<%BANNER%>

Weaving Power

Permanent Link: http://ufdc.ufl.edu/UFE0041474/00001

Material Information

Title: Weaving Power Displacement, Territory and Indigenous Knowledge Systems Across Three Kaiabi Groups in the Brazilian Amazon
Physical Description: 1 online resource (544 p.)
Language: english
Creator: Athayde, Simone
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: basketry, biocultural, brazilian, cultural, designs, displacement, indigenous, kaiabi, resilience, systems, territory, textiles, xingu
Interdisciplinary Ecology -- Dissertations, Academic -- UF
Genre: Interdisciplinary Ecology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Indigenous lands are responsible for the conservation of nearly 21% of the Brazilian Amazon, and are the main barrier against deforestation. This research contributes to the understanding of the relationship between political empowerment, socio-cultural resilience and territorial control among Amazonian indigenous peoples. I sought to explore which factors may lead to the persistence or loss of indigenous knowledge after geographical displacement. I argue that cultural and environmental resilience are intertwined, so that where traditional knowledge is maintained, there will be greater territorial control among indigenous peoples in the Amazon. I apply a systems approach to explore the effects of historical, environmental, political, socio-economic and cultural factors in their interaction with a specific domain of indigenous knowledge. The Kaiabi speak a language of the Tupi-Guarani stock and are great agriculturalists and basket makers. The majority of the population was transferred by the Brazilian government from their ancestral territory in the Tapajo acutes River basin to the Xingu Park region from 1950 to 1966. Two small groups remained in the ancestral land, one on the Peixes River and other on the Teles Pires River. Transfer to the park brought changes in Kaiabi social and political organisation and in the access to and management of natural resources. Forty years after the transfer, the Xingu Kaiabi have adapted to the new conditions, creating mechanisms for cultural perpetuation and territorial control. By contrast, the groups who remained in the ancestral land lost many aspects of their traditional lore. Cross-sectional and longitudinal studies were carried out in order to compare the dynamics of knowledge related to basketry and textile weaving among 114 men and 110 women in four villages across three Kaiabi groups in a five year period. Factors that led to the cultural peristence and political empowerment of the Xingu Kaiabi in contrast with the other two groups were also explored, as well as peoples? perspectives on the role and work of political organizations. Results show that knowledge is being both innovated and eroded among the Kaiabi, and that agency, leadership and innovation are critical assets in enabling cultural resilience. Whereas in Xingu and Teles Pires areas there is innovation and new knowledge being generated- with younger generations still learning- in Rio dos Peixes, knowledge and native language have been eroded at a faster pace. Promoting the persistence of native language and adapting schools to include both indigenous and western knowledge could help to reverse the process of loss of cultural resilience in Rio dos Peixes. Greater political empowerment in Xingu, allied to the development of community-based projects, has influenced indigenous knowledge perpetuation and mechanisms for its transmission. In the Kaiabi case, territorial isolation combined with political support and local leadership led to greater cultural and environmental resilience of the Xingu group in comparison to the other two areas.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Simone Athayde.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Schmink, Marianne C.
Local: Co-adviser: Heckenberger, Michael J.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041474:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041474/00001

Material Information

Title: Weaving Power Displacement, Territory and Indigenous Knowledge Systems Across Three Kaiabi Groups in the Brazilian Amazon
Physical Description: 1 online resource (544 p.)
Language: english
Creator: Athayde, Simone
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: basketry, biocultural, brazilian, cultural, designs, displacement, indigenous, kaiabi, resilience, systems, territory, textiles, xingu
Interdisciplinary Ecology -- Dissertations, Academic -- UF
Genre: Interdisciplinary Ecology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Indigenous lands are responsible for the conservation of nearly 21% of the Brazilian Amazon, and are the main barrier against deforestation. This research contributes to the understanding of the relationship between political empowerment, socio-cultural resilience and territorial control among Amazonian indigenous peoples. I sought to explore which factors may lead to the persistence or loss of indigenous knowledge after geographical displacement. I argue that cultural and environmental resilience are intertwined, so that where traditional knowledge is maintained, there will be greater territorial control among indigenous peoples in the Amazon. I apply a systems approach to explore the effects of historical, environmental, political, socio-economic and cultural factors in their interaction with a specific domain of indigenous knowledge. The Kaiabi speak a language of the Tupi-Guarani stock and are great agriculturalists and basket makers. The majority of the population was transferred by the Brazilian government from their ancestral territory in the Tapajo acutes River basin to the Xingu Park region from 1950 to 1966. Two small groups remained in the ancestral land, one on the Peixes River and other on the Teles Pires River. Transfer to the park brought changes in Kaiabi social and political organisation and in the access to and management of natural resources. Forty years after the transfer, the Xingu Kaiabi have adapted to the new conditions, creating mechanisms for cultural perpetuation and territorial control. By contrast, the groups who remained in the ancestral land lost many aspects of their traditional lore. Cross-sectional and longitudinal studies were carried out in order to compare the dynamics of knowledge related to basketry and textile weaving among 114 men and 110 women in four villages across three Kaiabi groups in a five year period. Factors that led to the cultural peristence and political empowerment of the Xingu Kaiabi in contrast with the other two groups were also explored, as well as peoples? perspectives on the role and work of political organizations. Results show that knowledge is being both innovated and eroded among the Kaiabi, and that agency, leadership and innovation are critical assets in enabling cultural resilience. Whereas in Xingu and Teles Pires areas there is innovation and new knowledge being generated- with younger generations still learning- in Rio dos Peixes, knowledge and native language have been eroded at a faster pace. Promoting the persistence of native language and adapting schools to include both indigenous and western knowledge could help to reverse the process of loss of cultural resilience in Rio dos Peixes. Greater political empowerment in Xingu, allied to the development of community-based projects, has influenced indigenous knowledge perpetuation and mechanisms for its transmission. In the Kaiabi case, territorial isolation combined with political support and local leadership led to greater cultural and environmental resilience of the Xingu group in comparison to the other two areas.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Simone Athayde.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Schmink, Marianne C.
Local: Co-adviser: Heckenberger, Michael J.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041474:00001


This item has the following downloads:


Full Text





WEAVING POWER: DISPLACEMENT, TERRITORY AND INDIGENOUS KNOWLEDGE
SYSTEMS ACROSS THREE KAIABI GROUPS IN THE BRAZILIAN AMAZON





















By

SIMONE FERREIRA DE ATHAYDE


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2010


































2010 Simone Ferreira de Athayde

































To my family, for all the love, care, and support, and to the Kaiabi people, who welcomed me in
their lands and inspired me to learn another perspective about the world.









ACKNOWLEDGMENTS

This dissertation is the result of a journey often years of work among the Kaiabi people

from the Brazilian Amazon, and builds on my previous master thesis. During all these years I

have counted on the support, inspiration and collaboration of many institutions, professors,

colleagues, family and friends whom I would like to acknowledge.

This research would not have existed as such if my friend Aturi (Jowosipep) Kaiabi had

not come to my shack in Diauarum Post in 1999 and asked me if I could help him to write a

project. Seven years later, we were sharing the satisfaction to hear that the "Kaiabi Araa" project

was one of the winners of an award from the Ministry of Culture in Brazil. I want to sincerely

thank my friend Aturi for having inspired me, for his trust, good spirit, and capacity of

leadership. I also want to thank him for the care and hospitality during my prolonged stay in

Tuiarare village in Xingu Park. If this dissertation could be co-authored, I would share the

autorship with him.

I wish to thank the Kaiabi people for having welcomed and allowed me to work in Xingu,

Rio dos Peixes and Teles Pires, especially the communities from Capivara, Tuiarare, Tatuy Post,

Novo Horizonte and Kururuzinho villages. Thanks to Mairawe, Alupa, Tari, Makupa and

Yanukula from ATIX, for all the help, collaboration and trust. Thanks to the chiefs Jywapan

(Capivara), Aturi (Tuiarare), Murua'i (Rio dos Peixes) and Hatu (Kururuzinho) for their

hospitality and interest. Thanks also to my dedicated field assistants who helped me conducting

and translating interviews: Sirakup (Capivara), Aturi (Tuiarare), Jywapan and Pipala (who

helped me conduct interviews in Rio dos Peixes) and Eroit (Kururuzinho). Thanks to my friend

Canisio and his wife Morea'i for the hospitality in Rio dos Peixes. Thanks to Eroit and his family

for the hospitality and great food in Kururuzinho. A special thanks to some women who









participated as leaders and textile teachers in the Kaiabi Araa project: Mytang (Tuiarare), Maru

(Caicara), Cunhaete (Kururuzinho), Juaruu (PIV Manito) and Josiane (Kururuzinho).

I thank my family for having believed and supported me during my life as a jungle

person, remembering my bank account password after I spent five months in the jungle without

touching my wallet. Especially, my gratitude and love for my father Mauro, my mother Eunice

and my sister Miriam. Thanks to Geraldo Silva for the friendship, for his commitment to the

Kaiabi people and for the dedication to our son Adriano. I wish to thank God for sending my son

Adriano my way and for protecting us during our Amazonian travels. I'm thankful to the boy

who had the privilege to spend his baby years at Xingu among the indians, and who

accompanied me during all my fieldwork stages: facing challenges, crossing rivers, getting

tropical viruses, fishing piranhas, riding planes, trucks, boats and buses in the middle of the

Amazon frontier.

I wish to thank many institutions that provided financial support to enable me to complete

this work. Thanks to the Instituto Socioambiental (ISA), my home institution, who supported

different stages of this research and the preliminary versions of the Kaiabi Araa book. Thanks to

the "Conselho Nacional de Desenvolvimento Cientifico e Tecnol6gico" (CNPq), the Brazilian

national agency for science and technology, which supported my research through a four-year

doctoral grant. To the "Instituto de Educacgo do Brasil" (IEB), for the support through the

BECA Program. Thanks to the Tropical Conservation and Development Program (TCD) and to

the Center for Latin American Studies at the University of Florida (UF), which supported

preliminary fieldwork stages through different scholarships. Thanks to the School of Natural

Resources and the Environment (SNRE) at UF for the support, especially during the final stages

of the writing of this dissertation. Thanks to the Amazon Conservation Leadership Program









(ACLI) at University of Florida, funded by Gordon and Betty Moore Foundation, for the

financial support. I thank the American Association for University Women (AAUW) for the

support through the concession of a prestigious international fellowship.

I met Dr. Marianne Schmink nine years ago in Rio Branco, Acre, Brazil. At that time, I

already enjoyed her smile and sympathy. Destiny worked its way in getting me to undertake my

doctoral program at University of Florida, under her supervision. I consider myself privileged to

have been her student and wish to express my sincere gratitude for all the attention,

encouragement, care and support she has devoted to me during my stay at UF. I wish to

acknowledge the important role that Marianne has played as TCD Director since the creation of

the Program. The environment and support provided by UF and TCD makes you feel at home,

even miles away from your home country. I feel that I could not have come to a better place to

take my doctoral program.

I want to acknowledge and thank other UF faculty who have supported and encouraged

my research. I wish to thank Dr. Robert Buschbacher for his constant dedication and support, as

well as for his leadership and extreme devotion in running programs such as the ACLI (Amazon

Conservation Leadership Initiative), which has enabled Latin American students to access

quality education at UF. Thanks to Dr. Michael Heckenberger, my Xingu colleague, for all the

exchange of ideas and advice related to my work with the Kaiabi. To Dr. Emilio Bruna for his

guidance in co-writing a paper with me on participatory management of arumd. To Dr. Anthony

Oliver-Smith for the great classes on economic anthropology and rural peoples, and for his

constant encouragement. My gratitude to Dr. Russel Bernard for the incredible course on

research design in anthropology, as well as for his availability, dedication and enthusiasm related

to my research. To Dr. Stephen Humphrey, SNRE graduate director, for his support since the









beginning of my program. Thanks also to Cathy Ritchie and Meisha Wade from SNRE for the

support. My gratitude to Dr. Karen Kainer, Jon Dain, Patricia Sampaio and Margarita Gandia

from TCD/Center for Latin American Studies for the inspiration, attention and support. My

appreciation and gratitude to Dr. Mary Allegretti for teaching a wonderful course on Amazonian

social movements, and for helping me with the writing of one of the chapters of my dissertation.

I'm grateful to national and international museums which have collaborated with the

Kaiabi people through the donation of photographs of objects of Kaiabi material culture,

especially baskets and textiles. Specifically, thanks to: the "Museu de Arqueologia e Etnologia"

from University of Sdo Paulo (MAE -USP); the Museum of Cultures in Basel, Switzerland, in

the person of Alexander Brust, head of the America's department; the Museum of Ethnology of

Berlin, Germany; and the National Museum of Rio de Janeiro. Thanks to Dr. Georg Grinberg,

for having collected many baskets among the Kaiabi in 1966 and deposited them in ethnographic

museums. I appreciate his effort in visiting the Kaiabi 40 years after the finishing of his

dissertation.

I wish to acknowledge the support of previous colleagues and professors with whom I

interacted during my Masters program in England. Thanks to Dr. Roy Ellen, my former advisor,

for his dedication; to Dr. Raj Puri, for the guidance in data analysis; and to Dr. Gary Martin, for

his interest and encouragement. I remember when Gary suggested that I should do research about

people learning from the Kaiabi Araa book about seven years ago in Canterbury. A special

thanks to Dr. Miguel Alexiades for his years of dedication and guidance in co-writing a chapter

with me for his book on mobility and migration in indigenous Amazonia.

My sincere gratitude and appreciation to my non-indigenous Xingu colleagues and

friends, without those my life in Xingu would not have been so pleasant, safe and adventurous.









Thanks to my former boss and friend Andre Villas B8as from Instituto Socioambiental for his

continuous care, encouragement and support, besides his incredible dedication to the work with

indigenous peoples in the Amazon. Thanks also to all my "xinguano" friends from ISA: Katia

Ono, Paulo Junqueira, Paula Menezes, Rosana Gasparini, Maria Cristina Troncarelli (Bimba),

Estela WOrker, Marcus Schmidt (Tope), Renata Farias and Angelise Pimenta for all the years of

fun, collaboration, friendship and collective baths in the Xingu River. Thanks to Jodo Pavese for

the friendship and dedication in the production of the excellent documentary Kaiabi Araa. I'm

grateful to Frederico de Oliveira and Francisco Stuchi for their support and for their work with

the Teles Pires Kaiabi.

So many friends have helped me during my PhD journey that I might forget to list and

thank some of them here. For the friends that I forget to list, please accept my excuses and my

gratitude. To my Portuguese friend Joana Rodrigues, thanks for the fun times spent during

fieldwork in Xingu and for her dedication to the work with the Yudja people. My sincere

gratitude to Doriam Borges and Ludmila Ribeiro for their help with data analysis and for

teaching me to work with the SPSS Program. Thanks to my friends in Gainesville (some already

finished their programs) for the friendship, fun and care: the "superpoderosas" Diana Alvira

(Tita) and Amy Duchelle; Arika Virapongse; Omaira Bolanos; Ane Alencar; Morena Maia; Eric

Carvalho; Ricardo and Denyse Mello; Lucimar Souza; Margareth Buschbacher; Pio and Jenny

Saqui; Wendy-Lin Bartels; Shoana Humphries; Valerio Gomes; Leonardo Pacheco and all the

Brazilians that play in the "Boteco do Pacheco" band. Finally, I wish to thank my boyfriend Gary

for his love, dedication and enthusiasm, especially during the final stages of the writing of this

dissertation.









TABLE OF CONTENTS

page

A CK N O W LED G M EN T S ................................................................. ........... ............. .....

L IST O F TA B L E S ....................... ........................................................13

LIST OF FIGURES ................................. .. .... ..... ................. 14

LIST OF ABBREVIATION S ................ ... ........... ........................................ 19

A B S T R A C T ............ ................... .................. ........................................ 2 2

CHAPTER

1 INTRODUCTION .................. .................. ................................ ............ 24

The Kaiabi: Diaspora, Resilience and Change of an Amazonian People .............................24
Weaving Indigenous and Scientific Knowledge through Collaboration..............................26
Fields of Inquiry and Possible Applications...................................................... ............... 29
Indigenous Knowledge Systems and Biocultural Diversity................ .............. ....29
Social-Ecological Systems, Resilience and Change............................................ 34
Indigenous Knowledge, Displacement and Place ................................. ............... 38
The Political Ecology of Empowerm ent ..................................................... ..................43
A Process Perspective on Indigenous Knowledge and Cultural Change ......................46
Brief Chronological Sketch of Research Done Among the Kaiabi ................. .............50
Personal N am es and O rthography ................................................ .............................. 53
Structure of Dissertation .................. ..................................... ....... .......... 55

2 FIELD W O R K A N D M ETH O D S......................................... ............................................62

Previous Fieldw ork ............. ........ ............. .. .... .. ... .......... ............. 62
Fieldwork Carried Out from 2002 to 2004 ................................................ ............... 63
Fieldwork Carried Out from 2006 to 2007 ................................................ ............... 65
Objectives, Research Questions and Hypotheses ............................................................ 66
R research D design ........................................................... ................ 72
D ata A n a ly sis .................................................................................................................... 7 5

3 THE KAIABI IN TAPAJOS AND THE TRANSFER TO XINGU PARK ..........................79

Intro du action ....................................................................................................................... 7 9
The K aiabi in the Tapaj6s Region ..................................................................... 80
Contact with the Villas B8as Brothers........................ ...... ........................ 87
The Indigenous Noble Savage and the Creation of the Xingu Park......................................90
'Xingu has Everything an Indian Could Wish': the Transfer to the Xingu Park ...................94
Conclusion ................ ......... .....................................................100









4 THE KAIABI AFTER THE DIASPORA: ADAPTATION AND LAND STRUGGLES
IN NEW TERRITORIAL CONFIGURATIONS ............................................ ...............104

Introdu action .................. ...................................... ......... .. ... ............................ 104
The Ones who Left: The Adaptation to Xingu Park after the Transfer...............................105
The Ones who Remained: The Adaptation of Rio dos Peixes and Teles Pires Groups
and Processes of Land D em arcation .............................. ...................................110
R io dos Peixes ........................................................................ ............... 111
T eles P ires ............... ... ......... ......... .......... ...................117
Indigenous Land Demarcation in Brazil, and Kaiabi Land Claims..................................122
B atel o Indigenous Land ............... .................................................. ................. .....126
T I K ayabi in T eles P ires ....................................................................... .... 13 1
A rraias R iver in X ingu R egion.............................................. ............................ 135
C onclu sion .......... .... ..... .....................................................138

5 ENVIRONMENTAL CONTRASTS, ADAPTATION AFTER DISPLACEMENT,
AND NATURAL RESOURCE MANAGEMENT ACROSS THREE KAIABI LANDS..145

Introduction ............ ... .. ............. .. .... ...... ........ .................. ........... 145
Environmental Contrasts between the Three Areas................................... ............... 146
Environmental Conservation Status and Threats............................................ .................. 149
Environmental Adaptation, Natural Resource Use and Management among the Xingu
K aiab i ............................................................................................1 5 7
C o n c lu sio n ................... ......................................................... ................ 1 6 6

6 CURRENT TERRITORIAL ORGANIZATION AND POLITICAL
CONFIGURATION ACROSS THREE KAIABI LANDS...............................................175

Introduction .... ............... ........... ..... .. ....... ..... .......................... 175
Overview of the Areas Currently Occupied by the Kaiabi............................176
X ingu Indigenous P ark ........................... ................................................. ............ 177
Apiaka-Kaiabi Indigenous Land in Rio dos Peixes...................................................... 182
Kayabi Indigenous Land in Teles Pires...................................................................... 185
N otes on the D em ography of the K aiabi ..................................... ......................... .......... 189
The Political Empowerment of the Xingu Kaiabi ............................................................191
The Formation of an Indigenous Political Leader ................................................. 196
The Creation and Sustaining of A TIX .................................... .................................... 201
Community-based Projects and Initiatives...... .............. ................................ 209
The K um ana Project ................... .... ...... ..... .............. ........ .. .. .... .. ........ .... 211
T he K aiabi A raa P project ..................................................................... ... ..................2 12
The M unuw i Project ..................................... .............. ........................ 215
Benefits and Disadvantages of Community-based Projects in Xingu...........................216
Itaok and Kawaip Associations in Rio dos Peixes and Teles Pires .....................................222
Peoples' Perceptions on the Work of Kaiabi Indigenous Associations..............................225
C o n clu sio n ............................ ........................................... 2 3 0





10









7 COMPARATIVE SOCIO-ECONOMIC ANALYSIS OF FOUR KAIABI VILLAGES....245

In tro d u ctio n .......... ................... ....................................................... 2 4 5
Overview of Villages Organization .......................................................... .............. 248
Capivara and Tuiarare Villages at Xingu Park.............................................................248
Tatuy Post and Novo Horizonte Village in Rio dos Peixes........................................252
Kururuzinho Post and Village in Teles Pires ..................................... .................255
Comparative Socioeconomic Analyses of Four Villages ...............................................257
People interview ed and A ge Classes........... ........................................ ................257
Places of B irth ..................... ................. ................ .................... ....... 259
Interethnic M ixing .......... ...... .. ........ ...................................... .................... 260
Functions in the Village, Sources and Stability of Income .......................................262
Travel and Dislocations Between the Villages............................ ............... .........268
Form al Schooling ................................... ............... .. ........... 271
Language Proficiency ......... .. ........ .. .... ... ............................ ...............279
Comparisons and Relationships between Socioeconomic Variables ................................284
C conclusion ......... .... .............. .................................... ...........................286

8 THE DYNAMICS OF WEAVING KNOWLEDGE ACROSS FOUR KAIABI
V IL L A G E S ............................................................... ......................... 13

Introduction ....... ...................... ......... ...... ............................... 313
W leaving Culture into M material Objects............................ ............................... ........... 315
Types and U ses of Basketry and Textile Item s ........................................ .....................320
Tw ill-plaited D designed B askets......... .......................... ................... ............... 323
Stages of Basketry W eaving.................................................................................... 324
Natural Resources Used in Basketry and Textile Production................................... ......326
Weaving Knowledge: Social Meaning, Transmission, Distribution and Change ................333
Social M meaning and Traditional Learning M echanisms..................... .....................333
Knowledge Distribution of W oven Item s................................................................336
Knowledge Distribution and Transmission of Designed Basketry and Textiles...........341
Age of learning ......................................... ........ ....................... 343
Reviewing models and mechanisms of knowledge transmission ..........................345
Distribution of knowledge on graphic designs...................................................... 351
Weaving Cosmology, Shamanism and Symbol ........................................... ...............367
Sham ans, baskets, cotton and textiles....................................... ......................... 372
Taboos related to weaving basketry and textiles .............. ...................... ............... 374
B asketry, Textiles and M markets ................................................. ..............................375
Relationship between Weaving Knowledge, Language Proficiency and Socio-economic
A aspects ..................................... ...................................... 380
C o n clu sio n .. ....... ..... ........ .. ............. .. ............................................................3 9 1

9 CONCLUSIONS: A SYSTEMS APPROACH TO THE STUDY OF CULTURAL
CH AN GE ..................................................................................... 433

Collapse and Renewal: Displacement, Socio-Ecological Systems and the Kaiabi .............433
Indigenous Knowledge, Displacement and Place............... ...... ..................... 440









Empowerment, Western Institutions and Territorial Control............................................443
Processual Studies on Indigenous Knowledge Systems.............. .... .................446
Applications and Policy Recomm endations ........................................ ...... ............... 455
F u tu re S tep s ............................................................................. 4 5 7

APPENDIX

A CODEBOOK USED IN THIS RESEARCH ............................................. .....................460

B NAMES AND MEANINGS OF BASKETRY AND TEXTILE DESIGNS .......................465

C CODES AND POSSIBLE NAMES FOR BASKETRY AND TEXTILE DESIGNS .........468

D CATALOGUE OF KAIABI BASKETRY DESIGNS ................................................ 474

E CATALOGUE OF KAIABI TEXTILE DESIGNS .................................. ...............510

L IS T O F R E F E R E N C E S .................................................................................. ..................... 19

B IO G R A PH IC A L SK E T C H ......................................................................... ... .....................544









LIST OF TABLES


Table page

5-1 Plant resources of cultural and economic importance mentioned in free listings
conducted with men and women in four Kaiabi villages.....................................171

6-1 Evolution of Kaiabi population through time. ............... ... ............. ................. 237

6-2 Synthesis of the results of the Kaiabi Araa project, in terms of objects produced and
people who have participated in the four workshops developed in Kururuzinho and
Tuiarare villages between 2004 and 2006. ........................................... ............... 238

6-3 Comparison between political participation, perception of indigenous organizations'
work and preference for future projects between four Kaiabi villages............................239

7-1 Number and gender of people interviewed in four Kaiabi villages..............................299

7-2 Functions and occupations in the villages. ........................................ ............... 303

7-3 Levels of stability of income for men and women in four Kaiabi villages......................306

7-4 Level of formal schooling for people interviewed in four Kaiabi villages......................309

7-5 Proficiency in native language for people interviewed in four Kaiabi villages.............310

7-6 Results of logistical regression analysis for selected socio-economic variables,
having language proficiency as the dependent variable. ...........................................312

8-1 Basketry items and textile related items produced by Kaiabi men..............................397

8-2 Main natural resources used in Kaiabi basketry and textiles................. .....................404

8-3 Plants currently used as arumd substitutes by the Kaiabi people..............................405

8-4 Basketry items and textile related items produced by Kaiabi men with respective
frequencies in each village by weavers and by total of men included in the sampling...406

8-5 Textile item s produced by Kaiabi women ............................................ ............... 409

8-6 Textile items produced by Kaiabi women with respective frequencies in each village
by weavers and by total of women included in the sampling .............. ... ...............410

8-7 Results of logistical regression analysis for selected socio-economic variables,
having weaving knowledge as the dependent variable ..................................................430









LIST OF FIGURES


Figure page

1-1 The adaptive cycle, used to understand change and reorganization of socio-
ecological systems. Source: Gunderson and Holling (2002). .................... ..............59

1-2 A possible representation of a systemic approach to the study of a cultural domain, in
this case w leaving know ledge ...................................................................................60

1-3 Another possible representation of a systems approach to the study of factors
involved in the resilience and change of Kaiabi cultural system................... ..............61

3-1 Map showing the location of Kaiabi ancestral territory and the demographic situation
of the group betw een 1955 and 1966 .................................................................. 102

3-2 Map of Xingu Park and location of Arraias River close to the Tywape Vigilance Post
(PIV ) in the north western region. ............................................ ............................ 103

4-1 Location of indigenous lands in Brazil. In the red polygon, location of Xingu Park
(PIX), Apiaka-Kaiabi Land, TI Kayabi in Teles Pires and TI Batelao (claimed). ..........141

4-2 Detail of the location of the three areas occupied by the Kaiabi in Mato Grosso: PI
Xingu, TI Apiaka-Kayabi and TI Kayabi, marked with red triangles ...........................142

4-3 Map of delimitation of TI Batelao in Rio dos Peixes region, showing the location of
old K aiabi villages. ...................................................................... 143

4-4 Map of delimitation of TI Kayabi in Teles Pires region ................................................144

5-1 Indigenous lands (green), federal and state protected areas (blue and brown) and
deforestation in the Brazilian Am azon. ........................................ ....................... 169

5-2 Map showing the municipalities that surround Xingu Indigenous Park..........................170

5-3 Selected plant resources most cited by women and by men interviewed in four
K aiabi villages. ......................................................................... 174

6-1 Location of the indigenous lands currently occupied by the Kaiabi: Xingu Indigenous
Park, TI Apiaka-Kaiabi (Rio dos Peixes) and TI Kayabi (Teles Pires). ........................234

6-2 Map of Xingu Indigenous Park. It is possible to distinguish the upper Xingu region
and PI Leonardo, the middle section and PI Pavuru and the northern portion or lower
section of the Park where PI Diauarum is located. Source: ISA (2009 b).......................235

6-3 Evolution of Kaiabi population in the three areas occupied by the group, from 1955
to 2007 ......................................................... .................................236









6-4 Degree of participation in political meetings amongst men (M-male) and women (F-
female) in four Kaiabi villages. CA=Capivara; TU=Tuiarare; RP=Rio dos Peixes;
TP= T eles Pires. ...........................................................................242

6-5 Perceptions of people (M-male and F-female) about the work carried out by local
organizations in Capivara (CA), Tuiarare (TU), Rio dos Peixes (RP) and Teles Pires
(TP) villages ................................... ............................ .... ...... ........ 243

6-6 Preference of communities for future projects to be developed by indigenous
organizations in four Kaiabi villages. CA-Capivara; TU-Tuiarare; RP-Rio dos
Peixes; TP-Teles Pires. ............................................. ......................... 244

7-1 Location of Capivara and Tuiarare villages on the northern portion of Xingu Park.
Source: Instituto Socioambiental, 2008. ............................................... ............... 290

7-2 Map of houses of Capivara village in 2002, around the soccer field.............................291

7-3 Map of Capivara village in 2007, showing the old village (front) with the school
(yellow building) and the new village (back). Drawing by Sirakup Kaiabi ...................292

7-4 Map of Tuiarare village in 2002 showing houses, health unit and school. Drawing by
Tamakari Kaiabi............ ... ...... ............. ....... ............ 293

7-5 Map of Tuiarare village in 2007. Drawing by Apurina and Piraju ..............................294

7-6 Map of the planned new Tuiarare village (2007) located on the back area of the old
village, with the "house of culture" or school of culture in the center, surrounded by
residents' houses. Drawing by Apurina and Piraju............................... ............... 295

7-7 Map of Rio dos Peixes Post (Tatuy) showing the river, the road which leads to Juara
and the path to Novo Horizonte village, located adjacent to the post. Drawing by
Sim one Athayde, 2007 .................. ....................................... .............. 296

7-8 Map of the houses in Novo Horizonte village at Rio dos Peixes. Drawing by Simone
A thayde, 2007. ...........................................................................297

7-9 Map of Kururuzinho Post and village at Kayabi Indigenous Land in Teles Pires river.
The airplane landing strip is located at the back of the village. Drawing by Tangeu'i
K aiab i, 2 0 0 7 .......................................................................... 2 9 8

7-10 Age pyramids for people interviewed in each age class by gender in four Kaiabi
villages. Age classes based on Athayde (2003). .................................... ...............300

7-11 Age pyramids for the Xingu Kaiabi population, from 1970 to 1999. Source: Pagliaro
(2 0 0 5) ............... ............................ ................................................3 0 1

7-12 Places of birth for people interviewed in four Kaiabi villages. .................................302









7-13 Total monthly income by village, in dollars, from waged officers, retirement
pensions and family pensions. CA=Capivara; TU=Tuiarare; RP=Rio dos Peixes; TP=
T eles P ires. ........................................................................... .. 3 04

7-14 Sources of income for people interviewed in four Kaiabi villages. CA=Capivara;
TU=Tuiarare; RP=Rio dos Peixes; TP=Teles Pires................................. ... ................ .305

7-15 Percentage of women and men interviewed in four villages, travelling between the
three Kaiabi areas. CA=Capivara; TU=Tuiarare; RP=Rio dos Peixes; TP=Teles
P ires. F =fem ale; M =m ale. .................................................................... .....................307

7-16 Reasons for travels between the three Kaiabi areas among women and men
interviewed in four villages. CA=Capivara; TU=Tuiarare; RP=Rio dos Peixes;
TP=Teles Pires. F=fem ale; M =m ale. ........................................ .......................... 308

7-17 A comparison of selected socioeconomic variables between four Kaiabi villages.
CA- Capivara; TU- Tuiarare; RP- Rio dos Peixes; TP- Teles Pires..............................311

8-1 Graphic representation of domains or modules of knowledge involved in weaving
knowledge, considered in this research. Broad categories considered: 1) Form; 2)
Function; 3) M materials; 4) M earnings; 5) Designs. ................................ ..................395

8-2 Types of Kaiabi baskets grouped according to semantic categories and uses, with
distinction of use by men, women or both. .............................. ................................. 396

8-3 Stages of basket w leaving. ....................................................................... ....................402

8-4 Kaiabi ethnomathemathics: counting, grouping and design structure. The design
woven in the center of the basket square is the "I'yp" pattern (Mendes, 2001). .............403

8-5 Distribution of men who weave each type of basketry item by village. CA-Capivara;
TU Tuiarare; RP Rio dos Peixes; TP Teles Pires. ............. .................................. 411

8-6 Distribution of women who weave each type of textile item by village. CA-Capivara;
TU Tuiarare; RP -Rio dos Peixes; TP Teles Pires. .............................................412

8-7 Percentage of men in all villages, who weave each type of basketry item..................413

8-8 Percentage of women in all villages, who weave each type of textile item.....................414

8-9 Percentage of women and men who weave designed baskets and textiles in four
Kaiabi villages. CA- Capivara; TU Tuiarare; RP Rio dos Peixes; TP Teles
P ire s......... ....... ... .......... ................. ................................................4 15

8-10 Age of learning designed baskets and textiles among female (F) and male (M)
weavers in four Kaiabi villages. CA- Capivara; TU Tuiarare; RP Rio dos Peixes;
T P T eles P ires. ........................................................................ 4 16









8-11 Typology of modes of cultural transmission developed by Cavalli-Sforza and
Feldman (1981) and Hewlett and Cavalli-Sforza (1986) ...............................................417

8-12 Ways of learning to weave designed baskets among Kaiabi men in two villages in
X ingu in 2002. ...........................................................................4 18

8-13 Ways of learning to weave designed baskets and textiles among Kaiabi women and
m an in all villages in 2007. ......................... ...... .... ......... ...............419

8-14 A model for the study of indigenous knowledge domains and mechanisms of
transmission based on the Kaiabi case. Adapted and expanded from Cavalli-Sforza
and Feldman (1981) and Hewlett and Cavalli-Sforza (1986) .......................................420

8-15 Distribution of knowledge on different basketry designs among Kaiabi men,
comparing total men interviewed and only basketry weavers......................................421

8-16 Distribution of knowledge on designs among men from Capivara and Tuiarare
villages interviewed in 2002 and 2007. ........................................ ....................... 422

8-17 Percentage of men who weave different designs in each village, grouped by classes. ...423

8-18 Competence scores on consensus analysis regarding agreement on naming the
different basketry designs among men in four Kaiabi villages. CA- Capivara (red);
TU-Tuiarare (blue); RP Rio dos Peixes (orange); and KU Kururuzinho (green-
T ele s P ire s)......................................................................... 4 2 4

8-19 Similarity between men from Capivara (red) and Tuiarare (blue) villages regarding
names given to basketry designs. Non-metric multidimensional scaling (MDS)
analysis of similarity.......................................................... 425

8-20 Similarity between men from Rio dos Peixes (orange) and Kururuzinho (green)
villages regarding names given to basketry designs. Non-metric multidimensional
scaling (M D S) analysis of sim ilarity. ........................................ .......................... 426

8-21 Distribution of knowledge on textile designs among Kaiabi women in four villages.....427

8-22 Competence scores on consensus analysis regarding agreement on naming the
different textile designs among women in four Kaiabi villages. CA- Capivara (red);
TU-Tuiarare (blue); RP Rio dos Peixes (orange); and KU Kururuzinho (green-
Teles Pires)............... 428
T e le s P ire s) ....................................................................................................................... 4 2 8

8-23 Linear regressions showing relationship between age and basketry knowledge among
Kaiabi men: A) Relationship between age and number of basketry items woven by
Kaiabi men; B) Knowledge of names for basketry designs; C) Ability to weave (or
use) of basketry designs. .......................... .................. ........................ 429









8-24 A comparison of selected socioeconomic variables, including language proficiency
and weaving knowledge (ability to weave) between four Kaiabi villages. CA-
Capivara; TU- Tuiarare; RP- Rio dos Peixes; TP- Teles Pires................................431

8-25 A model for understanding interactions among variables related to Kaiabi language
proficiency and weaving knowledge. ........................................ .......................... 432

9-1 A model based on the adaptive cycle from the socio-ecological systems theory,
showing collapse, reorganization, exploitation and conservation events applied to the
history of K aiabi people .................. ..................................... .. ........ .... 459









LIST OF ABBREVIATIONS

ADR Administracgo regional da FUNAI FUNAI's regional administration

AIK Associaco Indigena KIsedj6 (Suya) Kis6dj6 Indigenous Association

AIKK Associaco Indigena Kawaip Kaiabi Kawaip Kaiabi Indigenous Association

AIS Agente Indigena de Saude Indigenous health agent

ATIX Associacgo Terra Indigena Xingu, Xingu Indigenous Land Association

AVA Associacgo Vida e Ambiente Association life and environment

BIE Bilingual intercultural education

CIMI Conselho Indigenista Missionario Indigenist missionary council

CEMAT Centrais Eletricas Matogrossenses Electrical company of Mato Grosso

COIAB Coordenaqco das Organizac6es Indigenas da Amaz6nia Brasileira, Coordination
of Indigenous Organizations of Brazilian Amazon

COICA Coordinadora de las Organizaciones Indigenas de la Cuenca Amazonica,
Coordination of Indigenous Organizations of the Amazon Basin

CONOMALI Companhia Colonizadora do Noroeste Mato-Grossense

DFDR Development-forced displacement and resettlement

DOU Diario official da Unido Brazilian federal government official newspaper

DSEI XINGU Distrito de Saude Especial Indigena do Xingu Xingu Indigenous Health Unit

EPM Escola Paulista de Medicina Sdo Paulo School of Medicine

FAB Forca Aerea Nacional Brazilian Aeronautic Army

FBC Fundacgo Brasil-Central, Central Brazil Foundation

FMV Fundacgo Mata Virgem Mata Virgem Foundation

FUNAI Fundacgo Nacional do Indio, National Indian Foundation

FUNASA Fundacgo Nacional de Saude National Health Agency

GT Grupo de trabalho para quest6es de terras indigenas (FUNAI) Working group
for study of Indigenous lands claims









GTME Grupo de Trabalho Missionario Evangelico Group of evangelic missionaire
work

IBAMA Insttituto Nacional do Meio Ambiente e dos Recursos Naturais Renovaveis -
Brazilian Environmental Agency

INCRA Instituto Nacional de Colonizacgo e Reforma Agraria National Institute for
Colonization and Agrarian Reform

IK Indigenous knowledge

IPAM Instituto de Pesquisa Ambiental da Amaz6nia Amazonian Institute for
Environmental Research

ISA Instituto Socioambiental Social-Environmental Institute

ITAOK Associacgo Indigena Itaok (Kaiabi) Itaok Indigenous Association

MEC Ministerio da Educacgo Ministry of Education

MINC Ministerio da Cultura Brazilian Ministry of Culture

MMA Ministerio do Meio Ambiente Brazilian Minsitry of the Environment

NGO Non-governmental Organisation

NRF Norwegian Rainforest Foundation

NTFP Non timber Forest products

OPAN Operacgo Anchieta, Anchieta Operation

PDPI Projetos Demonstrativos dos Povos Indigenas Indigenous Demonstrative
Projects

PI Posto Indigena Indigenous Post

PIV Posto Indigena de Vigildncia Indigenous Vigilance Post

PIX Parque Indigena do Xingu Xingu Indigenous Park

PPTAL Projeto Integrado de Proteco as Populaces e Terras Indigenas da Amaz6nia
Legal

PROESI Programa de Educacgo Superior Indigena Intercultural Program for indigenous
intercultural education

SEMA Secretaria do Meio Ambiente Secretary of the Environment









SEDUC Secretaria de Estado de Educacgo do Mato Grosso Secretary of Education of
Mato Grosso State

SIL Sociedade Intemacional de Linguistica Summer Institute of Linguistics

SPI Servigo de Protecgo ao indio Indian Protection Service

SPU Servigo de Patrim6nio da Uniao, Service of Union Patrimony

SUDECO Superintend6ncia do Desenvolvimento do Centro Oeste Agency for
Development of the Center-western region of Brazil

TEK Traditional Environmental (or Ecological) Knowledge

TI Terra Indigena, Indigenous Land

UNEMAT Universidade do Estado de Mato Grosso University of Mato Grosso State

UNESCO United Nations Educational, Scientific and Cultural Organization

UNI Uniao das Nac6es Indigenas Indigenous Nations Union

UNIFESP Universidade Federal de Sao Paulo Federal University of Sao Paulo

Y IKATU XINGU Campanha Agua Bonita do Xingu Xingu River Campaign









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

WEAVING POWER: DISPLACEMENT, TERRITORY AND INDIGENOUS KNOWLEDGE
SYSTEMS ACROSS THREE KAIABI GROUPS IN THE BRAZILIAN AMAZON

By

Simone Ferreira de Athayde

August 2010

Chair: Marianne Schmink
Major: Interdisciplinary Ecology

Indigenous lands are responsible for the conservation of nearly 21% of the Brazilian

Amazon, and are the main barrier against deforestation. This research contributes to the

understanding of the relationship between political empowerment, socio-cultural resilience and

territorial control among Amazonian indigenous peoples. I sought to explore which factors may

lead to the persistence or loss of indigenous knowledge after geographical displacement. I argue

that cultural and environmental resilience are intertwined, so that where traditional knowledge is

maintained, there will be greater territorial control among indigenous peoples in the Amazon. I

apply a systems approach to explore the effects of historical, environmental, political, socio-

economic and cultural factors in their interaction with a specific domain of indigenous

knowledge.

The Kaiabi speak a language of the Tupi-Guarani stock and are great agriculturalists and

basket makers. The majority of the population was transferred by the Brazilian government from

their ancestral territory in the Tapaj6s River basin to the Xingu Park region from 1950 to 1966.

Two small groups remained in the ancestral land, one on the Peixes River and other on the Teles

Pires River. Transfer to the park brought changes in Kaiabi social and political organisation and

in the access to and management of natural resources. Forty years after the transfer, the Xingu









Kaiabi have adapted to the new conditions, creating mechanisms for cultural perpetuation and

territorial control. By contrast, the groups who remained in the ancestral land lost many aspects

of their traditional lore.

Cross-sectional and longitudinal studies were carried out in order to compare the dynamics

of knowledge related to basketry and textile weaving among 114 men and 110 women in four

villages across three Kaiabi groups in a five year period. Factors that led to the cultural

persistence and political empowerment of the Xingu Kaiabi in contrast with the other two groups

were also explored, as well as peoples' perspectives on the role and work of political

organizations.

Results show that knowledge is being both innovated and eroded among the Kaiabi, and

that agency, leadership and innovation are critical assets in enabling cultural resilience. Whereas

in Xingu and Teles Pires areas there is innovation and new knowledge being generated- with

younger generations still learning- in Rio dos Peixes, knowledge and native language have been

eroded at a faster pace. Promoting the persistence of native language and adapting schools to

include both indigenous and western knowledge could help to reverse the process of loss of

cultural resilience in Rio dos Peixes. Greater political empowerment in Xingu, allied to the

development of community-based projects, has influenced indigenous knowledge perpetuation

and mechanisms for its transmission. In the Kaiabi case, territorial isolation combined with

political support and local leadership led to greater cultural and environmental resilience of the

Xingu group in comparison to the other two areas.









CHAPTER 1
INTRODUCTION

The Kaiabi: Diaspora, Resilience and Change of an Amazonian People 1

The Kaiabi are a Tupi-Guarani speaking people who originally occupied several

tributaries of the Tapaj6s River in the southern Brazilian Amazon. In light of an imminent

government-sponsored agricultural colonization scheme in the area, and in order to minimize

conflicts with rubber tappers and settlers, the Brazilian federal government induced the

relocation of the Kaiabi several hundred kilometers to the southeast, to what is now the Xingu

Indigenous Park (Grinberg, 2004). While most Kaiabi relocated to Xingu between 1950 and

1966, two smaller groups refused to leave, and still live in or close to their ancestral region, one

in Rio dos Peixes and the other along the Teles Pires River.

After the diaspora, the process of re-settlement, fragmentation, sedentarization and

contact with other indigenous peoples and with non-indigenous social and economic systems

has created rapid and profound changes in territorial and social organisation amongst the three

Kaiabi groups, including shifts in their knowledge systems and natural resource management

practices. The three groups to which we are going to continuously refer in this dissertation are

hereby named the Xingu Kaiabi, the Rio dos Peixes or Tatuy Kaiabi and the Teles Pires or Para

Kaiabi.

The recently resettled Xingu Kaiabi have undergone a process of social and political

revitalization enhanced by the unique conditions found at Xingu Park and by access to

international funding and technical support that came to Amazonia with the environmentalist

movement (Oakdale, 1996; 2004; Athayde et al., 2009). Currently occupying an area of nearly


1 Parts of this dissertation are based on my previous writings, and include sections of my Master's thesis (Athayde,
2003) and of published book chapters and papers, mainly from Athayde (21 I14); Senra et al. (21 I14); Silva and
Athayde (2. 1'4); Athayde et al. (2006) and Athayde et al. (2009).









one million hectares in the northern portion of the Xingu Park, they have developed strong

leadership skills, maintaining, in a changing and adaptive way, a wide repertoire of traditional

practices and institutions, including their language, crop plants, material culture, as well as

festivals and rituals. In an apparent contradiction, Kaiabi groups who remained in the ancestral

region are numerically inferior to the Xingu Kaiabi, and have experienced a higher degree of

assimilation into Brazilian non-indigenous society and loss of traditional knowledge.

The Kaiabi have a sophisticated agriculture and culinary, as well as an elaborated

material culture, in which the most technically and artistically developed items are the designed

and painted twill-plaited baskets (Ribeiro, 1986; Athayde, 2003; Athayde et al., 2009). These

baskets have been produced by Kaiabi men according to different social meanings and

economic purposes, which have been shifting along with the changes faced by the group after

their transfer to Xingu Park. Basketry designs have became strong symbols of Kaiabi identity,

and have been applied in other artifacts such as painted benches, club handles and sculpted

gourds, as well as body painting. While in Xingu Park basketry knowledge has been at the same

time eroded, perpetuated and innovated, it seems that in the other two Kaiabi areas, knowledge

of this cultural domain is eroding at a faster pace.

Contradictions and paradoxes are more of a rule than an exception when it comes to

cultural persistence and change among indigenous peoples in the contemporary world. Whereas

in Xingu Park the Kaiabi have managed to perpetuate their basketry knowledge, they have been

dealing with a critical lack of the main natural fiber used in the weaving work (Ischnosiphon

gracilis, Maranthaceae), named "arumd" in Portuguese (Athayde et al., 2006). This fiber is

abundant in the other two Kaiabi areas, where the knowledge has been lost. In Xingu, there is









knowledge but no raw material: in the ancestral land, there is plenty of raw material, but lack of

knowledge.

Kaiabi women traditionally produced textiles hammocks and straps for carrying babies -

using the twined weaving technique (Ribeiro, 1984/1985). After 1970, there was a technical

shift and innovation amongst Xingu women, in which they began to use a different weaving

technique (twilled), applying the basketry designs woven by men in the textiles. There was no

parallel to this shift by women residing in the other two Kaiabi areas, who have lost the

knowledge of weaving to different extents.

In this dissertation, I take elements from the theories and fields of historical ecology,

ethnoecology, cognitive anthropology, political ecology, systems ecology and ethnography to

advance the understanding of processes of indigenous knowledge systems persistence and

change after displacement. Why and through which mechanisms have the Xingu Kaiabi retained

their traditional lore, whereas the other two groups have lost it in different degrees? The main

question to be explored in this research relates to the processes, mechanisms and factors

involved in indigenous peoples' cultural retention or resilience and erosion or loss.

Weaving Indigenous and Scientific Knowledge through Collaboration

This dissertation is based on information collected during ten years of work as a

practitioner and associated researcher for the Xingu Program of the NGO Instituto

Socioambiental (ISA) among the Kaiabi at Xingu Park, and on visits to other Kaiabi lands for

short periods of time, one in 1999, when I spent one month with the Kaiabi at the Rio dos Peixes

village (Silva et al., 2000; Senra et al., 2001) and the other in September of 2004, when I

participated in travel to the Kururuzinho Indigenous Post at Kaiabi land in Para, with a group of

Kaiabi from Xingu (ATIX, 2004).









In 1999, I was in my palm thatched hut at Diauarum Post (Xingu Park) when I received

the visit of Aturi Kaiabi (his current new name is Jowosipep, which means Tortoise), at that

time teacher at Tuiarare school and prominent political leader. He asked me to help him to write

a project for FUNAI's educational department, to promote the revitalization of basketry and

textile knowledge among the Kaiabi in Xingu and the Teles Pires Kaiabi, who reside in the

Kururuzinho Post and village. According to him, weaving knowledge was being lost in Xingu,

and in Teles Pires the situation was even more critical, with very few men still knowing how to

weave baskets and no women knowing how to weave a hammock anymore. He wanted to

develop a project with the Kaiabi from Teles Pires, because he was born there, and his father

took him to Xingu when he was only ten years old. Besides this, there are other kinship ties

between Tuiarare village in Xingu and Kururuzinho village in Teles Pires. It would also be too

difficult and ambitious to include the Rio dos Peixes Kaiabi in this first community-based

project.

Aturi had been talking to Eroit Kaiabi, a teacher in Kururuzinho village, about his idea of

promoting cultural exchange between the two groups: a group of Kaiabi teachers (men and

women) from Xingu would dislocate to Para to teach people there how to weave baskets and

textile items (hammocks, straps for carrying babies etc). They would also promote weaving

workshops for men and women, especially youth, in the Xingu Park, as well as activities for

management of the arumd plant. I found the idea compelling, spectacular: "I am in", I told Aturi

at that moment without envisioning that ten years later I would be telling this story to you. At

that time, I had already fallen in love with Kaiabi baskets, and had taken photos of baskets in

different Kaiabi villages. As a biologist, I had also been coordinating a participatory project for

research and management of the arumd plant (Athayde et al., 2006). What motivated me the









most was the fact that I would be working for them on their own idea, their project, from the

beginning to the end. This project was later approved by the PDPI (Projetos Demonstrativos dos

Povos Indigenas) in 2004, lasting two years, but the whole process took seven years. The

project, named Kaiabi Araa, was one of the winners of the "Indigenous Cultures" award from

the Brazilian Ministry of Culture (MINC) in 2007.

Later on, I thought that western science would have a place and a role in the project, in

the investigation of the processes of knowledge creation, erosion, transmission, distribution,

innovation and change related to weaving practices among Kaiabi men and women. In my

Master's dissertation in Ethnobotany defended in 2003 at University of Kent, England, I

investigated basketry knowledge among Kaiabi men in two Xingu villages: Capivara and

Tuiarare. However, when I finished the first thesis, the project had not begun yet; it was still

under consideration by the PDPI. In 2004, we traveled to Kururuzinho village in Teles Pires

along with ten Kaiabi men and women (weaving teachers) from Xingu. It was the first time I

had visited the Kaiabi from Para, as they are called, because the two other areas where the

Kaiabi live (Xingu and Rio dos Peixes) are in Mato Grosso.

Since my collaboration with the Kaiabi began, we have been organizing educational

books on basketry and textiles, containing myths, stories and drawings related to baskets and

textile production and products, as well as photos of basketry designs taken by myself and by

other researchers, some of them acquired through donation by national and international

museums. These books have helped the Kaiabi to keep and regain the knowledge of basketry

designs, due to the fact that some weavers (men and women alike) are able to reproduce the

design in the baskets or textiles by looking at the photos in the book. A quantitative









investigation on the ability of men and women to learn from the books was also part of this PhD

project, and is presented in Chapter 8.

After getting to know all three Kaiabi groups, I thought that research comparing the

situation of knowledge systems between them more than forty years after the transfer would be

very interesting, almost required, even with all the challenges (and miles of dirty roads and

rivers to be covered on buses, trucks and boats) ahead. My nine year old son Adriano has

accompanied me during all this process and also during my fieldwork travels to all Kaiabi areas.

This would be the first scientific research done including all three Kaiabi groups, giving

voice to Kaiabi women, and also the first longitudinal study done among the Kaiabi, since I

used baseline data collected in 2002 to compare with data collected in 2007. The cultural

domain was already chosen: weaving knowledge, basketry and textiles, men and women.

Nevertheless, this research also sheds light on other realms of scientific inquiry, which would

ultimately help to explain why, how, and to what extent, the Xingu Kaiabi became more

culturally resilient than the other two groups.

Fields of Inquiry and Possible Applications

Indigenous Knowledge Systems and Biocultural Diversity

In anthropology and related fields of inquiry, research on indigenous knowledge (IK)

systems has had multiple ramifications and applications in the last decades such as: sustainable

development; economic use of biodiversity; natural resource management and conservation;

intellectual property rights issues; political empowerment and grassroots movements; from

intercultural education and indigenous schooling to processes of advocacy, in the creation and

implementation of public policies concerning indigenous rights (Sillitoe et al., 1998; Berkes,

1999; Berkes et al., 2000, Colchester, 2000; Ellen and Harris, 2000; McGovern, 2000; Laird,









2002; Posey, 2004; Hall, 2006; Nugent, 2006; Posey and Balick, 2006; Sillitoe, 2007; Zent, 2009

a).

Berkes et al. (2000), define traditional or indigenous knowledge as cumulative and

adaptive by nature, tested by trial-and-error and transmitted through generations orally or by

shared practical experiences. One important difference between IK and western science is that

indigenous knowledge is largely dependent on social structure and mechanisms. Mechanisms

for intergenerational transmission of knowledge are embedded in social structure, which is also

dynamic and changes according to internal and external historical factors. I prefer to use the

wording indigenous knowledge instead of "traditional knowledge," which might have

misleading interpretations such as being perceived as contrary to modern or scientific

phenomena, or as denoting simple, primitive, anachronistic, irrational, stagnant, outdated or

immutable knowledge (Kearney, 19996; Ellen and Harris, 2000; Zent and Maffi, 2007). I like

the emphasis given by Zent (2009 a) to the term "traditional," which would refer to "cumulative

result of the collective historical experience of groups and individuals, usually handed down

from previous generations through customary modes of transmission." For him, the concept of

traditional knowledge would also relate to it being embodied in language, classifications,

beliefs, values, rituals, social institutions and daily practices. Traditional might also be used

when referring to knowledge held by other local peoples, who cannot be defined as indigenous,

but hold specific knowledge which is based on their close relationship with nature and the place

where they live: riverine communities, rubber tappers, "caboclos" or "mestizos," and others.

Berkes and Folke (2002) mention that the word traditional might be used to mean historical and

cultural continuity, but without losing sight of the fact that all societies are dynamic and in a

constant process of change; thus what is traditional is continuously redefined.









As environmental conservation awareness and crisis have exponentially progressed in the

last decades, greater attention and resources have been directed, among policy makers,

researchers and development practitioners, to the domain of environmental knowledge, also

called traditional environmental knowledge (TEK). TEK would have a big role to play in

natural resource management, biodiversity conservation and development of the Amazon and

elsewhere (Posey, 1999; Balick, 2006; Hall, 2006; Sillitoe, 2007; Zent and Maffi, 2007). Hall

(2006:328) considers TEK as a "component of social capital for promoting economic progress

and supplying environmental services" which has been neglected by official planners and

policy-makers. Cultural understandings of the environment might be instrumental in nature

conservation initiatives and programs, providing knowledge of species requirements, ecosystem

dynamics, sustainable harvesting levels and ecological interactions (Berkes and Folke, 2002;

Pilgrim et al., 2008). According to Brush (1993), TEK reflects the ecological adaptation of

humans to diverse environmental settings, thus it can serve as a ground for the development of

initiatives to conserve biological diversity. Furthermore, biodiversity conservation and

management projects have been more successful when local knowledge was incorporated in the

process (Carroll and Meffe, 1994; Athayde et al., 2006; Zent and Maffi, 2007). Sillitoe (2007)

emphasizes that including indigenous knowledge in the context of participatory development is

likely to produce more successful interventions, since these will be culturally significant and at

the same time contribute to empower people that will participate in their formulation and

implementation.

Research and theoretical approaches developed in the past few decades in the fields of

ethnolinguistics, ecosystem health, agroforestry, agriculture, ethnobotany, ethnobiology, systems

theory, cultural and environmental anthropology have reinforced the idea of interdependence









between cultural and biological diversity (Inglis, 1993; Balee, 1994; Carroll and Mefe, 1994;

Denevan, 2001; Maffi, 2001; Gunderson & Holling, 2002; Harmon, 2002; Heckenberger et al.,

2003; Zent and Maffi, 2007). Scientific adoption of the idea of co-evolution between systems of

people and nature or social-cultural and ecological systems (Walters, 1986), led to the

development of the concept of"biocultural diversity," to highlight the historical interactions and

therefore mutual effects and impacts of human cultures on nature and vice-versa (Harmon, 2002;

Maffi, 2005; Maffi, 2007; Zent and Maffi, 2007). Therefore, if environmental knowledge is

eroded, there is greater probability that this would cause a negative effect on the ecosystem. In

the opposite direction, where more vigorous systems of TEK are maintained, the probability of

finding stronger indicators of ecosystem health is greater (Inglis, 1993; Zent and Maffi, 2007).

For cognitive anthropologists, culture can be interpreted and studied as "a partially

shared understanding of the world" (Boster, 1986:429). Under that model, culture could be

defined as shared knowledge, organized in cultural domains (Weller, 1987; Ellen, 2009).

Previous researchers have studied indigenous knowledge systems focusing in specific domains,

mainly medical knowledge (Garro, 1986; Soto, 2004); and ecological knowledge (Boster, 1986;

Zent, 1999; Reyes-Garcia, 2001; Hunn, 2002; Wyndham, 2002; Rival, 2009 and many others).

Technical knowledge, or knowledge related to object production, use and meaning, has been

studied in the Amazon by Wilbert (1975); Riviere (1992); Van Velthem (1998; 2001) and

Chemela (2008), among others. Ellen (2009) presents research on Nuaulu basket-making from

Indonesia, in which he reviews and criticizes the concept and fixity of cultural domain.

According to him, boundaries attributed to domains by anthropologists are many times

artificially defined, easy to challenge and, most important, have neglected or denied the

overlap between domains. Furthermore, the domains chosen or defined by researchers might









not reflect local categorization of knowledge. Ignoring or not taking into account overlap

between domains might limit our capacity to understand mechanisms and processes of

knowledge transmission. He gives an example of a plant species (eg. Ficus sp), which might

simultaneously be a member of various domains identified as"a plant, a food, a medicine, a

construction material, an element in landscape, a totem and so on." Therefore, changes in

knowledge transmission in one domain can affect others, positively or negatively. He states that

knowledge transmission "must be understood in terms of overlapping knowledge of non-

mutually exclusive domains." (Ellen, op cit., 246).

This research is not specifically about traditional environmental knowledge. However, it

includes aspects of Kaiabi environmental knowledge as it relates to mechanisms of adaptation to

Xingu Park and to the knowledge of plant species of cultural and economic importance in the

three areas, including those used in basketry and textiles production (Chapters 5 and 8). In regard

to the concept of biocultural diversity, I seek to verify to what extent the vitality of a given

domain of IK (in this case, weaving knowledge), might be used as an indicator of conservation

status. In other words, where knowledge (and by extension, culture) is maintained, territorial

control and thus biodiversity conservation will be greater than where knowledge is eroded. On

the other hand, this is a two-way road, taking in account that ownership of the land, and more

important, territorial control, especially in Brazil, often happen through political mobilization,

empowerment and access to resources. Therefore, when land rights are assured, the potential of

social-cultural reproduction might be grater (Surrales and Hierro, 2005). If this argument is

plausible, linguistic and artistic knowledge might be used as proxy elements to indicate the

vitality of both ecological and environmental knowledge systems among indigenous peoples

(topic explored in Chapters 7 and 8). This may have practical implications for development









projects, in which more flexibility could be given to indigenous community-based initiatives

who might want to prioritize cultural revitalization over strict conservation and natural resource

management actions (Pilgrim et al., 2009).

Social-Ecological Systems, Resilience and Change

The maintenance of cultural diversity and the knowledge, innovations and outlooks it

contains increases the capacity of human systems to adapt and cope with change (Brush, 1993;

Maffi, 1999; Gunderson & Holling, 2002; Harmon, 2002). Gadgil et al. (2003) sustain that local

knowledge (of any resource-user group) is relevant in systems of adaptive management. Zent

and Maffi (2007:2) and Dei et al. (2000) state that, "by maintaining their traditional knowledge

and technologies, people give themselves more options, greater control over their lives, and

greater leverage with which to negotiate the process of development and change more on their

own terms." If indigenous knowledge systems can be understood as social-cultural capital for

the future (Hall, 2006), the greater their flexibility and capacity to adapt while still ensuring

their transmission through generations and thus their perpetuation or conservation the greater

the power and potential of indigenous peoples to achieve self-determination and control over

land and resources. Similarly, the biodiversity of a specific place can be viewed as natural

capital the greater the biodiversity, the greater capacity to generate products, services and

adapt to change.

Berkes and Folke (2002) consider that the distinction between social and natural systems

is artificial and arbitrary. They, along with a group of scientists involved in interdisciplinary

research of the complex links between people and nature, have developed the concept of social-

ecological systems to refer to integrated and multi-scaled processes of humans and nature,

constantly interacting and adapting in time and space (Berkes et al., 2003).









Applied to social-ecological systems, the idea of resilience comes to rescue us from

submersion in the paradox of understanding change in continuity and vice-versa. The resilience

concept, applied originally to solid mechanics, is "the property of a material to absorb energy

when it is deformed elastically and then, upon unloading, to have this energy recovered"

(Wikipedia, 2008) or the capacity to absorb change, impact or stress and then return to a

previous state of dynamic stability. According to Holling and Gunderson (2002), the concept of

resilience applied to ecological systems has had different interpretations in the ecological

literature. They note that there is always a tension between efficiency and persistence, constancy

and change, or predictability and unpredictability. One definition which is suitable to

understand cultural change in indigenous societies is "the magnitude of disturbance that can be

absorbed before the system changes its structure by changing the variables and processes that

control behaviour" (Holling and Gunderson, 2002:28). Walker et al. (2004) define resilience as

"the capacity of a system to absorb disturbance and reorganize while undergoing change so as to

still retain essentially the same function, structure, identity, and feedbacks."

Greater resilience provides complex systems with the possibility and ability to persist in

response to shocks and disturbances (Gunderson, 2003). According to Walker et al. (2004),

adaptability in social-ecological systems can be interpreted as the capacity of actors in a system

to influence resilience. Therefore, we could interpret cultural resilience as the capacity of human

societies to absorb change and adapt to novel situations without collapsing or losing their

identity as a group. According to Gunderson (2003), social systems have a unique property of

responding to uncertainty through the generation of novelty, which would be key in dealing

with surprises or crises. Again, resilience brings both continuity and innovation in adapting to

change and also allows for processes of uncertainty through human agency and capacity to









innovate, thus changing a possible prediction of the future. The greater the social-cultural and

natural capital, the greater the resilience or capacity of adaptation to change, or in other words,

the capacity of perpetuation of a social-ecological system under shifting conditions.

Gunderson and Holling (2002) and their collaborators, along with theories about the

functioning of socio-ecological systems, have developed the concept of adaptive management

and the adaptive cycle. According to them, any given socio-ecological system may go through

phases of exploitation, conservation, release and reorganization (Figure 1-1). This concept is

easy to understand when we think about the process of forest recovery or secondary succession

after a fire event, for instance. The fire event is the phase of release, shock or creative

destruction, forcing that system to change. The re-growth of vegetation represents the

reorganization phase, in which innovation and restructuring occurs. The system undergoes an

exploitation phase of competition and accumulation of nutrients and biomass in which

connectedness and stability increase towards the stage of conservation, in which a grown forest

is formed. Despite the fact that the ecosystem properties might be the same before and after the

release event, this forest is not exactly the same as it used to be before. This concept has been

applied to different fields, not only in socio-ecological systems, but in the functioning of any

system that might present a cyclic functioning. In this case, I want to apply these ideas to the

dynamics of indigenous knowledge systems (see Chapter 9).

Going back to our statement in the previous section, as social-ecological systems are

interlinked, and mutually interacting with each other, a logical conclusion would be that

conserving knowledge would ultimately mean conserving nature, through an intricate web of

social, institutional, economic and political interactions which scientists aim to decipher,

understand and interpret. Our worldviews affect our decisions, which affect our actions, which









affect other people's lives, which affect the environment, in a chain reaction. Applying these

ideas to the Kaiabi case, we might observe that, in comparison to other Amazonian indigenous

groups (see Fausto, 2001, about the Parakand people), they have been actively engaged in

transforming their destiny through their own agency upon historical facts to which they were

subjects. Therefore, they might be seen as agents of continuity and change, reproducing and re-

working their traditional knowledge and forms of organisation in new territorial, socio-cultural,

political and economic contexts and configurations (Oakdale, 1996).

In this dissertation, I adopt a systems approach for the study of a specific cultural domain,

considering factors and elements that might affect that sub-system in different ways. The

domain of weaving knowledge might be viewed as a sub-system nested into territorial, political,

ecological, socio-economic, institutional and cultural systems (Figures 1-2 and 1-3). The

integrated study of the factors involved in social-ecological resilience, political empowerment

and territorial control among the Kaiabi will provide a ground for further research and action

concerning indigenous peoples' development and conservation programs. Research on Kaiabi

history, land struggles, socio-economic and political systems (presented in Chapters 3, 4 and 6)

is not the main subject of this dissertation, and it would be too ambitious (even an impossible

mission) to present an in-depth portrait and discussion of the consequences of induced

displacement in Kaiabi social, economic and political organization among all groups. My

intention here is two-fold: 1) to present a brief chronological description of each Kaiabi group's

history after displacement, using interviewees' testimonies and secondary data; and 2) to present

a comparative description of the current socio-economic, political and environmental landscape

in the areas occupied by each group. I believe that this information is needed for us to put the

puzzle pieces together in trying to unveil what caused cultural and environmental resilience









among the Xingu Kaiabi as compared with the other places. This information might serve as a

background for future comparative research on the Kaiabi, as well as for the elaboration of

development projects by the Kaiabi and/or their partners.

One interesting aspect to explore, which was previously studied among the Xingu Kaiabi

by anthropologist Suzanne Oakdale (1996:4), is the interplay between external and internal

structures of identity, agency and power among them, which ultimately has shaped their history

and trajectory along with broader scale Brazilian national politics. Following Oakdale's idea but

taking another theoretical path, building on the previous section, my approach to this topic

relates to the possibility to link agency, power, knowledge, and social-ecological resilience.

Indigenous Knowledge, Displacement and Place

Numerous scholars have emphasized the tight interdependence between indigenous

knowledge and place. Indigenous knowledge has commonly been referred to as "local

knowledge," reflecting the mutual relationships embedded in the natural-cultural domain of a

given place (Balee, 1994; 1998; Berkes, 1999; Casimir and Rao, 1992; Gooch 1998; Hunn,

1999; Nazarea, 1999; Ellen and Harris, 2000). Hunn (1999) states that it is the "rootedness" or

local feature of IK that makes it fragile and therefore susceptible to be eroded or lost.

If on one side there is an obvious interdependence between knowledge and place, on the

other there is the paradox of both humans and the environment being mobile, adaptive, dynamic

and interchangeable by essence. History, as well as possibilities and constraints imposed by

Brazilian political circumstances, challenges the understanding of how indigenous societies have

adapted and changed in the Amazon (Roosevelt, 1994; Ramos, 1998; Little, 1999, 2001;

Heckenberger, 2001; 2005).









Feld and Basso (1996) describe the evolution of approaches to the study of place in

anthropology and related sciences. While in the 1980s there was a tendency to focus on the sense

of "rootedness" in place, subsequent work has embraced a theorization of place from "its

contestation and its linkage to local and global power relations" (Feld and Basso, 1996:4).

Renewed theoretical interpretations of people and place reflect the widespread conditions of

exile, displacement, diasporas and conflicts over borders, as well as struggles of indigenous

peoples for ancestral homelands, land rights and retention of sacred places (Feld and Basso, op.

cit.; Harvey and Thompson, 2005). Anthropologists have broadened the philosophical notions of

sense of place, and came to view places as sites of contestation, power struggles and

displacement as histories of annexation, assimilation and resistance. The development of studies

of place redirected our understanding of stabilized territories and cultural boundedness to a more

flexible one, where place is viewed as a more fluid entity, as spaces beyond cultures, as

ethnoscapes being constantly reconstructed in a dynamic and oscillatory way (Gupta and

Fergusson, 1992; Hirsh and O'Hanlon, 1995; Feld and Basso, 1996; Alexiades, 2009).

The history of development of the Brazilian Amazon is permeated by stories of

government-led forced displacement, relocation and resettlement of indigenous and other local

communities in order to give place for settlement and mining projects, dam and road

construction, and creation of protected areas (Aspelin and Santos, 1981; Little, 2001; Sohn,

2009). In the 1960s and 1970s, the urge to occupy and develop the presumed "emptiness" of the

Amazon by the military government caused the displacement and relocation of innumerable

indigenous groups (Cunha and Salzano, 1992; Menezes, 2000; Little, 2001). The process of

forced displacement and resettlement of local people resulting from development forces is a

political and global phenomenon, which has happened in many other countries besides Brazil.









According to Oliver-Smith (2009:3), "more people were involuntarily displaced in the 20th

Century than in any other moment in recorded history."

In Brazil, processes of displacement and relocation of indigenous peoples have resulted

in catastrophic experiments involving eviction of communities from their traditional lands with

which they had strong spiritual ties; exposure of the communities to epidemics and abuses;

restrictions on or impediments to their access to strategic natural resources; and changes to their

social and political organization (Aspelin and Santos, 1981). The Panara, the Ikpeng (Txicao)

and the Kisedj (Suya) are among other indigenous groups that have been displaced and

transferred to Xingu Park, comparable to the Kaiabi case (Menezes, 2000). In Chapter 4 I will go

back to these examples, in which movements to return to ancestral territories, always present,

were more or less successful.

The geographical dislocation of the Kaiabi can be considered a case of "development-

forced displacement and resettlement" or DFDR (Oliver-Smith, 2009:3). According to Oliver-

Smith (2009), there is a general failure of governments in planning, funding and training people

to support DFDR processes, resulting in the impoverishment of the displaced. These deficiencies

in DFDR, combined with struggles for survival, keeping cultural identities and securing human

rights (such as rights to land, forest, water, as well as spiritual, cultural and moral rights) have

led to resistance movements of all sorts, in an attempt to reassert both the logic and the sense of

control over their lives and futures, both as individuals and as a social group (Oliver-Smith,

1996). People affected by DFDR often develop innovative strategies to defend their rights in

negotiations with the state, developing new sources and forms of political power (Oliver-Smith,

2009).









Theories on social change and risks accompanying development-forced displacement

have been developed by scholars and practitioners such as Thayer Scudder and Michael Cernea.

According to Scudder (2009), DFDR accelerates social change: changes in behavior patterns,

institutions, belief systems and other social domains are compressed into a shorter time span.

Scudder has also found, comparing different development-forced displacement situations, that

"individuals and sociocultural systems are affected and respond in remarkably similar ways

throughout the resettlement process." (Scudder, 2009:3)

Scudder (1981) developed a four-stage stress model spanning over two generations to

characterize DFDR. According to him, most community members are affected by and respond to

DFDR in similar ways during the first two generations following the displacement event,

irrespective of cultural, geographical and political differences. According to Scudder (2009:30)

these stages are:

* Stage 1: Planning for resettlement before physical removal.

* Stage 2: Coping with the initial drop in living standards that tends to follow removal.

* Stage 3: Initiating economic development and community-formation activities necessary
for improving the living standards of first-generation resettlers.

* Stage 4: Handing over a sustainable resettlement process to the second generation of
resettlers and to nonproj ect authority institutions.

Interesting enough, these stages can be compared to the 4 stages of the adaptive cycle

from social-ecological systems theory (see Figure 1-1). Stage 1 would be comparable to the

release phase of adaptive cycle, in which the disturbance event is the displacement. Stage 2

would be passing from the release to the exploitation phase, in which people are facing a

"drop" or collapse in their social and territorial organization after the collapse, but starting to

reorganize. Stage 3 would be the exploitation phase, in which the community is organizing

and working to improve the standards of quality of life for their descendents. Stage 4 would









be the conservation stage, in which a certain level of stability and sustainability is achieved

and handed over to the second generation.

Cernea and McDowell (2000) identified eight risks that human groups might face when

exposed to DFDR events: a) landlessness; b)joblessness; c) homelessness; d) marginalization; e)

increased morbity/mortality; f) food insecurity; g) lack of access to common property; h) social

disarticulation. Applied to the Kaiabi case, people who went to Xingu and/or those who stayed

have faced these risks in greater or lesser intensity. This question will be further explored in

Chapter 4.

Indigenous responses to displacement can be compared to those of refugees that belong

to a certain ethnic group. According to Camino and Krufeld (1994:XV) "the process of

displacement may entail the creation of new and changed identities under old labels." The

authors affirm that in the case of refugees, many populations strive to formulate innovations that

are grounded in their own ethnic background and social organization. Regarding the Kaiabi,

considering the fact that the displaced group retained knowledge, whereas the ones who

remained lost it, I argue that there are factors more relevant to cultural persistence than

attachment to land. Forced displacement might not unequivocally lead to just the erosion of

indigenous knowledge. Given favourable circumstances and factors, indigenous societies can

recover from the losses following displacement, recreating and reaffirming their cultural

patrimony and identity in new territorial configurations (Oliver-Smith, 2002).

