central Brazilian societies b u t for other societies whose kinship systems are reportedly of the Crow or O m a h a type. T h e Kayapo, whose social arrangements are outlined in the paper by Joan Bamberger, contrast with the Krikati, Kraho, and Apinay6 in that their age-grade system is still a vital part of their social organization. T h i s gives Kayapo society a rather different cast a n d has markedly affected their own style of dual organiza­ tion. T h i s was ideally and traditionally a matter of the opposition between the two men's houses, which any properly constituted vil­ lage was supposed to have. But in recent times n o such village has actually existed, and this opposition has been transformed into a 1

Introduction

• 11

political clash between the age-grades of m a t u r e men (with chil­ dren) and i m m a t u r e men (still childless). Bamberger focuses on Kayapo politics and discusses the fluidity and divisiveness of their communities in terms of Hirschman's concepts of exit, voice, a n d loyalty. T w o papers by Terence T u r n e r follow. In the first, he outlines a general theory of dual organization which applies, a fortiori, to all the Central Brazilian peoples. H e maintains that, t h r o u g h o u t Central Brazil, men seek to control other m e n by retaining their own daughters after marriage a n d d o m i n a t i n g their in-marrying sons-in-law. Uxorilocality is thus essentially a political instrument, b u t it is also part of a pattern which generates moiety a n d age-class systems. T h e s e serve both as models of and as regulating mech­ anisms for the passage of individuals from their natal to their affinal families. In his second paper, he applies this a r g u m e n t to the Kayapo, a n d shows how their society functions as a feedback system in which these institutions shape each other. I myself take a different view, as my paper on the Central Ge and my concluding essay should make clear. I do not think that the elaborate a n d pervasive dualism to be found a m o n g the Ge and the Bororo is usefully regarded as being modeled on their means of detaching individuals from their natal families. N o r do I feel that uxorilocality is primarily a political matter. In fact, my paper on the Central Ge argues that for them, the only patrilineal societies in Ceneral Brazil, uxorilocality is a distinct embarrass­ ment. T h e y practice it, I believe, for the same symbolic reasons that make it c o m m o n to all the other societies of Central Brazil. My .paper deals, therefore, with how the Central Ge b l u r the sharp distinction between the central forum and the peripheral domes­ tic sphere by their patrilineal descent system, and then proceeds to a general discussion of dual organization in Central Brazil. I also take u p the issue of whether the Sherente can usefully b e said to practice patrilateral cross-cousin marriage of a prescriptive or any other kind and consider the implications of my data, a n d of the arguments of my fellow contributors, for the discussions surround­ ing the Crow and O m a h a type of kinship systems. Finally, Christopher Crocker's paper on the Bororo takes u p vir­ tually all of the issues discussed by the other contributors, for the Bororo appear to ring the changes on all the institutions of the Ge-Bororo complex. T h e y too make a distinction between the physical and social aspects of the individual, one which affects their ideas about descent and consequently the n a t u r e of their de­ scent groups. T h e y too sharply distinguish between the male and the female sphere a n d build elaborate men's houses in the middle

12 • David May bury-Lew is

of their villages where the m e n spend most of their time. Crocker's paper explores their rationale for g r o u p i n g m e n together in a curi­ ous sort of matriliny and then investigates the complicated dialec­ tics of the relationship between moieties and their constituent clans. It is clear that a major concern of all of these papers is the topic which anthropologists loosely refer to as "kinship." T h i s is partly because the idiom of kinship is central to these peoples' own way of t h i n k i n g about social relationships. Yet an investiga­ tion of kinship terminologies also proved to be a useful tool for getting at the thinking b e h i n d their social classifications. W e would argue that the G e and the Bororo belong to that large class of societies whose kinship terminologies cannot be elucidated ex­ cept as social matrices and in relation to other systems of classifi­ cation. It is in this sort of society that the study of kinship is most fruitful; b u t it is also in this sort of socilty that what we study as " k i n s h i p " is furthest removed from the ordinary or folk notions of kinship pertaining in the societies from which most anthro­ pologists come. Hence the paradox which has bedeviled " k i n s h i p " studies: they are most fruitful when they deal with p h e n o m e n a which are least like kinship. I n this book, we have treated " k i n s h i p " as a branch of cultural classification. Each c o n t r i b u t o r has made a systematic effort to understand how social and cultural domains relate to each other in the society u n d e r study. T h e idea is not altogether novel, and interestingly enough Mervyn Meggitt (1972) carried out an analy­ sis along similar lines for the Australians, whose kinship systems have so often been compared with those of the Central Brazilians. Yet this sort of total analysis has for the most part been either fit­ fully or intuitively carried o u t in anthropology. W e have tried here to show what a more systematic application of these ideas can achieve and believe that we have produced significant results. T h e reworking of Central Brazilian ethnology has already been men­ tioned, b u t the implications of our work go beyond Central Brazil. W e have found many of the concepts a n d classifications used by students of " k i n s h i p " to be inadequate or misleading. It follows, if we are right, that o u r results should lead to the modification or discarding of these concepts and to major revisions in the theory of "kinship." Finally, a word about the title of this volume. It was the dual organization of the Central Brazilian peoples which first caught the attention of anthropologists in general. It is to their dual or­ ganization that the contributors to this book r e t u r n again and again. T h e Ge and the Bororo have a binary view of the universe.

Introduction

• 13

T h e y state q u i t e explicitly t h a t t h e i r societies are i m b u e d w i t h oppositions, because o p p o s i t i o n is i m m a n e n t in the n a t u r e of things. T h e essays in this v o l u m e show h o w each society strives to create a h a r m o n i o u s synthesis o u t of t h e a n t i t h e t i c a l ideas, cate­ gories, a n d i n s t i t u t i o n s t h a t c o n s t i t u t e its way of life. Of course everyday life c a n n o t b e so easily o r so neatly c h a n n e l e d . T h e d y n a m i c s of social action in C e n t r a l Brazil p u t severe strains on t h e systems. I n c e r t a i n cases, these strains a p p e a r to b e m o v i n g t h e m o u t of their dualistic m o d e altogether. Yet t h e G e a n d the B o r o r o as we k n e w t h e m w e r e still t r y i n g to live by t h e i r tradi­ tional philosophies. T h e y w e r e still t r y i n g to achieve t h e synthesis of opposites, to create b a l a n c e a n d h a r m o n y by o p p o s i n g institu­ tions. I n t h a t sense, they were truly dialectical societies.

THE KRIKATI

T

HE K R I K A T I (also known as Caracati) are located in the south­ west of the state of Maranhao. They inhabit six small villages in two clusters, one close to the town of Montes Altos and the other around the town of Amarante. These in their turn lie to the east of Imperatriz, through which the Belem-Brasilia highway passes. T h e Krikatf's precise history is difficult to trace since there has been considerable fission and fusion among Eastern Timbira groups in the last hundred years (see Nimuendaju 1946:13-36). In fact, Nimuen­ daju reported that the Krikati were on the verge of extinction in 1930. Yet they have succeeded in reconstituting themselves. In the sixties they numbered between 350 and 400 individuals, but the population has now increased to about 600. They claim to be an amalgam of various Eastern Timbira peoples who were once separate, such as the Ronhugati, the Pihugati, and the Pukobke. T h e Krikati have frequent contact with other Northern Ge peoples, especially the Apaniekra, the Ramkokamekra, the Krah6, and the Apinaye. They also have a great deal to do with the Guajajara, of Tupian linguistic stock, who live intermingled with the local Brazil­ ians, though sometimes retaining their identity as Indians. Above all, the Krikati are in uncomfortably close contact with the Brazilians themselves. They are surrounded by Brazilian backw oodsmen. Small "roads" carry traffic right past their villages and they are only about forty kilometers from the Bel^m-Brasilia highway. It is this proximity rather than the efforts of missionaries or the Indian Protection Service (which did not until recently maintain a post for the tribe) which has been the major force for change in their lives. Although they still maintain themselves by subsistence farming and gathering the wild products of the region, a cash income is becoming increasingly important to them. Yet they have few systematic means of earning money. At the same time, their uneasy relations with the local Brazilians are always threatening to erupt into conflict over land or over the alleged killing of the Brazilians' cattle. r

15

j[ I Cycles and Trends in Krikati Naming Practices Jean Lave

Introduction T H E S O C I A L O R G A N I Z A T I O N of the Krikati Indians is u n u s u a l a m o n g the world's societies because of the kinds of rela­ tionships which are utilized in conveying social i n f o r m a t i o n a b o u t individuals a n d r e c r u i t i n g t h e m to formal c o m m u n i t y w i d e groups. Relationships of k i n s h i p a n d m a r r i a g e organize most do­ mestic g r o u p activities, b u t relationships based on t h e transmis­ sion of personal names are equally i m p o r t a n t . T h e latter are used to recruit m e m b e r s to three different moiety systems a n d n u m e r ­ ous ceremonial societies. T h e y a r e used to t r a n s m i t ceremonial roles, to establish responsibility for ceremonial training, a n d even as t h e basis for establishing an a d d i t i o n a l system of ceremonial relationships, k n o w n as formal friendship relations. Both k i n s h i p a n d n a m i n g relationships h e l p to m a i n t a i n t h e c o n t i n u i t y of Krikati society over time. But the way in which the Krikati repre­ sent this c o n t i n u i t y to themselves is t h r o u g h the name-based trans­ mission of social identities from name-givers to name-receivers. N e i t h e r bilateral kindreds n o r the corporate moiety systems of the Krikati are assigned these functions. N a m i n g a n d k i n s h i p rela­ tionships organize different d o m a i n s of social life: this division of kin-based a n d name-based organizational devices is itself a major feature of Krikati social organization, a n d m u c h of the organiza­ tional significance of each set of principles comes from the con­ trasts a n d complex interrelations between them. Since the system is u n u s u a l , the first task to be u n d e r t a k e n here is to present the rules of n a m e transmission a n d the ways in which they serve as i m p o r t a n t organizational devices for t h e Krikati. T h a t is, I will try to describe how the system works. T h e second p r o b l e m to be discussed is why the Krikati attach greater i m p o r t a n c e to personal n a m e transmission t h a n d o other 16

Cycles and Trends in Krikati Naming Practices • 17

closely related Ge groups. T h e importance of name transmission diminishes, while that of kin-based lineages a n d clans increases, as we move from the N o r t h e r n G e groups south to the Central Ge and Bororo. It is also t r u e that the Krikati place more organiza­ tional emphasis on n a m e transmission than any other g r o u p among the Eastern T i m b i r a . It seems likely that the extraordinary importance of n a m i n g for the Krikati is the result of social change over the last thirty years. T h e deterioration of the traditional ageset system appears to be a key change which led to increasing re­ liance on n a m i n g as the primary basis of ceremonial organization. T h e lapsing age-set system also may have led to the comparatively strong emphasis in the mid-sixties on domestic clusters as the focus of domestic and political activity. Evidence to support these propo­ sitions comes from a close comparison of field data collected among the Krikati in the mid-sixties w ith C u r t Nimuendajii's data on the Eastern T i m b i r a groups, which he gathered in the late thirties. r

Krikati Naming Practices and Their Relations to Other Aspects of Social Organization T h e Krikati relationship d o m a i n is called mekwu. At the most general level, mekwu contrasts the Krikati to strangers, spirits, a n d monsters (kakrit). At a less inclusive level, mekwu applies to all residents of one's own village a n d the closest neighboring village (about two h u n d r e d people) as opposed to residents of other villages. T h i s is the g r o u p within which all members are cus­ tomarily referred to and addressed by relationship terms. At the level of interpersonal relations, the term mekwu may be used to refer to close kin in contrast to individuals or domestic groups with whom the speaker is quarreling. Informants included as mekwu all two h u n d r e d residents of the local Krikati villages when asked specifically about each individual in turn. T h e y also conformed to their own ideals by referring to almost all of the residents by relationship terms. Criteria reported by informants for assigning referential terms were q u i t e varied, including m a n y other relations in addition to genealogical ones: "1 call X 'term Y' because she is my m o t h e r " ; " X is the mother of Y"; "Y is my child"; b u t also, " X gave his names to Y"; " X gave meat to Y in the last ceremony and Y gave rice to X " ; " I call X 'term Z' because X has to look out for Y b u t cannot speak to her— it's because of names"; "X's name-giver is married to Y." In addi­ tion to consanguineal and affinal kin relations, the domain in­ cludes relationships of formal friendship, ceremonial trading part­ nerships, a n d n a m i n g relationships.

18 • Jean Lave

Formal friend relationships are acquired t h r o u g h n a m i n g . Ego's formal friends are the formal friends of his or her name-giver. As t h e formal friends a c q u i r e n e w name-receivers, these, too, become ego's formal friends. F o r m a l friends may be of either sex, b u t opposite-sex formal friendships are m a i n t a i n e d m o r e elaborately t h a n same-sex formal friendships. Formal friends m u s t refrain from using each other's names in reference or address a n d must n o t meet face to face or speak to each other. Sexual relations and m a r r i a g e b e t w e e n formal friends are forbidden. Emphasis is placed on t h e obligation of adults to give r i t u a l protection a n d h e l p to their y o u n g formal friends whenever the latter are threat­ ened with ceremonial violence. Parents pay their children's for­ mal friends for this assistance. Adult formal friends have few obligations to each other except to protest a n d d e m a n d payment from a n y o n e w h o kills a n animal whose ( n a m e a formal friend bears. T h e s e protests are today carried out in a rough j o k i n g man­ ner and n o payments are actually m a d e . Ceremonial t r a d i n g p a r t n e r relations are very narrowly defined, although t r a d i n g p a r t n e r s refer to and address each other at all times only as t r a d i n g partners, kttzvuure. T h i s terminology is established when a woman a n d a m a n exchange cooked garden produce for game in one specific yearly ceremony. It continues so l o n g as they exchange each year. T h e relationship is transmitted from the w o m a n to h e r daughters and, according to a r u l e of parallel transmission, from the m a n to his sons. Each d a u g h t e r trades with a single son of the mother's t r a d i n g partner. I n contrast to ceremonial t r a d i n g p a r t n e r relations, transmission of personal names a n d the relationship established t h r o u g h n a m e transmission are elaborate, complex, a n d of central i m p o r t a n c e in Krlkati social organization. I will now describe these relationships in detail. NAMES AND THE RULES OF N A M E TRANSMISSION

T h e Krikati possess a b o u t eight h u n d r e d personal names. Each Krikati acquires at b i r t h a set of two to fifteen names, which the individual has the right to use d u r i n g his o r her lifetime, a n d to transmit to certain others after reaching a d u l t h o o d . Each person is customarily addressed a n d referred to by one of these names. Ideally, a name-set is p e r m a n e n t , u n a m b i g u o u s , and indivisible. If asked to recite t h e name-set, a Krikati m a n or w o m a n will check with an expert—the person w h o gave the names—to m a k e sure the list is complete a n d correct. In practice, however, persons with the same name-set frequently r e p o r t substantially different sets of names, even after consulting their name-givers. Such dis-

Cycles and Trends in Krlkati Naming Practices • 19

crepancies arise in various ways. O n e or m o r e names may be for­ gotten. A n i c k n a m e may b e included in one individual's name-set b u t n o t be a d d e d to the name-set of other individuals w h o have the same names. Contrary to the rules, two persons sometimes in­ sist o n giving their names to t h e same child, a n d the child eventuallv transmits a c o m b i n a t i o n of names from b o t h sets. A n d sometimes, again contrary to the rules, a name-set is split a n d the two parts transmitted separately. In spite of Hux in the c o n t e n t of name-sets, t h e Krlkati persist in the belief that name-sets a r e i m m u t a b l e . It may b e that the belief goes u n c h a l l e n g e d partly because the act of n a m e trans­ mission is itself sufficient to p r o d u c e precise transmission of par­ ticular ceremonial affiliations, rights, a n d responsibilities. T h e name-receiver is constrained to follow exactly in the name-giver's footsteps. T h e r e f o r e , once t h e individual has names, the names themselves a r e irrelevant, except for their use in identifying indi­ viduals. R e m e m b e r i n g all of one's names is not necessary as long as everyone realizes, t h r o u g h frequent ceremonial activity, with w h o m t h e names are shared. T h e r e is n o rationale for i n c l u d i n g names in name-sets o n the basis of similarity of m e a n i n g . A l t h o u g h most names have mean­ ings, frequently the names of animals a n d plants, I never encoun­ tered a situation in which someone tried to reconstruct o n e n a m e in a name-set from the m e a n i n g of a n o t h e r n a m e in that nameset. Connections between names seem to be inferred from k n o w n connections between people. T h e a p p a r e n t absence of semantic connections between names in a name-set is e x p l a i n a b l e in terms of the process by which newnames are a d d e d to name-sets. New names are added to the system in the form of nicknames referring to u n u s u a l events, usually of a sexual or excretory n a t u r e . T h e s e nicknames start o u t as long, vividly detailed descriptive phrases, b u t within weeks are short­ ened to a few words. Over the next m o n t h s or years the n a m e will be shortened even further to a word, or a string of syllables formed from the e n d of one word a n d the b e g i n n i n g of t h e next. In this way, and also because people eventually forget the events or per­ sonal peculiarities t h e names memorialize, nicknames lose their specific m e a n i n g a n d are incorporated into the name-set. Some names belong to men, some to w o m e n . A m a n ' s name-set occasionally transmitted to a w o m a n or vice versa, b u t this is considered to b e only a h o l d i n g operation, a n d the recipient is expected to pass the names to someone of a p p r o p r i a t e sex at the earliest o p p o r t u n i t y .