I hope that this dissertation contributes to the theories of cultural emplacement and

displacement, through the exploration of the factors and mechanisms through which indigenous

peoples react, adapt and resist to voluntary, forced or induced geographical movements (Feld

and Basso, 1996; Cernea and McDowell, 2000; Chatty and Colchester, 2002; Harvey and









Thompson, 2005; Oliver-Smith, 2009; Scudder, 2009; Alexiades, 2009). Through the

reconstruction of the history of Kaiabi diaspora and its social, economic, environmental and

cultural consequences, I intend to contribute to the processes of land claims and struggles in

which the three Kaiabi groups are involved (Chapter 4).

The Political Ecology of Empowerment

Knowledge and power are intrinsically correlated, fluid and dynamically changing.

Among Amazonian indigenous peoples, knowledge systems, territorial configurations and

political organization structures have shifted in unprecedented ways after hundreds of years of

contact and interaction with western societies and institutions (Hemming, 1978; Roosevelt, 1994;

Jackson, 1994; Posey, 1994; Little, 2001; Heckenberger, 2005; Alexiades, 2009). In this

research, to better explore the mutual relationships between political empowerment,

environmental conservation and the dynamics of indigenous knowledge systems, I use some

elements from the theoretical field of political ecology applied to the history of Amazonian

indigenous peoples in general and to the Kaiabi in particular.

Concerning the role of historical change and external influences on social-ecological

systems, the field of political ecology shares certain characteristics and concerns with political

economy. According to Bates and Lees (1996:9), both examine "the role of power relations in

determining human uses of the environment". Political ecology explores natural resource access

and utilization under the logic of capitalism and the diverse economic situations development has

brought everywhere (Wolf, 1972; 1982 Biersack, 1999). Slocum and Thomas-Slayter (1995:4)

define empowerment as a "process through which individuals, as well as local groups and

communities, identify and shape their lives and the kind of society in which they live." It relates

to the ability of a people to mobilize and influence change, on the basis of their access to

knowledge, political processes and financial, social and natural resources. According to Narayan









(2005:5), empowerment relates to "the expansion of assets and capabilities of poor people to

participate in, negotiate with, influence, control and hold accountable institutions that affect their

lives." Political empowerment, development of local institutions, land ownership and by

extension environmental conservation have been historically intertwined in Amazonian

indigenous landscapes.

Some later research on political ecology has had practical applications, such as in the

fields of conservation, resistance and social movements and human rights (Peet and Watts,

1996). Scholarship in political ecology and indigenous knowledge systems has also tackled

issues related to the (mis-) representations of indigenous peoples as innate conservationists or

"noble-savages," or on the contrary, as harmful to nature under certain circumstances (Redford

and Stearman, 1993; Peres, 1994; Conklin and Graham, 1995; Peres and Terborgh, 1995).

Results of work done in this field have had practical implications for the development of public

policies towards conservation of tropical forests in the Amazon and elsewhere (Colchester, 2000;

Schwartzman et al., 2000).

In the Amazon, starting in the 1970s, the development of indigenous grassroots

movements and the partnership forged between indigenous peoples and the environmental

agenda broadly set up by international scientific, politic and activist instances, have had great

impact on public policies, law making, management of indigenous lands and, of course, on the

empowerment of indigenous organizations (Ramos, 1998; Albert, 2004; 2005). Albert (2004)

introduces the concept of "the projects market," to highlight the importance of international

funding, done mainly through projects, as the main strategy to support indigenous people's needs

in terms of economic alternatives, territorial management and cultural revitalization.









The alliances with environmentalism and the capacity of Amazonian indigenous peoples

to create a public image consonant with the objectives of sustainable development and

conservation has been a theme of discussion in the academy (Conklin, 1997; Fisher, 1994;

Jackson, 1994; 1995). Authors have analyzed the ways in which indigenous Amazonian peoples

have resisted, appropriated, assumed and/or internalized the image created for them by western

institutions such as the state, NGOs, academy, churches or others (Turner, 1991; Fisher, 1994;

Jackson, 1994, 1995; Conklin and Graham, 1995; Gray, 1997; Sahlins, 1997; Oakdale, 2004;

Albert, 2005). In the Kaiabi case, specifically in Xingu, political leaders have been successful in

the articulation of their agenda in consonance with nature conservation objectives of their

donors, partners or funding agencies. Kaiabi history is permeated by paradoxes of becoming

something defined outside with and for others, and simultaneously keeping ancestral forms of

organization and identity in and for themselves (Oakdale, 1996; 2004).

For this research, I focus on two main interrelated topics linked to the political ecology

approach applied to the Kaiabi case, presented in Chapter 6, which involve: 1) How and with

what consequences have the indigenous grassroots movements in Brazil interacted with the

international environmentalist movement (Ramos, 1998; Albert, 2005); and 2) How have

indigenous peoples responded to and appropriated institutions, resources, norms and values in

the construction of their identity and management of their lands (Albert, 2005; Hierro, 2005;

Hierro and Surrales, 2005; Oakdale, 2005). I present a comparative analysis of the process of

Kaiabi political empowerment in Xingu, through the indigenous Association ATIX (Associacgo

terra Indigena Xingu) contrasted with the difficulties faced by the other political organizations in

Rio dos Peixes (Itaok Association) and Teles Pires (Kawaip Association). I also present and

discuss results of interviews done with men and women from the three Kaiabi groups, related to









the situation, role and potentialities of the indigenous organizations ATIX (Xingu), Itaok (Rio

dos Peixes) and Kawaip (Teles Pires).

A Process Perspective on Indigenous Knowledge and Cultural Change

Taking into account the magnitude and possible unfolding of the field, the development

of a general theory of indigenous knowledge systems and how they change would be too

ambitious and even impossible (Nugent, 2006). Zent (2009 a) identifies seven related phases that

have arisen in the past fifty years and that are still present in contemporary anthropological

research on indigenous knowledge systems: 1) environmental ethnoscience; 2) theoretical

cognitive ethnobiology; 3) modeling the relationship between knowledge and behavior; 4) the

significance of IK for sustainable development and conservation of nature; 5) debates about the

valuation, exploitation, and compensation of IK; 6) IK as a critical ecopolitical discourse, and 7)

processual perspectives of IK.

What Zent (2009 a: 12) called processuall perspectives on indigenous knowledge," would

include the following questions: "how it is created, what does the learning process entail, who

passes it on to whom, in what situations and contexts does transmission occur, why is it lost or

changed, what is the social organization of knowledge, how do social relationships regulate the

flow of information, how do use patterns and contexts affect knowledge, what social and

ecological factors promote its conservation or extinction." He suggests that due to the relative

immaturity of this perspective, most of the conclusions reached so far are limited to the particular

local settings in which the studies were carried out, and thus so far there has been virtually no

development of a more general or theoretical understanding of such processes. In another paper

directed to the development of a "Traditional Environmental Knowledge Vitality Index", Zent

and Maffi (2007) mention the scarcity of longitudinal studies on indigenous knowledge systems

transmission and change.









This dissertation is related to phases 4,6 and 7 defined by Zent (2009 a): 4) the

significance of IK for sustainable development and conservation, reflecting a broader

perspective "woven" throughout this research; 6) IK and ecopolitics, approached in Chapter 6

and 7) processual perspectives of IK, developed in Chapters 7 and 8. I understand processual

study as research that comprises a continuous time frame, through which one can understand

how change is occurring as a process, in a continuous set of events, rather than a snapshot taken

at a given moment in time. As an example, Ricardo Godoy, Victoria Reyes-Garcia and their

associates have carried out processual studies on indigenous knowledge and its many facets

among the Tsimane indigenous people in Bolivia for many years (Reyes-Garcia, 2001; 2005;

Godoy, 2001; Godoy et al., 2005). Building on previous research done with the Kaiabi

throughout ten years, I present a processual study of weaving knowledge among Kaiabi men

and women, in which I address the questions and gaps of knowledge stated by Zent (2009 a). It

is worth mentioning that I take one sub-domain of weaving knowledge to carry out an in-depth

analysis, related to the ability to weave designed basketry and textiles. This domain consists of

specialized knowledge, and as such, might present important differences in its patterns of

knowledge creation, transmission, erosion and innovation.

Research on indigenous knowledge systems has shown the highly variable distribution of

knowledge in a given ethnic group, according to diverse factors such as environment, gender,

age, kin groups, schooling and social position (Boster, 1986; Boster et al., 1987; Philips and

Gentry, 1993; Nabhan and St. Antoine, 1993; Ellen and Harris, 2000; Hunn, 2002; Wilbert,

2002; Zarger, 2002; Howard, 2003). In spite of variations, several studies have shown that intra-

cultural variation is patterned in a given indigenous group (Ellen, 1979; Boster, 1985; Weller,

1983; 1987; Reyes-Garcia, 2001; Reyes-Garcia et al., 2005). For instance, mechanisms of









knowledge acquisition and transmission differ according to gender, since they are related to the

different roles men and women play in the society (Boster, 1985; 1986, Howard, 2003).

The inclusion of women and knowledge related to textiles in this research opens up the

possibility to compare mechanisms of knowledge acquisition, transmission, distribution and

erosion between men and women. How do these mechanisms differ between genders? Are there

common trends which can be observed for men and women alike? Furthermore, the great

majority of previous research done amongst the Kaiabi has focused on men's knowledge, having

mostly relied on men as the main informants (Griinberg, 1970; Travassos, 1984; Rodrigues,

1993; Oakdale, 1996; Schmidt, 2000; Tomass, 2006). Howard (2003) discusses the

underappreciated role of indigenous women's knowledge in ethnographic research and

conservation projects around the globe. In this research, Kaiabi women have had the opportunity

to express themselves not only in relation to weaving knowledge, but also to give their opinions

related to political organisation, consequences of the transfer to Xingu, and prospective

development projects in which they would like to participate.

Weaving knowledge, especially related to graphic designs, has social and symbolic

meaning which prevents its mastery by everyone or by a large number of people (Athayde,

2003; Athayde et al., 2009). To understand how knowledge is learned, distributed and

transmitted among the Kaiabi, I partially adopt the concept of culture as shared knowledge

(Boster, 1986), applied to the domain of weaving knowledge. I want to contribute to the

understanding of intracultural similarity and variation (Boster, 1987; Weller, 1987) using

statistical models and consensus analysis (Romney et al., 1986). I also review and expand the

model proposed by Ellen (2009, see Chapter 8) to the study of overlapping categories in

basketry-making as a cultural domain. Besides general questions on how knowledge is created









and transmitted, how it varies between genders and other social categories and how it changes,

this research presents a review of models for knowledge or cultural transmission developed by

cognitive anthropologists, specifically of the model proposed by Cavalli-Sforza and Feldman

(1981).

The interactions between socio-economic aspects and cultural shifts are also explored and

compared among the three groups, and presented in Chapters 7 and 8 (Maffi, 2001). I sought to

deepen the understanding and open a debate on how social organization and thus mechanisms of

knowledge transmission are shifting among indigenous peoples, through their growing

participation in and assimilation of western institutions, such as schools, political organizations,

markets and development projects (Turner, 1991; Posey, 1994; Jackson, 1994; 1995; Gray, 1997;

Oakdale, 1996; 2004; Godoy, 2001; Godoy et al., 2005; Reyes-Garcia et al., 2005). Specifically,

this research advances some questions and models explored by other authors, relating changes in

indigenous knowledge domains (including language proficiency) with selected socio-economic

aspects such as market integration, schooling, and age (Zent, 1999; 2001; Godoy, 2001; Hill,

2001; Hunn, 2001; Reyes-Garcia, 2001; Reyes-Garcia et al., 2005; Athayde et al., 2009). It is

expected that older people will retain more knowledge and language proficiency in comparison

to younger people (Phillips and Gentry, 1993; Zent, 1999; Athayde, 2003; Athayde et al., 2009).

Additionally, it is expected that greater levels of market integration and formal schooling lead to

knowledge erosion (Nabhan and St. Antoine, 1993; Zent, 1999; Godoy, 2001; Godoy et al.,

2005; Reyes-Garcia, 2001; 2005). Beyond confirming or denying these postulates, this research

contributes to the understanding of how these processes happen, interact and affect the social-

political configuration and indigenous people's knowledge systems (Reyes-Garcia, 2001).









Since 2004, the Kaiabi from Xingu are carrying out a community-based project named

Kaiabi Araa, for the revitalisation of basketry and textiles weaving involving an exchange

between men and women from Xingu Park's villages and from Kururuzinho village in the

ancestral land at Teles Pires River (ATIX, 2004). Thus, one contribution of this research will be

an evaluation of the impact of this project in the distribution of knowledge between men and

women in Xingu Park and in Kururuzinho village (Teles Pires River), by comparing the data

obtained in 2007 to the 2002 baseline data for Xingu Park (mainly Tuiarare village) and the 2004

data for Teles Pires area (people were interviewed there before the project started).

While the Xingu Kaiabi have maintained and recreated basketry and textiles weaving

practices in spite of the lack of raw materials such as aruma, the two other groups of Kaiabi

have experienced a loss of knowledge in terms of this particular tradition. Prior to 2004, only

four men and one woman still knew how to weave in the Kururuzinho village (Teles Pires). In

Rio dos Peixes, according to testimonies given by Xingu Kaiabi informants, artistic knowledge

on basketry and textiles has been apparently lost. Through this research, we will be able to

evaluate if and to what extent knowledge is really lost there, and to determine the role that

kinship links and constant visits of Xingu relatives might play in knowledge revitalization. The

differences and dynamics related to weaving among the three Kaiabi groups also recur in other

aspects of their lore, including language, crop varieties, festivals and other handicraft

production. Research on the social dynamics of knowledge related to basketry and textiles thus

might provide insights into other realms of social change accompanying the distinct historical

trajectories produced and followed by different Kaiabi groups.

Brief Chronological Sketch of Research Done Among the Kaiabi

A chronology of the more relevant bibliographical sources about the Kaiabi begins with

the explorer Ant6nio Pyrineus de Souza in 1916, who wrote one of the first publications about









the group (Souza, 1916). He also collected valuable items of Kaiabi Material Culture, which are

deposited at the Museu Nacional (National Museum) in Rio de Janeiro. Among other first

accounting on the Kaiabi, we should mention the expeditions of the German explorer Karl Von

den Steinen, who published a detailed story of his expeditions in Mato Grosso and Para, and the

indigenous peoples with whom he established contact or got information about through other

groups (Steinen, 1940).

Max Schmidt (1942), a German anthropologist, was the first person to conduct more

detailed ethnographic research amongst the Kaiabi from 1927 to 1929. Schmidt brings valuable

information on the relationship between the Kaiabi and SPI Posts (Simdo Lopes and Pedro

Dantas) at that time. He also collected a variety of objects, which were deposited in the Museum

of Ethnology in Berlin. After him, Father Joao Dornstauder worked for the Anchieta mission for

twenty-five years among the Kaiabi in the Rio dos Peixes region, and wrote valuable information

on the group's traditional territory and the situation of this group before and after the transfer to

Xingu Park (Dornstauder, 1955; 1981). Georg Grunberg, an Austrian ethnologist, worked with

the Kaiabi of Rio dos Peixes before the transfer to Xingu Park, writing a detailed historical

description and ethnography of the group (Grunberg, 1970). A chapter portraying the actual

situation of the Kaiabi today accompanied the publication of his dissertation as a book in

Portuguese in 2004 (Grunberg, 2004; Senra, et al., 2004).

Claudio and Orlando Villas B8as (1989) presented the situation of the Kaiabi at Sao

Manoel or Teles Pires river prior to the transfer to Xingu Park. Berta Ribeiro (1979; 1986)

carried out ethnographic work on Kaiabi basketry in the Xingu Park. In the 70's and 80's,

Elizabeth Travassos (1984; 1993), wrote about Kaiabi music (Jowosi festival) and shamanism,

and Mariana Ferreira did research on indigenous peoples and schooling in Brazil, with a focus on









Xingu indigenous peoples and among them, the Kaiabi (Ferreira, 1992). Rodrigues (1993) did

research on Kaiabi agroforestry management among the Xingu Kaiabi in the Capivara village. In

1996, Suzanne Oakdale presented her doctoral dissertation on Kaiabi agency and identity

construction, working with the Kaiabi at Xingu Park, with a focus on Tuiatare village (Oakdale,

1996; 2005). Father Bartolome Melia (1993) presented a description of the situation of the Rio

dos Peixes group before and after the transfer of part of the Kaiabi population to Xingu Park.

Various anthropologists have carried out research among the three Kaiabi groups,

working for FUNAI on the identification of Kaiabi ancestral lands as part of land claims and

disputes processes. Among these, Patricia Rodrigues (1994) was responsible for the

anthropological report on the enlargement of the Kaiabi territory in the lower Teles Pires region

(where Kururuzinho Post is located). Klinton Senra worked on the preliminary identification of

a land claim of the Xingu Kaiabi related to the northwestern region of the Xingu Park (at the

Arraias river, see Chapter 4 on land claims) and also coordinated the identification of the

Bateldo land in the Rio dos Peixes region (Senra, 2001; 2002).

Regarding linguistic studies, Rose Dobson and Helga Weiss worked with the Kaiabi at

Rio dos Peixes and at Xingu Park in the 1960s, 1970s and 1980s (Dobson, 1998; Weiss, 2005,

see next section) and later on Patricia Souza researched aspects of Kaiabi language among the

Xingu Kaiabi (Souza, 2004). Among the Xingu Kaiabi, Pagliaro (2002; 2005) wrote about

Kaiabi demographic recovery after the transfer to Xingu Park, and Jackeline Mendes did

research on ethnomathematics (Mendes, 2001).

Beginning in the 1990's, as part of the Xingu Program from Instituto Socioambiental

(ISA), Geraldo Silva worked extensively on Kaiabi agrodiversity knowledge and management,

and natural resource management at Xingu Park (Silva, 1999; Silva and Athayde, 1999; Silva et









al., 2000; Silva, 2002a; Silva, 2002b; Silva, 2004). He wrote his dissertation on knowledge

related to peanut diversity and conservation among the Xingu Kaiabi (Silva, 2009). Marcus

Schmidt (2001) carried out research on Kaiabi silviculture and natural resources management for

his Master's dissertation. I have carried out research amongst the three Kaiabi groups (with a

focus on the Xingu Kaiabi) on material culture, basketry, textiles and natural resource

management (Athayde, 1998; 2001; 2003; 2004; Athayde et al., 2006; 2009).

Aturi Kaiabi did a research paper for his undergraduate course conclusion at

Universidade do Estado de Mato Grosso (UNEMAT) about basketry and textiles (Kaiabi, 2006).

More recently, Lea Tomass (2006) has worked on research of Kaiabi festivals, with a focus on

the Jowosi festival among the Xingu Kaiabi, which is also the subject of her doctoral research.

Amongst the Kaiabi from Teles Pires, Francisco Stuchi has done research on ethnoarcheology

for his master's thesis (Stuchi, 2010), and Frederico Oliveira (2008) has carried out his doctoral

research among the Teles Pires group at the Kururuzinho post and village, on territoriality and

land claim struggles.

Personal Names and Orthography

The Kaiabi speak a language from the Tupi stock, in the Tupi-Guarani linguistic family

(Rodrigues, 1986). In the existent literature concerning the Kaiabi, there have been many

different ways of spelling the name of the group: Cajahis, Cajabis, Kajabi, Caiabis, Cayabi,

Kayabi (Souza, 2004). For this dissertation, I will adopt the spelling as "Kaiabi," which is

adopted and in use by the teachers at Xingu Park (Souza, 2004).

A grammar and orthography for the Kaiabi language began to be developed in the 1960's, with

the work of linguists Rose Dobson and Helga Weiss from the Summer Institute of Linguistics or

Sociedade Internacional de Lingiiistica (SIL), who worked at the Tatuy indigenous post at Rio dos

Peixes and later on at Xingu Park after the transfer of the group, already in the 1970's (Dobson,









1973; 1988; 1997; Weiss, 1998; 2005). Later on, Kaiabi teachers continued to develop the grammar

and orthography through courses and workshops of the program for training of indigenous teachers

carried out in Xingu Park initially by FMV/AVA, which was taken over by ISA in 1995, with some

activities of accompanying indigenous teachers at the village still underway (Troncarelli et al.,

2003). Kaiabi teachers from Xingu Park, with the support from ISA's team and from the linguist

Patricia de Oliveira Borges e Souza, organized the first book for literacy training in the Kaiabi

language in 1999 (Troncarelli, 1999). Deriving from her work with the Kaiabi at Xingu Park,

Patricia wrote her Master's thesis on aspects of the Kaiabi language (Souza, 2004). Some teachers

trained through ISA's Program have entered the Indigenous University Program (Terceiro Grau

Indigena Program, UNEMAT2), and since then have continued working in the development of

Kaiabi written language.

In this dissertation, when using Kaiabi words or transcribing narratives, I tried to observe the

grammatical forms and orthography with the greatest accuracy possible. However, as Kaiabi and

other indigenous languages' orthographies are still in construction and debate, there might be

mistakes and incongruences in my Kaiabi writing, for which I apologize in advance to the Kaiabi

and to the general readers.

The names for basketry designs were written, translated and reviewed with the participation of

my teacher and friend Jowosipep (Aturi) Kaiabi, chief and teacher of Tuiarare village. The names for

basketry and textiles items derive from workshops developed in Tuiarare village, as part of the

Kaiabi Araa project (ATIX, 2004; 2005). Portuguese names and technique designations are based on

Berta Ribeiro's work on Brazilian indigenous peoples' basketry and textiles (Ribeiro, 1984/85;

Ribeiro, 1985; Ribeiro, 1986). The names for plants and natural resources mentioned in the

2 Universidade do Estado de Mato Grosso, which implemented the first Indigenous University Program in Brazil in
2001 (UNEMAT, 2008).









dissertation were based on previous reports and documents reviewed by indigenous teachers

(Athayde, 1998; Silva and Athayde, 1999; Silva and Athayde, 2004; Athayde et al., 2006).

Concerning the names for people who participated in the interviews and fieldwork, I opted

for asking each interviewee or participant how they would prefer to be named: by their real Kaiabi or

Portuguese names, or by a fake name, created in order to protect their privacy, as is usual in

ethnographic writings. Most people whom I interviewed in the three Kaiabi lands preferred to be

referred to by their real Kaiabi names in my dissertation, but said that I could also use their

Portuguese names to make easier to identify them. In Rio dos Peixes, where people usually have a

Portuguese and a Kaiabi name, they always preferred to be referred to by their Kaiabi name, even if

this name is seldom used by them. I also asked permission to publicize the contents of the interviews

and testimonies, which was part of the prior informed consent and in an ethical agreement signed by

me and by the Kaiabi.

Structure of Dissertation

This dissertation is organized in a book format, structured in two parts and nine chapters,

which follow progressive historical and scale levels, from Kaiabi ancestral territory to current lands,

villages, households and individuals. I adopt a systems approach to explore and understand the

dynamics of knowledge systems across three Kaiabi groups, analyzing historical, environmental,

socio-economic, political and cultural factors that influence the persistence, loss and change of

indigenous knowledge through time.

The first part includes Chapters 1 to 5 and contains a general introduction to the dissertation,

where I present theories, methods, and the historical, territorial and ecological context of the Kaiabi

people before and after their displacement and transfer to Xingu Park. Chapters 1 and 2 include

theoretical approaches, fields of inquiry and possible applications of the research, as well as the

methods employed in data collection and analysis, arguments to be discussed and hypotheses to be









tested. The subsequent chapters are organized in a similar structure, containing an introduction, in

which I present the main aspects, arguments and questions discussed in the chapter, and a small

conclusion at the end, summarizing results, findings and discussion.

In Chapter 3, I present a description of the Kaiabi lifestyle in the Tapaj6s region before their

contact with the Villas B8as brothers, the creation of Xingu Park by the Brazilian government and

the process of transfer of the majority of the group to Xingu Park.

Chapter 4 contains a description of the process of adaptation to Xingu, as well as the struggle

of the two groups who remained around the ancestral territories in Rio dos Peixes and Teles Pires

rivers regions. In this chapter, I also include an overview of the process of demarcation of

indigenous lands in Brazil, and how this relates to current Kaiabi land claims.

In Chapter 5, I include a description of environmental context and contrasts between the three

Kaiabi lands considered in this study: Xingu Indigenous Park, Apiaka-Kaiabi indigenous land in Rio

dos Peixes and Kayabi Indigenous Land in Teles Pires. In this chapter, I explore the question of what

mechanisms of ecological adaptation the Xingu Kaiabi have developed after their transfer to a new

land and distinct environment.

The second part of this dissertation includes Chapters 6 to 8 in which I present results of

cross-sectional and longitudinal studies regarding political empowerment and indigenous knowledge

systems among four Kaiabi villages and Chapter 9, which is the general conclusion.

Chapter 6 describes the current territorial organization and political configuration across the

three Kaiabi lands. The main question explored in this chapter relates to the factors which led to the

demographic expansion and concomitant political empowerment of the Xingu Kaiabi in contrast

with the other two groups. I describe the process of political empowerment of the Xingu Kaiabi, as it

relates to indigenous grassroots movements in Brazil. I present information on the political









organizations among the three Kaiabi groups, namely ATIX in Xingu, Itaok in Rio dos Peixes and

Kawaip in Teles Pires. I describe some community-based projects developed by Xingu communities,

including the Kaiabi Araa project, related to the revitalization of weaving knowledge among the

Xingu and Teles Pires Kaiabi. Finally, I present results of interviews about people's perceptions and

expectations on the role and work of indigenous associations in each village.

Chapter 7 includes a comparative socio-economic analysis of the four villages included in

this study: Capivara and Tuiarare in Xingu, Tatuy in Rio dos Peixes and Kururuzinho in Teles Pires.

In this chapter, I present a comparative analysis of socio-economic variables that might be influential

in knowledge systems and thus in the cultural resilience of Kaiabi people. I want to contribute to the

understanding of socioeconomic factors that underlie processes of cultural change among the Kaiabi,

and of the mechanisms through which they function. I test hypotheses linked to the effects of market

integration and formal schooling on indigenous knowledge systems. My main dependent variable is

language proficiency, which I sought to compare to the behaviour of weaving knowledge in the

subsequent chapter.

In Chapter 8, I go one more level down in scale and look at specificities linked to the domain

of weaving knowledge related to basketry and textiles among men and women in four villages. I

focus on two main aspects of weaving knowledge, exploring mechanisms of knowledge creation,

transmission, distribution, and change, and the influence of western institutions on the dynamics of

weaving knowledge across men and women in four villages. I present results of a longitudinal study

comparing data collected in 2002 and 2007 among men in two Xingu villages. I test hypotheses and

discuss the results attempting to integrate aspects discussed in previous chapters and how they relate

to the dynamics of weaving knowledge.









Chapter 9 is a general conclusion of the dissertation, in which I review theories, methods,

arguments and questions posed in the introduction in the light of the results and information

presented throughout the document. I include discussions relevant to each main field of inquiry

approached, as well as implications of my research for theory, research, practice and policy making

related to indigenous peoples cultural and environmental resilience in the light of rapid change.












Adaptive Cycle in 2D


Connectedness

Source: Panarchy, 2002, p. 34.

Figure 1-1. The adaptive cycle, used to understand change and reorganization of socio-ecological
systems. Source: Gunderson and Holling (2002).























Weaving Knowledge
Uses Function Symbolic
Form Materials Designs /


Figure 1-2. A possible representation of a systemic approach to the study of a cultural domain, in
this case weaving knowledge.











Displacement
Mobility


Cultural Resilience
(IK) A


Ecological
Conditions


Western
Institutions


Political
Context


O 1 Socio-economic
organization




/
Territorial control &
rights


Figure 1-3. Another possible representation of a systems approach to the study of factors
involved in the resilience and change of Kaiabi cultural system.









CHAPTER 2
FIELDWORK AND METHODS

Previous Fieldwork

For this dissertation, I used published and unpublished data that I have been collecting

during the last ten years at different Kaiabi villages and lands as part of my work as a practitioner

for the Instituto Socioambiental ISA. I also carried out specific fieldwork during 2006 (May-

November) and in 2007 (March-July).

I have been recording myths and stories related to basketry and textiles weaving and to

the natural resources used to make them, as well as data on the collection, preparation and

transformation of the natural resources into different kinds of baskets. I have also been

collecting information on the process of transfer of the Kaiabi to Xingu, and on natural resources

of cultural and economic importance to the Kaiabi, occurring in all three areas currently

occupied by the group. These data have been recorded in different ways: a) participant

observation in the villages; b) semi-structured interviews conducted with men and women at

selected villages; c) collection of narratives and testimonies with elders and others; d) through

training courses and workshops; and e) through field walks in different ecosystems to survey

natural resources.

I have been photographing and collecting photographs of different kinds of baskets and

textiles in the villages, in three Brazilian ethnographic museums located in Sao Paulo and Rio de

Janeiro', in the Museum of Ethnology in Berlin, Germany and in the Museum of Cultures in

Basel, Switzerland. I also included photographs of baskets taken by Georg Grinberg, an

anthropologist who worked with the Kaiabi people in Rio dos Peixes in 1966, before their


1 surveyed the MAE/USP Museu de Arqueologia e Etnologia da Universidade de Sio Paulo and Klinton Senra,
an anthropologist who has worked with Kaiabi people, carried out surveys at the Museu Nacional and Museu do
Indio in Rio de Janeiro.









transfer to Xingu Indigenous Park (Grunberg and Grunberg, 1967; Grunberg, 970). With the

material resulting from this photographic documentation, I organized an educational book for the

Kaiabi people, containing the pictures of the baskets, the myths and stories told and written by

them, and information about the natural resource ecology (Athayde, 2006). The organization of

this book allowed me to have a fairly complete catalog of almost all graphic designs woven in

the painted baskets, which was used in the interviews.

In 1999, I participated in travel to the Rio dos Peixes village with a group of Xingu

Kaiabi who were interested in visiting relatives, visiting old villages in the ancestral land, and

collecting seeds and saplings of natural resources which occur in the ancestral region but are

scarce in the Xingu Park region (Silva et al., 2000). It was an opportunity to get to know the

group of Kaiabi who stayed in Rio dos Peixes, and also to collect socio-economic and

environmental data valuable to the present research. At that time, I was in my third month of

pregnancy. When I went back to Rio dos Peixes in 2007, I took my seven year old boy with me,

which the Kaiabi there really appreciated.

Fieldwork Carried Out from 2002 to 2004

In 2002 and 2003, as part of my MSc in Ethnobotany at University of Kent (Athayde,

2003), I carried out fieldwork among Tuiarare and Capivara residents at Xingu Park. I was

investigating the variation of knowledge related to basketry and textile weaving among the

people of the two villages. Indigenous teachers, elders and representatives of the group of

indigenous agents for management of natural resources participated both in the questionnaire

design and in the interviewing process. We then structured a questionnaire to carry out semi-

structured and structured interviews with every man and woman aged fifteen years old and over

in the two villages, and with key informants and shamans in other Kaiabi villages and in the

Diauarum Indigenous Post.









For the purpose of that thesis I decided to work only with the men, leaving the interviews

with the women for a further stage of research. At that time, I selected Tuiarare and Capivara

villages for comparison, because these were the biggest Kaiabi villages in the Xingu Park and the

people who live there used to inhabit two different places located in their ancestral land before

the transfer to Xingu: people from Tuiarare came from the Teles Pires region, and people from

Capivara came from Rio dos Peixes.

In each village, I selected the household as the basic unit of analysis. In every household,

I interviewed the chief or the head of the household and all other men who lived in the same

house: sons, sons-in-law, nephews, brothers-in-law, etc.

I collected basic social data about each person, specific data on basketry weaving, and

data on natural resources use and management. I recorded all the interviews using a tape

recorder. In the second stage, I showed sheets containing photographs of different graphic

designs in a random order and I asked people to say which design a particular person was known

to weave. I assigned a code to each graphic design presented to the interviewed people. For other

designs which people were known not to weave, I asked if they knew the names and then

recorded the answers according to the codes. For some specific persons, such as shamans, elders

or very good weavers, I asked additional questions relating to cosmological and spiritual aspects

of basket weaving.

During May and June of 2003, I went back to some Kaiabi villages to fill in gaps related

to kinship and knowledge of basketry graphic designs. I worked in Tuiarare, Capivara and

Sobradinho and I also worked with Aturi Kaiabi on the correction of orthography in the native

language and on the translation of each graphic pattern name.









To input my quantitative and qualitative data on Kaiabi basketry weaving, I structured a

database using Microsoft Access. I assigned a symbol to each village, household and person,

which was defined as the primary key in the database. I carried out some basic statistics to

compare Tuiarare and Capivara villages: number of people interviewed, place of birth, age,

kinship. I also used Microsoft Excel to run statistical analysis, as well as SPSS for Windows, to

run the Anova linear regression analysis to relate uses and names of basketry designs to age of

the interviewees. I used Anthropac software to run specific statistical analysis such as consensus

analysis, MDS scaling and cluster, related to the knowledge of graphic designs, the names given

to them and the sequence in which they are learned2

In 2004, I participated in a trip to Kururuzinho village in the Teles Pires River along with

ten men and women from Xingu, as part of the Kaiabi Araa project (ATIX, 2004; 2005). At that

time, I collected demographic and socio-economic data about the Kururuzinho community and

also specific information on knowledge related to basketry and textiles, applying the same

questionnaire used for my MS research in Capivara and Tuiarare villages in 2002, in all

households, with men and women over 15 years of age.

Fieldwork Carried Out from 2006 to 2007

This research involved the participation of indigenous environmental managers3 in Xingu

Park and of teachers, women, men, shamans and youth in three Kaiabi lands: Xingu Park, Rio

dos Peixes and Kururuzinho village at Teles Pires river. Fieldwork stages occurred in the

summer of 2006 and during 2007. Background data were used to carry out a longitudinal study



2 1 will explain better what these techniques are and why I used them when I describe the statistical analyses I've
done for this dissertation.
3 There is a project in development by ISA and ATIX since 2000, for training of indigenous environmental
managers to work with management and conservation of natural resources at Xingu Park. This includes 25
representatives of four indigenous peoples working in nearly 18 villages (Silva et al., 2002).









(Bernard, 2006) on indigenous artistic knowledge change among the Xingu Kaiabi comprising a

five-year time span (2002-2007), in which past and current mechanisms of knowledge creation,

transmission and change related to basketry and textiles weaving were investigated amongst

Kaiabi men and women respectively. The interviews done at different villages in Xingu

Indigenous Park in 2002 and in Kururuzinho village (Teles Pires) in 2004 were adapted and

repeated in 2007, and the same data collection instruments were applied in Rio dos Peixes. The

study included four villages: two in Xingu Park (Tuiarare and Capivara), the Rio dos Peixes or

Tatuy village and post at Rio dos Peixes in Mato Grosso, and the Kururuzinho village and post in

Teles Pires River, Para.

In the summer of 2006, my doctoral proposal was presented to members of Kaiabi

communities, ATIX and ISA personnel in Sdo Paulo and in the Xingu Indigenous Park.

Preliminary data on the kinship structure and socio-economic characteristics of the Kaiabi were

also collected during this period.

Objectives, Research Questions and Hypotheses

In this section, I present the main objectives, research questions and hypotheses of this

research. Some questions and hypotheses are specifically treated in one or more chapters in this

dissertation, whereas others are broad and therefore are not linked to a specific chapter, but

approached throughout this document.

Objectives

1. Contribute to the understanding of the factors and processes related to knowledge creation,
conservation, distribution and change amongst indigenous societies.

2. Compare the effects of geographical displacement on the dynamics of indigenous knowledge,
political empowerment and territorial control across three Kaiabi groups.

3. Evaluate the impact of community-based projects on knowledge transmission and
distribution across two Kaiabi groups (Xingu Park and Teles Pires).









4. Contribute to processes of cultural revitalisation and land claims in which the Kaiabi are
currently engaged.

Research Questions

General

1. Which factors might enable or constrain cultural and environmental resilience among
Amazonian indigenous peoples?

2. What factors are involved in the persistence, loss and innovation of indigenous knowledge
among indigenous societies?

3. What are the mechanisms of social and ecological adaptation which might be developed by
indigenous peoples after a displacement situation?

Specific

1. How, even under unfavourable ecological conditions, did the displaced Kaiabi manage to
adapt and respond to the conditions faced in Xingu Park, being today influential in the Park's
politics? In contrast, why do the political organizations struggle to thrive in the other two
areas? (Chapters 3, 4, 5 and 6)

2. How have the Kaiabi appropriated western institutions (such as indigenous associations,
projects and schools) in the processes of construction of their identity and political
empowerment? (Chapters 6, 7 and 8)

3. To what extent might the vitality of a given domain of IK (language proficiency, weaving
knowledge), be used as an indicator of environmental conservation status? (Chapters 7 and 8)

4. Why have the displaced Kaiabi retained and innovated their knowledge on basketry and
textiles, whereas the other groups who remained in the ancestral land have lost it? To what
extent have the other two groups lost their traditional lore on weaving? (Chapter 8)

5. How has market integration interfered in the dynamics of weaving knowledge among the
three Kaiabi groups? (Chapters 7 and 8)

6. What might be the impact of community-based projects on cultural revitalization among
indigenous peoples? (Chapters 6 and 8)

7. Which adaptive mechanisms have the Xingu Kaiabi used to deal with environmental
constraints caused by the transfer? (Chapter 5)

8. Which socioeconomic factors underlie processes of cultural shift among the Kaiabi and
through which mechanisms do they operate? (Chapters 7 and 8)









Arguments and Hypotheses

This research includes two main arguments and five working hypotheses described

below. The hypotheses differ in scale and are linked to two main theoretical fields explored in

this study, namely: 1) Processual studies in the acquisition, distribution and transmission of IK

(Ross, 2004; Zent, 2009 b), which can be congregated in the general field of Cognitive

Anthropology; and 2) Indigenous knowledge and western institutions, which can be linked to

political ecology and development studies (e.g. Sillitoe, 2007, among many others mentioned in

introductory chapter). My two first hypotheses are concerned with variation of knowledge within

and between each of the four communities studied. The other three explore patterns and

variations between the three groups. For each argument and hypothesis, I included a theoretical

background statement taken from the introductory part of this chapter (in the "fields of inquiry"

sub-sections).

Argument 1: Political empowerment enables cultural resilience. Theory or scientific

corollary: Since the 1970s, political empowerment of indigenous organizations through alliances

with environmental conservation institutions has been the main strategy in supporting social-

cultural and territorial sustainability of indigenous peoples in the Amazon (Ramos, 1998; Albert,

2004; 2005).

Argument 1: I argue that greater political empowerment has led to greater cultural

resilience (capacity to absorb change) among the three Kaiabi groups. In other words,

empowerment enhances the capacity of promoting cultural continuity.

Argument 2: Indigenous knowledge and environmental conservation are

intertwined. Theory or scientific corollary: There is an historical and co-evolutionary

interdependence between cultural and biological diversity on planet Earth (Inglis, 1993; Balee,









1994; Carroll and Mefe, 1994; Denevan, 2001; Maffi, 2001; Gunderson & Holling, 2002;

Harmon, 2002; Heckenberger et al., 2003; Zent and Maffi, 2007). Where more vigorous systems

of indigenous knowledge are maintained, the probability of finding stronger indicators of

ecosystem health is greater (Inglis, 1993; Zent and Maffi, 2007). When land rights are assured,

the potential of social-cultural reproduction might be greater (Surrales and Hierro, 2005).

Argument 2: I argue that where knowledge (and by extension, culture) is maintained,

territorial control and thus biodiversity conservation will be greater than where knowledge is

eroded. Therefore, it is expected that more robust systems of weaving knowledge correspond to

greater territorial control and consequently to greater environmental conservation (see definitions

below) across the three Kaiabi groups.

HI and H2: Indigenous knowledge distribution is patterned. Theory or scientific

corollary: Research on indigenous knowledge systems has shown the highly variable distribution

of knowledge in a given ethnic group, according to diverse factors such as environment, gender,

age, kin groups, schooling and social position (Boster and Weller, 1987; Philips and Gentry,

1993; Nabhan and St. Antoine, 1993; Ellen and Harris, 2000; Hunn, 2002; Wilbert, 2002; Zarger,

2002; Howard, 2003). Several studies have shown that intra-cultural variation is patterned in a

given indigenous group (Ellen, 1979; Boster, 1986; Weller, 1983; 1987; Reyes-Garcia, 2001;

Reyes-Garcia et al., 2005). For instance, mechanisms of knowledge acquisition and transmission

differ according to gender, since they are related to the different roles men and women play in

the society (Boster, 1986, Howard, 2003; Silva, 2009). Also, it is expected that older people will

retain more knowledge and language proficiency in comparison to younger people (Phillips and

Gentry, 1993; Zent, 1999; Athayde, 2003; Rijal, 2008; Eyssartier et al., 2008).









Hypotheses: Among the Kaiabi, I go beyond accepted corollary that indigenous

knowledge is unevenly distributed within a given group. I hypothesize that:

HI: Elders retain deeper and different knowledge when compared to younger people, and

H2: Men and women use different mechanisms in knowledge creation and transmission.

H3, H4 and H5: Mechanisms for indigenous knowledge transmission are differently

affected by western institutions. Theory or scientific corollary: Social organization and thus

mechanisms of knowledge transmission are shifting among indigenous peoples, through their

growing participation in and assimilation of western institutions, such as schools, political

organizations, markets and development projects (Turner, 1991; Posey, 1994; Jackson, 1994;

1995; Gray, 1997; Oakdale, 1996; 2004; Godoy, 2001; Godoy et al., 2005; Reyes-Garcia et al.,

2005; Silva, 2009; Zent, 2009 b). It is expected that greater levels of market integration and

formal schooling lead to knowledge erosion (Nabhan and St. Antoine, 1993; Zent, 1999; Benz,

2000; Godoy, 2001; Godoy et al., 2005; Reyes-Garcia, 2001; 2005).

H3: I hypothesize that greater levels of market integration lead to erosion of weaving

knowledge across three Kaiabi groups.

H4: I hypothesize that formal schooling erodes indigenous knowledge: higher levels of

formal schooling leads to lower levels of weaving knowledge among men and women.

H5: Role of community-based projects: I hypothesize that the Kaiabi Araa project is

responsible for an increased number of basketry and textiles weavers among the younger

generations. In villages which are participating in the project, there are more youth who know

how to weave now in comparison to 2002 and to other villages which did not participate in the

project.









Key theoretical-methodological definitions:


1. Cultural resilience I define cultural resilience here as the capacity of human societies to
absorb change and adapt to novel situations without collapsing or losing their identity as
a group (my adaptation, based on Gunderson, 2003). This is a very subjective concept;
therefore we need concrete ways to operationalize it to be able to compare it within a
given cultural group. We can extend the concept to embrace the capacity of retention,
innovation and recreation of ethnic identity and of cultural domains, such as native
language, agrobiodiversity, agricultural practices, environmental knowledge, medical
knowledge, artistic knowledge, among others. For this research, cultural resilience will be
operationalized through measurements of indigenous knowledge domains, such as
proficiency in the native language and level of weaving knowledge (see next sections).

2. Environmental conservation and management Applied to indigenous lands, this
concept relates to the degree of conservation of a given territory (deforestation levels,
patchiness, threats, invasions), as well as its extension. It is operationalized here as the
capacity of indigenous peoples to control, monitor, access and use natural resources
existing in a defined geographical space. Also it relates to the legal status of the land, in
terms of Brazilian law (stages of recognition of an indigenous land, see Chapter 4).

3. Political empowerment The ability of a people to mobilize and influence change, on the
basis of their access to knowledge, political processes and financial, social and natural
resources (Slocum et al., 1995). For this research, political empowerment will be
operationalized through: a) degree of access and capacity of mobilization of financial
and technical resources, by each Kaiabi group, in the management of their territories; b)
degree of participation and information of community residents in political processes
(such as participation in meetings and knowledge on the role of indigenous
organizations).

4. Market integration Integration of indigenous peoples with market economy can take
diverse approaches and therefore be studied for specific purposes using different methods
(see Godoy, 2001 for a comprehensive analysis on indigenous peoples, forests and
markets). In this research, I'm interested in exploring the effects of market integration on
a specific domain of indigenous knowledge, i.e., weaving knowledge of basketry and
textiles. To do this, I estimate the level of income stability (cash income) of each
interviewee (see next sections on data collection and analysis). I also use qualitative data
on the distance and form of relationship with markets in each village (Godoy, 2001;
Reyes-Garcia, 2001): for example, by knowing which products are sold, by whom, how
and with what frequency.

5. Weaving knowledge This is a specific domain of indigenous knowledge, related to the
ability to weave basketry and textiles among men and women (Athayde et al., 2009). To
operationalize this concept and measure it as a variable, I used three measurements (see
next sections): a) number of items of basketry or textiles items an individual masters; b)
number and complexity of graphic designs that an individual masters; c) capacity of an
individual to name basketry and textiles designs.









Research Design

Structured, semi-structured and open-ended interviews were carried out among the

residents men and women over 15 years old of the four villages included in the research, as

well as with key informants such as shamans, political leaders and elderly residents in other

villages. I used the same age criterion adopted in my previous master thesis, to be

methodologically consistent (Athayde, 2003). Reyes-Garcia (2001) also used 15 years old as the

age criterion to include individuals in research on effects of market integration on ethnobotanical

knowledge among Tsimane indigenous people from Bolivia. A copy of the codebook used to

input data in my database is presented in Appendix A. Narratives and conversations which were

recorded previously were also used. Semi-structured interviews included four types of

information: 1) Environmental management and land claims; 2) Political empowerment; 3)

Socio-economic aspects; and 4) Knowledge on basketry and textiles weaving. Below, we

describe each type of information collected through the individual interviews.

1. Environmental contrasts, conservation and management Data related to this topic are
mostly presented in Chapters 4 and 5. Secondary and primary data were collected in
order to: a) compare environmental differences among the three areas; b) estimate the
conservation status of each of the three areas; c) explore mechanisms for environmental
adaptation developed by the Xingu Kaiabi; d) survey plant species of cultural importance
among residents of four villages; and e) describe each group's involvement in official
land claiming processes. To survey plant species of cultural importance I conducted a
free list exercise, in which I asked each participant to list at least five most important
plant species that occur in the ancestral land. I complemented this free list exercise with
open-ended interviews carried out with elders and key informants, asking them to
mention important plant resources for the Kaiabi that only occur in the ancestral areas, or
that occur sparsely in Xingu.

2. Political empowerment Data related to this topic are mostly presented on Chapter 6. An
outline was developed to collect specific data on political empowerment and the situation
of political organizations in each land (Xingu ATIX; Rio dos Peixes Itaok and Teles
Pires Kawaip), which included collection of secondary, qualitative and quantitative data
on: a) access to resources and opportunities (both financial and technical support); b)
participation of community residents in political meetings and processes; c) type and
degree of knowledge on the work performed by each indigenous organization; d)









preference for projects to be developed by each association in the future; and e) problems
and challenges faced by each organization.

3. Socio-economic data -_Socio-economic data were collected through semi-structured
interviews conducted across the three Kaiabi groups. Socio-economic data are mainly
presented and discussed on Chapter 7 and 8. Data collected included the following
themes and variables:

* Brief life history in the case of elders and shamans, in order to understand historical
facts, mobility patterns and processes of learning and transmission of knowledge.

* Estimated age it is expected that older people will be more knowledgeable than younger
people. Data on age were collected and/or confirmed through consultation of secondary
data and of health service records.

* Kinship this also included interethnic marriages between the Kaiabi and other indigenous
peoples as well as with non-indigenous.

* Integration with market economy this was assessed through estimates of cash and
stability of income and types of economic activities, functions and professions developed
in each village. It was estimated through interviews and, in the case of FUNAI officers,
health agents, retired people and teachers, through their official salary reports. I created a
variable to represent the stability of income of each individual with four levels: 1 Waged
officer; 2 Retirement pension; 3 Bolsa-familia (Family pension, a program of Brazilian
government); 4 Irregular does not have a monthly wage and does not receive any
government benefit.

* Language proficiency this was measured through a proficiency scale with 4 levels for
speaking and understanding abilities: 0- none (does not speak and does not understand); 1-
little (understands but does not speak); 2- regular (speaks and understands, but not fluent);
4- good (fluent).

* Schooling for formal schooling, there were also four levels considered: 1- Illiterate; 2 -
Elementary (complete or incomplete); 3 High school (complete or incomplete); 4 -
University (complete or incomplete).

* Travels and dislocations in order to evaluate the nature and frequency of contacts and
exchanges between the three areas, I asked people how many times they had travelled
between their villages and the other two Kaiabi areas and for what purposes. For instance, I
asked people in Rio dos Peixes how many times they had travelled to Xingu Park and for
what reason. The same procedure was adopted for the other areas and villages.

* Participation in the Kaiabi Araa project to compare the distribution of artistic knowledge
between individuals and villages that are participating in the project and villages that are
not. This enables assessment of the impact of the project on weaving knowledge.









In order to be able to carry out some statistical analysis, I had to transform some variables

that had four levels of measurement such as language proficiency, income stability and

schooling in only two levels. The reason is that there weren't enough people in each of the

categories to allow some statistical procedures such as correlations and logistic regression

analysis. Thus, I transformed these variables in only two levels, with yes or no responses: for

example, if the person did or did not have proficiency in the Kaiabi language, income stability or

formal schooling.

4. Longitudinal and cross-sectional studies on weaving knowledge transmission and
distribution- Data related to this topic are presented on Chapter 8. This included semi-
structured interviews on knowledge acquisition and transmission related to basketry and
textiles weaving conducted amongst men and women over 15 years old at Xingu Park,
Rio dos Peixes and Teles Pires areas. A catalog containing photos of basketry and textiles
designs was used to collect data on names and uses of the designs, as well as on the
sequence of learning. The photos contain the same codes used for the interviews done in
2002 at Xingu park and in 2004 at Kururuzinho village (Teles Pires). The main aspects
and variables considered were:

* age at which began to learn (when?);

* how the person learned (how?);

* with whom the person learned (who?);

* knowledge on design names;

* knowledge on techniques used in weaving;

* ability to weave the design (uses);

* knowledge on mythical or special attributes of the objects (quantitative data, to assess
differences between elders and shamans in relation to other people);

* ability to weave different types of basket and textile items.

* ability to represent graphic designs in other artefacts to assess innovation of artistic
knowledge;

* participation in the workshops of the Kaiabi Araa project, and basketry designs and/or
basketry and textiles objects learned at these events;









* knowledge on natural resources used in basketry and textiles weaving which species are
used, where they are collected, which varieties of cotton are used, which cotton varieties
each woman has in the family garden plot, which cotton varieties are declining.

Data Analysis

I'm aware of the risks associated with reduction of ethnographic data and complex socio-

ecological systems to quantitative data and numbers. Therefore, I attempt to use qualitative

methods as complementary to the ethnographic information available about the Kaiabi, in order

to better understand socioeconomic, cultural, political and environmental aspects of each of the

three groups involved in this study. Data analysis done for this research consisted in a

combination of quantitative univariate, bivariate and multivariate methods and models (Agresti

and Finlay, 1997; Bernard, 2006). In addition, I used the cultural consensus model to understand

the variability of knowledge related to basketry and textiles designs among men and women

across the three groups (Romney et al., 1986; Weller, 1987).

A database on Microsoft Excell was structured in order to enter data for the different

units of analysis to be considered. Other software include SPSS to run descriptive, univariate,

bivariate and multivariate statistical analyses. Ucinet was used to run consensus analysis, MDS

(multidimensional calling) and cluster.

A comparison between the results of the analyses of the data on basketry and textiles

knowledge collected in 2002 and 2004 in Xingu Park and in Teles Pires amongst men and

women, with the data collected in 2007, was carried out in order to assess changes in

mechanisms of: 1) learning and transmission (for example, from kin groups to books and

community-based projects); 2) distribution (did distribution of knowledge change in a five year

period? How?); 3) innovation (how, where, who and why innovation is taking place?); and 4)

degree of artistic knowledge (What is the situation of different villages in relation to the degree









of knowledge and endangered designs?). Below, I summarize the methods and models developed

to test the hypotheses explored in this research:

a) Univariate analysis- Univariate analysis was used as a first step to examine variables in
depth. This included coding, cleaning data, running measures of central tendency and
visualizing distributions.

b) Bivariate methods Chi-square tests were used to verify an existing association between the
variables. Correlation tests were used to evaluate the strength of the relationship between two
variables.

c) Multivariate methods A set of statistical procedures is used in multivariate analysis in order
to test hypotheses about "how independent and dependent variables are related, based on a
theory of causation" (Bernard, 2006: 649). Multivariate analysis was used to test hypotheses,
studying the combined effect of two or more independent variables on the dependent
variables, which in this case were language proficiency and weaving knowledge.

* Logistic regression model To better understand the interaction of different socioeconomic
variables with the dependent variable, language proficiency, I used a logistic regression
model, which provides a calculation of the probability that a category (dependent variable)
depends on the values of the explanatory (independent) variables. This analysis describes
the structure of the association among a set of qualitative response variables (Agresti and
Finlay, 1997).
* Multidimensional scaling (MDS) and cluster multivariate visual techniques were used to
tease out underlying relationships among a set of observation. A non-metric MDS analysis
was carried out in order to identify similar groups among the participants, in relation to
their ability to weave the basketry or textiles designs. Like MDS, cluster analysis is a
descriptive and visual technique for exploring relations among items in a matrix of
similarity. It comes from the consensus analysis, using UCINET Program.

d) Consensus analysis Consensus analysis is a technique developed in cognitive anthropology
for the analysis of structured interviews, based in the theoretical and methodological
premise that culture can be studied and measured by the degree of shared knowledge,
between members of a group, related to a given cultural domain (Romney et al., 1986).
According to Caulkins and Hyatt (1999), consensus analysis produces three main results:

* a measure of the degree of agreement between informants about a domain of knowledge,
belief or practice;

* the "culturally correct" information about a domain according to the pooled answers of the
informants;

* a score for each informant representing that person's knowledge of the domain.









Consensus analysis was used to compare participants based on the names they give for the

different designs, verifying the extent to which Kaiabi men and women share knowledge on

names given to designs and teasing out relationships amongst the participants in terms of how

much they agree on the different names that a design might have. I adopted the working concept

proposed by Weller (1987), of cultural knowledge as a pool of information or elements. In this

case, shared knowledge "is represented by the proportion of those elements that are shared or

held in common between an individual and the pool" (Weller, op.cit:180). Shared knowledge on

names for basketry designs can be inferred by the proportion of names for any given design that

are shared or held in common between a man or a woman and the group. In this case, there is no

correct answer and every individual's knowledge will be estimated by how much his or her

knowledge of designs names fits into the group's overall knowledge.

One problem faced here was that because this is a specialized knowledge domain, most

people interviewed in the four villages didn't know the names for designed baskets or textiles.

Therefore, to be able to run the consensus analysis, instead of assigning zero for "don't know"

answers, I assigned random values to them, different than those assigned to the responses for the

designs. I also had to reduce my sample to include people that knew the name for at least 10% of

the 36 designs for men and 10% of the 9 designs for women (Weller, 1987). This data

transformation resulted in the inclusion of 29 men with 14 basketry designs and 34 women with

3 textile designs in the consensus analysis.

For the data presented in this dissertation, I considered Tatuy Post and Novo Horizonte

village as one single village in Rio dos Peixes area. This merging was done because both places

are adjacent and very similar in terms of social organization, family composition and economic

activities and also to facilitate and improve data analysis.









I used a standard conversion rate from Brazilian currency (R$ Real) unit to the dollar (in

which 1U$=2.00 R$).









CHAPTER 3
THE KAIABI IN TAPAJOS AND THE TRANSFER TO XINGU PARK

Introduction

The history of the Kaiabi people is interlinked with the history of Amazonian conquest,

development and conservation. Mobility, territoriality and cosmography are key concepts to be

considered in the analysis of Amazonian indigenous peoples' history and identity construction

prior to and after contact with westerners. Little (2001 : 4) describes territoriality as "the

collective effort of a social group to identify with, occupy, use and establish control over the

specific parcel of their biophysical environment that serves as their homeland or territory."

Before contact, indigenous Amazonian groups were accustomed to different kinds of

movement, from nomadic to hunting and gathering expeditions, or group migrations, to fleeing

or fighting other enemies, procurement of sacred sites, mobility up and down the river, or simply

traveling around (Viveiros de Castro, 1992; Grunberg, 2004; Alexiades, 2008). Clastres (1995),

analyzing prophetism in Tupi-Guarani migrations, studied how myth, oratory, music and

shamanism interacted in the Tupi restlessness for identity construction, and in search of "the land

without evil" (Fernandes, 1963; Clastres, 1995).

These multiple migrations were followed by processes of reterritorialization (Little,

2001). In the process of establishing and maintaining human territories, a mutual influence

occurs between the earth and its inhabitants. The study of these mutual relationships,

"encompassing collective and historically contingent identities and environmental knowledge

systems developed by a social group" is what Little (2001 : 5), after Boas and other scholars,

defined as cosmography. It also includes the symbolic and affective relationships with land and

nature. Therefore, cosmography and cosmology interplay and express the ways by which a group

relates to its territory and reproduces its social order through time and space.









In this chapter, based on archive research and testimonies, I present a chronological

description of the Kaiabi social and geographical organization in the Tapaj6s region, as well as

the process of encounter and involvement with the rubber tappers in the first half of the 20th

century. Next, I describe the encounter with the Villas-B8as brother and the process of transfer

of the majority of the group to the Xingu Park.

The Kaiabi in the Tapaj6s Region

The Kaiabi are a Tupi-Guarani speaking people who originally occupied several

tributaries of the Tapaj6s River in the southern Brazilian Amazon. Oral and written accounts on

the Kaiabi suggest that by the beginning of the nineteenth century, they were divided into two

main and interconnected groups, linked to tributaries of the Tapaj6s, the Teles Pires, the Arinos

and the Rio dos Peixes, a tributary of Arinos (Steinen 1940; Villas B8as and Villas B8as, 1989).

According to Joao Dornstauder, a catholic priest who worked almost 30 years among the Kaiabi,

they reported that: "the area occupied by our ancestors was the entire extension of Rio dos

Peixes, between the Juruena and the Arinos, going up until the headwaters of the Rio dos Peixes

and from there joining with the upper Teles Pires, along whose riverbanks and tributaries there

were villages, going down as far as the Peixoto de Azevedo river, from where there was

communication with the Rio dos Peixes, up to the Jawary stream, from where they cross to the

Coata stream headwaters" (Figure 3-1, Dornstauder, 1985: 5) In this region, they had contact

with the Munduruku and Apiaka to the north; with the Rikbaktsa at Arinos; with the Tapayuna

who inhabited the Arinos and the Rio do Sangue; with the Panara in the Peixoto de Azevedo

River; and with the Bakairy to the south, at the headwaters of the Teles Pires (Steinen 1940;

Dornstauder 1985; Dal Poz 1996; Grinberg, 2004).

Mobility, extensive within and between the Rio dos Peixes and Teles Pires Rivers, was

linked to extended foraging expeditions, as well as war and trade with neighbours (Grinberg









2004). Individual settlements, consisting of an extended family group organised around a male

patriarch or chief, called wyryat, were located between four to seven kilometres from the nearest

grouping (Dornstauder 1955; Steinen 1940). The Kaiabi, similarly to many other Tupi-Guarani

groups, have a uxorilocal1 post-marital residential system, which reinforces the affinity between

fathers-in-law and sons-in-law (Senra, 1999). The extended uxorilocal family has constituted the

basic unit of the social, economic and political Kaiabi structure (Grtinberg 2004; Senra et al.,

2004).

According to Mairaw6 Kaiabi (1981), there were various Kaiabi villages along the upper,

middle and lower Teles Pires and its tributaries, as well as along the Arinos and Rio dos Peixes.

This was a strategy to hide from their enemies, mainly the Munduruku. In fact, warfare between

the Kaiabi and neighboring groups such as the Bakairi, the Munduruku and the Apiaka, was

reportedly the main factor leading to depopulation and village relocation at that time (Grinberg

2004; Villas B8as and Villas B8as 1989). During war times, the Kaiabi used to capture women

and children from enemy villages.

Subsistence was based on a sophisticated swidden-fallow agricultural system, which

included root crops such as manioc, taro, yams and sweet potatoes, as well as other crops such as

maize, peanuts and bananas (Grinberg 2004). The Kaiabi had used, and still continue to use a

criterion for establishment of their villages based on soil fertility and on the occurrence of black

earth2 terrapreta anthropogenic soil (Villas B8as and Villas B8as, 1989; Rodrigues, 1993;



1 Uxorilocal residential system is common to many Amazonian and Tupi indigenous societies and refers to the
husband moving to live in his wife's group after marriage (Viveiros de Castro, 1992).
2 Named "Terra Preta de Indio" in Portuguese, these anthropogenic soils ("anthrosols") have been formed by the
deposition of ashes and debris of past human occupation, representing the most important vestige of pre-historic
Amerindian populations (Petersen et al., 2001). They are an inheritance left by these ancient indigenous cultures,
since they have high fertility for agriculture, and have been used in shifting agriculture by contemporaneous
indigenous peoples for centuries (Denevan, 2001).









Silva and Athayde, 1999). According to them, there are many plants which are 'terra preta'

indicators, which thrive on these special soils as a result of the management done by ancient

indigenous populations (Silva and Athayde, 1999; Schmidt, 2001).

The exchange of knowledge, information, people, plants and objects among Amazonian

indigenous peoples has played an important role in the construction of identity, both prior to and

after contact with westerners (Alexiades, 2007; Alexiades and Peluso, 2007; Zent, 2007).

According to Kupeap (a knowledgeable elder born at Rio dos Peixes, son of the great Temeioni

chief), Kaiabi men learned how to weave some basketry designs from the Apiaka, one of their

fiercest enemies. Likewise, the main Kaiabi festival, "Jowosi," is related to war. In the past,

Jowosi festivals usually lasted for many months, with families and other kin exchanging visits,

labour, knowledge and history encoded in Jowosi chants (Travassos, 1984; Oakdale, 1996).

Jowosi lyrics are improvised by the men and great warriors, and repeated many times by the

women, men singing and dancing in front and a group of women following behind. In ancient

times, enemy skulls were used in Jowosi as trophies by men to celebrate war victories

(Travassos, 1984).

According to Senra (1999:15) "...the Kaiabi conceive the cosmos as divided in different

superposed layers, inhabited by supernatural beings or spirits." There are the lords or owners of

the animals, the dangerous "ajdng" and "mama'e", who can cause diseases and steal people's

souls and the "ma'it," the shamans who live in the heaven. There are also the ancestral mythical

heroes, who taught the Kaiabi everything they know. Amongst these, Tuiarare is considered the

creator of Kaiabi people. He was both a great warrior and a shaman. Shamanism is a

fundamental aspect of Kaiabi culture. The shamans are the intermediaries between the natural









and the supernatural world. The Kaiabi shamans can be men or women, but the most powerful

are men.

By the late nineteenth century, when the first encounters occurred between Kaiabi and

Brazilian (government officials and explorers looking for minerals) and European expeditions,

the Kaiabi were living in several dispersed settlements. Between 1884 and 1887, Karl Von den

Steinen the well-known German explorer coordinated the two first German expeditions in

Xingu. His field reports are the first written registers about the Kaiabi, which he called "Kayabi"

(Steinen, 1940). He encountered two Kaiabi women prisoners of the Bakairi in the upper Teles

Pires (also known as Paranatinga in Tupi-Guarani, "white river"). Steinen3 reported that the

Kaiabi had the monopoly of stone axes in the upper Paranatinga region. According to what some

Bakairi told Steinen at that time, the Kaiabi used as self-designation "Parua" and spoke a

language similar to the Kamayura from the Kulisehu River in the upper Xingu region.

Nevertheless, it was impossible to confirm if this self-designation was correct, because none of

the Kaiabi remember about this nowadays.

In 1889, the Brazilian officials Lourengo Telles Pires and Oscar Miranda de Oliveira

Miranda carried out an expedition to the Paranatinga to take measures of the river.

Unfortunately, their boat sank in a rapids near the parallel 100 S. Only six people survived. The

famous Brazilian official Marechal Cdndido Rondon suggested that the name of the river be

changed to Teles Pires in honour of Lourengo Telles Pires (Rondon, 1916, cited in Griinberg,

2004).





3 Many of the artifacts collected by Steinen at that time are deposited in the collections of the Museum of Ethnology
in Berlin, Germany. Recently, the museum sent copies of photos of all Kaiabi objects back to the Kaiabi, via their
local association, ATIX.









After the short and intermittent encounters with the non-indigenous Brazilians and

Europeans during the nineteen century, Kaiabi in the Teles Pires and Rio dos Peixes rivers

established permanent contact with Brazilian national society in the late nineteenth century and

mid-twentieth century, respectively (Villas B8as and Villas B8as 1989). At that time, mobility

was also related to the possibility of getting industrialized goods, and even to establish contact

with the non-indigenous, who would provide the Kaiabi with unknown goods and valuable tools,

such as machetes and firearms. Xupe, an elder in the Tuiarare village, still remembers that

"everybody was crazy about tools and axes, people were going after the whites to get these

tools."

In 1861, rubber tappers were seen at the mouth of the Teles Pires, in the Juruena River

(Chandless, 1861, cited in Grfinberg, 2004). Between 1877 and 1907, large numbers of tapayin

('white people' or Brazilian nationals) -many from the drought-stricken north-eastern region of

Brazil- poured into the Teles Pires to work rubber (Hevea brasiliensis). There were two fronts of

penetration: in the Arinos-Juruena region, in the lower Teles Pires, and in the upper Teles Pires

(Grfinberg 2004). Enemies of the Kaiabi, such as the Bakairi and the Apiaka, were 'pacified' at

this time and began to work in several of the seringais (rubber-tapping estates). The Teles Pires

Kaiabi began to acquire steel axes, machetes and firearms through exchanges and wars with

these groups, which they in turn subsequently exchanged with the Rio dos Peixes Kaiabi.

Increased conflicts with the rubber tappers, and the effect of epidemics, led to population decline

and to increased mobility among the Kaiabi. The Apiaka, also a Tupi-Guarani group with whom

the Kaiabi share territorial rights in Rio dos Peixes and Teles Pires nowadays, were almost

totally exterminated by rubber tappers at the beginning of the 20th century. In 1912, there was a









register of a total population of only twenty Apiaka survivors (Campos, 1936, cited in Grunberg,

2004).

Aiming to pacify the Kaiabi, the Indian Protection Service (SPI), the first governmental

institution created to 'pacify and protect' the Brazilian indians, established the Pedro Dantas post

on the Teles Pires river in 1922. Two years later, this post was totally destroyed by a massive

attack by the Kaiabi, in which two SPI officers died. In 1929 the post was re-established 10 Km

to the south. Kaiabi military resistance to SPI actions and to rubber tappers finally ceased around

1936, at which point they became one of the last groups in the upper Tapaj s basin to become

engaged in rubber tapping (Grunberg 2004).

One of the first ethnologists to study the Kaiabi was the German Max Schmidt. He spent

only two months with the Kaiabi in the Pedro Dantas Post, due to the difficult relationship with

the Kaiabi and sickness (he got malaria several times). However, his publications turned out to

be very relevant to determine Kaiabi linguistic affiliation in the Tupi-Guarani family, in addition

to the ethnological information, photos and artefacts collected4 (Schmidt, 1929; 1942).

The renewed international demand for rubber during the Second World War coincided

with the establishment of a second SPI post in the Teles Pires in 1941. Encouraged, sometimes

forced, by SPI officers, many Kaiabi men from upper Teles Pires villages began to work in the

seringais, with some families settling in the post in 1942. By 1953, most Teles Pires Kaiabi were

thus actively engaged in tapping rubber (Grunberg 2004). Even though the Rio dos Peixes region

was colonised later than the Teles Pires- the first rubber estate was established around 1951- by

1955, many Kaiabi from Rio dos Peixes were also working with rubber tappers (Grunberg 2004).

The Kaiabi worked through the aviamento system, receiving the necessary credit, food and

4 These artifacts, similarly to the ones collected by Von den Steinen, are also part of the collection of The Museum
of Ethnology in Berlin, Germany. I repatriated photos of this collection to the Kaiabi in 2004.









equipment in advance, and paying back later through the harvested rubber. In the seringais, men

learned Portuguese, and developed links with the non- indigenous market economy. Some people

began to travel to nearby cities, bringing manufactured goods back to the villages. Women began

to abandon the manufacture of traditional pottery, relying instead on metal pots (Villas B8as and

Villas B8as 1989).

The government agency SPI was set up to contact, provide certain kinds of assistance to

and, if necessary, assist the relocation of indigenous peoples. Indigenous groups were generally

and initially hostile to the SPI, sometimes destroying their posts and killing their officials, as the

Kaiabi did at Pedro Dantas Post in 1929 (Grinberg 2004). The FUNAI or National Indian

Foundation replaced the SPI in 1967-1968 following serious accusations of misconduct and

corruption. Nevertheless, a paternalistic approach to the protection of indigenous peoples is still

maintained to this day (Ramos 1998). According to accounts by Griinberg (2004), the situation

of SPI Posts servicing the Kaiabi population between 1940 and 1960 was precarious. Many

Indians would die at these posts because of lack of proper medical treatment and remedies. Also,

SPI officers contributed to attract labor force to the seringais and many times were abusive with

the Indians.

The occupation and colonization of the Arinos river in the Rio dos Peixes region inside

Kaiabi territory began in 1955, with clearing and measurements of the right bank of the middle

Arinos by the CONOMALI colonizer enterprise (Companhia Colonizadora Noroeste Mato-

Grossense Ltda). Dozens of families of agriculturalists, mostly German descendents, migrated

from Rio Grande do Sul to the newly created locality of Porto dos Gauchos, which still exists

(Grinberg 2004).









By the middle twentieth century, the Kaiabi were facing an increased encroachment of

their territory and simultaneous closer contact with non-indigenous rubber tappers, settlers,

miners and animal poachers. They became increasingly dependent on the goods provided by the

"patr6es" (owners of the rubber tapping estates) and also on medical relief provided by

missionaries and, to a lesser extent, by the SPI. However, after the decline of rubber extraction

due to a new cheaper Asian supply, there was no place for indigenous workers in the boom-burst

Amazonian frontier. At that time, the Kaiabi were living in scattered villages throughout the

upper and middle courses of the Teles Pires River. Some of them would go searching for safer

places to live, running away from rubber tappers and miners. Some Kaiabi men worked for cat

hunters, or fur trappers (named "gateiros" in Portuguese), which appeared in the region right

after rubber production collapsed.

Contact with the Villas B6as Brothers

In June of 1943, the Roncador-Xingu Expedition (Expedigco Roncador-Xingu) was

officially created by the Brazilian government to prepare the occupation of a "large empty"

region located in the central-western region of Brazil, comprising part of Goias, Mato Grosso

and Para states, including the area located at the Xingu river headwaters (Menezes, 2000). The

expedition's objectives were to explore the territory; discover natural richness, mainly minerals;

open roads; build landing stripes for airplanes; prepare the land to be occupied by several groups

of migrants coming from many places in Brazil and from Europe; and establish points of

radiotelegraphic communication (Menezes, 2000). The expedition was part of a greater

government plan, named "March to the West" ("Marcha para Oeste'). According to Villas-B8as

and Villas-B8as (1994), since the beginning of the twentieth century, the Brazilian government

had the idea to transfer the capital of Brazil from the coast (at that time, Rio de Janeiro) to the

interior of the country. After the First World War, land contractors began to sell plots of land in









the central region of Brazil, where the capital would be built. However, only during the Second

World War was the idea taken up again. The government wanted to take a step further, not only

transferring the capital, but also promoting the occupation and development of the interior or

"sertdo" (countryside, backwoods, bush). At that time, most of the nearly 43 million Brazilians

lived in the country's costal region. Months later, in October of 1943, the government

established the Fundagdo Brasil Central (FBC, 'Central Brazilian Foundation'), with the specific

function of colonizing or establishing small population settlements in strategic sites designated

by the Roncador-Xingu Expedition (Villas-B8as and Villas-B8as, 1994).

Various cities, villages and landing fields were established by FBC during more than two

decades of its existence. The Roncador-Xingu expedition founded the town of Xavantina in

1944, reaching the headwaters of Xingu river and constructing the Jacare landing field for the

Brazilian Air Force (Forca Aerea Brasileira, FAB) in 1947, inside the current boundaries of the

Xingu Indigenous Park (Menezes, 2000; Grunberg, 2004). The FBC had the power to acquire

lands given through state concessions, to contract enterprises, to share profits with them, and

even to decide on indigenous peoples' relocations. Everything was legalized and officialised by

the Brazilian government, through legal instruments such as decrees. FBC actions on land

occupation and destination generated numerous conflicts and confusion that persist to this day

(Menezes, 2000). Regarding indigenous lands, many times the FBC and SPI territorial interests

would overlap. Often, the FBC would impose its interests over SPI or over indigenous peoples'

claims, acting against land interdiction for the establishment of indigenous reservations or lands

(Menezes, 2000).









Working as sertanistas for the FBC, the Villas-B8as brothers5 had the mission of

pacifying and relocating indigenous peoples living in the region to be colonized by southern

Brazilian migrants. They contacted thousands of indians, belonging to different groups, who

lived in the presumed "emptiness" of central Brazil: the Xavante, the Kalapalo, the Ikpeng

(Txicdo), the Kamayuri and other upper-Xingu peoples, the Kayap6, the Panard, and the Kaiabi,

among others. The FBC and the Roncador-Xingu Expedition were extinct in 1967, with the FBC

substituted by the SUDECO (Superintend6ncia do Desenvolvimento do Centro Oeste).