20 • Jean Lave

RELATIONS BETWEEN NAME-GIVERS AND NAME-RECEIVERS

T h e Krikati say that they differ from wild animals a n d wild In­ dians because they have personal names. T h e y are very much con­ cerned that a newborn child should receive names immediately, a n d the ceremony is held just after the umbilical cord is cut. T h e child is provided with a social identity at approximately the same time he or she becomes a separate physical being. T h e child acquires all of his or her names at once. T h e ideal name-giver for a girl is her father's sister and for a boy his mother's brother. Ideally, firstborn son and firstborn daughter are paired from birth as future name-set exchangers. Each is expected to n a m e his or her partner's first child of the appropriate sex. Secondb o r n siblings are similarly paired as potential name-exchangers, a n d so on. In addition to n a m i n g each otherls children, sibling pairs give each other ceremonial services in return for gifts of food. T h e name-giver performs the ceremonial service and the namereceiver's parent pays for the service w ith food. Name-givers, both m e n and women, act as mentors to their name-receivers, guide them in ceremonial performances, teach them lore and songs, and teach them to r u n well. A major respon­ sibility is supervising the e n d u r a n c e training of boys so that they will be good log-carrying relay racers w hen they grow u p . T h e name-giver's obligations end when the name-receiver marries. A male name-giver goes h u n t i n g for his name-receiver and rep­ resents h i m or performs other services for him in ceremonies, and is repaid in meat pies. T h e name-giver must agree before a young m a n may marry. Unlike the father, who cannot formally contrib­ ute to family deliberations on the matter, the name-giver is for­ mally consulted. It is only after marriage that the relationship between giver a n d receiver becomes relatively egalitarian. A female name-giver takes her name-receiver to the stream to bathe early each m o r n i n g until the child can walk. (A male namegiver finds a close kinswoman to perform this duty for his namereceiver.) Each day the child's mother gives the bather a small gift of food as payment for the service. T o mark the end of the bath­ ing sessions, the parents give the name-giver a very large meat pie. Although a woman does n o t go h u n t i n g on behalf of her namereceiver, she performs special ceremonial services for the child and is repaid in meat pies. She prepares a cord girdle which the girl wears from the time she is a b o u t ten years old until her first pregnancy. It is not clear how much influence a woman has on her name-receiver's choice of marriage partner. T h e relationship ber

T

1

Cycles and Trends in Krikati Naming Practices • 21

tween female name-giver a n d her name-receiver is probably never as severely asymmetrical as that between males. T h e Krlkatf do not distinguish in speech between the formal mentor (the direct giver of the names) and the other older per­ sons who hold the same name-set. All are "name-givers" to ego. But distinctions between principal and subsidiary name-givers are certainly m a d e in action. W h i l e a serious name-giving m e n t o r cannot refuse to carry out his or her ceremonial responsibilities to the name-receiver, peripheral name-givers may elect to ignore the n a m i n g tie by addressing the child by n a m e and treating him or her like other children. In this way, a distant name-giver simply does not acknowledge what is at most not a very d e m a n d i n g rela­ tionship on either side. Also, while the Krikati are p r o u d of hav­ ing many name-receivers, "there are l i m i t s / ' they feel. It is ac­ ceptable, if o n e already has many name-receivers, to decide at the birth of a particular child not to treat him or her as a namereceiver. N o one can give names to more than one living child of any given person. T h u s although each person has a name-exchanger arranged from early childhood (the opposite-sex sibling nearest in birth order) , name-exchange relations with other siblings and distant relatives must be established in order to give names to more than one child and, reciprocally, in order to find names for one's own children. It is q u i t e common for a person to give names to one child of each opposite-sex sibling. It follows that each mem­ ber of a set of siblings receives names from a different sibling of his or her opposite-sex parent. Hence, the ceremonial affiliations of siblings are q u i t e heterogeneous. If one person changes his or her ceremonial affiliations, the change does n o t affect the ceremonial affiliations of siblings or any other close kin. W h i l e the affiliations within a g r o u p of siblings are unrelated, each set of female siblings should hold exactly the same inventory of name-sets and ceremonial affiliations as their father's sisters, taken as a set. And a set of male siblings should have the same inventory as their mother's brothers, taken as a set. T h i s reproduction of sibling-set characteristics across generations rarely works out, for a variety of practical reasons. But the Krikati have a special terminology for labeling same-sex sibling sets and are explicitly aware that their n a m i n g rules logically should lead to cross-generation identity between sets of same-sex siblings. NAMING RELATIONSHIP TERMINOLOGY

T h e social identity between name-giver and name-receiver is made additionally clear in n a m i n g relationship terminology (see

22 • Jean Lave

comaiyront

FIGURE

la. Naming terminology for male ego

Figures la and l b ) . It is possible to generate a »et of rewrite rules underlying the structure of n a m i n g relations (cf. Lounsbury 1964). T h e r e are four rules, which are the same for m e n and women except for a difference in the order of application. Rules 1 and 2 are a merging r u l e and a half-sibling rule. Rules 3 and 4 are Crow I I I a n d O m a h a III skewing rules, except that Crow I I I is applied before O m a h a I I I for males and in reverse order for females (see T a b l e 1). Egos of each sex use the rules which most accurately reflect the name-transmission relations in which they are involved. W h e t h e r Crow or O m a h a , the rules reflect a single principle: that name-giver a n d name-receiver share a single social identity. W h e r e the rules generate alternative usages (ikachui/ inchigrunto and ikamtele / inchungrunto), the Krikati express in­ difference as to which pair of reciprocals is chosen, and they occur in free variation. T h a t is, ego can emphasize the equation of him­ self with his name-giver (his mother's brother) a n d hence refer to his mother's brother's d a u g h t e r as ikachui, "daughter," o r h e can take the point of view that his mother's brother's daughter is

comaiyront

FIGURE

lb. Naming terminology for female ego

Cycles and Trends in Krikati Naming Practices • 23

TABLE 1. Relationship terminology rules 1. Collateral merging rule

$zc-> $c

£BC-» $C 2. Half-sibling rule FC -> sibling MC -> sibling 3, First skewing rule Male ego: Crow rule III $B . . . -* 9S . . . Corollary: . . . $Z . . . £M 4. Second skewing rule Male ego: Omaha rule III $Z . . . -+ $T> . . . Corollary: . . . $B -* . . . £ F

Female ego: Omaha rule III $Z . . . -> . .. Corollary: . . . 9 B - » . . . $ F Female ego: Crow rule III 9B . . . $S . . . Corollary: . . . $Z -» . . . 3M

his mother's name-receiver, socially identified with his mother, and refer to her as inchigrunto, "mother's name-receiver," or just inchi, "mother," for short. At the same time, from the mother's brother's daughter's point of view, the male is both her father's name-receiver, inchungrunto, a n d her name-giver's son, ikamtele. W h e n two Krikati are related by more than one type of rela­ tionship, n a m i n g relations take precedence over all b u t ceremonial trading partnerships a n d formal friendships as criteria for assign­ ing customary terms of reference. Even close kin use n a m i n g terms in preference to kinship terms. T h u s ego will refer to his sister as tuire, the same kin term he applies to his father's sister, if the father's sister n a m e d his sister. H e keeps the two kinswomen sep­ arate, referentially, by a p p e n d i n g their names to the term. T h e r e is also a narrower, b u t more formal, usage involving name-sharers and their parents-in-law. A m a n and his father-in-law address a n d refer to each other's name-receivers by what is ordi­ narily an honorific suffix attached to in-law terms, -ye. T h i s prac­ tice does not, however, extend to reciprocal use between the re­ spective name-receivers. Principles of n a m e transmission are used as a vehicle for ex­ panding the range of kin relations. Based on the social identity of name-giver and name-receiver and the broad interpretation of "giver" a n d "receiver" (which includes total strangers, if they have the same n a m e s ) , the rule is to call the kinsmen of one's name-sharer by the same terms one uses for one's kin and, re­ ciprocally, to treat the name-sharers of one's kinsmen as kinsmen. N a m i n g terminology is preferred in cases where there is un-

24 • Jean Lave

certainty over whether a clearly specifiable relationship exists be­ tween two individuals. T w o strangers or distantly related village members may compare the names of their close relatives u n t i l they find a correspondence a n d then treat each other accordingly. If a stranger comes to live in the village, a Krikati will transmit his or her name-set to the newcomer. Strangers thus quickly acquire names, a n d one relationship—that of name-giver/name-receiver. T h e y are in a position slowly to acquire n a m i n g a n d kin relation­ ships. But the emphasis on the immediate n a m i n g of strangers, and the practice of addressing and referring to them by name, is an apt reflection of their marginal social position. T h e use of names contrasts strongly with the use of n a m i n g ter­ minology in reference a n d address. N a m i n g terminology is used to communicate the existence of relationships which have been established on the basis of actual or putative bestowal of names. Names are consistently used in reference and address to label po­ sitions which have not yet been combined into specific relation­ ships. Therefore, they very often are applied to categories of people who are marginal to the society. T h e bestowal of names on strangers is one example. A n o t h e r in­ volves sexual relatedness. A person should marry or have sexual relations only with persons referred to by name, that is, those with whom n o relationship exists. Ideally, sexual ties should not be formed with those referred to by kin terms or n a m i n g terms. Since there is also great emphasis on referring to all members of the community by kin terms, to express a high degree of relatedness, the two rules frequently come into conflict. T o cope with the problem, people often switch from kin terminology to the use of personal names as a means of indicating sexual interest. If two people have sexual relations, however, they are then prohibited from addressing or referring to each other by n a m e , a n d only relational terms may be used. It is customary at this time for the m a n and woman to adopt teknonymous designations for each other. Each refers to the other as " p a r e n t of X , " where X stands for alter's name-exchanger, since that is the n a m e which will be given to the first child of each sex, should the two m a r r y . 2

NAME-HOLDING GROUPS

Since there is n o formal term in the Krikati language for a g r o u p of people who share the same name-set, I have coined the phrase "name-holding g r o u p . " T h e members of such A g r o u p have a great deal in common. T h e y share the same name*, sex, moiety affiliations, formal friends, a n d ceremonial roles. But they do not stand in uniform kin relationships with each other, since names

Cycles and Trends in Krikati Naming Practices • 25

may be transmitted through two or more unrelated chains of kin. Indeed, members of a name-holding g r o u p need not even share tribal affiliation, since strangers are integrated into Krikati social life by acquiring Krikati names. A m e m b e r of a name-holding g r o u p is considered to be a namegiver to all those members who are younger than himself or her­ self. Conversely, he or she is treated as a name-receiver by all older members of the group. T h e Krikati say that whenever names are transmitted, all persons with those names take part in the giving. W h e t h e r each person is aware of the transaction at the time is ir­ relevant. T h e same principle is used to establish appropriate be­ havior and terminology between strangers who happen to have the same names. In these cases, it is appropriate for the elder of the two to take the role of name-giver and for the younger to adopt the role of name-receiver. N o term exists for two people w ho share the same names a n d who are approximately equal in age a n d status. All relationships between holders of the same names are structured in terms of asymmetrical relations between a giver and a receiver. T h e r e is one principle which results in the internal structuring of name-holding groups, and it too is concerned with seniority. Name-holding groups are split into older and younger halves. Which half a m e m b e r belongs to depends on whether the individ­ ual has acquired at least one name-receiver. T h e younger half con­ sists of those who have received b u t not yet transmitted names. Members of the older half have all transmitted their names at least once. T h e Krikati d o not transmit their names until after they complete initiation. T h e notion of reciprocity between pairs of name-givers is so strong that people rarely bestow names o n a child until they are starting their own family. It is after initiation a n d marriage, therefore, that the strong respect owed by young name-receivers to the older m e m b e r s of their name-holding g r o u p abates considerably, as young members begin to bestow their ow n names, and to b r i n g new name-receivers into the name-holding group. Certain major ceremonial roles are occupied successively by each m e m b e r of a name-holding group. Most ceremonial roles may n o t be performed simultaneously by two members of the group, b u t must b e performed by each in order, according to age. T h e parents of the name-receiver are supposed to sponsor the par­ ticular ceremony twice in order to dispatch the child's responsibil­ ity, once when the child is very young a n d a second time as she or he nears puberty. Decisions as to who will sponsor the ceremony in a given year involve discussions about which n e w b o r n nameT

T

r

f

26 • Jean Lave

holders have n o t had the ceremony performed for them yet, and whether there are children who are a b o u t to finish initiation and who thus must fulfill their ceremonial responsibilities in the near future. A child who has completed both performances achieves the honorary title of " r i p e , " a n d o n e who fails to do so is " u n r i p e . " " R i p e " persons are felt to be dignified peace-makers of excep­ tional merit. U n d e r ideal conditions the division of name-holding groups into r i p e and u n r i p e corresponds to the division between name-givers and those w h o have n o t yet transmitted their names. Moiety Systems Name-holding groups are the basic b u i l d i n g blocks for three sets of moieties: K u i g a t i y e / H a r u n g a t i y e , K a p i / K a i k u l a , a n d plaza moieties a n d their constituent plaza groups (Figure 2 ) . T h e moiety affiliation of a name-holding g r o u p in one moiety system provides no clue to its affiliation in either of the others. Further, the rare Krlkati individuals who have changed moiety affiliations argue that switching their affiliation in one moiety set requires n o comparable change in the other moiety sets. T h e s e independent, p e r m a n e n t moiety systems have different characteristics, a n d are associated with different ceremonies. 1. K u i g a t i y e / H a r u n g a t i y e moieties are former age moieties, n o longer differentiated into age sets. T h e s e moieties are asso­ ciated respectively with east a n d west, high and low, ostrich (a b i r d which stays on the g r o u n d a n d can scarcely fly) a n d turtle (an animal of b o t h land a n d w a t e r ) . T h e s e moieties traditionally performed the marriage ceremonies, which are now obsolete, and they still perform ceremonies of m o u r n i n g . It is this set of affilia­ tions which the Krikati think of first when they discuss moieties. Kuigatiye and H a r u n g a t i y e each have chiefs, chosen by all of the men of the village. T h e y serve as a pair, a n d when o n e of the two quits or dies, a new pair is chosen. T h e chiefs direct ceremonies, and harangue the village early in the m o r n i n g a b o u t appropriate behavior a n d the day's activities. T h e j o b also involves providing hospitality for visitors to the village and adjudicating disputes between individuals or domestic clusters. (A domestic cluster is a g r o u p of houses located next to each other on the village circle, in which live the members of an uxorilocal extended family.) 2. K a p i / K a i k u l a moieties function d u r i n g several ceremonies and also d u r i n g fish-drugging expeditions. Behavior a n d personal likes and dislikes are felt to be influenced by this system. These moieties are associated with rainy season a n d dry season activities, respectively. T h e year is divided between the two moieties, which

(J) Kuigatiyc/Harungatiye moiety system

Harungatiyc

(2) Kuigatiye

Kapi/Kaikula moiety system

Kaikula

Kapi

Plaza moiety system

Plaza moiety 1

Plaza moiety 2

(S)

(4) (1) Moiety systems (2) Moieties (5)

(3) Plaza g r o u p s (4) Name-holding g r o u p s (5) M e m b e r s of name-holding g r o u p s (name-givers)

(6)

X l X2

>'l V2

Zl Z2

W l W2

Ul

(6) M e m b e r s of name-holding g r o u p s (name-receivers) Formal friendship relations

FIGURE 2. Schematic representation for KrTkati ceremonial organization

28 • Jean Lave

stand for rainy season and dry season, flood a n d drought, night a n d day, darkness a n d sunlight, cold and heat. T h e year is parti­ tioned between times when there is plenty of food b u t relative so­ cial isolation (rainy season) and times of scarcity b u t heightened social interaction (dry season). T h e forest (cold, wet, dark) is opposed to the savannah (dry, warm, and s u n n y ) , and directions (east and u p , versus west and down) are also opposed. T h e char­ acteristics of the two moieties are associated with the d i u r n a l cycle as well. T h u s , Kaikula prefer cold, dark nights, and perform cere­ monial activities at night, while Kapi prefer hot, sunny days and are ceremonially active d u r i n g the day. T h e Kapi chief is chosen from a m o n g the Kaikula m e n a n d vice versa. T h e s e chiefs are supposed to be overseers of the activities of the moiety which they govern. Each is appointed by the surviv­ ing chief when his opposite n u m b e r dies. T h e Kapi chief governs d u r i n g the day, the Kaikula chief at night. Kaikula, while being of the west, is also associated w ith the front of the body a n d the periphery of the village. Kapi is associated with the east and the back of the body, and with the plaza. T h e association of K a p i / Kaikula with the plaza/house circle opposition, or more generally w ith the c e n t e r / p e r i p h e r y dichotomy found elsewhere a m o n g the Ge, is much attenuated. It does n o t have a major place in the symbolism associated with this moiety system, a n d is not expressed in ceremonial action. 3. Plaza groups and plaza-group moieties are restricted to males. W h i l e the plaza moieties are u n n a m e d a n d practically inopera­ tive, the eight plaza groups play a considerable role in ceremonies. T h e y are paired across moieties for log-racing and paired in a dif­ ferent way for the exchange of meat pies (Figure 3 ) . Each plaza g r o u p has a meeting place in the plaza, special paint insignia, a cry, and a leader. Plaza groups are active d u r i n g the initiation ceremony a n d the ponhupru (maize) festival. Each g r o u p is headed by its eldest m e m b e r , w hose duties consist mostly of guiding ceremonial h u n t i n g expeditions. Like the other moiety systems, plaza-group moieties are composed of name-holding groups. But in this moiety system, the plaza groups constitute an additional level of organization, so that name-holding groups are aggregated into the eight plaza groups which in turn form the two moieties. T h e r e are also two temporary moiety systems which function d u r i n g specific ceremonies. In contrast to the p e r m a n e n t moieties, these systems are organized on principles other than n a m i n g . For the ceremony of zuu'tukrekrc, the population is divided into those who have more than one child (living or dead) and those with n o r

T

r

Cycles and Trends in Krikati Naming Practices • 29 Northwest Moiety Meat-pie trading

Southeast Moiety Log-racing opponents

Meat-pie trading

FIGURE 3. Pairing of plaza groups for ceremonies children or only one child. For the wu'tu ceremony, the popula­ tion is divided into kloahugatiye (water monsters, literally, kloaburiti palm Aw-leaves, from which the water monster masks are made) and ropgatiye (jaguars). T h e moieties are associated with west and east, respectively, and subdivided into several n a m e d so­ cieties. Men enter the society of their choice at the b e g i n n i n g of the festival. W o m e n enter the female counterpart of the society chosen by one of their formal friends and subsequently receive payment for having "followed" the m a n . T h e system of ceremonial organizations may appear confusingly complicated. But in practice there is n o ambiguity as to which affiliation is governing an individual's activities at any given time since n o two moiety systems ever function simultaneously. T h e segregation of ceremonial activities from domestic activities in­ sures that a person's m e m b e r s h i p in a particular moiety does not conflict with relationships outside the ceremonial context. T H E PUZZLE OF CORPORATE NAME-HOLDING GROUPS

Name-holding groups are sets of social positions. Each position !s labeled by a n a m e from a m o n g the c o m m o n set of names. At the same time, the members of each name-holding g r o u p are linked in ordered relations based on age and on a distinction be­ tween those w h o have given their names a n d those who have not. In fact, they are corporate groups with a changing membership, but with a continuous existence so long as the name-set survives.

SO • Jean Lave

T h e Krikati, however, do not conceptualize name-holding groups as corporate entities. Rather, they talk in terms of relations between individual name-givers and name-receivers. T h e y de­ scribe formal friend relationships in terms of links between in­ dividual name-givers a n d name-receivers, although it would be equally accurate to talk about a tie of formal friendship linking two name-holding groups. It is puzzling that the Krikati do not make use of the corporate properties of name-holding groups in characterizing their own society. However, this dyadic, rather than corporate, model of n a m i n g relationships makes better sense when viewed in the light of a n o t h e r problem: how to keep moiety systems clearly independent from one another, given that there is m o r e than one moiety system in the society, a n d given that these moiety systems perform very similar manifest functions. T h e Krikati have three moiety systems which recruit members through name transmission and which carry out primarily ceremonial functions. T h e actual integrity of name-holding groups across moieties (all members of a nameholding g r o u p have the same three moiety affiliations) drama­ tizes the interconnectedness of the three systems in ways that con­ tradict more important messages of separateness encoded in the complex metasystem of three moiety systems. (See Lave 1976 for an analysis of relations among the three systems.) T h e message of continuity is m u c h diluted if it is reduced to the level of special dyadic relations where an individual simply follows his or her name-giver into the three moieties with which that individual is affiliated. It is these considerations which have given shape to the local model of n a m i n g relationships among the Krikati. NAMING As THE BASIS OF SOCIAL CONTINUITY THROUGH T I M E

T h e Krikati are unusual in using n a m i n g relationships rather than kin ties as the basis for establishing social continuity through time. T h i s is brought about by downplaying the long-term con­ tinuity of intergenerational kin ties while emphasizing these same features in intergenerational n a m i n g relationships. T h u s genealogical amnesia is encouraged and required by the Krikati. Relations between the living and the dead are severed as quickly as possible. T h e living are enjoined to forget the dead. T a l k i n g or t h i n k i n g a b o u t them is viewed as dangerous, leading to the possibility that the dead person will wish to c o n t i n u e the relationship with the living by bringing a b o u t his or her death. T h e constant reduction of genealogical information makes pos­ sible the useful fiction that domestic clusters consist of separate kin groups. T h i s may be its major function. But it also eliminates

Cycles and Trends in Krikati Natning Practices • 31

the possibility of using genealogical ties to signify the broad order­ ing of time across generations. In contrast, continuity is a major preoccupation in the transmis­ sion of names. Name-giver and name-receiver are felt to be a single social personage. T h e name-receiver is said to take the place of, to become, the name-giver. Indeed, the name-receiver steps i n t o all the same ceremonial positions and roles as the name-giver, u n d e r the name-giver's tutelage. T h e term for name-receiver, comaiyrontj can be roughly glossed as "shoots that grow u p from a dying tree," a n d the Krikati say that name-receivers grow u p a r o u n d the name-giver just like new shoots. Passing on one's social identity is so important that if an individual dies w i t h o u t living name-re­ ceivers, it is felt to be appropriate to break o n e of the basic rules of name transmission: that parents never n a m e their own chil­ dren. T h e names of the deceased will be given by adult siblings of the dead person to their own children in order to preserve the names and the "place" of the deceased. T h i s sense of urgency and concern a b o u t the transmission of names reflects the crucial function of n a m i n g in relating past, present, and future. It is conveyed in other ways as well. Pairing opposite-sex siblings as future name-exchangers establishes the next incumbents of a pair of social identities a generation before they physically arrive on the scene. T h e Krikati do not d e p e n d on transmitting the social identity of someone long deceased, b u t rather, transmit identity from living adults to newborn babies. Name-givers are physically present when the transaction occurs and have substantial obligations d u r i n g the years in which the name-receiver is becoming a full-fledged replacement. T h e indi­ vidual transmits his or her social position to another almost im­ mediately u p o n coming into full possession of it. T h i s partially accounts for the Krikati emphasis o n the asymmetry of relations between name-giver a n d name-receiver, on who came first and who came after, for the distinction is crucial in m a i n t a i n i n g continuity through time. T h e contrast between n a m i n g a n d kin relationships has a n u m ­ ber of implications. T h e r e is a clear opposition between physical reproduction of the body, involving parenthood, kin relations, and domestic g r o u p activities, a n d the reproduction of social iden­ tity, involving n a m e transmission, n a m e relations, a n d ceremonial activities. For the Krikati, marriage is the last step in the process °f initiation into adulthood. Men a n d woman cannot marry or ^ansrnit their names u n t i l they complete initiation. T h e end of Vitiation thus marks the point at which it becomes possible to niarry and reproduce physically, on the one hand, and give names,

32 • Jean Lave

thus b r i n g i n g a b o u t social r e p r o d u c t i o n , on the o t h e r h a n d . It is m a r k e d by elaborate r i t u a l , n o t only because it symbolizes crucial events in individual lives, b u t because it represents the p o i n t at which a n e w g r o u p of i n d i v i d u a l s a r e ready to transfer their names, thus i n s u r i n g once again c o n t i n u i t y t h r o u g h t i m e for the c o m m u n i t y as a whole. I n i t i a t i o n / m a r r i a g e is also t h e p o i n t at which a m a n moves from his natal domestic g r o u p to j o i n his wife in establishing a new domestic g r o u p w i t h i n h e r n a t a l domestic cluster. F r o m the p o i n t of view of a n individual, this is a one-directional moye in the course of the life cycle. B u t from a sociocentric p o i n t of view, any individual is involved from b i r t h o n w a r d in o n e p r i m a r y set a n d o n e "shadow" set of relations, b o t h of which c o n t i n u e simultane­ ously t h r o u g h o u t life. T h i s occurs because names a n d social iden­ tities alternate between d o m e s t i c groups t h r o u g h the process of n a m e transmission. T h u s a c h i l d , as a name-receiver, acquires with his or h e r names a " s p o u s e " a n d " c h i l d r e n " (the name-giver's spouse a n d c h i l d r e n ) . T h i s is the reverse of the situation in e g o s natal domestic g r o u p , w h e r e h e or she is, of course, a child to his or h e r parents. W h e n a m a n marries, h e moves from his natal domestic cluster to reside in h i s wife's domestic cluster, taking his n a m e s with h i m . His s i s t m r e m a i n in their natal domestic clus­ ter, establishing new n u c l e a r families within it as they marry. T h e r e f o r e , when a m a n t r a n s m i t s his names to his sister's son, the names, a n d his social identity, are r e t u r n e d to the domestic g r o u p from which they came. A n d w h e n the w o m a n transmits h e r names, they leave her domestic cluster, creating new ties outside her do­ mestic g r o u p . T h e Krikati w ay of m a i n t a i n i n g c o n t i n u i t y through t i m e sug­ gests an image to m e of r e p e a t e d short cycles, r a t h e r than some sort of linear progression. Withini a lifetime a person m a t u r e s a n d his or h e r name-set gradually shifts from b e l o n g i n g to a child, w h o is a name-receiver only, to b e l o n g i n g to an a d u l t , w h o is capable of name-giving. N a m e s are t h e n transmitted, so that once again they b e l o n g to y o u n g children, a n d begin the cycle again. It is a short cycle in the sense that it goes only from b i r t h to m a t u r i t y . It is easy to imagine longer cycles in which names could be transmitted to a child from a person in o l d age, or soon after a death, or after a n u m b e r of generations, b u t the Krikati d o n o t use the system that way. Men's name-sets c a n be seen as cycling in space as well as in time; cycling, because they are taken to an affinal domestic cluster when the m a n marrys., b u t quickly r e t u r n e d to c h i l d r e n of his natal domestic cluster. T h e cycles are short for men's names. r

Cycles and Trends in Krikati Naming Practices • 33

For women's names the spatial cycle is not so clear-cut; it is only in a general sense that w o m e n ' s names could be said to circulate a m o n g domestic clusters. B u t t h e cycling t h r o u g h time is not sexlinked, a n d applies to all Krikati. T h e Loss of the Age-Set System I believe that some of the i m p o r t a n c e of n a m i n g in Krikati so­ cial organization can be e x p l a i n e d in terms of the loss of an ageset organization similar to that described for the T i m b i r a in 1935 by N i m u e n d a j i i (1946). Since t h e r e is n o description of the Kri­ kati age-set system in 1935, I t u r n to Nimuendajii's description of the T i m b i r a system as the closest a p p r o x i m a t i o n available. T h e T i m b i r a age-set system in 1935 consisted of four active agesets, divided equally between eastern a n d western moieties (Kuigatiye a n d H a r u n g a t i y e ) . T h e surviving m e m b e r s of older age-sets were c o m b i n e d in a councilors' "set" that m e t in the center of the plaza, undifferentiated by moiety m e m b e r s h i p . T h e absence of both age-set a n d moiety distinctions w i t h i n this g r o u p underscores that the councilors' position e m b o d i e d concern for the welfare of the c o m m u n i t y as a whole. T h e initiation period for an age-set s p a n n e d a b o u t ten years. N e w age-sets alternated between the eastern a n d western moieties, each moiety a c q u i r i n g a new age-set every twenty years. Each new age-set u s u r p e d the m e e t i n g place in t h e plaza of the n e x t older age-set in its moiety, the oldest age-set b e i n g p r o m o t e d to join the councilors in the center of the plaza. T h e activities engaged in by t h e age-sets in 1935 were primarily ceremonial. But an age-set d i d b u i l d houses for those of its mem­ bers heavily involved in p u b l i c service, a n d t h e m e m b e r s helped each o t h e r in harvesting, especially the rice c r o p ( N i m u e n d a j i i 1946:94). W h i l e N i m u e n d a j i i n o w h e r e discusses the process of village fission, W i l l i a m Crocker (1966) says that a m o n g the R a m kokamekra in the 1960s factions crystallized a r o u n d age-set leaders. N i m u e n d a j u (1946:93) emphasizes t h e u n u s u a l a u t h o r i t y of ageset leaders in the mid-thirties, although age-set activities were dwindling by that time: 3