According to Souza (1994:18), the expedition and the FBC produced 1,500 km of opened trails,

1,000 km of navigated rivers, 43 towns and villages, 19 landing fields, and contacted five

thousand indigenous. Orlando and CliudioVillas B8as spent forty years of their lives working for

the expedition and then in the administration of the Xingu Indigenous Park (Souza, 1994).

In 1949, during the first years of the Roncador-Xingu expedition, the Villas-B8as

brothers and their team arrived in the Teles Pires River by foot and by boat, opening a trail from

the Diauarum Post in Xingu to the Teles Pires river, in almost three months opening up the

"sertdo" (Villas B8as and Villas B8as, 1994). They had the mission to build a landing field

(Cachimbo) in the Teles Pires region, where they established a campsite some 12 km from the

river. Claudio and Orlando were bringing provisions from the river to the camp, when they

began to listen to many animal sounds: jaguars, monkeys, birds and even wild pigs. They found

it strange to hear the sound of so many animals together, and, they were immediately wary of

some unknown Indians who might be playing with them. They already knew that the Kaiabi

lived in villages near the place where the campsite was located. Claudio and Orlando began to

5 Sertanistas were government agents responsible for contacting and 'pacifying' indigenous groups during the
development of the Amazonian frontier, or the interior, also called 'sertio.' The Villas-B6as brothers- Leonardo,
Claudio and Orlando- became nationally famous for the pacification of various indigenous groups, and for the
creation of the Xingu Indigenous Park. Orlando Villas-B6as was the last brother to die, in 2002.









call: "Kaiabi, Kaiabi!" and the answer came immediately from the opposite bank in the form of a

strident scream. Then, two Kaiabi men in a bark canoe came to the middle of the river, observing

the white foreigners. The two Villas-B8as got their canoe and paddled in their direction to meet

them. But the Kaiabi became wary and went back to the riverbank, hiding in the bushes. The

Villas B8as went after them, pronouncing some words in the kamayura language (which they

knew was in the same linguistic family as the Kaiabi language), showing and offering their

machetes to them (Villas-B8as and Villas-B8as, 1989; 1994). The indians agreed to accompany

the Villas B8as to the campsite, where they were offered other gifts. One month after this first

contact, there were around fifty Kaiabi visiting the campsite ibidd).

The Villas B8as enlisted three Kaiabi as guides for their expedition to the northwest of

Mato Grosso State: Prepori, Cuiabano or Jurumuk and Ip66 (Villas-B8as and Villas-B8as 1989;

1994). At that time, the Kaiabi in the upper Teles Pires had a closer contact with rubber tappers,

being considered "tamed", as opposed to the Rio dos Peixes Kaiabi who were more isolated, the

so-called wild or untamed "Tatuy" Kaiabi (Sim6es, 1963; Villas-B8as and Villas-B8as, 1989).

The Indigenous Noble Savage and the Creation of the Xingu Park

The establishment of landing fields in different areas of the central Brazilian region by

the Roncador-Xingu Expedition and the FBC had a strategic and military objective, since these

fields were to be used in the military route linking Rio de Janeiro to Miami by the FAB as well

as by the US Air Force (Menezes, 2000). In the Xingu Tapaj6s region, the main landing fields

were located inside territories occupied by indigenous peoples: the Jacare Post in the upper

Xingu region and the Serra do Cachimbo landing field, near the Teles Pires river (ibid.). The


6 Prepori was a great political leader and shaman. Before him, his father was also a great Kaiabi chief (Mairaw6
Kaiabi, 1981). He was the last great Kaiabi shaman to die, in 2002. Jurumuk or Cuiabano was a knowledgeable elder
and also a shaman, who passed away in 2005. Ip6 is a retired FUNAI officer who now lives in Diauarum Post with
his family.









existence of an airport in the upper Xingu, allied with the sympathy of the Villas B6as to

journalists, scientists, film makers and politicians, opened up access by the academy and media

to the previously poorly- known and ignored Xingu indigenous groups: "the state of virginity of

the land and the men's primitivism cannot be ignored by scientific curiosity" (Menezes,

2000:73). The public image of Xingu as a natural and cultural sanctuary and paradise was

constructed and publicized by the media, being highly influential in the creation of the Park as a

postal card of Brazilian natural and cultural purity. The 'xinguanos' were portrayed as good and

happy human beings, not (yet) spoiled by western society's greed, viciousness and diseases

(Menezes, 2000; 2001). This image was so strongly constructed and reinforced in the last half

century, that it has persisted until now in political discourse, in scientific discourse, and even in

indigenous discourse, as indigenous peoples appropriate the image constructed for them to

defend their own interests (Conklin, 1997; Fisher, 1994; Jackson, 1995).

The indigenist practice of the Villas B6as had a marked influence on the territorial and

cultural re-organization of what is known today as the Xingu Park. A partnership and agreement

was established between the FBC and the Museu Nacional (National Museum) of Rio de Janeiro.

At that time, various naturalists, anthropologists, and other scientists, Brazilians and foreigners,

carried out research about the upper Xingu region and the indians. The amount of scientific

information produced, combined with media publicity and with pressure from indigenists,

scientists and the Villas B6as, created the basis for the first indigenous territory of large

proportions to be officially recognized by Brazilian law (Menezes, 2000).

The objective of the government during the developmentalistt" post World War II period

was the assimilation of the remaining indigenous population into the larger national society

(Ribeiro, 1995; Little 2001). The idea of "assimilation" was clearly expressed (and at the same









time contradicted by the idea of "cultural preservation") in the first paragraph of the first Indian

Statute from 19737: "This law regulates the judicial status of the indians and of indigenous

communities, with the goal of preserving their culture and progressively and harmoniously

integrating them into national society" (ISA, 2005). The Brazilian government used the

classification proposed by Ribeiro (1970) to determine if indigenous peoples were considered

isolated, in process of integration, or fully integrated with the national society. Before the

proposal of the creation of the Xingu Indigenous Park, the SPI policy towards indigenous lands

or "reserves" as they were called- was to establish fixed areas of small proportions to ensure the

sedentarization of indigenous groups until their complete assimilation by Brazilian society, as

national workers (Menezes, 2000). Therefore, it was congruent for the government to establish

small indigenous reserves which would enable only the physical -not cultural reproduction of

indigenous peoples, as they were considered human beings in transition towards full cultural

assimilation. This strategy would make possible the liberation of lands to be occupied by

Brazilian society, with the indigenous reserves serving as a source of cheap labor for the

economic expansion of the country (Oliveira-Filho, 1983).

In the beginning of the 1950s, the SPI was under pressure to regulate the Federal

Constitution of 1934, in regard to indigenous lands and reserves. According to Lima (1989),

cited in Menezes (2000), the governor of Parana state at that time, presented a Law project (Bill

n 245) to regulate Article 216 of the constitution, concerning indigenous peoples' ownership

rights over land. The project foresaw the titling of indigenous reserves, with private ownership

titles given to families instead of to the indigenous groups in a common property regime. This

project was adapted from the U.S. law of 1887. Fortunately, the president of the SPI did not

I will discuss this law in more detail in Chapter 6, when describing indigenous grassroots movements in the
Brazilian Amazon.









accept the project, and an SPI ethnologist, the well-known anthropologist Darcy Ribeiro (now

deceased) formulated a substitute for the Bill 245 of 1950 (Ribeiro, 1951; Menezes, 2000). The

formulation of this substitute by Darcy Ribeiro coincided with the time when the proposal for

creation of the Xingu Park was being formulated. The elaboration and presentation of this bill to

the government, with the subsequent presentation of the proposal for creation of the Xingu Park,

were instrumental in the history of development of indigenous territorial rights in Brazil. The

main innovation concerning indigenous rights was the idea of cultural, social and physical

reproduction, in contrast with the idea of assimilation. Also, indigenous lands would guarantee

eternal usufruct rights to the indigenous communities under a common property regime, as a

collective patrimony of the group (Ribeiro, 1951).

The creation of the Xingu Park was a source of conflict and debates, given its ambiguous

legal characteristics. Until 1952, the date of the first presentation to the Brazilian Congress of the

bill or decree which would create the Park ("Anteprojeto de Lei de 1952"), there was no legal

category in Brazilian law that would have characteristics of both a protected area (national park)

and an indigenous reserve. The president of the SPI at that time, Jose Maria da Gama Malcher,

based his argument for the category of indigenous areas of large proportions on the north

American "Wheeller-Howard" law (1934), which foresaw the creation of indigenous forestry

reserves that would preserve indigenous habitat and the environment as a whole (Menezes,

2000). Therefore, the SPI wanted to install the first experience of creation of an indigenous park

in Brazil. This would provide for the preservation of Indians and nature in a virgin or primary

state, in contrast to the mentality of assimilation of the Indians by the surrounding society in

development.









In reality, the approval of the project for the creation of the "Xingu National Park" would

only happen eleven years later in 1961 Official Decree 50.455 of 14 of April of 1961 after

many reviews and conflicts, and with a significant reduction of the original area proposed to be

encompassed by the Park. Its area at that time was of nearly 22,000 square kilometers or

2,200,000 hectares (Menezes, 2000). In the original project, most of the region encompassing the

Xingu headwaters' tributaries was inside the area of the Park, which was subsequently left out of

the official Park's limits. This fact has had numerous cultural and environmental consequences

for the sustainability of the Park, which we will discuss in the next sections. In contrast to the

creation of the Xingu Park, the Law Project of Darcy Ribeiro was never implemented as such.

The formal regulation of the category "indigenous land" in Brazilian law advanced only in 1973

with the first "Indian Statute," which was reviewed and modified by the Constitution of 1988

(ISA, 2007a).

'Xingu has Everything an Indian Could Wish': the Transfer to the Xingu Park

The Villas B8as had the mission of building a landing strip in the Rio dos Peixes region,

allowing for the removal of two Kaiabi from Rio dos Peixes who had a rare skin disease (Jorge

Lobo mycosis), in order to be studied by Noel Nutels, the medical doctor of the Roncador-Xingu

Expedition (Villas B8as and Villas B8as, 1989). Prepori and the others spent many days

travelling on foot, guiding the Villas B8as from their campsite at Teles Pires to the region where

the Tatuy Kaiabi lived. They established another campsite to attract the Tatuy Kaiabi, who were

very wary of the visitors, but got less nervous with the presence of their Teles Pires relatives.

According to Mairaw6 Kaiabi's testimony (1981), after the contact with the Tatuy Kaiabi

at Rio dos Peixes, Prepori and others went back to the Xingu with Claudio and Orlando on foot

and by river through the recently opened trail, spending one or two years in the Xingu. The upper









Xingu groups such as Kamayura, Kalapalo, Kuikuro and others had already been contacted and

"pacified" by the Villas B8as at that time.

The situation of the Kaiabi at the moment of their encounter with the Villas B8as was

precarious: lack of assistance by the SPI; several deaths provoked by epidemics (mainly

measles); conflicts with rubber tappers and gold miners; and plans for colonisation of the

territory by agricultural migrants (Grunberg, 2004). There were at least four big measles

epidemics in the Teles Pires region: 1927, 1932, 1943 and 1965. Aturi reported the death of

nearly 130 people in one of these epidemics (Ferreira, 1992; Lea, 1997).

The Villas-B8as brothers strongly encouraged the relocation of both Teles Pires and

Tatuy Kaiabi groups to the upper Xingu region, where there were abundant land and fish,

providing, in the words of Claudio Villas B8as 'everything that an Indian could wish for'

(Menezes 2000:288).

The leadership of Prepori Kaiabi on one hand and the strong personality of Claudio

Villas-B8as on the other, established the organizational bases for the transfer of the group. In the

lower Teles Pires region, Piu'ni (now deceased) acted as an intermediary between the Villas-

B8as and the Teles Pires group, encouraging them to move to the new land (Rodrigues, 1994).

The majority of the Kaiabi "accepted" to move to the new area, but some refused to go and

remained in the ancestral territory (Grunberg, 2004). Some went following their relatives; others,

such as Xupe, went because they liked to move around and travel. Most Kaiabi families

relocated to Xingu between 1950 and 1966 (Grunberg 2004). After travelling by canoe and on

foot for over two months, and covering almost 500 km (Silva et al., 2002), the first and largest

group coming from Teles Pires (nearly forty people) established a campsite in the Arraias river


8 The last great Kaiabi shaman to die in 2002.









in 1955, in what is now the northwestern portion of the Park (see Figure 3-2; Grunberg 2004;

Menezes 2000). During these travels, they would suffer from hunger and physical exhaustion.

They would take a supply of cassava flour and live basically from hunting, collecting fruits, and

extracting wild honey.

From 1961 to 1962, the majority of Kaiabi from the upper and middle Teles Pires was

already living in the Xingu (Grunberg, 2004). The last group, consisting of thirteen people from

the lower Teles Pires and thirty-one from Rio dos Peixes, arrived by plane in 1966 in the so-

called "Kaiabi Operation," accompanied by Georg Grunberg, an Austrian anthropologist who

worked with the Kaiabi at the Rio dos Peixes village from 1965 to 1966. Xupe told me that

when the plane arrived, people told him: "You can board" (his son Aturi, nowadays chief of

Tuiarare village, was around ten years old). He thought: "Oh, no, this time I'm going to fall."

Mairaw6 Kaiabi, a prominent Kaiabi political leader, was five years old when he came to

Xingu on foot with his parents, along with the group that established the campsite at the Arraias

River. According to him, in the beginning they had precious help from the Yudja people (also

called Juruna), who provided them shelter and food until they were able to produce crops and

construct the village. He told the story: "after the village was built, my father told my older

brother: go get your uncle [Sirawe's father] and your cousin [Jurumuk, deceased] to move here

with us. There was an 80 Km road near Marceldndia municipality, linking the upper Teles Pires

to the Manitsaua-micu river, which was opened by the Villas-B8as. After three months, they

arrived [Myauapan Sirawe's father, Jururmuk, Pi'u and others]. They came in a better situation

because the things were already there, the village was ready. They also brought seeds, peanuts

and other things, so it was easier. We lived there for five to six years at the mouth of the Arraias.

Then, in 1961 Claudio and Orlando came to the village to visit us. They told us that they already









had a project to delimit the Xingu Park area and that it would not encompass all of our village's

area. The limit would cut the Arraias River. Part of the area in which we used to get materials

such as taquari, honey and uruyp (arumd) was going to be left out of the Park's limits. There

were also some fruits there such as akusikanafu, akusityrywa, ywapiruru, kwanu'ywa, Brazil nut.

Today, that area is part of the big Ibicaba ranch. We used to go as far as Marcelandia to collect

Brazil nuts; those Brazil nut stands were all named by the Kaiabi. Then, the Villas-B8as asked us

to leave the place; they did not want people living close to the border. In 1962, we left there to

live here in Diauarum Post. There were already some 'whites' living here since 1945."

The view of Xingu as a land of fortune was not only portrayed in terms of natural

richness, but also, and no less important, in relation to access to industrialized goods. According

to different testimonies, Prepori and other Indians who went to Xingu beforehand in order to

check out the place told the Kaiabi in Rio dos Peixes and Teles Pires that in Xingu they would

get all they needed as "gifts." They would not even need to work to get clothes, pans, hooks,

gunpowder, machetes and other highly appreciated industrialized products.

Masi'a, one of the eldest Kaiabi still alive, who is the chief of one of the bigger families

in Tuiarare village, told me that before the transfer, he used to work for the rubber tappers and

also collect Brazil nuts for sale. He was not thinking about moving to another place. But Prepori

was always inviting people to come to the Xingu, saying that it was a good place for them. He

was not suffering there in Teles Pires, but Prepori and Claudio invited him to come. He told me a

somewhat sad but funny anecdote about his travel to the Xingu: before the trip, Prepori told them

to throw their shoes away, because there were a lot of shoes in the Xingu; they would get shoes

for free. Therefore, they threw their shoes away before travelling on foot to the Xingu, during

which they sorely missed their shoes, arriving with wounded feet.









Jywatu (also known as chief Atu) is the current chief of Kururuzinho village at Kayabi

Indigenous Land in Teles Pires River. His father, Kum, was one of few Kaiabi who refused to be

transferred to the Xingu, staying at Teles Pires until his death. According to Jywatu's testimony,

his father did not believe that the whites would give everything to the Kaiabi at Xingu. Jywatu

told me that the "tapayin" (non-indigenous in Kaiabi language) arrived there cheating them:

"Man, there [at Xingu] things are given to us for free: firearms, clothes, aluminium pans, etc.

You are working here because you want to. Better you go there, there things are given away."

My father knew that the "tapayin" lied; he said: "No, this is only to tame us such as they used to

do in SPI times....Do you know why they (the Villas-B8as) want to take the Indians out of here?

Because they are giving this land to their relatives. That's why they are taking us out. But I'm

not leaving here. I prefer to die here, I prefer that the white kills me right here. I know how the

whites cheat us. They are telling the whites to take us off of our land in order to pass this area on

to their relatives." Jywatu said that at the time of the transfer, everything was fine with them. The

rubber tappers used to abuse their women, but this happened only in the beginning; later they

stopped with these abuses. When they found out about the transfer to the Xingu, they had

everything. "Regaties [merchants who travelled in boats and used to sell or trade goods with the

indians and rubber tappers] used to pass by here by boat. They exchanged goods for balls of

rubber. My father used to have things; nowadays we don't have anything." Jywatu told me about

Fernandes, an Apiaka married to a Kaiabi who lives with his family in a village close to

Kururuzinho in Teles Pires: "Fernandes left all of his belongings to move to the Xingu: stereo

system, sewing machine, pans. Because there, things were given away. When he returned here,

he did not have anything."









Xupe, an elder in Tuiarare village, lost his parents early and was partly raised by the chief

of a rubber tapping estate (seringal) in Teles Pires. He told me how he liked the seringal's chief,

who used to give gifts to him and treated him well. He used to work tapping rubber and

exchanging it in the rubber estate store, which according to him had plenty of goods. When he

arrived in Xingu, he found it weird: "there was nothing (to buy); there was nothing ... there

were only gifts, and the people got used to not working. I did not work anymore."

This idea of not having to work anymore to get non-indigenous industrialized

merchandise can be curiously compared to a traditional Kaiabi myth, which talks about how

work was done in the mythical past: according to the myth (told by Tymd Kaiabi in Ferreira,

1994), in the past the Kaiabi did not need to work to get most things done. The machete would

work alone to prepare the garden plots for planting, the tucumd (black palm fruit) would be

prepared by itself to produce necklaces, and so on. Tuiarare, the creator of the Kaiabi people, had

two sons at that time. One was good and diligent; the other, named the Moon (Jay in Kaiabi,

masculine) was curious and stubborn, a bungler. The stubborn son always wanted to perform the

work that the objects would do by themselves. After his many mistakes, finally the objects (or

their spirits) would not perform the work of the humans anymore, who since that time have had

to work hard to get things done.

According to Mairaw6 Kaiabi, one of the main difficulties in Xingu Park after the

transfer was transportation of goods, and the amount, which wasn't enough for everybody:

"everything was in small amounts: salt, fishing line, soap. We had to share that small portion. A

fishing line of 100 meters had to be divided in four pieces. Soap bars had to be cut in the middle.

It was very hard. Salt for example, was /2 kg for each. The chiefs would get a little bit more, but









the ordinary people would get almost nothing. The first time they gave more things; they said

there would be plenty."

Conclusion

The Kaiabi occupied a large portion of land in the north-western portion of Mato Grosso

State. However, it is impossible to draw a line marking the exact limits of their homeland,

because those limits are very fluid and subject to change. First, natural resources needed for their

cultural and physical reproduction were scattered in a landscape in which, sometimes, they

would travel days to collect. Second, warfare and contacts with other indigenous peoples had

caused changes in the establishment of villages and temporary campsites.

The contact with Brazilians, initially explorers, then rubber tappers, followed by SPI

officers and finally culminating with the Villas B8as brothers who took them to the Xingu, also

caused various processes of geographical change and re-territorialisation (Albert, 2000; Little,

2001). These contacts affected the Kaiabi in different ways, from the procurement and

dependence on industrialized tools to the work in rubber tapping, in which they gradually learned

that they could not trust "the whites," in spite of enjoying what they had to offer.

Brazilian government' polices towards indigenous peoples in the 1950s and 1960s were

contradictory and based on the idea of their assimilation by the Brazilian society. This had

consequences in the history of creation of indigenous lands in the country, which in the

beginning were small pieces of land capable of providing the minimum conditions for the

physical reproduction of the groups. In this context, the creation of the Xingu as a park where

Indians and nature would be protected was an exception, and a relative victory due to the effort

of Villas B8as brothers and other collaborators. This happened in the middle of a "march to the

west," where government plans were to provide the occupation of the imagined big emptiness of

Amazonian region as fast as possible. The idealization of indigenous peoples living a pure life in









harmony with nature has survived in Brazilian and international collective imaginary until

nowadays (Menezes, 2000), and has had political and cultural consequences.

The transfer of the Kaiabi to Xingu park was not consensual nor an easy step to take by

the Kaiabi. Deluded with the promise of"a land without evil," where everything was given as a

gift, many Kaiabi regretted having abandoned their traditional land in Rio dos Peixes and Teles

Pires, but the return was not easy or recommended by the Villas-B6as. Therefore, the only option

was to construct a new life in the new land, which belonged to other indigenous peoples. For the

ones who refused to go, the episode of the transfer also represented the beginning of another

stage in their lives, in which encounters with treasure hunters attracted by Amazonian natural

richness would continue in a faster and permanent way.


























































v P,' el. lll I Q lN
* ; K.?a-


Figure 3-1. Map showing the location of Kaiabi ancestral territory and the demographic situation
of the group between 1955 and 1966. Source: Dornstauder (1985), reprinted in
Grunberg (2004).


4 IM II rh


) S soea a Bvrr psE istL













* TA Ieae
A iP .r Iainafai
LD Teur Indfwu


SIy!l dKAPOT

KREMORcKAO

s ./


2l
insIlut


PIVT

I


Irpinu I f .... m -
METYK-IRE/Kayap6
0 2D Km0
e SWlombilntali, 2002 i
I -"

PV JARINA1r

/ ,Plaru1KayaD6
Ma la/Kiabi "' L- PI TUIUIU
STuba Tub./IYudlA PakisamnNudA
PIV MANI TQ. T PIV RIO PRETO
,.. -KuruiwKaiabi p _< ,,i.ai ,aja
', ':'-', ,.' c 'z.,Capiv lral labi "
Aldela Velha Pr ana
SobradinhdKaiabi Ital/Kalabi Pl DIUAR M
F IMuitaralKaieabi *~ Aldeia Vea AranhwKis4djl
4Tuarare/Kajali I isaja
KWQ~a/Kiabi 'A
PIV TYWAPE *A- Ng90SOofKis4dj6
i Ilha Grand '?Kaiabi.
Tr5 rlnm1nau/Kaiabi i Barrarc Kaiabi
M'ygulkpeng PIPAVURU
Terra F retafrrumrai ...
Boa EsperlancTmrur i Mor MenA~anliuri Ngo"e.isdj6
Wur6 i PIV WAWI\KiAdj
Steinen A .NahuKua
ERRA NOVAS IpavuKarnalur6 A Kalapalo
War '. A i A
waurd A R .Matlpu
IV RONURO pI LEONARDO
Aldeia Velha Yawallapiti
A Kuikuro
A.
Aweli Aukuri

Mehinaki ATnguro in
SA PIVTANGURO
Kunu./kalapaIo

r 1PIV KULUENE

PIV BATOVI PIV KURISEVO .


i
a
1~7W
,-J L"^


ikl I
-I I


I

i";
fJ
at


Figure 3-2. Map of Xingu Park and location of Arraias River close to the Tywape Vigilance Post
(PIV) in the northwestern region. Source: ISA (2008 a).


7.


t









CHAPTER 4
THE KAIABI AFTER THE DIASPORA: ADAPTATION AND LAND STRUGGLES IN NEW
TERRITORIAL CONFIGURATIONS

Introduction

The history of the process of securing rights over land among Amazonian indigenous

peoples has had a direct effect on the adaptation of these societies to highly politicized and

contested landscapes (Hierro and Surrales, 2005; Albert, 2005). These were formally turned into

indigenous lands through the complex and sometimes exhausting process of land demarcation

in Brazil. To understand how indigenous peoples became empowered to secure their land rights

and at the same time provide for the conservation of such a significant portion of the Brazilian

Amazon, we have to go back to the 1970s, when indigenous social movements began to mobilize

and get organized in Brazil (Albert, 2000). Since the Indian Statute was approved in 1973, the

formal recognition of indigenous lands has had to follow certain administrative procedures, in

accordance with Article 19 of the Statute. These procedures stipulate the stages of the long

demarcation process. They are regulated by an Executive Decree and, over the years, have gone

through many alterations (ISA, 2008 c).

It was through indigenous peoples' struggles, supported by indigenists, activists, NGOs

and the Catholic Church, that the indigenous movement grew, got stronger and was able to

influence the Constitution of 1988 in their favor, which recognized their rights over land and to

their cultural identity (Ramos, 1998). In the Constitution of 1988, indigenous peoples were given

both the right to be culturally different, as well as the right to their lands: "The indigenous

peoples are the first and natural owners of the land. Indigenous lands are those permanently

inhabited by them, used for their productive activities, and indispensable for the preservation of

natural resources needed for their welfare and for their physical and social reproduction in

accordance to their uses, habits and traditions." (ISA, 2008 c) The Brazilian Constitution of 1988









is considered a benchmark and an example for other Latin American countries in terms of

recognition of indigenous peoples' rights over their lands. Today, the Brazilian Amazon is home

to the greatest cultural and environmental diversity of the Americas, where 222 indigenous

groups live in 430 indigenous lands, totaling almost 21% of the region (see Figure 4-1; ISA,

2008 c). According to Nepstad et al. (2006), indigenous lands are currently the main barrier

against deforestation and environmental fragmentation. The three Kaiabi groups have been

subjected to all types of political conflicts related to land struggles, which have influenced their

contemporary territoriality, the environmental degradation or conservation of their lands and

their relationships with the State and between each other.

In this chapter, I present an account of what happened to the three Kaiabi groups after the

displacement in terms of demography, socio-economic configurations and processes of

encounter with new agents and adaptation to new places. I also present a description of the

process of land demarcation among the Rio dos Peixes and Teles Pires groups, as well as of land

claims and struggles in which the three groups are currently involved. Despite the apparent

isolation, we should not underestimate the capacity of recovery and of forging alliances among

the three groups. As we are going to discuss in this chapter, kinship and alliances to advance land

claim struggles have re-united Kaiabi groups in new social, political, and territorial contexts.

The Ones who Left: The Adaptation to Xingu Park after the Transfer

The different social and environmental conditions in Xingu Park presented many

challenges to the Kaiabi in the years after their transfer. Conscious of living in 'the land of other

people' and missing the access to natural resources and to manufactured goods previously

enjoyed through their contact with rubber tappers, some Kaiabi initially expressed the wish to

return to their homeland. According to Mairaw6, "a lot of people wanted to return to Tatuy and

Teles Pires; many of them tried. But the others would warn them: If you go, you are going to









suffer, there will be no food, and something might happen to you. Kupeap1 wanted to go, but his

family did not let him. Claudio said that the land was already taken by the whites in the Tatuy

region. He knew what the government wanted with the land. How would we know how to

impede this, create a reservation? For them, the more indians at Xingu Park, the better."

Fernando Apiaka, married to Rosinha Kaiabi (Morefwa) currently resident of Teles Pires

close to Kururuzinho village, was one of the few people who went to Xingu Park and later

returned to the original area. He told of his adventures in trying to get back to Teles Pires, in a

traumatic trip which lasted eight months. He, his wife, and her father, mother and brother-in-law,

went to Xingu by plane with the others from the Sdo Benedito landing strip. They stayed there

from May to August, then returned on foot and arrived back to the Teles Pires area in January of

the following year. He mentions that he was used to buying things, and in Xingu he couldn't. He

went to the BR-080 highway, which crosses the northern portion of the Park, to exchange things

with the whites, but the Villas-B6as met him there and argued with him, telling him that he

should not do this and that he was teaching the others bad habits. He did not like this, and

decided to come back to Teles Pires. On the way back, their food and clothing ran out. Rubber

tappers and cat hunters helped them with food. His mother-in-law was sick; he had to carry her

on his shoulders part of the way.

Mairaw6 acknowledges both the good and bad aspects of the Villas-B6as' work, the

creation of Xingu Park and the transfer of the Kaiabi to Xingu: "There are things that we,

indians, are learning with you, non-indians. The world is not perfect; things are just like this, all

twisted ... the Villas-B6as got the indians, took them out of their land, but they did a good job.

What is good about this? This land (Xingu) is big and there are 14 to 17 tribes which get along

1 Son of the chief Temeioni who came to Xingu from Rio dos Peixes by plane. Kupeap is one of the most
knowledgeable Kaiabi elders, one of the most experienced basketry weavers still alive. He lives in Capivara village.









well; there is no more warfare. The whites do not understand each other, but here the indigenous

peoples did it. It was a good thing also because the Kaiabi population is increasing here, instead

of diminishing. On the other hand, what went wrong? Taking indigenous groups from one land

and bringing them to another .. like I say, you killed the truth; you killed the roots of the

people... For me, this is a crime, taking you out of your land; your history is there."

When the Kaiabi arrived at Xingu, they met the truly "xinguano" groups, which still

used (until now) to walk around naked, painted and adorned, in contrast with the "cabloclized2"

lifestyle of the Kaiabi, after many years of intense contact with the rubber tappers and other non-

indigenous individuals. The Kaiabi and the other groups who were transferred to Xingu are

considered outsiders by the upper Xingu tribes (see section on Park's description for a more

detailed account of upper Xingu groups), which have lived in the region for thousands of years

(Franchetto and Heckenberger, 2001). At the time of the transfer and until the movement of the

Kaiabi towards the permanent occupation of the north portion of the Park, the relationship

between them and upper Xingu groups was less conflictive than it is nowadays. In the Park, we

can observe a clear tendency to narrow the separation between north and south, not only in

geographical but in political, educational and health service aspects (see Chapter 6). Not only the

Kaiabi, but all the outsider tribes, newcomers to the Xingu, are afraid of witchcraft and of the

power of the upper Xingu shamans. The Kaiabi avoid attending the well-known upper xinguano

cultural ceremonies such as the jawari (a mock battle) or the kwarup (a mortary ritual), and

rarely play soccer with upper-Xingu teams, even if they are invited (Oakdale, 1996).





2 Caboclized here means more acculturated, in the sense that the Kaiabi were already dressing in non-indigenous
clothing; men were able to speak Portuguese; and they used manufactured products due to their closer contact with
rubber tappers.









Despite the resistance, interactions and cultural, economic and political exchanges

between newcomers and xinguanos were unavoidable and happened in different domains. For

instance, the Kaiabi brought to the Xingu many crop varieties that were unknown to the upper

Xingu tribes (Silva, 2004), learning with them how to prepare a different type of cassava

pancake (from cassava starch) and other cassava-based beverages and dishes. Upper Xingu

representatives have participated in Kaiabi political meetings and in the creation and functioning

of ATIX (Associaco Terra Indigena Xingu, see next chapter for a detailed description), a

political organization run by the Kaiabi since 1995. The Kaiabi have also exchanged handicrafts

for industrialized goods with upper-Xingu representatives during meetings and visits to posts and

villages. As Suzanne Oakdale noted in the 1990s, and still happens in the present, in spite of

being considered more acculturatedd" in contrast with the more "culturally pure" upper-Xingu

groups, the Kaiabi exchange coconut shell necklaces, baskets and feathers with the upper

xinguanos for industrialized products such as beads, dresses, blankets, towels and other goods.

In the Xingu, Kaiabi traditional settlement patterns, involving a number of dispersed and

relatively fixed small family units, each grouped around a chief (iI /i kl), eventually gave way to

larger villages located along the main waterways (Oakdale 1996). Some of these villages were,

and still are, located on old Suya3 villages, which used to occupy the northern portion of the Park

before the Kaiabi. The construction of bigger villages, in which several families with no close

kinship links would live together, was encouraged by the Park's administration in order to

improve access and services. This has, in some cases, resulted in social friction and conflicts, as

well as in the villages becoming increasingly fixed in space, and the people increasingly



3 The Suya are a G6 speaking group which has inhabited the Xingu and Suih-Miqu rivers' region for more than
1,000 years (Seeger, 1981). They have interacted with the upper Xingu groups, from which they have assimilated
many cultural traditions.









sedentary (Senra et al. 2004). The access to better health assistance, provided by the

EPM/UNIFESP (Escola Paulista de Medicina/Universidade Federal de Sao Paulo), combined

with social changes faced after the transfer, have caused significant demographic growth among

the Kaiabi (Villas-B8as, 2005; Pagliaro, 2005, see section on demography on Chapter 6).

Coupled with population growth, the process of village sedentarization has contributed to the

over- exploitation of fertile soils and to growing pressure on natural resources of cultural and

economic importance (Athayde et al. 2002; Senra et al. 2004).

Since I began to work in the Park in 1997, I have noticed that some elderly chiefs have

been moving away from the bigger villages to small villages they call "ranches," in which they

live with their nuclear and sometimes extended families. Nevertheless, some of them still keep

strong social and economic relationships with their relatives in the bigger villages. Kaiabi

patterns of occupation of space are dynamic, as they used to be in the past, before their transfer

to the new land. The difference is that they are now circumscribed to the Park's area, which

limits their mobility and trekking activities, as before the transfer they were used to often

changing village locations and performing long excursions to gather natural resources and

explore new territories (Lea, 1997; Grinberg, 2004). Even being geographically circumscribed to

the northern area of the Xingu Park, the Kaiabi still show a notable capacity to move nowadays.

The social landscape is quite dynamic, and can change significantly from one year to the other.

The opening of new smaller villages or "ranches" as they are called is also related to social

friction and conflicts between different families. In bigger villages, gossip, jealousy, small thefts,

adultery, rape, alcohol consumption and other social misbehaviours are more prone to happen

and more difficult to control than in smaller villages, where social control is greater.









Some elders mention that in Xingu they have lost their freedom: isolation and the

difficulty of traveling to nearby cities have meant the development of a dependency on chiefs

and relatives who possess greater mobility and flexibility. This situation has twisted social order

and hierarchic relations among the Xingu Kaiabi in different ways (Oakdale, 1996). For instance,

the power of the elders, who used to be chiefs or "wyryat" (male patriarchs), has somehow been

transferred to younger leaders, most of them paid officers, who travel to towns and bring

industrialized goods back to the community. This re-orientation of social organization and power

as a function of the possibility of accessing and distributing industrialized goods has also been

observed among upper Xingu groups and other Amazonian peoples (Fisher, 2000; Heckenberger,

2005).

The Ones who Remained: The Adaptation of Rio dos Peixes and Teles Pires Groups and
Processes of Land Demarcation

There are few published accounts of the situation of the other two Kaiabi groups who

remained in Rio dos Peixes and Teles Pires areas after the successive transfer episodes to Xingu,

which lasted nearly twenty years (see Figure 4-2 for the current geographical location of the

three Kaiabi groups). Among those, it is worth mentioning that anthropological reports done for

FUNAI as a result of official land claim processes are important documents which serve as

sources of historical information. For the Rio dos Peixes Kaiabi, the most relevant sources of

information describing their situation after the transfer are Wenzel (1983), Melia (1993) and

Senra (2002), who wrote an anthropological report for the process of claiming the Bateldo land.

More recently, Silva et al. (2000) prepared a report about a trip by me and some colleagues to the

Rio dos Peixes village in 1999. For the Teles Pires Kaiabi, the most comprehensive document

describing their situation after the transfer is the anthropological report that Patricia de

Mendonca Rodrigues (1994) organized for the FUNAI working group (GT) created for the









enlargement of the Kayabi land in Para. The following description is partly based on those

documents and partly derived from interviews and testimonies collected during my fieldwork in

both areas.

Rio dos Peixes

At Rio dos Peixes, on the occasion of the transfer to the Xingu, the community organized

a meeting and some seven families decided to stay behind. According to the testimony of Simdo

Kaiabi, one of the elders at Rio dos Peixes and the most experienced basketry weaver, people

who refused to go were the families of Tafut (elder, alive), Luis Pedro (elder, alive), Chico

(passed away, the first cacique or chief after the transfer), Joaquim (Tevit, elder, alive, living at

Xingu now), Gilberto (passed away), Kwasiari (passed away) and him.

In the first years after the last official transfer of the Rio dos Peixes Kaiabi to Xingu,

there was no contact with Xingu relatives. Mairawe told me that after the administration of

Xingu Park changed from the Villas-B8as brothers to Olympio Serra around 1974, the Xingu

Kaiabi communicated with Olympio about their lack of contact with their relatives in Rio dos

Peixes and their will to visit them. Finally, in 1978 (twelve years after the last transfer), the first

expedition to visit the Rio dos Peixes village was organized by the Xingu Kaiabi. A group of five

men from Xingu, including Mairawe, left Diauarum Post by plane and arrived at Juara. After

flying many hours trying to find the Rio dos Peixes village's landing strip, they landed at a ranch

and asked the ranchers if they knew the Indians who supposedly lived nearby. They told them

that it was close, they could land there and the landing strip was good. They took off again but

did not find the village and ended up landing at another rancher's strip. They were told that they

could go by boat, and that 30 minutes from there they would find a lot of Indians. They did not

have a boat so they had to use an old canoe. It took them the whole day and part of the night to

go up the river until they reached the village.









At that time, Father Joao Dornstauder4 was still alive, and he was in the village. The next

day they had a big meeting with the community: it was very emotional, the elderly meeting each

other, crying. The community told the group of visitors what happened after they left, that Father

Joao was helping them to write documents to claim Batelao land's back (see next sections), and

that they wanted to move back to the Batelao area, where the main villages were located in the

past, before the transfer. Community members told the visiting group that there was a process

initiated with FUNAI to regain control over the traditional land and that they did not want to stay

in Rio dos Peixes Post, because it was Munduruku5 territory. They wanted to go back to Batelao

region, and asked the group from Xingu to help them. But at that time, there was no possibility to

communicate by radio, phone or internet. After the group left, they lost contact again until the

1980s, when they began to sparsely communicate with each other via radio.

During the 1980s and 1990s, with the improvement of existing roads and construction of

new ones, linking Juara with Marceldndia and Sinop, the transit of people between Xingu and

Rio dos Peixes increased. Capivara villagers were the most frequent travellers, and some families

who were living in Xingu moved back to Rio dos Peixes. Canisio Kaiabi is a good example of

this movement between the two areas: he was born in a village in Batelao, went to Xingu with

his uncle Temeioni by plane in 1966, and was the chief of Capivara village until 1998, when he

formally asket to be transferred back to Rio dos Peixes. He told me that he did not adapt to

Xingu. Before his permanent move back, he used to travel from Xingu to Rio dos Peixes almost

every year, to visit relatives and collect natural resources (especially Brazil nut, arumd used to

4 Missionary that worked with the Kaiabi in the 1960s, see Chapter 1.
5 The Munduruku people speak an isolated language in the Tupi stock. They live in different territories and regions
in the states of Para, Amazon River and Mato Grosso. They were one of the fiercest enemies of the Kaiabi, who
mention war battles against them in their ceremonial chants performed in Jowosi festivals. They share the land with
the Kaiabi in the Apiaki-Kaiabi land in Rio dos Peixes, numbering nearly 64 people. They also share the land with
Kaiabi and Apiaka peoples in the Kayabi land at Teles Pires, Para, numbering 244 (Travassos, 1984; Ramos, 2003).









make baskets, and wood for bows) to bring back to Xingu. In some of these travels, he would

bring his sons and daughters, who eventually ended up getting married at Rio dos Peixes and

staying there.

The Anchieta Mission (which later became OPAN, Operacgo Anchieta), is a Catholic

Church group which has worked with indigenous peoples in Mato Grosso since 1938. It was the

main institution to provide support for education, health and political assistance to the Rio dos

Peixes Kaiabi for many years after the transfer, until FUNAI's post was created and FUNASA

(Fundacgo Nacional de Saude) began to provide health services to indigenous peoples in Brazil.

Some village residents lived several years in the Utiariti mission in Cuiaba, eventually returning

to the village (Melia, 1993; Grunberg, 2004). Lourdes, the widow of chief Chico, told me that it

was difficult to survive after the transfer: "at that time there was no contact with anybody; we did

not know how they were doing. We survived here because of the 'Missdo Anchieta', Father Joao

and then other priests and nuns. We had Portuguese classes, school. My daughters understand

our language but they don't speak; maybe they are ashamed to speak their own language." There

were always some priests and nuns in the village, who were also responsible for running the

school and teaching Portuguese to the residents. As Melia (1993) emphasizes, the Catholic

presence in the village was (and still is) very strong, and marked by the efforts to promote socio-

economic development in the community, such as handicraft workshops for the women, and

mechanics and carpentry for the men. The CIMI (Conselho Indigenista Missionario) assumed the

missionary work of OPAN in the village, and keeps a priest and a nun living there until the

present.

Catarina Katumait Kaiabi is one of the older women in Tatuy, who also still knows how

to weave hammocks and conserve crop plants in her home garden. She was born in a small









tributary of the Batelao River. She said that in the past they used to live in the interior of the

forest, not along the course of the main rivers. She mentioned that the first outsiders who came to

exploit resources in the area after the rubber tappers left were fur hunters or trappers (in

Portuguese, "gateiros"), who used to hunt and trade wild feline pelts, including ocelots and

jaguars. A lot of Kaiabi men worked for the "gateiros" during the time of the transfer to Xingu,

both in Rio dos Peixes and in Teles Pires area. They would travel far accompanying the

"gateiros" who would pay them in goods and food. Sometimes, deadly conflicts occurred on

these hunting expeditions. Catarina told me that the "gateiros" used to give alcoholic beverages

to Kaiabi men, also bringing liquor to the village. Fisher (2000) mentions the involvement of the

Bakaj a Xikrin (a Kayap6 group) with fur trappers after the decline of rubber tapping in the

Amazonian region. According to him, SPI officers were also involved in the trapping activity, as

they used to work with rubber patrons.

According to Tafut Kaiabi, the eldest man and last shaman still alive in the village, born

in a village in Batelao region, after the transfer there was a lot of mahogany in the Rio dos Peixes

area, with great pressure from non-indigenous loggers to exploit it. They were lacking financial

resources to buy tools, gasoline, food, and clothing. Therefore, the chief "Chico" decided to sell

timber to get money for the community. This began some three years after the transfer to Xingu

and lasted nearly four years, until chief Chico died. The next chief, named Kutap (Gilberto)

continued selling timber to the whites, until there was not one mahogany tree standing. He

mentioned that the Apiaka also used to sell timber in the area they controlled. The loggers would

bring non-indigenous food from the city to the villagers, who got used to it and began to use the

money from the timber to buy more industrialized goods and food from the cities nearby.









With the installation of ranches and towns close to the indigenous reservation (such as the

city of Juara, founded in 1973), the contact between the Kaiabi from Rio dos Peixes and

Brazilian non-indigenous society (in this case represented largely by ranchers, hunters, loggers

and migrants) increased exponentially. The commerce of products and exchange for

industrialized goods also increased, as well as the temporary work of Kaiabi men in the ranches.

After Father Jodo Dornstauder died, in the beginning of the 1980s, Rio dos Peixes villagers faced

difficulties to control their land and resources. He used to be the main interlocutor between the

Kaiabi and the outside society, gathering resources, medication, gasoline and other types of

support to the area. He also helped them to control problems and to carry out land demarcation

and land claim processes.

During the 1970s, due to the increased occupation of the surroundings of the area shared

by the Kaiabi, the Apiaka6 and the Munduruku by settlers, migrants and ranchers, it was

necessary to promote the official demarcation of the reserve (Melia, 1993). As I understand it,

the Kaiabi had little influence in the decision on the borders of the area and on other details

concerning the planning and demarcation processes, which were conducted by the Diamantino

mission Catholicc mission) and FUNAI, respectively. Several interventions and changes in the

extension and borders of the area followed the first proposal approved by FUNAI in 1968

(Decree 63.368 of 10/08/1968, redefined by Decree 74.477 of 08/20/1974). The first demarcation

occurred in 1975, with 25,160 ha destined to the Apiaka and 47,450 ha to the Kaiabi (FUNAI,




6 The Apiaka are a Tupi-Guarani speaking group which originally was one of the Kaiabi enemy tribes. Numbering
around 2.700 people in the mid-nineteenth century, they were almost exterminated on the occasion of the contact
with the "whites" by slaughter, epidemics and also fights with other tribes. By 1912, their population was reduced to
some 32 people. Nowadays, they live in the northern part of Mato Grosso state, scattered along the Arinos, Juruena
and Teles Pires rivers. However, the majority of the population (nearly 92 in 1999) shares the Kaiabi-Apiach land
with the Kaiabi and Munduruku in Rio dos Peixes area (Wenzel, 1999).









1983). This first delimitation was not preceded by an official FUNAI anthropological study as is

nowadays required by law.

In 1978, the Kaiabi and Apiaka initiated another process for the revision of the

boundaries of the land, which, according to them, did not include the sacred land in the region of

the Rio dos Peixes waterfalls, rubber and Brazil nut stands (seringais and castanhais in

Portuguese), bamboo stands (used for their arrows), and special areas for agriculture, hunting and

fishing (FUNAI, 1983). Years later, the state government gave to CEMAT (Centrais Eletricas

Matogrossenses) the right to explore the hydroelectric potential in the region of the Rio dos

Peixes Falls, overlapping with the area already claimed by the Kaiabi and Apiaka as their sacred

sites. In 1982, a conflict arose between the Kaiabi and CEMAT, with the Indians interrupting the

construction of a road through Porto dos Gauchos municipality, which would cut their ancestral

lands to reach the CEMAT construction area (FUNAI, 1987). FUNAI's response was faster after

this incident, with the creation, in the same year, of a study group (GT 88.118/83) to carry out

the study of the land requested by the Indians. In 1984, after many negotiations between

indigenous leaders, CEMAT, FUNAI and the Anchieta Mission, a consensus was reached, in

which the Kaiabi would agree to the construction of the hydroelectric plant above the falls,

which would not destroy the sacred falls, and FUNAI agreed to the demarcation of 42,000 ha in

addition to the 67,610 ha already legalized (FUNAI, 1987). Finally, the Apiaka-Kaiabi area was

demarcated and officially registered by the Decree 394 of 21/06/1991 with 109,245 ha. The

hydroelectric plant was never built (for political and economic reasons).

The demarcation process and the subsequent changes in the Apiaka-Kaiabi land limits

has entailed (until now, see section on land claiming for an update of this conflict and legal

struggles) a lot of conflict with neighbour ranchers, mainly with ranchers residing at Porto dos









Gauchos7 and Tabapord municipalities, which are located some 100 km from the reserve. Big

ranchers sometimes are also part of or very influential in local government, augmenting the

pressure on indigenous lands and impeding advancement of land claim. Even though the Kaiabi

were successful in the first land claim process related to the inclusion of the falls in their land,

today they occupy only a minimal portion of what used to be the heart of their land.

Teles Pires

The Teles Pires or Para Kaiabi are a group of people descended from the Kaiabi who

used to occupy the middle Teles Pires valley and the Rio dos Peixes. With the pressure of rubber

tappers, settlers and later on of miners, they progressively dislocated their villages to the north or

lower Teles Pires, eventually reaching the border between Mato Grosso and Para states, where

they settled at the beginning of the twentieth century (Rodrigues, 1994). Based on testimonies of

elders such as Corone8 who nowadays lives in Kururuzinho village, we can imagine a lot of

movement in Teles Pires by the Kaiabi before and after the transfer to Xingu. For instance,

Coroner was born in a Rio dos Peixes tributary, moved with his family to a little mountain chain

in Rio dos Peixes region (Serra dos Caiabis), then moved again to the banks of the Rio dos

Peixes, building a village there. When he was close to age 18, he went to work for CONOMALI,

opening land for colonization. There he met the "gateiros" (trappers), who offered him and his

relatives munitions in exchange for work hunting wild animals. He traveled for months with his

father, brother and two non-indigenous (whites) hunting wild cats and giant otters. Eventually,

they had conflicts with the whites and abandoned them, establishing a village in the lower Teles



7 Porto dos Gauchos municipality was founded in 1955, by migrants coming from the state of Rio Grande do Sul in
the south of Brazil (commonly called "gauchos"), who founded CONOMALI, a governmental enterprise destined to
support the colonization of the northwestern region of Mato Grosso State (Grtinberg, 2004).
8 Masi'a is his indigenous name, but I will use Corone to distinguish from the Masi'a from Tuiarar6 village in
Xingu.









Pires, already in Para. According to Ryp, Corone's wife, her family came to Para heading down

the Teles Pires River. She said that the whites wanted to "tame" the Indians, hiring them first to

work in the rubber tapping activity, and then in pelt hunting. Jywatu (nickname Atu), current

chief of the Kururuzinho village, is the son of a former great chief (Manoel, Kaiabi named Kum)

who was influential in the choice of not moving to Xingu. He used to move a lot from village to

village. He told me that they used to travel looking for fruits and other natural resources. When

the rubber tappers arrived, they scattered along the Teles Pires. Some established villages, and

began to give names to places... in his words, "they began to make history."

At the time of the transfer to Xingu, there were many Kaiabi villages scattered in the

middle and lower Teles Pires sectors.9 Corone said that he was traveling with some whites,

hunting in Santa Rosa River, when he found out that Claudio Villas-B8as had already taken

some of his relatives to Xingu. His brother Tamanauu had already left, and nowadays is a leader

and "wyryat" (see Chapter 3) in Ilha Grande village at Xingu Park. Coroner said: "I'm not going,

what am I going to do there?" Like him, five other families who lived close to the current

Kururuzinho Post and village refused to leave: Chico (Kupeuwu, shaman, lives between Teles

Pires and Xingu nowadays), Andre (Pyrea'i, Valdir's father), Jywatu's father (Kum), Wyrakatu

(Selma's father, passed away) and Teme (elder, deceased).

In the beginning of the 1970s, the Sao Benedito mining company coming from Sao Paulo

installed its mining field in the Sao Benedito River, right on the site of an old Kaiabi village. The

company, besides the mining operations, also exploited a Brazil nut stand (castanhal) in the




9 Here, I adopted the geographical division proposed by Rodrigues (1994:5): the upper Teles Pires ranges from its
headwaters until the Rio Verde delta; the middle Teles Pires goes from Rio Verde delta to the Sete Quedas Falls in
Para, natural division between Mato Grosso and Pard states; and the lower Teles Pires ranges from the Sete Quedas
falls until its delta, in which it merges with the Juruena to form the great Tapaj6s.









region and employed indigenous as a labor force, also buying cassava flour from them. Many

Kaiabi, Munduruku and Apiaka Indians worked in the mining operations at that time.

Also in 1970, after the last group (close to 73 people) was transferred to Xingu by plane,

the few families remaining, totaling around 15 people, were scared of being forced to move to

Xingu and also afraid of the threats from Sao Benedito company officers. They hid in the woods

for approximately two months (Rodrigues, 1994). It seems that Sao Benedito officers wanted all

the Kaiabi to move to Xingu. The remaining group, commanded by Manoel Kaiabi, decided to

settle in one of their old villages, on the right Teles Pires riverbank, a little further down from

where the Kururuzinho Post and village is located nowadays. In the place where the Post was

created, there was only one old man living at that time, named Teme. After Fernando Apiaka

returned from Xingu, he also settled nearby with his family. In 1973, a FUNAI officer registered

31 people living around Manoel's village (Rodrigues, 1994). This group suffered with the lack of

assistance from the government, and with invasion of their territory by trappers, miners and then

loggers. The boat traders (regat6es) also used to exploit the Indians through a system of

exchange of indigenous products (flour, fish, Brazil nut, rubber) for industrialized goods.

In 1976, the linguist Rose Dobson visited the Teles Pires Kaiabi at their new location,

reporting that they were back in their old village sites, such as the Kururu River. She said that the

situation of the group was good, and that they were planting manioc to produce flour to sell to

the Sao Benedito mining company and also to the new ranch called "Santa Rosa" recently

installed on the Santa Rosa River. According to her, there was plenty of fish and game (Dobson,

1976 cited by Rodrigues, 1994). The Kururuzinho Post and village were created in 1987 by

FUNAI, in response to old demands of the Kaiabi who had moved to this location after the last

transfer of their relatives to Xingu. The old "Kaiabi Post," located down the Teles Pires close to









the confluence with the Juruena, was transformed into the Teles Pires Post, coming to serve the

Munduruku, who lived in that region.

In the 1980s, the Teles Pires valley was invaded by thousands of gold prospectors

("garimpeiros") coming from different parts of Brazil (Rodrigues, 1994). Hundreds of rafts and

dredges were installed along the Teles Pires and other Tapaj6s tributaries such as Peixoto de

Azevedo, occupying Kaiabi's old villages and sacred sites such as the Morro do Jabuti, a rock

formation rich in limestone. Remains of these rafts can still be seen until now when one travels

along the Teles Pires going to the Kururuzinho Post and village. Various little towns appeared

with the gold at that time. The gold hunters were responsible for great environmental degradation

in the Teles Pires River. The dredges caused damages to the river such as silting, pollution and

the death of fish and other animals through contamination with mercury used in gold mining

operations. The Kaiabi told me that the river was "as white as milk" and that they could not drink

the river water anymore; they had to get water to drink from small streams and from other Teles

Pires tributaries. At the time of the gold prospecting fever in the region, some Kaiabi worked for

gold mining patrons and others used to sell fish, cassava flour, banana and other agricultural

products to the gold miners. The massive gold mining operations in the middle and lower Teles

Pires valley lasted almost twenty years. In the beginning of the 1990s, with the decrease in gold

prices, most mining operations stagnated in the center-western region of Brazil, causing

population reduction and the return to agriculture and logging activities (Azevedo and Delgado,

2002). In 2000, researchers conducted studies on mercury contamination in fish in the Teles

Pires and Kururuzinho rivers, and also among the Kaiabi, finding a high concentration of the

metal in the fish, the base of Kaiabi diet (CIMI, 2001).









The fight by the Kaiabi to secure rights to land in the lower Teles Pires began in the early

1940s, when an SPI officer solicited the concession, for the Kaiabi, of an area between the Prata

stream and the Sdo Benedito River, both tributaries of the Teles Pires, totaling around 1,790,000

ha according to updated calculations presented by Rodrigues (1994:107). As expected, taking

into account the government's land demarcation polices in force at that time, the government

agreed to give to the indians a smaller area of lesser economic value, extending only 166,500 ha

(Decree n 251, 03/11/1945).

In 1975, FUNAI authorized the demarcation of the land conceded in 1945, but the

company responsible for the demarcation, in a hidden agreement with some corrupt FUNAI

officers, left outside of the borders the area occupied by the mining company. The area

demarcated in 1975 had only 117,246 ha, not 166,500 ha as it should have been according to the

Decree of 1945 (Rodrigues, 1994). The area was officially registered in 1982 (Decree 87.842 of

11/22/1982), but FUNAI officers immediately recognized that anthropological studies were

needed for a clearer and more adequate definition of its limits. In 1987, a FUNAI anthropologist

proposed the interdiction of the land (52,500 ha) left outside the borders during the demarcation

process, which happened in 1990 (Rodrigues, 1994). In 1993, FUNAI constituted a new GT (GT

1137, 11/12/1993) coordinated by anthropologist Patricia M. Rodrigues, to review and identify

new borders for the Kayabi land. In 1994, she proposed a new demarcation of the Kayabi land,

extending its area to nearly 1,400,000 ha, recognized as "area of traditional occupation" by the

Kaiabi, in accordance with the Brazilian Constitution of 1988. However, almost 400,000 ha of

those proposed 1,400,000 ha overlapped with an area that belonged to the Brazilian Ministry of

Defense, which contested the report (ISA, 2002). The solution found by FUNAI and the Ministry

of Justice was to remove this area from the boundary of the indigenous land. Therefore, in 2002,









the Kayabi Indigenous land finally was declared by the Brazilian government through Decree

1.149 of 10/03/2002 with 1,053,000 ha, giving permanent possession to the Kayabi, Apiaka and

Munduruku Indians (ISA, 2002). However, since then, the demarcation process has been delayed

by various contestations and interventions, as we are going to discuss in the next section on

current land claims.

Indigenous Land Demarcation in Brazil, and Kaiabi Land Claims

In this section, I will present a description of the history of the process of establishing

indigenous lands in Brazil, followed by an account of the situation of three current Kaiabi land

claim struggles: 1) TI Bateldo in Rio dos Peixes (TI -Terra Indigena or Indigenous Land); 2)

enlargement of TI Kayabi in Teles Pires; and 3) creation of a new indigenous land adjacent to the

Xingu Park area, in the Arraias river, for the Xingu Kaiabi.

According to ISA (2008 c), the process of legal recognition of indigenous lands by the

Brazilian government includes various stages or phases: 1) Studies of identification of the land,

in which FUNAI chooses an anthropologist to organize a working group (GT -Grupo de

Trabalho) which will carry out studies about the land in question and produce a report

identifying the indigenous land, within a given time limit. 2) Approval, in which the report

produced by the GT has to be formally approved by the president of FUNAI, who will authorize

its publication in the DOU (Diario Oficial da Unido, the Federal Government's official

publication) and in the Diario Oficial of the state where the future TI will be located. The

publication must also be displayed in the local municipality. 3) Resolution of disputes: from the

beginning of the procedures up to 90 days after the publication of the report in the DOU, any

interested parties, including states and municipalities, may manifest themself by presenting

FUNAI their arguments, along with all pertinent evidence, with the aim of demanding

indemnification or to demonstrate inconsistencies, gaps or errors in the report. FUNAI then has









60 days, in addition to the 90, to elaborate opinions on the arguments of all interested parties and

hand over the process to the Ministry of Justice. 4) Declaration of boundaries of the indigenous

land: the Minister of Justice will have 30 days to: (a) emit a directive declaring the boundaries of

the area and determining the beginning of its physical demarcation; or (b) prescribe judicial

proceedings to be carried out for 90 additional days, or (c) disapprove the identification,

publishing a decision substantiated upon paragraph one of Article 231 of the Constitution. 5)

Physical demarcation: once the boundaries of the area are declared, FUNAI promotes its physical

demarcation. At this stage, INCRA (Instituto Nacional de Colonizacgo e Reforma Agraria -

National Institute for Colonization and Agrarian Reform) will give priority to the resettlement of

any non-Indian occupants of the TI. 6) Ratification: finally, the demarcation procedure must be

submitted to the President of the Republic for homologation by decree. 7) Registration: the

indigenous land will then be registered, within a maximum of 30 days after the ratification, with

the notary of the correspondent judicial district and in the SPU (Servigo de Patrim6nio da Unido,

Service of Union Patrimony).

Decree 1775 of January 8, 1996, by the President of Brazil has caused a large impact in

the process of creation of indigenous lands. Also known as the "Jobim Decree" -it was written by

then Minister of Justice Nelson Jobim the main transformation in the process of creation of

indigenous lands brought about by this decree was that any person or entity has the right to

contest the report produced by the GT and previously approved by FUNAI: "presenting to the

Federal agency for indigenous affairs (FUNAI) reasons instructed with all pertinent proofs, such

as land titles, expert reports, witnesses testimonies, photographs and maps, aiming to claim

compensation or to demonstrate corruptions or errors, total or partial, in the report produced by

the GT" (Brasil, 1996). This legal instrument has caused a great deal of conflict and delay in the









creation of indigenous lands, since after any contestation, FUNAI has to send another expert to

the area to carry out more anthropological and technical studies in order for the Ministry of

Justice to approve or disapprove the creation of the land and the terms under which it would

occur. Therefore, some indigenous lands in Brazil, such as the Kayabi Land on Teles Pires River,

might wait more than ten years to be created, in an endless cycle of contestation and response, in

which the victims of a slow bureaucratic process often are the indigenous peoples and the

environment.

Despite the successful intervention in the 1970s and 1980s, of the indigenous grassroots

movements in official public polices recognizing their rights over lands, the process of

indigenous lands recognition and demarcation in Brazil nowadays is still a very contested and

difficult one. We can think about indigenous lands in the Amazon as a mosaic quilt in which

every piece is in a different stage concerning: 1) the formal phases of demarcation and 2) the

post-registration dilemma of territorial management and governance. Lands in both situations

suffer various types of pressures over natural resources and mineral riches that occur inside or in

their surroundings, pressures which are clearly greater over lands not yet demarcated and

registered. The list of threats and conflicts is not short: invasions by miners, fishers, hunters,

loggers and settlers; environmental degradation or pollution; deforestation; territorial

overlapping with other conservation units; conflicts over land with the Brazilian army;

construction of roads or hydroelectric dams; etc. (ISA, 2008 c). The three Kaiabi groups have

suffered, to a greater or lesser extent, many of these threats and impacts on the resources of both

their already recognized lands and their claimed ones.

Shortly after the process of transfer of the Kaiabi to Xingu Park, the remaining groups

began to worry about securing their rights to land. At that time, in the beginning of the 1970s, the









Brazilian government polices towards indigenous peoples were still aimed towards their

assimilation to the national society (see Chapter 3). The occupation and development of the

empty Amazon was the government's main development objective. There was an obvious clash

between settlers, ranchers and indigenous peoples' interests, and little doubt of whose interests

the government would prioritize (Menezes, 2000). As I mentioned before, the trend at that time

was the establishment of small indigenous lands to enable only the physical reproduction of

indigenous peoples, since they were considered human beings in the process of cultural

assimilation. The creation of the two existing legally recognized Kaiabi lands (TI Apiaka-Kaiabi

and TI Kayabi), with the exception of Xingu Park which is a sui generis case was influenced

by the polices and constraints of that time. Sacred places, sites of occurrence of important natural

resources, and the graveyards of Kaiabi ancestors, which for them were the most obvious proof

of immemorial occupation, were left outside the boundaries of current lands.

Various other indigenous groups in Brazilian Amazon have faced the same challenges as

the Kaiabi, and followed comparable paths in terms of adaptation, resistance and attempts to

return to ancestral territories after displacement events related to Brazilian governmental

development policies. Displacement events due mainly to road construction (such as the Panara

case) and colonization projects (Kaiabi, Rikbatsa, Xavante) happened mostly around the 1970s,

with attempts, sometimes successful, to return to or assert rights over ancestral lands, mostly in

the 1990s (Pasca, 2002). The Panara suffered with the construction of the "Cuiaba-Santarem"

road, which cut across their traditional territory. The violence of the contact resulted in the

extermination of two-thirds of their population from epidemics and massacres. In 1975, they

were transferred by the Funai to Xingu Indigenous Park. However, in contrast with the Kaiabi,

they never really adapted to the new cultural and environmental landscape of the Park. Around









1995, after twenty years exiled, with support of indigenists and NGOs, the Panara won back part

of their ancestral land, where they built a new village. Besides this victory, of regaining rights

over a portion of their traditional land, in 2000 they were the first indigenous group in Brazilian

history to win a lawsuit against the Brazilian government and FUNAI, including monetary

compensation for material and moral damages caused by the contact (ISA, 2004). The case of the

Panara and other indigenous cases in the Amazon support the argument that indigenous peoples

may be more innovative and resilient than previously thought, challenging deterministic ideas of

their assimilation into the bigger national society.

Batelao Indigenous Land

In Rio dos Peixes, on the occasion of the creation of the Tatuy Post by priest Jodo

Dornstauder in 1960, some families relocated there (initially 30 people) from their ancestral

occupation sites located in the region of the Batelao River (named "Yaruu" by the Kaiabi, which

means "big boat"), a Rio dos Peixes' tributary. The families which remained in the Batelao river,

approximately 130 km upriver, lived under the leadership of Chief Temeoni10, and were part of

the group which was transferred to Xingu Park in 1966 (Senra, 2002). Shortly after the families

left, the Batelao region began to be occupied by migrants ranchers and settlers coming from

the south of Brazil. The group which remained, suffering with the impact of the transfer, did not

have the power to fight against the occupation of their ancestral land. Senra (2002) compared the

actions of missionaries with those of the Villas-B8as brothers: both encouraged the relocation of

the Kaiabi to other sites, provoking a neither consensual nor voluntary abandonment of the land

where, as they say, their relatives were buried. Besides the dislocation to the Tatuy Post, the

practice of taking children and youth to study in the Utiariti mission was another action of the


10 Temeoni was a great patriarch in the Batelao area. Most members of his family live today in Capivara village in
Xingu Park. He died shortly after the transfer to Xingu Park.









missionaries that contributed to the group's disaggregation, cultural loss and territory

abandonment (Senra, op. cit.).

The proposal of the missionaries for the creation of the existing Apiaka-Kayabi land did

not include Kaiabi's old villages and sites of occurrence of important natural resources. Later on,

in the beginning of the 1980s, with the threat of the Rio dos Peixes hydroelectric plant, the

Kaiabi restarted the struggle to get back at least part of their nuclear territory in Batelao River,

where the main villages were located until the 1960s (Wenzel, 1983). The fight for this land has

had the important support of some Xingu Kaiabi who relocated from Xingu Park back to Rio dos

Peixes, including Canisio's family and Yurumuk's family, which relocated at the end of the

1990s. After our travel with a group of Kaiabi from Xingu in 1999, including part of Yurumuk's

family, in which we visited several villages and sites in the Batelao area, the wish to have that

land back was strongly regained amongst both the Xingu and the Rio dos Peixes Kaiabi.

Yurumuk decided to abandon Maraka village in Xingu Park and went back to Rio dos Peixes

with his family with the main objective of fighting for the land where his father was buried. The

expeditions and visits of the Kaiabi to the area were always reason for worries and conflicts with

ranchers of Tabapord municipality. The climax of these conflicts happened with the permanent

disappearance and possible assassination of Owit Kaiabi, Yurumuk's eldest son, in January of

2003 (Barreto, 2003). Owit was very active in the fight for the land, and participated in the GT

instituted in 2001 to study the area under claim.

In 1996, before the event described above, FUNAI carried out a preliminary study of the

region, in response to a request made by a big rancher (Fazenda Tapena, from the Macisa

company) installed on the left bank of the Batelao river. The anthropologist responsible for this

study emphasized the urgent need for the creation of an official GT to deal specifically with the









Kaiabi claims to the Batelao region (Dal Poz, 1996). At that time, the process of environmental

degradation of that area was already occurring at a fast pace, with the intrusion of loggers and the

opening of new areas for cattle ranching.

FUNAI did not respond to that suggestion. Therefore, in 1998, the Xingu Kaiabi,

supported by other indigenous leaders from Xingu Park, had a personal meeting with Sullivan

Silvestre (president of FUNAI at that time) at Diauarum Post. On the occasion, they handed in a

document asking for a response from FUNAI concerning their old request of promoting official

studies of the Batelao land as well as the creation of a new indigenous land adjacent to the Xingu

Park. Part of this document is transcribed below, translated from Senra (2002: 8): "We Kaiabi

were transferred some 40 years ago from the land where we used to live and where our relatives

are buried. At the time of the transfer, many of us did not want to abandon the land and still

today, the elders think about going back to the region of Teles Pires and Tatuy [Rio dos Peixes],

where we used to live. Since the transfer, the non-indigenous who occupied our lands are

destroying the forest and dirtying the rivers. Today, that region, in the Rio dos Peixes and Teles

Pires rivers, is almost completely spoiled and occupied by the whites. The Kaiabi, with the

support of Xingu leaders, handed in a document to FUNAI presidency on 03/10/97, proposing

actions to repair the loss of our territory. However, FUNAI's president did not give us any

answer. After that, in a meeting at PI Diauarum in July of 1998, the President promised to sign a

decree creating a working group [GT] to study the situation and analyze the creation of an area

for the Kaiabi neighboring Xingu Park and also to recover areas close to TI Apiaka-Kayabi in

Rio dos Peixes. Until now, the working group was not created and we did not receive any

answer."









On that occasion, FUNAI's president promised to the Kaiabi that both GTs would be

immediately created. Unfortunately, shortly after that FUNAI President Silvestre died in a plane

crash. The next President, Marcio Lacerda, also promised to create the GT, but left FUNAI

shortly thereafter without making good on his promises. Finally, under the presidency of Carlos

Mares, the GT for identification of Bateldo land was created in 2001, and coordinated by

anthropologist Klinton Senra. The studies were carried out under pressure due to conflicts with

ranchers and residents of the area, who tried to prevent the anthropologist, accompanied by some

Kaiabi and other technicians, from entering the area. In the identification report, Senra (2002)

indicated the need to demarcate a piece of land in the Bateldo river for the Kaiabi, recognizing

the immemorial occupation of the place and the location of sacred sites and strategic natural

resource clusters. Senra (2002), accompanied by Kaiabi representatives, carried out a brief

zoning of the land in question, reaching an agreement upon the extension and location of the

proposed area, based both on current occupation and the status of degradation, and also on the

presence of sacred sites and ecosystems which would enable the physical and cultural

reproduction of the group, as required by law (Figure 4-3). Therefore, the area proposed includes

three ecological zones: 1) rivers, lagoons and their margins; 2) upland forested areas (named ka'a

rete by the Kaiabi) and 3) mountain and savanna fragments (Senra, 2002).

Many people I interviewed in Rio dos Peixes in 2007, especially the elders, mentioned

the importance of securing that piece of land for the Kaiabi. First of all, there was the traditional

land of the Kaiabi, where old villages, sacred sites and cemeteries of ancestors were located.

Second, because of the existence of natural resources (plants and animals) which were vital for

Kaiabi social reproduction and occurred only in that region, such as clusters of"taquari"

(kamai'yp, Guadua sp), the prized bamboo used for arrows which occur in little mounds in the









savanna ecosystem; the clay used by the women to make pottery in the past; different types of

honey produced by stingless bees which only occur there; various fruits; and more abundance of

game, especially monkeys and wild pigs. Also, some Kaiabi mentioned that in Batelao there is

more variety and abundance of fish in comparison to the Rio dos Peixes village area, because of

the presence of lagoons. Some fish species do not pass over the Rio dos Peixes falls, and are

therefore concentrated in the upper river and its tributaries. These include certain species of

"amuata", "acara", "traira", and "pacu." Finally, the soil there is very fertile, with big patches of

black earth anthropogenic soils (kofet, capoeira), which are highly appreciated for agriculture.

Raimundo (Jewi Kwasiari) told me that Batelao is the true Kaiabi land, and where they

are living now is other people's land. In the past, they used to come to this region to get "siriva"

for their bows and collect Brazil nuts. Canisio compared the Batelao area with Xingu: "The first

thing is that in Batelao there is good soil, which produces our crops well, very different than in

Xingu. Because we say that there the soil is not good. But in Batelao, you can plant in any place

and it will yield good production. And also there are plants which still grow from the past, such

as papaya, potato, urucum; I saw a lot. The only thing we did not have there was Brazil nuts. So,

in the past we usually came down here to get Brazil nuts and bows."

The identification report for the creation of the TI Batelao, with 117,050 ha (Figure 4-3),

was approved by FUNAI only in 2008 (Decree 787 of 04/25/2008), and it is currently open for

contestations. Contestations came even before the approval of this report. I personally witnessed

the legal contestation process opened by Tabapord and Porto dos Gauchos ranchers against the

creation of TI Batelao, which was opened in 1999 and is still in course. In the document, the

ranchers accused the Kaiabi of having abandoned their own land and exchanged their territory

for the area they occupy in Xingu Park. Among other inconsistencies, the ranchers deny that the









area is the site of immemorial Kaiabi occupation, and ask the judiciary authority to impede or

prohibit the entrance of Kaiabi residents to the area under contestation.

In June of 2007 there was a big meeting in the Diauarum Post, where Kaiabi indigenous

leaders along with ISA lawyers, FUNAI representatives, and other practitioners and authorities,

discussed the content of the lawsuit produced by Tabapord and Porto dos Gauchos ranchers

against the creation of the TI Batelao. Since then, ISA's lawyers have supported FUNAI's

lawyers responding to this process, which is still under FUNAI's and the Brazilian judicial

system's analysis. Another anthropologist was sent to the area to carry out further studies, but the

process continued to stagnate. It appears that it will be a challenge for the Kaiabi to win this

struggle and get at least a small piece of their traditional territory back.

TI Kayabi in Teles Pires

The situation of TI Kayabi in Teles Pires is also problematic, though it has gone one step

further in comparison to the TI Batelao. However, since the approval of the report resulting from

the GT carried out by Patricia Rodrigues in 1994 and the declaration of the extension of the

existing land with 1,053,000 ha by the federal government in 2002, different judicial processes

which have been filed by local ranchers and logging companies owning lands inside the area in

contestation have really delayed its demarcation. The existing land is located in the Para state, on

the right bank of the Teles Pires River, which is the natural border of the demarcated land. The

claimed land is adjacent to the existing one, located in Mato Grosso state, on the left bank of the

Teles Pires (Figure 4-4). The whole land in question is part of a mosaic of different indigenous

lands and protected areas such as ecological stations, state parks, federal parks and national

forests.









In August of 2004, I was at Kururuzinho Post and village along with a group of Xingu

Kaiabi. 11 One day after we arrived, an officer from PPTAL/FUNAI (Projeto Integrado de

Protecgo as Populac6es e Terras Indigenas da Amaz6nia Legal) got there and organized a

meeting to set up the details of the demarcation process, which supposedly would begin shortly,

with the subsequent arrival of officers from FUNAI and the Federal Police. The Kaiabi would

actively participate in the demarcation process, and part of the funds destined to this work had

already been transferred to the Kawaip association.12 During the meeting, the officer received a

phone call through the radio: somebody told him that the demarcation was suspended because

ranchers had opened a judicial process contesting the report approved by FUNAI and asking for

additional studies to prove that the area in question was traditionally occupied by the Kaiabi. The

anthropologist and priest Eugenio Wenzel was hired to carry out subsequent anthropological

studies in order to respond to this process. In his report, Wenzel (2005) concluded that the area in

question can be considered to be of traditional use and occupancy by the Kaiabi, taking into

account that they have been circulating in that region of the lower Teles Pires for more than fifty

years. Moreover, the area is needed for the physical and cultural reproduction of the Kaiabi, as is

stated in the Federal Constitution of 1988, because there are strategic natural resources which

only occur on the "Mato Grosso side."