4

It is they who actually govern the age classes, being possibly the only func­ tionaries who literally issue orders among the Canella, a task for which they are trained from the beginning. Only they have the right to summon their class fellows, who are obliged to obey the call and may not assemble without their leaders. Anyone who has dealings with a class, including the chiefs, must turn to its mamkyeti (leader). These leaders are subject only

34 • Jean Lave

to the council . . . Theoretically, the leaders are equals, but in practice, the abler of the two soon gains ascendancy without, however, completely eclipsing his colleague. In contrast to the chiefs, the mamkyeti exercise no authority whatever beyond their age class. In former times the mamkyeti led their class in war and in the chase, though always aided by one or several older men. Since warfare belongs completely to the past and communal hunting has dwindled in importance the present significance of these officers is slight—apart from ceremonial. T o d a y the Krikati have moieties called Kuigatiye and H a r u n g a tiye, the same names given by N i m u e n d a j u for both age-set moi­ eties and exogamous moieties. T h e Krikati moieties have cere­ monial functions, b u t are not divided into age-sets and d o not participate in the organization of domestic activities. T h e Krikati do not r e m e m b e r the names, meeting places, or members of ages£ts, b u t there is one g r o u p of m e n in their late forties who agree that they were initiated together in a series of ceremonies. T h e y are all Harungatiye. T h e probability is very small that all these m e n would belong to the same moiety if age had not been the principal recruiting mechanism for moiety membership. T h e r e is one younger g r o u p , men between the ages of thirty and forty, who went through at least one initiation ceremony together. T h e y ap­ pear to be divided a b o u t evenly between Kuigatiye and H a r u n g a ­ tiye. I would guess that it was some time d u r i n g the period be­ tween the initiation of the thirty- to forty-year-old g r o u p and the forty- to fifty-year-old g r o u p that age-sets ceased to function a n d m e m b e r s h i p in Kuigatiye and Harungatiye moieties became name-based. Ceremonies that a m o n g the T i m b i r a in 1935 were connected with age-sets persisted a m o n g the Krikati in 1965. O n e of these is an initiation ceremony. Krikati children first enter seclusion for a two- to three-month period between the ages of eight and twelve, a n d may enter a second time, d e p e n d i n g on their own inclinations and those of their parents. T h e ceremonies are held whenever there are two or more children whose parents wish to sponsor their initiation, and the whole c o m m u n i t y takes part. Each initiate has a companion, a special friend, who goes through the initiation at the same time, b u t large groups are no longer initiated together. T h e ceremony, ikrere, corresponds to the pepye age-set initiation ceremony described by N i m u e n d a j u (1946:170—171, 179-201) and is identical in many details. As a m o n g the T i m b i r a in 1935, chil­ dren are beaten at ceremonial times and sent regularly to bathe at dawn. Krikati men now in their thirties r e m e m b e r being hazed extensively until marriage. T h e y entered seclusion in relatively

Cycles and Trends in Krxkali Naming Practices • 35

large groups (six to t e n ) . T h e y also ended the seclusion period as a g r o u p a n d maintained various sex and food restrictions for some time afterward. Differences in the ceremonial m a r k i n g of life cycle events be­ tween 1935 and 1965 are mostly a matter of degree. T h e series of six or eight initiation ceremonies through which an age-set passed in 1935 has collapsed into a series of two, or at the most four, cere­ monies, performed for one or more pairs of children. T h e in­ formal aspects of initiation, including hazing and training for logracing, c o n t i n u e in slightly attenuated form u n t i l marriage. As a m o n g the T i m b i r a in 1935, the Kuigatiye and Harungatiye moiety system is the appropriate one for traditional marriage ceremonies a m o n g the Krikati, on the rare occasions when they are performed. In the past, formal marriage ceremonies marked the e n d of the initiation ceremonies, b u t did not necessarily signal the beginning of actual coresidence and cohabitation. T o d a y the mar­ riage ceremony is, more often than not, simply a public announce­ m e n t of actual coresidence and may occur some years after the in­ itiation rites. Ceremonies for the dead are, however, carried out in consistent a n d elaborate style by the K u i g a t i y e / H a r u n g a t i y e moi­ eties, again corresponding to the T i m b i r a practice in 1935. As in 1935, the major ceremonial activity of the year occurs in May and October, at transition periods between the rainy a n d dry seasons, m a r k i n g changes in the yearly cycle. Each ceremony is divided into two parts, a b e g i n n i n g and an end, both involving intensive ceremonial activity over a period of a few days. T h e two periods of a given ceremony involve q u i t e similar, repetitive events. T h e time in between the opening a n d closing is relatively quiet, a sort of diffuse "holiday season" with little specific cere­ monial activity. T h e closing festivities for one ceremony occur close together in time with the opening log race of the next cere­ mony, so that all the yearly cycle is given structure by the cycle of ceremonies. T h e economic a n d political activities associated with age-sets amor/g the T i m b i r a of 1935 are organized by the Krikati today according to domestic cluster affiliations. Planting a n d harvesting are coordinated a m o n g nuclear families within each domestic cluster. T h e ceremonies the T i m b i r a performed in 1935 to open the corn harvest season officially are still performed a m o n g the Krikati, b u t are not described as harvest ceremonies a n d do not affect harvesting. Ceremonial h u n t i n g trips are carried out by moieties—whichever pair is active for the particular ceremony. T h e m e n of each moiety, and their wives a n d children, travel to­ gether, make g r o u p decisions a b o u t which area to h u n t in, and so

36 • Jean Lave

on, although the game is not contributed to a c o m m o n pot. Raid­ ing, which a m o n g the T i m b i r a in 1935 was conducted by age-sets, has not occurred among the Krikati for many years. O n e of the oldest m e n in the c o m m u n i t y remembers a single raid when he was a young man. For the Krikati, domestic clusters are the basic units in communityw ide quarrels, and one or m o r e domestic clus­ ters may leave the village as a u n i t in the case of a c o m m u n i t y split. T h e r e are no such strong leadership positions as N i m u e n d a j u describes for the T i m b i r a in Krikati social organization today. T h e job of Kuigatiye or H a r u n g a t i y e chief has onerous responsi­ bilities and few rewards, and does not imply rights or powers to affect anybody's behavior except t h r o u g h moral persuasion. At least six living m e n have been chief at some time, a n d the Krikati insist that they are happy to q u i t such a difficult a n d thankless j o b . O t h e r "chiefs" have strictly ceremonial leadership roles, exercised d u r i n g the ceremonial periods when their moiety system is in ef­ fect. Leadership today, for groups smaller than the c o m m u n i t y as a whole, resides informally with a few m e n a n d women with force­ ful personalities who have considerable influence in their domestic clusters. T

Historical Trends: Organizational Changes Resulting from Loss of the Age-Set System Evidence has just been presented to support the proposition that the Krika-.f once had an age-set system a n d have lost it. T h e change was not an isolated o n e and m u c h of the rest of the paper is con­ cerned with tracing some of the implications of the loss of the ageset system. First of all, as the age-set system became less important, individ­ ualization of age-based relationships increased. T h i s came about through the increasing importance of name-transmission relations. N a m e transmission was not a new recruiting device for the Kri­ kati of 1965. N o r did it spring into existence only as the age-set system disappeared. Formerly, they coexisted. A m o n g the T i m b i r a in 1935, names were the basis for m e m b e r s h i p in plaza groups (and hence plaza-group moieties), rainy season/dry season moie­ ties, and men's societies. Some formal friends were acquired through name transmission. However, m e m b e r s h i p in age-set moieties (Kuigatiye and Harungatiye) was based on age, member­ ship in the tamhuk honorary organization was based on other principles, and formal friends could be acquired in ways other than naming. For the Krikati today, essentially all ceremonial or­ ganization, including K u i g a t i y e / H a r u n g a t i y e moiety m e m b e r s h i p

Cycles and Trends in Krikati Naming Practices • 37

and formal friendship relations, is based on naming. T h e only honorary positions (for example, r i p e / u n r i p e ) are based on per­ formance of special ceremonial roles, and the right to perform these roles derives only from possession of a particular name-set. T h e referential terminology for name-transmission relation­ ships does appear to be a recent development. Not a single such term is reported by N i m u e n d a j u (1946) . N i m u e n d a j u would probably have reported such terms if they h a d been in use in the late thirties, for h e would have heard them used constantly in ad­ dress; and as a stranger being incorporated into the community, he would have been taught them when (1946:110) he received names and n a m i n g relatives. T o some extent, the change in relative importance of n a m i n g and age-set activities is a quantitative one. Between 1935 a n d 1965 there appears to have been a decrease in the a m o u n t of time spent in organized ceremonial activities. W h e r e once log races a n d ageset meetings and other activities occurred every day, and there is daily involvement only just before, and during, periods of intense ceremonial activity. Age was the primary organizing principle for the age-set system. Age continues to be an organizing principle. But the units or­ ganized according to relative age have u n d e r g o n e a drastic change. N a m e transmission and the organizations based on naming, espe­ cially name-holding groups, might be characterized as individual­ ized versions of the age-set organization. A comparison of the ma­ jor features of age-sets and name-holding groups will help to make the point clear. I n both the system of name-holding groups and the age-set sys­ tem, a man changes status with age. Among the T i m b i r a in 1935, an age-set shifted from initiate to mature-man status, then to coun­ cilor status, with accompanying changes in meeting places, and so on. T h e Krikati in 1965 move from the status of name-receiver to that of name-giver. Both age-sets a n d name-holding groups have members of one sex only. Age-sets are ordered by age; individuals within a name-holding g r o u p are ordered by age. In both, the initialed are distinguished from the uninitiated—a distinction be­ tween sets in the one case and between individuals in the other. Initiation is a prerequisite for full participation in age-set activi­ ties, and completion of initiation confers the right to transmit names a m o n g the Krikati today. A Krikati child completes initia­ tion with a special companion, and they aid and support each other for the rest of their lives, b u t they are n o t caught u p w ith a set of peers in an extensive and absorbing set of group activities. Both age-sets and name-holding groups recruit members through 5

6

r

38 • Jean Lave

initiation ceremonies, b u t initiation was carried out for an ageset in 1935 and for individuals in 1965. In addition to individualization, there is a second consistent difference between 1935 a n d 1965 in the structure of interrelation­ ships. It is partially an outcome of the individualization. Since 1935 there has been increasing participation of women in cere­ monial activities, and the initiation of girls has become r o u t i n e . Krikati parents speak about sponsoring initiation ceremonies for their daughters with the same concern as for their sons. T h i s con­ trasts with Nimuendaju's description of age-set initiation, in w hich only two representative girls were secluded with the boys. Krikati women, as a group, now " o w n " ceremonies, including zou'tu, the most important ceremony of the Krikati. Groups which perform in this ceremony are rather similar to the T i m b i r a men's societies in 1935, b u t now there are both men's and women's so­ cieties, a n d societies of each sex have two opposite-sex associates. T h e ceremony for the dead is now performed by men for a de­ ceased male a n d by women for a deceased female. W h e n a m a n dies, the m e n of the c o m m u n i t y r u n with logs and the women sing and prepare a ceremonial meal. W h e n a woman dies, women r u n while men sing a n d prepare food. N o n e of these activities in which Krikati women participate are reported for the T i m b i r a of 1935 or for other groups of Ge. Since b o t h m e n and women received names, belonged to nameholding groups, and inherited ceremonial roles along with names in 1935, the increasing importance of names should have led to increasing involvement of women in ceremonial activities. N o other principle of T i m b i r a social organization exhibited this sex­ ual symmetry. W e can speculate about the processes which b r o u g h t about this increasing involvement. N i m u e n d a j u reports that T i m b i r a women joined one age-set moiety or the other for occasional racing purposes. T h i s connection could have been used as a new criterion for transmitting moiety m e m b e r s h i p along with names. Krikati women have m e m b e r s h i p in K u i g a t i y e / H a r u n g a tiye today on the basis of their names. Similarly, women might have acquired ceremonial roles by serving as associates to men's ceremonial groups. Girls were chosen as ivu'tu associates of the age-set moieties a m o n g the T i m b i r a in 1935, a n d today wu'tu is a special g i r l s ceremonial role. W h e n a girl had once acted a special ceremonial role, it would be consistent with the rules of n a m e transmission for other women to gain access to the role along with a name-set. Another way in which ceremonial roles might have come to be more homogeneously distributed in the population is through the occasional transmission, on a temporary T

Cycles and Trends in Krikati Naming Practices • 39

basis, of male names to females. U n d e r some circumstances (the chances are increased by epidemics and general d e p o p u l a t i o n ) , a girl might thus come to hold a special ceremonial role which once belonged exclusively to males. Today, for example, both males and females inherit the ceremonial role ru'rut, although their name-sets are said to b e different from each other. T h e third major historical change resulting from the end of age-sets was the strengthening of the political a n d economic posi­ tion of the domestic cluster. At the present time, it is the heads of domestic clusters w ho provide leadership for village factions. In the past, quarrels severe enough to split the village appear to have crystallized a r o u n d age-set leadership. It is not at all clear what process has led to this long-term change in the organization of factions. But in the late thirties, there was almost certainly some complex interaction of kin ties and age-set ties which influenced individual and family decisions about whom to live with when village splits occurred. T h e r e must have been conflict for a m a n in deciding whether to move his fam­ ily with his age-set, u n d e r its leader, or to follow the decision­ makers of his wife's domestic cluster. In general, the stronger men's allegiances to their age-sets, the m o r e likely that domestic clusters w ould split u p in the case of factional division. T h u s , a major implication of the demise of age-sets should be a strengthen­ ing of domestic clusters and, perhaps, sharper boundaries between them as they become the primary units in community disputes. Nimuendajii did not seem impressed by the integration of T i m bira domestic clusters in 1935 (1946:84, 161) : r

r

The members of a house community (domestic clusters) . . . are ob­ viously closer to one another than to outsiders at large, but they fail to constitute a definite economic unit among the Timbira. The plantations of the residents are ordinarily in juxtaposition, but every married woman has her own plots and the house community as such has no property apart from the domicile. However, the entire group will safeguard the individual's rights against any irjustice from outside . . . In this sense the safety of the individual rests'on his affiliation with the major family . . . On the other hand, the house community plays not the slightest role in ceremonial. A n d later: " T h e following legal principles emerge from the con­ crete instances: (1) Both the culprit a n d the aggrieved party are aided by the extended families; (2) the chiefs and councillors take pains to adjust quarrels as soon as possible; (3) for this pur­ pose they use the services either of specially gifted persons or of

40 • Jean Lave

the individuals linked by specially close ties to the parties in­ volved." While it is especially difficult to make comparative judgments when differences are a matter of degree, comparison of N i m u e n dajii's description with the data to be presented next seems to bear out the conclusion that Kiikatf domestic groups today are m o r e integrated a n d solidary units than such groups were among the T i m b i r a of 1935. Nimuendajii's assessment of the T i m b i r a is that in 1935 domes­ tic clusters were rather trivial a n d u n i m p o r t a n t social units. H e reaches this conclusion on the basis of two lines of argument. First, T i m b i r a house communities did not own property. Second, house communities took no part in ceremonial activity. But prop­ erty ownership may not be necessary for strong g r o u p cohesion. Krikati domestic clusters achieve strong, economic cohesion through coordination of activities and exchange of goods, neither of which requires g r o u p ownership of property. Further, the fact that domestic clusters did not act as units in ceremonial activity in 1935 does n o t distinguish them from the Krikati today. But the reason why domestic clusters did not take part in ceremonies is precisely that they were domestic groups. It has m u c h more to d o with separating domestic from ritual occasions a n d activities than it does with the cohesion of domestic units. R a t h e r than measure the Krikati today against N i m u e n d a j u ' s criteria, it may be useful to examine new evidence a b o u t the po­ litical and economic cohesion of Krikati domestic clusters in 1965. I will discuss first the economic activities of domestic clusters, then their political role. Finally, I will use patterns of kin-terminology usage to show the highly crystallized domestic cluster organization of Krikati villages. Domestic clusters today are economic units of great importance. Economic solidarity is demonstrated in the coordination of daily activities a n d the exchange of food. Each of the major gardens of the Krikati is occupied by a single domestic cluster. Garden work parties change in purpose and composition t h r o u g h o u t the grow­ ing cycle b u t always involve coordination of domestic cluster members' activities. Men cut the trees and undergrowth on the garden site. Each m a n (with the help of his u n m a r r i e d sons) cuts his own plot. After the whole g r o u p fires the cleared area, he builds a portion of the common fence within the agreed-upon domestic cluster garden. T h e women clear the small brush and prepare the g r o u n d for planting. Clearing and planting work parties involve all of the women of a domestic cluster, who go to the garden early in the m o r n i n g to work, each on her own plot,

Cycles and Trends in Krikati Naming Practices • 41

occasionally stopping to rest a n d gossip. Harvesting involves both men a n d women, using the same work pattern. Coordination of daily activities within a domestic cluster applies to gathering and h u n t i n g expeditions, to the new forays to do wage work, and to traveling, as well as to gardening. Frequently a g r o u p of m e n within a domestic cluster go h u n t i n g together. Each man takes h o m e what h e has killed, to be cleaned, divided, a n d distrib­ uted by his wife. W o m e n of a domestic cluster will set out to­ gether on a gathering expedition, each b r i n g i n g h o m e what she has personally collected. T h e r e is essentially n o coordination or cooperation across do­ mestic cluster boundaries. Each cluster acts independently. People enter arcs of the house circle other than their own only when mak­ ing specific visits to relatives. Similarly, no one enters the garden of any other cluster unless a m e m b e r invites h i m or her. W i t h i n the domestic cluster, nuclear families are linked in fre­ q u e n t and substantial exchanges of food, gossip, and other items. W h i l e productive activities are carried out through parallel, in­ dividual efforts of domestic cluster members, over half the food brought into the village by the person who labored to produce it is consumed bv other members of the domestic cluster. Each house has a small path leading away from the village. T h i s path prob­ ably minimizes the distance from the house to the nearest conceal­ ing cover. Even so, it is practically impossible for anyone to reach home carrying game or other food without being observed by other members of the domestic cluster. Immediately, a small child or other m e m b e r appears from each nuclear family in the cluster to d e m a n d a share. T h e family is lucky to save enough for a single meal, b u t they have reciprocal rights to the fruits of their fellow residents' labors. Domestic clusters d o not " o w n " garden land or houses, b u t in both cases they have use rights. Nuclear families in a domestic cluster make their garden plots together and build adjacent houses. So long as they c o n t i n u e to use them, other people do not have the right to do so. If a garden is abandoned while still usable, other (domestic cluster members have first rights to utilize it. W h e n a garden is n o longer plantable (after two years' u s e ) , it must lie fallow until regrowth has occurred (at least ten years), after which it is available to anyone who wishes to use it. Use rights in houses are more complicated. A m a n builds a house a n d he a n d his wife a n d children live in it. O t h e r nuclear families within the domestic cluster may join them after negotiat­ ing with the builder. If divorce or death removes the m a n from the house, the other residents continue to live there. If a house

42 • Jean Lave

were to be a b a n d o n e d by a family, other nuclear families in the same domestic cluster could move into it or utilize the house posts to build another; or (and this is the only way a n o n m e m b e r of the domestic cluster could utilize the house) a relative of the deceased —his sister, for example—might appropriate the house posts to build a house in her own domestic cluster arc. Since a domestic cluster very rarely abandons use of a house, a n d there is never any reason why anyone would want to move i n t o a village arc of which she or he was not a member, there d o n o t exist situations which would provide proof that domestic clusters have use b u t not ownership rights to houses. I believe, however, that it is relation­ ship a n d residence, rather than "ownership," that is involved. Domestic clusters are also political units today. Factional splits occur along domestic cluster lines. T h e two domestic clusters of which Sao Gregorio was composed in 1964 had separated by 1967. O n e domestic cluster still lived in Sao Gregorio, while the other lived near its garden. In a similar village split in Governador in 1964, one domestic cluster moved to a new site. Manipulations of the residence rule c o n t r i b u t e to the impression that domestic clusters are political units. But before describing variant uses of the rule, it will be helpful to present the normative rules of resi­ dence. A cluster comes to occupy an arc of the village circle as daugh­ ters grow u p and marry. W h e n the parents' house becomes too crowded, young couples receive permission to b u i l d separate dwellings to the right of the parental house. T h e youngest daugh­ ter resides in the parental household indefinitely. W h e n she and h e r husband have established a family, they request formal per­ mission to b u i l d a separate dwelling. O n this occasion, they take on the status of head of household regardless of the parents' reply to the request: either the young couple obtain permission to move out a n d build their own house, or permission is refused. I n the latter case, the young couple remain a n d the parents t u r n manage­ m e n t of the household over to them, b e c o m i n g "children to their children," as the Krikati say. I n practice, the residence rules are m o r e complex. T h e Krikati appear to arrange their houses where possible so that adult m e n who are heads of households live between their affinal and consanguineal domestic clusters. T h e houses of these m e n are transi­ tion points where two domestic clusters overlap, and members of both clusters are likely to treat the transition house as part of their domain. T h e m e n who were in the position of living in proximity to both their natal and affinal domestic clusters in 1965 were all ma-

Cycles and Trends in Krikati Naming Practices • 43

ture, with children approaching adulthood. Clearly, it is impos­ sible for a majority of m e n to arrange their residence in such a conveniently ambiguous way. In general, a man has little choice when young b u t to follow the uxorilocal residence rule or to be divorced, b u t his options increase with age a n d with the develop­ m e n t of his family of procreation. Older males seem to be able to place their households so as to align themselves with the politically strongest domestic cluster while m a i n t a i n i n g ties to the other clus­ ter as well. T h e s e individuals tend to play key roles as mediators in disputes between domestic clusters and to gain prestige from and the support of both. T h e fact that considerations of political ad­ vantage may be taken into account in locating houses on the vil­ lage circle supports the notion that domestic clusters have become increasingly important as political units. O t h e r indications of the increased strength of the domestic cluster may perhaps be seen in the a p p a r e n t decrease in the range of application of kin-relationship terminology a n d the absence of restriction to the domestic cluster of certain activities that a m o n g the T i m b i r a were public events. It seems reasonable that as the domestic cluster becomes the basis for factions it becomes less ap­ propriate for individuals to emphasize distant or uncertain kin­ ship ties as a m e t h o d for increasing the solidarity of their position in the community as a w hole. T h e Krikati comment frequently that young people today are r u d e and ignorant because they n o longer refer to everyone in the village by a kin term. I have data from three men on their referential use of kin terms for all mem­ bers of the community. T w o of the men w ere in their thirties, one about fifteen years of age. T h e fifteen-year-old h a d completed in­ itiation, was married, and like the thirty-year-olds was considered to be an adult by other members of the community. In many cases, the youngest m a n did not know how he ought to refer to people, while neither of the older m e n hesitated over the task at all. T h e two older m e n had n o customary term of reference or address for two and eleven members of the community, respectively; the y o u n e m a n claimed no knowledge for sixty-nine.) T h e simplest explanation is that in the fifteen to twenty years separating him from the older men, the range of effective application has, in fact, shrunk. r

r

r

Some activities w hich involved the T i m b i r a age-sets in domestic affairs in 1935 are carried out exclusively within domestic clusters among the Krfkatf of 1965. Marriage is such an occasion. Nametransmission ceremonies are also private now, while N i m u e n d a j u describes public name-bestowing ceremonies for the T i m b i r a of 1935. T h e T i m b i r a rituals of public reentry into full social par-

44 • Jean Lave

ticipation following childbirth, severe illness, m o u r n i n g , and long journeys are all absent. T h e age-set system integrated some ceremonial functions with some domestic ones. More important, it provided a strongly or­ ganized focus for men's activities which cut across domestic groups a n d m a d e it nearly impossible for domestic g r o u p loyalties to serve as the basis for either communitywide factional disputes or, ulti­ mately, the dispersion of the community. But the end of the ageset system did not lead to the disintegration of Krikati society. Principles of organization which had played a m i n o r role in earlier times were pressed into service, or perhaps expanded to fill an organizational vacuum, leading to a new version of the old system. T h a t this organizational change involved a new balance between kin-based a n d name-based relations is in n o way u n i q u e . In fact, ,the Ge societies all express different solutions to the problem of balancing interfamily relations in terms of the relative importance assigned to kin a n d n a m i n g relationships (see Chapter 5 a n d 6 and Lave, 1975, 1976). T h e new integration, in which individual ties of n a m i n g re­ placed age-based, highly solidary groups, seems to me a weaker, m o r e brittle system than the earlier form. It is unlikely, in the face of continued pressures from Brazilian settlers on the land, the re­ sources, a n d the culture of the Indians, that they will be able to repeat again the feat of b u i l d i n g a new integration out of the re­ mains of the present system. B u t as this volume demonstrates, the possibilities for c o m b i n i n g and r e c o m b i n i n g the structural features of Ge social organization are enormous. It is to be hoped that I, like N i m u e n d a j u in the late thirties, am over pessimistic a b o u t the generative powers of Krikati social organization.

THE KRAHO

T

HE K R A H 6 LIVE in the state of Goias on a reservation with an area of almost 3,200 square kilometers near the small town of Itacaja. They have been in peaceful contact with the Brazilians since 1809. In this region, the settlers live by subsistence agriculture and cattle rearing. However, their very presence has cut down the supply of game animals, on which the Indians formerly relied. Consequently, Indians sometimes kill steers belonging to the local cattlemen, and this produces ill-feeling between them and the Brazilians. This was the issue that led the cattlemen to attack two Krah6 villages in 1940 and kill at least twenty-three Indians. As a result the federal government adopted a series of measures for the protection of the Krah6. It pun­ ished the three ranchers responsible for the attack, demarcated a reser­ vation for the tribe in 1944, and established an Indian Protection Service post on it. In 1962-63 there were 519 Indians living in the villages on the reservation and 43 Indians (or people of Indian descent) living on the reservation but outside of the villages. By 1971 this population had increased to 586 in the villages and at least 49 on the reservation but outside the villages. There are Indians living in the Kraho villages today who are de­ scended from Sherente, Apaniekra, Apinaye, and Krenkateye. The sole Krenkateye village in the state of Maranhao was destroyed by settlers in 1913, and the survivors of the massacre scattered, some coming to live with the Kraho. The Krah6 nowadays visit the villages of the ApinayS, Apaniekra, and Sherente. It seems that their contacts with other Timbira peoples tend to strengthen their Indian traditions, par­ ticularly as regards songs and rituals. By contrast, their contacts with the Sherente tend to influence them to adopt Brazilian ways. Currently, the Krah6 practice subsistence agriculture, but this is not sufficient to feed them all year round. They gather the wild fruits of the region and also gather fruit from trees planted at the Indian Protec tiorf Service post or from trees abandoned by ranchers in the region. Hunting was once very important but is now restricted to the killing of small game, since the larger animals are vanishing. Some Krah6 are nowadays trying to rear cattle.

45

The Relationship System of the Kraho Julio Gezar Melatti Translated from the Portuguese by David

Maybury-Lewis

IN T H I S P A P E R I describe a n d analyze the relationship system of the Kraho using the data which I myself collected in the field. 1 shall try to show that in order to understand this relation­ ship system it is not enough to relate it to family organization, residential arrangements, or matrimonial choices. T h e system also shows a set of relationships w hich are ritual in nature, particularly the dualism which permeates most of Krah6 institutions. T h i s dualism is not limited to theories of oppositions between elements taken tw o at a time. One of its most notable features is that every opposition between two elements is opposed by another opposi­ tion w hich negates the first one. W e thus find an opposition of oppositions. Consequently, before proceeding to an examination of the re­ lationship terminology a n d the behavior connected with it I must briefly describe Krah sister's son succession than to the father-in-law —> son-in-law succession is an ideological factor which is reflected in the terminology. It indicates that ritual rela­ tions are accorded greater importance than economic ones. Mean­ while, it is established beyond d o u b t that name transmission does affect the terminology, so that female name transmission actually gives it an O m a h a character as used by some individuals. But while the transfer of female names proves that n a m e transmission affects relationship terminology, it also creates a problem. If the rules for the transfer of male a n d female names are the inverse of each other, why is the terminology so clearly of the Crow type? O n e answer to this question would be to suggest that the trans­ fer of male names was more important than that of female names and that female n a m e transmission was simply the inverse reflec­ tion of male n a m e transmission. T h i s hypothesis receives support from the fact that n a m e transmission is of great importance in fill­ ing ritual roles a n d d e t e r m i n i n g the m e m b e r s h i p of ritual groups and that m e n are clearly p r e d o m i n a n t in ritual activities. Further­ more, we are not dealing here with the only case of such inversion. We have already noted the inversion in the rules of marriage pre­ stations. So the Crow terminology would therefore reflect the rule of male n a m e transference. O n the other hand, personal names are closely related to the rule of residence. W h e n a man leaves his maternal household to take u p residence with his wife, he leaves b e h i n d part of his social personality, incarnated in his sister's son, to w h o m he preferen­ tially transmits his name. Moreover, h e takes with him to his newplace of residence part of the social personality of his sister, whose name is given to his own daughter. I n this way, the transfer of names acts like a compensation for the transfer of residence. Given the preferential rules for n a m e transmission, certain male names

78 • Julio Cezar Melatti

tend to a c c u m u l a t e in certain domestic groups or residential seg­ ments, while female names, which follow m e n as they move from o n e residence to a n o t h e r , tend to be spread t h r o u g h o u t K r a h o society. Perhaps the c o n c e n t r a t i o n of these male names tends to re­ inforce the individuality of these segments and to emphasize their c o n t i n u i t y via m o t h e r ' s b r o t h e r -> sister's son succession, which in t u r n contributes to the greater i m p o r t a n c e of m e n ' s names i n the domestic sphere. T o sum u p the a r g u m e n t : t h e residential segment is n o t only i m p o r t a n t in the r e g u l a t i o n of m a r r i a g e , since it tends to b e exogamous, b u t it is also the stable nucleus of a n individual's g r o u p of relatives. T h e rules for the transmission of m e n ' s names result in their t e n d i n g to a c c u m u l a t e in given residential segments. Since personal names are crucial in K r a h o r i t u a l a n d their trans­ mission also influences the use of the relationship terminology, it is only to be expected that m e n ' s names, the rules for whose trans­ fer correspond m o r e closely with the other rules of K r a h o society, should be the ones which most strongly influence the o r d e r i n g of the terminology itself. T h i s analysis of the K r a h o relationship system seems to have shown that it is n o t only related to their marriage rules, their resi­ dence rules, a n d their domestic organization. T o u n d e r s t a n d the system, o n e m u s t also take into account the dualism which i m b u e s all K r a h o institutions. T h u s the transmission of personal names appears to b e a nega­ tion of uxorilocality. T h e man's n a m e stays in his sister's house­ hold, while his sister's names goes with h i m to the household of his wife. T h i s is d u e to the fact that every K r a h 6 has two aspects: body a n d n a m e . Body a n d n a m e form yet a n o t h e r opposition which is reflected in the terminology a n d in the behavior associated with it, dis­ tinguishing relatives connected with procreation from relatives connected with n o m i n a t i o n . H e r e n o m i n a t i o n appears almost as the m i r r o r image of procreation. W h i l e h u s b a n d a n d wife pro­ create the bodies of their children, b r o t h e r a n d sister transmit the social personalities which the c h i l d r e n incarnate. It is as if n a m e transmission became a negation of the incest p r o h i b i t i o n . T h e relationship system also shows the K r a h o p e n c h a n t for op­ posing oppositions. T h e occasions for virilocality, the i n d e m n i t y paid for the sexual initiation of a boy, the fact that ipre is t h e re­ ciprocal of b o t h itxwiye a n d ipiaydye, the limited O m a h a char­ acteristics of their relationship terminology resulting from female