In addition to strategic natural resources, it is in the Mato Grosso portion that the Morro

do Jaboti (Tortoise Mound) is located,13 considered a sacred site by the Kaiabi and a valuable

site for miners' interests, since it is a big limestone reservoir. It is also the place of an old Kaiabi

11 We went there as part of a project for revitalization of basketry and textiles among the Xingu and the Teles Pires
groups (Kaiabi Araa project, mentioned in the Introduction).
12 The local political association of the Teles Pires Kaiabi; see Chapter 6 for more details.

13 The geographical location of this mound is S08058'10.2" and W57009'45.7". I personally visited the site in
2004, accompanied by some Kaiabi. On its top, there is a beautiful limestone cave, considered to be sacred by the
Kaiabi.









village, where some residents who went to Xingu used to live, such as Xupe (father of Aturi or

Jowosipep), nowadays resident at Tuiarare village. The main Teles Pires' tributaries existing in

the region are the Ximari, the Santa Rosa and the Piranha-Preta rivers, all located in the

contested area. In the Ximari, during the dry season, it is possible to collect tortoise and tortoise

eggs, foods very appreciated by the Kaiabi (Wenzel, 2005). In the Santa Rosa River, there is

more fish abundance in comparison to the Teles Pires, especially the fish named "matrinchd"

(Brycon lundii), highly appreciated by the Kaiabi and by sport fishing tourists who frequently

come to fish in the region. Concerning forest resources, the Kaiabi affirm that there are more

fruits and materials for handicrafts and construction on the Mato Grosso side. For instance, there

are bigger stands of Brazil nuts and copaibaa" (in Kaiabi Kupai 'yp, Copaifera sp, from which is

extracted medicinal oil) trees in Mato Grosso, and the "siriva" palm used for bows (see next

chapter on natural resources) occurs only there. Also, stands of"babacu" palm (in Kaiabi

Inatauu, Orbygniaphalerata), highly appreciated for its fruits and for roof thatching, only occur

in Mato Grosso.

Shortly after Wenzel presented his report to FUNAI, the group of ranchers did not accept

his findings and asked again for further investigation. They re-opened the lawsuit of contestation

in 2005, hiring another anthropologist from a consultancy company, who carried out some

additional studies and presented a counter-report in which he affirmed that the area under dispute

was never occupied by the Kaiabi. Therefore, this time FUNAI sent another anthropologist, Dr.

Marcos Paulo Schettino, from the federal Minsitry of Justice, to carry out further investigations

and respond to the counter-report produced by the anthropologist hired by the ranchers.14

Schettino (2005) organized a comprehensive report about the area and its situation, corroborating

14 I did not have access to this report, but its content was explained to me by personal communication from
Frederico Oliveira and by Kaiabi representatives.









the two previous studies carried out by Rodrigues (1994) and Wenzel (2005). According to him,

the area under dispute is of traditional occupancy by the Kaiabi, Apiaka and Munduruku Indians

since the beginning of the 20th century. The area is also needed for the physical and cultural

reproduction of the Kaiabi according to their habits and traditions, as is stated in the Federal

Constitution of 1988. Moreover, Schettino (2005) highlights that most of the land titles held by

ranchers and other non-indigenous occupants, authors of the lawsuit, are relatively recent, dating

from 1984, and that the effective possession of land, as well as logging and cattle raising

activities, began to occur during or after the conclusion of the studies done by Rodrigues (1994)

for FUNAI. He emphasized that in many cases the deforestation had occurred illegally, and that

the process of land demarcation had caused an enormous increase in deforestation (most often

practiced illegally) in recent years, between 2002 and 2005. He affirmed that: "the authors of the

lawsuit, under the threat of losing the land due to its demarcation as an indigenous land, were

driven more by the possibility of fast profit than properly of spontaneous use and occupation, as

a result of a natural interest in enjoying their properties...this type of occupation, many times

carried out disrespecting the law, has caused deep environmental impacts. It occurred initially in

forested areas considered intact until 1994 with primary vegetation and high biodiversity -

when the land was studied by FUNAI, as well as, in consequence, in headwaters of micro-

watersheds and their tributaries, constituting an environmental crime, unfortunately harming the

integrity of the indigenous territory still in the demarcation process." (Schettino, 2005:27)

The Kaiabi have not been passive in this dispute, having organized in the past ten years

innumerous meetings, protests, documents and visits to Brasilia to speak about the problem with

representatives from FUNAI and Ministry of Justice. The Kaiabi from Xingu, invited by Teles

Pires relatives, have also participated in the struggle over this land in different ways. Some of the









Xingu residents interviewed said that if the area were demarcated and registered for the Kaiabi,

they would move back there to ensure the protection of the land. The current President of

Kawaip Association is Tarawi Kaiabi, who moved to Teles Pires motivated by the wish to gain

back the area which had been occupied by his grandfather. In June of 2007, when I was at

Diauarum Post participating in Kaiabi meetings about land struggles, a FUNAI lawyer who was

also there announced that the area finally would be demarcated, and that the lawsuits had been

struck down. Just like a replay of the same sad movie, in June of 2008, when everything was

ready for the demarcation, another lawsuit process was triggered and the demarcation suspended

again. The inconstancy of FUNAI administration, with changes in Presidency every other month

or year, does not help to accelerate this already complex land demarcation process. The Kaiabi

are very unhappy with this situation and are not going to remain quiet.

Arraias River in Xingu Region

When the first groups of Kaiabi arrived in Xingu Park coming on foot from Rio dos

Peixes and Teles Pires, such as the family of Mairaw6, they established villages in the

northwestern portion of the Park, along the Arraias and the Manitsaua rivers. These groups lived

in and used that area for approximately ten years before moving closer to Diauarum Post to have

easier access to health services and other types of support. Later on, some Kaiabi families moved

to that region and established new villages, such as the Aipore (old Kururu), Jodo's15 "ranch"

and the Sobradinho village. The northwestern border of Xingu Park is one of the most threatened

due to the action of loggers and illegal fishing and hunting. Since ATIX began to develop

activities of patrolling the Park's borders, various vigilance or monitoring posts (PIV- Posto



15 JoAo Kaiabi is a shaman who used to live in the town of Marcelindia but moved back to the Park and opened a
new village, where he threatens people. We will talk more about his village and his curing ceremonies in Chapter 8,
when we discuss the Maraka healing festival where basketry and cotton are used.









Indigena de Vigilincia) have been established in strategic points around the Park's borders. One

of those is located on the Arraias River, the PI Tywape. From the beginning of the 1990s until

now, the Kaiabi have often had problems with invaders and loggers in that region.

In 1998, a group of fishermen was arrested by the Kaiabi close to the PIV Tywape and

brought to Diauarum post, where they were kept in captivity for a couple of days until the

authorities such as FUNAI and IBAMA came to Diauarum Post to have a meeting with Kaiabi

and other groups' leaders. As a result of negotiations deriving from this incident, ATIX and the

indians began to run the project for monitoring the park's borders, previously run by FUNAI. It

was on that occasion, during exhausting meetings held by the Kaiabi with other indigenous

groups' representatives, that the indians had the idea to claim a piece of land adjacent to the Park

in its northwestern portion. The creation of this land was justified by them for various reasons:

first, it would guarantee to the Kaiabi and to other peoples that live in the north portion of the

Park a reservoir of land and resources for their immediate and future use, taking into account that

the environment of that region is more similar to that of Kaiabi ancestral territory, in terms of

forest physiognomy and composition, compared to the area towards the south of the Park, where

most Kaiabi villages are located today. For instance, some species which do not occur or occur

sparsely in other park areas can be found in the forests of that region, such as the "arumd" used

in basketry and the "acai" palm fruit, both of great economic and cultural value for the Kaiabi.

Also, the good soils for agriculture (black earth, or "terras pretas") are being exhausted inside the

Park, which poses a threat to the future food security of the Kaiabi. Second, the creation of a

reserve there would reinforce the protection of that region which is more vulnerable to invasions

and environmental impacts than others. It would basically work as a buffer zone for the Park.

Third, the region is still relatively well conserved from the environmental standpoint, which









would also justify its special protection, given the alarming deforestation rates around Xingu

Park in particular and in Mato Grosso in general.

In 1998, the Kaiabi handed a document to FUNAI's President in which they laid out the

reasons for the establishment of a reserve on the Arraias river, together with other land claim

processes such as TI Bateldo-Rio dos Peixes and TI Kayabi-Teles Pires. In 2000, the new

FUNAI President Carlos Mares authorized a preliminary study of the area, which would inform

the decision on the creation or not of an official GT to identify the area in question. The studies

were carried out by the anthropologist Klinton Senra, accompanied by a group of Kaiabi, and a

report was delivered to FUNAI in 2001 (Senra, 2001). The main conclusion reached by this

preliminary study refers to the need to conduct a real GT to identify the area, which would be

justified by compliance with the principles established in the Federal Constitution of 1988

(mainly article 231) for the creation of indigenous lands. Primarily, there has been traditional use

of the land and natural resources by indigenous peoples such as the Kaiabi and Yudja at least

in the last fifty years. The law requires a traditional use of the land, which does not necessarily

mean an immemorial occupation, but the way that the resources and land are used such as

fishing, hunting, collecting and practicing agriculture (Senra, 2001). This area is relevant to

ensure the cultural and physical reproduction of indigenous peoples who were not originally

inhabitants of Xingu Park taking into account the occurrence of important forest resources,

game and fish. Another point brought up by Senra (2001) is that in the original project for the

creation of Xingu Park, the area to be destined for the Indians, was significantly bigger than the

land officially approved and demarcated. As indigenous peoples' populations grow, the pressure

over land and resources is increased. With the sedentarization of villages, the pressure over plant

and animal resources is augmented, as well as the environmental impacts and deforestation









carried out by ranchers, loggers and other resource users living and working in the surroundings

of the Park. Therefore, the extension of the Park's area through the creation of "satellite" smaller

areas in strategic sites around its borders would ultimately mean insurance, in medium and long

terms, for the maintenance of a balance between residents and resources (Senra, 2001). Today,

ten years after the first request and seven after the report, there is still no action taken by FUNAI

in response to this request. This request has also been used against the Kaiabi by ranchers from

the Tabapord region near the TI Batelao: they accuse the Kaiabi of having proposed to "trade" or

bargain their ancestral land in Batelao for this one claimed in the Arraias River, which is not

true.

Conclusion

While the Kaiabi who were transferred to Xingu Park were struggling to adapt to the new

socio-cultural and environmental conditions, the two other groups which remained in Tapaj6s'

watershed the Rio dos Peixes or Tatuy Kaiabi and the Teles Pires or Para Kaiabi were

struggling to survive and resist the occupation of their lands by non-indigenous invaders such as

loggers, miners, ranchers and hunters. The three groups began a new chapter in their history with

the transfer. The Xingu Kaiabi became more isolated from contact with non-indigenous society

than the other two groups. This was the policy implemented by the Villas-B8as brothers at that

time, and one which is followed still: "the more isolated the Indians remain, the better for the

preservation of their culture."

The Rio dos Peixes group suffered from lack of support, especially after the death of

priest Jodo Dornstauder. The Teles Pires group was the one that suffered most after they refused

to relocate, especially after the Sdo Benedito mining company established itself in the region

they used to occupy. Both groups that remained continued to work as assistants for non-

indigenous agents who came to the Amazon to exploit its natural richness: trappers, miners,









loggers. The arrangements for work were similar to those of rubber tapper patrons; therefore the

Kaiabi did not feel a big difference between the rubber boom and the new opportunities for

making money or goods in exchange for labour.

In contrast, the relocated people in the Xingu initially suffered from the lack of choice

and paid work, which would enable them to buy industrialized goods as they were used to

previously. The Xingu Kaiabi began to live under a highly paternalistic regime, getting goods as

gifts or exchanging handicrafts for goods with other indigenous groups or with non-indigenous

individuals. This has profound consequences in the way that the Xingu Kaiabi relate to work and

labor nowadays, and their eagerness for "gifts". A preliminary conclusion possible to be

envisioned at this time (still to be deepened as we progress in this dissertation) is that one of the

reasons which led the Kaiabi to keep and innovate their knowledge on handicrafts was the lack

of option in getting money or goods through labor, which they used to get when involved in

tapping rubber.

We can consider that none of the Kaiabi groups currently live in what was once their

traditional territory before the contact with non-indigenous Brazilians. The closest group are the

Rio dos Peixes Kaiabi, since the actual land includes a little portion of the traditional land, which

is located further up the river, beyond the Rio dos Peixes waterfall. The Teles Pires Kaiabi are

also living in an area to which they were pushed by the rubber tappers, settlers and ranchers,

located further down the Teles Pires river from the areas they occupied in the upper Teles Pires

on the occasion of contact with whites. Nevertheless, Kaiabi groups have constructed their

destinies and inscribed history in the new landscapes they occupied before and after the

displacement (Santos-Granero, 2005). In this regard, Schama (1995) considers that landscapes

are the result of human agency applied to natural settings over time. This idea has practical









implications in the demarcation of indigenous lands in Brazil, since the writing of history for

indigenous peoples might mean "traditional use" in the 1988 Constitution. To clarify this

metaphor, different interviewees mentioned the fact that after living in a specific place, they

would begin to give names to places and to groupings of resources (e.g. castanhais, Brazil nut

stands). Jywatu and Mairawe referred to the "making of history" in a given place over a given

amount of time. This "writing of history in the landscape" (Santos-Granero, 2005:186) or place

making, can be compared with the idea of traditional use, not necessarily immemorial, but one

that is alive in memory and oral tradition.

In the Kaiabi case, mobility and sedentarism seem to interplay through history, before

and after contact with westerners, where the concept of territoriality itself may be interpreted as

in constant movement (Athayde et al., 2009). The Kaiabi have configured new territories,

embedded in highly politicized, institutionalized and contested spaces such as Xingu Park

(Oakdale, 1996; Menezes, 2000). Land claiming processes have been very traumatic, long-term

and frustrating experiences among the Kaiabi. Ancestral cemeteries and sacred places are now

inside ranches and towns. On the other hand, these struggles have enabled new alliances,

strengthened local institutions, and helped build leadership among the three groups.

































Legenda
I Estados

Terras Indigenas
i51tuagao '*
SDeclarada
Confir-mada
Delirnmitada
| | Hoimologada
SI Regularizada n .d MinraduMEiAnblm

Figure 4-1. Location of indigenous lands in Brazil. In the red polygon, location of Xingu Park
(PIX), Apiaka-Kaiabi Land, TI Kayabi in Teles Pires and TI Bateldo (claimed).










Si--;^ i jca a II
S -' : -
I 'Pfl if Ir
MA 0 I, ** .
I .NE .... LJLr I.,1I 1 ,,- II l ',
.I

N r..0 L', i I ',.r i; '1 r t
-I *rji r T I -A"C r

N1FerTo in[ Lin, 11 Yilrj ,,.I- ]'il'i-i I- Ir-' tE 'i "
T-ir -' L- -. ."i A ,

S' ) IT :I IL
.i I ,n I '- .
-f 1 li-ie.!- :* I i* ,: :-..
Si`l L -. A -_

Ir .\...I.. 2 T~"9-' .Ti--- --- L .


r r- r
,l-. --.-.. .. !t" C T I "*.
I.J bJ.-4 I ..T- ,l-,r rll.h .i -


I Tl -"TI u I. [
..... .- .i u.. -......,,,ui-,
I .

T' i Tii Ir I
^ .. -r ;i r .'.. -., -, T n .t -. -I r,,_ ,


-.r- '' -"T ; *- .TI-.1,, i

.,L- l I- I- *' I .
"- .t T ". 1 ; '




Figure 4-2. Detail of the location of the three areas occupied by the Kaiabi in Mato Grosso: PI
Xingu, TI Apiaka-Kayabi and TI Kayabi, marked with red triangles. Source: Adapted
from ISA (2008a).
from ISA (2008a).

























-1, a,


.



-la- .*>


ci _,
.-


Lu.


-C-?


tc r-


fll -fl mc.


I r Pwcaa si~davic4

j ~i r 'iaI I tMLif
.M; t-L~ W4LtlttW
----- ix~-)uwe feiu BW i



:--1 sKsrr uf^ rrri^Aj
_, : ,aii~~iB"


.... .. C. a. I A-.
,f 7r -, i


- ,t"; a r~'kr

'Ir., -,4'Lup I- --:Sl~


Figure 4-3. Map of delimitation of TI Batelao in Rio dos Peixes region, showing the location of


old Kaiabi villages.

























143


*45


-n J


"*^a ;,i-


i~----L *II;QI

li':'Nrg.:ogl.


T4rt


iYlli
~U;i~h
----- I
aru~~-
--


.~i--"
-r*-8



















* i ; v : I I&


I


a


cr
-p.


~*, 4'.



9 C4
q I. */
4~"1 ?, ,,,
-^*n ^
I ^-'.


st-w CONVEN ID

C t --W- ~u ULYL
igasi n ne= ansm~ mum r

IF "aS *MinoHFwist:|Oftla jiCnmCTB
a A nunCUJ m
.PcIarnmben 0.maurs
ma--a. wwant m
a war WRyow

'ImrrrALA ~ e a)i~h~
samanuams.,..C -


9Z 'tf


/



%fl

9


r%
I






"%
x / .
-,, ; -"* ,i




-1,











N .
9 *. ,..,
9
!e "- ** r
'/ C 1"i






d. / 1~ d

1 ^
-' '"'"r

sp '

9' *


II IPTCIDa DA JVBTIC*
FIr*OAQAC MNACtrOAL GO INDID FUNAl
N P"TBQO A DC AI* Uuna. UaROiAr.0l OA


KA YAl DELITJTAqA*I

A PIIA i,.I frltr
JeVIACAeA-PAS* -PA I-

,If ;- .. I,, -. f -., r.


Figure 4-4. Map of delimitation of TI Kayabi in Teles Pires region.














144









CHAPTER 5
ENVIRONMENTAL CONTRASTS, ADAPTATION AFTER DISPLACEMENT, AND
NATURAL RESOURCE MANAGEMENT ACROSS THREE KAIABI LANDS

Introduction

Indigenous lands in Brazil show the lowest levels of deforestation in comparison with

other legally protected areas (COIAB and TNC, 2006). A recent study done by COIAB

(Coordination of Indigenous Organizations of Brazilian Amazon) and the NGO TNC (The

Nature Conservancy) reports that inside the Indigenous lands the deforestation rate is around

1.1% in comparison with 1.52% in federal wildlife protected areas (COIAB and TNC, 2006).

These data are important to ground the biocultural resilience concept, which states that biological

and cultural diversity are intrinsically correlated and co-evolve in history (Pretty et al., 2009).

Along with extractive reserves and other protected areas, indigenous lands constitute green

corridors that enable conservation at species, community, ecosystem and landscape levels.

Additionally, indigenous lands provide numerous environmental services to the broader

Brazilian society, such as clean water, climate stabilization and biodiversity protection.

The richness of natural resources present in indigenous land is coveted by many social

actors who participate in Amazonian occupation and development. Therefore, indigenous lands

are also viewed as a reserve of resources, that might be exploited as opportunities appear. The

exploitation of resources inside indigenous lands with a commercial purpose is prohibited by

law, but this has not prevented indigenous peoples from reaching agreements with non-

indigenous individuals for logging, mining, fishing and hunting in their lands.

After the struggle over land rights is solved, indigenous peoples need to be able to

manage and protect that territory and its resources. In theory, this should be supported by the

state, through FUNAI's actions and projects. In practice, government and FUNAI structures are

insufficient to carry out this difficult task. The sustainable management of indigenous lands in









Amazonia is a critical subject on the environmental conservation agenda (and lately, of the

climate change issue).

Among the Kaiabi, geographical displacement caused a disruption in the traditional

systems of managing natural resources, and also meant a loss in access to resources of cultural

and economical importance. From the environmental standpoint, the Xingu is significantly

different from the other two areas originally occupied by the Kaiabi. While in the Xingu the

Kaiabi were adapting to the different environment and to the lack of choice in terms of resource

availability, the two other groups in the traditional lands were subjected to different pressures

and exploitation of resources by external agents.

This chapter presents an overview of the environmental landscape of each land occupied

by the Kaiabi, including a description of their particular status and threats in terms of

environmental conservation. The main question I aim to answer here relates to the mechanisms

developed by the Xingu Kaiabi to adapt to the environmental constraints faced after the transfer.

I also present the results of a free listing exercise carried out with men and women across the

three groups, on selected plants of cultural and economic importance which occur in their

ancestral territories.

Environmental Contrasts between the Three Areas

The process of adaptation to Xingu has been constrained by the climatic, geomorphic and

ecological differences between Xingu Park and the areas traditionally occupied by the Kaiabi in

the Tapaj6s watershed. Both Xingu and Tapaj6s rivers flow from south to north into the

Amazon. The "dos Peixes" river has the totality of its course within Mato Grosso state. Its delta

is located in the Arinos river, a Juruena affluent, which is a Tapaj6s tributary. The relief is flatter

in the Xingu, a tertiary plain, in contrast with the older, hilly and rocky landscape of the ancestral

region.









According to Koppen's classification, the climate in the Xingu is tropical, including two

types: Am tropical monsoon climate with monsoon rains and Aw tropical or savanna climate,

with wet summers and dry winters. It shows a transition between the domains of the Amazon

rain forest and the Brazilian central plateau, which is covered with savannas, and has two

delimited seasons (Radam Brasil, 1981). The total annual precipitation oscillates between 2,000

and 2,750 mm (increasing from southeast to northwest) and the median annual temperatures

vary between 24 and 260C. In spite of the high relative humidity, from 80 to 85%, there are two

periods with distinct pluvial precipitation indices: the rainy season, which occurs during

spring/summer, concentrating more than 80% of the total rainfall, and the dry period, between

April and September, with low pluvial indexes, where one may register two consecutive months

without any rain (Radam Brasil, op. cit.). This seasonal variation has great influence over

indigenous peoples who live in the Xingu and whose agricultural, fishing, hunting and collecting

calendars as well as cultural festivals depend on the dry-wet binomial seasonality.

In the region comprised by Rio dos Peixes and Teles Pires areas in the Tapaj6s

watershed, the climate is classified as Am, hot and humid, with monsoon rains and a small dry

period observed during the winter. The annual pluvial rate is high, close to 2500-2750 mm

annually, with less intensity of rainfall between May and August. The relative air humidity is

around 85% and the medium temperature is 24C (Radam Brasil, 1980).

Xingu Park is located in an ecological transitional zone between savannas, semi-

deciduous forests and lowland tropical forests, in which a mosaic of various types of ecosystems

make up the landscape (Radam Brasil, 1981). Towards the south, there are the rivers which form

the Xingu, which make up a sub-basin including rivers such as the Von den Stein, Jatoba,

Ronuro, Batovi, Kurisevo and Kuluene, this last one being the main Xingu tributary, when it









meets the Batovi-Ronuro. To the south, in the upper Xingu area, savannas predominate, whereas

to the north, forest patches are bigger in extension, with the majority of the trees losing their

leaves during the dry season (Silva and Athayde, 2004). Plant species typical to the Amazonian

region do not occur, or occur sparsely, within the Park's limits, such as the rubber tree (Hevea

brasiliensis), mahogany (Swietenia macrophylla) and the Brazil nut tree (Bertholletia excelsa.

The extension of flooded riparian forests is greater in Xingu, while the river landscape lacks the

rocks and rapids of the Tapaj6s River. In contrast to the Xingu, the forest appearance, structure

and composition in the Tapaj6s are typically Amazonian. Xingu is poorer in resources of

economic importance for both extractivism and commercial timber exploitation (Radam Brasil

1980; 1981; Silva and Athayde 2004). At Rio dos Peixes and Teles Pires, savannas and semi-

deciduous forests occur in the form of contact with other forest formations, especially with the

Amazonian rainforest. The savannas have always been used by the Kaiabi for hunting and for

collection of a variety of fruits and specific resources such as the "taquari," a bamboo used to

make arrows (Guadua sp, which does not occur in Xingu Park), found only in rocky areas within

the small mountain range area denominated "Serra dos Caiabis" in the Rio dos Peixes region

(Silva and Athayde, 1999). In both Rio dos Peixes and Teles Pires areas we can find Amazonian

species of economic importance, such as Brazil nut, rubber, valuable wood species such as

mahogany (Swietenia macrophylla), fruits such as cocoa (Theobroma cacao) and a variety of

palms such as "acai" (Euterpeprecatoria), "pupunha-brava" (Bactris macana), "pataua"

(Oenocarpus bataua) and others.

The environmental contrasts between the areas are also reflected in the faunal

composition. According to the Kaiabi, in the ancestral area there is more quantity and diversity

of mammals and birds than in the Xingu, which means greater game availability. Some animal









species which do not occur in the Park are the "arara-canga" or little red macaw (Ara macao), as

well as other parrot and macaw species, used for the confection of feather headbands and arrows;

some stingless bee species, whose honey is highly appreciated by the Kaiabi; and mollusks

whose shells were used in the past to make necklaces and other adornments (Silva and Athayde,

2004). Regarding fish species, in the Xingu and in Teles Pires there is more availability and

diversity of fish than in Rio dos Peixes. The tambaqui fish (Colossoma macropomum) for

example, occurs in Teles Pires but does not occur in Xingu or Rio dos Peixes. The Kaiabi

mention that in Rio dos Peixes there is great availability of wild pigs (Tayassu tajacu), whereas

in Teles Pires, it is apparently easier to hunt a tapir (Tapirus terrestris), a highly appreciated

game. Monkeys, present in all three areas, especially the capuchin monkey (Cebus apella) and

the spider monkey (Atelespaniscus), are also very important in the Kaiabi diet, representing a

good source of protein during the rainy season, when it becomes more difficult to fish because of

the higher level of river waters. Jaguars (Panthera onca) used to attack Kaiabi camps in the past

and kill people, mainly children and women. The Kaiabi still use jaguar teeth and claws to

produce ceremonial necklaces, which can be used only by a great shaman or chief. Jaguars can

still be found at Xingu and Teles Pires, but are very scarce in Rio dos Peixes.

Environmental Conservation Status and Threats

The Apiaka-Kaiabi land in Rio dos Peixes (109,245 ha) is significantly smaller than the

Kayabi land in Teles Pires (declared 1,053.000 ha, see Chapter 6), and the environment has been

subject to greater pressure, disturbance and deforestation. The differences between the two areas

are magnified by management practices and pressure over specific resources. Furthermore, the

heart of Kaiabi territory in Rio dos Peixes, in the Batelao river, was left out of the borders of

Apiaka-Kaiabi land in the process of its demarcation. Resources present only in the Batelao area,









which have symbolic and cultural importance for the Kaiabi, are the special clay1 used by the

women to make pottery in the past and the mollusks used in necklace and ritual adornments

(Silva et al., 1999). The Kaiabi also share this small portion of land with the Munduruku and

Apiaka groups, making the pressure on important economic resources, such as Brazil nut and

fish, even greater. Therefore, plant and animal resources once abundant in that area are

nowadays scarce, and some are located outside the borders of the area. Kayabi land in Teles

Pires shows a similar situation in terms of access to strategic resources: the area located in the

Mato Grosso portion of the Teles Pires river was also left out in the demarcation process. It is in

this area that the most important cultural and economic natural resources occur, such as big

stands of Brazil nut trees, fruits, and other materials used for handicrafts and house construction.

Nowadays, the Apiaka-Kaiabi area is a little refuge of great ecosystem patchiness in the

middle of a cattle-ranching landscape. Travelling to the area, one can observe large cattle ranches

over the land that was once visited by Kaiabi and other groups to gather resources such as the

Brazil nut. The only trees standing are palms and isolated Brazil nut trees in the middle of the

pastures. In the expedition carried out by a group of Xingu Kaiabi to the Rio dos Peixes river in

1999 in which I participated, we were able to verify that many ranches and properties established

along the Rio dos Peixes and Batelao rivers do not comply with the Brazilian environmental law

regarding the protection of riparian forests covering the riverbanks (Law 4.771, of 09/15/1965;

Silva et al., 2000). The deforestation and occupation of areas close to riverbanks causes major





1In 1999, I participated in a historical trip to the Rio dos Peixes with a group of Kaiabi from Xingu and some non-
indigenous researchers. We visited historical sites in the Batelao region, and we tried to find the place from which
they used to extract the clay used to make pottery in the past, in an excursion which lasted many hours (Silva et al.,
1999; Senra et al., 2000). An elder who lived in Rio dos Peixes and had moved to Capivara village in Xingu was the
one who presumably knew where the clay spot was located, but we could not find it and we ended up giving up the
mission.









environmental impacts such as river silting, pollution, erosion and loss of habitat for plant and

animal species.

The main environmental pressures faced by the Rio dos Peixes Kaiabi today are invasion

and encroachment of their land by ranchers; exhaustion of fish resources (taking into account the

natural low fish availability and the fact that some people sell them in towns); river pollution and

siltation by cattle-ranching operations on the banks of the Rio dos Peixes and its tributaries; the

actions of logging companies in the area claimed on the Batelao river; and the planned

construction of the Rio dos Peixes hydroelectric plant (see Chapter 4). The establishment of

small groupings of families near the Rio dos Peixes waterfall is a strategy adopted by the group

to have greater control over that area, which is more prone to invasions.

The situation of the Kayabi land in Teles Pires is critical from a conservation standpoint.

While the process of extension or revision of land borders has dragged on for more than ten

years, loggers, ranchers and big companies who have their land lots in the area to de demarcated

for the Kayabi are chopping down what is left of the once exuberant forest. Teles Pires river is

still contaminated with mercury from gold mining operations, posing an extra threat to the

Kaiabi, Apiaka and Munduruku groups who depend on fishing as the main protein source of their

diet. Logging, agriculture and cattle ranching are the main economic activities of Alta Floresta,

Apiacas and Paranaita municipalities in Mato Grosso and Jacareacanga municipality in Para,

which have direct and indirect effects on the Teles Pires Kaiabi and their contested territory.

Illegal commercial logging is the economic activity which has caused by far the strongest

impacts inside the already demarcated land, as well as on the land in process of demarcation.

According to data collected by SEMA-MT (Secretaria do Meio Ambiente Mato Grosso,

2008) for a period of two years, between 2006 and 2008, the volume of timber legally exploited









in Alta Floresta municipality was 123,318 m3 and of 279,026 for Apiacas municipality, whose

area is adjacent to the Kayabi land, worth close to $US 2,508,631 and $US 6,987,344

respectively. There are about 90 species commercially exploited for timber in the region, of

which the main ones in terms of volume exploited are "amescla" (Protium heptaphyllum);

"angelim-pedra" (Hymenolobium excelsum); "cambara" (Vochysia sp); "cedrinho" (Erisma

uncinatum); "garapeira" (Apuleia sp); "champagne" (Dypterix sp); "jatoba" (Hymenaea

courbaril); "itauba" (Mezilaurus itauba); "macaranduba" (Manilkara huberi); and "sumauma"

(Ceiba pentandra), among others (SEMA, 2008).

Another economic activity which brings some environmental impacts, but also economic

benefits for the Kaiabi, is the sport fishing practiced along the indigenous land borders in the

Teles Pires River. As I am going to discuss in the next chapter, the main current source of

income of these Kaiabi is the money that they receive from the fishing resorts.

Conflicts and contested properties are abundant on one of the last continuous fragments

of forest over 1,000,000 ha to be demarcated for indigenous peoples in the Amazon. For

instance, the multinational company BRASCAN acquired an area of 136,000 ha (former Santa

Rosa ranch) in the Santa Rosa river, inside the land to be demarcated. BRASCAN officers and

representatives tried to reach agreements with the Kaiabi concerning the use of that area by them

until the demarcation battle is over (Rodrigues, 2004; Wenzel, 2005).

The Kaiabi have received different monetary proposals from companies, loggers and

miners concerning the possibility of use and exploitation of the resources contained in their area.

These proposals are sometimes very attractive and have caused social conflicts and divergence of

opinions among the members of the Kururuzinho post and village. Jodo Kaiabi, in an interview

given to some IBAMA (Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais









Renovaveis) representatives, gave vent to his feelings about the complicated situation of the

Teles Pires Kaiabi: "we are cloistered, ranches there, settlers, loggers, miners, everybody

surrounding us, therefore we are like in an island. We are fighting to secure the land for

ourselves, so we don't mix with the white men ... we are fighting to impede invaders inside the

area, we don't want conflicts with them. We are fighting to have the land demarcated." (IBAMA,

2004: 2)

In the last few years, IBAMA and Brazilian Federal Police have carried out patroling

operations to identify and arrest illegal loggers in the area, but the lack of a better structure for

carrying out this complicated work by the government agency makes the patroling work very

inefficient. In 2006, IBAMA and the Federal Police carried out what was known as the "Kaiabi

operation," fining ranchers, loggers, settlers and even government officers operating illegally

inside the indigenous land, in the municipalities of Apiacas, Paranaita and Jacareacanga. IBAMA

fined 27 people, totaling R$ 34,217,500 (close to $US 17,000,000), apprehended two tractors

and two chainsaws, and seized close to 30,000 ha of deforestation (Midianews, 2006). As part of

the same operation, the Brazilian Federal Police arrested 43 people accused of being involved in

environmental crimes in Mato Grosso indigenous lands, including IBAMA's ex-superintendent

in Mato Grosso, one SEMA-MT officer, along with illegal land grabbers, ranchers and logging

company owners (ISA, 2006).

Similar to the problems and pressures faced by the Kaiabi living in Rio dos Peixes and

Teles Pires lands, the Xingu Kaiabi are "locked" in an immense island of green in the middle of

various types of environmentally degrading economic activities. In the south, towards the

dominance of savannas and drier climate, soy cultivation and cattle ranching are the main

economic activities performed by small, medium and large farmers and land owners. In many









areas, there is a combination of cattle raising, soy, and millet plantations. Beyond the Park's

northern borders, the main economic activity is logging combined with big cattle ranches

(Sanches and Villas-B8as, 2005). Other activities practiced around the Xingu Park are sport

fishing (in fishing resorts) and rice cultivation. These activities have caused different types of

pressure for the peoples living inside the Park's limits, such as river pollution, overfishing,

invasions and social conflicts. Similarly to what happened in the other two areas, the most

important areas from the environmental standpoint were left out of the borders of the Xingu park

when it was created through the Villas-B8as' pressure in 1961: these comprise no less than the

headwaters of Xingu river, with its various tributaries and constituents (Menezes, 2000).

In the 1970s and 1980s, the surroundings of Xingu Park began to be occupied by large

agriculture and cattle ranches established by migrants coming from the south of Brazil,

encouraged by economic incentives provided by the Brazilian government at that time (Sanches

and Villas-B8as, 2005). The rapid occupation of the region contributed to the founding of

villages, towns and regional capitals, such as the city of Sinop, located in the heart of the Kaiabi

ancestral territory. The discovery of the timber potential in the region also contributed to its

urban development, especially with the simultaneous construction of big highways such as the

Cuiaba-Santarem (BR-163). Therefore, most municipalities located in the surroundings of the

Xingu Park were recently established, in the last 30-40 years (Figure 5-2). The first settlements

were located in the current towns of Sao Jose do Xingu and Sao Felix do Araguaia, followed by

Canarana and Agua Boa. Sinop, Vera, Santa Carmem and Marcelandia were founded after the

opening of the BR-163 road (Sanches and Villas-B8as, 2005).

Soy expansion in the region of the Xingu river headwaters has meant reason for worries

among indigenous peoples inside the Park's borders, together with environmentalists. This









expansion began in the 1990s, as a result of both government economic incentives and the strong

international markets for soy. Many rural producers migrated from the south to the north of the

country, attracted by settlement projects and economic subsidies. Still in the 1990s, big

multinational companies such as Bunge (1996) and Cargill (1997) arrived in the region

surrounding the Xingu park to buy grain and construct storage facilities in municipalities

neighboring the Park (ISA, 2003). It's obvious that to a certain extent, soy expansion happened

at the cost of cutting down forests. This does not mean that all soy expansion happens illegally

and should be avoided or prohibited, but that there is a real trade-off between increasing

production and decreasing forests. Mato Grosso occupies the first place in soy production in

Brazil, and soy production is also being expanded to the West region of the Park, in

municipalities in which logging is still the main economic activity, such as Sinop, Vera, Claudia,

Marcelindia and Nova Ubiratd (ISA, 2003).

In the 1994-2000 period, deforestation rates in the region of the Xingu watershed

increased by 40%, except in the Xingu Park and other indigenous lands, which hold the lowest

deforestation levels (ISA, 2003; TNC, 2006). Indigenous peoples living inside the Park haven't

been passive in relation to the soy and deforestation advance close to the park's borders. ATIX

(Associacgo Terra Indigena Xingu, see next chapter section about its creation and development)

has carried out a program for patrolling the Xingu park borders since 1997. ATIX officers and

community representatives have carried out various expeditions to the surroundings of the Park

in the last ten years. Through the support of ISA, IBAMA, FUNAI and other institutions, the

Indians have tried to establish a dialogue with cattle ranchers and soy producers that occupy the

surroundings of the Park, intending to minimize the impact of these activities inside the Park's

borders.









Once I witnessed Mairawe's speech during a meeting, struck by his cunning when

showing the map of the Xingu Park inside Mato Grosso "upside down". This makes sense to the

indians, because for them what matters is the direction in which the river flows: therefore, the

upper Xingu would be up in the map (south for us) and the lower Xingu would be down in the

map (north for us). Turning the map upside down makes sense because one can see all the

deforestation concentrated around the Xingu headwaters in the southern park's border region

magnified or highlighted with potential harm for the indigenous peoples who live inside the

"green island." The forests which occur mostly in the southern region of Xingu Park are are

currently considered one of the most endangered Amazonian ecosystems. The climate regime,

with marked seasonal droughts, makes this type of forest more prone to burn through successive

fire events every year (IPAM, 2008).

Worries about the situation of the Xingu headwaters' region have increased the attention

given to the region by NGOs and by Brazilian government state and federal environmental

agencies. NGOS such as ISA and IPAM (Instituto de Pesquisa Ambiental da Amaz6nia) have

developed initiatives to diagnose problems and to foster dialogue among the stakeholders

involved in occupation, development, degradation and conservation of the watershed. For

instance, IPAM has a project called "Projeto Savanizagco" (which means "savanization" or

desertification), in which a series of experiments using controlled fire have been developed to

deepen scientific understanding on the probability of conversion of Amazonian transition forests

into savanna vegetation, through the continuous vegetation burning for cattle ranching and/or soy

production (IPAM, 2008). In 2004, the NGO Instituto Socioambiental (ISA) promoted the first

meeting about the Xingu headwaters in the city of Canarana, to encourage the development of

dialogues, policies and actions, involving different actors (Indians, ranchers, soy producers,









settlers, loggers) and institutions (NGOs, government, civil society organizations and

movements, etc), to discuss the environmental conservation and sustainable development of the

region. From this first meeting, ISA created the Ykatu-Xingu Campaign (in the kamaiura

language, Y=river, water; katu=good, pretty, healthy). The lines of action of this campaign

(which has already carried out various projects and initiatives in the region) are the protection of

indigenous lands' rights; the promotion of the economic viability of rural settlements; the

reduction of the cost and the amount of time needed to recover riparian forests (through

reforestation and natural resource management); and the provision of basic sanitation services to

the cities of the region (ISA, 2008).

The coordinated efforts of these and other institutions working in the Xingu region have

meant an opportunity to carry out governance experiments and establish a more fluid and

qualified relationship between those inside and outside of indigenous lands in the Brazilian

Amazon. The resources mobilized through these initiatives have also directly and indirectly

affected indigenous organizations and Kaiabi communities living in the Xingu Park, in contrast

to the other two Kaiabi lands, where such a strong institutional force to contend with

environmental degradation is not present.

Environmental Adaptation, Natural Resource Use and Management among the Xingu
Kaiabi

Given the environmental contrasts and differences in climatic characteristics and floristic

and faunal composition between Kaiabi ancestral territory in Tapaj6s and the new territory in the

Xingu, displacement has meant dramatic changes in access to and management of natural

resources by the Xingu Kaiabi. A wide range of foods, materials for technology and

commercially valuable forest products- of both plant and animal origin- are missing in the









Xingu. As the elders say, referring to the lower diversity and amounts of birds and mammals, "in

the Xingu, the forest is poor."

Adaptation to the new environmental conditions has happened, in the last fifty years,

through diverse mechanisms and strategies, some consciously developed, some institutionally

induced, some community coordinated, some unstructured or unconscious responses to the lack

of choice. I identified seven mechanisms of environmental adaptation among the Xingu Kaiabi:

1) knowledge innovation in development of nomenclature for ecological zones and species; 2)

increase in diversity of resources used for different purposes (e.g. to build canoes) due to village

sedentarization and scarcity of important forest resources; 3) agrobiodiversity conservation and

recuperation of crop diversity; 4) travel to ancestral land to collect resources; 5) substitution by

other local species; 6) exchange of varieties and seeds among families, villages and other ethnic

groups; 7) semi-domestication or intentional management through experiments for planting

and protecting key resources.

One mechanism is the development of names and terms for new ecological zones, plant

and animal species which do not exist in the ancestral lands (Schmidt, 2001; Silva and Athayde,

2004). The naming process seems to happen always in reference to a known similar ecosystem or

plant. The presence of water is one of the criteria adopted by the Kaiabi to name ecological

zones. Therefore, "ka'a rete" (ka'a=forest; ete=true, meaning literally true or real forest) is the

forest that is never flooded whereas "yapopep" is the riparian forest which floods in the rainy

season (y=river, water). Forest appearance or dominance of one plant species is another criterion:

"ji" is the general name for grass and also the name used for grassy savannas Campopo" in

Portuguese), "jfsing" is the shrubby-tree savannas or "cerrado" in Portuguese. The Kaiabi

differentiate secondary forests which suffered past human manipulation through agricultural









practices, and which are suitable for agriculture in the present because of their fertile soils (Silva

and Athayde, 2004).

The difficulty to find fertile soils for agriculture in the Xingu Park has meant the

development of a system of classification and management of "black earth" anthropogenic soils

(Petersen et al., 2001), which are the best soils for agriculture in general and for more demanding

crops (such as peanuts and maize) in particular. The Kaiabi distinguish mature forests over these

soils by looking at species composition and variety of animals. According to them, these patches

of forest, which they call "kofet rarete" are richer in useful species2. Historically, the main

reason for Kaiabi establishment of villages in a geographic region was the presence of forests

over black earth soils (Rodrigues, 1993; Schmidt, 2001; Silva, 2002 a; 2004).

Heckenberger et al. (2003) argued that the forests that grow over the black earth soils

might have a different floristic composition when compared to other types of forests, resulting

from the co-evolution of indigenous management and ecological dynamics. These authors

presented recent results of archaeological research carried out with the Kuikuro people in upper-

Xingu river, identifying the occurrence of complex indigenous settlements in the last thousand

years, suggesting that the Kuikuro might have produced deep alterations in the forests and other

local ecosystems. More recently, in the last forty years, the Kaiabi have also contributed to shape

forest ecology and composition around the villages, through shifting agriculture and management

of secondary succession (Silva, 2002; 2004).


2 Here, I translated a paragraph of a text written by four Kaiabi teachers during a training course in ecology at Xingu
Park. Capoeira (Portuguese name) and Kofet (Kaiabi name, Ko=agricultural plot, fet=something that occurred in the
past) mean patches of forest over black earth soils. "The 'capoeira' is very important for our society. In the "kofet"
there are a lot ofjatobW, barriguda, inajh and other trees that like to live in the places where the soil is black. These
forests were made by the ancient indigenous peoples and are good places to plant our crops. The 'kofet' is a place
used for agriculture in the past, where there are many fruits that are food for the animals and for the people. You
find many animals and game there. This is why the capoeira is so precious for us, the Kaiabi." Matari, Jamanary,
Takapeju'i and Jemy Kaiabi indigenous teachers, second training course for indigenous environmental managers,
Xingu Indigenous Park. Translated from Silva (2002 b: 18).









According to the Kaiabi, in their ancestral land at Rio dos Peixes and Bateldo rivers, there

was more availability of black earth and good soils for agriculture. They recognize plant species

that are indicators of black earth forests, and they say that normally they find black earth spots

when they are hunting, because there is also greater animal diversity in these sites, theoretically

associated with greater fruit availability. However, due to the sedentarization of indigenous

villages and to the impossibility of practicing agriculture in more remote places far from the

villages, the areas of black earth near villages have been overexploited through increased

pressure of shifting cultivation, with reduced time between swidden-fallow cycles, thus causing

retardation of the successional process and consequent loss of soil fertility (Silva, 2002 a). This

situation poses a challenge to both agrodiversity and biodiversity conservation, since more

demanding crop plant varieties do not grow well on overexploited, less fertile soils. Thus, the

process of recovery following swidden-fallow agriculture might be threatened, compromising the

dynamic cycles of disturbance and recovery that characterize tropical forests under indigenous

management (Gunderson and Holling, 2002).

Even with the constraints regarding drier climate, limited black earth availability, village

sedentarization and changes in social organisation for agricultural production (Silva, 2009), the

Xingu Kaiabi have managed to keep and even increase the diversity of crop plants that they

have cultivated in the last centuries (Silva, 1992; 2004; 2009). Crop species and varieties were

taken to the Xingu by Kaiabi families at the time of the transfer, but the movement of varieties

between Xingu and ancestral lands has not stagnated. Known for their sophisticated agricultural

system and culinary traditions, the Kaiabi have developed new varieties of peanuts in the last 40

years after the transfer to the Xingu through a combination of shamanism and community-based









projects3 (Silva, 2002; 2004; 2009). By contrast, people living in the Rio dos Peixes and Teles

Pires areas have lost the knowledge and use of traditional peanut varieties to different degrees,

especially amongst the younger generations. Curiously, what has happened at Rio dos Peixes is

that families who were living in the Xingu and moved back there took crop plant varieties with

them and spread those among some kin residents. For example, Jurumuk's (deceased, also

named "Cuiabano") family decided to move back to Rio dos Peixes after we visited the area in

1999. Jurumuk's widow, named Jeru'a, brought cotton seeds and some peanut varieties from the

Xingu back to Rio dos Peixes. After they returned to the Xingu, people in Rio dos Peixes still

kept these new varieties. Another case is Canisio's family, which moved back to Rio dos Peixes

after living many years in Xingu around 1998.

There is a constant movement of people and crop varieties back and forth between Xingu

and Rio dos Peixes, which happens to a lesser extent between the Xingu and Teles Pires. We

can actually infer that when whole families move to another territory, including older people,

the transfer of knowledge (comprising different indigenous knowledge domains such as

environmental, artistic, or agricultural) and materials or resources (be they material culture

objects and or crop plants, for instance) is greater then when only young or middle-aged men

move, which is the case in Teles Pires. Interestingly enough, Xingu now functions as a reserve

of crop varieties, as in the other two areas the diversity of traditional crops is smaller.

Concerning forest resources, the most important plants used for crafting material culture

objects, and also appreciated as edible fruits, are absent or occur sparsely throughout the Xingu

parks' drier ecologic zones. A free list exercise of the most used plant species present in the

3 It is estimated that they have around 40 varieties of peanuts now. Peanut seeds were initially brought by the
families who came on foot to the Xingu Park (Silva, 2004). The Kaiabi developed a community-based project,
called Monowi (which means peanut) from 2004 to 2006. In this project, supported by the PDPI program and
carried out through ATIX with ISA personnel support, the Kaiabi aimed to promote the multiplication and
distribution of peanut varieties among different Kaiabi villages in the Xingu Park (see next chapter).









ancestral lands was carried out with a total of 224 people (110 men and 114 women)

interviewed in four villages among the three Kaiabi groups. The results include a list of 92

ethnospecies used as fruits or materials for handicrafts, such as fibres, dyes, wood and resins

(Table 5-1). The results differed for men and women, reflecting the economic and cultural

importance that each gender gives to plant resources based on their uses. Figure 5-3 shows the

differences between genders concerning the species of fruits and materials most used for crafts

and construction. Amongst the women, the first most used species was the "tucumd" palm

(tukuma 'yp, Astrocaryum aculeatum), cited in 45% of the cases. Women use the rigid black nut

of this palm to manufacture beautiful necklaces, bracelets and rings which are sold or

exchanged. For the men, the most cited species was the "pupunha-brava" or "siriva" (yryp,

Bactris macana), cited in 59% of the cases. This palm also does not occur in Xingu Park: it is

typical of forests located over rock formations, such as in Teles Pires area. The black wood

extracted from its spiny trunk is the most appreciated for bow-making, besides its use also for

crafting clubs and spindles. The Xingu Kaiabi have substituted this resource by other palms

such as the "tucumd" (Astrocaryum aculeatum) and the "bacaba" (Oenocarpus bacaba), but

they say that they are of inferior quality, especially for bows, because of the high elasticity of

"siriva" wood.

The third most cited resource for women and men alike, mentioned in 44% of the total

cases, was the "Castanha-do-Para" or Brazil nut ("ywaete"- ywa=fruit; ete=true, real;

Bertholetia excelsa), which does not occur at all in the park region. The loss of access to this

resource meant a big economic and cultural impact for the Xingu Kaiabi after the displacement.

Brazil nut is highly appreciated by the Kaiabi since immemorial times, and the seeds' oil is used

in traditional culinary. It is worth mentioning here (more details will follow in the next chapter)









that Brazil nut extraction is the most important cash economic activity practiced in Rio dos

Peixes today. In Teles Pires, the bigger stands of Brazil nut are located in the Mato Grosso state,

in the area currently under contestation. More than once I have witnessed people coming from

Rio dos Peixes to Xingu bringing bags of Brazil nuts as gifts for their relatives.

The interviewees mentioned a total of fourteen palm species, the most important family

of plants producing fruits, fibre, oil and thatching materials among Amazonian indigenous

peoples in general (Balick and Beck, 1990). The acai berry (Euterpe oleraceae) was the second

in the overall resources list, mentioned by 69 people, in 43% of the cases. It scored second for

the women and fourth for the men (42.5% and 44% respectively). The fruits of this palm are

highly appreciated in the preparation of juice and porridges, and have a high nutritional value.

In Xingu Park, very few representatives of "acai" palm occur in some forest formations towards

the north-western region, where the climate is more humid.

A particularly important plant species missed in Xingu is uruyp (Ischnosiphon gracilis,

Marantaceae, aruma in Portuguese), the main fibre used in Kaiabi basketry. This herb forms

clumps in swampy and periodically flooded areas throughout lowland Amazonia (Andersson

1977). The fibre from various Ischnosiphon species is removed from the external surface of the

stems and used for basketry weaving by many South American indigenous and traditional

peoples (Balee, 1994; FOIRN/ISA, 2000; Guss, 1989; Milliken et al., 1992; Nakazono, 2000;

Ribeiro, 1985; Van Velthem, 2001). The main objects produced both for subsistence and for

commercialization are baskets, mats, war club adornments, sieves, bracelets, and headdresses.

The importance of"aruma" for the Kaiabi is demonstrated by its high score in the free list

assessment, appearing second on the men's list (with a relative frequency of 52%) and in

thirteenth on the women's list (relative frequency of 16%). More details on aruma occurrence









and management are found in Chapter 8, in the section about natural resources used in basketry

and textiles.

Another important species absent in the Xingu is the bamboo used by the Kaiabi to make

hunting and ceremonial arrows (Guadua spp). It scored fifth in the men's resources list (39%).

This resource occurs only in very specific sites of Rio dos Peixes and Teles Pires areas: in Rio

dos Peixes, it occurs outside the borders of the indigenous reserve, close to the Bateldo river. In

Teles Pires, it occurs inside the area designated for the army, which they can visit and collect

the resource with official permission. Among other species mentioned are wild fruits which do

not occur or occur sparsely in Xingu. During the interviews, people mentioned that one of the

things they missed most in the Xingu was the quantity and diversity of fruits available,

especially at Teles Pires. Among these, we can mention "api" varieties (Naucleopsis sp), cocoa

tree (Theobroma cacao), "uxi" (Endopleura uchi), "pajura" (Couepia bracteosa) and others.

From the results, we can infer that while for men the most used resources are materials

for handicrafts, specifically those plants used for making bows, arrows and baskets, for women,

that the most used and prized resources are fruits (with the exception of the tucumd palm, cited in

first place). Fruits were also mentioned by men, but the relevance given to materials used for

handicrafts was greater. These results reflect both the dependence on those objects for

subsistence activities, such as hunting and fishing (bow and arrow) and also the value given to

basketry in both cultural and economic terms, since these items (especially the painted baskets

with graphic designs) have been extensively commercialized and traded by the Xingu Kaiabi

after the transfer. On the other hand, women cited "tucuma" palm in first place, which also

reflects the economic value both exchange and market values- of objects produced from this

palm.









The Xingu Kaiabi have compensated for the environmental shortcomings faced after the

transfer in a number of ways. First, people, mostly men, occasionally travel to Rio dos Peixes

and Teles Pires to visit their relatives and collect plants such as aruma, taquari and siriva. On

their return from these trips, they sometimes also bring back seeds and cuttings to plant in the

Xingu. Given the drier climate, however, some of these attempts notably Brazil nut, cocoa and

the siriva palm- have not been very successful. Another alternative is the use of substitutes, as

they have been experimenting for "aruma", "siriva" and wild fruits. In fact, there is a two-way

process going on between Xingu and the other two lands. Xingu residents or people visiting from

Rio dos Peixes and/or Teles Pires have taken seeds and seedlings of crop plants back to those

areas. Similarly, Xingu residents have also collected materials and brought seeds of forest

resources from ancestral areas to the Xingu.

The role of institutions in shaping natural resource practices among indigenous peoples

might be illustrated with the Kaiabi example. Institute Socioambiental has carried out a project

(in which I used to work) for development of economic alternatives and management of natural

resources among Xingu indigenous peoples since 1996. Through this project, the Kaiabi have

participated in a formal ecological research project (focusing on the "aruma" plant, see Athayde

et al., 2006) and in the conduct of different management practices with problematic or desired

resources, mostly plant species. The idea is that once you live in a delimited land, the resources

available inside it are also finite and might be subject to increased pressure, especially those used

to make products oriented to the market, such as "aruma", "tucuma" and "inaj (Athayde,

2000). In addition, the stock of resources used in subsistence activities might also be

compromised by their continued use in a scenario of population growth and village

sedentarization (COICA, 1996; Athayde et al., 2002). Therefore, applying traditional









management techniques would not suffice to ensure the sustained use of the resource, which

would demand adaptations in livelihood strategies (Castellanet and Jordan, 2002). It is in this

context that ecological science might be useful, sometimes necessary, to support territorial

management amongst indigenous peoples in circumscribed territories. However, every research

project or participatory management activity requires institutional and monetary arrangements,

which also influence local forms of social and political organisation. My reflection is that these

projects contribute to the formalisation of management techniques and that often the results of

combined western and traditional natural resource management practices are only effective over

the long-term; therefore they might not be perceived as tangible results by local communities.

Nevertheless, these projects might also contribute to raise awareness within the community about

the status of strategic resources and prospects for their adaptive management. The possibilities

brought by the adaptive management concept and practices (Stankey et al., 2005) might be good

platforms or starting points to be combined with participatory research in the achievement of

sustainable use of resources by indigenous communities living in a circumscribed, politicized,

institutionalized and market- oriented context. Participatory projects for natural resource

management among indigenous peoples may function, at the same time, as agents of social

change, adaptation of traditional practices, and sources of empowerment of local communities

(Klooster, 2002; Athayde et al., 2006).

Conclusion

Geographical displacement has presented challenges for the three Kaiabi groups alike, in

terms of environmental conservation and natural resource management. The climatic conditions,

the appearance and structure of vegetation differ greatly between the Tapaj6s watershed where

the Kaiabi used to live and the Xingu watershed. In the Xingu, climate is drier and forests lack

many Amazonian elements important for Kaiabi physical and cultural reproduction.









Environmental conservation is a concern for the three areas, but it is more critical in Rio

dos Peixes and Teles Pires, since the area is smaller and the institutional support for territory

monitoring and patroling is basically inexistent in Rio dos Peixes and still very weak in Teles

Pires. In the Xingu, the coordinated effort of institutions and Indians have resulted in successful

partnerships for territorial control and monitoring, where actions span from inside to beyond the

Park's borders. The main conservation threats faced by Kaiabi groups today are logging, cattle

ranching, soy expansion and hydroelectric plant construction.

The Kaiabi have developed at least seven different adaptation mechanisms to deal with

the environmental constraints faced at Xingu after the transfer: 1) knowledge innovation or

creation; 2) increase in diversity of resources used for different purposes due to scarcity and

village sedentarization; 3) conscious protection, multiplication and development of new crop

varieties, enhanced by community-based projects; 4) substitution of strategic resources by

others of similar quality; 5) travelling to ancestral land or to other areas to collect strategic

resources; 6) exchange of crop varieties between the three groups; and 7) development of

management practices for specific resources, through institutional support and collaboration

between western and traditional knowledge. All of these mechanisms combined have conferred

ecological resilience to the Xingu Kaiabi. Also, the isolation and the distance to industrialized

products and to non-indigenous agents have probably reinforced the dependence on natural

resources among the Xingu group, in contrast to the other two groups. This is particularly true

in relation to crop plants: the Xingu Kaiabi have not only conserved the traditional peanut

varieties, but also developed new ones. In the other areas, the two groups have experienced loss

of traditional crop varieties since they became more oriented and dependent on industrialized

food bought in nearby towns.









The results of a free list exercise carried out with men and women in the three areas show

a difference in species mentioned by the two genders. The common result is that resources used

for the manufacture of products oriented to the market scored higher than resources used solely

for subsistence, especially in the men's case. However, many times market and use value

overlap. Some resources which scored high for both genders were Brazil nut, "acai" palm,

"siriva" palm, "aruma" and fruits such as wild cocoa and "api".

In the past, Amazonian indigenous societies were very mobile; nowadays they are

becoming increasingly sedentary (Alexiades, 1999; Senra, 2004; Milliken, 2006; Zent, 1999).

The problem of sedentarization of indigenous peoples brings new challenges to anthropologists

and human ecologists who try to understand social-cultural responses to ecological

characteristics and vice-versa. Regarding the Kaiabi, their ecological adaptation to Xingu Park

is also an example of knowledge innovation and adaptation, in spite of ecological constraints

augmented by the sedentarization of villages. Among the Xingu Kaiabi, adaptive management

of natural resources has occurred through conscious development and monitoring of new

management practices through collaboration with "western-science oriented" research, such as

the Monowi project, for recovery of peanut varieties, and the "aruma" project.



















COL~MIA


PARA,


0 r



TOCANTINS



Conservation Units, Indigenous
Reserves, and Deforestation in the Amazon DF
Brauin .*
Dhf-tnboh M Sbh CchrhAlab L -AM.
Fderal crmcnsrtloAnu za Indnou irtumr OIAS
200 a 2DD 400 Kilomelers MAt .- O/
DOSUL N -.. ASGERAIS

Figure 5-1. Indigenous lands (green), federal and state protected areas (blue and brown) and
deforestation in the Brazilian Amazon. Source: EDF (2001).























































STerras indigenas
O Llmlu da Bacia do do Xlngu


Figure 5-2. Map showing the municipalities that surround Xingu Indigenous Park. Source: ISA
(2008 b).


I *LOOLIZgA










Table 5-1. Plant resources of cultural and economic importance mentioned in free listings
conducted with men and women in four Kaiabi villages*.
Name in Kaiabi Name in Portuguese/ Scientific name Type of N of Freq.
English resource citations citations
(%)


Ywaete


Jujywa
Yryp/Wyrapat
Akusikanafuf

Uruyp ete
Api
Ka'au wywa
Tukuma'yp

Pataud
Ywapiruru
Kamai'yp
'Nga
Pinowa'yp

Inata'yp
Akusityrywa
Simuku'a
Myrysy'wa

Uruyp kuruk
Uxi
Ka'a si'a
Ywapirang
Pajura
Pinowauu
Juba
Kwae'ma
Uruyp piremi
Tamemuri
Kwanu'ywa
Awai
Pinnowa'i
Wyrawuru'a
Apiuu
Araityranauu


Castanha-do-
Para/Brazil nut
Agai
Pupunha-brava
Mao-de-cachorro,
mao de cutia
Arumm verdadeiro
Api, chimico
Cacau/Wild cocoa
Tucuma

Pataud


Flecha, Taquari
Ingi
Bacaba

Inaja
Mao-de-jaboti

Buriti

Aruma rugoso
Uxi
Abuta
Cafe-de-macaco
Pajura
Bacabao
Juba

Aruma

Fruta de gaviao
Pequid
Bacabinha


Api grande
Bacuri grande


Bertholetia excelsa FR 71 44.10


Euterpe oleraceae
Bactris macana


Ischnosiphon sp
Naucleopsis sp
Theobroma cacao
Astrocaryum
aculeatum
Jessenia bataua


Guadua sp
Inga spp
Oenocarpus bacaba

Maximiliana maripa



Mauritiaflexuosa

Ischnosiphon gracilis
Endopleura uchi
Abuta grandiflora

Couepia bracteosa
Oenocarpus sp



Ischnosiphon spp



Caryocar brasiliense
Oenocarpus sp

Cf Naucleopsis sp
Platonia insignis


FR
HA,WO
FR

HA,FI
FR
FR
HA,FR,W
O
FR
FR
HA
FR
HA,
WO,FR
HA,TH
FR
HA
HA, FI,
PA
HA, FI
FR
FR
FR
FR
HA, WO
FR
HA
HA, FI
FR
FR
FR
FR, HA
HA
FR
FR


42.86
37.89
36.02

35.40
32.92
31.06
31.06

26.71
22.98
22.36
21.12
20.50

19.88
17.39
14.29
13.66

13.04
12.42
9.32
9.32
8.70
7.45
6.83
6.83
6.83
6.21
5.59
5.59
5.59
5.59
4.97
4.97










Table 5-1. Continued
Name in Kaiabi Name in Portuguese/ Scientific name Type of N of Freq.
English resource citations citations
(%)


Jatobi
Babacu
Api
Bacuri


Hymenaea spp
Attalea speciosa

Platonia sp


Caja
Caju-do-mato


Ywy taga
Juta'yp
Inatauu
Apiawijup
Araityrana
Jakarana
Kaja'ywa
Kaju'u
Kyryma
Pariri
Taipawa
Ywasi'asi'u
Apinan
Capim-navalha
Jutaiwa'i
Kyrymauu
Marakuja
Panakuwa
Uruywuu
Ywa'yp
pirang/ngua'yp
Kupajyp
Olho-de-boi
Ang'ram
Tangrerewa
Yrem
Ywasi'asi'i
Abiu
Ameiwyt
Asianguauu
Asi'a ywa
Ata
lang'yp
Irupirangi
Irupiwat
Jakurupe'yp
Jasimuku'a


Spondias sp
Anacardium giganteum


Pouteria pairiry


Api

Jatobazinho


Maracuj a-do-mato


Aruma grande
Falso pau-Brazil


Copaiba
Olho-de-boi
Pororoca
Murici
Mogno

Abiu
Cip6-imbe




Itafiba


Passiflora nitida


Copaifera sp
Ormosia arborea


Byrsonima sp
Swietenia macrophylla

Pouteria caimito
Philodendron imbe




Mezilaurus itauba


i 7 A Q


FR
FR,TH
FR
FR
FR
FR
FR
FR
FR
HA, FI
FR
FR
HA
FR
FR
FR
HA, FI
HA, FI
WO


RE
HA
HA
FR
WO
FR
FR
HA,FI
FR
FR
FR
WO
FR
FR
HA, WO
FR


3.73
3.73
3.73
2.48
2.48
2.48
2.48
2.48
2.48
2.48
2.48
2.48
1.86
1.86
1.86
1.86
1.86
1.86
1.86
1.86


1.86
1.24
1.24
1.24
1.24
1.24
0.62
0.62
0.62
0.62
0.62
0.62
0.62
0.62
0.62
0.62


Pariri
Cip6










Table 5-1. Continued
Name in Kaiabi Name in Portuguese/ Scientific name Type of N of Freq.
English resource citations citations
(%)
Jowosiupy -FR 1 0.62
Jujywa'i Acai-pequeno Euterpe sp FR 1 0.62
Kamaiynan Falso-taquari -HA 1 0.62
Kawiang'yp WO 1 0.62
Kwarykyta'ywa -FR 1 0.62
Mukai'yp Caraua, caroa Neoglaziovia variegata HA, FI 1 0.62
Muri'i Muricizinho Byrsonima sp FR 1 0.62
Panakupe'yp WO 1 0.62
Parajuba'a -FR 1 0.62
Pasi'yp Paxiuba Socratea exorrhiza WO 1 0.62
Pokowy'a Banana-brava, Heliconia sp HA,FI 1 0.62
Pacova
Takamu Salacia cfspectabilis FR 1 0.62
Takwarete Taquara HA 1 0.62
Tamakrea HA 1 0.62
Tamemuriuu FR 1 0.62
Tapi'u ywa FR 1 0.62
Tukuma'i Tucunzinho Astrocaryum sp HA,FR 1 0.62
Wyrapat pytang -WO 1 0.62
Ywauni FR 1 0.62
Ywyjupe -RE, HA 1 0.62
Ywy mu'a -FR 1 0.62
Ywy wuu -FR 1 0.62
*Organized by order of most cited (for both men and women). FR fruits; HA used for
handicrafts; WO- wood; FI fiber; RE resin; TH leaf used for thatching. Sources: fieldwork;
Silva and Athayde (1999); INBRAPI (2008).














Uruyp piremi


Kamaiyp


Wyrawuru'a :::::::: :


Uxi


Akusityrywa :.:


Myrysy'wa


Y'ryp


Kwae'ma ::::::::::::|


Uruyp wete -. .


Ywapiuru

*men
6 Pinowa
Women

Simuku'a


Nga : : : : : : : : .


Ka'au wywa


Pataua





Api .


Akusikanafu .............


Ywaete ::: :: :: :: ::: :: :: :: :


Jujywa ..........................

Tukuma'yp .............. .... ...


0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00 55.00 60.00 65.00

frequency of citations (o)



Figure 5-3. Selected plant resources most cited by women and by men interviewed in four Kaiabi

villages.









CHAPTER 6
CURRENT TERRITORIAL ORGANIZATION AND POLITICAL CONFIGURATION
ACROSS THREE KAIABI LANDS

Introduction

This chapter presents an overview of current territorial organization of the three areas

occupied by the Kaiabi in the present Rio dos Peixes, Xingu and Teles Pires including

general information on demography, infra-structure and socioeconomic aspects, with a focus

on institutional and political configurations. The information provided in this chapter is a

chronological continuation of Chapters 3 and 4, setting the stage for the specific comparisons

on socioeconomic and weaving knowledge aspects which follow in Chapters 7 and 8.

The main questions explored in this chapter relate to the factors which led to the

demographic expansion and to the political empowerment of the Xingu Kaiabi in contrast

with the other two groups. More broadly, I seek to contribute to the current debate of how

Amazonian indigenous peoples have responded to and appropriated institutions, resources,

systems of norms (regulations, laws, polices) and values (symbols, such as the Amazon in the

view of conservationists) in the construction of their identities and management of territories

(Gray, 1997; Ramos, 1998; Albert, 2005; Hierro, 2005; Hierro and Surrales, 2005; Oakdale,

2005). I also explore people's perceptions about the work developed by the three Kaiabi

organizations. I present the results of interviews done in four Kaiabi communities (men and

women) living in the three areas considered in this study, exploring their understanding,

participation, and expectations related to the role and work of the indigenous organizations

ATIX (Xingu), Itaok (Rio dos Peixes) and Kawaip (Teles Pires).

I briefly describe the development of the indigenous grassroots movement in Brazil,

as it is tied to prospects for Amazonian environmental conservation and the empowerment of

the Xingu Kaiabi, through the formation of salient political leaders. The legalization of









indigenous organizations after the Constitution of 1988 brought several changes in the

process of dialogue and interaction between indigenous peoples, the State and other

institutions. In this context, the creation and development of ATIX (Associacgo Terra

Indigena Xingu) brought not only empowerment to the Xingu Kaiabi, but also changed the

way in which other ethnic groups living in Xingu began to relate with the outside and to

manage and patrol their own territory. However, there might be trade-offs and pitfalls in the

assimilation of western institutions by indigenous peoples, which are discussed in this

chapter, based on the Kaiabi experience. The process of creation and implementation of

indigenous organizations in Rio dos Peixes and Teles Pires is described and compared to

ATIX, and obstacles faced by the Kaiabi in the administration of these institutions are

identified.

Finally, I present information on the development of community-based projects

among the Kaiabi (including the Kaiabi Araa project for revitalisation of basketry and

textiles knowledge), enabled by funding opportunities coming from national and

international political-environmental agendas for the Amazon. This information is

accompanied by a brief critique of opportunities and constraints involved in the adoption of

the "projects culture" by the Kaiabi. I hope that the information provided in this chapter

serves to improve the relationships of Kaiabi indigenous organizations among themselves

and between them and their constituencies (communities). I also hope that this information is

useful for donors, practitioners, policy-makers and researchers working with development

and conservation among Amazonian indigenous peoples.

Overview of the Areas Currently Occupied by the Kaiabi

Currently, the Kaiabi live in three indigenous lands: the Xingu Indigenous Park, the

Apiaka-Kaiabi indigenous land in Rio dos Peixes, and the Kayabi indigenous land in Teles Pires









River (Figure 6-1). In all areas, the territory is shared to greater or lesser extent with other

indigenous groups: in Xingu, with thirteen other groups; in Rio dos Peixes and Teles Pires

mainly with Apiaka and Munduruku residents. In the next sections, I present an overall

description of each land occupied by the Kaiabi.

Xingu Indigenous Park

Xingu Park has a current area of 2,642,003 million ha (26,420 square kilometers), where

nearly 4,700 indians live (ISA, 2008 a; Figure 6-2). The Kaiabi are the most numerous of the

fourteen indigenous groups sharing the Park's territory, totaling nearly 1,000 individuals in 2007.

The Park is located in the northeastern region of Mato Grosso State, surrounded by 11

municipalities, with which the Indians relate in different intensities. Two main dirt roads provide

terrestrial access to the Park, both accessible through Canarana town. The road by the Culuene

river is more used by the upper-Xingu groups, while the road by the Suya-Migu river is used by

the groups which live in the middle and north sections of the Park, including the Kaiabi. There

are also two other access points to the Park used by the Kaiabi: through the BR-080, a road

which crosses the Xingu river at the Park's northernmost limit, and through the municipality of

Marceldndia, also to the north, located close to Manitsaua-Migu river's margins. A typical travel

route would come from Brasilia (DF) by bus in a 16-17 hour trip to Canarana, then take a 5-8

hour truck ride to the Suya-Migu port at Ngosoko or Beira-Rio villages (where some Suya live),

followed by a 3-4 hour boat trip on the Suya-Migu until reaching the Xingu and finally the

Diauarum Post.

The Park can be divided in three main sectors: the upper Xingu region or south portion,

the middle section, and the lower or north portion of the Park. In each sector, there is a FUNAI

administrative unit or post, where the health and administrative services and offices are

centralized: the Leonardo Post in the upper Xingu portion, the Pavuru Post in the middle sector,









and the Diauarum Post in the north portion (see Figure 6-2), which serves the Kaiabi and where

ATIX's headquarters is located. The park is administered by FUNAI regional office (ADR

Xingu- Administracgo Regional do Xingu) located in Brasilia, where the Park's director and his

assistants work. The Park has been administered by Indians since the "Guerra da Balsa" incident

(described in the following sections about Mairaw6's political trajectory), in 1985. The current

administrator is Tamalui Mehinaku.

The fourteen groups which share Xingu Park's territory are divided among the three

sectors. Next to each group's name, I give the estimated population for 2002 in parentheses (ISA,

2008 a; Menezes, 2008). In the upper portion we can find the upper Xingu cultural complex,

which includes those peoples that have occupied that region for hundreds of years, and speak

different languages: Aweti (138), Kalapalo (417), Kamaiura (355), Kuikuro (415), Matipu (119),

Mehinaku (199), Nafukwa (105), Trumai (120), Waura (321) and Yawalapiti (208). Bruna

Franchetto and Michael Heckenberger (2001) have organized a volume which includes detailed

information on upper-Xingu groups' history and culture. According to Heckenberger (2001),

Xingu river's upper region, encompassing all the tributaries which will merge to form the Xingu,

was once inhabited by thousands of Indians from different linguistic affiliations. The

concentration (by migration and geographical dislocations) of some groups in the same

geographical region for hundreds of years, and the relative isolation from other tribes or from the

Brazilian non-indigenous society, in addition to more complex ethnological factors, led to the

formation of what is known as the upper-Xingu cultural pluralism or the upper-Xingu cultural

complex (Galvdo, 1953; Heckenberger, 2001; 2005). Upper-Xingu groups have in common some

social organization structures and cultural traits such as festivals (e.g. Kwarup, the festival to

render homage to the dead), shaman practices, adornment practices and body paintings, and









some material culture objects (Galvao, 1953). It is estimated that at the time of the first contacts

with the non-indigenous, such as when the expedition of Karl Von den Steinen reached the upper

Xingu in 1884, there were some 3,000 people living in that region. Later on, with closer contact

with Brazilian nationals and the outbreak of epidemics (mainly measles), the population dropped

to 574 people in 1954. In spite of efforts to improve the health situation, conditions continued to

be precarious, and the upper Xingu population reached its lowest level in 1965, with only 542

people (Heckenberger, 2001; ISA, 2008a). Since then, improvements in access to health and

sanitation services and to financial and technical resources have enabled demographic

recuperation and growth among upper-Xingu groups.