The Relationship

System of the Kraho • 79

name transmission, all contradict the distinction between male and female and insist on the equality of the two sexes. Perhaps the relationship designated by the terms hopin a n d hopintxwdi should also be considered a case of opposing opposi­ tions. It is, as we have seen, associated with consanguinity (pro­ hibition of incest, extreme solidarity) and also with affinity (ob­ ligation to repay goods and services and aggression, albeit indirect) . Perhaps this relationship has the effect of opposing the affirmation that consanguines are different from affines by another which insists that they are the same. But what is the rationale for these oppositions a m o n g opposi­ tions which mark the entire Kraho social system a n d their relation­ ship system too? Every opposition between two elements is also in some way a comparison of them. Naturally, if we are dealing with distinct elements, there is some difference between them. But it is only elements which have something in c o m m o n which are compared. For the Kraho, then, the opposition of oppositions is a way of stating that when two distinct elements are compared, they must have something in common. T h e oppositions of oppositions also seem to have some effect on Kraho behavior. Since they coun­ teract the absolute n a t u r e of social rules, they are in all prob­ ability the reason for the fact that Kraho behavior is n o t at all rigidly determined by those same rules.

THE APINAYE

I

N 1971 there were about 350 Apinaye living in two villages near Tocantinopolis in the far north of the state of Goids, and there are now over four hundred. The first contacts between the Apinay£ and Portuguese settlers date from the eighteenth century, but it was not until the late nineteenth century that there was much pressure on the Indians. At that time, a wave of immigrants who lived by cattle rearing and the gathering of wild products reached the Tocantins River. Since these immigrants did not produce anything of great value to the outside world, their pene­ tration into the area was relatively weak and the Apinaye" managed to come to terms with it. T h u s modern Apinaye have godparents among the local settlers and vice versa, even though there may be tension be­ tween them, usually as a result of the disappearance of cattle, which is systematically blamed on the Indians. Apart from their contacts with the backwoodsmen of the region and with the citizens of Tocantinbpolis (who do their best to ignore their inconvenient presence in the area), the Apinaye' have regular contact with the Krikati and with the Krah6. Since 1970, their traditional territories have been occupied by cattle ranchers and settlers who earn their living by gathering babassu nuts. Meanwhile, the work on the Trans-Amazon Highway has led the Apinay£ to await the legal demarcation of their lands with understandable anxiety. The Indian Protection Service maintains a post for the Apinay6 and has recently been encouraging them to manufacture handicrafts for sale to tourists. The FUNAI arranges to distribute these to the cities in the south of Brazil. Current conflicts between the Apinaye: and the settlers in the region revolve around the issue of land. T h e Indians have been severely pres­ sured of late by the massive occupation of their territories and the al­ most total disappearance of game. They now live by engaging in subsistence agriculture and gathering wild products of the region, espe­ cially babassu nuts, which bring them a cash income. They are handi­ capped by having no Brazilian representative to speak for them at the level of the municipio and the state. T h e Indian Protection Service is a federal institution and thus represents them at the less immediate national level. A casual visitor to modern Apinaye villages might think that these people had totally adopted the life-style of the local Brazilians. This is

81

82 •

TheApinayS

not so, however. In spite of the forces which have been pushing the Apinaye* in the direction of Brazilian society for over a century, they still maintain their traditional way of life, especially as regards religion and their system of social interrelationships.

The Apinaye Relationship System: Terminology and Ideology 1

Roberto da Matta Translated from the Portuguese by /. Christopher and David Maybury-Lewis

Crocker

Introduction T H E A P I N A Y E M A D E their a p p e a r a n c e in a n t h r o p o ­ logical theory in 1937 when C u r t N i m u e n d a j i i a n d R o b e r t H . Lowie, in an article entitled " T h e D u a l Organizations of the R a m k o k a m e k r a of N o r t h e r n Brazil," m e n t i o n e d this W e s t e r n T i m b i r a t r i b e . T h e Apinay£ were referred to as a g r o u p which d i d n o t practice the sororate, in the context of a discussion where Lowie c o m p a r e d various forms of m a r r i a g e with his usual preci­ sion (1937:579). T h e inclusion of the Apinaye in a work devoted to the R a m k o k a m e k r a appears accidental, b u t it is i m p o r t a n t to note that, as early as the thirties, Lowie was using the available e t h n o g r a p h i c data o n t h e Central Brazilian I n d i a n s to c o m p a r e certain aspects of their social organization. H e paid special atten­ tion to two p r o b l e m s which, from t h e n on, w o u l d b e f u n d a m e n t a l issues in all discussions of the Ge. I refer to the dualism expressed in the symbolic a n d ritual order of all these tribes a n d , para­ doxically, the lack of any prescription or even preference for mar­ riage between cross-cousins. Lowie, f u r t h e r m o r e , was the first to n o t e the structural simi­ larity b e t w e e n R a m k o k a m e k r a k i n s h i p terminology a n d the ter­ minologies of certain N o r t h A m e r i c a n tribes (Crow , Choctaw, a n d H o p i ) , a n d to seek forms of preferential m a r r i a g e which m i g h t indicate h o w the dualism expressed in R a m k o k a m e k r a r i t u a l was i m p l e m e n t e d in their social organization. T w o years after this work, C u r t N i m u e n d a j i i published his book The Apinaye (1939), a n d in 1940 Lowie explicitly took u p a crucial aspect of the e t h n o g r a p h y by stating: " T h e four mar­ riage-regulating kiye of the Apinay£ d o n o t conform to any type 2

T

83

84 • Roberto da Matta

of organization hitherto described. Descent differs according to sex, sons following the father, daughters the mother. T h e kiyi are thus neither clans n o r Australoid sections, though there is the pseudo-Australoid r u l e that an individual may marry into only one of the three other kiyi" H e concludes, somew hat perversely, "Incidentally, once more to confound the diffusionists, whence does this curious institution come? So long as it is known from just one tribe, shall we n o t simply treat it as a creature of Apinaye culture?" (1940:428). T h i s was the origin of "the Apinay6 anomaly," After 1940, various anthropologists tried to interpret this system of four m a t r i m o n i a l "groups." Jules H e n r y called attention to inter-kiye endogamy. T h i s resulted from the descent rule which separated the sexes and a marriage rule specifying that a m a n of Kiye A had to choose a spouse from kiye B, a m a n of B, from kiye C, and so on, u n t i l the circle was closed with a m a n of D taking a wife from kiye A. T h e result, as H e n r y perceived, was four other "groups," now formed by the men of kiye A and the women of kiye B, m e n of B a n d women of C, m e n of C and women of D, and, finally, men of D and women of A (Henry 1940; after May­ bury-Lewis 1960). In 1942, Kroeber used the same material, call­ ing attention to the fact that the two-by-two g r o u p i n g of the four kiyi produced a system of latent moieties (cf. Kroeber 1942; after Maybury-Lewis 1960). In 1949, tw o classic w orks in the study of kinship systems ap­ peared, a n d both dealt with the Apinay^ and their curious mar­ riage system. George Peter Murdock characterized the Apinay£ as a society possessing a cyclical marriage system with matrilineal descent, but, in a most confusing note, suggested an evolutionary sequence for this tribe (Murdock 1949:242-243, 332). Claude L^vi-Strauss in his Les structures elementaires de la parente used the "kiye marriage system" to illustrate the relativity of concepts of exogamy a n d endogamy (1949:61, 2 8 7 ) . L£vi-Strauss made another attempt in 1952 (reprinted in 1963), w hen, in a pioneer­ ing article, he once again used the Apinaye to demonstrate that actually the dualism of the Ge and Bororo was m u c h more appar­ ent than real. T h e ethnologists who had studied these societies, argued Levi-Strauss, took the dualistic model because it was the natives' own model. It was only among the Apinay£ that this ap­ parent dualism could be seen in its proper perspective without the importance that the natives themselves lent it. H e therefore used the Apinaye marriage system to confirm his analysis of Bororo social organization, which, he argued, was really based on the endogamy of a system of three superimposed and hierarchir

r

r

T

The Apinay4 Relationship

System • 85

cally arranged groups. Since the kiye system led equally to a latent endogamy, L£vi-Strauss commented, " T o regard the Bororo as an endogamous society is so startling that we should hesitate even to consider this possibility had n o t an analogous conclusion already been drawn tor the Apinay£ by three different authors working independently with documents collected by N i m u e n d a j u " (1963: 125). T h e Apinay£ anomaly seemed to have been resolved by L6viStrauss, b u t in 1960 David Maybury-Lewis raised the problem anew a n d attempted a total, detailed analysis of the Apinay^ sys­ tem. In his article, Maybury-Lewis took a direction opposite to that of Levi-Strauss a n d sought to demonstrate, t h r o u g h statistical reasoning and a study of the relationship terminology, that the kiye were a secondary institution in the Apinaye social order, that they did n o t regulate marriage, and, furthermore, that the best hypothesis to be applied to the Apinay^ material was that this so­ ciety operated within a two-section system. After Maybury-Lewis's work, three other attempts were m a d e to resolve the Apinay£ anomaly: one by R o b i n Fox (1967), another by R. T . Zuidema (1969), and a third by Floyd Lounsbury and Harold Scheffler (1971). An analysis of all these interpretations reveals some constants. First, there is a successive a b a n d o n m e n t of the primary sources as the interpretations of the Apinaye system follow o n e another (this is the case with Fox, 1967). It is as if the analyst were suffering from a "passion for models," so that, seeing his diagrams operate on paper, h e instituted them as the t r u e Apinaye reality. Second, the interpretations are based on paradigms derived from other ethnographic sources and are applied to the Apinaye so as to ac­ commodate NimuendaJL'fs data (as the analyses of Maybury-Lewis [1960] and Zuidema [1969] s h o w ) . T h a t which cannot be included in the paradigm is interpreted either as an error on the eth­ nographer's part or as a consequence of social change. T h e third is common to all sociological analysis where the native ideology is scorned a n d the analyst's perspective is correspondingly ethno­ centric: having many variables to choose from, the analyst arbi­ trarily selects those which best fit his model. Fourth, as a result of all these factors, the system is reduced to a m o r e or less arbitrary and disjointed conjunction of components (rules of residence, descent, cousin terms, and so on) and is so classified. T h a t is why it is essential to examine the explicit beliefs of a group, for it is these beliefs which demonstrate the significance (or lack of it) of each of these components. In this context, it is appropriate to re­ call the words of D u m o n t . W r i t i n g on India a n d dealing with a

86 • Roberto da Matta

situation analogous to the one I am discussing here, he said, "Another way of remaining shut in u p o n ourselves consists in as­ suming from the outset that ideas, beliefs and values—in a word, ideology—have a secondary place in social life, and can be ex­ plained by, or reduced to, other aspects of society" (1970:3) . It was through consideration of this I n d i a n lesson that I re­ solved, in this analysis, to contrapose terminology a n d ideology. T h e s e expressions refer not only to the verbal conventions em­ ployed by the Apinaye to classify persons and social relations, b u t also to the anchoring of this set of terms or categories in a system of ideas capable of being explicated a n d / o r discussed by the na­ tives themselves. Terminological systems are generally studied in the opposite m a n n e r . It is felt that the terminology reveals the ideology, or at the least aids in its examination, and this examina­ tion, in so-called formal analyses, is usually most limited. Here, the reverse direction is taken: ideology is taken as the basis for the analysis of the terminology. P u t this way, it is clear that the purpose of this paper is to de­ scribe and interpret the ideology which informs the ensemble of terms employed by the Apinay^ to classify persons a n d social rela­ tions. I will seek to avoid, as m u c h as possible, treating this ensem­ ble as "kinship terms," again for an "ideological" reason, since the terms, as I shall try to show, form a c o n t i n u u m in which it is not easy to segregate terms w hich can be given genealogical and biological referents from those which cannot. Since social anthro­ pology has never been able to define precisely what is m e a n t by " k i n s h i p " or "kinship system," I prefer to transfer my analytical attention to a "relationship system" rather than arbitrarily to treat certain categories as "kinship terms" a n d others as "Active kinship terms," "ritual kinship terms," a n d so on, as is c o m m o n in the c u r r e n t literature. In this essay, my point of d e p a r t u r e is a set of terms and its ide­ ological foundation. T o each set of terms (or terminological sub­ system) corresponds an ideology relevant to some d o m a i n of reality, as it is defined and carved out by the Apinay£. T o move, then, from an u n d e r l y i n g ideology to a terminological system is to attempt the construction of an Apinaye ontology, that is, a system of categorization which sustains this dimension of their culture. r

Terminology T h e terminological systems used by the Apinaye to classify so­ cial relations can be divided into two parts. T h e first is made u p of relationship terms for general or more inclusive categories, with n o explicit genealogical referents; these I call comprehensive

The Apinaye'Relationship

System • 87

terms. T h e second part involves those terms conventionally called "kinship terms," for the majority have an explicit genealogical a n d / o r biological referent. T h e s e are the terms most commonly used by the Apinay£, and they are as much a justification of be­ havior toward given persons as a m o d e of address. T h e s e I call specific terms, a n d they are divided below into four groups. In the following discussion, the five terminological subsystems are desig­ nated by R o m a n numerals, terms themselves by Arabic n u m b e r s , except the four terms constituting the comprehensive terms. I. COMPREHENSIVE TERMS

a. kwoya; b. kwoya kumrendy; c. kwoya kaag (or kwoya and d. kwoya ket

puro),

All b u t the first of these expressions are formed from the word kwoya and an additional suffix. T h u s the word kumrendy signifies " t r u e , " "real," in contrast to kaag or puro, which means "false," "imitation," and to ket, a negative particle. T h e expression kwoya, furthermore, may be combined with other terms, as in, for exam­ ple, iprom kwoya, which designates relatives by marriage (iprom = wife). N o n e of these terms, including kwoya itself, is ever used in address. T h e y are fundamentally reference categories, em­ ployed to specify a person or, more frequently, categories of per­ sons a n d of social relations. Contemporary Apinay^ gloss the word kwoya into Portuguese as parente ("relative") or men povo ("my p e o p l e " ) . Just as the word parente applies to all kindred or just the nuclear family in Brazilian usage, the term kwoya has a semantic field which ex­ pands a n d contracts according to the vicissitudes of daily life and as a function of what is being emphasized or specified in a par­ ticular context. T h e term kwoya can thus apply to the inhabitants of a given village, in contradistinction to the inhabitants of another village or tribe (the kwoya ket). More rarely, it can also be used to refer to the members of a residential g r o u p (the kwoya kumrendy), in opposition to other residential groups of the same village, the kwoya kaag. Since the category kivoya has no specific reference to social groups on the ground, it lends itself to this classificatory flexibility, as do o u r own terms "relative" or "family." It is possible, then, to make a preliminary ordering of such cate­ gories in the following m a n n e r : (A) kwoya in opposition to kwoya ket; (B) kwoya kumrendy in opposition to kwoya kaag. T h e s e expressions allow a greater or lesser precision in defining any social field to which the terms are applied. In its more general

88 • Roberto da Matta

sense (A) , the category kwoya denotes all the social relations char­ acterized by a sentiment or duty of solidarity, actual or potential. T h e expression thus establishes a contrast with an area where this sentiment or duty does not exist. In its more specific sense ( B ) , the expression amounts to a gradation w ithin a more limited so­ cial field. T h u s kwoya kumrendy specifies persons whose residence is nearby a n d w ith w h o m genealogical relations can be traced easily; persons who support one in disputes a n d with w hom one freely exchanges gossip, food, and labor. In short, kwoya kum­ rendy denotes primarily, b u t not exclusively, persons who rec­ ognize categorical obligations toward each other. T h e s e are peo­ ple who, according to one informant, "one does n o t need to ask, for they always offer hospitality." But the category kwoya kaag (or puro) applies to such areas as adoptive relations and persons w ho are not included within the preceding terms. W h i l e it is possible to exchange meat and work w ith them, this occurs unsystematically and discontinuously. As can be seen, the specifications of these categories are vague. T h e y are applied to fields of relationships and do not generate any absolute dichotomy between kumrendy and kaag which might correspond w ith discrete and well-defined social groups, capable of corporate action. Consequently, in each Apinay^ village there is a continual flow of individuals from one category to another, since there are always kwoya ket in the process of being trans­ formed into kwoya (through marriage or common residence) and kwoya kaag metamorphosing into kumrendy through name-giving, friendship, food, and service exchange. T h e categories, then, over­ lap in a way which might be approximately represented by a Venn diagram. Let us now proceed to examine the n a t u r e of the over­ lapping a n d how the Apinay£ are able to manipulate this general system of relational terms. Every Apinay6 invariably claims that he has a great n u m b e r of kwoya. But when the subcategories of kwoya to w hich the mem­ bers of a village belong are specified, it becomes clear that the n u m ­ ber of kwoya kumrendy a n d kwoya ket is q u i t e small, while the n u m b e r of kwoya kaag is large, composing the majority of a given ego's kwoya. I n fact, a detailed inquiry into this topic utilizing six­ teen Apinaye individuals of both sexes and different ages clearly substantiated their normative statements, In a universe of 152 persons, only 20 percent were classified as kiuoya kumrendy, while 75 percent were included in the kwoya kaag category; the re­ m a i n d e r were composed of kwoya ket. These figures obtained from sixteen informants are very consistent with verbalizations ob­ tained less systematically. r

T

T

r

T

T

T

3

The Apinayd Relationship

System • 89

However, a detailed investigation of the kwoya kumrendy rec­ ognized by various informants shows that it is possible to establish another distinction within this subclass itself. T h i s is because the Apinay£ distinguish betw een kwoya kumrendy with whom restric­ tions on food exchanges and behavior (piangri) are obligatory, and those other kwoya kumrendy toward whom such obligations do not exist. T h e Apinay^ sometimes call the former g r o u p "birth relatives" or "relatives whose blood is near," using the expression kabro atpen burog. T h e s e persons are invariably members of the nuclear family of origin a n d / o r of marriage of the informants (depending on their marital status) a n d are thus segregated from the other kwoya kumrendy, w h o constitute an " u n m a r k e d " class, with n o special term. T h e class kwoya kumrendy is thus subdi­ vided into two classes: one which includes the members of a given ego's family of origin or of marriage (or both) a n d another which includes other " t r u e relatives." Aflines, especially those who reside with ego or who are connected to his wife or her h u s b a n d through primary ties, are considered kwoya kumrendy of the type toward whom no restrictions obtain. T h e ideological basis for this distinction is discussed below. For the m o m e n t , it is sufficient to point out that, in its most general use, the category kwoya refers essentially to a nucleus made u p of people whose interrelationship is seen by the Apinay^ in terms of physiological or biological ties or bonds of c o m m o n substance. T h i s does not m e a n that any given ego's family of origin or of marriage is inevitably defined in this rigid fashion. In fact, rela­ tions of respect or common substance frequently include persons connected to a certain ego by more distant ties. But one quickly discovers that their inclusion in this category is d u e to the fact that they have had an intense intimacy with ego. In all cases of this type, the Apinaye justification is the statement that such persons have come to smell alike (me-kutxe) through having slept to­ gether for a long time. A n d this demonstrates that some social relationships can be initiated from outside the kwoya kumrendy class and subsequently become an integral part of it, since such social relations can be justified through an idiom of common substance. In fact, a m o n g the Apinaye, this idiom of substance is the best translation for what we term " k i n s h i p " or, rather, "con­ sanguinity." I n the Apinay6 system, therefore, there is a set of obligations a n d / o r restrictions which obtain among a restricted g r o u p of people (an average of 6.5 for each of 7 i n f o r m a n t s ) . T h i s g r o u p consists of the nuclear family of origin or of marriage of a given ego or, rather, of those persons with whom a given ego maintains r

90 • Roberto da Matt a

systematic a n d constant b u t n o t exclusive physiological rela­ tionships. T h e s e general categories can thus b e seen in terms of two di­ mensions of contrast. O n e of t h e m separates classes diametrically, establishing a division which m i g h t b e e q u i v a l e n t to t h e classic dichotomy between " u s " a n d " t h e m . " T h e other, however, estab­ lishes a series of gradations o u t w a r d from a restricted circle of per­ sons. T h u s t h e kwoya a r e subdivided again i n t o "persons toward w h o m ego observes restrictions in cases of sickness o r b i r t h " a n d "persons tow ard w h o m such precautions a r e n o t t a k e n . " Grada­ tions c a n b e distinguished a m o n g kwoya, in a series of discrimina­ tions d e p a r t i n g from a central nucleus o r primary g r o u p a r o u n d ego, b u t the d i m e n s i o n which makes these distinctions is o n e in­ volving n a t u r a l , physiological, or o t h e r c o m p o n e n t s of c o m m o n substance. T h e s e terms c a n therefore b e ranged in a series as follows: kwoya ket kwoya kwoya kaag kwoya kum­ r

rendy

kwoya "of restrictions"

ego.

T h e r e is n o difficulty in distinguishing the poles of t h e con­ t i n u u m , b u t there a r e problems in a n a priori specification of t h e classes in the m i d d l e of the series. T h i s goes to show t h e difference between these categories a n d actual social groups, b u t also dem­ onstrates that these categories a r e always used from a n egocentric perspective. T h e categories presented above cover areas which intersect, b u t w i t h i n certain limits. A n Apinaye is thus able to m a n i p u l a t e his classificatory system u p to a certain point. F o r example: a l t h o u g h in cases of conflict h e can change t h e classification of certain rela­ tives considered kumrendy, this is n o t possible if these kumrendy relatives a r e m e m b e r s of his family of origin (his father, his m o t h e r , a n d his u t e r i n e siblings). T h e same holds t r u e for his chil­ d r e n . T h u s restrictions can b e relaxed, b u t t h e relationship is life­ long. In reverse fashion, an Apinay£ can transform some of his kwoya ket i n t o kumrendy, b u t there a r e limits to such transforma­ tions, for they can only take place after regular a n d intense social bonds have been formed between ego a n d t h e persons previously classed as kwoya ket, which very rarely h a p p e n s a n d then usually in cases of a d o p t i o n . I n s u m , I t h i n k it i m p o r t a n t to stress t h e following aspects of this general terminology: 1. T h e terms cover social fields a n d relations; they d o n o t refer to social groups. 2. T h e only kwoya who can be u n a m b i g u o u s l y specified a r e the m e m b e r s of t h e nuclear family of origin or marriage directly tied

The Apinayi

Relationship

System • 91

to ego, that is, his genitors, u t e r i n e siblings, wife, a n d children; and interaction with these people is justified through an ideology founded on notions of c o m m o n substance. 3. It is precisely because this indivisible a n d life-long nucleus of persons with w h o m a given ego has categorical relations exists that t h e m a n i p u l a t i o n of t h e system occurs within certain limits. 4. Besides this g r o u p which shares a c o m m o n substance, there are also intense social relations which can b e spoken of as belong­

ing to the class kwoya kumrendy, b u t which lack a biological foundation; these are relations created a n d m a i n t a i n e d by com­ m o n residence, marriage, n o m i n a t i o n , a n d exchanges of food, pres­ ents, a n d services. II. SPECIFIC T E R M S : COMMON SUBSTANCE

In a d d i t i o n to t h e comprehensive terms, the Apinay£ use a sys­ tem of terms to classify specific relationships, such as those of c o m m o n substance o r initiation r i t u a l friends. T h e s e terms a r e used to classify relations a n d persons situated w i t h i n the field de­ fined as kwoya. T h e r e are n o terms for relations w i t h i n the con­ trasting field, the kwoya ket. Such people are called by personal names o r by terms referring to special relationships. F o r example, if a person is classified as a "formal friend," t h e circle of relatives a r o u n d h i m can b e designated by terms a p p r o p r i a t e to this type or relation, as will b e seen below. If t h e person is a spouse, there is a system of terms used to define such relations, or rather, a system for relations established by marriage, also presented below. 1. pam: F, FB, MH, MZH . . . T h i s term is applied only referentially, w i t h o u t further speci­

fications. T h e term txun, in its variations txun-re o r txun-ti (where re = d i m i n u t i v e particle, ti = a u g m e n t i v e particle, b o t h referring to t h e physical size of a p e r s o n ) , is used as t h e vocative term. Some informants used the Brazilian papai, l i m i t i n g its application to the

genitor o r pam kumrendy. T h e most frequent use of papai occurs in those u n d e r twenty. Both pam a n d txun may b e used w i t h i n t h e categories kumrendy a n d kaag, when it is wished to signify the dis­ tance or specific position of a pam in relation to a given ego. T h e txun form may also b e used with a proper n a m e , as well as with other suffixes (such as prek = old or prin = y o u n g ) , as the voca­ tive o r the necessity for i n d i v i d u a t i o n may r e q u i r e . I t is also pos­ sible to employ the descriptive t e r m nipeitxo ("he w h o m a d e me") to designate t h e genitor a n d / o r t h e "adoptive father" (who

92 • Roberto da Matta

arranged n a m e s ) , with the following distinction: nipeitxo is used here as a reference term for the genitor b u t as a term of address for the name-arranger. T h e use of this descriptive form will become clear in a m o m e n t . For the genitor, the vocative form is always a nickname, that is, an individualizing t e r m whose use does n o t ex­ tend beyond the boundaries of a very m u c h reduced g r o u p of per­ sons. It is also possible to use the term to-re to address the genitor, thus indicating intimacy and affection. W h e n a pam dies, the par­ ticle pinrog is used immediately after the term. T h i s particle is interpreted by the Apinay£ as a death marker. Besides these forms, the Apinay6 use the term me-papam (where me = collective) to refer to the sun or, nowadays, to God, a n d sometimes to the Chief of the Goiania Inspectorate or to the President of Brazil. 2. na: M, MZ, FW, FBW T h i s term is used in reference. T h e term dyil in its various forms is used as the vocative which results from the addition of particles (ti, re, prek, kumrendy, or kaag). A proper n a m e may also be used to distinguish an individual in this category. As with the previous term, the expression kator-txo ("she from w h o m I came") may be used to indicate the genetrix or to refer to the "adopted m o t h e r " (who arranged names for ego) or to the wife of the name-arranger. Again, as with the preceding term, the genetrix is never addressed by a term, b u t addressed by a nickname used only by the members of her nuclear family or by those persons resident in her primary group. W h e n a na dies, the term pabotxoi-ti is used to refer to her. 3a. to: B, MS, FS, MZS, FBS 3b. tody: Z, MD, FD, MZD, FBD To and tody are the vocative forms. Referentially, they are used w ith ti (augmentative) a n d re (diminutive) to indicate physical size. W i t h kumrendy or kaag they indicate social distance. T h e form kambu is used in address by a female ego a n d the form pigkwa by a male ego. Like the genitors, u t e r i n e siblings are al­ ways addressed by a proper n a m e or nickname, the preceding forms being used much more for adoptive or ceremonial siblings, in contexts where social distance is emphasized. T h e term pinrog is added after these forms to indicate that the "sister" or " b r o t h e r " in question is dead. T h e expressions kro-ti and kro-re are used, respectively, for the older brother and younger brother, b u t they are never employed as vocatives for uterine brothers, who are al­ ways addressed by n a m e . T

The Apinayi Relationship

System • 93

4. kra: S, D, BD, BS (man sp.); ZS, ZD, S, D (worn, sp.) T h e term is used as a vocative and as reference. T h e forms akatxoi and akante-re are used, b u t very rarely, to distinguish, re­ spectively, a male from a female kra. However, ego's own children are always addressed by names or nicknames, never by relation­ ship terms. As with the preceeding terms, it is possible to use this category to specify social distance, through the addition of kum­ rendy and kaag. 5. geti (ngeti):

FF, MF, MB, FZH

T h i s term is used both vocatively and referentially. W h e n a person in this category passes his personal names to a male ego, the expression krd-tum may be used referentially or vocatively for h i m . Now, the reciprocal term is commonly kra-duw. T h e form turn can often be used, b u t it is always accom­ panied by the suffixes re, ti, or prek, or by the proper n a m e of the krd-tum. T h e Apinaye have a verbal and behavioral tendency to treat as krd-tum the name-giver or other persons of the same gen­ eration as the parents. Meanwhile, the term geti implies old age, as in the term pinget (age-class of "grandfathers" or "old people"), or imbre-geti, a term used as reference for the father-in-law (see term 8 ) . W h e n a geti dies, the term pinrog is a p p e n d e d after geti to indicate death. Some informants of various ages use the Brazil­ ian form vovo or vovo (without making the vowel contrast which differentiates male from female in Portuguese) to translate this term, a n d often use it in address. T h e term can also b e accom­ panied by the specification kumrendy or kaag. 6. tui (tukatui):

MM, FM, FZ, MBW

T h e term is used for both address and reference. As with the other terms, the particles prek, prin, ti, and re, or kumrendy or kaag can be used to indicate differences or specify a particular tui. T h e same end can be achieved by using the proper name after the term. W h e n , however, a tui gives her names to a child in the cate­ gory tamtxua (specified b e l o w ) , there is n o change of terms, as in the previous male case. W h e n a tui dies, the term pinrog is added as a suffix to indicate death. 7. tamtxua: SD, SS, DD, ZS, ZD, WBS, WBD (man sp.); SS, SD, DD, DS, BS, BD, HZS, HZD (worn, sp.) T h e term is used for b o t h address and reference. It is possible to distinguish the sexes by using the expression parxiote for a

94 • Roberto da Matt a

female tamtxua a n d apare for a male tamtxua. Yet the term is rarely used in address for tamtxua closely related to Ego (his grandsons, for e x a m p l e ) . For persons thus related, the proper name or nickname is used. T h i s term also is used with pinrog to refer to a deceased person. III. SPECIFIC TERMS: AFFINAL

8. im bre-geti: W F (man sp.); H F (worn, sp.) Imbre-geti is a reference term. T h e vocative form is tukoya. T h e Apinay£ never use personal names as terms of address for affines, always calling them by the appropriate relationship term. Similarly, such terms are rarely used with the suffixes re or ti, a n d are never employed w ith the categories kumrendy or kaag. T h e y are therefore situated outside of the general system of classification. T h i s holds true for all the following terms. r

9. papam-geti

(address and reference): W M (man sp.) ; H M (worn, sp.) 10. improm:

W

T h e term is only referential. T h e h u s b a n d never calls his wife iprom, always using her n a m e a n d / o r nickname, especially after the marriage has been stabilized through the advent of children. 11. idpien:

H

T h e term is only referential. As with term 10, a wife always addresses her husband by n a m e or nickname, especially after they have children. 12. papany

(address) : WZ, BW

T h e referential form is ipany. IS. imbre: WB Imbre is a reference term. T h e vocative form is imboi. 14. idpienhon:

ZH

T h e term is used for both address a n d reference and is always ac­ companied with the suffixes re or ti. T h e term tuko can be equally used for address. 15. txoiti: SW, BSW, ZSW, BW (worn, sp.) T h e term is used for both address and reference.