The middle portion of the Park is inhabited by the Ikpeng (319) and by some Trumai and

Kaiabi communities. The Ikpeng speak a language in the Karib family, and came to live in the

upper Xingu region (outside the actual Park's borders and before the Park was created) at the

beginning of the 20th century, living in a state of war with their upper xinguano neighbors

(Menget, 2003). Contact with the non-indigenous society happened in the beginning of the

1960s, and had disastrous consequences for their population, which was reduced to less than half

as a result of diseases and killings. At that time, they were pacified by the Villas-B8as brothers

and transferred to the Xingu Park, similarly to what happened to the Kaiabi (Menget, 2003).

The Suya or KIs6dj6 (self-designation, 334) are the only G6 speaking group in Xingu

Park. Since 2001, most of the population lives in the southeastern section of the Park, in the

Wawi Indigenous Land, which was annexed to the Park's area in 1998 (Seeger, 2003). Since

their arrival in the Xingu region (probably in the second half of the 19th century), as a

consequence of contact with other Xingu groups and, primarily, with those of the so-called









"Upper-Xingu cultural complex," they have adopted many new cultural features and

technologies (Seeger, 1981; 2003).

Finally, the northern portion is occupied mostly by Kaiabi (around 1,000) villages, but also

by a few Yudja (or Juruna, 248) villages and by some mixed Kaiabi-Yudja villages. The Yudja

(commonly called "Juruna"), speak an isolated language in the Tupi stock. Their ancestral area

was located in Para State, from where they fled in the beginning of the 20th century, due to

conflicts with settlers and rubber tappers. They first settled around the upper-Xingu region, but

later on moved towards the north of the Park in consequence of conflicts with upper-Xingu tribes

(Lima, 2001).

Communication between villages is commonly by radio, and transportation is by river,

mostly through the use of motor boats run by gasoline. Big ferrys run by diesel engines transport

heavy loads and gasoline to the villages and indigenous posts. Usually, villages have solar panels

for the functioning of the radios. Posts and some villages have diesel generators which provide

electricity for the houses at night and for the functioning of health units as needed. Gasoline is

the single most valued and difficult to control commodity in the whole Xingu Park. Internet

access is available at ATIX headquarters in Diauarum Post. A few villages (such as Ngojwere-

Kis6dj6 and Tuiarare-Kaiabi) have public phones installed through the support of local

municipalities.

The local economy in the Park is a mix of different activities practiced to lesser or greater

extent by different groups: a) subsistence-oriented activities such as agriculture, fishing, hunting

and forest extraction; b) NTFP (non-timber forest products) production for use and/or market

commercialization, such as handicrafts, honey, fruits and oils; c) labor employment (such as

FUNAI and ATIX officers, teachers or health professionals); d) temporary paid work done in the









villages or posts (such as workers for house construction, cooks and fishermen who work for

training courses or meetings) and; e) temporary work done outside the Park's limits (such as in

towns outside the Park or in ranches, etc).

Health services for Xingu Park inhabitants are provided by the UNIFESP/EPM

(Universidade Federal de Sao Paulo/Escola Paulista de Medicina), through an agreement

between them and FUNAI and later FUNASA (Fundacgo Nacional de Saude), which was first

signed in 1966 (Villas-B8as, 2005). The DSEI Xingu (Distrito Sanitario Especial Indigena do

Xingu) was created in 1999, by a contract established between FUNASA and UNIFESP, where

many Indians occupy administrative posts (Rodrigues, 2005). UNIFESP has carried out a process

of training indigenous health agents since the 1980s, formalized in the 1990s, culminating with

the graduation of the first group of Indigenous Nurse Assistants in Brazil in 2001 (Mendonca,

2005, Rodrigues, 2005). These AIS agentss indigenas de saude, indigenous health agents) have

been hired as health professionals to work in their communities at Xingu Park by different

institutions, mainly by FUNASA (totaling 45 AIS working in Xingu Park in 2005), but also by

UNIFESP/EPM, by FUNAI and by some municipalities. In 2005, there were close to 80 indians

working as AIS or in administrative positions related to health services in Xingu Park (Oliveira,

2005).

The NGO Instituto Socioambiental1 (ISA), has been working with indigenous peoples at

Xingu Park since its creation in 1995. Counting on different sources of funding, from which the

Norwegian Rainforest Foundation (NRF) has been the most solid and continuous, ISA has


1 The creation of ISA in 1995 resulted from a fusion between the extinct CEDI (Centro Ecumenico de
Documentaqgo e Informaqio) the educational NGO Aqgo Educativa and some people coming from the
environmental NGO SOS Mata AtlAntica. ISA's mission is to coordinate research, action and public policy actions
to protect Brazilian's cultural and biological diversity. Currently, ISA develops programs and projects in different
Brazilian regions, and also works with public polices and documentation through its two main offices located in Sio
Paulo and Brasilia. More information can be found at www.socioambiental.org.









carried out projects for capacity building of local indigenous organizations; training of

indigenous teachers and supporting local schools; patroling of park's borders; developing

economic alternatives (especially honey and handicrafts production); natural resource

management activities; and lately, promoting dialogue between indigenous peoples and their

neighbors outside the Park through the Y Ikatu Xingu Campaign.2 ISA has worked in the south

and in the north portions of the Park, but it has concentrated most of its projects and activities

with peoples who occupy the middle and north sectors, such as the Ikpeng, the Kaiabi, the Yudj a

(Juruna) and the Kisedj (Suya) peoples. I will comment more on ISA's actions in the Park in the

following sections and chapters.

Besides FUNAI, UNIFESP/EPM, and ISA, other non-indigenous institutions working in

the Park are IBAMA (Instituto Nacional do Meio Ambiente e dos Recursos Naturais Renovaveis

- Brazilian environmental agency), which has promoted training for indigenous environmental

agents to patrol the Park's limits; SEDUC-MT (Secretaria Estadual de Educacgo do Estado do

Mato Grosso, Mato Grosso's educational board), which has assumed the administration of the

Park's educational program for teachers and schools; and some university researchers, linguists

and anthropologists who have worked on different subjects involving the Park's peoples.

ApiakA-Kaiabi Indigenous Land in Rio dos Peixes

Apiaka-Kaiabi indigenous land is located in Rio dos Peixes, northwestern Mato Grosso

State with 109,245 ha (ISA, 2008a). It takes approximately three hours by taxi or truck to travel

from Juara to the Tatuy village/post, and the distance is nearly 100 km. There is a small town

midway, named Aguas Claras, where there is a public school, restaurant, bar and people who sell

fuel.


2 www.yikatuxingu.org.br









In 2007, there were six Kaiabi villages in the area: the Tatuy Post and village and Novo

Horizonte village adjacent to the Post; Figueirinha village; and Itu, Vale Verde and Nova

Esperanca close to the Rio dos Peixes waterfall, which were created to protect the borders of the

area, after it was enlarged. There are around two to five families living in each of these small

groupings surrounding the waterfall. They have some infra-structure (radio, school, health

officer), assistance from FUNAI and the health district, as well as a road which connects them

with Juara municipality.

The Apiaka village named Mairob is located on the left margin of Rio dos Peixes, around

30 minutes by boat from the Tatuy Post, going downriver. There is also a road connecting the

Tatuy Post with the Mairob village. There are some Kaiabi married to Apiaka living in that

village. The Munduruku village is located upstream, before Tatuy Post and village.

The area is linked to FUNAI's regional administration (ADR) in Juina- MT, which

sporadically sends a diesel allowance to support activities of land patrolling. According to

Raimundo Kaiabi, local resident in Rio dos Peixes, patrolling activities are carried out by a group

of indians who patrol different sectors of the land three times a month, by car and by boat. The

health service is provided by FUNASA (Fundacgo Nacional de Saude), which keeps a house in

Juara to accommodate people who need treatment in the town. Various persons in the village

work as FUNASA health assistants and car drivers. There are two trucks to take sick people to

Juara. Catholic priests and nuns regularly visit the village. Priests and nuns from CIMI (Conselho

Missionario Indigenista) provide further assistance in the villages, continuing a tradition which

began with priest Jodo Dornstauder in the 1950s and 1960s. Often, the priest in charge celebrates

a mass on Sundays.









The Tatuy Post's school is linked to Juara municipality, and covers the first grades of the

elementary level (1st to 4th grades). After the completion of the elementary level, the students

have to leave the village to study in Aguas Claras or Juara. The Kaiabi already requested the

installation of facilities and hiring of teachers to provide education from 5th to 8th grades in the

village, and are awaiting a response from Juara's educational council. In 2007, there were 78

students registered in Tatuy's school. Esmeraldo and Cesarina, married, are the teachers who run

the school, but neither is fluent in the Kaiabi language. They told me that every Friday there is a

class for literacy in the Kaiabi language. Esmeraldo has participated in training courses in Xingu

Park twice, but after that he was not invited anymore. Cesarina has never participated in such

courses. Every day before class the students pray, according to Catholic habits. Students who

finish the first four grades of the elementary level at the village's school normally continue

studying in Aguas Claras. There is a school bus from Juara's municipality (in very bad

conditions) which transports the students to and from Aguas Claras every day.

In the Rio dos Peixes area, there is also a mix of subsistence (agriculture, fishing and

hunting) and market economic activities, but in comparison to Xingu, there is a preponderance of

market activities. The main economic activity is the sale of Brazil nut to middle-men (named

"castanheiros") who come to the village to buy the commodity, at very low prices. November is

the month when the Brazil nut production is higher and the price better. The "castanheiros" pay

R$ 1.50 (around US$ 0.75) per kg in the high season. After the harvest, the price begins to drop,

and according to local informants, reached R$ 0.80/kg in May of 2006. Nobody wants to sell

Brazil nut for such a low price; therefore they get involved in other activities for income

generation out of Brazil nut harvest season. The Kaiabi already have an order of 40,000 kg of

Brazil nuts for the next harvest. Other economic activities are the sale of crop plants (such as









maize, bananas, sweet potatoes and manioc) in Juara and the commercialization of handicrafts -

mostly clubs (men) and seed jewelry (women) to middlemen or to non-indigenous who visit the

village (see Chapter 7). Some of the village dwellers sell fish to others or in the market, and

others work in temporary jobs in neighboring ranches or in house construction in the villages.

There is also production of artisanal brooms made for sale in the town. They sell each broom in

Juara's market for about R$ 5,00 to 7,00 (around US$ 2.50-3.50).

Kayabi Indigenous Land in Teles Pires

Kayabi Indigenous Land is located on the border of Para and Mato Grosso (Jacareacanga

and Apiacas municipalities, respectively) states, with a declared extension of 1,053,000 ha. The

total population, including Munduruku and Apiaka groups, was around 640 people in 2006

(WWF, 2006). The location in which most Kaiabi live, in and around Kururuzinho Post and

village, is located in Para state, but the area under demarcation is located on the opposite bank of

the Teles Pires, in Mato Grosso state (see Chapter 4). In spite of physically residing in Para, the

Kaiabi maintain economic and political relationships with Mato Grosso state, specifically with

the town of Alta Floresta, in which the headquarters of their local organization Kawaip is

located. It takes some five to six hours by truck to travel from Alta Floresta, passing by Apiacas

municipality, to the margin of Sdo Benedito River after crossing the Teles Pires by ferry. From

there, it is necessary to take a boat and travel for three to five hours more before arriving in

Kururuzinho Post and village.

There are two indigenous posts in the area: PI Kururuzinho Kaiabi and the PI Teles

Pires, the old Kaiabi post, nowadays occupied mainly by Munduruku. There are around ten

villages, split between Kaiabi, Apiaka and Munduruku residents. According to data provided by

FUNAI (2007) and a recent report from WWF (2006), in 2006-2007 there were around 155

Kaiabi (FUNAI, 2007), 90 Apiaka and 420 Munduruku residents in the area. Some villages are









mixed Munduruku-Kaiabi or Apiaka-Kaiabi, in which the most common form of intermarriage is

an Apiaka or Munduruku male married to a Kaiabi woman. Remanso do Coelho and Siqueirinha

villages were established by Munduruku men (Vitorino and Albertino, respectively) both married

to Kaiabi women (Regina and Ines Kujdir6). They also have houses in the Kururuzinho Post

and village, and divide their time between the two locations. Minhocucu village is run by

Fernando Apiaka, married to Rosinha Kaiabi.

With the problem of land claims and demarcation, some Kaiabi established villages in the

area to be demarcated, as a strategy to show that the land is being occupied and in order to avoid

more invasions and land grabbing. For instance, Kurune Kaiabi recently established the Sao

Benedito village on the Sao Benedito river, Valdir (Kua'mi) created the Cachoeirinha village on

the Ximari river and Ze (Ywyrup) has built a house in the Dinossauro old village, where some

agricultural fields from other families are placed. Para Kaiabi came from Xingu to live in Teles

Pires with his family in 2007 and also established a new village close to Kururuzinho Post,

named Tucuma. There is also Mairowi village, whose residents are mostly Apiaka, and Sapezal,

which is Munduruku. In spite of the land being shared by Kaiabi, Apiaka and Munduruku

peoples, the residents have established boundaries for the piece of land and rivers that should be

used and respected by each group. As in Xingu and in Rio dos Peixes, the agreement regarding

each group's territorial limits is verbal and not represented in maps. However, sometimes there

are conflicts among the Kaiabi and other groups because of trespassing or not respecting the

verbal agreements.

The area is linked to FUNAI's regional administration (ADR) in Colider- MT. Cl6vis

Nunes is FUNAI's officer and chief of the Post (for more than 20 years), keeping residency in

Alta Floresta. According to Iracildo Munduruku, FUNAI's support in the form of resources or









gasoline is practically inexistent nowadays. According to Kaiabi residents' testimonies, the last

time their motor boat broke, a gold miner helped them to fix it, instead of FUNAI.

The health service is provided by FUNASA (Fundacgo Nacional de Saude), which keeps

a house in Colider to accommodate people who need treatment in the town. There is always a

non -indigenous nurse in Kururuzinho village, and once in a while dentists also visit the villages.

Jodo Kaiabi and Iracildo Munduruku are indigenous health agents (AIS) in the village, both

employed by FUNASA.

The village's school belongs to the municipality of Jacareacanga in Para, which makes its

administration difficult for the Kaiabi, since they interact much more with Alta Floresta

municipality in Mato Grosso. The school has stopped functioning for four years, and only in

October of 2007 was the school finally transferred to Mato Grosso state's administration. The

community was planning to have classes restarting in 2008. Meanwhile, the few waged workers

in the village usually have sent their kids to study and live in Alta Floresta.

Since 1999, the Kaiabi in Kururuzinho receive financial support from three fishing

resorts located in Teles Pires' tributaries: Thaimacu (www.thaimacu.com.br), Mantega and Santa

Rosa, this last one placed inside the land to be demarcated, and owned by the ex-mayor of Alta

Floresta town. They have contracts with these fishing resorts, which use part of the river that is

located inside the indigenous land for transportation of tourists. The resorts pay a monthly

stipend to the Kawaip Association (a percentage of their profit in each month), plus a quota of

gasoline and diesel. This money (according to Iracildo Munduruku, around R$ 4.000,00/month -

US$ 2,000) is administered by the Kawaip Association, and has been used to maintain the house

of the association in the city, their truck, and also to help out families in need. Some villagers get

involved in temporary work for the fishing resorts, working as boat drivers or in cleaning.









According to Ywyrup (Jose), they get around R$ 50,00 (US$ 25) for one day of work driving

boats for the fishing resorts.

Miners have also offered money and fuel (diesel and gasoline) to the Kaiabi so they can

mine for gold in the river surrounding the village (in spite of it being an illegal activity). There

was a big discussion going on in the village about the proposal brought up by one gold miner.

The village was divided about the question of whether they would authorize the miner to carry

out mining operations inside the area. In fact, this should, by law, be controlled by FUNAI and

IBAMA, and not by the Indians themselves.

Besides subsistence oriented activities such as agriculture, fishing, hunting and forest

extraction, other economic activities practiced by the Teles Pires Kaiabi are the sale of tucumd

and seed jewelry manufactured by the women in neighboring towns (mainly Alta Floresta) and to

tourists (especially fishermen who stay in fishing resorts) who occasionally visit the village. A

few men also sell clubs, baskets, bows, arrows and other handicrafts in the towns or to visitors.

Some residents collect Brazil nut to sell in the town of Alta Floresta. Fernando Apiaka, resident

of Minhocucu village, told me that he still sells cassava flour and Brazil nut in Alta Floresta.

Brazil nut's price is around R$ 9.00 a can, which yields around 15 kg, giving a unit price of

around R$ 1.50/kg (US$ 0.75), roughly the same as practiced in Rio dos Peixes.

The constant changes in land occupation which have happened to a greater or lesser

extent in the three areas occupied by the Kaiabi, accompanied by the movement between

villages, establishment of new villages and posts and interethnic marriages give us an idea of the

continued mobility of the Kaiabi people, even inside restricted areas, which is the case now,

post- transfer and post-state regulations. In Chapter 7, I will come back to this topic when talking

about exchanges and travels between the three areas.









Notes on the Demography of the Kaiabi

The demographic situation of the Kaiabi, in steep decline before and after the transfer to

Xingu, showed a strong reversal from the 1970's onwards (see Figure 6-3 and Table 6-1;

Pagliaro, 2005). The first census of Kaiabi population was done by priest Joao Dornstauder in

1955 (Grunberg, 2004). At that time, he estimated the population at 341 people, distributed in

several villages in Teles Pires (103 people), Rio dos Peixes (108 people) and some who had

already moved to Xingu (40 people), in addition to the residents of SPI posts such as Jose

Bezerra (45 people) and Kaiabi post (45 people). There were still other Kaiabi dispersed in the

rubber tapper landholdings in both regions and some at the indigenous post "Fraternidade"

(Catholic Church) in Barra dos Bugres (Melia, 1993).

During the 1960s, before the transfer to Xingu, many Kaiabi died of measles epidemics,

especially in the Teles Pires region, where access to health services was very precarious in

comparison to the Rio dos Peixes, where they had the support of Missdo Anchieta and priest

Joao Dornstauder. At that time, the decadence of SPI posts in Teles Pires caused the Kaiabi to

travel to the "Cururu Mission" in the Cururu River to exchange products. There, some of them

got infected with the measles virus and brought the epidemics to the villages (Rodrigues, 1994).

According to Joao and Jywatu, residents of Kururuzinho post and village, after the measles

epidemics, only four small villages remained in the Teles Pires area. With the transfer of the

majority of the group to Xingu, around thirty people remained in Teles Pires (Rodrigues, 1994).

In Rio dos Peixes, there were around 53 people remaining in 1966, after the main transfer events

to Xingu (Melia, 1993).

Between 1970 and 1999, Kaiabi population in Xingu increased from 204 to 758

inhabitants, meaning an absolute growth of 4.5% a year (Pagliaro, 2005). This rate is very high

when compared to the demographic rate of Brazilian population in general, which for the year









2000 was 1.64% (IBGE, 2000). This rate can also be considered high for indigenous populations

and it is higher than the demographic growth of other Xingu indigenous groups, which show an

average growth rate of 3.5% a year (Rodrigues, 2001). The Kaiabi are currently the most

numerous group in Xingu, with a population totaling around 1226 in 2006 (ISA, 2007). After the

1990s, the mortality levels stabilized at a lower plateau in comparison to previous decades. The

fecundity rate also increased in the 1990s, with each woman in reproductive life having an

average of 9.5 children. According to Heloisa Pagliaro, who studied the demography of the

Xingu Kaiabi (2002; 2005), the main factors causing this population growth are related to: better

health assistance service; the sheltered situation and relative isolation found in Xingu; and the

desire to recover their socio-cultural structure after the transfer.

Among the other two Kaiabi groups, a more modest tendency of population growth can

be observed, with the Rio dos Peixes Kaiabi totaling around 266 and the Teles Pires 155 in 2006.

It is worth mentioning that in Rio dos Peixes there are more interethnic marriages between the

Kaiabi and other groups in comparison to Teles Pires or Xingu. This, allied with other factors

such as access to better health services, as well as the migration back of some Kaiabi (especially

men) who were living in Xingu but returned to the ancestral areas, caused an increase in both

Rio dos Peixes and Teles Pires population, which is more significant in Rio dos Peixes.

According to the last estimate summing the three Kaiabi areas, they total 1647 in 2006, in

comparison to 341 in 1955. In 51 years, the population grew significantly. It is worth mentioning

that the sum of 1647 individuals does not include the Kaiabi living in the cities such as Canarana,

Juara or Alta Floresta, which is not insignificant, taking into account that many Kaiabi are now

working as health agents, FUNAI officers or officers of indigenous associations (mainly ATIX in

3 Which has been provided by the Escola Paulista de Medicina (EPM) and Universidade Federal de Sio Paulo
(UNIFESP) to the peoples of Xingu Indigenous Park since 1966 (see previous section on Xingu Park for details).









Xingu and Kawaip in Teles Pires), that have established offices in towns nearby the indigenous

lands.

The Political Empowerment of the Xingu Kaiabi

The socioeconomic and political changes brought by the transfer of the Kaiabi to Xingu have

had important implications not only for the Kaiabi, but also for other indigenous peoples living in

Xingu Park and, more broadly, for strategies for management and protection of indigenous lands in

the Amazon. In the last fifty years or so, there has been interplay between the dynamics of creation

and change in broader national political structures and opportunities, with the development of

indigenous grassroots movements and of strategic alliances with conservation institutions. The

Kaiabi have been both objects and subjects in this process, especially through the formation of

political leaders and the establishment of local organizations.

This section is devoted to the exploration of agents, events and processes which contributed

to the political empowerment of the Kaiabi, especially the Xingu group, in contrast to the other two

areas (Rio dos Peixes and Teles Pires). Through the personal testimony of Mairaw6 Kaiabi, an

important political leader, we can understand how he and the Xingu Kaiabi have been affected by

external agents (such as other indigenous leaders, NGO practitioners, and government officers),

connected to contexts and structures such as legal frameworks, the "projects market" and other

financial opportunities for Amazonian indigenous lands deriving from the conservation agenda.

Indigenous Grassroots Movements and Environmental Conservation in Brazil

After the military took over the Brazilian government in 1964, prospects for the

development of the Amazon were based on programs for geopolitical integration, demographic

migration for land occupation, and economic growth (Albert, 2005). According to Schmink and

Wood (1992) the model of development at that time (mostly between 1964 and 1985) relied on

political repression and on the centralization of power in the federal government's hands. The









wealth of untapped resources in the Amazon associated with sparsely populated areas made the

region a suitable place to attract investment capital and absorb surplus labor from other parts of

Brazil (Schmink and Woods, 1992). Struggles and competition over land and resources were

magnified, and included (until today in some regions) diverse actors such as the State itself;

ranchers; land grabbers; corporate enterprises (private sector); banks; logging and mining

companies; gold planners; small farmers; rubber tappers; landless peasants and, of course,

indigenous peoples (Schmink and Wood, op. cit.; Albert, op. cit.).

Brazilian military government instituted legal instruments to, at the same time, curb the

problem of indigenous identities and territories, and respond to media, civil society and

international donors' denunciations and requests. For instance, in 1966 the government signed

Convention 107 of the International Labor Organization (ILO) regarding indigenous peoples'

rights. One year later, the former SPI (Servigo de Protecgo ao Indio) was substituted by the

current FUNAI (National Indian Foundation) responding to international denunciations and

condemnation of SPI's actions towards indigenous peoples (Albert, 2005). In 1973, the military

government approved the "Indian Statute," a law concerning the rights of indigenous peoples

over land and based on the intent of assimilation of indigenous peoples in the Brazilian national

society or labor-force (see Chapter 3; Ribeiro, 1970). According to Albert (2005), this law has

been under revision by the Brazilian government with participation of indigenous leaders since

1991, in order to address current concerns (such as intellectual property rights, policies for

natural resource use and environmental protection) and also to bring some provisions and

regulations into conformity with the Constitution of 1988 (see below).

It was in this context that the formation of the indigenous grassroots movement occurred,

beginning in the 1970s, initially as a response to the Indian Statute's and government's policies









towards indigenous struggles for land and citizenship rights (Ramos, 1988; Albert, 2000). In the

beginning of the 1980s, the first pan-Brazilian indigenous organization was founded (UNI, Unido

das Nac6es Indigenas). Indigenous representatives and leadership, especially the Kayap6 people,

played an important role in the process of writing and approval of the Brazilian Constitution of

1988, which is considered a benchmark in the Latin American indigenous rights movement

(Ramos, 1994; ISA, 2009 a). In the Constitution of 1988, indigenous peoples were given both the

right to be culturally different, as well as the right over their lands. The struggle of indigenous

peoples to influence the text of the Constitution had the strategic support of institutions and

people such as the Catholic Church (through CIMI), NGOs, anthropologists, environmentalists

and indigenists (Ramos, 1998).

While indigenous leaders were getting involved and familiarised with formal Brazilian

politics though the participation in the planning and writing of the new constitution, the recently

resettled Kaiabi were adapting and getting acquainted with the novelty of living in Xingu Park.

The ability of some Kaiabi to relate with the non-indigenous and to speak Portuguese made them

important allies of the Villas-B8as brothers in Xingu Park's administration tasks. During the

1970s and 1980s, some of the men began to work for the Villas-B8as brothers, assisting in the

administration of the Park and helping to contact isolated indigenous groups, such as the Kayap6,

the Txicdo, the Panara and the Arara (Oakdale 1996; Menezes, 2000). Through their previous

contact with rubber tappers and their work with the Villas-B8as brothers, the Kaiabi acquired

valuable social skills. In addition to learning Portuguese, some men received formal education in

the FUNAI school at Diauarum Post. Working as boat drivers, assistants and, eventually, as

chiefs of villages and Posts, the Kaiabi began to gain increased control and influence over the

northern portion of Xingu Park (Lea, 1997). Some Kaiabi men became FUNAI officers (see next









section on Mairaw&s' testimony), living and working between Brasilia and Xingu Park (Oakdale

1996), or working with NGOs outside the Park. This contributed to the formation of several

important political leaders.

In parallel with the development of the indigenous grassroots movement in Brazil and the

formation of indigenous leaders, beginning in the 1960s and 1970s, the world witnessed the rise

of what was called the "environmentalist" movement, which gained real political force in the late

1980s (Little, 2001). For indigenous and other traditional peoples, it was an opportunity to link

human rights to environmental protection. As Keamey (1996: 107) highlighted, "sustainability is

viewed as a movement to the political aspirations of indigenous peoples to defend their cultural

and political autonomy against the designs of the modem nation-state to make them disappear

from history whether by assimilation or some other form of ethnocide. The global environmental

movement has thus become linked to the defense of human rights and the self-determination of

indigenous peoples in ways that perhaps no one foresaw before 1970."

The end of the 1980s and beginning of the 1990s was a moment of particular importance

for indigenous peoples in Brazil. The fortunate combination between indigenous grassroots

movements with the environmental movement during and after the approval of the Constitution

has had many social, economic and political impacts in indigenous lands in the Amazon. Inspired

by indigenous peoples' struggles, the rubber-tappers political movement reinforced the

importance to integrate human rights and social welfare with sound natural resource use in

alternatives for sustainable development of the region (Cunha and Almeida, 2000). The

Convention on Biological Diversity and the Agenda 21 approved in the 1992 Earth Summit in

Rio de Janeiro explicitly recognized the important role played by indigenous and traditional

communities in environmental conservation (Cunha and Almeida, 2000). A unique political









opportunity emerged, including participation in international political networks, for indigenous

peoples to access all kinds of resources to support their struggles, building and strengthening

transcommunity networks nationally and internationally (Yashar, 2005).

It was in this scenario that many international NGOs and governmental institutions began

to consider working with indigenous peoples to secure the world's largest tropical forest area. As

indigenous leaders became prominent in the international arena of environmental and human

rights activism, new forms of transnational and transcultural encounters and alliances emerged

(Conklin and Graham, 1995). However, along with the opportunities, resources and support that

came to indigenous hands through the new pact with environmentalism, indigenous peoples have

also gained a continued dependence on funding provided by these international organizations for

the development of local projects (Fisher, 1997; Albert, 2005).

One partnership that had a great effect on the Kaiabi and other peoples living in Xingu

Park began with the singer Sting teaming with Trudie Styler "in response to a direct request for

help from chief Raoni, a Kayap6 Indian leader in Brazil, seeking help in the fight to protect his

peoples' land and culture." (Rainforest Foundation US, 2009) This movement had international

repercussion and massive media coverage, triggering the creation of the Rainforest Foundation in

1989, which today has branches in US, UK and Norway. At that time, the Mektyktire Kayapo

(group commanded by Raoni) were still living in Xingu Park. In Rainforest Foundation's

mission statement, they refer to the intrinsic link between indigenous peoples and rainforest

conservation: "Our mission is to support Indigenous and traditional populations of the rainforest

in their efforts to protect their environment and fulfill their rights by assisting them." (Rainforest

Foundation US, 2009) Since 1995, the Norwegian Rainforest Foundation (NRF) has been one of

the main agencies giving financial support to the Kaiabi and other peoples in Xingu Park.









The Formation of an Indigenous Political Leader

Mairawe Kaiabi has been an important political leader among the Kaiabi. He was

president of ATIX (Associacgo Terra Indigena Xingu) for three consecutive mandates, he has

coordinated the project for monitoring Xingu Park's borders, and at present he is ATIX' vice-

president. I interviewed Mairawe on June 28 of 2007, and asked him to tell me how he became a

political leader. His history is intertwined with Xingu Park's history, and helps us to understand

the process of empowerment of the Xingu Kaiabi.

Anthropologist Berta Ribeiro met Mairawe in 1977, when he was 28 years old. She

described him as a "self made man" (Ribeiro, 1979:103) in contrast to Aritana Yawalapiti, a

prominent chief and political leader in the upper Xingu region: "both hold dialogue with the non-

indigenous and also with their own people; they circulate well in and between the two worlds".

While Aritana is an aristocrat, coming from an important family of chiefs, Mairawe is an

ordinary man who conquered the place he holds by his own effort and means.

According to Mairawe's testimony, when he moved from the village in which he used to

live with his parents in Arraias River to the Diauarum Post, he was around eight years old. He

became friends with a white boy and began to help the Villas-B6as brothers with the work in the

kitchen. He would carry water for them, and take care of their bird. The younger Villas-B6as

brother, Alvaro, began to teach him to read and write in Portuguese. His father had already died

and his mother always asked him not to travel away from the Post, away from her. Once he was

invited to go on a boat trip to Leonardo Post (around nine hours by boat from the Diauarum Post)

and the driver told him he could come back with him the next day, but this was not true, and he

ended up staying there for a couple of months. His mother was sick and died before he was able

to see her again.









He came to Diauarum but went back to Leonardo Post to continue working with Alvaro.

He would participate in every meeting that the Villas-B8as held with the non-indigenous: "I was

in every meeting that Orlando and Claudio had with the whites, listening, participating. They

talked a lot about land titling around the Park; I did not know what that meant .... Then, in

1962, a group left, commanded by Prepori, to bring another group of Kaiabi coming from Teles

Pires ... here in Diauarum it was full of Kaiabi. The first Kaiabi village was opened by Prepori

in the site where Capivara village is located today. Then, they began to build other villages, they

began to spread out. The Villas-B8as wanted the Kaiabi close to Diauarum. They were worried

about people leaving to visit the cities; they did not like it. I wanted to work .. then they called

me to work with transportation at Leonardo again. I led an expedition from Leonardo to Garapu,

about 30 days of travel. After that I would already know how to run and fix boat engines. I

stayed there for 3-4 years. When it was 1966, another group of Kaiabi arrived from Tatuy:

Temeioni, Tapa, Domingos, Takaperun, Tewit, bringing the kids. Then I came to Diauarum to

help to take care of them. I ended up staying here and getting married to a Suya woman. Claudio

was organizing the expedition to contact the Panara people, and he asked me to take care of the

Post (Diauarum) while he was gone. Then it started, everything was on me. A lot of

responsibility, even applying injections in people. I did not get paid to work. Only in 1973 was I

hired by FUNAI. There was way more fish in the river at that time compared with nowadays ...

then I stayed here, right? Then the political part began, I had to do everything: health, visitors,

politics. It was easier to talk to my people, the Indians. I had more difficulty with the whites. It

was good, because today I can see that I did a good job for the Indians and whites alike."

At the time that Mairawe was working as the chief of Diauarum Post, FUNAI had begun

to hire indians as officers, but there was no indian hired as a Post chief. Mairawe was probably









the first in Brazil. There were many Indians being hired to work in the construction of the

Transamazon road. A FUNAI airplane pilot became Mairawe's friend and helped him with the

formal hiring process and contract. From 1973 to 1985 he worked as the chief of Diauarum Post:

"I stayed here commanding the post....then I thought that I needed people to help me, Tapai6,

Atu, Tangue, Pai6, Tuim, Ip6. I helped them to get hired as officers also. I already had my

experience, so for them it was easier."

According to Mairawe's accounting, in 1984 a conflict between some indigenous

peoples from Xingu Park and the Brazilian government over the control of the crossing of the

Xingu river by ferry on the route of the road BR-080 (which was working since 1970) had

important consequences for the Kayap6 people and for political leaders who participated in this

process alike. The so-called "Guerra da Balsa" (Ferry war) lasted about two months. The Kayap6

seized and held the big boat which was used to cross cars and trucks from one bank of the Xingu

River to the other. With the delimitation of the Xingu Park, areas inhabited by the Mektyktire

Kayap6, known as "Jarind" and "Pururi" (where the important Kayap6 chief Raoni used to live

with his group), located to the north of the BR-080, were left out of the Park's area. The Villas-

B8as brothers wanted the Mektutyre Kayap6 to leave those areas and move inside the Park.

Raoni agreed and moved to the village called "Kretire", nowadays a Kaiabi village named

"Caicara." But two other important Kayap6 leaders refused to leave their lands and initiated a

movement for the demarcation of their territories and to gain control over the crossing of the BR-

080 road. At that time, there were a lot of conflicts in the crossing of the BR-080, with frequent

assassinations of Indians and non Indians alike. The Kayap6 kept the boat and FUNAI officers

captive, including Claudio Villas-B8as. The Kayap6 asked for help from other indigenous

groups.









Ip6 Kaiabi (deceased), who was a FUNAI officer in Xingu, gave his testimony about this

incident to Mariana Ferreira in 1984: "It was hard, but beautiful to see all the Xingu tribes

fighting together, everybody united. Everybody helped: the Suya, the Kaiabi, the Txucarramae

[Kayap6], the Txicao [Ikpeng] and the Krenakore [Panara]. Now, the Park's director is our

relative. The Park's director is Megaron. He understands the life of the Indians, because he is

also an Indian. This is good for us. In this way, Xingu Park, which has been full of 'caraibas'

(non-indigenous), is slowly becoming a place where the Indians can solve their own problems."

(Ferreira, 1992:210)

Therefore, the indians had a proposal to present to the government in Brasilia, Brazil's

Federal District. A group of political leaders, Mairaw6 among them, went to talk to the Minister

of the Interior, Mario Andreazza. It was Mairaw' s first experience of negotiating with the

government. They eventually reached a middle-ground in the negotiations, which involved the

resignation of FUNAI's president (at that time, Otavio Ferreira Lima) and the placement of

Megaron Mektyktire as the first indigenous administrator of Xingu Park in 1985 (Lea, 1997). In

1991, the Kayap6 gained official recognition of part of their traditional territory, with the

homologation of the Capoto-Jarina indigenous land (adjacent to Xingu Park), with 634.915 ha

(ISA, 2008 a).

Mairaw6 revealed how this incident definitively influenced his political life, as a process

of learning from other leaders' experiences, as he says, with his "indian professors." He told me

that "during these stories, these meetings, these hard moments, many stories, memories come

alive, what happened with each people, not only Kayap6, but also Kaiabi, Suya, we reveal

everything that happened with our people, what happened, how they suffered." Then, Claudio









told us: "You have capacity; you conducted the war without anybody's influence. Maybe I

would not have done what you did. So now, I want an Indian as the Park's administrator."

Megaron Kayap6, the new Xingu Park administrator, asked Mairawe to work as his

assistant in Brasilia, where he lived for about five years, returning to Xingu to work as chief of

Diauarum Post, for five years more. It was then around 1995, when a group of Xingu leaders,

assisted by indigenists and by non-indigenous organizations, decided to create the indigenous

organization ATIX (Associacgo Terra Indigena Xingu). Mairawe was president of ATIX during

three consecutive mandates, from 1995 until 2004, nine years. Nowadays, he still works as a

political leader and is ATIX' vice-president.

The social, cultural and political roles played by Kaiabi leaders in the present may be

compared to those of, in a not so distant past, warriors, and in a mythical past, to those of Kaiabi

mythical heroes. The warrior figure has had a very important place in Kaiabi culture, for it was

during warfare that the warriors learned things from the enemies and used to bring goods,

women and children to the villages (Travassos, 1984; Grunberg, 2004). It seems that, at least

partially, the role of the warrior is being replaced with that of the political leader. The leaders

travel and build bridges between worlds, fight for their security, bring goods and knowledge, are

brave, strong and restless. Their main weapon is the word, spoken and written.

The process of identity construction among the Xingu Kaiabi has followed paths

connected to their specific history and cosmology and with the cosmology of the Tupi-Guarani in

a broader sense. As Viveiros de Castro (1992:3) wrote about the Arawete, and Oakdale

(1996:14) emphasized for the Kaiabi, it seems that in Tupi-Guarani societies: the center is

outside, its identity is elsewhere and the "other" is not a mirror for man, but his destiny. In the

myth of creation of Kaiabi people, the ancestral hero Tuiarare, who taught the Kaiabi everything









they know, travelled all the time, interacting with other peoples, marrying a woman from another

tribe and ending up killing their enemies commanded by his brother-in-law (ATIX, 2006 a). He

was a great warrior and shaman and at the same time a great basket maker. Interestingly enough,

in coming to the Xingu river in mythical times, Tuiarare discovered some natural resources that

are used by Kaiabi people today.

The enemies are often celebrated and portrayed as a source of knowledge to be learned,

adapted and incorporated as their own. The process of "learning from the enemy' identified by

Viveiros de Castro (1986; 1992) for other Tupi-guarani societies, happens historically in many

spheres of Kaiabi culture and social life, and can be interpreted through the long trajectory of the

Kaiabi from the ancestral land to Xingu Park. The Kaiabi have shown a special ability to learn

and adapt knowledge, skills, concepts and even institutions of the western society according to

their own interests (Senra, 2004). I concur with Gray (1997:198), referring to the Arakmbut

people from Peru, when he says that: "any future development work with the Arakmbut therefore

has to combine a re-strengthening of the old ways with the introduction of the new. In this way

they will be able to blend their own cultural identity with the outside power of the Amiko

(whites)." Mairawe is a leader who shows the capacity to adapt and translate non-indigenous

knowledge and political systems to Kaiabi reality.

In the next section, I will explain the process of creation of ATIX, and the role that

Mairawe along with other indigenous leaders has had in managing and sustaining this new

institution, as they say, "borrowed" from the non indians.

The Creation and Sustaining of ATIX

After the Constitution of 1988, indigenous organizations flourished in Brazil due to the

possibility of constituting legally recognized civil society organizations. This means that they









could now be legal representatives and defenders of their own rights. According to Albert

(2000), other factors that were responsible for the "boom" of indigenous organizations after the

Constitution were the decentralization of international funding for human rights and

conservation; the alliances between indigenous peoples and environmentalists; the multiplication

of social-environmental national and international NGOs; and the opening of the "projects

market" after the 1992 Rio Earth Summit with the concept of sustainable development through

local action. Brazilian indigenous organizations have proved to be critical as agents for dialogue

and negotiation between indigenous peoples and governmental and non-governmental

institutions, in mobilizing forces against construction of dams and other large-scale development

projects, in gaining autonomy for territorial management, and in accessing financial and political

support for their projects (Albert, 2005). In 2000, there were around 300 indigenous

organizations in the Brazilian Amazon (Albert, 2000).

In 1994, the Fundaq(o Mata Virgem, a Brazilian branch of the international

environmental NGO Rainforest Foundation, began to work directly with the Kaiabi, Yudja, Suya

and Ikpeng peoples in Xingu Park, developing a series of educational and development projects

which were subsequently adopted by the Brazilian NGO Instituto Socioambiental (ISA). In 1995,

with the support of ISA, UNIFESP and other institutions, a group of Kaiabi, Yudj a and Suya

created the indigenous organization ATIX (Associacgo Terra Indigena Xingu Xingu Indigenous

Land Association), whose purpose is to represent and organize the indigenous communities in

the northern portion of the park, advancing human rights and developing projects of common

interest. In spite of its multi-ethnic composition its council includes representatives from all

indigenous peoples in the Park- ATIX has been historically dominated by the Kaiabi, with

Mairaw6's family playing a prominent role in the organization's growth and administration.









Through ATIX, the Xingu Kaiabi have been actively involved in the Park's territorial

management, including a diverse range of activities such as: running their health and education

services; coordinating the patroling of the Park's borders; organizing land rights claims; and

implementing development projects, including cultural revitalisation, natural resource

management and conservation activities. These projects have been carried out in partnership with

ISA and other NGOs, as well as with governmental institutions, and have been funded by

national and international institutions. Through ATIX and these processes, indigenous people in

the park have acquired a degree of political autonomy from FUNAI and the state.

The task of assimilating a new institution which would serve the Indians' interests, but at

the same time would bring another type of social-political organization to Kaiabi communities in

Xingu, has not been an easy one. Nor has it been easy to carry out so many activities and

projects. ATIX was the first indigenous organisation in Xingu Park to be formally created, which

has endured more than ten years in activity and growth. This is in great part the consequence of

first, the continuous financial support from NRF (Norwegian Rainforest Foundation) during

ATIX's fourteen years of existence, and second, the close assistance and training provided by

ISA during all these years (enabled by continued funding provided by NRF and ISA's

commitment). In other words, continued access to technical and financial support is an important

factor in ATIX's successful trajectory. However, access to resources per se would not suffice to

guarantee the sustaining of the organization through time. The leadership and dedication of

Mairaw6 Kaiabi and his family have been crucial in the organization's performance and

endurance. A multicultural environment also provides a stronger social control of the territory.

ATIX's political council composed of representatives of the 14 groups that live in the Park has

had an important role in promoting discussions and negotiations about any operation or proposal









that can influence peoples or natural resources inside the Park. Related to this, Langer (2003)

observes that "in almost all cases the strength of the indigenous movement has come from the

alliance of separate ethnic groups."

In my view, another factor that has contributed to the organization's empowerment (and

by extension, to the empowerment of the Xingu Kaiabi) vis-a-vis the other ethnic groups living

in the Park is the fact that ATIX undertook the hard mission to coordinate the monitoring of the

Park's borders, which was its first autonomous project. I was in Diauarum Post when the

deceased former FUNAI president Sullivan Silvestre signed the first official agreement between

FUNAI and an indigenous organisation, delegating to ATIX powers to carry out work which was

previously (and officially) coordinated by FUNAI and the Brazilian Federal Police. Mairaw6

was coordinator of this project for many years, having promoted innumerous meetings and

expeditions involving all the fourteen indigenous groups living in the Park. This coordination of

activities to protect a common territory or patrimony has contributed to confer visibility and

power to ATIX (Revista Veja, 30/06/1999). Mairaw6 and his team have also been successful in

establishing partnerships with other organizations, besides ISA, in the development of this

project, such as the national environmental agency IBAMA (see Chapter 5). ATIX's Borders

Monitoring Project has been used as a model for territorial protection of other indigenous lands

in the Amazon. Nevertheless, it is worth mentioning that the project has undergone many

difficulties and obstacles, such as bureaucratic problems in renewing FUNAI's contract, lack of

organization of chiefs of monitoring posts, pressures for exploration of resources from loggers,

ranchers and other park neighbors, among other difficulties.

I have had the opportunity to accompany the development of the association since 1997,

only two years after its creation until 2007, the last time I visited the Park. In the beginning, a









great part of the efforts were destined to promote capacity building activities for ATIX's

personnel. Controlling ATIX's many accounts and spreadsheets has been a challenge for the few

Kaiabi trained in those skills (only two people), who tend to alternate the function of financial

director between each other through successive mandates. In the beginning, many youth wanted

to be part of ATIX's team. However, after some years, people realized that the small salary or

gratitude received for their work in ATIX does not compensate for the dedication and effort

expected from them in the association's work. This problem which also happens among

unwaged teachers, beekeepers and other community workers has been the reason few people

are willing to work for the organisation. Who is going to fish, hunt and plant for their families?

Do they have to rely on industrialized food now that they are ATIX's officers? Also, becoming

an ATIX officer means establishing residence in Diauarum Post, where the soil is overexploited

and fish are (compared to other villages) more scarce. Therefore, people who get involved in

ATIX's work are mostly sons of FUNAI officers who already reside in Diauarum Post, or, more

rarely, young Kaiabi full of ideology and willingness to become a political leader.

Today, ATIX has its headquarters in the Diauarum post and keeps an office in the nearby

town of Canarana MT, both equipped with computers, printers, GPS, office supplies, maps,

publications and other materials. ATIX' annual agenda is overwhelming with activities, meetings

and travels. Frequently, political leaders such as Mairaw6 are invited to participate in political

meetings at national and international levels. ATIX is a member of broader indigenous

organizations and networks, such as COIAB at the national level (Coordenacgo das

Organizac6es Indigenas da Amaz6nia Brasileira, Coordination of Indigenous Organizations of

the Brazilian Amazon) and COICA at the international level (Coordinadora de las

Organizaciones Indigenas de la Cuenca Amaz6nica, Coordination of Indigenous Organizations









of the Amazon Basin). This overflow of activities brings a trade-off between being a political

leader and an ordinary Kaiabi person. In other words, sometimes the leader is regarded as a

"guardian" of a culture that he cannot pursue. For instance, many (with exceptions) political

leaders don't know how to weave baskets, which is Mairaw6's case. The type of lifestyle

demanded by their activity prevents them from spending more time with their family, doing, as

Mairaw6 is used to saying, "indian things" (coisas de indio) such as fishing, hunting, planting,

weaving and others.

The Norwegian Rainforest Foundation (RFN) has a program for building up networks

between the projects funded by them. Through this program, ATIX's representatives have had

the opportunity to interact with other indigenous peoples and organizations in the Amazon,

strengthening their "transcommunity networks" and their capacity to understand the Brazilian

political scenario towards indigenous peoples and how their distant relatives have responded to it

(Yashar, 2005). ISA and RFN have worked with the premise of a gradual transfer of power to

ATIX through time. ATIX's personnel now administrator part of the funds that were originally

managed by ISA, and have also taken on the responsibility of coordinating, executing and

accounting for projects, such as Borders Monitoring and Honeybee Production. In the last years,

ATIX has been fully responsible for other projects such as infra-structure (construction of wells

in villages), in partnership with FUNASA, and community-based cultural revitalization projects,

such as the Munuwi and the Kaiabi Araa (see next sections). It is worth mentioning that in spite

of the Kaiabi people's wish to achieve full autonomy to run projects and activities, there has

been continued technical assistance (to a greater or lesser extent) from ISA's personnel to ATIX

in the administration of these projects.









ATIX's ascendance and success, along with interethnic conflicts, have encouraged other

Xingu groups to create their own associations. Despite ATIX's multiethnic composition since its

creation, counting on a political council represented by other Xingu groups, the management of

such a multi-ethnic institution is very hard on the ground, taking into account the image of peace,

timelessness, purity and lack of conflict created for Xingu Park (Menezes, 2000; Oakdale, 2005)

exists only in TV series and postal cards. In reality, ATIX has been viewed by other Xingu

peoples as a pan-indian association, but whose ethnic essence is Kaiabi. Furthermore, ATIX has

been viewed as dealing with problems and projects at a broader scale, and therefore incapable of

giving to the village's local communities the attention and response to demands they would like.

This boom of creation of local associations is at the same time good for ATIX, because it

decentralizes and takes pressure off the association's shoulders, but on the other hand it can get

out of control, since every single village and people in Xingu might want (and of course has the

right to) have their own organization, not knowing what exactly this means and how to manage

it. For instance, the Yudja created their own association in 2001 (Yarikayu), which is basically

dealing with cultural projects, and the Kis6dj 6 created their association in 2005 (Associacgo

Indigena KiTsdj AIK). Among the Kaiabi, Tuiarare villagers, commanded by chief and teacher

Aturi (Jowosipep), also decided to create their own Association (Associacgo Tapawia), in 2005.

The Kaiabi have shown a great capacity to frame their identity through the appropriation

of concepts and discourses produced by western science in their social organization, self-

determination and political struggle (Sahlins, 1997; Hornborg, 1998; Kurkiala, 1998; Oakdale,

2004; 2005; Senra, 2004). Many authors have referred to the capacity of indigenous peoples to

construct self-representations based on externally constructed notions and concepts such as

culture, Indian identity, and the noble-savage debate as one means through which they can access









the resources they need to maintain their own reproduction (Colchester, 1994; Fisher, 1994;

Jackson, 1991, 1995; Conklin, 1997, 2002; Conklin and Graham 1995; Turner, 1991, 2000;

Langer, 2003; Albert, 2005; Oakdale, 2004; 2005). Albert (2005:200) refers to the process of

"adaptive resistance" in which indigenous peoples' identities and mechanisms for social and

cultural reproduction come to depend "as much on repertoires of legitimation imposed by

developing states and advocacy organizations as on their own political-symbolic resources."

This adaptive resistance process might bring both innovations and resources but also

contradictions and conflicts to indigenous peoples, as is the case with the Kaiabi. The territorial

confinement and the cultural changes and paradoxes which they have had to face with the

transfer to Xingu have produced shifts in the way they present themselves to other peoples and

external agents (Oakdale, 1996; 2005). Suzanne Oakdale (1996), in her dissertation about the

Xingu Kaiabi, explores the conformity of the discourse (and practices) of some Kaiabi men to

the pre-conceived and established view of Xingu Park as an untouched paradise where Indians

and nature will live forever in harmony. According to her accounting (Oakdale, 1996:31), once

one Kaiabi man and political leader "told his community that they should not mention that they

kill animals or eat meat when representatives of an international NGO came to visit. He was

concerned that animal-killing, meat-eating Indians would not fit the image of original purity that

has been mapped on to the Park, and that the Kaiabi would be seen as already too impure, too

fully inserted into the process of history to merit foreign capital." At the same time, the Kaiabi

have engaged in several processes of "cultural revitalization" which might be partially related to

the fact that they are regarded as acculturatedd" and caboclized by their upper-Xingu neighbours.

In fact, the Kaiabi have appropriated the notion of culture as an instrument, or as an object, as

Senra (2004) pointed out. They have re-worked the concept of culture as an ideal(ized) set of









knowledge and traits which should be pursued by any ordinary person, so the person would

"have culture" (Senra, 2004). Under such conceptions, there has been a clash between older and

younger generations and between what is and what is not the authentic "Kaiabi culture." is

(Oakdale, 1996; 2004)

If on one side the assimilation of western institutions causes local conflict, on the other

the political power and economic benefits of long term international funding can provide the

structural basis for social strengthening through cultural "revival" initiatives. Amongst the

Kayap6, Turner (1991) highlighted the culturally revitalising effects of political activism and

increased power in strengthening self-awareness and pride in their traditions. The Kaiabi have

not only dealt with assimilation and appropriation of discourses and institutions in the

configuration of their Xingu identity and empowerment, but also with western modes of

knowledge transmission such as schools and projects, which is the topic of the next section.

Community-based Projects and Initiatives

The dramatic increase of indigenous organizations in the Brazilian Amazon region after

the Constitution of 1988 was accompanied by the need for financial, technical and political

resources to sustain these new local institutions. According to Albert (2004), the majority of

these organizations have relied on external resources allocated in specific projects designed for

local development such as territorial management; institutional maintenance; organization of

political events and assemblies; health and education programs; initiatives for commercialization

of products; and initiatives for cultural revitalization and publicity.

This boom of indigenous organizations and the opening of what Bruce Albert named the

"projects market" has happened in association with favorable national and international policies

and facts. In the international arena, the main factors which contributed to the establishment of

the "projects market" through local initiatives were the globalization of issues related to the









environment and human rights, allied to the expansion of NGOs willing to work with local

peoples in the implementation of conservation and development projects, also known as ICDPs -

Integrated Conservation and Development Projects (Wells, Brandon et al., 1992). The

decentralization of international cooperation and the political movement around the conservation

of the Amazon provided a flow of information, financial resources and institutions willing to

work locally with traditional communities and indigenous peoples with the aim of promoting

sustainable development, attempting to reconcile community development with sound natural

resource management (Albert, 2000). Nationally, in spite of the capability given to the

organizations of becoming official civil society entities, the State has, in many circumstances,

distanced itself from the responsibilities linked to the management of indigenous lands in the

country. This deccurs in part of budget cuts in the indigenist administration initiated by the

military government.

In Xingu, development and conservation initiatives have also been filtered through the

access and appropriation of resources coming through projects. This is true not only for ATIX,

but also for other indigenous organizations. The Indians have had to learn the project language,

get trained in computer programs and accounting, keep good communication with partners and

donors, and prepare various types of documents as required by their funding institutions. They

have had to learn that working with projects means more than preparing lists of items with a

budget. This process has been a great challenge for the few ATIX officers trained to work with

computers. In the next sections, I describe some of the first community-based projects developed

by or through ATIX4 for cultural revitalization, including the "Kaiabi Araa" project related to

basketry and textiles weaving knowledge, whose results are further explored in Chapter 8. In the


4 Currently there are other projects for cultural revitalization in development though ATIX, but with which I have no
familiarity.









end, I present a brief discussion on the benefits and disadvantages faced by ATIX through its

engagement in the projects market in the last ten years.

The KumanA Project

Since its creation in 1995, ATIX has implemented several projects, funded by

governmental and non-governmental, national and international institutions. According to

Mairaw6 Kaiabi, one of the main objectives of the organization has been to promote cultural

awareness/revitalization activities for Xingu peoples. In other words, they were creating an

institution borrowed from the non-indigenous to preserve their indigenousness. The first project

envisioned by the Kaiabi, initiated in 1997, was named "Kumana Project" (kumand=type of

bean), and had the aim of promoting cultural revitalization activities related to hadicraft

production. The goal was to implement a series of "Culture Schools" in several villages, to

constitute spaces through which the elders would teach the youth to make traditional handicrafts.

The objects produced in each school of culture would then be evaluated and sold, in order to

generate cash to sustain the functioning of the schools and the visits of experienced artisans to

each village to conduct handicraft workshops and evaluation of the school's work. The idea was

good, but the resources weren't sufficient to finance all the activities proposed in the project,

which to complicate things further, included three different peoples, the Kaiabi, the Yudja and

the Kisedj6. However, it lasted one year, and in some villages it had a significant effect. For

instance, in 1998, in a period of four months, the families of Kururu (Kaiabi) village (nowadays

named Aipore), produced various handicraft items worth around R$ 5.000,00 (US$ 2,500) in the

school of culture named "Panakii" (a type of basket, see Chapter 8). Part of this production was

sold directly to middlemen who used to come to the village to buy, and the rest was sold through

ATIX in its first experience of selling handicrafts directly to stores in Sao Paulo (Athayde, 1998;

1999).









I was able to observe that the Kumana project had an impact in the cultural revitalization

of some items of Kaiabi material culture which were not being produced regularly, such as

necklaces of inaj (palm, Maximiliana maripa) produced by women among others. On the other

hand, it also enabled creativity and innovations, such as the carved wooden benches traditionally

produced by the men, which came to be painted with graphic designs adapted from their rich

repertoire of basketry designs. Besides this, the possibility to exchange art for cash was a good

incentive for the youth to learn how to produce traditional handicrafts with the elders.

The Kaiabi Araa Project

In 2004, two local communities began to develop "cultural revitalisation" community-

based projects through ATIX. They both had their projects approved by the PDPI program

(Projetos Demonstrativos dos Povos Indigenas/PPG7), which is concerned with indigenous

capacity building, to plan, write and run their own projects with minimal external assistance.

In Tuiarare village, Aturi (Jowosipep) conceptualised and coordinated the Kaiabi Araa5

project, mentioned previously in the introduction. In Kwaruj a village, Aturi's brother-in-law

named Tuiarajup conceptualized and coordinated the Munuwi project (next section). This was a

project for the revitalization of basketry and textile weaving knowledge that also included the

Kaiabi from Para (Teles Pires). The project, which lasted seven years from its conception until

its conclusion, was financed by PDPI and executed through ATIX. The total cost of the project

was R$ 108.523,90 (around US$ 50,000). The main activities carried out through the execution

of the project from 2004 to 2006 were: 1) workshops to teach basketry and textiles weaving to

men and women in Teles Pires (Kururuzinho village) and at Xingu Park (Tuiarare village); 2)

activities for management of the "arumd" plant, including transplanting of saplings from Teles


5 The word Araa means graphic design in the Kaiabi language.









Pires area to Xingu park6; 3) visit to two ethnographic museums located in Goiania (GO), with a

survey of Kaiabi objects, as part of an initiative of surveying Kaiabi ethnographic collections in

national and international museums; 4) workshops in the Tuiarare village school to produce

educational books and materials, such as the book on Kaiabi basketry and the book on Kaiabi

textiles (ATIX, 2006 a; 2006 b); 5) production of a video documentary project7; and 6)

construction of a traditional house named the "Culture School" in Tuiarare village (ATIX, 2006

c; Kaiabi and Kaiabi, 2006).

The Kaiabi Araa project included four weaving workshops, with the participation of 12

basketry teachers, 15 textiles teachers, 63 basketry students (males) and 48 textile students

(females), and the production of 76 designed baskets, besides other items (ATIX, 2006 c, Table

6-2). The project was one of the winners of the Indigenous Cultures Award 2007 Ministry of

Culture/Brazil (Premio Culturas Indigenas Edigco Angelo Cret. Ministerio da Cultura -

MINC), and the Tuiarare community received around US$ 7,500 to be used in cultural

revitalization initiatives. It is worth mentioning that the practical workshops which happened in

the villages were entirely organized by the communities and had no previous structural

arrangements, when compared to the workshops normally organized by ISA or EPM in the

villages. Each person would arrive and begin to work with his or her teacher, surrounded by

children, women breastfeeding while weaving and young men listening to music in their

headphones. In addition to the twill-plaited baskets with graphic designs, men and youth

participating in the project also learned to make different types of baskets, such as the cylindrical

basket named "Tamakari" and the clubs with woven handles (see Chapter 8). During one


6 Ischnosiphon gracilis (Marantaceae), main fiber used for production of baskets with graphic designs whose
populations are scarce in Xingu Park. See chapters 5 and 8 for more information.
7 See excerpt of Kaiabi Araa documentary on: http://www.youtube.com/watch?v=LRPbH4IvwZA









workshop at Tuiarare village, an experienced elder (named Karauu) produced another type of

cylindrical basket (Yriokote 'em) which according to the participants, nobody had seen or done

before. An analysis of the effects of this project in knowledge transmission and cultural

revitalization among the Tuiarare and Kururuzinho villages is presented in Chapter 8.

The project was considered to be very successful by the participant communities and by

the Kaiabi in general, taking into account that it was one of the first village projects developed

through ATIX. However, there were many obstacles and difficulties to be overcome in the

process of its development. One of them relates to the difficulty of communication between the

coordinators (Aturi in Xingu and Eroit in Teles Pires) and ATIX's team in Canarana. ATIX was

supposed to send monthly accounting reports to PDPI, in order to be able to receive the next

deposit in the project's bank account. They had difficulties in communicating to each other,

preparing the account reports, keeping the receipts of all expenses, and sending the reports to

PDPI on time. ATIX's financial director at that time (Tariajup Kaiabi) was overloaded with all

the other ATIX activities. ISA's personnel and I helped ATIX's team and villagers in many

situations in order to keep the project going in good shape. Another difficulty encountered was

the lack of raw material (arumd) to be used in the workshops, which was addressed through the

use of other plants as substitutes. Participants also mentioned in the evaluation that they needed

more time to really learn how to make the baskets and textiles objects taught through the project.

However, the cost of each workshop was very high, considering that the project paid for

industrialized food bought in the cities (there weren't enough fish and agricultural products to

sustain nearly 40 visitors in a village for 10 days) and also for each person's work in the

workshop (such as fishermen, cooks, and boat drivers). See the following section with a brief









critique on the "projects culture" for further analysis on the constraints of relying on them for

achieving cultural resilience among the Kaiabi.

The Munuwi Project

Another community-based project developed with PDPI funds and executed through

ATIX was conceived by the chief of Kwaruja village, Tuiarajup Kaiabi (nicknamed Tuiat). He is

the son of Prepori (see Chapter 3), the charismatic leader and shaman who played a decisive role

in the transfer of the group to Xingu Park. He was also responsible, along with other elders, for

the transport of most of the Kaiabi crop varieties from Rio dos Peixes to Xingu Park, on foot

(Silva and Athayde, 2002). After that, he turned out to be a great shaman, being able to

communicate with the Mait (spirits from heaven), curing diseases and making predictions.

Tuiajarup, today nearly 45 years old, followed the steps of his father also becoming a shaman.

Since the death of his father in 2000, Tuiat along with his wife Wici6, his brother Arupajup and

his nephew Sirawan (teacher) began to carry out activities for the recuperation, multiplication

and exchange of crop plant varieties between the Kaiabi at Xingu Park, with a focus on peanut

varieties. These efforts ended up in the PDPI project named "Munuwi" (in Kaiabi, peanut), with

"in situ" management of more than 30 peanut varieties that constitute one of the most important

element of Kaiabi cultural and sacred patrimony (Silva, 2002 a). Women from other villages

come to plant and harvest peanut seeds planted in special fertile land plots. The seeds are then

stored and distributed to other villages. Geraldo Silva, an agronomist who also worked for ISA

several years, has accompanied and advised the Kaiabi in the planning and execution of this

project since its inception. His PhD research includes an analysis of peanut management and

conservation among the Kaiabi (Silva, 2009). Tuiat has travelled to different places in and out of

Brazil to publicize his project. He became nationally known for his efforts in agrobiodiversity

conservation through a documentary exhibited by the television network "Rede Globo" in 2003.









Benefits and Disadvantages of Community-based Projects in Xingu

In the Amazon, community-based initiatives for conservation and development among

traditional and indigenous peoples mostly have been carried out through the so called projects

market or "Integrated Conservation and Development Projects" ICDPs (Wells, Brandon et al.,

1992; Albert, 2000; 2005). In fact, this market sustains not only communities and their local

institutions, but contributes substantially to the budget of NGOs (for many of them it is the main

source of funding) and governmental agencies alike. This entails a complex system of formal and

informal social and institutional interactions, or social networks, which range from local to

regional, national and global scales (Agrawal and Gibson, 1999; Mitraud, 2001).

Among indigenous peoples, access to and dependence on the projects market might bring

both benefits and disadvantages. Analyses of the constraints that indigenous peoples face when

entering in this type of social-political-economical framework, especially coming from their own

perspective, are critical in efforts to adjust and improve public polices directed towards

Amazonian development. According to Brown and Wychoff-Baird (1994), conservation

organizations and donor institutions have a lot to learn and adjust based on project evaluations

and analyses.

In the case of the Xingu and the Kaiabi, the access to resources (financial, technical,

institutional) coming from projects has been essential for ATIX's existence and activities since it

was created in 1995. Not only ATIX, but all other indigenous associations in Xingu have become

dependent on the projects market. For the indians, working through projects facilitates the

distribution of functions and resources needed to carry out sustained work. It makes it easier for

them to control highly disputed and problematic resources such as gasoline for motor boats and

food for workshops and meetings. Thus, from the organizational perspective, projects might be a

useful tool or instrument for indigenous peoples. Therefore, projects should be viewed as a









means to achieve a needed goal through specific work, not as lists of items to be purchased and

brought to the villages. This simple distinction has been discussed a lot with ATIX's team and

other political leaders in Xingu, since they got used to Villas-B8as and FUNAI's paternalistic

approach of bringing gifts and responding to lists of demands prepared by villagers (Oakdale,

1996). The projects culture has produced changes in the way Xingu Indians relate to external

actors and institutions. Nowadays, they (or some of them) are required to be creative, trained,

literate and legally represented by an institution to access funding to work with their

communities.

For the Kaiabi, running work through projects has also brought changes in social

organization structures and in power relations, sometimes entailing disputes and conflicts within

the communities. Oakdale (1996), carrying out research with the Kaiabi in the beginning of the

1990s, perceived changes in the social organization of villages, with a shift of power from elders

to younger leaders. She explains that elders have delegated authority to younger men, because

they speak fluent Portuguese, have greater facility in communicating with the outside world, and

therefore are better prepared to access goods and services for the villages. At present, many

bigger villages such as Capivara and Tuiarare, are commanded by younger chiefs (caciques).

Normally, these chiefs have kinship ties with elderly, reflecting an attempt to maintain traditional

social order but in new social organization structures. Therefore, what was regarded in the past

as a needed quality or ability in a chief or leader, of providing traditional food, objects and goods

for the village, has shifted in the present to the ability of younger leaders to attract resources

(monetary, objects, projects) and agents/services (NGO's practitioners, FUNAI officers, health

agents etc). The entry and relative dependence on the projects market, beginning in the 1990s,

has magnified these shifts in the perception and pursuit of power amongst generations (Oakdale,









1996; 2005). For example, the perceived success of the Kaiabi Araa project brought status and

power to the young chief Aturi (Jowosipep) in Tuiarare village.

Charged with the obligation to adjust to the project framework, indigenous peoples are

becoming increasingly dependent on technology and bureaucracy. Indigenous officers

responsible for writing and organizing documents, using computers and communicating with

outside actors, have become distanced from the indigenous way of life, to become indigenous

bureaucrats. Working under the premise that indigenous peoples have the right to self-

determination and thus of conquering a viable level of independence from non-indigenous

agents, we might have to accept these shifts in indigenous ways of life as part of the empowering

process.

The establishment of frameworks, networks and funding to promote sustainable

development through the projects market has entailed an increased dependence of local peoples

on external funding and sometimes on external agendas (Albert, 2000; 2005). According to

Fisher (1997), indigenous peoples might have to compete among themselves to access resources

that are not available for everyone, and many times to accept agendas and demands that are not

constructed with them or do not reflect their needs and interests. Rosengren (2003) states that the

new political organization brought by indigenous organizations might cause power conflicts

amongst indigenous communities, and increased vulnerability through dependence on external

demands and lack of power for making decisions. Some roles played by indigenous peoples,

many times with the support of national NGOs, tend to substitute State roles, which many times

do not have the technical or financial resources needed to perform that work.

Specifically in the case of ATIX, there has been a generalized lack of understanding

about ATIX's role, functions and work among the Kaiabi themselves and among other









indigenous peoples resident in Xingu. They tend to confound roles and duties related to each

institution which works with them, including FUNAI, ISA, EPM and others. It has been a

challenge for the Kaiabi to appropriate and assimilate ATIX as an institution of their own,

especially in the villages. We will come back to this question in the following section on

communities' perceptions of Kaiabi associations' work and role.

Another problem or constraint in project development is the lack of clear definition and

examination of the concept of "community" by external institutions and agents. According to

Agrawal and Gibson (1999:640), some common assumptions about communities held both by

the scientific community and development agents include the idealization of a social group that

has: "small size, territorial fixity, group homogeneity and shared understandings and identities."

The idealized view of community as a black box or as a homogenous social arrangement has also

been appropriated by indigenous communities at Xingu Park, causing social frictions and

compromising the results of the so-called community-based projects. Among the Kaiabi, the kin

group represented by an extended family traditionally was the basic social unit before the

transfer of the group to Xingu (Grunberg, 2004). After the transfer, as discussed above, the

organization of different kin groups in the same village has produced different social strains

(Oakdale, 1996). Thus, it was easy for the Indians to appropriate the concept of community to

refer to a congregated multi-family village. However, this concept might be used both for good

and for bad. For instance, power imbalances and misappropriations may be legitimized through

the use of "community" as benefiting from resources that in reality are being channeled to a

specific family or household. Or, leaders may speak in the name of the community without really

reflecting people's opinions and needs. Another problem relates to money or objects which









belong to "the community," causing wariness and conflicts related to who controls or uses these

resources.

Agrawal and Gibson (1999:629), in their analysis of the use of the concept of community

in projects and actions for natural resource management and conservation, suggest that actors

involved in community-based conservation and development should adopt a political approach,

focusing on institutions rather than on community: "identifying and working with multiple

interests and actors within communities, on how these actors influence decision-making and on

the internal and external institutions that shape the decision-making process." Among the

Indians, a good leader should also be able to deal with internal differences within the

community, reconciling interests and trying to reach an acceptable level of agreement (or

consensus) about a determined question. Therefore, a young chief might have to promote

numerous meetings in his community before taking any decision about a specific project. These

meetings should be included in project planning and development. Indigenous organizations,

political leaders and NGOs alike should be careful in being accountable to their constituents and

keeping transparency in their activities, especially regarding monetary expenditures.

Practitioners and donors should learn to recognize and work with the intrinsic differences, power

structures and possible emerging conflicts resulting when planning and executing conservation

and development projects amongst indigenous peoples.

The perceived dependency on external resources to develop indigenous traditional

activities might be another pitfall which comes with the projects market. In addition, if viewed as

new institutions entering in the villages' organization, projects may reinforce current trends

towards formalization and institutionalization of indigenous knowledge and practices (see

Chapter 8 for a more in-depth discussion on this). For instance, regarding both Munzwi and









Kaiabi Araa projects, the Indians may think that the only way to recuperate traditional crops or

to revitalize traditional knowledge is through projects and therefore access to external technical

and financial resources. I participated in a meeting in Tuiarare village where the villagers and

participants were discussing this topic. According to them, traditional ways of knowledge

transmission and traditional cultural practices should continue to occur independently of the

projects, which should be viewed as a tool to improve existing initiatives or traditional practices,

not as their substitute.

Whereas the list of contradictions and pitfalls related to the assimilation of the projects

culture among indigenous peoples is long, this is the framework in place and thus the one with

which stakeholders involved in community-based development have to learn to work, adapt and

adjust. More than once I have heard leaders affirming the need to keep both cultural systems

working together: that of the "whites" and the indigenous one, even if the distinction between

those has become blurred and nuanced. Positive outcomes of these community-based projects

and initiatives are not always instantly visible, perceived or quantifiable by practitioners,

researchers and donors involved. In the case of the Kaiabi, positive outcomes of these projects

include improvements in the communication inside and between different villages, with the

coordination of efforts and work towards the valorization of Kaiabi knowledge and culture. This

brings pride and interest to the people involved, including to the youth, who were usually not

interested enough in learning aspects of their own cultural inheritance (Turner, 1991). Besides

this, through these initiatives, the Indians are able to advance their own agendas and become

gradually more independent from non-indigenous agents. These initiatives are empowering to

indigenous peoples, since they allow more freedom of choice, relative independence from the

State to access and use resources, and also bring cultural pride to the communities involved.









Itaok and Kawaip Associations in Rio dos Peixes and Teles Pires

The Itaok Association (meaning rock or stone house; Ita=rock; ok=house) was created in

1999. At that time, it congregated the Apiaka, the Munduruku and the Kaiabi groups. However,

after some years, people realized that it wasn't possible to have all three groups involved in the

association; it was not working well. Therefore, in 2002 Itaok came to be composed solely of the

Kaiabi people living in Apiaka-Kaiabi land. The president in 2007 was Jewi Kwasiari

(Raimundo), and the vice-President was Faustino Tukuma Kaiabi, son of Catarina.

The association has run only two projects, both of them with accounting problems. The

first one was a project for Brazil nut processing and commercialization (R$ 30.000; US$

15,000), which involved all the three groups (Kaiabi, Apiaka and Munduruku), supported by the

Brazilian Ministry of Environment (MMA, Ministerio do Meio Ambiente). The only tangible

result of this project was the construction of a house (nowadays in very bad condition) for drying

the harvested Brazil nut. The Brazil nut project was receiving support from the GTME (Grupo de

Trabalho Missionario Evangelico), but the person who used to help them with the project

disappeared, never contacting the indians again. The second project was about improving

agriculture in the areas where wood has been extracted in the past, planting maize, rice, beans

and other crops. It received support from the Dutch government (R$ 6.000; US$ 3,000), but the

results were very modest.

President Raimundo said that they have had many difficulties running the association.

First, because the community does not want to contribute to the association, so he had to pay for

things with his personal funds when needed. Second, because they do not understand the

"whites" politics. He was once invited to participate in an ATIX assembly at Xingu Park.

Faustino, the vice-president, affirmed that the association has problems because people don't









know how to work in the organization. Both agreed that they need capacity-building and

technical assistance in order to be able to run the association. Wilson, the health agent in the

village and Tafut's (the older man and last shaman in the village) grandson said that there is lack

of dialogue between the association's personnel and the community: "the community needs to

understand what their role in the association is." Sebastiao said that the president of the

association has to be together with the community, talking, promoting meetings every day,

otherwise it does not work. Elections for changing the directors of the association were planned

for 2007.

Kawaip association (AIKK Associadao Indigena Kawaip Kaiabi) was created in 1999, to

legally represent the Teles Pires Kaiabi and to make it easier for them to raise and manage

resources and also to run their own projects. The first president was Ywyrup (Jose Kaiabi) from

1999 to 2002, and then Iracildo Munduruku from 2002 to 2006. Tarawi assumed the presidency

in 2007. He was born at Ilha Grande village in Xingu Park and got married in Kururuzinho. He is

living and studying in Alta Floresta, and has actively participated in the fight for the expansion

of the land to the Mato Grosso side. He has also been successful in mobilizing his relatives from

Xingu, especially from Ilha Grande village, to join them in political articulation concerning Teles

Pires' land struggles.

Tangeu'i (Arlindo), one of the village's teachers, commented that they could be trained

by the Xingu people on how to work with an indigenous association. According to Iracildo

Munduruku, former president, Cl6vis Nunes, FUNAI's officer and chief of Kururuzinho Post,

always encouraged the creation of the association. "What FUNAI could not do, the association

could," says Iracildo. A great example is the process of land demarcation. In 2004, the PPTAL

program for indigenous lands demarcation in Brazil was about to begin the process of land









demarcation for the extension of Kayabi land in Teles Pires. Part of the money for the

demarcation project was deposited in Kawaip's bank account, but the project never started

because of the legal contestation process which is still in course (see Chapter 4). Eroit, teacher

and vice-President of the association, said that Kawaip's biggest difficulty was not having run

any project, even small. He thinks that they need a person who can help them to write a funding

proposal. Some IBAMA representatives have helped them to review the association's statute

(IBAMA, 2004). He thinks that now it is going to be easier for them to organize the association's

work since the association represents only the Kaiabi, leaving aside the Munduruku and the

Apiaka peoples. "We have many plans...the first project we want is beekeeping, the second will

be about indigenous pepper."

After the Xingu and the Teles Pires Kaiabi groups got reconnected following the transfer

to Xingu, there has been an informal, kinship-based process of social and political exchange

between the two groups. This process happens between the Xingu Kaiabi and the two other

groups Rio dos Peixes and Teles Pires, but not between the last two. Makupa Kaiabi, former

ATIX's president and Eroit's brother (both sons of Masi'a elder from Tuiarare), has visited

Kururuzinho village to help in land struggles actions and also to explain to them about what

might be the work of an indigenous association. In the other direction, Eroit has traveled to

Xingu on many opportunities and for many different reasons, such as to visit his father Masi'a, to

participate in courses for training of indigenous teachers, to conduct health treatment with

shamans or to participate in political meetings.

While in Xingu the Kaiabi have had access to technical and financial resources, in the

other two areas they have struggled to create and maintain their local organizations. It is hard to

access and manage resources without a combination of proper assistance, support and internal









leadership. Land struggles have contributed to develop political cooperation and flow of

resources, information, people and knowledge between the three Kaiabi groups. Through their

accumulation of political power, the Xingu Kaiabi have subsequently been able to help the other

Kaiabi groups in territorial rights claims. However, ATIX representatives in Xingu have felt

overwhelmed by the amount of work related to ATIX's management and projects, therefore not

having much availability to help their relatives from other areas.

Peoples' Perceptions on the Work of Kaiabi Indigenous Associations

In this section, I present data on the understanding, participation and expectations of

people (men and women over 15 years old) interviewed in Kaiabi communities in Xingu

(Capivara and Tuiarare), Rio dos Peixes (Tatuy and Novo Horizonte) and Kururuzinho (Teles

Pires), regarding the role, current work and possible future projects of Kaiabi associations

(ATIX, Itaok and Kawaip respectively). These interviews were carried out as part of my research

on socio-economic factors and weaving knowledge, presented in Chapters 7 and 8 respectively,

and included a total of 224 people in the four villages, divided in 110 men and 114 women (see

chapter 7 and Table 7-1). I plan to return these results to the communities and leaders of Kaiabi

associations, so they can better understand communities' perceptions about their work and take

them into account in their future actions.

The results of the interviews carried out among Capivara and Tuiarare villages' residents

show that women know much less about the function and work performed by ATIX when

compared to men (Table 6-3). They also participate less in political meetings and as members of

political councils. For instance, in Capivara only 17% of women interviewed have participated in

ATIX's political meetings, versus 60% of men. In Capivara, men and women participate more in









ATIX's meetings (total 39%) than in Tuiarare (total 24%). This is probably related to the closer

proximity of Capivara village to the Diauarum Post.

In Rio dos Peixes and Teles Pires, people's participation and perceptions concerning the

local organizations Itaok and Kawaip differ from those observed in Xingu. For instance, women

participate more in political meetings at Rio dos Peixes (35%) and Kururuzinho (36%) in

comparison to Capivara (17%) and Tuiarare (9%). In Rio dos Peixes, this can be attributed to the

fact that the majority of women have some level of literacy (see Chapter 7) in comparison to the

other areas, besides the fact that in Rio dos Peixes they are also struggling for land demarcation,

having frequent community meetings to debate problems related to land being claimed. The

participation of men in political meetings at Teles Pires is greater (81%) than in Rio dos Peixes

(47%). This is probably related to the land struggle process in Teles Pires, which has been more

intense than in Rio dos Peixes, besides the fact that Kururuzinho's population is smaller than Rio

dos Peixes. Overall, political participation is greater in Teles Pires (57%) in comparison to the

three other villages (Figure 6-4).

The interviews identified six categories of work or functions that are developed through

the local indigenous associations: 1) Patrolling of land borders; 2) Development of projects; 3)

Cultural revitalisation; 4) Political articulation; 5) Commercialisation of products; 6) Land

struggles. Some respondents said that they did not know what work was carried out by local

organizations. There is a common trend registered for all four villages, in which the great

majority of women don't know what the associations' work or functions are. The "don't know"

answer was given by 87% of women in Capivara, 86% in Tuiarare, 85% in Rio dos Peixes and

89% in Teles Pires. For men, "not knowing what the work of associations is" also scored high,

especially in Rio dos Peixes (61%), but less than for the women (see Figure 6-5). According to









the interviewees in Xingu villages, the main work carried out by ATIX in the Park is that of

patrolling Xingu Park's borders (16% in Capivara and 20% in Tuiarare), which was initiated by

Mairaw6 through ATIX in 1998. Another function that scored high for both villages is the

commercialization of products (18% in Capivara and 15% in Tuiarare). In Capivara, 20% of

people mentioned that ATIX's main work is the development of projects.

In Rio dos Peixes and Teles Pires, where the work of patrolling lands' borders is not so

significant in comparison to Xingu, the main work or role for the associations is the development

of projects: 19% in Rio dos Peixes and 35% in Teles Pires. The second main work of

associations mentioned in Rio dos Peixes and Teles Pires was that of commercialization of

products in Rio dos Peixes (9%), and land struggles in Teles Pires (17%). The greater link to

markets and the dependence on sale of handicrafts, agricultural products and Brazil nuts to

consolidate the family income in Rio dos Peixes (Chapter 7) is reflected in their expectations that

the association will undertake more control and thus improve the organization of commercial

activities among them.

It is easier to visualize the differences between men and women and among villages by

looking at Figure 6-5. In general, men know more and in more detail about the work of

organizations. Women's perception about the work of organizations is more diversified in Xingu

(Capivara and Tuiarare villages). In Rio dos Peixes and Teles Pires, women basically refer to

projects and commercialization of products (only in Rio dos Peixes) or they don't know about it.

This makes us infer that in Xingu the association works more closely with the community and

also that women have more access to information about ATIX's work, while in the other areas,

the associations' work is either incipient (such as in the case of Rio dos Peixes) or is performed

with little involvement of women (communication and participation). Therefore, in all three









areas, political leaders and associations' teams need to direct more attention to women, creating

mechanisms through which they can have better access to information and more opportunities

for participation.

Peoples' preferences for future projects to be developed through the local organizations

also differ among villages, but less between genders than the perceptions of associations' work.

Thirteen different ideas were identified for projects to be developed by the associations in the

four villages studied (Table 6-3), but four of them were the most cited by both men and women

alike: 1) handicrafts; 2) traditional crops; 3) Brazil nut; and 4) beekeeping. Projects for

handicrafts production and commercialization scored 37% in Capivara; 29% in Tuiarare; 50% in

Rio dos Peixes and 35% in Teles Pires (Figure 6-6). In fact, handicrafts' trade is one of the most

important sources of family income in all four villages, with the exception of Brazil nut

commercialization in Rio dos Peixes. This might also be related, in the case of Capivara and

Tuiarare, to the previous experiences they have had with cultural revitalisation projects regarding

handicrafts, such as the Kumand and the Kaiabi Araa projects. Management and revitalisation of

traditional agriculture and crops varieties also scored high for all villages, especially among men.

There is a current generalised worry among the Kaiabi, related to the perceived loss of their

agrobiodiversity (Silva, 2009). In Rio dos Peixes, 10% of the women and 44% of men, totalling

26%, expressed the expectation that the Itaok association should develop a project for the

organisation of Brazil nut collection and commercialization. In Teles Pires, people already had

mentioned the interest in a project for honey production and commercialization, similar to that

developed by ATIX and ISA in Xingu.

The number of people who don't know which projects) they would like to see being

developed in their villages is greater among the women, and it is also overall greater in Xingu









(Capivara-30%, Tuiarare-40%) in comparison to the other villages (Rio dos Peixes-16%, Teles

Pires-7%). Overall, the same people who don't know about indigenous organizations' work also

responded that they did not know which project they would like to participate in.

I believe that there are both needs and opportunities for greater interaction between the

three Kaiabi organizations, which could learn from and collaborate with each other in different

ways. For instance, ATIX could provide information and technical assistance to the Kawaip

association in the development of a beekeeping project. ATIX's personnel could provide

information and training to both Itaok and Kawaip associations in project planning and

development, as already indicated by people who work in these organizations. Maybe Kaiabi

organizations need a project to be able to understand and experience working through projects.8

In Xingu, where I have more experience observing people's reactions to the

establishment of a new indigenous institution, I have perceived that communities in the villages

frequently misunderstand the role and function of ATIX as representing their own needs. They

often say that ATIX is now "their father," and being so, it is expected to provide the financial

support and the goods that they need. People tend to confuse the role of ATIX with that of

FUNAI, known by its paternalistic posture towards Xingu people, especially during the Villas-

B8as administration. They also criticize ATIX as portraying ISA's goals and agenda, not theirs.

ATIX representatives are aware of this problem and have directed efforts to make more visits to

the villages to talk with the communities.

The role and work of different institutions present in an indigenous land should be

constantly reaffirmed to the communities. Related to this, Rosengren (2003) mentions that the



8 It is worth mentioning here that the decision of what to do with the information generated by this research in
relation to Kaiabi organizations will be put in their hands through the process of returning results to the
communities.









assimilation of a "westernized" model of institution can cause increased local conflict and

vulnerability among indigenous peoples. He identified two parallel models for constructing

identity amongst the Matsigenka people in Peru: one comes from Matsigenka cosmogony, by

which they conceptualize the world. The other is the way that Matsigenka represent themselves

in the political arena, through local organizations. He affirms that "local organizations based on

ethnic belonging emerge, expand and then start to experience problems in the relationship

between leaders and their constituencies" Rosengren (2003:237).

The political ascendancy of the Kaiabi in Xingu, and especially of some individuals, has

strained social relations both among the Kaiabi and with other groups in the upper and middle

sectors of the Park. The relationship between the Kaiabi- newcomers in Xingu- and the upper

Xingu groups whose ancestors have occupied the Xingu headwaters for at least one thousand

years (Oakdale 1996; 2005; Heckenberger 2001; 2005) is complex. Due to the unique conditions

of Xingu Park and to the status of these xinguano tribes as 'authentic Brazilian Indians,' the

upper Xingu groups have received greater attention from the media, anthropologists and

politicians (Menezes 2000; Oakdale 1996). Some of their rituals, such as the kwaryp

('celebration of the dead'), are nationally famous and receive important visitors every year. The

Kaiabi have in the past felt powerless and marginalised by these more traditional and emplaced

upper Xingu groups, who generally view the Kaiabi as 'caboclised' newcomers (Oakdale 1996).

These dynamics have probably contributed to driving the Kaiabi into the process of political

empowerment and cultural revitalization in which they are currently engaged.

Conclusion

The relocation of the Kaiabi to Xingu Park has enabled them to access knowledge,

technical and financial resources otherwise unavailable, which provided a platform for the

renewal of their identity and for their political empowerment. The Kaiabi, through leadership and









persistence, have been relatively successful in the appropriation of Western institutions and in

taking advantage of opportunities provided by the concomitant development of indigenous

grassroots movements and the environmentalist movement in Brazil (Little, 2001; Albert, 2000;

2005; Oakdale, 1996; 2005). In other words, some Kaiabi leaders have had the previous

historical experience, the knowledge and the leadership needed to better take advantage of the

opportunities offered to them. The previous experience with rubber tappers made some Kaiabi

men instrumental in assisting the Villas-Boas brothers in the administration of Xingu Park. At

the same time, the growing contact with the outside world, facilitated by the emblematic

reputation of Xingu Park, as well as the experiences resultant from involvement with other

indigenous groups (especially the Kayap6) in political struggles at national levels, contributed to

the "capacity-building" of Kaiabi leaders, such as Mairawe.

The demographic growth of the Xingu Kaiabi a few decades after the transfer, helps to

explain the differences between the three Kaiabi groups. Sheltered and assisted by a good health

service, maybe one of the best available to indigenous peoples in the country, the Kaiabi are the

biggest population in Xingu Park. This has implications also for the cultural persistence of the

group, since in Xingu there is a reserve of people who still speak the native language and stick to

cultural traditions.

The creation of ATIX is an internal historical landmark for the Kaiabi as a socio-political

experiment in which new roles, identities and actors take action. It was made possible through

the favorable political moment and opportunity provided by the multiplication of Amazonian

indigenous organizations after the Constitution of 1988 and the alliance of indigenous peoples

with the environmentalist movement (Conklin and Graham, 1995; Kearney, 1996; Ramos, 1998;

Albert, 2000). ATIX has counted on technical support from ISA and financial support from the









Norwegian Rainforest Foundation since its creation. This long-term relationship with partners

and donor institutions was critical in providing the financial resources and the capacity-building

needed to sustain and power up ATIX since its creation fourteen years ago.

However, the assimilation of western social structures and institutions has potential for

both the good and bad. At the same time that the Kaiabi are better prepared to deal with external

circumstances, they are exposed to internal conflicts due to clashes between the traditional and

the new forms of power through social and political organization, and between old and new

generations (Oakdale, 1996; Rosengren, 2003). ATIX, Itaok and Kwaip associations are also

dependent on international funds to survive, and very few donors are willing to provide

institutional support instead of applying resources in specific projects. In a way, the partnership

between indigenous peoples and conservation has meant both freedom (from the state) and

dependence (on international political conjunctures). As Albert (2005:205) stated, "Amerindian

strategies regarding identity and territory are inscribed within international political conjectures

that set out their conditions of possibility, sustained their emergence, and delineated the range of

their implementation."

People's perceptions about the role of Kaiabi indigenous organizations in four villages

differ among genders and villages. One common feature among all villages analyzed is the lack

of women's participation and access to information related to the work of indigenous

organizations, as well as about preference for future projects to be developed in their villages. In

Xingu, the work of patrolling Park's borders is recognized in both Capivara and Tuiarare

villages, besides development of projects and product commercialization. In Rio dos Peixes,

Itaok's work is too incipient to be recognized by the community, and in Teles Pires, Kawaip's

work is regarded as linked mostly to projects and land struggles.









Whereas ATIX has had access to different types of resources during its fourteen years of

existence, the other two associations at Rio dos Peixes (Itaok) and Teles Pires (Kawaip) have

struggled to continue existing in spite of lack of access to funding opportunities, training,

technical advice and qualified personnel. However, ATIX's empowerment has trickled down to

the other two Kaiabi associations, through kinship ties and related to the processes of land claims

in which Rio dos Peixes and Teles Pires groups have been involved.

In the case of ATIX, Mairaw6's leadership was key in ensuring good political

relationships with neighboring groups in Xingu in spite of conflicts and jealousy that might

accompany any escalade to power. Therefore, access to resources per se might not ensure

political empowerment of grassroots cultural minorities. Leadership, training, persistence,

transparency and long term funding are important ingredients in the formation, strengthening and

endurance of indigenous organizations in the Brazilian Amazon.


























Th APIAKA-KAYABI




\4Q


S'1


4i




PARQUE dOAGENA
DO XINGU


..1 a_ -.
4P
SAreas inhabited by the Kaiabi in Brazil -
XBug I ltl Par Ti A, la.K. iWr. anW Tl KI.at. "
Other indigenous lads -



INSTITUTO SOCICAMBrENTAL, 2004


Figure 6-1. Location of the indigenous lands currently occupied by the Kaiabi: Xingu Indigenous
Park, TI Apiaka-Kaiabi (Rio dos Peixes) and TI Kayabi (Teles Pires). Adapted from
Griinberg (2004) and Athayde et al. (2009).


L

I,
z-1


9
e
"I


'I





















































Figure 6-2. Map of Xingu Indigenous Park. It is possible to distinguish the upper Xingu region
and PI Leonardo, the middle section and PI Pavuru and the northern portion or lower
section of the Park where PI Diauarum is located. Source: ISA (2009 b).






235











1800
1600 1647
1400
S-- Xingu Park
1200 -
S1102 -m- Rio dos Peixes
3 1000 -
-A- Teles Pires
600- --X-- Others
600 -
-- Total
400- 308 T ot
200 -
0




Year/range

Figure 6-3. Evolution of Kaiabi population in the three areas occupied by the group, from 1955
to 2007. Sources of information: Dornstauder, 1955; Rangel, 1987; Melia, 1993;
UNIFESP, 2004; FUNAI, 2004; Baruzzi, 2005; Pagliaro, 2005; FUNAI, 2007; ISA,
2007.









Table 6-1. Evolution of Kaiabi population through time.
Year/range Number of people
Xingu Park Rio dos Peixes Teles Pires Others Total
1955 40 108 148 45 341
1966 179 53 56 20 308
1970 213 70 -
1979 330 95 -
1985 435 -
1989 526 130 -
1995 656 -
1999 767 265 70 1102
2002 958 -
2003/2004 1000 -121 -
2006/2007* 1226 266 155 1647
*Total number of Kaiabi for 2006/2007 is estimated and might not include Kaiabi living in cities
outside the indigenous lands. Data sources: Dornstauder (1955); Rangel (1987); Melia (1993);
UNIFESP (2004; 2007); Baruzzi (2005); Pagliaro (2005); FUNAI (2004; 2007); FUNASA
(2007); ISA (2007).









Table 6-2. Synthesis of the results of the Kaiabi Araa project, in terms of objects produced and
people who have participated in the four workshops developed in Kururuzinho and
Tuiarare villages between 2004 and 2006.

Products and participants Villages and workshops TOTAL

Kururuzinho Tuiarare Tuiarare Tuiarare
2004 2004 2005 2006
Twill-plaited baskets with 20 31 25 0 76
graphic designs


Fans


Cylindrical basket

Adorned clubs (woven
handles)

Hammock with graphics
designs

Simple hammock

Strap for carrying babies

Bags

Basketry teachers (male)


Textiles teachers (female) 2

Basketry students (male) 12

Textiles students (female) 13
Source: Adapted from ATIX (2006 c).











Table 6-3. Comparison between political participation, perception of indigenous organizations' work and preference for future
projects between four Kaiabi villages. F=female; M=male; T=total.
Villages Capivara Tuiarard Rio dos Peixes Teles Pires
Gender/Total F M T F M T F M T F M T
N/% N % N % N % N % N % N % N % N % N % N % N % N %
Political
participation
Participation in
political
meetings
4 16.67 15 60.00 19 38.78 2 9.09 9 39.13 11 24.44 14 35.00 17 47.22 31 40.79 10 35.71 21 80.77 31 57.41
Participation in
political councils 2 8.33 1 4.00 3 6.12 1 4.55 7 30.43 8 17.78 3 7.50 8 22.22 11 14.47 1 3.57 10 38.46 11 20.37


Role of
indigenous
organizations
Patrolling of land
borders

Projects

Cultural
revitalization

Political
articulation

Commercialisation
of products

Land Struggles


1 4.17 7 28.00 8 16.33

1 4.17 9 36.00 10 20.41


0 0.00 6 24.00 6 12.24


1 4.17 5 20.00 6 12.24


2 8.33 7 28.00 9 18.37


4.17
87.50


Don't know


3 12.00 4 8.16
10 40.00 31 63.27


0 0.00 9 39.13 9 20.00 0 0.00 0 0.00 0 0.00

0 0.00 2 8.70 2 4.44 4 10.00 11 30.56 15 19.74


1 4.55 4 17.39 5 11.11 0 0.00 1 2.78 1 1.32


0 0.00 5 21.74 5 11.11 0 0.00 2 5.56 2 2.63


3 13.64 4 17.39 7 15.56 1 2.50 6 16.67 7 9.21


0.00
86.36


1 4.35 1 2.22 0 0.00 0 0.00 0 0.00
13 56.52 32 71.11 34 85.00 22 61.11 56 73.68


0 0.00 1 3.85 1 1.85

3 10.71 16 61.54 19 35.19


0 0.00 3 11.54 3 5.56


0 0.00 4 15.38 4 7.41


0 0.00 1 3.85 1 1.85

0 0.00 9 34.62 9 16.67
25 89.29 10 38.46 35 64.81





Capivara Tuiarard Rio dos Peixes Teles Pires
M T F M T F M T F M T
N % N % N % N % N % N % N % N % N % N % N %


Table 6-3. Continued
Villages
Gender F
N/% N %
Preference
for future
projects

Agriculture/
traditional
crop
1 4.17
Natural
resource
management
0 0.00
Handicrafts 1
1 45.83
Musical
revitalization
0 0.00
Beekeping
0 0.00
Brazil nut
0 0.00
Cattle raising
0 0.00
Festivals
0 0.00
Political
1 4.17
Salt
production


24.00 6

28.00 18


12.24

36.73



2.04

2.04

0

0

2.04

4.08


2 9.09 8 34.78 10 22.22 3 7.50 10 27.78 13 17.11


0.00 6 26.09

31.82 6 26.09


1 4.35

4 17.39

0 0.00

0 0.00

1 4.35

1 4.35


13.33 0 0.00 1 2.78 1 1.32

28.89 22 55.00 16 44.44 38 50.00


2.22

11.11

0.00

0.00

2.22

2.22


6 21.43 5 19.23 11 20.37


3.57 3

46.43 6


0 0.00 1 2.78 1 1.32

2 5.00 6 16.67 8 10.53

4 10.00 16 44.44 20 26.32

1 2.50 1 2.78 2 2.63

0 0.00 1 2.78 1 1.32

0 0.00 0 0.00 0 0.00


11.54 4

23.08 19


0.00

57.69

7.69

0.00

0.00

7.69


7.41

35.19



0.00

33.33

7.41

0.00

0.00

3.70


0 0.00 0 0.00 0 0.00 0 0.00 1 4.35 1 2.22 0 0.00 0 0.00 0 0.00


11 44.00 12 24.49


0 0.00 0 0.00 0 0.00











Table 6-3. Continued
Villages Capivara Tuiarard Rio dos Peixes Teles Pires
Gender F M T F M T F M T F M T
N/% N % N % N % N % N % N % N % N % N % N % N % N %
Preference
for future
projects


0 0.00 0 0.00 0 0.00

0 0.00 0 0.00 0 0.00


0 0.00 0 0.00 0 0.00 0 0.00 0 0.00 0 0.00 0 0.00 2 7.69 2 3.70

0 0.00 0 0.00 0 0.00 1 2.50 0 0.00 1 1.32 0 0.00 0 0.00 0 0.00


0 0.00 0 0.00 0 0.00 0 0.00 0 0.00 0 0.00 0 0.00 1 2.78 1 1.32 0 0.00 0 0.00 0 0.00
Don't know 11 45.83 4 16.00 15 30.61 14 63.64 4 17.39 18 40.00 11 27.50 1 2.78 12 15.79 2 7.14 2 7.69 4 7.41


Pepper
production

Education

Fish















M

F

M



<" M
F

M



0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00
% of participation in political meetings

Figure 6-4. Degree of participation in political meetings amongst men (M-male) and women (F-
female) in four Kaiabi villages. CA=Capivara; TU=Tuiarare; RP=Rio dos Peixes;
TP= Teles Pires.














M


F


M Patrolling of land borders
:* Projects
-F O Cultural revitalisation
0 Political articulation
M i Pi Commercialisation of products
Land Struggles
F 0 Don't know

M




0 20 40 60 80 100 120 140 160 180 200
% of people interviewed

Figure 6-5. Perceptions of people (M-male and F-female) about the work carried out by local
organizations in Capivara (CA), Tuiarare (TU), Rio dos Peixes (RP) and Teles Pires
(TP) villages.















Indigenous salt production

Indigenous pepper production

Fish

Education

Cattle raising

Brazil nut

Festivals

Musical revitalisation

Beekeping

Political

Natural resource management

Agriculture/traditional crop

Don't know

Handicrafts


0

I-


STP
O RP
SCTU
* CA


0.00 10.00 20.00 30.00 40.00 50.00

Percentage of people interviewed in each village (%)


60.00


Figure 6-6. Preference of communities for future projects to be developed by indigenous
organizations in four Kaiabi villages. CA-Capivara; TU-Tuiarare; RP-Rio dos Peixes;
TP-Teles Pires.


r----









CHAPTER 7
COMPARATIVE SOCIO-ECONOMIC ANALYSIS OF FOUR KAIABI VILLAGES

Introduction

In the previous chapter, I presented information on the infra-structure, institutional and

political configuration of each land occupied by the Kaiabi, on a broader scale, without

considering villages' and individuals' specificities. This chapter goes one step down in scale,

containing socio-economic descriptions and analyses at the village and individual levels, and

providing a comparison among the four Kaiabi villages included in this research. The exploration

of the socio-economic variables considered in this chapter will enable a more in-depth

understanding of the factors and processes which underlie the different trajectories followed by

the three Kaiabi groups, in terms of cultural resilience and change. The main socio-economic

data collected and analysed were place of birth; interethnic mixing; function in the village;

stability and sources of income; proficiency in the native language; schooling; and degree of

contact and interactions between the four villages.

Multiple questions might be formulated concerning the complex relationships between

socioeconomic factors and cultural resilience. In this chapter, I take language proficiency as an

indicator of cultural vitality or robustness (dependent variable) and analyse the effect or

relationship of other socio-economic variables on it (independent or explanatory variables). I

carry out an exploratory analizys of the effects formal schooling, gender, stability of income and

age might have on language proficiency across the four villages. Then, in Chapter 8, I carry out

the same analyses and hypotheses testing using weaving knowledge as the dependent variable.

For the analyses, I considered language proficiency as a proxy for indigenous knowledge (Maffi,

2001), conducting a preliminary test of the following hypotheses (detailed in Chapter 2):









HI: Elders retain deeper and different knowledge when compared to younger people.

Therefore, it would be expected that elders are more proficient in the language than youth.

H3: Greater levels of market integration erodes indigenous knowledge. It is expected that

greater income stability will result in lower language proficiency.

H4:Formal schooling erodes IK. Here, higher levels of formal schooling would be

correlated to lower language proficiency.

Some specific questions to be asked are: a) What is the effect of formal schooling on

language vitality? b) Is language proficiency correlated with age? c) Which Kaiabi group retains

greater language vitality? Why? Specific questions are stated in each sub-section of this chapter,

which is divided by socioeconomic variables. Then, in the next chapter, I will compare these

socioeconomic variables with weaving knowledge, which I treat as another indicator of cultural

vitality. To better understand the multiple interactions among socioeconomic and cultural

processes, I used a set of quantitative statistical methods, including chi-square and correlation

tests, bivariate and multivariate linear and logistical regression.

The topics explored in this and in the next chapter have been studied by authors who have

worked with different domains of knowledge and cultural change amongst indigenous and other

traditional peoples, using both qualitative and quantitative approaches (Boster, 1986; Garro,

1986; Nabhan and St Antoine, 1993; Alexiades, 1999; Zent, 1999; 2001; Godoy et al., 1998;

Godoy, 2001; Reyes-Garcia, 2001; Reyes-Garcia et al., 2005; Hill, 2001; Hunn, 2001; Ross,

2002; Zarger and Stepp, 2004; and others). Concurring with Maffi (2001:13), my broader

objective here is to "advance the understanding of socioeconomic factors which underlie

processes of culture shift among indigenous peoples, and through which mechanisms they


1 See Chapter 2 for detailed description on methods for data analysis used in this research.









operate." For instance, Victoria Reyes-Garcia (2001; 2005) studied the effects of market

integration on environmental knowledge systems among the Tsimane in lowland Bolivia. Gary

Nabhan and Sara St. Antoine (1993) compared the effects of lack of contact with nature and

changes in oral transmission of knowledge among indigenous and non-indigenous children in the

Sonoran desert in the US. Stanford Zent (1999; 2001) studied the relationships between social

variables (age, schooling and bilingual ability) and ethnobotanical knowledge amongst the Piaroa

indigenous people of Venezuela. All of these studies have shown that knowledge loss or change

is never a linear process evenly distributed throughout an indigenous population. Therefore, the

main theoretical contribution of this and those studies is not just proving that market integration

or formal schooling changes and or erodes language proficiency and indigenous knowledge, but

how these processes happen, interact and affect the social-political configuration and self-

determination of indigenous peoples (Reyes-Garcia, 2001).

The socioeconomic factors or, as I am treating them, "variables," are intertwined and

interactively shape and are shaped by history and by internal and external forms of social,

political and institutional organization. Therefore, it is often difficult to isolate socioeconomic

factors and distinguish which ones are more influential (and why) in processes of cultural

change. For instance, the access to fixed monthly wages by a person may permit that this

individual provides formal schooling for his or her children, which in turn will affect their

capacity to retain or lose traditional knowledge. The closer proximity to cities, such as in the case

of the Rio dos Peixes group, enables closer contact with the market economy and also with the

non- indigenous lifestyle, including substitution of the native language by Brazilian Portuguese,

the national language.









This chapter is organized in two main parts: the first part contains a description of

location, infra-structure and general socioeconomic organisation of each of the four villages

considered in the research: Capivara and Tuiarare villages at Xingu Park, Tatuy post and village

in Rio dos Peixes (including Novo Horizonte village as part of Tatuy) and Kururuzinho post and

village in Teles Pires river. In Chapter 6, I described each of the three lands in which the villages

are located. In this and in the next chapter, I refer to the villages and subsequently, household

and individuals. The second part of this chapter contains comparative socioeconomic analyses of

the four villages, and thus is divided by the main independent and dependent socioeconomic

variables explored in this research.

Overview of Villages Organization

Capivara and Tuiarare Villages at Xingu Park

Both Capivara and Tuiarare villages are located in the northern portion of Xingu

Indigenous Park (Figure 7-1). Capivara village is placed on the right riverbank of Xingu River,

about 20 minutes by boat to the north of Diauarum Post. It was created around 1977, comprising

the extended family linked to the great chief Temeioni (deceased), who lived at Rio dos Peixes

and died at Xingu Park. His older son, Kupeap (around 77 years old), is one of the oldest Kaiabi

men still alive and is very knowledgeable about Kaiabi culture. When Temeioni's family first

arrived in the Park, they lived some years in and around Diauarum Post. Then, they established a

village in front of the place where Capivara is located today. Canisio (who lived in Xingu but

moved back to Rio dos Peixes around 1997) and Kupeianim (another of Temeioni's sons and

Kupeap's brother) decided to open Capivara village, inviting other people who came from Rio

dos Peixes to join them. It is one of the biggest Kaiabi villages in Xingu Park, where nearly 145

people live nowadays. Currently, the head of the village is Jywapan (Temeioni's nephew).









The village has one motorboat, a radio and health unit buildings, a school, public toilets

and a beekeeper's office. There are three teachers, Awatat, Jemy and Sirakup, working in the

elementary school. After nearly 30 years living in the same area, in 2006 Capivara residents

decided to transfer their houses to an adjacent area in the back of the village, leaving in the

village the services and institutional buildings such as radio, school, honey house and meeting

plaza (see Figures 7-2 and 7-3 drawn by Capivara's students, which show a map with the houses'

spatial disposition in the old and in the new village). When I visited the new village in June of

2007, they were in the process of building the new houses, and they were planning to promote a

Jowosi (traditional Kaiabi festival) to celebrate the inauguration of the new village. In the new

village, they were finishing the construction of a traditional Kaiabi festival house.

There is a significant two-way movement of people between Capivara and Rio dos

Peixes/Novo Horizonte villages. For instance, there are people from Capivara who moved to Rio

dos Peixes, such as Canisio and part of his family (some of his many sons still live in Capivara);

and also people who used to live in Capivara, moved to Rio dos Peixes and decided to come

back to Capivara in Xingu. This is the case of Tewit (78 years old), an elder also very

knowledgeable in basketry weaving, who moved to Rio dos Peixes with his wife but did not

want to stay there apart from his family, coming back to Xingu after a couple of years of residing

in Rio dos Peixes. Another example is Pofat and Rosilda. Pofat is one of Kupeianin's children,

who is married to Rosilda from Rio dos Peixes and has travelled more than seven times between

there and Xingu. Often, longer or extended travels and exchanges between the two places happen

as a consequence of marriages or kinship ties2




2 See following section on travels and dislocations between the villages.









Capivara residents are considering creating a village association to manage community

projects and to facilitate access to technical and financial resources, but this did not happen yet.

One of the biggest problems confronted by residents is the lack of fertile soils for agriculture.

The sedentary lifestyle and the fact that they have been living in the same area for nearly 30

years have caused the depletion of good soils and other resources used in house construction

(such as the inaja palm used for thatching) and in handicrafts production. The residents say that

some more demanding crops such as maize, peanut varieties and bananas do not grow well in the

over-exploited red soils still available around the village. As a result, they have had to search for

"capoeiras" (black earth spots of fertile soil, see Chapter 5) in areas distant from the village,

which demands a lot of hard work and sometimes creates conflicts and competition between

some families. The first community-based project chosen by Capivara villagers to be presented

to the PDPI program was related to the recovery of exhausted "capoeiras," or of areas suitable

for agriculture.

Tuiarare village is located on the right riverbank of the Xingu, southwards of Diauarum

Post, nearly one and a half hours by boat. In the beginning, Tuiarare was a small village

comprising Masia's (one of the elders) family, who came from the Teles Pires River to Xingu.

Masia (around 82 years old) is the father of Eroit Kaiabi, teacher and political leader in

Kururuzinho village in Teles Pires. Afterwards, other families decided to join them and to build a

bigger village. This happened around 1987. The current chief is Aturi Kaiabi (now named

Jowosipep, 46 years old), who accumulates the functions of chief, teacher, political leader and

president of the village's association, named Tapawia. Aturi is a very active and entrepreneurial

person. He was responsible for the conceptualization and coordination of the Kaiabi Araa

project. Aturi's father, named Xupe (75 years old), also came from Teles Pires.









The village has a school, a combined radio and health unit building, two motorboats,

public toilets and a beekeeper's office. There is also a public phone in the village, installed with

the support of the mayor of Quer6ncia municipality, to which the village officially belongs.

There are two indigenous teachers, Aturi and Pikuruk. Currently, there are close to 165 people

living in the village.

In Tuiarare, similarly with what has happened in Capivara, the residents are also

relocating the houses to the area in the back of the village. When I visited the village in June of

2007, some of the residents, such as Masia's family, were already living in their new houses in

the back of the village. Figure 7-4 shows the configuration of the village in 2002 and Figures 7-5

and 7-6 present the current and the planned village in 2007. In the new village, a traditional

Kaiabi house was also built, with resources from the PDPI project Kaiabi Araa. This house has

been used as a "cultural school", in which activities such as crafts workshops, festivals and

community meetings have been promoted.

In comparison to Capivara and Rio dos Peixes, the movement between Tuiarare and

Teles Pires is less intense. Chico (Moyt, 62 years old), is a shaman who has been living between

Tuiarare and Kururuzinho in Teles Pires in the last couple of years. Travels between the two

areas happen more often in Masi'a's family, with visits by Eroit and his family to Xingu and also

by Makupa (Eroit's brother) to Kururuzinho.

Comparable to what is happening in Capivara, in Tuiarare's region there has been a

growing scarcity of good soils for agricultural production, and of some key resources used for

house construction and for handicraft production. This is also a consequence of the village's

sedentarization, and concern has been expressed by the residents.









In both villages, the main economic activities are fishing and hunting, almost exclusively

done by men, and agriculture, carried out by men and women, with family and some communal

agricultural plots. Handicrafts production and sale plays an important role in the family

economy. Tuiarare's women are very active in producing palm fruit jewelry for exchange and

sale. As we are going to detail later, waged work has increasingly played an important role in the

villages' economy and in the access to industrialized food and goods.

Tatuy Post and Novo Horizonte Village in Rio dos Peixes3

In Rio dos Peixes, Novo Horizonte village is adjacent to Tatuy Post and uses the post's

services and facilities. There are around 266 people living in the Tatuy Post and in the Novo

Horizonte village altogether (FUNASA, 2006), but not all of them are Kaiabi. There are

interethnic marriages, especially between Kaiabi, Apiaka and Munduruku, which share the land.

There is also a non-indigenous woman married to a Kaiabi man living in the village. Novo

Horizonte village was recently constructed, around 2006. The health district approved a project

for construction of sanitary units, so the residents decided to build a new village adjacent to the

Tatuy post. In Figures 7-7 and 7-8, I present maps with the spatial distribution of houses and

facilities in these villages in 2007. The village's chief is the young Murua'i (20 years old) born in

Capivara village in Xingu, son of the deceased Takaperun and Juwea'i. His mother Juwea'i still

lives in Capivara village, but she was in Rio dos Peixes on the occasion of my visit, undertaking

medical treatment in Juara and visiting her son. Murua'i is an example of a man who came to

Rio dos Peixes to visit relatives, got married and did not return to Xingu anymore.




3 For the data presented in this dissertation, I considered Tatuy Post and Novo Horizonte village to be one unique
village in Rio dos Peixes area. This merging was done to facilitate and improve data analysis, taking into account
that both places are adjacent and very similar in terms of social organization, family composition and economic
activities (see Chapter 2 for details).









The post's infrastructure is composed of an energy generator operated by diesel, water

wels, radio, health unit and school. There is also a house used for meetings, festivities and

occasionally for celebration of mass by priests who live in or visit the village.

People still plant garden farming plots (roca), cultivating cassava, peanuts, bananas,

sweet potatoes, taro and other crops, but the diversity of crop plants cultivated by Rio dos Peixes

residents is smaller than that of Xingu villages. Some families do not pursue a specific garden

plot away from the village, therefore cultivating cassava and few other crops and fruit trees in

their homegardens. This trend, of substituting an agricultural family plot (named "roca" in

Portuguese or "ko" in Kaiabi) by a homegarden, was observed in Xingu only at Diauarum Post,

but not in the villages. It is also absent in Kururuzinho village in Teles Pires.

The village differs from those of Xingu in many aspects, but at a first glance, one can

perceive that there is a clear tendency of atomization of the domestic units around nuclear

families (Senra, 2002). This does not necessarily mean that the extended family has lost its social

role at Rio dos Peixes. The contact with Brazilian society and the influence of the missionaries

probably contributed to the abandonment of the tradition of living in big collective houses, where

an extended family co-habit, as still can be observed in Xingu's villages. However, even residing

in a small nuclear family, the role of the extended family in social organization has still prevailed

within most families, as Senra (2002) noted.

The village' economy is dependent on seasonal products and market opportunities. It

comprises a mix of Brazil nut commercialization, waged work (temporary and permanent), sales

of handicrafts (especially by women), and sales of fish and agricultural products at Juara's

markets.









In the Rio dos Peixes region there is still game availability, mainly wild pigs. Almost all the

families receive bolsa-familia4 from the health district. Amongst the elders, some receive

retirement pensions from the Brazilian government.

There are few elders in the village, and they are the ones who still retain traditional

knowledge: they still speak Kaiabi language and know how to produce traditional crafts. Tafut

(around 89 years old) is the oldest person interviewed in all four Kaiabi villages and the last

shaman alive in Rio dos Peixes. He has several relatives in Xingu Park, but he said he is too old

to travel. Unfortunately, he has a problem in his eyes and does not weave baskets or do other

types of handicrafts. Simao (around 65 years old; his Kaiabi name was Kwapdn when he was a

child), is one of the few men in Rio dos Peixes who still weaves designed baskets. He was born

at Wyrapatyp5 village in the Batelao river region, the same village where Kupeap from Capivara

village was born. Most of Simao's relatives in Xingu live in Capivara village. Luiz Pedro (his

Kaiabi name is Matariowy and he is around 79 years old), another elder, lived several years in

the Utiariti Catholic mission. He does not do handicrafts, but he still speaks Kaiabi language and

cultivates traditional crops. His relatives in Xingu also live in Capivara village. Among the

women, Catarina (Katumait, 77 years old) is one of the oldest in the village. She knows how to

weave hammocks and keeps a homegarden where she plants cotton and other crops, including

peanuts. Paulina (Morekatu, 69 years old), Luiz Pedro's wife, is fluent in Kaibi language and

knows how to weave baskets and hammocks. Katu (68 years old) is the sister of Corone (an elder

who lives in Kururuzinho, Teles Pires), Tamanauu (elder and political leader in Ilha Grande




4 Brazilian government sponsored program for low income families.
5 Wyrapatyp means "place where there is a lot of sii a", the palm wood used by Kaiabi men to make bows.
Wyrapat=bow; typ=resource concentration.









village, Xingu Park) and Jeru'a (mother of the man who was assassinated in Rio dos Peixes see

page 82, Chapter 4). She is also fluent in the language and knows how to weave textiles.

Kururuzinho Post and Village in Teles Pires

Kururuzinho Post is located some five hours down Teles Pires river from the Port of Sdo

Benedito River. There were around 155 people living in the village in 2007, belonging to nine

main families (see Figure 7-9 for a map of the village in 2007, drawn by Tangeu'i, one of the

teachers). The basic infrastructure comprises a diesel energy generator, a school (needing

renovation), a system for water supply from a nearby stream, a health unit, a radio room, public

bathrooms, two motor boats and a big house for lodging and meetings. The village's school is

still linked to the municipality of Jacareacanga in Para, which has posed many difficulties in its

administration by the Kaiabi.

Similarly to what happens in Rio dos Peixes, there are interethnic marriages between the

Kaiabi, Apiaka and Munduruku peoples. In all cases, these marriages occur between a Kaiabi

woman and an Apiaka or Munduruku man. The children would normally speak the mother's

language, but in this case and also in Rio dos Peixes, interethnic marriages would increase the

erosion of knowledge of native languages, taking into account that it is easier for the child to

communicate with his parents in Portuguese (I will go back to this issue in the next section,

related to language proficiency comparing the three Kaiabi groups). The chief of the village is

Jywatu (61 years old), son of Kun, a very important leader who refused to go to Xingu. Jywatu

or Atu (nickname) is the older brother of Jodo (41 years old, kaiabi name Mairawi), indigenous

health agent in the village.

Households differ in composition and social organisation. There are more traditional

families, in which Kaiabi language is still spoken, and extended family social ties and obligations

observed, which is the case of Corone's family (Eroit's father in law). In contrast, there are









families, especially in the case of interethnic marriages, in which an atomization of households

around a nuclear family also happens, comparable to the Rio dos Peixes situation. However, this

is not as widespread as it is in Rio dos Peixes, where the majority of nuclear families live in

separated houses.

In Kururuzinho, the Kaiabi maintain a system of family agricultural plots located in

different sites around and away from the village. In comparison to Rio dos Peixes, families still

consume more traditional than industrialized food bought in the city. However, there are some

families, especially of waged officers, that depend more on industrialized food and therefore do

not pursue big agricultural plots for traditional crop cultivation.

Similar to Rio dos Peixes, the local economy is a mixture of subsistence and cash

oriented activities. Most men fish and hunt, and some get involved in temporary work, mostly in

nearby fishing resorts. Waged workers are mainly employed by the health service (FUNASA) or

as school teachers. Some men also sell Brazil nut in the town of Alta Floresta. A few families are

involved in handicraft production for commercialization to tourists who visit the village or for

sales in nearby towns such as Alta Floresta, especially Corone's family and Joao's family. In

Corone's family, most women make palm fruit (tucumd) jewelry and other seed jewelry for sale.

Joao's family specialty is the clubs with woven handles and baskets. Joao is a great artisan in

Kururuzinho, and he is also teaching and encouraging his children to learn basketry. Some

families prepare cassava flour and other agricultural products for occasional commercialization

in the towns. Many families have applied for and are receiving retirement pensions from the

Brazilian government.

There are fewer elders in Kururuzinho in comparison to Capivara, Tuiarare and Rio dos

Peixes villages. Amongst them, Corone (74 years old) and his wife Ryp (67 years old) are the









eldest, followed by Andre (Pyreai, 69 years old), the chief Jywatu (61), Maria (61) and Albertino

Munduruku (60). There is no shaman in the village, so when people need shaman treatment, they

have to travel to Xingu Park. This has happened several times with Eroit and his family; Chico

used to live in Kururuzinho, but his shaman's powers are limited and he has been living in

Tuiarare (Xingu) in the last couple of years.

Comparative Socioeconomic Analyses of Four Villages

People interviewed and Age Classes

For this research, I interviewed a total of 224 people, divided between 110 men and 114

women (Table 7-1). I visited 78 households, 13 in Capivara, 14 in Tuiarare, 37 in Rio dos Peixes

and 14 in Teles Pires (Kururuzinho). A similar number of men and women were interviewed in

each village: between 22 and 28 in Capivara, Tuiarare and Kururuzinho, and 36 men and 40

women interviewed in Rio dos Peixes, where the population is bigger. An estimate of the total

number of residents for all villages in 2007 would sum 728 people. A rough estimative of the

people interviewed in relation to the total of residents, including those under 15 years of age,

gives us 31% If we subtracted the number of residents under 15 for all villages, this percentage

would certainly be much higher.

Six age classes were defined for people interviewed in the four Kaiabi villages, adapted

from Athayde (2003). Taking in account that the criterion for inclusion in the interviews was

men and women over 15 years of age (see Chapter 2), the age classes below 15 years were not

represented in the age pyramids constructed for this study (Figure 7- 10). In all villages, the

number of people interviewed in each age class reflects the current population structure of the

Xingu Kaiabi, previously studied by Pagliaro (2005). She found that after thirty years, from 1970

to 1999, there was a reversion from a demographic regime of high levels of mortality and birth

rates, characterizing stability or even population decrease, to one in which the mortality declines









and birth rates increase and are maintained high, leading to a phase of population growth and

rejuvenation (Pagliaro, 2005:218, see Figure 7-11 for age pyramids for the Xingu Kaiabi over a

thirty year period). Pagliaro affirms that demographic studies carried out with other Brazilian

indigenous peoples (such as the Mucajai Yanomami, the Xavante, the Waura and the

Tenetehara) show a growth in the proportion of the population under 15 years of age, which is a

sign of increase in population growth rates and population recovery. The age pyramids found in

this study are also consistent with the distribution of the indigenous population in rural areas in

the center-western region of Brazil (IBGE, 2005).

From the age pyramids for the people interviewed (Figure 7-10), we can possibly infer

the same tendency of population growth as documented in Xingu also occurring in Rio dos

Peixes and Teles Pires areas: more people in the lower age classes, gradually decreasing to the

top. In Xingu, there are fewer elder women in the age class of 66 years old or older in

comparison to the other two areas. Looking to the age pyramids presented by Pagliaro (2005,

Figure 7-11), it is possible to verify that this tendency of fewer women in the age class over 60

years old was very significant in 1970 and 1979, gradually reversing towards 1999. This might

be related to the higher mortality of women after the transfer to Xingu, due to health

complications, stress, and adaptation problems faced with the transfer. It seems that in the case of

the Kaiabi, men have greater than women, as there are very few elderly females, especially in

Xingu Park.

Concerning gender and age, in first world countries (and also in some third world

countries such as Brazil) women show greater longevity in comparison to men, the opposite of

what happens with the Kaiabi. This might be related to several factors such as the high mortality

of Kaiabi people before the transfer to Xingu due to measles epidemics; problems of high









fecundity of indigenous women (making them more succeptible to labor problems or

gynecological diseases); previous conflicts with rubber tappers; and maybe more susceptibility to

some non-indigenous diseases. However, there is not enough data or information available to

support these inferences.

The pyramid format, in which the base is bigger, gradually decreasing towards the top

containing older individuals, is also a characteristic of the Brazilian population as a whole and of

third world countries in general. In first world or industrialized countries, the population

pyramids do not show such a pyramidal distribution. For instance, in European countries in

general, the population distribution graph shows an enlargement in the middle sections with

smaller sizes in lower and upper portions (US/IDB, 2008). This means lower birth rates and also

death rates, so the population is relatively stable or even declining.

Places of Birth

From the results reporting places of birth for the interviewed people, it is possible to

confirm the close relationship of Capivara residents with Rio dos Peixes and that of Tuiarare

residents with Teles Pires (see Figure 7-12). However, in Capivara this relationship is greater

than in Tuiarare, with 57% of people born in Capivara and 39% born in Rio dos Peixes. In fact, a

great part of the population over 35 years of age in Capivara was born in Rio dos Peixes,

belonging to the nuclear and extended family of the great chief Temeioni. In Tuiarare, this

proportion is smaller: 60% of people were born in Tuiarare compared to 29% born in Teles Pires.

In Capivara, only one person was born in Teles Pires (2%) whereas in Tuiarare, 3 people were

born in Rio dos Peixes (6%).

There is a similar trend observed in Rio dos Peixes and Teles Pires, concerning people

who were born in Xingu Park and later on moved to these areas. In Rio dos Peixes, 14% of

interviewed people were born in Xingu, whereas in Teles Pires this percentage is 13%. The great









majority of people born in Xingu who later moved back to the ancestral areas are men. This

aspect will be discussed in the following sections, but shows us important differences between

genders that have influence on social-economic organization and on the dynamics of knowledge

systems among the three areas.

Interethnic Mixing

Interethnic marriages between the Kaiabi and other indigenous groups or non-indigenous

persons have happened in greater intensity in villages in Rio dos Peixes and Teles Pires in

comparison to Xingu villages. In Xingu, the only descendant of an interethnic marriage is Tapi,

representing 2% of the people interviewed, who lives in Tuiarare village and is an indigenous

health agent. In Capivara, there is no interethnic mixing among the interviewed people. I know

one couple there in which the woman is Yudja, but they were absent from the village at the time

I was conducting interviews.

In Rio dos Peixes, 13 people (17%) are from another ethnicity or descendants from

interethnic marriages. This total comprises people of Munduruku, Satere-Maue and non-

indigenous origins. There are three people descendants from unrelated Munduruku parents who

got married to Kaiabi women or men, such as Otavio, born in Teles Pires, son of Maria with a

deceased Munduruku man, who travelled to Rio dos Peixes and ended up getting married and

staying there. The other two are descendants of Munduruku families who live in Rio dos Peixes

area. There are four people of Satere-Maue origin, related to Adelaide, a Satere-Maue woman

married to Ataide Kaiabi (son of the deceased chief Chico). The other six people are relatives of

Jocelina Franca, a non-indigenous woman married to Jodo Franca Kaiabi.

In Kururuzinho, the mixing between Kaiabi with other ethnicities is surprisingly greater

than in Rio dos Peixes: 20 people (37%) there are of Munduruku, Apiaka or of non-indigenous

origin. These people often belong to a family in which the head is a Munduruku or Apiaka man









married to a Kaiabi woman. This is the case of Vitorino and Albertino Munduruku and their

families, both of them pursuing their own villages or "ranches" apart from Kururuzinho Post and

village, but also keeping a house in Kururuzinho, spending time there for health treatment,

participating in meetings or for any other reason. Apiaka people in Kururuzinho are related to the

family of Fernando (five people), married to a Kaiabi woman, who also keep a ranch or village

apart from the post. In Kururuzinho, there are only two people descendant of non-indigenous

mixed with Kaiabi.

It is curious to note that in Rio dos Peixes the trend is the opposite of Kururuzinho, with

mixing occurring through marriage between a non-Kaiabi woman with a Kaiabi man. This might

have, among other consequences, a particular effect on the language proficiency of the children,

because amongst most Amazonian indigenous peoples, the children generally speak the language

of the mother (Rodrigues, 1986). In Teles Pires, the effects of epidemics, mobility, transfer to

Xingu and involvement with the Sdo Benedito mining company have all combined to result in

greater interethnic mixing. It is important to mention that in spite of belonging to other groups,

when it comes to political choices and decisions, the Apiaka and Munduruku married to Kaiabi

women often agree with the community in which they are living.

Interethnic mixing might affect cultural resilience in different ways. It might reinforce

tendencies towards the loss of proficiency in the native language, through the predominance of

use of Portuguese to communicate, since the parents would speak different languages. In

addition, it may also interfere and weaken processes of indigenous knowledge transmission,

since the parents would belong to two different cultural systems. When the extended family does

not accept the marriage, mixing would also lead to segregation and nuclearization of extended

families. It is worth mentioning that, in the process of cultural change, the greater the interethinic









mixing and the greater the sociocultural distance from the extended family, the closer indigenous

peoples get to the non-indigenous "caboclo" or "mestizo" way of life (Adams et al., 2008).

Montenegro and Stephens (2006), talking about the effects of interethnic mixing on

indigenous peoples' health in Latin America, use the term "relative isolation" to refer not only to

geographic barriers, but also to cultural and language barriers which might impede inter-ethnic

exchanges and enable mixings among indigenous peoples. Therefore, we can think of geographic

and cultural barriers present in Xingu which have prevented inter-ethnic mixing there in

comparison to the other lands occupied by the Kaiabi. This cultural barrier that I'm talking about

may be an inner natural defense of Kaiabi ethnicity against mixing, and thus losing their group

identity (Oakdale, 1996; 2005; Menezes, 2000).

Functions in the Village, Sources and Stability of Income

This section presents data on occupations and income in the four villages. Sources and

stability of income are variables used as proxies for degree of market integration. Eleven

functions or occupations were identified for people resident in Capivara, Tuiarare, Rio dos

Peixes and Kururuzinho villages/posts (Table 7-2). In Capivara there are seven types of

occupation, in Tuiarare nine and in Rio dos Peixes and Kururuzinho six each. The majority of

people (78.68%) in the four villages do not have a specific function or profession, especially

among the women. For Tuiarare and Kururuzinho, no women interviewed have a special

occupation. In Capivara, there are two women who participate in political meetings (8%), as

political representatives, and in Rio dos Peixes, four women (10%) are involved in waged work,

two as teachers and two as health agents. Whereas in Xingu women have increasingly claimed

and achieved more political power than in the other areas, in Rio dos Peixes women have come

to work as waged officers in education and health services, in positions normally occupied by

men in Xingu.









Men have been engaged in different functions in the villages, from more traditional and

informal ones such as shaman, represented by only one man in Tuiarare (Chico) and one in Rio

dos Peixes (Tafut), to formal waged positions such as indigenous teacher or indigenous health

agent. Currently, waged work in all sites is mostly related to health and education services. Other

occupations held by men in Xingu are beekeeper6 and other assigned informal functions such as

boat driver and radio operator.

The sources of income, or money-oriented economic activities for the people

interviewed, reflect the situation of occupations and professions. The majority of people (78%)

in the four villages do not pursue a regular source of income, thus depending on a combination of

different activities to earn money used for basic needs such as clothing, industrialized food

(especially in Rio dos Peixes and Kururuzinho), fishing materials (hooks and lines), and tools

(machete, axes, needles for sewing etc). Among the four levels of stability of income defined for

this research (Table 7-3), in Xingu there are only two levels: waged officers or irregular income,

with the exception of Tewit, who has retirement pension from the time he lived in Rio dos Peixes

(see below). In Capivara, the six people who have income stability (12%) receive from R$ 288 to

R$ 1.152 (U$144 to U$576) monthly. This village's total monthly income is R$ 3.450

(U$1,725). In Tuiarare, three men are waged officers and receive from R$ 380 to R$ 990 (U$190

to U$495) monthly, with a total village monthly income of R$2.280 (U$1,140). In Rio dos

Peixes, the amount of people on wages or any other type of regular income is greater: 31 people

(41%), receive from R$ 45 to R$ 1080 (U$22 to U$540), totalling R$ 7.879 monthly (U$3,939).

In Kururuzinho, there are 10 people (18.5%) with regular income, receiving from R$351




6 Related to the beekeeping project which ATIX and ISA have been carrying out since 1997 (see chapter 6).









(U$175) to R$390 (U$195), totalling R$ 3.629 monthly (U$1,814). Therefore, monthly income

in Rio dos Peixes is at least double the amount of all the other villages (Figure 7-13).

All women in Xingu present irregular or no stability of income, whereas in Rio dos

Peixes the majority of women (60%) have access to wages (10%), retirement pension (15%) and

family pension (35%). Thus, only 16 women out of 40 have no income stability (40%). In Teles

Pires, three women receive retirement pension (10.71%), whereas all others present irregular

income stability (89.29%). Whereas in Xingu only 11 people in both Capivara and Tuiarare

villages have income stability (11.70%), outside Xingu in Rio dos Peixes and Kururuzinho

villages altogether, 41 people have income stability through wages, retirement pension and/or

family pension (31.54%).

Related to this, the Pearson Chi-square (x2) correlating residence inside and outside of

Xingu with irregular versus regular (stable) income was significant (p<0.001) and the Pearson's

correlation was r=0.232 (p<0.001). This means that the distribution of regular or irregular

income stability is not a chance event, but depends on whether or not somebody lives in Xingu.

The estimated odds ratio for the relationship between living inside or outside Xingu and having

income stability or not for both genders was 3.76, which means that people living outside Xingu

have 3.76 more chances of having income stability than people living in one of the two Xingu

villages analysed (Capivara or Tuiarare).

Nine types of sources of income were identified among the Kaiabi in the four villages

included in this research (Figure 7-14). The main source of family income in all villages is

production and sale of handicrafts: it was mentioned by 67% of the people in Capivara, 71% in

Tuiarare, 75% in Rio dos Peixes and 59% in Teles Pires. People often rely on more than one

source of income and therefore combine sales of handicrafts, family pension and other types of









activities, such as temporary work, to compose the household monthly income, which can

fluctuate significantly throughout the year. Most women rely on sales of handicrafts as their

unique source of income or for exchange with other women. In all villages, women produce fruit

and seed jewellery for sale to tourists, researchers, health agents and occasionally middlemen

that come to the villages. Sometimes, they also send their production along with some relative,

generally a man, who travels to the cities and supposedly brings the money or products back to

them. In Rio dos Peixes and Kururuzinho villages it is easier for the women to travel to nearby

cities and personally sell their production. In Xingu, given the greater distance to nearby cities, it

is more difficult for women to travel, especially if it is not for health related reasons.

Another economic activity generally performed by men which is growing in all areas in

recent years is involvement in temporary work. It was mentioned by 18% to 24% of men in all

villages. This type of work may involve being hired by relatives or villagers to help build their

houses, or work for non-indigenous neighbours such as in fishing resorts (Teles Pires) or ranches

(Rio dos Peixes). Family support was also considered part of the sources of income, because

members of nuclear families of waged officers benefit from their monthly wages. This is

currently an important source of income for people in Capivara (37%), Tuiarare (33%) and Teles

Pires (29%) but not so significant in Rio dos Peixes (6%).

The fact that in Rio dos Peixes people rely less on family support might be related to two

aspects: first of all, as mentioned before, in that village there has been an ongoing process of

nuclearization of extended families and households, with each household becoming physically,

socially and economically bounded by a nuclear family: wife, husband and children. Some social

and economic consequences of this process may be exemplified by the Xingu Kaiabi referring to









the "Tatuy" as they name the Rio dos Peixes group: "there, they even sell fish to their own

relatives."

The closer contact with and reliance on the market economy by the Rio dos Peixes group

has brought changes in the social organisation and in the role of the extended family in providing

support for its members. Nevertheless, this process of nuclearization and social change derived

from greater contact with the market economy and by extension with the non-indigenous

lifestyle cannot be defined as linear, and is not widespread and homogenous among all families.

Among more traditional families in Rio dos Peixes which often are composed of people who

have lived in Xingu for a variable amount of time- the nuclearization process is not as visible as

it is with families that have always lived in Rio dos Peixes, or with families with mixed

ethnicities. For instance, Canisio and Morea'i, who moved from Xingu to Rio dos Peixes in 1998

and have most of their sons and daughters around them, still keep a more traditional extended

family system, even with most of their married children living in separate houses. From what I

could observe while I was living in their house, there is still substantial sharing of agricultural

production, fish and game amongst family members. While there are indications that this process

of family nuclearization is happening, a more in-depth understanding would require further

investigation of Rio dos Peixes social organisation.

The second factor related to less reliance on family support in Rio dos Peixes might be

linked to the greater access to family pension there, cited by 20% of people. The "Bolsa Familia"

(family pension), a Brazilian government program destined to support low income families, is

still absent in the other Kaiabi lands. In spite of the low monthly support provided by this

program (around R$ 100 or U$ 50 per month per household), it makes a difference for the

families in Rio dos Peixes, who use the money generally for buying clothes and industrialized









food in Juara. The same happens with retirement pensions. In Rio dos Peixes, 10% of people

receive retirement pension from the Brazilian government whereas in Kururuzinho this

percentage is 13%. In Xingu, normally the elders do not receive retirement pensions. In

Capivara, the only person who receives retirement pension is Tewit, who was living in Rio dos

Peixes and got his paperwork done there. Getting retirement benefits by age is a right of every

Brazilian citizen (65 years of age for men and 60 for women). However, the paperwork involved

in the process of acquiring the benefit is mountainous and dependent on the municipality to

which the village is officially linked.

In Rio dos Peixes, the number of sources of income is greater compared to all the other

villages. Despite being a seasonal activity, Brazil nut extraction and commercialization is an

important source of income there, practiced by 35% of men. In 2006, they harvested around 50

tons of Brazil nut, whereas in 2007, the harvest was around 70 tons, according to Sebastiao

Kaiabi. On average, each family collects around 40 kg/month during the main season, which

ranges from October through January.

The sale of agricultural products is also greater in Rio dos Peixes, practiced by 22% of

the people interviewed. This is related to the closest proximity of the village to the market, in

comparison to the other villages. In Kururuzinho, sale of agricultural products was mentioned by

7% of people, and it is mostly related to sale of manioc flour in Alta Floresta. In Kururuzinho,

the lower investment or opportunities in market oriented activities might be related to the greater

distance to the closest town (Alta Floresta), but also to the fact that the village receives financial

support from fishing resorts, which sometimes is used to buy industrialized food, clothing and

other basic needs for the families. Finally, it is curious to note that honey production and

commercialization appears as a source of income in Capivara and Tuiarare villages at Xingu









Park. The honey production project began around 12 years ago with ISA's support, and its

administration hs been gradually transferred to the Kaiabi, through a honey cooperative. They

have sold the product to different buyers (including Pdo-de-Acucar, a big supermarket in Sao

Paulo), and part of the money from the sales goes back to the villages to be reinvested in

materials and sometimes to pay the work of the beekeepers. Therefore, this is a relatively new

source of income that already appears in the data collected, along with the other established ones.

Travel and Dislocations Between the Villages

This section presents information on travel and dislocations between Kaiabi villages

according to interviewee's estimate of the number of trips taken to the other locations and the

main reasons for these travels. When people move or travel, they often bring new information

and knowledge, which is transmitted to others in the new location. Exploring the ways in which

movement of people, knowledge and sometimes natural resources (such as crop seeds for

instance) occur is important to understand the mechanisms through which indigenous knowledge

is transmitted and distributed, and other cultural and political interactions among Kaiabi groups.

Some specific questions which could be asked here are: How do the villages in and outside of

Xingu interact? What is the frequency of travel between these villages? What are the main

reasons for these trips? Which factors have enabled more contact between the three Kaiabi

groups?

One of the unexpected results of this research was the realization that there is a lot of

interaction and contact between the three Kaiabi groups, more than I expected or was aware of.

The increased contact between Xingu Park and the other areas, especially since the 1990s, can be

tied to factors such as: a) improvement in infra-structure for travel in the region, through the

construction of roads and opening of new bus lines; b) the increased access to financial resources

by Kaiabi communities coming from wages, funding for political meetings, training









opportunities, projects, land demarcation struggles, etc; and c) the greater access to funding

opportunities provided mainly by health and environmental organizations (in Teles Pires, also

through the funds provided by fishing resources). Kaiabi communities have been able to acquire

trucks and boats, as well as the expensive fuel needed for the distant travel.

Similarly to what we have found out concerning political participation, functions in the

village, and stability of income, there is a significant difference between genders regarding

frequency of and reasons for travel between the four Kaiabi villages analyzed (Figures 7-15 and

7-16). As expected, men travel much more than women, and for different reasons. In Xingu,

there is more travelling between Capivara and Rio dos Peixes villages than between Tuiarare and

Rio dos Peixes or Teles Pires. In Capivara, 42% of women and 72% of men mentioned that they

have travelled to Rio dos Peixes, totalling 57% of people interviewed. In Tuiarare, men have

travelled both to Rio dos Peixes (39%) and Teles Pires (30%), but fewer women have travelled,

totalling 4% to Rio dos Peixes and 13% to Teles Pires. The increased number of travels between

Tuiarare and Tele Pires is related both to kinship ties and also to the process of land demarcation

in which the Teles Pires Kaiabi have been involved and have asked their Xingu relatives to take

part.

In Rio dos Peixes, the majority of people who have travelled to Xingu went to Capivara

village (47%), and the difference between women and men travelling is smaller in comparison to

all other villages: 37% of women and 58% of men. Few trips were taken by Rio dos Peixes

residents to Kururuzinho village (5% of interviewees), which shows that the ties between Rio

dos Peixes and Teles Pires areas are weaker than between these areas and Xingu. Kururuzinho is

the place with greatest number of interviewees travelling: 65% of people travelled to Xingu and

a surprising percentage of 52% of people have travelled to Rio dos Peixes. Also, in Kururuzinho,









the difference between men and women travelling is not so big in comparison to Tuiarare, for

instance. There, 50% of women and 81% of men interviewed had travelled to Xingu. It seems

that it is easier for people in Teles Pires to travel to visit their relatives in Rio dos Peixes than

vice-versa. This might be linked to the acquisition of the community truck in Kururuzinho, as

well as to opportunities and demands for travel, such as relatives' sicknesses or political events.

Also, it is worth clarifying that the travels from Teles Pires to Xingu have targeted mainly Ilha

Grande village, where the majority of Kururuzinho's relatives live. However, a significant

number of trips have been taken to Tuiarare village. Some families travel more than the others,

especially when there is a couple married with split relatives between the two areas, such as the

case of Pofat and Rosilda. Pofat's parents (Kupeianim and Re'a) live in Xingu, and Rosilda's

mother lives in Rio dos Peixes (her dad is deceased and her mom is Maria, not interviewed).

Seven main reasons were identified for people to dislocate between the areas. In addition

to travelling with more frequency, men also travel for a more varied set of reasons, as we can

observe in Figure 7-16. The main reason for travelling in all four villages is to visit relatives:

77% of people in Capivara, 79% in Tuiarare, 93% in Rio dos Peixes and 82% in Kururuzinho.

Another important reason for travels between Xingu and the other two areas is to collect and

bring natural resources unavailable in the Park, needed for the sociocultural reproduction of the

Kaiabi (see Chapter 5). It is worth highlighting that these reasons for travel often happen

together, especially in the case of Capivara village: you go visit your relative and take advantage

of the travel to bring some natural resources back to Xingu. Travel for political reasons was also

mentioned by a significant number of people, mainly in Tuiarare and Teles Pires areas, related to

land demarcation struggles. Other reasons given for travel were to work, due to marriage with

people from other villages, and to participate in courses and projects.









Based on peoples' testimonies, it is curious to note that these travels may last from a few

weeks to months and even years. Sometimes, people travel to visit a relative and end up getting

married in the other place, returning to their original villages to gather belongings and

permanently moving to the new village. This happened with many of Canisio's sons, who used

to live in Xingu, but ended up getting married in Rio dos Peixes, permanently moving there.

People may use money coming from wages to travel, or they might save money from temporary

work or sale of products. Sometimes, they travel with support from a relative, and other times

people are sponsored by projects (such as training of indigenous teachers and Kaiabi Araa

project) or by funding available for political struggles (such as land demarcation struggles in

Teles Pires).

The results show that kinship is the single most important tie linking the three Kaiabi

areas. Even with the isolation and lack of contact that happened in the years immediately

following the displacement, existing kinship ties have been re-strengthened and new ones have

been constructed, mainly through marriages between people from Xingu and the other two areas.

Also, political and land demarcation struggles as well as training events, have created new

opportunities for interaction and exchange of information between the areas. During these visits

and movements between the three areas, a great deal of exchange of information, knowledge,

resources and skills occur, which might have affected the distribution of knowledge of different

cultural domains among the three areas. This idea will be revisited in Chapter 8, in the discussion

on acquisition and distribution of weaving knowledge among the Kaiabi.

Formal Schooling

School is a western science-based institution with which indigenous peoples around the

world have interacted since the establishment of regular contact with westerners. In Brazilian

Amazon, as well as in other Amazonian countries such as Peru, schooling and literacy among









indigenous peoples often began through the Catholic church and the work of national and

international missionaries (Aikman, 2003; Grupioni, 2006). In that context, the school was one

of the main instruments used for assimilation of indigenous peoples into the Brazilian society, in

consonance with the national policy towards indigenous peoples in place until the constitutional

reform of 1988 (Ramos, 1998, see Chapter 6 for details). Therefore, at that time, formal

education available to indigenous peoples was (and still is in some places) an instrument of

cultural homogenization in which indigenous groups would abandon their native languages,

lifestyles, knowledge systems and differentiated identities in favour of the predominant national

system (Grupioni, 2006).

According to Grupioni (2006), one of the strategies used to impose the supremacy of

western knowledge and of the national language (Brazilian Portuguese) over indigenous

languages and knowledge systems was to create indigenous camps ("internatos") where

indigenous children would remain for various years, being educated in Portuguese and in which

the use of native languages was forbidden. This happened with the Kaiabi through the Utiariti

Catholic mission (Grunberg, 2004). Many men and women, especially at Rio dos Peixes, studied

at the Utiariti mission in Cuiaba. Some missionaries of the Summer Institute of Linguistics (SIL)

have also remained in Rio dos Peixes for different periods of time studying the Kaiabi language

and producing materials for the study of the language, as well as a translation of parts of the

Bible in Tupi-Guarani.

Starting in the 1970s, the process of national re-democratization triggered by the rise of

Brazilian civil grassroots movements, and among them, indigenous grassroots movements,

brought awareness of the need for a reform of Brazilian policies regarding formal education's

objectives, contents and methods directed to indigenous peoples. For Grupioni (2006), the main









idea which has been advanced since that time is that the school as a western institution could be

appropriated and adapted by indigenous peoples, who would bring new meanings and objectives

to it, transforming the school into an instrument in their favour.

Through the Brazilian constitution of 1988, indigenous peoples were granted the right to

remain Indians by holding rights over their lands, identities, social organisation, lifestyles,

languages and knowledge systems (Ramos, 1998). The constitution also advanced the

recognition that indigenous peoples could use their native languages and learning mechanisms in

the process of formal schooling. Since then, other policy instruments have been developed,

through the Ministry of Education, towards the implementation of an indigenous educational

system in which the Indians could be trained to work as teachers in their communities, and,

which could ultimately contribute to the maintenance of indigenous cultural diversity rather than

to the outdated idea of cultural homogenization. However, in spite of the development of a

national system for indigenous education, there are still many challenges and constraints to be

overcome in the implementation of such a system in Brazil (Grupioni, 2006).

According to Aikman (1999:19), intercultural and bi-cultural education are terms that

have been used throughout Latin America to refer to indigenous educational programs whose

"main aim is to use education as a means of strengthening indigenous language and culture and

of providing the skills and competence needed for fuller participation in the wider society."

Aikman states that intercultural education in Latin American countries has been predominantly

conceived for primary schooling, in which the foundation of the learning process is literacy, from

which the concept of curriculum derives. Since indigenous traditions and knowledge are based

on mostly oral and experiential learning, indigenous formal schooling, even under the

intercultural premise, is founded on non-indigenous structures, thus bringing challenges to the









full appropriation of this western institution among indigenous peoples, comparable to the

appropriation of indigenous associations (Chapter 6).

The Kaiabi case is a good example of the enormous contrasts still in place regarding

indigenous education in Brazil, ranging from the assimilation to the intercultural education

paradigms. School has been appropriated differently by communities residing in each Kaiabi

area, affecting local knowledge systems including language proficiency in the native language -

in different ways. Schooling began with the work of Catholic missionaries (Anchieta Mission or

Missdo Anchieta) in Rio dos Peixes in the 1960s. Father Dornstauder used to send children and

youth to undertake health treatments and also to study in the Utiariti mission in Cuiaba. In

December of 1965, Georg Grunberg (2004) registered 23 people (children, youth and two

families) living in Utiariti. After the transfer to Xingu Park, people who remained in Rio dos

Peixes continued to have access to education, until nowadays, but no bi-lingual intercultural

education process was ever implemented. Therefore, the educational system still available for the

Rio dos Peixes community probably hasn't changed much since the missionaries, with the

extensive use of Portuguese language and the curriculum reflecting western science concepts.