The ApinayS Relationship

16. tukoti:

System • 95

DH

T h e term is used for both address a n d reference., 17. ponmre-geti:

HM

T h e term is used for b o t h address a n d reference. 18. ponmre:

ZH

T h e term is used for both address a n d reference. I V . SPECIFIC TERMS: "FORMAL FRIENDS"

19. krd-geti (masc.); kra-gedy (fern.) T h e s e terms are used in reference a n d address for a person who establishes highly ceremonial ties with ego, giving h i m a set of ornaments for the chest, arms, and legs, in addition to the insignia of one of the Ipgnotxoine or Krenotx6ine moieties. These terms do not p e r m i t any modification or specification in terms of physi­ cal size or relative age. Formal friends are never addressed by their personal names or nicknames. 19a. pa-kra T h i s is the reciprocal of the terms given above, used in address and reference toward a formal friend younger than ego. T h i s re­ lationship also affects the children of the "formal friends." T h e children of persons thus related treat each other reciprocally as ikamde-re or ikamde-ti (the suffixes indicating relative a g e ) , if they are b o t h male. If they are of opposite sexes, they treat each other as improm-ti (man speaking with the daughter of his krdgeti or pa-kra) and as ikande-ti (woman speaking with the son of her krd-geti or pa-kra) . T h e s e are both terms of reference and address. V . SPECIFIC TERMS: INITIATION RITUAL FRIENDS

20. krd-txua T h e term is used to classify persons of the same age who par­ ticipate together in the rituals of male initiation, in their first or second phases (respectively, peb kaag or peb kumrendy) (cf. N i m u e n d a j u 1939:37 ff., 56 ff., and 1956:chap. 9 ) . T h e selection of these friends is made through personal preference, b u t the terms are employed for life, replacing personal names or the re­ lationship terms discussed previously. As with terms 19 and 19a, this class does not allow any gradations. Friends thus related are of r

96 • Roberto da Matt a

the same age (or, to be m o r e precise, of the same age-class), and their relations are symmetrical, as witnessed by the modern Apinaye term for them: companheiros (companions) . Occasionally, the relatives of krd-txua are affected by the relationship, so that two people who are krd-txua to each other call each other's chil­ dren kra. All these terms are invariably preceded by the possessive id m

( = Y) • T h i s is the complete list of the Apinay^ relationship terms. As their primary and partial definitions reveal, these terms refer to various subsystems of social relations. In fact, the presentation of the system itself implied a decision on my part to segregate them into subsystems. A cursory analysis of such terminological sub­ systems reveals certain features which indicate that these subdi­ visions do not violate Apinaye classificatory thought. For example, subsystems I I a n d III, unlike the others, are genealogically de­ termined. Besides this subsystem I is somewhat relative, being—as I define it—an all-inclusive system of classificatory relations. But subsystems II and I I I are automatic in their application, being for that very reason determined biologically a n d genealogically. Sub­ system IV (that of "formal friends") is difficult to translate genea­ logically, and V (that of krd-txua) is frankly dependent on in­ dividual choice. Similarly, subsystem II can be used with particles or categories which differentiate persons a n d social relations (the particles re, ti, prek, prin, a n d the categories kumrendy a n d kaag), b u t subsystems III, IV, a n d V do not permit such nuances. T h e same thing holds true with reference to the address terms which accompany such terminological fields. W h i l e subsystem II permits the use of personal names in address, this does n o t h a p p e n with subsystems I I I , IV, a n d V, which are thus m u c h more formalized. These subsystems, consquently, always supersede other relation­ ships and in so doing indicate new social relations. I argue here that, in an orthodox analysis, the terms and their morphological distinctions would guide the analyst. Undoubtedly, then, the Apinay^ relationship terminology would be promptly subdivided u n d e r the classic headings "consanguineal terms" a n d "affinal terms," leaving aside the two last subsystems, which refer to "formal friends" a n d "companions." But such a focus would certainly raise various problems. For example, how would one explain the extensive use of the term geti for "consanguines" a n d certain affines, as in terms 8 and 9? Or, again, how could one elucidate the use of the word krd ( = head) in so many terms? Are these relations to be regarded as extensions of primary ones, that

The Apinaye Relationship

System • 97

is, those which possess an indisputable genealogical determina­ tion? But, then, would the sole common denominator of such terms be a linguistic one? It was in seeking to avoid such problems that I resolved to let my analysis be guided by a study of ideology. If, in fact, the Apinay£ terminological system is subdivided into specfic fields, then there ought to exist corresponding ideologies for each one of these subsystems. Such ideologies would be explicated by infor­ mants in terms of rules of etiquette, prescriptions for behavior, dis­ tinctions between address a n d reference forms, beliefs relative to the h u m a n body a n d procreation, a n d so on—in a word, through a system of signs which in t u r n would indicate the fixed a n d irre­ ducible points of Apinay£ social order. It is these sets of signs that I have been calling ideologies. And it is these ideologies which refer not just to Apinaye relationship terms, b u t to their entire cosmos. Let us, then, examine the terms as a function of these sys­ tems of ideas. Ideologies I n the Apinayd terminological system, there are various terms which specify members of the nuclear family. T h i s holds as m u c h for the comprehensive terms as for the terms applied to the genitors of a given ego. I shall, therefore, start this section by taking as my d e p a r t u r e point the nuclear family and its ideology. Among the Apinaye, each nuclear family has its own plot of g r o u n d to cultivate, and is the only socially recognized u n i t for the production of children. Marriage is strictly monogamous and spouses have mutually exclusive rights to each other's domestic and sexual services. T o each nuclear family corresponds a specific area: either a house or a special area within a house. Like all the Ge, the Apinaye practice uxorilocal residence, so that a large num­ ber of households in their two villages contain more than one nu­ clear family. W h e n this happens, the ideal is that the residential g r o u p operates as a unit, at least for the execution of specified duties. But, even so, the nuclear families remain independent in eating a n d sleeping, two activities by which the Apinay6 define domestic life. Young unmarrieds of both sexes do not have the right to build a house, and the m e n do not possess a p e r m a n e n t residential structure in the middle of the village, such as occurs a m o n g the Bororo and the Kayap6. Apinaye houses are arranged in a circle a r o u n d a public plaza so that the opposition between these two areas, houses and plaza, a n d the groups which have rights to them is clearly expressed. In

98 • Roberto da Matta

fact, the sphere of the family, economics, and socialization of the young is a domestic and daily order; while the ceremonial order, along with the groups formed to conduct it, is a public order essentially situated outside daily routine. While the first is regu­ lated by persons who maintain intense a n d lifelong social rela­ tions, the second is formed by groups which come together sporadically and whose membership can vary from occasion to occasion. T h e Apinaye domestic order is marked by two clearly expressed ideologies. T h e first, discussed immediately below, is expressed in so-called avoidance rules or rules of etiquette. T h e second, dis­ cussed next, has been treated in the anthropological literature, especially in South American ethnology, u n d e r the title of "couvade," which I here call precautions, abstinence, or restric­ tions. Last to be discussed in this section is the etiquette of cere­ monial relationships. ETIQUETTE

Every time an Apinay£ speaks of marriage, he is always careful to enumerate the rules of etiquette w hich characterize conjugal life in his society. It is evident that these rules grow out of the asymmetry produced by uxorilocal residence, which com­ pels men to leave their natal houses to their brothers-in-law and move to the houses of their wives. But this, obviously, is not all. In fact, the problem is not only that of having to live physically close to persons who are socially distant (affines). It is also that of creating within an originally unified nuclear family a social field for a new nuclear family, since most Apinay6 households contain more than one, a n d of accommodating the relative politi­ cal and economic independence of the latter. T h e problem there­ fore has equal repercussions for the affines. Actually, the rules of Apinay£ etiquette are of two types: to­ ward the parents and the brothers of his wife, a husband must behave in a ceremonial a n d restrained manner. H e never calls them by their proper names or by their nicknames, b u t always uses the appropriate relationship terms. H e never speaks to them, unless prompted by a question. And even when he has succeeded in the h u n t , h e remains silent u n t i l his parents-in-law a n d wife have cleaned the animal a n d cut u p the meat to distribute it. Apinaye etiquette stipulates that the father-in-law receive, butcher, and distribute the meat, always separating a portion for his son-in-law's parents; b u t the son-in-law says n o t h i n g and merely waits for his older affines to handle the matter. Similarly, a r

r

The Apinaye Relationship

System • 99

husband never enters the rooms of his parents-in-law or brothersin-law, nor does he have the right to use their weapons or posses­ sions. These are avoidance rules par excellence. T h e r e are, however, other rules, m o r e circumscribed than (and in a certain sense opposed to) these. A husband ought to a n d can converse freely with his wife's sister a n d treat her children affec­ tionately, in the same way as he would his own wife and her chil­ dren, because the relationship between them should, at the outset, be characterized by cordiality and restraint, virtues m u c h admired by the Apinay6. In fact, when relations between a m a n and his papany (see term 12) are discussed, the Apinay^ always say that such a woman is almost the same thing as a true wife. T h e reverse of these rules also holds true. A father-in-law or brother-in-law never enters his affine's room, nor takes his personal possessions without permission. T h i s reciprocal avoidance be­ havior shows that the tension of residence change is felt by both sides, that the h u s b a n d as much as the original residents of the household g r o u p suffer with the former's change of residence. In the Apinay^ case, this is clearly seen because there is no rigid definition of domestic authority. Since unilineal descent groups do not exist, either the father-in-law, mother-in-law, the husband, or the brother-in-law can occupy an important position within the domestic g r o u p a n d be the key point of articulation of its mem­ bers' social activities. In some domestic groups the extended family is coordinated by an old man, in others by a woman, and in still others by a middle-aged husband. Since this coordination generally extends to neighboring houses, residential clusters tend to form, with important political consequences. Therefore, the presence of a strange man in a house generates tensions not just for him, b u t for all his affines. These paradoxes are expressed by the Apinay£ through an ideology which underlies such avoidance relations: the ideology of piam. All these relations are marked by piam, and this cate­ gory defines qualities of shame, respect, a n d / o r social distance implicit in certain social relationships. In fact, the ensemble of prescriptions expressed by this category is impressive. Persons who have piam cannot joke, they cannot speak lightly of each other, nor fight, ask each other favors, call each other by name, look each other in the eye, walk or bathe together, have sexual rela­ tions, urinate together, talk to each other, or observe ceremonial restrictions for each other. Persons thus united can, however, carry on formal transactions, an activity always marked by piam for the Apinay^. T h e prescriptions determined by piam apply, with greater or 7

100 • Roberto da Matta

lesser intensity, to all Apinay^ social relationships. T h e y d o not exist in relations between persons of the same age, same sex, and same nuclear family (two uterine brothers or sisters), b u t begin to exist between a genitor and his children a n d even m o r e be­ tween a genetrix and her son, the reverse being t r u e for a woman in relation with her u t e r i n e brothers, her father, a n d her mother. Piam is, then, correlated with biological differences of sex a n d age, gradually increasing as other discriminations are added to these two fundamental dimensions of contrast in the Apinay£ social system. Consequently, every time that persons of different sex, age, and social groups enter into formal transactions, their relations are marked by piam. Consistent with this custom, the passage from one of these categories to another is marked by some type of precaution. Young initiates ready to change their class follow alimentary regulations of the same kind as those obeyed by a m a n who has become sick through contact with an animal and by a h u s b a n d a n d wife who are in direct contact with a newborn child (I will r e t u r n to this topic b e l o w ) . T h e greater the difference be­ tween the fields a n d the categories in contact, the greater the piam; the less the difference, the less the piam. T h e corollary of this ideology is the expression piam ket, used to castigate those persons who egotistically cease to follow the more i m p o r t a n t pre­ scriptions of Apinay£ culture. Actually, a person "without piam" is equated with dogs, animals who u n d e r s t a n d all that is said, b u t nonetheless continue to act antisocially. A person being piam ket is therefore an individual who does not obey the n a t u r a l order of things, since a world without piam is a world in disorder—a uni­ verse where the fundamental categories of Apinay£ society would be confused and its social fields scrambled. But in the ideology of Apinaye etiquette, it is not only the husband a n d his affines who have relationships marked by piam. A woman also has piam, n o t only with regard to her parents and brothers (but n o t her sisters), b u t also toward her husband. H e r piam is centered in the fact of her public admission that she has p e r m a n e n t sexual relations with a m a n . W h i l e the Apinaye say that the piam of a woman is equal to that of her husband, vis-a-vis their respective affines, I suspect that the woman has m u c h m o r e piam with h e r own genitors and close relatives. T h i s is d u e to the physical (and social) distance between her and her affines, who generally live on the opposite side of the village. Perhaps it is because of this that the Apinay£ always discuss these rules of etiquette from the point of view of the husband, even when the informant is a woman. T h e ideology of Apinaye etiquette rules marks a social field

The Apinaye Relationship System • 101

which was opened by marriage. T h i s field is initially segregated by avoidance rules which rigidly separate the husband from his affines of the opposite sex and separate h i m from his wife and her m o t h e r in a more attenuated way. But as the marriage stabilizes through time, especially with the arrival of children, piam di­ minishes with opposite-sex affines and disappears with the wife. T h i s d i m i n u t i o n signifies n o t only that the husband and wife have adjusted to the marriage, b u t also that they are capable of creating a potentially independent zone inside an original nuclear family. An uxorilocal extended family is thus formed, and the group, when the piam is equilibrated, can operate as a u n i t in various social affairs. In other words, intensive piam expresses social relations in which persons who represent discrete a n d / o r segregated social fields enter into contact. T h e relations marked by piam at its greatest intensity are relations in which the agents have clearly defined matters to transact; they are therefore ex­ change relations. In the case of relationships between a man and his affines, these matters are made clear at the beginning of mar­ riage: he h u n t s a n d works for his parents-in-law and brothers-inlaw, receiving in exchange the exclusive domestic and sexual services of his wife. But to the degree that his nuclear family is established and integrated into an extended family, the cycle of these relations becomes a long one (Sahlins 1965), with conse­ q u e n t d i m i n u t i o n of piam between the man and his affines. Avoidance rules are strong and intense at the m o m e n t of marriage, when h e and his wife create their own, potentially i n d e p e n d e n t area. "COUVADE"

Another ideological component of the nuclear family is the fact that this g r o u p is conceived by the Apinay