Formal schooling at Xingu Park started in the 1970s through the creation of the first

school at Leonardo Post, administered by FUNAI and with a non-indigenous teacher. In

Diauarum Post, the school was implemented in the 1980s, also under FUNAI's auspices, but

working in an intermittent way, always depending on non-indigenous teachers hired by FUNAI

(ISA, 2009 d). At the end of the 1980s, most schools in PIX posts were inactive, due to the lack

of teachers. Therefore, some young men who had studied in these schools began to informally

teach Portuguese language classes in their villages. At that time, they did not have any

institutional support. In 1994, the Fundaco Mata Virgem (NGO administered by the Norwegian









Rainforest Foundation-NRF, see Chapter 6) started to conduct the first course or process of

training indigenous teachers in Xingu Park. This project was then taken over by Instituto

Socioambiental in 1996, with support from NRF, SEDUC-MT (State Secretary of Education

from Mato Grosso), MEC (Ministry of Education) and FUNAI. The project's main mission was

to develop a multi-lingual intercultural education process, valuing the diverse indigenous

cultures represented in the Park, combining western and indigenous knowledge and providing

literacy in both indigenous and Portuguese language (ISA, 2009 d). Between 1999 and 2003, 38

teachers were trained by the projects. Among them, 19 began to attend the first Brazilian

Indigenous University program implemented in 2001 by the UNEMAT (Universidade Estadual

do Mato Grosso, State University of Mato Grosso) (UNEMAT, 2008).

ISA's project for training of indigenous teachers at Xingu Park became a national model

for indigenous education in Brazil, training 81 teachers of 40 schools and nearly 1,358 students,

wining various awards and prizes. Through the project, alphabets for some of the 14 languages

spoken in the Park were created, writing skills of teachers in languages with existing alphabets -

such as the Kaiabi were developed, and 25 educational books were organised (ISA, 2009 d).

Kaiabi teachers from Kururuzinho village (Eroit and Arlindo- Tangeu'i) at Teles Pires have also

participated in some courses through this project. Some indigenous teachers trained through this

process have also become salient political leaders in their communities, such as Aturi

(Jowosipep) Kaiabi, the protagonist of the educational process of weaving revitalisation through

the Project Kaiabi Araa.

Formal schooling is a powerful institution that might affect the maintenance, change

and/or erosion of indigenous knowledge systems (Nabhan and St. Antoine, 1993). Therefore, a

comparison of the level of schooling in each of the four Kaiabi villages provides crucial









information for understanding the processes and mechanisms through which indigenous

knowledge is perpetuated or eroded. Beyond the exploration of the differences in formal

schooling in terms of gender, age and level in each village, I'm interested in understanding how

the level and type of schooling might affect and be affected by other socio-economic variables,

such as income stability and language proficiency. These multiple comparisons are developed in

the last section of this chapter.

For this research, four categories were assigned to estimate the level of formal schooling

among the interviewees: 1) no instruction; 2) elementary (corresponding to Brazilian "1. grau);

3) secondary (corresponding to Brazilian "2. grau"); and 4) university (corresponding to

Brazilian "grau universitario"). For some statistical analyses such as chi-square, correlations and

logistic regression, I transformed the schooling variable into two levels: 1) no instruction; 2) any

level of formal instruction. I did this because in some of the four categories established initially

there were very few people, making it difficult and even impossible to run some analyses. People

were assigned to the levels in which they had at least one year of instruction, but this does not

mean that they had completed that level.

Again, the results show similar patterns between Capivara and Tuiarare villages in Xingu,

with Teles Pires and Rio dos Peixes showing greater differences in terms of the number of

people in each level, and gender contrasts (Table 7-4). The big difference between Capivara and

Tuiarare villages is that in Capivara there are many more women having elementary schooling

(42%), whereas in Tuiarare there is only one (5%); therefore 95% of women have no instruction

in that village. This is probably due to the fact that some girls have attended school at Diauarum

Post, which is closer to the village. Amongst the men, the results are similar for both Xingu

villages, and balanced between no instruction and elementary: 48% of men in Capivara have no









instruction, against 44% in Tuiarare, and 44% of men in Capivara and 39% in Tuiarare have

attended elementary school. Nobody attended secondary school in Capivara, and only three men

did in Tuiarare. These are sons of waged officers who were sent to study in nearby cities, in this

case, So Jose do Xingu. The only three people attending university in all villages are two

teachers from Capivara village (Jemy and Sirakup) and Aturi (Jowosipep), teacher and chief of

Tuiarare village, who were initially trained by ISA's education project in Xingu Park.

In Rio dos Peixes, there is a balance between genders in terms of formal schooling: 75%

of men and 72% of women interviewed have attended elementary school. Very few people

(roughly 5% of both men and women) have had secondary education, and none have attended

university. Most people who finish the fourth grade of elementary school in the village attend

subsequent grades at Aguas Claras, a nearby town. The school's curriculum follows the

guidelines for public school education set up by MEC in Brazil. According to the teachers

Esmeraldo and Casarina, they have tried to include some aspects of Kaiabi culture in the

curriculum, but the general approach is to replicate the same education systems present in public

schools outside indigenous lands. They also have religion classes, inherited from the strong

presence of the church in the village since the 1970s. There is no literacy training in the Kaiabi

language, apart from one time slot in which they teach aspects of the Kaiabi language to

substitute for an English class, since neither teacher speaks English.

In Kururuzinho, there is also a balance between genders concerning attendance at the

elementary level of schooling: 61% of men and 75% of women belong to this group. Only four

men have had access to secondary education and nobody has had university training. When the

school was still working, approximately five years ago, the teachers Eroit and Tangeu'i tried to

replicate the same model of differential indigenous education developed in Xingu, from the









experiences they had of participating in some training courses there. However, they had no

support or monitoring of their work, which makes it difficult to evaluate the quality of education

for people in the village. They have adapted the curriculum and used the books developed in

Xingu Park, and have provided literacy training in the Kaiabi language for the students. The

community has been fighting for four years to transfer the school's administration from

Jacareacanga municipality in Para to the State Secretary of Education of Mato Grosso (SEDUC-

MT). Meanwhile, the school's activities have been suspended, and only in October of 2007 did

the school finally pass to Mato Grosso State administration; it was going to resume classes in

2008. The few waged workers in the village often send their kids to study and live in Alta

Floresta.

The Chi-square relating village and schooling ( 2=36.493) using two levels for schooling

was significant (p<0.001). This means that level of schooling is statistically different by village.

We can group the Xingu villages together, where people have less formal instruction, especially

women, and then cluster Rio dos Peixes and Teles Pires showing more people with formal

schooling, especially at the elementary level. Formal schooling is negatively correlated with age

(r=-0.573, p<0.001): the older people are, the less formal schooling they have. For Rio dos

Peixes and Teles Pires villages, where there are people with no proficiency in the Kaiabi

language, the Pearson's correlation is also negative (r=-0.298, p<0.001), meaning that the more

formal schooling people have, the less proficient in the native language they are. The test for

correlation between schooling and stability of income was not significant.

Taking into account the differences in curriculum and type of education between Xingu

Park and mainly in Rio dos Peixes (in Kururuzinho they have tried to replicate the Xingu

experience), it is hard to compare even the elementary levels of schooling between the two areas.









In Xingu, besides the educational program, there have been various other training and capacity

building activities and opportunities. ATIX has also participated in the administration of

indigenous schools at the Park, making the connection between indigenous education and

indigenous politics stronger. The presence of an intercultural education program in Xingu has

also certainly contributed to the valorisation and perpetuation of different aspects of Kaiabi

traditional knowledge, within various cultural domains. This suggests that we should not

underestimate the effects of institutional arrangements on the cultural resilience of Amazonian

peoples.

Language Proficiency

Brazil is a country of great linguistic diversity, with an estimated 190 indigenous

languages, the majority of these represented in the Amazonian region (ISA, 2009 c). According

to Rodrigues (1986), there are two main linguistic stocks in Brazil, the Tupi and the Macro-G6.

Besides these, there are another 19 linguistic families which are not grouped in stocks. The Tupi

stock is divided in ten families, which are subsequently divided in languages and dialects

(Rodrigues, op cit).

The Kaiabi speak a language of the Tupi stock, in the Tupi-Guarani family, the biggest

family of this stock, with nearly 20 languages. According to the recently released "Interactive

Atlas of the World's Languages in Danger" organized by UNESCO (UNESCO, 2009), all

indigenous languages in Brazil are considered to be in danger of disappearing, fitting in one of

the five levels of endangerment defined by the organization: 1) unsafe, 2) definitely endangered;

3) severely endangered; 4) critically endangered; 5) extinct. The Kaiabi language is considered

"unsafe" in this classification, with approximately 1,000 speakers.

Language is a crucial component of indigenous knowledge, and has been used as a proxy

for estimating cultural diversity at different scales (Hunn, 2001: Maffi, 2001; 2005). As Maffi









(2001:6) points out, "while not all knowledge may be linguistically encoded, language does

represent the main tool for humans to elaborate, maintain, develop and transmit knowledge."

When a language is moribund or extinct, there is an irreplaceable (and many times irrecoverable)

loss of knowledge such as historical, spiritual and ecological knowledge that has been embodied

in the language for centuries (Hill, 2001; UNESCO, 2009). The loss of this patrimony represents

not only a loss or shift of a specific group's identity, but also diminishes the adaptational strength

of humans, since it lowers the pool of knowledge and information from which we can draw

(Bernard, 1992). In other words, it means a reduction in the world's "human" capital.

Some of the questions that I want to answer with this research relate to the extent to

which knowledge is being maintained or lost among the four Kaiabi villages, and broadly

between Xingu residents and non-Xingu residents. Is the native language being lost outside

Xingu? If the answer is positive, to what extent and what are the possible causes? What is the

relationship between language proficiency and the other socioeconomic aspects that characterize

each of the areas considered in this study?

The measure of language proficiency for people interviewed in four Kaiabi villages

included a scale of four levels: 1) no proficiency (does not understand nor speak); 2) low

proficiency (understands but doesn't speak); 3) moderate (understands and speaks, but not

fluent); and 4) fluent. For some of the statistical analyses (similarly to what was done for

schooling), I reduced the variable language proficiency into two levels, joining levels 1 and 2 in

"no or low proficiency" and 3 and 4 in "proficient". In spite of the relative arbitrariness of this

method, it does give us an estimative of how much the native indigenous language has been

maintained or lost among the three Kaiabi lands. However, since in Xingu villages (Capivara and

Tuiarare) all the residents show from moderate to fluent levels of proficiency, some of the









statistical analyses, including the logistic regression, were run only for Rio dos Peixes and

Kururuzinho villages, leaving Xingu out.

From the results, we conclude that Kaiabi language in Rio dos Peixes has been eroded at

a faster pace in comparison to Teles Pires and Xingu (Table 7-5). In Rio dos Peixes, the majority

of people interviewed show low proficiency in Kaiabi language (66%). Only 30% of women and

39% of men are proficient. According to Lourdes, one of the oldest women in the village and

mother of twelve kids, the nuns and priests from the "Missdo Anchieta" Catholicc church, CIMI)

taught Portuguese classes to the community since the time when Father Joao Dornstauder left, in

the 1970s:"there were Portuguese classes, there was a school, the nuns giving classes.., my

daughters understand the language but they don't speak, I don't know if they are ashamed to

speak." Esmeraldo and Cesarina, teachers at Rio dos Peixes school, are also not fluent in the

Kaiabi language.

The situation in Kururuzinho (Teles Pires), in spite of being better than in Rio dos Peixes,

is also reason for concern regarding language erosion: only 50% of the men and 39% of women

are proficient. I ran a chi-square test to verify if there was a statistical correlation between gender

and language proficiency, but it was not significant. In both places, the most frequent level of

proficiency among the non-proficient people is the low level, which indicates that it is easier to

understand than to speak the language. In fact, from my experience working with the Kaiabi and

with indigenous peoples in general, a person will only be considered fluent in the language if he

or she has learned it since childhood. It is very shameful and reason for mockery among them to

speak the language or pronounce words in the wrong way. Even if the person can speak

relatively well, he or she will refrain from speaking, especially in public or in the presence of

elders.









In Xingu, it seems that the physical (geographical isolation) and cultural barriers are

preventing people from losing proficiency in Kaiabi language. In both Xingu villages, the

number of people who present a moderate level of fluency is really low: only three people in

Capivara and one person in Tuiarare. Therefore, in Capivara, 94% of people are fluent and in

Tuiarare, 98%. The few cases of people who are not fluent in both villages relate to people

coming from Rio dos Peixes (even in Tuiarare, there is a woman who came from Rio dos Peixes

and married a Kaiabi man) and getting married to Xingu men or women. It is also worth

mentioning that these few cases in Xingu all fall in the "moderate" level of proficiency, which

means that they can speak and understand the language fairly well. There are no cases in the low

and/or no levels of proficiency.

In fact, there is a huge difference between language proficiency inside and outside Xingu,

shown in Table 7.5 and in Figure 7-17. According to Hill (2001), we can consider that in Rio dos

Peixes, the Kaiabi language is moribund, because the children do not speak it anymore. During

the time I spent in Rio dos Peixes, I was able to observe that the children communicate among

themselves and with their parents in Portuguese. The same trend is happening in Kururuzinho,

where the youth and the children tend to communicate in Portuguese, whereas among the adults

some use the language and some don't. The fact that there is more interethnic mixing at Teles

Pires also contributes to the widespread use of Portuguese.

Correlation and logistic regression statistical tests were used to study the relationship

between language proficiency and other variables (see next section) outside Xingu, taking

language proficiency as the dependent variable. Language proficiency in Rio dos Peixes and

Kururuzinho villages is positively correlated with age (r=0.398, p<0.001), meaning that

proficiency in the native language increases with age, and also that there are more elders than









youth who are proficient, especially outside Xingu, where the intergenerational transmission of

language seems to be compromised or interrupted.

The estimated odds of language proficiency for people over 50 years old outside Xingu

are very high, compared to youth between 15-25 years old: people over 50 years of age have a

91.9% greater chance of being proficient than those in the age class of 15-25 years. Similar

results related to age and knowledge were registered by Phillips and Gentry (1993) for Peruvian

peasants and by Zent (1999; 2001) for a group of Piaroa indigenous people in Venezuela.

Whereas these studies specifically targeted ethnobotanical knowledge, both authors found that

there is a positive correlation between age and knowledge. My point here is that it is reasonable

to take language proficiency as an indicator of knowledge, or as a proxy for knowledge

(whatever domain it might be), which will show comparable statistical behaviour as other

knowledge variables (such as ethnobotanical, crop knowledge or artistic-weaving). This

argument and these studies are going to be revisited in the next chapter, in the analysis of

weaving knowledge in relation to other variables, including age.

Xingu Park is a place of multilingualism, where fourteen different indigenous languages

are spoken.7 The different ethnic groups who live in Xingu use the Portuguese language to

communicate with each other and also with the non-indigenous. Also, there is an increased need

for learning and becoming fluent in Portuguese, spoken and written, coming from the historical

insertion and participation of Xingu in the indigenous national political agenda. Whereas this

might lead to more and more Portuguese speakers, in the villages the official languages used are





7 The only language affiliated with the Kaiabi language in Xingu Park is spoken by the Kamayurd people from the
upper Xingu region, which is also classified in the Tupi stock, Tupi-Guarani family. However, due to differences in
meaning and pronunciation, both groups use the Portuguese language to speak to each other.









still the indigenous ones. The isolation, the low interethnic mixing and the ethnic pride are all

factors that have contributed to keep the native languages alive.

Comparisons and Relationships between Socioeconomic Variables

Through the various statistical analyses carried out for better understanding the

relationships between selected socioeconomic variables, it was possible to test some of the

hypotheses designed for this research, as well as to identify variables which are more influential

in explaining cultural change among the Kaiabi. The differences between the Xingu group and

the other two Rio dos Peixes and Teles Pires are significant for the three main variables

analysed here: stability of income, schooling and language proficiency. Figure 7-17 shows a

comparison between various socioeconomic variables for the four villages included in the

research.

First of all, gender was a variable that did not show statistical significance when

correlated to other socioeconomic variables considered here. The only statistical correlation we

found for gender, explained in Chapter 6, was that women participate less in political meetings

when compared to men. Therefore, in spite of the apparent differences in, for instance, stability

of income and schooling for women and men living in and out of Xingu Park, these are not

statistically significant.

Stability of income, a variable used as a proxy to measure linkage to market economy and

also level of access to waged income, showed statistical interaction with age class, location and

language proficiency. As expected, the older the person, the less stable their income, and also

people living outside Xingu have more stable incomes than people living in Xingu, in which

there are fewer waged officers. This is related to greater employment opportunities in the health

and education sectors in Rio dos Peixes, and also to access to government programs such as









family and retirement pensions, besides the closest proximity to markets. In the logistic

regression run for comparing language proficiency (dependent variable) outside Xingu with the

other variables, stability of income was the second most important independent variable after

age, showing a negative relationship with proficiency (p<0.01, see table 7-6). The odds ratio

transformed into percentage indicates that a person with a more stable income has a 71% less

chance to be proficient in the Kaiabi language, in comparison with people with no income

stability. This indicates that even when taking Xingu villages out, people who have greater

access to wages and thus closer contact with market economy, tend to present lower levels of

language proficiency. If I test H3 considering that language proficiency as a proxy for

indigenous knowledge, I accept this hypothesis here, since greater income stability leads to less

language proficiency.

Several factors interact to produce these results, which can be roughly exemplified taking

Rio dos Peixes as a model of cultural erosion compared to Xingu as a model of cultural

resilience. Integration to market economy and thus increased access to wages and to income

stability is a process that might come along with historical determinants, greater proximity to

towns, interethnic mixing, schooling, adoption of non-indigenous lifestyles, lack of strong

leadership, among other factors. This multi-variable interaction makes it challenging to

objectively understand the link between socioeconomic variables and a person's traditional

knowledge, as Reyes-Garcia and her collaborators (2005) have also found for the Tsimane'

people in the Bolivian Amazon. As it is not really possible to isolate the combined effects of

socioeconomic variables, it is probably more useful to treat them as clusters leading to more or

less cultural resilience.









Age is the variable that showed more interactions and significant statistical correlations in

comparison with all the others. Age was negatively correlated to stability of income and

schooling, and positively correlated with proficiency in Kaiabi language. Results of the logistic

regression taking language proficiency as the dependent variable against gender, schooling,

stability of income and age, show that age class is the variable that most affects language

proficiency, followed by stability of income (Table 7-6). Schooling and gender were not

important in this model. Age and income stability were the variables that showed more influence

in language proficiency in the logistic regression model considering all the variables.

As expected, the older the person, the less formal schooling and the less stable the

income, the greater their native language proficiency. This means that there is a significant

difference between the elders and the younger generations in terms of socioeconomic profile, and

indicates that the gap between these generations might be a strong factor affecting cultural

change and loss of cultural resilience. This is especially important taking into account that people

over 50 years old are the least represented in terms of population, being at the top of the age

pyramid. Additionally, when comparing the generation gap among the Kaiabi, we have to take

into account the displacement event to which they were subjected, especially in the ancestral

areas, where fewer older people remained and are still alive.

Conclusion

The villages considered in this study present some interesting commonalities and

differences, in terms of social organisation and socioeconomic profile. In Xingu, Capivara and

Tuiarare show similarities regarding the socioeconomic variables analysed here, such as

schooling, stability of income and language proficiency. However, they also present differences

based on kinship ties, leadership, gender participation and access to resources. Within each of the

four villages, there is a great heterogeneity in terms of access to wages, schooling and political









opportunities. Therefore, when conducting work or research with any "community," we had

better be aware of the heterogeneity we are dealing with.

Gender as a variable did not show statistical significance in the bi and multivariate

analyses carried out for this chapter. However, through the interviews and results, it is possible to

visualize differences between men and women, and among women, for the different variables

and villages considered. Women in Xingu do not work as waged officers and also don't have

access to income such as bolsa-familia and retirement pensions, as happens in Rio dos Peixes

and in Kururuzinho. Also, women travel less than men, and in the case of Xingu, do not have

formal schooling, except for a few cases in Capivara. Women in Rio dos Peixes show less

proficiency in the native language when compared to men, and also higher attendance to formal

schooling.

The socioeconomic drivers of cultural change and loss of resilience interact at different

scales and intensities among the villages. Age is the most important and influential variable. The

elders are the ones who present more fluency in the language, but in contrast have no income

stability and generally no schooling. The generation gap seems to be greater in Rio dos Peixes

and Kururuzinho in comparison to Xingu. In this model, stability of income was best variable to

predict cultural erosion, measured by proficiency in the native language. Schooling was not

statistically significant in the logistical regression model developed for this data, but looking at

Figure 7-17 we can see that greater stability of income, schooling and interethnic mixing,

associated with lower proficiency in native language. From the villages' socioeconomic profiles

and analyses it's possible to confirm that the greater the socioeconomic heterogeneity within a

social group, the more integrated to the market economy (Reyes-Garcia et al, 2005). This is

especially true when comparing the two extremes: Rio dos Peixes and Xingu. In Rio dos Peixes,









the mix of economic activities and varied ethnic and cultural background of residents indicate a

transition from the Kaiabi traditional lifestyle, to the adoption of that of non-indigenous peoples

("caboclos") inhabiting forest-river ecosystems in the Amazon (Adams et al., 2008). Among

some features that indicate a process of cultural shift or loss of traditional cultural resilience for

the Rio dos Peixes group, we could mention the nuclearization of families; the loss of traditional

language; the substitution of garden plots by homegardens; the loss of agrobiodiversity; the

greater integration to market economy; and the selling of products or services to relatives.

Taking the case of Rio dos Peixes as an extreme does not mean that in Teles Pires or

Xingu the process of cultural shift is not occurring. It is just occurring under other circumstances

and at a different pace. It is inevitable that indigenous peoples get integrated in the market

economy, as (almost) the whole world needs cash to buy products needed for day-by-day living.

Nevertheless, the institutional, political and geographic filters, as well as the cultural pride

present in Xingu, prevent integration, impeding deeper shifts and thus loss of resilience. In Teles

Pires, the situation is intermediate between Rio dos Peixes and Xingu. Kinship ties are stronger

because of village size and family composition, in spite of some interethnic mixing. Also, Xingu

and Teles Pires ties have been greatly strengthened with the participation of Xingu relatives in

the process of claiming land in Teles Pires, as well as marriages between Ilha Grande and

Kururuzinho villagers, which does not happen at the same intensity among Rio dos Peixes and

Capivara. The transition or "caboclization" process that is happening in Rio dos Peixes should

not be regarded as linear and predictable, though. Changes in political and institutional

configurations, as well as contacts and exchanges within a cultural group can affect and in some

cases change the course of history. The Xingu Kaiabi are living proof of this.









In spite of these results being interesting, the fact that in Xingu villages nearly all the

residents are proficient in the language makes our comparisons limited. Also, the way language

proficiency was measured and analysed, in two categories, is very broad and vague. To better

estimate language proficiency, we would have to develop an instrument to test and apply among

the various participants. Despite our data and analyses limitations, this socioeconomic analysis is

valuable because, first of all, it provides a socioeconomic panorama of all villages, establishing

ground for further research and comparisons. Secondly, it gives us information on the factors that

underlie processes of culture shift and associated mechanisms. Weaving knowledge, the topic of

the next chapter and the cultural domain chosen for deepening our analyses and comparisons

among the three Kaiabi groups, will cover some of the gaps and clarify doubts related to the

choice of language proficiency as a proxy for cultural vitality and resilience.























































---'I -Ir't- M".4

Figure 7-1. Location of Capivara and Tuiarare villages on the northern portion of Xingu Park.
Source: Instituto Socioambiental, 2008.
















-. .P :F


"* ,
, .. S -.M t .*:. '' *.. "


I na -""~~


,, _' ,.11

*.... a






,fcju rrr IrOD
j-P STC O ) .... :
.fiitrer SruPb itrbdb e 1



OnfV 5 Ulf -f r-
T i* rL
rI .Y


Ci -;


C-~


'rVN


: -





7 .a lla .l










Figure 7-2. Map of houses of Capivara village in 2002, around the soccer field.
















4449 -Alrt u t


a -Z-:
rr
;~- ;. C
.. ,<^ 'Pi 6
*^ a-^ -&

SBQ a a\


C;i
gjh C4
~L Tw


Figure 7-3. Map of Capivara village in 2007, showing the old village (front) with the school
(yellow building) and the new village (back). Drawing by Sirakup Kaiabi.


- c!j


"N


CJ









o r* r

^B~~~ Amn ip f^


6 .1


1,4

'.1

i .
Sj.


1 IITI L


:[ ..
r"
: 1


!I
"(-^



i *! \l '


*:.


Figure 7-4. Map of Tuiarare village in 2002 showing houses, health unit and school. Drawing by
Tamakari Kaiabi.














3 > -4A


I


'K~rs


P~t -:


-J


I _


Figure 7-5. Map of Tuiarare village in 2007. Drawing by Apurina and Piraju.































294


C-


/ -
I
"i


UU ^3


\_.,'


r

9ni~


I;~b
--












i ,, !










^- 6I







(Ni









Figure 7-6. Map of the planned new Tuiarare village (2007) located on the back area of the old
village, with the "house of culture" or school of culture in the center, surrounded by
residents' houses. Drawing by Apurina and Piraju.





















295























-
I
'Y .-


I;


Figure 7-7. Map of Rio dos Peixes Post (Tatuy) showing the river, the road which leads to Juara

and the path to Novo Horizonte village, located adjacent to the post. Drawing by

Simone Athayde, 2007.








































296


I' .*;


I A


c-- ----

i




























- I-


A

i. )J


Figure 7-8. Map of the houses in Novo Horizonte village at Rio dos Peixes. Drawing by Simone

Athayde, 2007.

















































297


;I
.i
r r 1,1
ri


.. r
'i' i' i, "F i.:


'ii 'it
; ,"-


- ,. ,










SJ *te Pt


re

f mIdRfl

12 jo&~t


set-..


%,re iw k 13 nt&
ftb t A -brI-*C$ $ It4
^__ rfl. ire


Figure 7-9. Map of Kururuzinho Post and village at Kayabi Indigenous Land in Teles Pires river.
The airplane landing strip is located at the back of the village. Drawing by Tangeu'i
Kaiabi, 2007.


t>iH e/*
^w^A ^f.






thti* ^*/
t*tT /f
BE3^[ S^





T JU~enldc

'I f*TrfiiM-s/e


I8


3 t ln tf
C MY"r

9,4 *
S j -


W









Table 7-1. Number and gender of people interviewed in four Kaiabi villages.
People interviewed Capivara Tuiarare Rio dos Teles Total
Peixes Pires
Number of residents* 144 163 266 155 728


Number of households 13 14 37 14 78


Number of people 49 45 76 54 224
interviewed

Number of men 25 23 36 26 110
interviewed

Number of women 24 22 40 28 114
interviewed

% of interviewees in the 21.90 20.10 33.90 24.10 100
research

% of interviewees in 34.03 27.61 28.57 34.84 30.76
relation to the total of
residents
Total of interviewees was 224. Sampling included men and women over 15 years old.
*Actual number of residents based on data collected by ISA's personnel in Xingu, FUNASA in
Rio dos Peixes and FUNAI in Teles Pires for the year of 2007.
































CA TU



100 83 36 > 115


5 56-6 83 Male 107 115
FemaleFemale Male

50 46- 6 36
46-5




553333 6


375 278



RP TP


Figure 7-10. Age pyramids for people interviewed in each age class by gender in four Kaiabi
villages. Age classes based on Athayde (2003).





















197o


IU '
Qin4






0 ii 50


544541




1024 &

-1.4 5
-15 xl' -


EHamens
m02-de
E] M- h.


b 5 In is


lI ou..em
n nMulaee"


a S i 15i


0 a 54


IU I I





4 -M A


ha4



0.4

.15 -10 -5


4 5 10 13L.r


5 lllal
O unnuma


Figure 7-11. Age pyramids for the Xingu Kaiabi population, from

Pagliaro (2005).


1970 to 1999. Source:





























Capivara


Tuiarare


Rio dos Peixes


B Other
[I Teles Pires
E Rio dos Peixes
* Xingu Park


Teles Pires


Villages of residents

Figure 7-12. Places of birth for people interviewed in four Kaiabi villages.


100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%


3~VI











Table 7-2. Functions and occupations in the villages. CA= Capivara; TU=Tuiarar6; RP= Rio dos Peixes; TP = Teles Pires.


o 0 -
n a,, ^" a, "
~~~~~~~~a ^ S % % a .
C) S .-S ^ ) ^^ ^ ^8 io S S
00 5 -ctio~Q .
t l cl .2 au 5 'j
> UZ 3 U o > c- m E2 0 -a -
CA F n 0 0 0 0 0 2 0 0 0 22 0 0 0 24


% 0
M n 4
% 16
T n 4
% 8.16
TU F n 0
% 0
M N 1
% 4.35
T n 1
% 2.22
RP F n 2
% 5
M n 1
% 2.78
T n 3
% 3.95
TP F n 0
% 0
M n 3
% 11.54
T n 3
% 5.56


0
1
4
1
2.04
0
0
3
13.04
3
6.67
2
5
1
2.78
3
3.95
0
0
2
7.69
2
3.7


0
1
4
1
2.04
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0


8.33
0
0
2
4.08
0
0
1
4.35
1
2.22
0
0
2
5.56
2
2.63
0
0
3
11.54
3
5.56


0
1
4
1
2.04
0
0
1
4.35
1
2.22
0
0
1
2.78
1
1.32
0
0
1
3.85
1
1.85


0
0
0
0
0
0
0
1
4.35
1
2.22
0
0
0
0
0
0
0
0
0
0
0
0


0
0
0
0
0
0
0
1
4.35
1
2.22
0
0
1
2.78
1
1.32
0
0
0
0
0
0


91.67
14
56
36
73.47
22
100
12
52.17
34
75.56
36
90
28
77.78
64
84.21
0
0
16
61.54
16
29.63


0
1
4
1
2.04
0
0
1
4.35
1
2.22
0
0
0
0
0
0
28
100
0
0
28
51.85


0
1
4
1
2.04
0
0
1
4.35
1
2.22
0
0
0
0
0
0
0
0
0
0
0
0


0
0
0
0
0
0
0
0
0
0
0
0
0
2
5.56
2
2.63
0
0
1
3.85
1
1.85











4500 -3939.82
4000
S3500
3000
2500 -
2000 1725.16 1814.65
a 1500 1140.25
0 1000 -
500 -
0
0 -- -------------------- -
CA TU RP TP
Village


Figure 7-13. Total monthly income by village, in dollars, from waged officers, retirement
pensions and family pensions. CA=Capivara; TU=Tuiarare; RP=Rio dos Peixes; TP:
Teles Pires.










Brazil nut
Honey production
Family support
Temporary work
Retirement pension
Family pension
Agriculture
Wages
Handicrafts


0.00


20.C


)0 40.00 6C

% of people/village


Figure 7-14. Sources of income for people interviewed in four Kaiabi villages. CA=Capivara;
TU=Tuiarare; RP=Rio dos Peixes; TP=Teles Pires.


0 TP
* RP
E TU
0 CA


).00


80.00


17











Table 7-3. Levels of stability of income for men and women in four Kaiabi villages.
Waged officer Retirement pension Family pension Irregular
Village
F % F % F % F %
Capivara Male 5 20.00 1 4.00 0 0.00 19 76.00
Female 0 0.00 0 0.00 0 0.00 24 100.00
Total 5 10.20 1 2.04 0 0.00 43 87.76


Tuiarard Male 3 13.04 0 0.00 0 0.00 20 86.96
Female 0 0.00 0 0.00 0 0.00 22 100.00
Total 3 6.67 0 0.00 0 0.00 42 93.33


Rio dosPeixes Male 4 11.11 3 8.33 0 0.00 29 80.56
Female 4 10.00 6 15.00 14 35.00 16 40.00
Total 8 10.53 9 11.84 14 18.42 45 59.21


Teles Pires Male 3 11.54 4 15.38 0 0 19 73.08
Female 0 0.00 3 10.71 0 0.00 25 89.29
Total 3 5.56 7 12.96 0 0.00 44 81.48














M

F


F

M


* Xingu-RP
O Xingu-TP
E RP-TP


20.00


40.00


60.00


80.00


100.00


120.00


% of people


Figure 7-15. Percentage of women and men interviewed in four villages, travelling between the
three Kaiabi areas. CA=Capivara; TU=Tuiarare; RP=Rio dos Peixes; TP=Teles Pires.
F=female; M=male.


ME=

IONE



















E 1. Visit relatives
E 2. Fetch natural resources
o 3. Political
0 4. Work
E 5. Marriage
E 6. Participation in training courses
E 7. Participation in projects
O 8. Other


100%


0% 20%


Reasons for travels (%)


Figure 7-16. Reasons for travels between the three Kaiabi areas among women and men
interviewed in four villages. CA=Capivara; TU=Tuiarare; RP=Rio dos Peixes;
TP=Teles Pires. F=female; M=male.


M IUB


F


M

F

M

F

M

F


u
0
o<
03
> F





U


n









Table 7-4. Level of formal schooling for people interviewed in four Kaiabi villages.
Village Gender No instruction Elementary Secondary University
Freq % Freq % Freq % Freq %


Capivara




Tuiarare




Rio dos Peixes


Teles Pires


Male
Female
Total

Male
Female
Total

Male
Female
Total

Male
Female
Total


12 4
14 5
26 5


8.00
8.33
3.06


43.48
95.45
68.89

19.44
22.50
21.05

23.08
25.00
24.07


11
10
21


0
0
0


2
0
2


44.00
41.67
42.86

39.13
4.545
22.22

75.00
72.50
73.68

61.54
75.00
68.52


8.00
0.00
4.08

4.35
0.00
2.22

0
0
0


13.04
0
6.67

5.56
5.00
5.26

15.38
0
7.41









Table 7-5. Proficiency in native (Kaiabi) language for people interviewed in four Kaiabi villages.
Village Gender None (0) Low (1) Moderate (2) Fluent (3)
Freq % Freq % Freq % Freq %
Capivara Male 0 0 0 0 1 4.00 24 96.00
Female 0 0 0 0 2 8.33 22 91.67
Total 0 0 0 0 3 12.33 46 93.88


Tuiarare




Rio dos Peixes


Teles Pires


Male
Female
Total

Male
Female
Total

Male
Female
Total


2.78
17.50
10.53

3.85
3.57
3.70


50.00
40.00
44.74

38.46
32.14
35.19


0
4.55
2.22

8.33
12.50
10.53

7.69
25.00
16.67


100.00
95.45
97.78

38.89
30.00
34.21

50.00
39.29
44.44






















* Participation in political meetings
O Interethnic mixing
* Formal Schooling (elementary)
* Language proficiency i '!. in'!i
* Stability of income (stable)


20.00 40.00 60.00 80.00
% of people


100.00


Figure 7-17. A comparison of selected socioeconomic variables between four Kaiabi villages.
CA- Capivara; TU- Tuiarare; RP- Rio dos Peixes; TP- Teles Pires.


~1


RP

M
O


I


EMEM









Table 7-6. Results of logistical regression analysis for selected socio-economic variables, having
language proficiency as the dependent variable.
Variables/parameters B S.E. Wald df Sig. Exp(B)
Step 1 Gender (1) -.309 .409 .570 1 .450 .734

Schooling(l) -.257 .748 .118 1 .731 .773

Stability of income(l) -1.211 .515 5.527 1 .019 .298

Age class (over 50 years)(1) -2.513 1.045 5.784 1 .016 .081

Age class (25-50) (2) -3.647 1.124 10.522 1 .001 .026

Constant 3.420 .919 13.853 1 .001 30.561









CHAPTER 8
THE DYNAMICS OF WEAVING KNOWLEDGE ACROSS FOUR KAIABI VILLAGES

Introduction

This chapter contains complementary analyses of questions and findings presented

throughout this document, on the multiple and systemic interactions between indigenous

knowledge, cultural resilience and socioeconomic, political and territorial processes among

indigenous peoples. As I postulate in the general introduction, cultural resilience reflects the

capacity of adaptation and persistence of a given cultural group in response to changes (Holling

and Gunderson, 2002; Gunderson, 2003, my interpretation). One can explain mechanisms

involved in cultural resilience qualitatively and quantitatively, and here I intend to do both, in a

complementary fashion. To do this, I study a cultural domain which, integrated with other

domains and topics, provides a perspective on the mechanisms involved in the perpetuation and

change of indigenous knowledge systems across three Kaiabi groups. Weaving knowledge on

basketry and textiles is analyzed and compared at various spatial and temporal scales, among

genders, individuals, households, kin groups, villages and finally between the three groups

included in this study. Among the Kaiabi, weaving is an ideal knowledge domain to investigate

mechanisms of knowledge transmission and change, given the central importance that basketry

and textiles have in their history and in their social, symbolic and economic life.

Taking into account the previous background of this research and some of the questions

presented in the general introduction, I will be working under two main scientific inquisitive and

theoretical fields in this chapter: the first, related to processual studies in IK, focusing on which

mechanisms are involved in knowledge distribution, transmission and change (Ross, 2002; 2004;

Zent, 2009a); and the second, investigating the effects of the growing participation in and

appropriation of western institutions (including schools, markets, indigenous organizations,









development projects) by indigenous peoples, on their knowledge systems (Turner, 1991; Posey,

1994; Jackson, 1994; 1995; Gray, 1997; Oakdale, 1996; 2004; Maffi, 2001; Zent, 2009b). As

Ross (2002) states, the study of intracultural differences within a given ethnic group can trigger

insights on how processes of knowledge formation and transmission might be changing in new

contexts such as globalization. In this chapter, I will be testing the five hypotheses postulated in

the general introduction, as follows:

* HI: Indigenous knowledge and age: elders retain deeper and different knowledge when
compared to younger people;

* H2: Indigenous knowledge and gender: men and women use different mechanisms in
knowledge creation and transmission;

* H3: Indigenous knowledge and markets: greater levels of market integration (proxied by
income stability) lead to erosion of weaving knowledge across three Kaiabi groups.

* H4: Indigenous knowledge and formal schooling: formal schooling erodes indigenous
knowledge: higher levels of formal schooling leads to lower levels of weaving knowledge
among men and women.

* H5: Role of community-based projects: the Kaiabi Araa project is responsible for an
increased number of basketry and textiles weavers among the younger generations. In
villages which are participating in the project, there are more youth who know how to
weave now in comparison to 2002 and to other villages which are not participating in the
project.

A processual perspective (Zent, 2009 a, see general introduction) on Kaiabi weaving

knowledge is presented through a five-year longitudinal analysis on the distribution of

knowledge among male residents of Capivara and Tuiarare villages in Xingu. This research

builds on results found in previous studies (Athayde, 2003; Athayde et al., 2009) in which I

detected trends and changes in knowledge transmission, distribution and on the social meaning

of weaving among the Kaiabi. In each section of this chapter, I intend to review and deepen

those findings, as well as expand these analyses to weaving knowledge held by women, and

among distinct Kaiabi groups (Xingu, Rio dos Peixes and Teles Pires). I compare knowledge of









names for design motifs among more knowledgeable individuals, using consensus statistical

analysis. I review models of cultural transmission presented by other authors (Cavalli-Sforza and

Feldman, 1981; Hewlett and Cavalli-Sforza, 1986; Ohmagari and Berkes, 1997). I also looked at

mechanisms which allow or constrain the retention of weaving knowledge.

To understand and organize the domain of weaving knowledge, I adapted the

classification proposed by Ellen (2009). He suggested that to study knowledge transmission

within a cognitive domain, a scholar needs to take into account other domains that might affect

knowledge in the domain you are considering. I agree with that author in that the classification of

culture in domains is a western science artifact that many times does not correspond to local

ways of organizing knowledge. Ellen (2009:245) proposes for the study of Nuaulu basket-

weaving knowledge, the integration of overlapping domains or "modules" of knowledge, which I

adopted and modified for this research in basketry and textiles knowledge, as follows (see Figure

8-1):

1. Form knowledge of different types of baskets (or textiles) items ability to weave the
different types and domain of techniques, including preparation of materials;

2. Function knowledge of different basketry and textiles uses;

3. Materials ethnoecological knowledge related to identification, management, extraction
and preparation of plant resources used in weaving;

4. Meanings knowledge of symbolic elements related to basketry and textiles weaving,
such as mythic and spiritual meanings;

5. Motifs knowledge of graphic designs (patterns), ability to weave and name the designs.
Each of these sub-domains or modules will be explored throughout the sections of this
chapter.

Weaving Culture into Material Objects

Studies on the systems of objects produced by human societies, generally named

material culture, may approach them from different theoretical perspectives and can take diverse









exploratory directions. Material culture studies among indigenous societies may involve, for

instance, archeological research on remnants of objects at archeological sites; anthropological

research on various aspects related to form (techniques of production and decoration); function

(uses, meanings); social organization of production, re-production, appropriation and

commercialization of these objects; and ecological research on natural resources used in the

production of objects (Ribeiro, 1987 a).

According to Chernela (2008), material culture objects may work as signifiers of history,

or as agents in the construction of history, having consequences in the social and political life of

indigenous peoples. The objects bring with them a web of meanings, reflecting aspects of the

ecology, economy and the lifestyle of indigenous peoples (Ribeiro, 1987 a). They can work as

stimuli upon which human societies reflect and restate their culture, through material and

symbolic representations (Ribeiro, 1987 a; Ross, 2004). Van Velthem (2004) states that the

multiplicity of representations, techniques, and meanings encoded in indigenous people's artistic

material manifestations should lead us to talk about indigenous "arts" in the plural, commonly

referred to as indigenous art in the singular. In her study about the system of material culture of

the Wayana people from the Brazilian Amazon, she considers that the elements form, function

and decoration are intertwined with present and primeval times and with the individual and his

social group, having real and symbolic transformative capacities. The object has the power to

transform animals and plants into food (e.g. baskets, graters, and squeezers), youth into adults

(e.g. earrings, tattoos, and piercings), and ordinary men in supernatural beings (e.g. masks,

adornments and other ritual objects).

Being directly connected to the social dimensions, indigenous arts bring up messages

about the social position of their producers and users (Vidal, 1992; Van Velthem, 2004).









Mastership in basketry and textile weaving among the Kaiabi is valued by the social group, who

confer special status to expert weavers. It's interesting to note that many times a basket or a

textile object are made not to be used by the producer, but to be integrated in the social and

kinship systems in which that person lives. Just as with the meanings, the use of objects is

determined by the social and economic systems which characterize each indigenous society. The

object might be kept by the producer, given away, exchanged or, more recently,

commercialised.

According to Ribeiro (1983), the increased contact of indigenous peoples with western

society and the market economy have caused profound transformations in their material culture

systems: objects that traditionally constituted production for the inside, including exchanges

with other groups, came to be produced for the outside, for exchange with money. At this point,

we should make a distinction between material culture, indigenous art and handicraft. Any

indigenous object which integrates the larger material culture system might be considered a

work of art for its aesthetic and technical qualities (Van Velthem, 2004). Ribeiro (1983)

distinguishes handicraft as the objects from the material culture system that are intentionally

produced for the market.

If you look at a Kaiabi basket or hammock with a naive eye, you might not perceive the

historical, ethnical, mythical, technical and ecological meanings and values attached to that

apparently "simple" object. Guss (1989:162), describing basketry weaving among the Yekuana

people in Venezuelan Amazon talks about a "mutual reflexivity", where meaning is continually

being created in the basket's designs, from a shared context of forms. Each aspect of a basket or

textile object such as design, material, technique and function reviews and incorporates the

messages of the other members of a given ethnic group.









When a Yekuana, a Wayana or a Kaiabi weaves or uses a basket, the range of meanings

evoked go far beyond the choice of design, the preparation of the materials or its routine and/or

ritual use (Guss, 1989). They are reproducing their history, ethnicity and social organization in

the process, such as in a mirror. Likewise, changes that might happen in a society in

consequence of external agents and institutions (e.g. markets, missionaries, schools, cultural

projects) are also reflected in the production of objects in different ways: prioritizing production

of some objects over others; causing revitalization of objects threatened with disappearing;

valuing certain graphic designs and adornment styles; and, very importantly, enabling

innovation in the creation of new designs, new objects or new ways to express a traditional

technique and/or design, for instance. These aspects are going to be discussed and explored for

the Kaiabi case in the sub-sections that follow.

Objects that compose Kaiabi material culture have been accompanying the historical

trajectory of the group, especially after the transfer to Xingu. In this process, gender roles

switched in the production of objects; some objects disappeared; new ones were incorporated

into the system; and others have changed as a result of new techniques learned, interethnic

contact or market integration (Athayde, 1998). This is in fact a very dynamic process, influenced

by internal and external forces, which nobody controls or can predict the direction of. One thing

we know for sure is that there will be knowledge erosion, change and innovation or creation. If

we compare the list of items produced by the Kaiabi in 1966, documented by the anthropologist

Georg Grunberg1 (Grunberg and Grunberg, 1967; Grunberg, 2004), with recent surveys of

Kaiabi material culture (Athayde, 1998), the panorama is quite different.




1 In 1966, while working at Rio dos Peixes, Griinberg organized a fairly complete collection of Kaiabi objects,
which is deposited at the Museum of Cultures in Basel, Switzerland.









Kaiabi material culture is composed of objects used in body ornamentation (elaborated

necklaces made of palm fruit, bracelets, rings, earrings, belts and feathered headdresses); hunting

and fishing artefacts (bows and arrows, clubs, fishing traps); and objects produced for domestic

use (wooden benches, gourds and calabashes, baskets, textiles). Production of jewellery items

has gone through changes due to interethnic contact, market integration and of course, in

consequence of the displacement to Xingu Park. Necklaces, bracelets and rings originally made

by men, came to be produced almost solely by women. The sale and exchange of these objects is

an important source of family income and access to industrialized products by the women. The

production of bows, arrows and clubs has been limited by the lack or scarcity of natural

resources traditionally used in the confection of these items (see Chapter 5). Therefore, substitute

resources have been used, as well as traditional materials gathered in trips from Xingu to the

ancestral areas in Rio dos Peixes and/or Teles Pires. Pottery production can be considered extinct

among the Kaiabi, in spite of recent efforts to try to revitalize it. When Grunberg visited the

Kaiabi in 1966, he already observed the rarity of Kaiabi pottery pans, being substituted by

industrialized pans.

In Xingu Park, the Kaiabi have developed innovations in textile production and in

painting of wooden-carved benches, mainly through the contact with the Yudja people's

techniques (description of innovations on textile production at Xingu Park follows in the next

sections). Most of the items that have gone through innovation processes are destined for market

commercialization. In the next sections, I will be focusing on basketry and textile items, as an

example of the dynamics of artefact production reflecting social-economic changes among the

three Kaiabi groups.









Types and Uses of Basketry and Textile Items

According to Ellen (2009), technical definitions of basketry might vary depending on the

field of a specialist. Here I adopt his simplified definition of basketry as "a container created by

weaving semi-rigid vegetable fibres" (Ellen, 2009:248). For Wendrich (1991), the category of

basketry comprises: baskets, bags and mats; brushes and brooms; hurdles; wattle-and-daub

constructions, sandals, hats and belts. Basketry is the most elaborate group of items of Kaiabi

material culture (Grunberg and Grunberg 1967; Grunberg 2004). Indigenous systems of

classification of material culture items often differ from those of the western science, in terms of

covert categories and its ramifications. For instance, Ellen (2009) found that among the Nuaulu

people from Indonesia there is a category that covers a group of objects we would designate as

baskets. For the Kaiabi, the criteria used to classify objects refer mainly to the type of use of the

object, and also to the material used in its confection. Sometimes, they might even name a plant

by the name of the object produced from it. This is the case, from example, of arrow-cane

species used in arrow production that may be called just "arrow". A Kaiabi covert category that

encompasses many types of baskets is defined by the presence of the prefix "yri", which means

container. Therefore, yripem is used for basketry strainers, while yrfifuku (yri=container;

fuku=long) is a long basket and yrCipejuap (yri=container, basket; juap=grass, young leaf) is a

basketry container made of tucumd (Astrocaryum aculeatum) palm leaf. In Figure 8-2, I present a

diagram grouping Kaiabi baskets by covert category, by use and gender of use.

During workshops of the Kaiabi Araa project, 21 different items that compose the rich

Kaiabi basketry repertoire were identified (Table 8-1). Some of these objects are used solely by

men (5 or 24%), others are mostly used by women (6 or 29%), but the majority may be used by

both (10 or 48%). Some are also used by the shamans and in rituals. The most basic type of

basket, and the easiest to learn, is the myayta, made of vine fiber, used by everyone to transport









food from the garden plot to the village. Every married man should be able to make a myayta.

Another basket of relatively easy manufacture is the simple fan (tapekwap) made of young

tucumd palm leaf. It is used mainly to cover food and to make fire. There is a more complex fan

(named tapekwajowai) with two handles, which only a few men know how to make. Men also

produce beautiful woven handles for wooden clubs, which are nowadays used by them in

political events and during festivals. In all villages studied, men produce clubs for sale. They

even make little ones which can be used as keychains. In Rio dos Peixes, I saw a lot of clubs in

the houses, but the majority had no woven ornament, because few men there still know how to

weave (see next sections). There is a diversity of baskets made of tucuma and inaja palm

(Maximilliana maripa) leaves and used as containers. One of them, named "yri7okote 'em", was

done in a project workshop after many years of relative disappearance by Karauu, the only elder

who still remembers how to weave it.

The different types of basketry strainers, used mainly by women to prepare food, reflect

the sophistication of Kaiabi agriculture and culinary. There is a small sieve strainer for fine grain

flour (yriipemeaii), as well as a large sieve strainer (yriipemeauu). There is also an interesting

type of basket, made of inaja palm leaf petiole, used exclusively to store peanuts (juyp munuwi

)yi i'). Its name reflects both the material used in its confection (juyp=grass or fiber) and its use

(munuwi=peanut). As I mentioned, the Kaiabi cherish an astonishing diversity of peanuts, maybe

one of the greatest in the world (Silva, 2009). Fans and strainers can also be used by the shamans

to preach (see following section on shamanism). The twill-plaited painted baskets in which the

men weave different graphic designs, might be generally named "yi /ily''/", or might be called

araa, which means "design". Other items that compose the basketry repertoire are woven

supports for headfeathers, and woven adornments for arrows and fish traps, both used by men.









The different types of baskets are made according to different weaving techniques, which

are also classified differently by different authors. For this dissertation, I adopt the classification

proposed by Adovasio (1977) and adapted by Ribeiro (1985). Adovasio (1977) divides basketry

techniques into three main groups: twined, coiled and twilled. According to Ribeiro (1985), in

Brazilian indigenous basketry the most important technique is willing, which is subsequently

classified into four groups by the author: checker; wicker; twill-plaiting; and lattice work or open

work. Subtypes of the willing technique are represented in Kaiabi basketry, such as checker

work (fans, mats); wicker (panakiiawet, jesi 'a); lattice or open work (l,,'y' t'i) and twill-plaiting

(strainers, simple and painted, such as the twill-plaited araa baskets).

Textile weaving is primarily done by women, with men playing a role in manufacturing

some accessories used by them in weaving, such as the spindle (e ym), produced from palm

wood and turtle shell; the textile comb (taitypypykap), made from wood; and the wooden loom

(taity retykap). In contrast to basketry, the Kaiabi textiles group comprises only six items, three

of them developed after the transfer of the group to Xingu (Table 8-2): strap for carrying babies;

three types of hammocks; woven bags and belts. Bags and belts are innovations developed in

Xingu for use in festivals, trade and sale.

Before the transfer to Xingu Park, Kaiabi women used to weave hammocks (taity) and

straps for carrying babies (tupai) using a 'twined' technique and a ground loom commonly

employed by other Tupi groups, and which included no designs (Ribeiro 1984/85). Through

contact and inter-marriage with Yudja people in Xingu, Kaiabi women learned the 'twilled'

technique, which allows weaving of complex geometric designs ibidd). They also began to use

the frame loom and the comb to weave, almost abandoning the simpler ground loom, in which no

comb was used.









In addition to learning a new weaving technique, women have also begun to incorporate

designs previously used only in baskets. Initially, men taught the women to copy the basketry

designs into textile pieces. Some women subsequently began to copy other designs from baskets

or, more recently, from printed photographs. At least five new different graphic motifs have been

incorporated through this way into women's textiles. The Kaiabi have been copying their

basketry designs into other objects as well, including bags, clubs and benches, all of which are

made for sale (Athayde, 1998).

Comparing basketry and textile weaving among men and women, we can conclude that

while less innovation was developed in basketry, where no new technique was developed, in

textile there was a significant development of techniques and incorporation of items to the

traditional repertoire. It is also interesting to note that, in spite of learning a new technique from

another indigenous group (the Yudj a), Kaiabi women learned the designs represented in the

twill-plaited baskets from the men and have represented them in the new items they came to

produce, such as straps for babies, hammocks, bags and belts. The innovation developed by the

men was the application of graphic designs in other objects and in body painting.

Twill-plaited Designed Baskets

Kaiabi basketry encounters its greatest expression in the twilled-plaited baskets and

woven club handles, both made by men. Through a symbolic language woven into nearly thirty

different named graphic designs, the making of baskets reflects diverse aspects of Kaiabi

cosmology and social organization (Ribeiro 1987 a; 1987 b; Athayde 2003; Athayde et al.,

2009). Specific names are used for each kind of basket: iparupit are simple strainers used by

women to sieve cassava flour and some beverages (yripemeaii, yripemeauu), araa are painted

baskets with different woven designs, while the panakii is a kind of decorated 'backpack'

traditionally used to carry hammocks during travels. Designed baskets are traditionally used by









women to store cotton while spinning as well as to store small objects. However, baskets are also

used as an ornament and symbol of prestige, placed hanging from ceilings of artisans' houses.

Grunberg (2004) mentions that painted baskets used to be the most important item of the dowry,

given to the bride. Due to the beauty and artistic value of these baskets, the Kaiabi have traded

them with other indigenous peoples, as well as increasingly destined their production to the

market.

Stages of Basketry Weaving

It takes five to ten days to make a decorated araa basket, depending on the size and

design. In a previous work (Athayde, 2003) I identified six main stages involved in the process

of weaving a painted basket (Figure 8-3):

1. Harvesting the natural resources. The natural resources used to make baskets occur in
different ecological zones. It can take time to collect the main fiber "arum'", or its
substitutes, the dye used for painting and the vine used for the rim of the baskets. The
ways through which the Kaiabi harvest and manage these resources are described in the
section on natural resources.

2. Preparation of fiber strands. After arriving home, it is important to separate the stems and
pith (removing the inner soft part) as soon as possible. With a knife, men prepare the
stems and put them to dry in the sun. They leave them to dry for one day. Then it is time
to measure the strands and divide them. They do it with the help of hands and mouth,
trying to keep all the strands the same width. Then they prepare a bunch of strands in
order to begin to weave. The rest, the strands of worst quality, mainly the thicker ones,
are used to make other kinds of baskets, such as "yrupemeauu" (a basket used to sieve
cassava flour).

3. Weaving. Before beginning to weave, a person needs to decide which design is going to
be created, to initiate the counting. The beginning of the basket, or initial point is called
"i )1' )'pingap" or "i 'yp", which means a "way or a path to follow". "I'yp" is also the
name for a basket design woven in the simple unpainted baskets. The weft is arranged,
composed by a group of strands with the rough side up and another with the smooth side
up. When the square is done, some pieces have different lenght. So, they measure and cut
off the tips. Then they can begin to weave the rim.

The Kaiabi counting system is based on grouping numbers. They count up to five, then

they group the numbers they know. For example, 8=4+4; 10=5+5. Mendes (2001), in a study on









the ethnomathematics of the Xingu indigenous peoples, referred to the practice of counting by

grouping elements in the process of weaving a design in a basket. Through conversations with

Aturi Kaiabi, she realized that in basketry weaving, the counting follows a group order, coming

from a symmetric division. This way of counting, which can encompass a huge amount of

different combinations, producing different designs, is what we can call Kaiabi mathematics

(Mendes, 2001). The designs also represent "numbers" for the Kaiabi. For example, to weave the

design "'yp", one of the first designs they learn, they use 17 strands in the vertical position.

From the centre of the design, or first point, they weave two groups of three and one group of

two strands for each side of the weaving square, as shown in Figure 8-4.

4. Tying up the rims. Basket makers need to seek good vines in order to prepare the rim.
They use a double rim. The upper part must be shorter and the lower part longer. These
two parts are tied with "arum'". One part of the basket enters into the rim to make the
concavity. Then the four corners of the basket are tied with the rest of the rim. After
everything is ready, they tie it with cotton. It takes about two days to tie a big basket.

5. Painting the basket. A reddish dye is applied to the surface of the basket using the hands.
Four or five layers of dye are applied, so that it fully adheres to the basket. Baskets are
then put out to dry, preferably in the hot sun, for one entire day. The dye does not adhere
on the outer or rough side of the strands, but to the inner or smooth side.

6. Scraping and finishing. To remove the dye, men use a wooden stick or a brush and scrape
the basket, getting it off the rough stems, to which it does not stick, revealing the finished
design.

Ribeiro (1980), researching a collection of 960 baskets from different Brazilian

indigenous groups, found that the painting technique used by the Kaiabi is shared only by two

other indigenous groups: the Tapirape and the Paresi. This technique is called posterior painting

or "scraped" painting. The more common technique is to paint the strands before weaving. She

suggests that the sharing of these technical styles and of some designs is an indication of

historical contact between the groups, as well as of the possibility that they could have learned









some techniques from each other. Some Kaiabi design patterns are also present in Bakairi indian

baskets.

The different designs reproduced in these baskets represent a symbolic language that is

unique to the Kaiabi people, materializing and transporting history, mythology and ethnicity

through time (Ribeiro, 1987b; Athayde, 2003). Grunberg (1970/2004) identified 12 graphic

designs for the Kaiabi baskets during his stay with the group in the 1960s. As I mentioned in the

methods chapter, some of the photos taken by Georg Grunberg and the material organized by

Berta Ribeiro (1987b) were recuperated and incorporated in the photographic catalogue of

basketry designs I have been organizing for the Kaiabi in the last ten years. So far, I have

documented 29 designs used for the twill-plaited baskets, including variations, and 7 designs

applied to the panakfi backpack basket, totalling 36 designs. The process of learning the designs,

as well mythical aspects and knowledge distribution related to them, will be explored in further

sections.

Natural Resources Used in Basketry and Textile Production

The Kaiabi use ten main species in basketry and textiles weaving, without taking into

account "arumd" substitutes, which are presented in a separate table (see Tables 8-2 and 8-3).

The activity of collecting and preparing the plants for use can take many days, taking into

account that sometimes men need to go to distant places, far from the village, to get the materials

they need. The process of harvesting these plant species can be combined with other forest

activities, mainly hunting and collecting fruits. Therefore, a man may go to the swidden-fallow

plots or to the forest with the specific intent of collecting resources to weave a basket or he might

be hunting and find a vine or fiber that he would use later, and bring it home.

In Chapter 5, I mentioned the environmental contrasts between the ancestral territory and

the Xingu Park, to which the majority of the Kaiabi population was transferred around fifty years









ago. As a result of their relocation to Xingu Park, they lost access to many important plant and

crop species, including those used for weaving baskets and cotton varieties used in textile

production. A particularly important plant resource missed in Xingu is uruyp (Ischnosiphon

gracilis, Marantaceae, arumd in Portuguese), the main fibre used in Kaiabi designed baskets.

This herb forms clumps in swampy and periodically flooded areas throughout lowland Amazonia

(Andersson 1977, 1984), but occurs only in very scattered and small populations in the

northwest, more humid, region of the Park. Moreover, the most prized variety of aruma, uruyp

ete ('truly aruma'), does not grow at all in Xingu and can only be found in the ancestral territory

in the Tapaj6s (Athayde et al. 2006). The Kaiabi say that uruyp ete is of best quality for the

baskets, and the uruyp kuruk is classified aa a second class resource. According to them, in the

ancestral area, there are large "aruma" clusters, that are not so restricted to special types of

habitats as in Xingu, and which, as they affirm, "can be found virtually anywhere."

To gather the "aruma" stems, the men have to decide which plants are ready to be

harvested. They can tell by the thickness, height and color of the stems. If the plant is not mature,

the stems get soft and break. Most of the time they collect the stems above the first node to allow

the plant to re-sprout. They cut the stems with a machete, the leaves are removed and the stalks

are bundled and fastened.

Due to the scarcity of aruma in Xingu, the Kaiabi started to search for substitutes for

weaving the body of the baskets, currently using at least six substitutes for aruma (Table 8-3).

The use of substitutes is interesting in this case because it helps to prevent the erosion of

basketry knowledge due to the lack of arumd. Most, such as wywa (arrow cane, Gynerium

sagittatum) are weaker and cannot be properly dyed. Arumd' s superior quality and unique

cosmological importance means that weavers are reluctant to use substitutes. Kaiabi creation









narratives tell how the ancestral hero and shaman Tuiarare used to spend hours weaving baskets

in his hammock. Under a pile of discarded arumd behind his hammock lived a larva, which

during the night transformed itself into a beautiful woman, who became his wife. Some elders

said that this larva is the 'owner' of aruma, and that it has taken care of the plant until now.

There is also a graphic design called 'worm' or 'larva' (yok), which is probably related to this

narrative (Athayde et al. 2006). According to shamans and elders, the aruma plant has an owner,

a spirit who takes care of the plant. Kupeap told me that the name of this spirit is akwaap or

akwaawi.

The symbolic meaning attached to weaving and arumd is traced back to the epic travels

of the Kaiabi cultural hero Tuiarare, who ran away with a piece of the skin from an ancestral

supernatural snake, subsequently learning how to weave baskets using aruma by copying the

designs represented on its skin (see following section on shamanism). The unique social and

symbolic importance of arumd is shared by many other Amazonian indigenous groups, including

the Wayana, Baniwa, Yekuana, Desana, Warao and Aparai (Guss 1989; Van Velthem 2001;

Wilbert, 1975). The link bestowed by these groups between arumd and the 'skins,' of primordial

humans or of supernatural beings, also endows the baskets made with this plant with a unique

meaning and power (Van Velthem 2001).

Concerned about the scarcity of arumd in Xingu, Kaiabi representatives have been

developing activities for cultural revitalization and management of natural resources used in

basketry production in partnership with the Brazilian NGO Instituto Socioambiental and the local

organization Associacgo Terra Indigena Xingu. From 1999 to 2006, these groups conducted

participatory and collaborative research on use and management of aruma with Kaiabi

communities in the northern region of Xingu Park. This work included ethnobotanical research









on harvesting techniques, the documentation of plant characteristics, and investigation of

mythical aspects related to the plant and baskets. The research activities were developed with

Kaiabi communities and young men who have been trained to work as natural resource managers

in their villages, and who were participating in the coordination of the activities (Athayde et al.,

2006). In 2004, saplings of uruyp kuruk, uruy wete and uruyppirem 'i (another variety) were

brought by truck from Teles Pires area to Tuiarare village in Xingu, as part of the activities

developed under the Kaiabi Araa project. The saplings grew very well in the place that Tuiarare

villagers assigned to plant them, close to a little stream. However, there were few saplings,

which were not enough to support basketry weaving activity in the village, especially if part of

the production is destined for the external market. In spite of efforts made with ISA's support to

bring this wild plant into sustainable management and cultivation, these are experimental

activities that won't support the continuous production of basketry by the Kaiabi. Every non-

timber forest product (NTFPs) has its own ecology, limitations and constraints when it comes to

sustainable management. In the case of arumd, management is constrained by ecological

characteristics of the plant such as slow growth, special habitat requirements, and reproduction

that probably is constrained by the lack of water in Xingu (extended dry season) (Athayde et al.,

2006).

Palm species are also important in basketry weaving among the Kaiabi, especially the

tucuma (Astrocaryum aculeatum) and the inaj (Maximilliana maripa). These species are

normally associated with swidden-fallow practices and might be present on old fallow plots or

secondary forests around the villages. They are also used for many other purposes, such as for

house thatch, jewelry, food and oil and as substitutes for the "siriva" (Bactris macana) wood,

which does not occur in Xingu, in the confection of spindles and clubs.









Another important species that is rare within the Park is "jemore 'yp" (jequitiba,

Cariniana sp, Lecythidaceae). This tree occurs sparsely in the forests along small streams,

mainly in Capivara and Kururu villages. Kaiabi extract the thick red resin present in the bark of

this tree to dye the baskets, first by hitting the tree many times with an axe to loosen the bark.

Then they remove it with the help of a machete and axe until the amount of bark needed is

released from the tree. They cut the fiber and squeeze it into rolls in a container. The dye is then

stored and ready to use. The Kaiabi need a large amount of dye in order to decorate baskets,

since they require four to five layers of application for each basket. To obtain two liters of dye,

they have to extract at least three meters of bark. When the bark is removed from the entire

circumference of the tree, it may die. Even when it does not, it can take almost five years for the

tree to recover the bark. The Kaiabi use a substitute for "jemore 'yp", named "ujupe" (Myrcia

deflexa, Myrtaceae) which is the same dye that the women use to make calabashes impermeable.

This tree is commonly found in old swidden-fallow plots and young secondary forests close to

the villages. Again, they say that the quality is not so good as the "jequitiba" dye, but it is an

option for villages where it does not occur.

To make the rim of the baskets, the Kaiabi use two species of vine, both called

"37 iq pep,,iu' (which literally means "rim of the basket"). The extracted part of the vine is

brought to the village and the outer part removed using a knife. It is dried in the sun for one day,

and is then ready for use.

According to Silva (2002 a), there are two traditional varieties of cotton Gossipium

barbadens, (Malvaceae), cultivated by the Kaiabi: amyneju owising white cotton, and amyneju

pytang = owiwytang brown cotton. The third type found in some Xingu villages is a variety of

G. hirsutum, the most common and commercial cotton, that the Kaiabi got from the non-









indigenous. They name it amyneju piamuku, because of its long fibers. Kaiabi women still

cultivate traditional cotton varieties to weave textiles in Xingu and in Rio dos Peixes. Seeds of

the traditional cotton varieties were brought from the Tapaj s at the time of the transfer to

Xingu. People in Rio dos Peixes told me that the brown cotton variety no longer exists there.

Catarina, the most traditional elder in Rio dos Peixes, said that she brought a cotton variety from

the Bakairi and planted in her homegarden, but apparently it is darker than the white variety

normally cultivated by the Kaiabi. In Kururuzinho (Teles Pires), only one woman has the white

variety planted in her family farm garden plot.

In the interviews, I asked women whether they had cotton planted in their farm plots, and

which variety of cotton they had. The results provide an estimate of the situation of cotton

varieties in all four villages, taking into account that many times a woman has her farming plot

together with a relative, and also that cotton production may vary from one year to the other. In

Capivara, 71% (17) of the women reported that they have cotton planted in their farming plot.

All of them have the white variety (amynejusing), and five of them have both the white and the

brown variety planted. In Tuiarare, only 27% (6) of women interviewed have cotton in their

farming plots, and of those, only one has brown cotton. In Rio dos Peixes, 22% of women still

have cotton planted, all of the white variety. In Kururuzinho village at Teles Pires, only one

woman has cotton planted (3.5%). Her name is Sara, young daughter of the artisan Jodo, who

participated in the workshops of Kaiabi Araa project and got cotton seeds from Xingu relatives

to plant in her plot. The majority of women who have cotton in their plots are older, mostly over

30 and 40 years, since younger women share the plot with their mothers. From these results, it

seems that the brown cotton variety (amynejupytang) is really threatened with disappearing

among the Kaiabi. We don't know about the situation of this variety in other Xingu villages, but









it looks like women tend to substitute its use by the colored industrialized cotton bought in

cities. Xingu works as a reserve of cotton varieties, as in the other two areas cotton cultivation is

almost lost.

According to Jurufet from Capivara village, women take care of the cotton and keep the

cotton seeds. Men and women can participate in the planting time, which occurs during the rainy

season, mainly in September and October, when the Api tree is producing fruits. Taangap from

Capivara told me that when it rains for the first time after the dry season is over, they plant corn.

When the corn is growing (around 8 inches) they start to plant the cotton, putting two seeds in

each hole. The harvest occurs during the dry season (between May and August). Juwete said that

other women (relatives) help each other during the harvest period. They have to harvest it soon,

before the rain starts, otherwise there is a certain type of larvae that eats the leaves. If you plant

too much, there is not time enough to harvest and you might end up losing production and effort.

Also, when you gin the cotton, taking the seeds out, if you keep them until the next season, they

get rotten. Therefore, you have to harvest a little to have seeds right before planting again.

Women told me that the cotton plant produces only once in the old fallow plot, after which the

plant dies, because the trees grow and produce shade, and larvae eat the leaves and also the

seeds.

Many women prefer to buy industrialized cotton to weave big hammocks, given the

amount of time and work that manual spinning requires. Thus, I could say that dependence on

industrialized cotton is greater in Teles Pires, then in Rio dos Peixes and lastly in Xingu.

Moreover, in order to weave the cotton straps, hammocks, bags and belts with colourful

contrasting designs borrowed from basketry, women need to use industrialized colourful cotton

bought in cities, which has become an important exchange item.









Weaving Knowledge: Social Meaning, Transmission, Distribution and Change

In this section, I will present results related to weaving knowledge from the interviews

conducted with all men and women over 15 years of age in four Kaiabi villages, two in Xingu,

one Rio dos Peixes and one in Teles Pires. The longitudinal analysis involving data gathered for

this research (2007) with data collected in 2002 for my previous Master's thesis is going to be

focused on men's knowledge, specifically of the designed baskets, since I don't have baseline

data for the women. Also, in my previous research, I did not include knowledge on the other

types of baskets, or ability to weave them, which was included in this study. I will explore

mechanisms, processes and trends involved in Kaiabi weaving knowledge distribution,

transmission and change between genders and across villages.

Social Meaning and Traditional Learning Mechanisms

According to Silva (2000), processes of production and reproduction of material culture

objects by indigenous peoples reinforce principles of their social organization; of the

construction of the individual as a person; of the knowledge of using and managing natural

resources; and in the consolidation and maintenance of kinship ties. Among the Kaiabi, as well

as with other Amazonian indigenous peoples, mastering skills in weaving baskets and textiles is

considered necessary in the preparation of a young adult to be ready to get married (Riviere,

1992; Crickmay, 2002). Husbands produced baskets for their parents-in-law as part of their

bridal service (Athayde, 2003). Comparably, a young woman ready for marriage is expected to

know how to weave hammocks and straps for carrying babies. Traditional mechanisms for

learning to weave among men and women involve observation, copying, doing, undoing, and

repeating (Silva, 2000; Crickmay, 2002; Zarger, 2002). In the Andean regions for instance, the

concept of learning is related to experience (Crickmay, 2002). Learning starts early in life with

observation of parents or of other members of the social group performing a given skill. As









learning to weave happens mostly in the household for girls and includes bush skills and

weaving skills for boys, we can expect differences in the teaching-learning experience among

genders.

Pelissier (1991) mentions that skills such as building a canoe or weaving a hammock are

comparable to norms and roles, and are learned in daily experience. For the designs, it also

entails mastering a particular system of counting (Mendes 2001). Counting is important in the

beginning of the learning process, but after a man or woman masters the skill of weaving, they

don't need to count anymore. Prochaska (1990:113) mentions about Andean weavers that

numbers are important to the beginning weaver, but as they become recognized as expert

weavers "they don't know how to count, because they already know how to weave."

In spite of the similarities that can be found regarding the social meaning and traditional

learning of basketry and textile items, there are some differences between genders that I will

highlight during this and subsequent sections. The description of traditional modes of learning

and knowledge transmission initially presented here refers to an ideal situation, which has

changed significantly in the real world. I will contrast the traditional modes of learning and

transmission with the actual distribution of knowledge and changes in learning and transmission

that have been taking place among the Kaiabi since the diaspora. Hewlett and Cavalli-Sforza

(1986:922) define cultural transmission as "a process of social reproduction in which the

culture's technological knowledge, behaviour, patterns, cosmological beliefs, etc. are

communicated and acquired."

Traditionally, in the past, young men acquired basketry-related skills and knowledge

while accompanying their male kin- notably father, brothers, brothers-in-law, uncles and

cousins- in their daily tasks. Starting around 10 years of age, boys would start learning by









accompanying their male kin on forest excursions, thus getting exposure and familiarity with the

natural resources used in the production of the different types of baskets. While in the forest,

they would observe and learn how to extract and prepare the natural resources, beginning with

the simpler disposable baskets such as the vine basket for cassava transportation (myayta). Silva

(2000) also noted this mechanism of learning basketry among the Kayap6-Xikrin in the Amazon.

Apart from the myatyta, considered the easiest basket to be made, the young Kaiabi would also

start to learn to weave fans (tapekwap). Young men were expected to learn at least basic types of

baskets to provide for their families.

Concerning twill-plaited baskets, a twelve-year old would know how to harvest arumd,

prepare the strands for weaving, and weave simpler, unpainted baskets containing designs such

as i 'yp, ipirien orjarukang, and inimo eta (Athayde 2003; Athayde et al., 2009). After mastering

weaving of simple baskets for his wife and mother-in-law, he would then learn the more complex

designs according to his own will and ability. There is a certain sequence in which basketry

designs should be learned, from the simpler to more complex designs.

According to the interviews, Kaiabi girls used to begin to learn to weave cotton by the

age of seven or eight, before many boys learned basket-weaving. In contrast with boys, who

would start learning from other kin during visits to the forest, girls would often learn to spin

cotton in their households from their mothers, grandmothers, aunts or other female relatives

(Crickmay, 2002). In the beginning, learning happens naturally as play or a game, in which little

girls imitate the adults and weave little "toy" pieces for their imaginary babies. Fabiola Silva

(2000), describing pottery making and learning among the Assurini women (also Tupi-guarani as

the Kaiabi), registered the same process of miniature construction among young girls learning to

make their first pottery pans. Among the Kaiabi, after her first menstruation, a girl would learn to









weave hammocks as part of her preparation to get married. While in the beginning of the

weaving apprenticeship the girl would be learning more directly from closer kin women such as

her mother and grandmother, in later stages of learning to weave hammocks, including mastering

of different designs, the young woman could learn from other kin and even unrelated women.

This is also true for men and basketry apprenticeship. As with men and baskets, being a skilled

weaver was a requirement for marriage, and is still a source of pride and prestige. In contrast, not

knowing to weave the basic items required to provide for his or her family is reason for shame

and mockery among the Kaiabi. During my fieldwork and interviewing process, many times I

witnessed people making fun of others by saying: "she doesn't even know to spin cotton;" or

"my husband does not weave for me. When I need a basket, I have to ask my uncle or other

men." In spite of the perceived importance of knowing how to weave until present times,

changes have happened in the motivations for learning, and social meaning and value of

knowing. This can be observed especially with the more elaborate items produced for sale. If in

the past a skilled basketry weaver was admired for his capacity to provide for his family,

nowadays he is still admired, but increasingly for his access and capacity to participate in the

market. Many men interviewed in the three Kaiabi lands mentioned that they want to learn to

weave baskets because they can sell them, instead of providing them for their female kin.

Knowledge Distribution of Woven Items

Weaving knowledge is unevenly distributed within and between the four villages

considered in this study. Distribution differences are mostly related to age, gender, kinship and

village. As expected, in Xingu villages, both men and women weave more basketry and textiles

items (Tables 8-4 and 8-6; Figures 8-5 to 8-7), and more people have the ability to weave

designed baskets and textiles. Considering basketry as a whole, including all the 21 items

assigned for the category, Rio dos Peixes is better off than Teles Pires in terms of number of men









who are able to weave any of those objects. This was in part a surprise for me, because I

expected weaving knowledge to be almost lost there, given the situation of the village in terms of

socioeconomic factors such as language loss, proximity to town, ethnic mixture and market

influence, explored in the last chapter. Observing Figure 8-5 it is possible to see that Tuiarare is

the place where more villagers know the most items, followed by Capivara, Rio dos Peixes and

lastly Teles Pires.

Older men weave more types of baskets. The elder Tewit from Capivara (who moved

back to the village in 2007 after living with his family in Rio dos Peixes for approximately 5

years) knows how to weave 19 of the 21 items (90%) that compose Kaiabi basketry repertoire. In

Capivara, another knowledgeable men is Kupeap, who can weave 12 (57%) of the 21 items. In

Tuiarare, there are six men who can be considered specialists in basketry, all of them weaving

more than 12 objects. Aturi is the one that knows how to weave the most objects in the village

(17 or 81%), followed by the middle aged Miau'i (76%), the shaman Chico (71%), elders Masi'a

and Xupe (both with 66% of the items) and the younger Tare'i (62%). In Rio dos Peixes, men

who know to weave more basketry items are the middle-aged Raimundo (86%); the elder and

shaman Tafut (86%); Simdo, another elder and specialist in basketry designs (71%) and Moacir,

who lived in Xingu for a long time, with 66%. In Teles Pires, there are only two experts in

basketry; the most knowledgeable is the elder Corone (knowing to weave 66% of the basketry

items), followed by Jodo (52%).

Regarding the different items of basketry, the simple vine basket (myayta) is the item

most men weave in all villages, totalling 55% of all men interviewed. In second place comes the

woven club (muap), with 51%. In Rio dos Peixes, 72% of men know how to produce the club,

but the majority wove the handle in a very simple way. The baskets used as strainers by women









are also popular among the men, being produced by nearly 35% of them. Mastering the ability to

weave these strainers was a required skill to get married in past times. Other popular objects

produced by men are the palm-leaf fan, the basket used for storing peanuts and, in sixth place,

the twill-plaited designed baskets or araa, known by only 25% of all men interviewed (see

Figure 8-8). It is interesting to note that, apart from the club, the most popular items are the

baskets done by men and used by women for food transportation and processing. The club is an

exception in this case, because nowadays it is produced mostly for the market. More elaborated

baskets such as the backpack basket (panakf) and the little palm leaf bag (yrfokote 'em) are in

danger of disappearing among the Kaiabi. There is only one person or 0.9% among the

interviewees who can produce the panakf (Tewit, from Capivara). Raimundo, 41 years old and

living in Rio dos Peixes (son of the deceased Kwasiari, a great basket-maker in Rio dos Peixes)

told me he knows how to make the yriokote 'em. Related to this, Ohmagari and Berkes (1997)

observed among the Cree indigenous peoples of Canada that skills which are more essential for

livelihoods were the ones which still prevailed and were being actively transmitted within the

population. Among the Kaiabi, baskets that are more commonly used and which don't have a

comparable industrial substitute are still being produced by the men for women to use.

We have to consider the dynamic process of knowledge transmission between the

villages and kin groups. Many men that live in Rio dos Peixes today spent variable amounts of

time in Xingu, and this also happens in Teles Pires. The simple fact of the existence in the village

of a master in such and such an object may trigger the will to learn in other men. If the person is

constantly weaving, some young men may get interested in learning by observing the producing

process and the object ready to be used or commercialised. For instance, I know that Canisio, the









ex-chief of Capivara village, when he moved to Rio dos Peixes, brought knowledge on artifacts

and baskets and started to spread it to his sons and other kin groups.

Among the women, there are only six items that compose the textile repertoire in

comparison to 21 for basketry. What we have here is a more kin-oriented process of knowledge

distribution and transmission in comparison to basketry among the men. Also, Capivara is the

village where overall knowledge on textiles is greater, apart from the woven bag, which was a

new invention promoted in workshops among Tuiarare women (Figure 8-6). In Capivara, five

women know how to produce five of the six textile items. Four of these women are mother and

daughter. For instance, the wife of the health agent in Capivara (who does not weave designed

baskets) is a very knowledgeable weaver, who knows how to make different items and designs

and teaches her daughters. Capivara is the village where there is more cotton being planted, and

where the brown variety is still cultivated. In Tuiarare, only one woman knows how to weave all

the six items (Moreru). Zulmira, an older woman married to Xupe, knows how to make five of

the six items and was teacher at the workshops of the Kaiabi Araa project that happened in the

village. In Rio dos Peixes, four women have the skills to make four of the six textile items. Three

of them lived in Xingu for varied periods of time, therefore they know how to make plain and

designed hammocks. Suzana, daughter of the knowledgeable elder Catarina, is among the few

women who are still able to make the traditional hammock, in which a ground loom was used.

Again, Teles Pires is the place where the majority of women did not know how to weave before

the classes given by Xingu women at the workshops of the Project Kaiabi Araa. Therefore, the

women who learned in those workshops do not know how to spin cotton. There are two girls in

Teles Pires who can weave four textile items. Looking at figure 8-6, we can observe that all the









items are produced in all the villages, differently for men, in which some items were mastered by

only one individual.

The item that more women in all villages know how to make is the strap for carrying

babies or "tupai," with 52% of all women interviewed being able to weave it (Table 8-6; Figure

8-9). This makes sense, taking into account that the tupai is a common item, used daily by

mothers to carry their babies. Also, many times the young women learn to make the tupai first,

including some simpler designs, coming to learn to weave hammocks afterwards. More women

know to make simple plain hammocks (41%) compared to the designed hammocks, which 31%

of women know. The other three items are the traditional hammock (known by 19%), of which

knowledge is in process of disappearing among Kaiabi women, since it has been substituted by

the simple and designed hammocks; the woven belt (18%), a novelty developed in Capivara

village, made by women for men to use in festivals; and lastly the woven bag with graphic

designs, developed by Tuiarare women and made mostly for sale (17%). We can speculate that

while knowledge of the traditional hammock is going to be lost, knowledge of belts and bags

will increase, since these items also can be produced for the market.

At this point, we can conclude that knowledge on textiles in more evenly distributed

within the Kaiabi people than knowledge on baskets; more kin-oriented and dependent; and more

resilient or less prone to disappear, since it is more embedded in kinship and social structure and

includes less sophisticated items. Also, the fact that women can use industrialized cotton to

produce textiles is an advantage in comparison to men who need to harvest and process natural

resources to produce different baskets, especially the twill-plaited painted baskets for which they

prefer to use the scarce arumd as the raw fiber. In Rio dos Peixes, weaving knowledge still

persists among some women, who are proud of their work. What also helped to conserve









knowledge there was the work of Catholic missionary nuns, who, according to some

interviewees, used to promote weaving workshops for the women. In both cases, weaving

knowledge still persists in Rio dos Peixes, contrary to expectations that knowledge would be

largely lost. In Teles Pires, it seems that external incentive and access to market may be the most

important factors in improving weaving ability among men and women who are still in learning

ages.

Knowledge Distribution and Transmission of Designed Basketry and Textiles

This section explores knowledge distribution and transmission of selected items of

basketry and textile repertoires. I chose to carry out more in depth research on knowledge

associated with designed baskets and hammocks because of the following factors: a) request

from the Kaiabi for me to advise them in a cultural revitalization project involving designed

basketry and textiles; b) existing data from previous research done in 2002 and 2004; and c)

social, cultural and economic importance of these artifacts. In this research, I repeated interviews

done with men in 2002 at Tuiarare and Capivara villages, added Rio dos Peixes and Kururuzinho

villages, and also added women's knowledge. I will be presenting the results and, to the extent

possible, comparing with my previous findings. In 2002, I interviewed 20 men in Capivara

village and 25 in Tuiarare (Athayde, 2003). From these, 12 were interviewed again in Capivara

in 2007 and 15 in Tuiarare, which gives us more than half of the same people interviewed in a 5-

year interval. However, when looking specifically to weavers that master designed baskets, there

were only 7 men in Capivara and 6 in Tuiarare who were interviewed twice. This is a limiting

factor to consider for quantitative data comparisons between the two years. Nevertheless,

qualitative data available, and previous findings, can be explored in more detail, and the process

of observing changes that have occurred in the communities during this time interval provided

fertile material that was explored in this research.









According to Boster (1987), patterns of intracultural knowledge distribution may depend

on domain specificities, social contexts and learning opportunities. For men, knowing to weave

designed baskets can be considered specialized knowledge, that is, knowledge which is not

shared among all individuals in an ethnic group, or that is not absolutely required to get married.

For women, given the importance of designed hammocks nowadays, especially in Xingu

villages, knowing to weave designed hammocks is increasingly a symbol of status and readiness

to support a family. However, because of the availability of industrialized hammocks

everywhere, weaving hammocks would not be indispensable for living. Whatever the practical

value of these items, we cannot underestimate the power of beauty and identity which they bring

to the Kaiabi. Therefore, besides being beautiful, useful and appreciated artifacts, designed

baskets and hammocks also carry a symbolic iconographic language that is unique to the Kaiabi

people (Ribeiro, 1986; Athayde et al., 2009). If we consider knowledge in the weaving domain

as encompassing form, function, materials, meaning and designs (Ellen, 2009, my adaptation),

evaluating whether people are transmitting and capturing all these components may be a

challenge, especially regarding the social and spiritual meanings attached to the object. Mairaw6

Kaiabi once called attention to this fact, mentioning that nowadays a young man might be

learning to weave a basket, but might not be capturing the deeper meaning that is embedded in

that object. It's good to have that in mind.

Differently to what we observed when considering the whole group of items that

constitute basketry and textiles, knowledge on special items such as designed baskets and

hammocks is much lower in Rio dos Peixes (Figure 8-9). While the majority of people know

how to weave these items in Capivara and Tuiarare villages, in Rio dos Peixes only 7.5% of

women and 14% of men master this skill. Results for Capivara and Tuiarare are comparable and









very close: while in Tuiarare there are more women who weave designed hammocks (82%), in

Capivara more men know designed baskets (68%). The results also show that more women

weave hammocks than men weave baskets. This is probably related to what was discussed

above, on specialized knowledge and on the way knowledge is transmitted, as part of household

dynamics and more embedded in social organization in the case of the women and hammocks. In

Teles Pires, the percentage of women and men who weave designed items is close (32% for

women and 34.6% for men). These percentages increased significantly from 2004 to 2007, since

before the workshops of the Kaiabi Araa project, no women in the village.knew to weave textiles

I will come back to this point in the next sections, exploring mechanisms of cultural

transmission.

Age of learning

This section details when people start to learn to weave designed baskets and textiles.

Even if the learning process starts early in childhood with observation and copy of simple woven

items, there is a time when men and women actually learn to weave graphic designs into objects.

This should happen, as previous stated, in the stages preceding marriage, so that the person has

an asset to offer to his or her partner.

To better capture the variation in the ability to learn to weave designed baskets and

textiles, I assigned five age classes to classify people according to the age with which they learn

to weave: 1) From 10-15 years of age; 2) From 16-21 years; 3) 22-30 years; 4) 31-39 years and

5) 40 years old and up. Learning to weave twill-plaited baskets and designed hammocks starts at

an early age, with boys and girls accompanying their parents and relatives in daily tasks and

excursions to the forest (in the case of boys). According to the weavers interviewed in each

village, most people learn between 10 and 15 years of age, and the great majority learn before

they are 21 years old (Figure 8-10). This is especially true for the women in all villages except









Tuiarare, where the ages of learning are more distributed between age classes and where there is

a woman who learned to weave designed hammocks at the age of 40 during a workshop of the

Kaiabi Araa project. It is curious to note in Figure 8-10 that the villages in which the project

Kaiabi Araa happened (Tuiarare and Kururuzinho-Teles Pires) are the ones with more people

learning at later ages.

Results found here are corroborated by Silva (2009) studying knowledge related to

agrobiodiversity, specifically knowledge related to peanut varieties among the Kaiabi in Xingu.

He found that 91% of men and women interviewed reported that they learn about peanut

varieties between 8-15 years of age. Zent (2009b) presents data comparing TEK transmission

among four Joti communities in Venezuela, finding that in three of them there is an increase of

competence by age among younger people up until 20 years of age, with no further significant

acquisition after that. Ohmagari and Berkes (1997) also found that the age range between 13 and

15 years is the most important period for the mastery of most of the bush skills, before the age of

marriage.

Compared to data collected in 2002, age classes mentioned by men from Capivara in

2002 differ from those of 2007, in which four age classes are represented instead of two

registered for 2007. Discrepancies in men's answers about the age of learning to weave baskets

are found for both Capivara and Tuiarare villages, including discrepancies for the same

interviewee. This shows that data related to age of learning can be very changeable among

indigenous peoples. However, in spite of discrepancies, we can conclude that learning to weave

happens at early stages of transitioning from childhood to teenager for both men and women, and

that for women this apprenticeship happens even earlier than for men. Also, there are changes in

age of learning in villages where strategies other than traditional mechanisms have been used for









encouraging learning, such as the workshops promoted by the Kaiabi Araa project in Tuiarare

and Kururuzinho.

Reviewing models and mechanisms of knowledge transmission

As previously described, mechanisms for learning in indigenous societies are mainly

based on socialization into a group, observation, copying, trial-and-error and learning by doing

(Crickmay, 2002; Bowser and Patton, 2008; Athayde et al., 2009). Transmission can occur

within the household, between parents and child; among peers; between an older teacher and a

younger apprentice; between a teacher and his/her students (such as in indigenous or non-

indigenous formal schools); between various elders and one student; and also, as I'm going to

propose, between many teachers and many apprentices. I contrast the results of this research with

existing models of cultural transmission and propose a new approach that better captures the

specificities and innovations that indigenous peoples have developed to transmit technological

and artistic knowledge.

In the cultural evolution model developed by Cavalli-Sforza and Feldman (1981), there

are two stages in the transmission process, namely awareness, requiring the existence of a signal

(e.g. through observation), and acceptance or learning itself. The authors developed models of

cultural transmission (between transmitter and transmitted) in human societies, later reviewed by

Hewlett and Cavalli-Sforza (1986), grouped in five main modes of cultural transmission (Figure

8-11):

1. Vertical or parent-to-child closest to biological transmission, it occurs between
generations and can include one or both genitors in the transmission of a given skill or
knowledge. According to the authors, acceptance of innovation in this mode is of
intermediate difficulty and cultural evolution is slow.

2. Horizontal or contagious transmission happens between any two individuals
independently of their relationship. It can happen between same age peers, related or
unrelated. The acceptance of innovation is easy, and cultural evolution can be rapid.









3. Oblique transmission occurs between generations other than parents, such as between a
related or unrelated older person (of similar age range as the person's parents) to a
younger person.

4. One-to-many one main transmitter passes knowledge to many transmittees. This relates
to teacher-leader or media (TV, internet, books) type of transmission and it is dominant
nowadays, with advances in technology such as internet. Acceptance of innovation is
easy and cultural evolution might be rapid.

5. Concerted or many-to-one This type happens between older members to a younger
member of the same social group. It is assumed that all transmitters act in concert, so that
the influence is reciprocally reinforced. According to Hewlett and Cavalli-Sforza (1986),
this mode, involving a high level of social control, would lead to very difficult acceptance
of innovation and thus would be considered the most conservative in terms of cultural
evolution.

In 2002, 30% of the male basketry weavers interviewed in Capivara and Tuiarare village

told me that they learned to weave designed baskets from their fathers (see Figure 8-12). Vertical

transmission by parents was also reported by Silva (2009) as the main mechanism for

transmission of knowledge about peanut varieties among the Xingu Kaiabi, reported by 52% of

his informants. Ohmagari and Berkes (1997) also found that parents, especially the mothers,

were the main teachers among Cree women, and that grandmothers were also important. The

importance of vertical transmission (parent-child) has also been registered in other studies

(Ruddle and Chesterfield, 1977; Hewlett and Cavalli-Sforza, 1986).

Interestingly, the same percentage (30%) said that they had "learned alone, nobody taught

them" (Athayde 2003). In fact, much of the work a weaver does is copying and counting on a

ready or discarded basket; therefore a lot of the learning comes from the apprentice's own

interest and ability. Sometimes, men learn by copying from a basket or a piece of basket

discarded in the village's garbage. Other ways of learning mentioned in that research were from

friends (20%), brother (10%), uncle (5%) and grandfather (5%). In 2007, a similar tendency is

maintained, with father and mother appearing as the main transmitters of weaving knowledge for

the women (30.77%) and men (28.89) interviewed in all villages. "Alone" is again the second









category of learning by copying and trial-error. Oakdale (1996:112), working with men in

Tuiarare village, registered that there is a tendency, mainly amongst younger generations and

young leaders, to state that nobody taught them how to do things in general, that they

"understood and were able to do things by themselves."

In 2007, a new category named "learning in the workshop" appears in third place for

learning designed baskets and textiles among the Kaiabi, with a relatively high score (20.04% for

women and 26.67% for men, see Figure 8-13) considering that it was mentioned in only two

villages. This category relates to the weaving workshops promoted by the Kaiabi Araa

community-based project (see Chapter 6) in Tuiarare and Kururuzinho villages from 2004 to

2006. In these workshops, organized by the villagers under the supervision of the coordinator

Aturi Kaiabi (Jowosipep), the learning process occurred in a very innovative and interesting way,

which does not fit in the typology presented by Cavalli-Sforza and his contributors. I am naming

it as a "many-to-many" mode of transmission, in which many teachers, elders or not,

independently of age, teach many students and also might learn from each other. Teachers and

students may come from different villages, increasing their sense of ethnicity and creating closer

ties between villages. This mode of learning can be also named "collaborative learning", since it

happens in a community of practice, whose members are consciously participating in a learning

activity with a shared sense of group identity (Bowser and Patton, 2008; Ellen, 2009).

It is intriguing to think that first of all, this model of collaborative learning was developed

by the Indians themselves, not copied from an existing western model. Second, in this type of

workshop there is more freedom to innovate and also to learn from anybody; it is even possible

for an older person to learn from a younger teacher who may be more experienced in one

technique or on certain basketry types or designs. Finally, it provides social learning









opportunities otherwise unavailable in the traditional system of knowledge transmission which

used to generally occur in the household, between parents and children, within kinship groups or

among unrelated kin but never within a bigger group of specialists and apprentices united for

sharing knowledge of a specific domain.

In a workshop carried out at Tuiarare village in 2006, an elder from another village

named Karauu made a basket for which only he still remembers the technique (yrfiokote 'em),

and which is in danger of disappearing. Its confection was observed by many workshop

participants, coming from various villages. Among the women, through the workshops, new

designs were copied from baskets, learned from the book and from other women, and new items

were learnt and started to be made by more women, such as bags and belts. Innovation in this

type of transmission may be even quicker than in the "one-to-many" mode, because knowledge

is exchanged within the community of practice.

Takako (2004) presents a model of generational cultural transmission adapted from Boyd

and Richerson (1985) named "the recursion model," in which there are mature and immature

populations. The immature individual learns within a determined knowledge domain, and comes

to constitute the mature population. When there is insufficient transmission/learning, individuals

from the mature population die and there is less ingress from the immature population into the

mature pool, causing knowledge erosion or loss.

In 2002, using the typology of cultural transmission from Cavalli-Sforza and Feldman

(1981), I noted that the direction of knowledge transmission among the Xingu Kaiabi is changing

from vertical (parent-child) to more oblique (between generations but outside of the parent-child

relationship) and horizontal (within the same generation, and often across different kin groups).

In a previous paper, we discussed the need to review and refine Cavalli-Sforza and Feldman's









typology of cultural transmission, taking into account qualitative changes that accompany

learning and transmission amidst social change (Athayde et al, 2009). We should also distinguish

between 'direct' modes of transmission- taking place through observation or learning from

others, and 'indirect' transmission- which occurs from such 'virtual' intermediate sources as

books or films. Moreover, we suggested that the category of 'horizontal' transmission would

need to be more carefully unpacked among the Kaiabi, given the intensity and diversity of

exchanges that occur between different kinds of non-kin (Athayde et al, 2009).

Another innovative mechanism used by the Kaiabi to learn different basketry designs is

copying from photographs, drawings and books. When I first arrived in Xingu Park in 1997, I

noted that some men in the villages had copies of an article published by Berta Ribeiro in 1986,

which contains drawings of basketry designs further developed from Georg Grunberg's previous

work (Grunberg, and Grunberg, 1967; Grunberg, 1994). They were using those drawings to learn

new graphic designs. At that time, I started to take photographs and document every basket in the

village that contained a different graphic design. Complementarily, I started to collect

photographs archived in national and international museums and libraries. As stated previously, I

have organized a book on basketry with the Kaiabi that includes these photos, and also texts,

stories and myths produced by them about basketry and natural resources used in their

production. This book has been used in the villages for learning new designs both by men and

women. Surprisingly, when I visited Rio dos Peixes in 2007, I found one of those books I

organized in the house of one of the elders. Somebody from Xingu Park took the book there and

gave it to Simdo, the most knowledgeable basket maker. He told me that he learned thirteen new

designs from the book. It's amazing to realize how fast knowledge and information can move

between Xingu and other Kaiabi lands. In Kururuzinho, people already have copies of the book,









taken there during the workshop of the Kaiabi Araa project. Thus, this book in another example

of using new media to transmit traditional knowledge. It would be inserted in the "one-to-many"

type of transmission. I quantified how many designs people have learned from the book, and will

present this in the next section.

It is worth remembering that apprenticeship of a given technique or design might also

happen through intercultural contact and interaction. For instance, Ribeiro (1980) discusses the

exchange of basketry graphic motifs, techniques and technology between the Kaiabi and the

Tapirape, Bakairi, and Paresi in the Tapaj6s, before their relocation to Xingu. These exchanges

may have increased the number of designs and their intricateness. Some Kaiabi men, for

example, affirm that the ta 'agap design, considered one of the core designs in Kaiabi basketry,

was acquired from the Apiaka (Athayde 2003; Grunberg and Grunberg 1967; Ribeiro 1987b).

We can conclude that the Kaiabi today are using seven mechanisms for knowledge

transmission, synthesized in the model presented in Figure 8-14. I suggest that this model best

captures the changes and innovations that indigenous peoples have achieved through increased

exposure to, integration with and assimilation of non-indigenous knowledge, institutions and

politics. The seven mechanisms may be further classified according to their occurrence in or

outside kinship groups and also if they occur within the same cultural group (intracultural) or

between ethnic groups (intercultural, including non-indigenous).

Suzanne Oakdale (1996:110) presents an interpretation of intergenerational change

among the Kaiabi, in which, according to the author, alternate views on capability and

knowledge interact in a single cosmological process. She named them progressive and

degenerative trajectories. In the progressive trajectory, "each generation knows more and

understands the world better than the last." People may be using this discourse when referring to









the fact that nobody taught them basketry or textiles, that they have learnt by their own effort and

ability. In the contrasting degenerative trajectory, "human capability and understanding are

diminishing with each passing generation." In other words, there would be an ideal primeval

time, where "pure" Kaiabi knowledge existed, which cannot be rescued anymore. These two

views may be also interpreted as the traditional and the new in any ethnic group where there are

old and new values interacting in the continuous re-production of a human society. For instance,

the fact that the Kaiabi are using and adapting new institutions and mechanisms to transmit

traditional (with innovations) knowledge on basketry and textiles can illustrate this apparent

antagonism. The ability to adapt, assimilate and create conditions for intertwining traditional

knowledge in contemporary contexts is an asset that provides resilience and endurance of

indigenous knowledge systems.

Distribution of knowledge on graphic designs

The Kaiabi have a repertoire of basketry designs traditionally depicted in twill-plaited

baskets, the special backpack basket, club handles and gourd engravings. Some Kaiabi tattoos

using basketry designs have been made since ancient times (Schmidt, 1942; Ribeiro, 1987b;

Grinberg, 1994). They started to innovate and use the basketry motifs in other objects after the

transfer to Xingu Park. Observing Yudj a textiles, and during workshops promoted in Xingu Park

by FUNAI in the 1970s, women started to weave basketry designs in hammocks, belts straps for

carrying babies and later on in bags for sale (Ribeiro, 1984/85). In this case, many times the

husband would start to teach his wife to count and weave the design, and then she would learn

other designs with her female peers or by herself. Some women have also used the book

(photograph catalogue of basketry designs from Kaiabi Araa project) to learn new designs. This

section explores men's knowledge on designs in more depth than women's, taking into account

that the origin of designs comes from basketry; that there are a greater number and complexity of









designs; and that only recently did women develop knowledge on designs. Furthermore, the

study carried out in 2002 included only men.

The designs used in basketry have become powerful symbols of ethnic identity, and are

used by the Kaiabi as a way of reinforcing their identity and indigenousness in relation both to

other indigenous groups and to the non-indigenous society at large (Athayde et al., 2009). During

festivals and political events, for example, men now paint their bodies with basketry designs, and

political leaders carry the large wooden clubs with woven handles, which are also decorated with

these designs. Chemela (2008) suggests that the symbols contained in indigenous basketry (in

the case of the Desana) may be compared to the heraldic symbols that identified sociopolitical

entities in Europe.

Men started to copy the technique of painting wooden carved benches from the Yudj a

people. Painted benches started to be made around 1998 by men in some villages, mostly for

sale. The innovation of producing painted benches was developed in the context of the Kumana

project, the first community-based project run by ATIX (see Chapter 6).

In 2002, I analyzed knowledge on 27 basketry and 6 panaki (backpack basket) designs

(33 total) among 25 men in Tuiarare village and 20 men in Capivara, totaling 45 participants.

Amongst them, 44% knew how to weave baskets, either simple or designed baskets. In that

research, I used consensus analysis to evaluate the distribution and level of knowledge on

designs among the participants, as well as to compare people in the two villages, according to the

designs they knew how to weave. At that time, I called the ability to weave the design "use of the

design," and I also measured the ability to name the designs (Athayde, 2003). I classified men in

four groups of expertise based on their knowledge of designs. I found out that there was a

significant difference between the two villages, in which Capivara had a greater proportion of









weavers in comparison to Tuiarare, but most people knew simpler designs, with exception of

Kupeap, an elder expert in basketry weaving. In Tuiarare, there were fewer weavers but the

diversity of designs used was greater.

For this research, I expanded the number of designs in the catalogue to 29 basketry and 7

panakfi designs totaling 36 designs for the men and 9 textile designs for the women. A total of 43

men and 46 women in all villages who testified they can weave at least one design participated in

this analysis. In Appendixes B and C, I present tables with the codes, names (with respective

ratings) and translation of meanings for the different designs. In Appendix D and E, I include a

translated copy of the photographic catalogue of basketry and textile designs used in this

research with their respective codes, which is part of the Kaiabi Araa educational book. I

interviewed men and women on their ability to use or weave the designs and also on their

knowledge of the name(s) given to the design

I want to clarify that my analysis on knowledge of basketry and textile designs has as a

basic assumption the fact that knowledge on designs is specialized knowledge, and thus it should

not be shared by all individuals of the same group (in this case, Kaiabi). As in any knowledge

domain, it is obvious that there will be intracultural variation (Boster, 1987; Weller, 1987).

Intracultural variation or informant variation has been studied for a multitude of cultural domains

(Boster, 1987). What is different about knowledge on designs in contrast to some other cultural

domains, such as TEK for instance, is that you cannot expect that a young 18 year old man will

have domain of many basketry designs, whereas you expect that he holds knowledge on

preparing a field for planting or on some important plant resources, such as those used for house

construction.









Romney, Weller and Batchelder (1986:313) suggested that we might define culture as

shared knowledge, since culture would consist of "what people have to learn as distinct from

their biological heritage." The level of agreement on any given knowledge domain, between a

group of the same cultural background, would ultimately indicate level of knowledge on that

domain. This hypothesis was tested by Boster (1986) studying the distribution of knowledge

related to manioc classification among Aguaruna men and women. Some important conclusions

from that study were that women know more about manioc than men, and that women in the

same kin and residential groups were more similar in knowledge in comparison to nonrelated

women. The author also concluded that the more a participant agreed with the others, the more

knowledge or competence he or she had about manioc.

Adopting the model of culture as consensus, I wish to analyze here the extent to which

Kaiabi men and women share knowledge on names given to designs, teasing out relationships

amongst the participants in terms of how much they agree on the different names that a design

might have. In 2002, I found out that there was a tendency for generalization of names and

perpetuation of simpler basketry designs among the youth, while fewer elders retained the

knowledge of more complex designs. I registered that a design can have many names, and that

the more complex designs are the ones for which that there is less agreement on names. At that

time, only one elder (Kupeap, from Capivara) was able to name all the thirty designs registered

in a fairly complete photographic catalogue.

According to Weller (1987), in a given cultural system, an individual's knowledge may

be considered a sample from the larger pool of cultural knowledge. Agreement among

individuals can be interpreted as a function of the extent to which they share similar information.

The author further developed the cultural consensus model and methods, presenting two models









for analyzing data on knowledge domains: one related to questions that produce dichotomous

choices or "yes-no" responses. The second, adopted in this research, conceptualizes cultural

knowledge as a pool of information or elements. "Shared knowledge is represented by the

proportion of those elements that are shared or held in common between an individual and the

pool" (Weller, op.cit: 180). In other words, shared knowledge on names for basketry designs can

be inferred by the proportion of names for any given design that are shared or held in common

between a man or a woman and the group. In this case, there is no correct answer; every

individual's knowledge will be estimated by how much his or her knowledge of designs names

fits into the groups' overall knowledge.

For this analysis, as described in Chapter 2, I took only men and women who knew

designs, and from those I took those who knew to weave at least 10% or more of the 36 designs.

I did the same procedure for the women. Thus, 29 men and 14 basketry designs and 34 women

with three textile designs were included in the consensus analysis.

Men's knowledge on using and naming designs. Knowledge on basketry designs is

unevenly distributed among men. Some people learn just the basic designs and techniques,

enough to provide for their family, while others get specialized in the diversity of designs. I

remember what chief Aturi (Jowosipep) told me once, that knowledge on designs is like studying

in western society: some people only finish elementary school while others complete a PhD.

During the last workshop of the Kaiabi Araa project in Tuiarare village, another friend from

Muitara village said that a master in basketry weaving has to be appreciated for his skill, and that

the domain of different designs is not for everybody. In other words, mastery in the different and

more complicated designs should and cannot be achieved by every Kaiabi men, and maybe this









is also true for the women. Some people are more knowledgeable in one cultural domain, such as

crop management, others are good hunters, others are great basketry weavers.

Normally, a man would learn to weave simpler designs, gradually mastering more

complex designs according to his interest. Often, the first design learned is the design coded

PE26 -"i'yp" (meaning i, mine, yp, tree; my tree or my trunk, a path to follow) applied to non-

painted baskets. This design was the only one mastered by all men interviewed in Capivara and

Tuiarare villages in 2002. In the present research, only four men did not weave the i 'yp design.

Three of them are young men (15 to 26 years old) who learned to weave during workshops or

from the book. The other is Eroit from Kururzinho village, who started to weave painted baskets

using the "awasiayj" (maize grain) design. Some people start to learn designs using the simpler

designs awasiayj (PE17 and PE18), awarapypot (PE23, meaning fox footprint) or another design

mostly used in non-painted baskets named ipirien orjarukang (PE24, path or rib bones). Another

very popular design is the kururu 'i or little frog (PE16). Some men told me that if they learn

some difficult designs first they will never master the simpler designs. For instance, if they learn

the design PE10 (named jowiterian, inimo eta oryok) when they are still young, before learning

other designs, they won't be able to learn new designs anymore. According to them, only older

men should learn and weave this design.

The distribution of knowledge on the use (ability to weave) of different designs is shown

in Figure 8-15, contrasting all men interviewed with the weavers. In 2007, there were 14 design

weavers in Capivara, 15 in Tuiarare, 9 in Kururuzinho and 5 in Rio dos Peixes. In Kururuzinho,

there were four design weavers in 2004, previously to the Kaiabi Araa workshops; therefore, five

men learned during the workshops. From the current distribution (2007), it is possible to confirm

that the simpler designs, which people normally learn first, are the ones that more people master.









The design that most people know is the PE 26 or i 'yp (35% of total men and 85% of weavers),

followed by the PE 18, awasiayj and PE 23, awarapypot (both known by 21% of men and 50%

of weavers). Another type of awasiayj (PE9) and the kururu 'i (PE16) are known by 14.5% of

men and 35% of weavers. The designs that are most in danger of disappearing are the same

registered in 2002. All panakii designs are endangered, mastered by only one or two people.

Some men are able to weave panakfi designs in baskets (see PE 29, kwasiaruu, for an example).

Among these, the design PA4 or kwasiarapat (variation of basketry design meaning opened

arms), PA5 orpanakifkupe (applied on the back part of panakfi basket) and PA7 or kwasiaruu

(big image) are the ones that fewer people know how to weave. This is expected, taking into

account that only one man in Xingu still retains the technique of weaving this basket. For the

basketry designs, the ones that fewer people know how to weave are mainly derived from the

general form named ta 'agap (which means image or person), from PE1 to PE15. Among these,

PE4 (tangajopep, image of person divided in two or twins) is the most endangered, mastered by

only 8.70% of basketry weavers and 3.65% of all.

It's interesting to note that the majority of designs which are threatened of being

forgotten are from baskets that are placed in museum collections, many of them collected by

Georg Grunberg in 1966. If these baskets had not been collected, stored and preserved in

museums, we would probably never know about the diversity of graphic designs mastered by the

Kaiabi. The work of surveying, documenting and returning these collections to the Kaiabi is

known as visual repatriation. There are many ethnographic collections of indigenous peoples

placed in museums around the world that are unknown by the group to which they belong.

Comparing the knowledge on use of designs from the exact same men interviewed in

2002 and in 2007 in Capivara and Tuiarare villages, it is possible to register an increase of









knowledge on designs (see Figure 8-16), despite some inconsistencies derived from participants

who stated that they knew the design in 2002 and then reported that they didn't know to weave

the design in 2007. For the 33 designs considered in 2002, 25 men had knowledge increased

from 2002 to 2007 (76% of the designs); 8 remained the same (24%) and 7 had knowledge

decrease (21%). Thus, there was a significant increase in knowledge on designs after the

development of Kaiabi Araa project. Two men from Capivara and two from Tuiarare learned

from three to ten designs in a five year period. Towaju'i (28 years old) learned three designs; he

was the only man from Capivara village who participated in workshops of Kaiabi Araa project.

Tare'a (32 years old) was already an interested basketry weaver in 2002, and learned four new

designs in five years, two of those from the book. Men from Tuiarare are the ones who learned

the most designs in the period. For instance, Myau'i (55 years old) is an experienced basketry

weaver who participated in all workshops of Kaiabi Araa project as a teacher. He learned nine

new designs in five years, four of those from the book. Tare'i (29 years old) was the winner in

terms of new designs learned, with 10 new designs, seven of those copied from the book. Thus,

while in Tuiarare there was greater increase of knowledge, in Capivara there was a more discrete

increase. A big difference in Capivara was the return to the village of Tewit Kaiabi, who was

living in Rio dos Peixes for around five years and decided to return to Xingu to stay closer to his

family. He is the weaver who knows the most designs in this research, reporting being able to

weave 31 of the 36 designs presented to him in the photographic catalogue. No one knew all of

the 36 designs. In 2002, Kupeap was the most knowledgeable weaver in the village, knowing 24

designs. It is worth remembering that knowledge on designs does not reflect knowledge on

basketry in a broader sense.









Men who may be considered specialists, being able to weave ten or more designs, are

older in Capivara than in other villages. Also, it is interesting to note that some of the specialists

in designed baskets also know how to weave more types of baskets. In Capivara, there are three

experts mentioned above, namely Tewit (78 years old, 31 designs); Kupeap (77 years old, 24

designs) and Tare'a (32 years old, 11 designs). In Tuiarare, experts are younger and include

Tare'i (29 years old, 29 designs); Myau'i (55 years old, 24 designs); Aturi (46 years old, 12

designs); and Makupa (47 years old, 10 designs). In Rio dos Peixes, there are currently three

experts, namely Simdo (65 years of age, 22 designs); Raimundo (41 years old, 16 designs); and

Moacir (56 years old, 12 designs). In Kururuzinho, only Corone (74 years old, 11 designs)

knows more than 10 designs. To better visualize the distribution of knowledge on designs among

the villages, I divided the number of designs in six classes: 1 to 5 designs; 6-10; 11-15; 16-20;

21-25 and more than 25 designs (see Figure 8-17). The distribution of knowledge on designs

shows that Capivara and Tuiarare are close on the percentage of weavers that fit in each class of

knowledge of designs, with men in Tuiarare knowing slightly more designs than men in

Capivara. Also, it seems that whereas in Teles Pires (Kururuzinho) knowledge is still starting,

with more men knowing fewer designs, in Rio dos Peixes men know more designs, but there is

no new knowledge being generated. A few men there that already knew how to weave are

keeping their knowledge and learning some new designs from the book, but no new apprentices

are being generated in the community.

Both young and older men have had their knowledge increased in the five year period

between the two researches, and the majority of men who have increased their knowledge used

the book to learn one or more designs. Thus, learning from the book is another mechanism that

has increased knowledge of designs among the Kaiabi. The increase of available photocopies of









the Kaiabi Araa book containing photos and names of the various designs has promoted an

increase in the knowledge of different designs among young and older men alike. Simdo learned

13 designs from the book that was brought from Xingu to Rio dos Peixes. Sirejup, a young man

from Capivara village (21 years of age), told me that he actually learned to weave from the

beginning using the book, weaving two designs nowadays. A total of 16 people, representing

7.14% of all the men and women interviewed, mentioned that they have learned some designs

from the book. Among them, three are women and thirteen are men. They mentioned twenty

basketry designs learned from the book, four panakii designs and five textile designs. Most men

who mentioned learning from the book are from Capivara (5 men), Rio dos Peixes (4 men),

Kururuzinho (2) and Tuiarare (2). Therefore, despite not participating in the workshops of Kaiabi

Araa project, men in Capivara and Rio dos Peixes villages have used the book to learn new

designs. Men from other villages in Xingu have also learned from the book, which was

distributed to all villages. The designs that were copied from the book are mainly the most

difficult ones, the "taa'gap" based designs. The design PE5, named Ta 'agafu 'a or Ta 'agafu 'a

jakunaap (person in shape of cross), is the one more people mentioned to have learned from the

book (4 people). It was taken from the collection of the Archeology and Ethnology Museum of

Sdo Paulo University (MAE/USP). The book is an instrument for knowledge transmission on

basketry designs that has contributed to the revitalization of lost knowledge on more complex

designs.

The different basketry designs can have a great variation of possible names. Araa is the

general name for "design." For the 36 designs, there are a total of 139 names mentioned, with a

medium number of 3.86 names for each design. All designs have two or more names, the

majority of them between two and four (see Appendix II). The design with the most names is









PE10 with nine names, named as inimoeta (many threads) by 8.16% of men or as yok (larvae) by

4.08% and so on. Other designs with great variation belong to the general shape "ta'agap"

meaning mythical figure, whose variations encompass the most complex designs. Some ta 'agap

types with a greater number of names include PE14, with seven names; PE 5 also with seven

names; and PE6 and PE2 with two names each. The tendency observed in 2002, of having less

agreement and more possible names for the ta 'agap type of designs, has been also registered in

2007. When a man does not know the detailed name for a ta 'agap design (for instance, ta 'agap

tayt mythical figure with children), they only mention "ta'agap". Sometimes, the ta 'agap

design is also referred to by the general name for design, araa. This causes simplification and

loss of diversity of names, and thus loss of knowledge. Some designs, such as PA4 used in the

panakfi basket, have all the possible names mentioned in the same percentage (2.04%). In such a

case, it's hard and even impossible to determine which name is the correct one. In contrast, there

are some designs which have a specific name mentioned by a lot of people. The designs

considered easier to weave are also the ones with more agreement about the names. Among

them, the most consistent is PE 26 or i yp, whose same name is used by 69.39% of respondents;

followed by variations of awasiayj or corn seed -PE9, PE17 and PE 18- whose same name was

mentioned by 46.94%, 44.90% and 38.78% respectively. Other designs with a higher level of

agreement between names are PE 16 or kururu 'i (little frog, also popular) with 40.82%; PE 25 or

jowosiape (tortoise shell) with 32.65%; PE 23 or awarapypot (fox footprint) with 30.61%; and

PE 21 or moiafu 'a (rolled up snake) with 20.41%.

Variation in naming may be due to place of birth and residence, kinship and/or exposure

to different weavers. Elders are mentioned as the ones who know the names, and when

somebody is in doubt about a name, they suggest "va perguntar ao velho go talk to the old









man." In the Kaiabi Araa book, we included all the names mentioned by different weavers for

each design, since we did not want to favor one specific name, as if there should be only one

"correct" name. Silva (2009) registered the same pattern of variation in names for peanut

varieties among the Kaiabi in Xingu Park. He obtained a list of 133 names given to 17 traditional

varieties of peanuts by 286 informants, including men and women. There was more agreement

on names for more common and widespread varieties, as opposed to newer or rare varieties,

comparable to the concordance on simpler basketry designs and disagreement on the ta'agap

group of designs. He also registered that older women were more knowledgeable in naming

peanut varieties when compared to men.

The consensus analysis for 14 designs among 29 men revealed that, as expected, experts

in weaving designs were also knowledgeable in naming them, as shown in Figure 8.18. In

general, men that were expert weavers also appeared as being more in agreement with each other

in the consensus analysis on naming designs. The large eigenratio (9.906) indicates a good fit to

the consensus model.

Tewit from Capivara village was the first in the competence score list with 0.886.

Following him, five men from Tuiarare show higher competence scores in the consensus analysis

(see the colors for different villages represented in Figure 8-18). They are experts who have also

scored high in the number of designs that they can weave. It is possible to infer that these men

who show greater agreement with Tewit concerning names for the designs have learned some

names from the book. The nine higher competence scores are shared between Tuiarare and

Capivara villages. Tewit is the only older man, followed by middle-aged and younger men, such

as Tare'a (Capivara, 32 years old) and Tare'i (Tuiarare, 29 years old). Kupeap is another elder

from Capivara who was the first in the competence score in 2002. However, this time he had a









lower competence score due to the different names he gave to the designs, which were not in

high agreement with Tewit and the others. Older men from Capivara and Tuiarare who did not

score high on the competence test were the ones who weaved few designs and gave very general

names for the ta'agap designs, such as Chico (lowest score) from Tuiarare, who responded

"araa" for the designs for which he did not know the name.

Valdir, from Kururuzinho village, scored 10th in the competence list, which indicates that

he has a fairly good knowledge of basketry designs; even greater than Corone, who is older and

knows how to weave more designs. In general, men from Teles Pires scored higher and are in

more agreement with the Xingu experts than men from Rio dos Peixes. This might reflect the

impacts of Kaiabi Araa project on knowledge of uses and names of basketry designs in Teles

Pires, and, again, the use of the book to learn, especially among the youth. Among men from

Teles Pires participating in the consensus analysis there are also youth, such as Elenildo (18

years of age) and Arlindo (29 years old). This does not happen in Rio dos Peixes, where the few

expert weavers are mostly middle-aged or older men. Men from Rio dos Peixes who scored

relatively high in the competence list are Moacir and Simao. In spite of the fact that Simao

named 26 designs and Moacir named only thirteen, the names given by Moacir were in more

concordance to the majority than the names given by Simao. Moacir lived in Xingu for a long

time before coming to Rio dos Peixes, and also learned to weave (and name) baskets there,

whereas Simao always lived in Rio dos Peixes. Other men from Rio dos Peixes included in the

consensus analysis were Tafut (elder and shaman); Canisio (lived in Xingu); and Raimundo (son

of an expert basket-maker).

Figures 8-19 and 8-20 show a spatial representation of the similarity between men from

different villages regarding naming of basketry designs, from the non-metric multidimensional









scaling analysis of similarity (MDS). It is possible to verify in Figure 8-21 for Capivara (red) and

Tuiarare villages (blue dots) that the experts are clustered around the point "0" of the axes,

whereas men who don't know much or do not agree on the names informed by the majority are

sparsely placed around it. We can also visualize that expertise exists in both Capivara and

Tuiarare villages; thus the difference between them is not great. In Figure 8-22, representing Rio

dos Peixes (orange) and Kururuzinho (green) villages, there is more disagreement among the

weavers, with a little cluster formed around Elenildo, Valdir (Elenildo's father) and Jodo from

Kururuzinho. Coroner, in spite of knowing to weave many designs and having named 15 of them,

is probably confused about their names, given his disaggregated position in the graph. Canisio

and Tafut know few designs, and probably did not know the names for many of them.

These results are interesting because they reveal that in spite of the fact that older men are

more knowledgeable in many domains, including proficiency in the native language, names of

designs are also mastered by younger and middle-aged men. There is also more agreement and

coherence between Xingu weavers in comparison to the other groups. This is true for both

Capivara and Tuiarare villages, independently of the fact that Kaiabi Araa happened in Tuiarare

and not in Capivara. Actually, it is very positive that there are men in Capivara keeping and

transmitting weaving knowledge which increases the resilience of basketry knowledge instead

of having it concentrated in only one village. It is also very positive that knowledge in

Kururuzinho has increased after the Kaiabi Araa project, with younger men learning to weave

and name the designs. This also increases cultural resilience, since knowledge is being

transmitted between generations. On the other hand, the lack of younger men learning and

knowing how to weave and name baskets in Rio dos Peixes is a sign of knowledge erosion, thus

loss of resilience. I accept hypothesis H5 (see introduction) that the Kaiabi Araa project is









responsible for an increased number of basketry weavers among the younger generations and

also for an increased knowledge on the designs, enhanced by the book.

Women's knowledge on using and naming designs. Regarding women's knowledge on

textile designs, there were 18 design weavers in Capivara and Tuiarare, 8 in Kururuzinho and

only 2 in Rio dos Peixes. It seems that when women first started to copy the designs from

baskets into textiles, they learned the easy design named awasiayj (corn or maize seed or grain,

TEl, PE17), which is known by 89.6% of weavers and 38.6% of all women interviewed (Figure

8-21). This design also has some variations, for the "grain" might be small or large, inside or

outside a frame (see Appendix IV). Jarukang or ipirien (TE2, PE24) is the second most known

design, with 63% of weavers and 27% of interviewed women. This design is preferably used in

hammocks, while the awasiayj is applied to hammocks, straps for carrying children, bags, and

other objects. I personally witnessed the learning of a new basketry design by the women in

Tuiarare in 2004. I saw a bag woven with the kururu 'i or little frog pattern and really wanted to

purchase it; I wanted to know who was the artisan who had made it. Morerfi (nicknamed More)

is the weaver from Tuiarare who copied this design from the book, and then various other

women were avid to learn the new design, which started to spread to other women and other

villages in the workshops of the Kaiabi Araa project.

I was able to observe that women, especially in the same kin group, exchange knowledge

on textiles more easily than men, for in the household there is a lot of time spent talking,

chatting, making jewelry and other handicrafts and also weaving textiles. Furthermore, the strap

for carrying babies is an extremely useful and important item that almost every woman who is or

is going to become pregnant has. I carried my son on those straps for many months, and they are

really convenient, besides the esthetic and symbolic value. Being of a later age does not









necessarily mean increased knowledge on textiles designs, for this is a technique developed more

or less recently.

From the consensus analysis results, six women from Tuaiarare are in 100% agreement

on the names given to the three textile designs, presenting the same competence score (0.953;

Figure 8-22). They are from Tuiarare village, but not necessarily from the same kinship group. If

we had done the same interviews in 2002, before the workshops of the Kaiabi Araa project

happened, maybe these scores would be different, with more mixing between Capivara and

Tuiarare villages, since the project encouraged learning of new designs by Tuiarare women.

Also, women from Kururuzinho village would not appear in the competence scores list, since

they also learned in 2004 during the Kaiabi Araa workshop with participants and textile teachers

from Xingu. Three women from Kururuzinho are represented in the competence list, Diana,

Cunhaete and Josiane. All of them learned during the workshops of Kaiabi Araa project and are

nowadays producing woven bags and belts for sale. Only two women from Rio dos Peixes

appear in the competence score list: Morea'i and Katuryp. Both came from Xingu to Rio dos

Peixes, thus bringing knowledge on designs with them. Similarly to what is happening for

basketry knowledge, the lack of younger women in Rio dos Peixes knowing how to use and

name the designs is also a sign of knowledge erosion or lack of transmission, thus leading to a

possible permanent loss of weaving knowledge in that village.

These results let us conclude that knowledge on designs is not necessarily linked to

kinship, but to a person's interest and to the amount and availability of objects and people who

make them in the village where you live. Since the main way of learning is by observation, the

more baskets and textile items available in the village for copy, the more stimulus and

opportunities to copy and learn a person will have. Kinship as a mechanism of transmission









seems to be more significant among women than for men and more important in other cultural

domains in contrast to specialized weaving knowledge. I argue that the availability of designs in

the village as well as opportunities to learn will have a greater effect on the pool of designs

known than will kinship relations. This reflects what is going on now, but does not mean that it

has been like this in the past. It is worth remembering that the main transmission mechanism for

starting to learn basketry and textile weaving is still through kinship, mainly vertically by mother

and father. However, after learning the basics, the development of knowledge on designs may

happen through different mechanisms, inside or outside the kinship system, including innovative

ways to learn through institutions and instruments coming from the non-indigenous society such

as projects and books.

Weaving Cosmology, Shamanism and Symbol2

Shamanism plays a crucial role in the production and reproduction of Kaiabi social and

cultural organization, as happens with many other indigenous groups in the Amazon (Seeger,

1981; Travassos, 1984; Viveiros de Castro, 1992; Wright, 1998). Nowadays, there are fewer

shamans among the Kaiabi than in the past, and according to informants, they are not as

powerful as they used to be. There are few men and women shamans in Xingu, only one shaman

in Rio dos Peixes, and no shaman in Teles Pires (Kururuzinho). Therefore, when people from

Teles Pires or Rio dos Peixes need to consult or be treated by a shaman, they have to go to

Xingu. Sometimes, they can consult with shamans from other ethnic groups.

Learning and creating in Kaiabi society is related to shamanism. Designed baskets and

now many other objects adorned with distinctive Kaiabi graphic designs carry the memory of





2 Parts of this section were adapted from Athayde et al (2009).









ancestors, mythical times and mythical "others." A Kaiabi ancestral hero named Tuiarare3 was a

great shaman and basketry weaver. According to a myth, he learned to weave painted baskets by

stealing a piece of snake skin from the house of a dangerous snake when visiting the Xingu River

in ancestral times. He took that piece home, copied the design and learned by himself, then

taught the Kaiabi people (Athayde, 2006). Many knowledgeable men interviewed mentioned that

the Kaiabi people first learned to weave baskets from Tuiarare. Therefore, Tuiarare first learned

indirectly from the snake skin, which has the design imprinted. Snakes are powerful supernatural

entities among many Amazonian groups, mainly related to acts of discovery and creation (Van

Velthem, 1998; 2001). Curiously enough, the mechanism that Tuiarare used to learn is

reproduced in Kaiabi everyday life. Every time a man gets an old basket or a piece of basket

from the trash to observe and copy, learning alone, he is reproducing the myth. Guss (1989:93)

stated that "myths of origin serve as the perfect paradigm of transformation, symbolically

depicting the daily operation of culture." Even learning from the book is linked to an ancestral

mechanism of learning, by copy. The photographs of the book can be thought of as new snake

skins from which people can learn. Besides highlighting the relationship between travel, mobility

and the acquisition of knowledge and power, the story of Tuiarare underscore the relationship

between basketry and shamanism.

The relationship between basketry and shamanism has been shown by ethnographers

such as David Guss (1989) and Johannes Wilbert (1975), the former among the Yekuana and the

latter among the Warao, both Venezuelan groups. In the Yekuana myth of origin of painted

3 The great ancestral hero, creator of Kaiabi people. There is a myth in which Tuiarare went on an expedition to
Xingu river (Wywa'y) to collect a kind of bamboo (cana-brava, Gynerium sagitatum) in order to make arrows. He
walked around a lot, and he discovered many natural resources important for the Kaiabi people during this travel.
On the way back to his home, he arrived in a village; it was the "snake" village. After discussing with the snake for
one entire night, he spoke the name of a great hawk known as a snake eater, and then he left the house carrying a
piece of painted basket, from which he learned how to weave baskets and transmitted it to the Kaiabi people
(Athayde, 2006).









baskets (waja), the twill-plaited baskets are discovered in a shaman's (the master of the spider

monkeys) big woven pouch and the symbols woven into them assure a means to the supernatural.

By weaving the supernatural world into designs applied to material objects, the Yekuana

symbolically gain control over supernatural, mortal forces, "without whose integration society

cannot be maintained" (Guss, 1989:105). Wilbert (1975) describes the process by which a master

basket maker is transformed into a shaman through the practice of his art. In this case, the

craftsman acquired shamanic powers strictly for personal advancement rather than for practicing

shamanism. Ribeiro (1986), in her article on the symbolism of Kaiabi basketry designs, shows

how the set of graphic motifs represented in the painted baskets works as an iconography and

thus as a symbolic language which helps to construct and to reinforce the group's identity. She

affirmed that some motifs such as the "ta 'agap" and the "kururu" might represent supernatural

beings present in Kaiabi mythology and cosmology, sometimes with human and/or animal

attributes at the same time. Thus, in representing supernatural beings in the baskets, the Kaiabi

are, at the same time, gaining control over supernatural forces (such as showing supremacy over

the snake) while transmitting and reaffirming their mythology and ancestral knowledge in the

present.

For the Kaiabi, cotton was "born" from a woman through the work of a shaman.

According to a myth, an ancestral woman gave birth to cotton seeds, which came out of her

vagina while a shaman was praying. This myth, with variations, was told to me by various

people, including the elder Kupeap from Capivara. Catarina from Rio dos Peixes remembered

that her mom told her about the story of the origin of cotton. A woman gave birth to a child, and

also to the cotton seeds at the same time. They took the seeds, washed and put them to dry in the

sun, and then they planted them. She told me that the name of a white variety of cotton called









"kwaramyneju" (meaning cotton of the sun), comes from this myth. Another myth tells the story

of Kupeirup, the ancestral female shaman who originated all Kaiabi crop plants, including

cotton. Thus, cotton origins were already linked to female power and agency, which are

maintained today by female work in planting, harvesting, keeping and weaving cotton into

textiles.

The incorporation of the 'other'- often an enemy- either literally, as in the consumption of

flesh, or metaphorically, as in the appropriation of particular skills, is a theme widely shared

among Tupi-Guarani speakers in Amazonia (Van Velthen 2001; Viveiros de Castro 1992).

Kaiabi mechanisms of learning and transmitting weaving knowledge on techniques, designs and

meanings illustrate very well the capacity of appropriating of and learning from "others,"

transforming others' knowledge, skills and even skins into their own. In the myth told above, in

ancestral times, the shaman Tuiarare first learned to weave baskets by copying the design from

the snake's skin, a mythical supernatural enemy. Then, he taught the Kaiabi people how to

weave designed baskets. In other myths, the Kaiabi got designs from supernatural beings such as

frogs and forest spirits (Ribeiro, 1986). In the myth of creation of the Kaiabi, the uruyp (arumd),

fiber used in basketry, had supernatural powers, hiding a larvae (cor6) who at night transformed

itself in Tuiarare's wife. There is a design, named "Cor6", (PE27, yok, yogajurat) which depicts

this larvae. More recently, the Kaiabi got some designs, especially the "ta 'agap" types, from the

Apiaka, another Tupi-Guarani group and fierce enemy of the Kaiabi in the past. The Kaiabi

women knew to weave simple cotton hammocks before they went to Xingu. At Xingu Park, they

learned to weave designed hammocks with the Yudja women in the 1970s. Then, Kaiabi men

taught the women how to weave the designs of baskets in the hammocks. Kaiabi men also

"borrowed" the technique of painting wooden carved benches from Yudj a men, starting to depict









basketry symbols on the benches. In 2004, a group of men and women from Xingu went to the

Kayabi Land in Para to teach their relatives to weave baskets and hammocks. Lately, the Kaiabi

have used a book with photographs of designs to keep their knowledge on designs. It is thus in

the mythical and social relations with "others" that knowledge, symbol and identity are created,

taught, transformed, and transmitted within Kaiabi society.

Shamans' dreams are intimately linked to creation, discovery and symbol development

such as creation of new designs or painting patterns and naming beings. People mentioned that

the shaman learned many designs by dreaming and also that names for the designs were given to

the shaman by spirits during dreams. Kupeap told me that the design named "jowosiape" (PE 25,

tortoise shell) came to the Kaiabi from a dream that some person (probably a shaman, he did not

know) had about the body of a spirit from the water (karauat), with a painting similar to a turtle

shell. They then began to weave this pattern and somebody named it. This is also very true for

the discovery and naming of new peanut varieties among the Kaiabi. According to testimony

given to Geraldo Silva by the shaman Tuiarajup Kaiabi from Kwaruja village4, he learned about

new peanut (Arachis hypogea) varieties during his dreams, from the female ancestral hero

Kureirup, who according to a myth, gave origin to the Kaiabi crop plants by being burnt in a

farming field (Silva, 2002 b; 2009). Tuiarajup mentioned that Kupeirup's spirit also taught him

the correct names for the peanut varieties that he was developing in his village. He said that, in

general, the names for crop varieties are learned from the elders, who learned from the ancestors,

who learned directly with the spirits owner of plants or animals (Silva, 2009).

For the Kaiabi, as for other Amazonian peoples, shamanism still seems to play a crucial

role in the generation and maintenance of biological and cultural diversity (Salick et al., 1997).

4 Tuiarajup is shaman and chief of Kwaruja village, and coordinated the community-based project Munuwi for the
revitalization of peanut varieties among the Kaiabi (see chapter 6 and also Silva, 2009).









The social valuation placed by some Kaiabi on diversity and complexity is not restricted to

baskets, and extends to other domains, notably crops. Just as basket-makers derive considerable

prestige from their mastery over a diversity of designs, some Kaiabi men and women have

accumulated a surprising diversity of crops. As with basketry, the ability to generate new forms -

whether weaving designs or crop varieties is linked to shamanistic power and agency (Athayde,

2006; Silva, 2009). I concur with other authors that the ability to generate and keep diversity,

both cultural and biological, is an asset that confers capacity to adapt and resist to shocks.

Therefore, we can think of diversity (biological, cultural, linguistic) as a pool of genetic and

cultural information that increases the resilience of a system, conferring greater capacity to resist

or recover from crisis (Maffi, 2007; Pretty et al, 2009).

Shamans, baskets, cotton and textiles

The Kaiabi believe that baskets and textiles have their spiritual master or owner, and that

they are living beings, rather than ordinary objects. They also may carry characteristics of their

owner. Once, somebody told me that I should not let my baby sleep in other people's hammocks,

because babies are very susceptible and could get inflicted with the hammock's owner's energies

and even illnesses. When a person dies, she or he is wrapped in her or his hammock and is buried

with all their personal belongings. Kupeap told me that sometimes, when a man dies, their

relatives might place a painted basket in his tomb, to prevent their relatives from dreaming of

him. For this purpose, they would only use big painted baskets with the ta 'agap motif.

Baskets and textiles are used in shamanic practices, or might be given as a gift or

payment for shaman's services. From the interviews, 24 people (14 women and 10 men, 10.7%

of the total) reported that they have used baskets or textiles to pay for shaman services.

According to Jurufet (Kupeap's wife), elder from the Capivara village, when the shaman asks

for a hammock, something bad might happen if the person does not give it to him. Jemy, teacher









in Capivara village, mentioned that shamans use the squared big sieve type of basket

(yriipemeauu) to collect the patient's illness and send it away. They use the small sieve strainer

(yrifpemeaii) to collect the soul of the person, and the fan (tapekwap) to preach to the patient and

send the bad spirits or illness away. Kupeap affirmed that there is a type of basket that only

shamans can use. It is named "yrifpefuku", and it is stretched with a handle. Only old men can

make this basket. When spirits of the water (named as karauat) are harming the person, the

shaman leaves this basket for him, by the river. Nobody can say the name of this basket in the

river or even in a dream; otherwise the "karauat" can be harmful. The shamans use this basket to

preach and to keep their belongings.

In 2007, while doing fieldwork, I had the opportunity to visit the village of a Kaiabi

shaman who was living in a nearby town and had recently moved back to the Park. His

Portuguese name was Joao and his Kaiabi name was Jawami 'u which means literally "jaguar's

food". Jawami'u became a shaman after an encounter with a jaguar in the jungle, in which he

was almost eaten by the animal, but resisted, fought and survived. After that incident, he started

to develop dreaming and shamanic powers. While living in the town of Marceldndia, he started

to practice shaman herbalism and cure. He married a Kaiabi woman and moved back to Xingu

Park, establishing his village, named "Fazenda do Joao" (Jodo's ranch) in the Manitsaua-micu

River, a Xingu tributary. In his village, he continued his work and has become a respected Kaiabi

shaman, maybe one of the more powerful ones relating to curing people5. During my visit to the

village, I was very lucky given that, for the first time after working with the Kaiabi for ten years,

I was able to witness a curing ritual named "Maraka," in which the shaman, accompanied by



5 Kaiabi shamans can be specialized in different aspects, such as curing, harnessing crop plants, protecting against
snake bites, spirits of the river, etc. (Silva, 2009). According to Travassos (1984), each Kaiabi shaman might have
his/her own repertory associated with a set of spirits.









other men and family relatives of the ill person, sings and preaches for long hours and even days

in a circle. In this special event, Joao was performing the Maraka ritual to help cure all the

patients present in his village. What struck me was the special use of cotton and baskets in this

ritual. Carefully woven cotton yarns were arranged in the form of a web hanging from the ceiling

of the big house, the two ends of which fell inside two big baskets placed on two painted benches

in the middle of the preaching salon. The cotton web was used to capture the soul of the patient,

while the baskets functioned as receptacles for the soul, so that the shaman could direct them

back to the ill person who had lost or had his or her soul stolen by a spirit. During the ritual, Joao

would teach the participants about the dangers of the forest and how people should behave in

order to avoid getting sick or harmed by bad spirits. I found interesting the association between

cotton and baskets, thinking that they could represent female and male forces in action, as well as

the antagonistic forces of life and death, health and illness. In a way, baskets and cotton in this

context were used to bridge the communication between the mundane and the supernatural

world.

Taboos related to weaving basketry and textiles

Among Kaiabi men and women, taboos related to weaving are mostly linked to

pregnancy or early stages of children's development. When the woman is pregnant, the husband

cannot weave baskets, because that could cause her to have difficulties during the labour process.

They believe that the act of weaving causes blocking of the passage for the baby. If the husband

is weaving a basket and she is having difficulties giving birth, they have to untie the whole

basket to help the baby "pass." The same thing applies to pregnant women; they should not

weave or roll cotton yar, otherwise the baby might have difficulties coming out of the vagina.

They believe the vagina "closes," metaphorically referring to closing the weft in weaving. If the

woman is rolling yarns, the umbilical cord might get around the baby's neck, suffocating the









child. People said that women can only spin cotton during pregnancy, and other relatives can roll

up the yarns or weave textiles.

Basketry, Textiles and Markets

My approach to the study of a cultural domain has been the development of a systems

framework, including the identification and exploration of the different factors that might affect

the creation, distribution and transmission of knowledge related to that given domain. In

previous chapters and sections, I have analyzed historical, environmental, political, socio-

economical and cultural (including spiritual, described in the last section on shamanism) aspects

that affect weaving knowledge among three Kaiabi groups. In this section, I want to explore how

access to markets has influenced weaving knowledge among the Kaiabi. This is different than

market integration, which was treated as a socio-economic variable proxied by stability of

income and discussed in Chapter 7 and in the next section. Here, I will describe how and why the

destination of baskets and textiles to market has contributed to the retention or erosion of

weaving knowledge among the Kaiabi.

Studies of indigenous material culture and of how they have became commodities in the

market economy open up an avenue to understand the intricate relationships between indigenous

and capitalist economic systems. Artifacts have been the principal goods exchanged by Brazilian

indigenous societies within themselves as well as with the westerners during initial contacts. The

commoditization of indigenous handicrafts and the variation in their market value and demand

over time have produced changes in indigenous peoples' social organization, in the mechanisms

of artistic knowledge transmission and distribution, and in systems of use and management of

natural resources used as raw materials (Ribeiro, 1983; 1987a; Newton, 1987).

During the 1970s, FUNAI created the Artindia Program, a system for marketing

indigenous handicrafts in Brazil through the establishment of a network of shops in the principal









urban centers. FUNAI employees began to buy indigenous artifacts directly in the villages and

take them to be sold in the cities. Besides this market opportunity, indigenous peoples have sold

handicrafts to middle men who come to the villages to buy their production and sell in big cities

such as Sdo Paulo and Rio de Janeiro. On the one hand, these initiatives have contributed to

shape and orient the production of handicrafts to the market. On the other, they have provided a

source of income at the family level among different indigenous communities. Some limitations

and problems of marketing of handicrafts by indigenous peoples relate to the dependence on

middle-men and brokers; the under valuation of the objects; and the production of more

marketable items at the expense of other objects with use but no market value.

The possibility to sell their artistic production contributed to both the perpetuation and

change of social traditions and artistic knowledge among many indigenous groups, inside and

beyond Brazil's borders (Ribeiro, 1983; Newton, 1987; Nash, 1993; Krokvin, 1998; Duncan,

2000; Athayde, 2004). While market integration may cause disruptions and conflicts regarding

traditional social organization structures among indigenous societies, in the case of handicrafts

production this integration might lead to conservation and innovation of artistic knowledge

(Ribeiro, 1983;1985; Korokvin, 1998; Athayde, 2004). Godoy (2001), referring to indigenous

environmental knowledge, argues that markets may erode knowledge of some plants but they are

likely to provide greater retention of others. According to him, indigenous peoples will engage in

markets and thus improve their knowledge in producing goods for which they enjoy a

comparative advantage, or in other words, people specialize in activities at which they are best.

The determinants of value in a market economy are supply and demand, social status and

value. The market value of symbolic goods is established not only by social status, but also by

the symbolic value of the commodity in addition to the cost of materials, labor and transportation









(Duncan, 2000:220). Among Native American indigenous peoples, traditional arts and crafts also

became commodities for the market during the 20th century, but rather than becoming

functional, they are more symbolic of the survival and continuity of indigenous cultures

(Duncan, 2000).

Given Kaiabi mastery in basketry weaving, as well as the symbolic meaning of the

graphic designs as icons of their ethnic identity, these objects have been historically their most

valued exchange items. They had exchanged the baskets between themselves, with other

indigenous groups, and with non-indigenous since the first contacts. After the Kaiabi were

transferred to Xingu Park, they began to sell handicrafts to non-indigenous peoples who used to

work in or visit the Park such as anthropologists, biologists, medical professionals and FUNAI

officers.

The Xingu Kaiabi have sold a lot of baskets and clubs for the Artindia Program, as well

as to middle men who come to the Park, especially to buy handicrafts. Since 19986, they have

also sold handicrafts through their local association, ATIX, which has a little store in their office

in Canarana, the nearest town. ATIX has also sold handicrafts to regular buyers who have shops

in Sao Paulo. It is more advantageous for the individuals to sell through the local organization,

because they receive a higher value for the baskets. However, there are still many people in

Xingu who individually take baskets and textiles to sell in nearby or big cities. Women travel

less than men, and generally give their handicraft production to husbands or relatives to sell and

bring back money or products for them. In Rio dos Peixes, people sell handicrafts in the village

or in the nearby town Juara. Men sell mainly clubs (sometimes with woven handles) and women

sell seed necklaces. In Kururuzinho, they sell handicrafts to people who come to the village, or in

6 1 helped to develop a program for training and commercialization of handicrafts through ATIX among Xingu
indigenous peoples. Some of the challenges and achievements of this program were described in Athayde (1998).









the nearby town of Alta Floresta. After the Kaiabi Araa project, with the increase of production

of basketry and textiles in the village, people have also sold baskets and textiles in the

headquarters of Kawaip association in Alta Floresta.

Currently, the main paths for basket and textiles exchange or commercialization amongst

the Kaiabi are (Athayde, 2003; 2004):

* Baskets are given as gifts to women in the family (wife, mother-in-law, aunt, mother,
grandmother, sister);

* Textiles are made to be used by nuclear or extended family members;

* Baskets and textiles are given as payment for a shaman's healing services;

* Baskets and textiles are exchanged for industrial products and indigenous products, among
Kaiabi people, with other indigenous peoples from Xingu Park and with non-indigenous
persons who work or visit the park;

* Baskets and textiles are sold commercially to non-indians who live and/or work at Xingu
Park; to middle-men who come to the park to buy handicrafts, to the local organizations
ATIX and Kawaip, and to shops placed in the cities.

Due to their beauty and scarcity, Kaiabi baskets can fetch good prices in the Brazilian

crafts market. A medium-sized good-quality painted basket can be sold for up to U$ 20.00 to

U$30.00 in specialized shops in Sdo Paulo. The market demand for painted baskets has

profoundly impacted the social and ecological aspects of basketry. For one thing, over-harvesting

of the already limited stocks of arumd in Xingu meant that by the 1980s, production of painted

baskets was again in decline. In addition, because baskets provided a unique and highly valued

source of income, weavers prioritized production for sale over local consumption. Even today,

women complain that the men are not producing even the simpler undecorated baskets for them

to use in daily tasks. Rather, men prefer to use whatever aruma they can find to produce higher-

value painted baskets for sale. While some weavers have adapted to scarcity of aruma by using

poorer-quality substitutes, others notably elders have refused. The commercialization and









increasing demand for baskets in the context of spatial dislocation then, has undermined the

viability of basketry by contributing to resource over-exploitation (Athayde et al., 2006). Being

very time-consuming to make, hammocks are too expensive- around US$200 for a sizeable

market, while the demand for textiles is also quite limited. The recently developed woven bags

using basketry designs have been sold by women from Tuiarare and Kururzinho villages for US$

15 to 20, depending on the size and quality of work. Bags are an interesting product for the

market, since they are much less time consuming than hammocks, are easy to weave, have a

practical use, and are appreciated by non-indigenous women. In addition, they don't cause over-

exploitation of the raw materials as in the case of the baskets.

While market demand and participation is leading to the perpetuation of Kaiabi graphic

design repertoire among the Xingu Kaiabi, there is a constant tension between traditional and

new. It seems that the younger generation tend to innovate for the market, as is happening with

the wooden painted benches and woven bags, while the elders are more attached to the older and

traditional weaving techniques and design repertoire. It is also interesting to note that what is

done for the market can be adapted, while what is done for use has to follow traditions. This

contrast and tension between the old and the new, or tradition and innovation in relation to the

market economy, has also been observed by Duncan (2000) among "mestizo" women in

Colombia. While major changes are not encouraged, there is a tendency for younger generations

to innovate for the market. Among the Hopi women from Arizona, while major changes in

basketry making did not occur because of the strong social meaning and traditions this art has

had historically, production for the market has led to the development of more complicated and

intricate designs in plaques and deep baskets (Teiwes, 1996).









The commoditisation of Kaiabi baskets has clearly transformed their social value in a

number of ways. Men used to learn to weave between the ages of twelve and fifteen, and baskets

were used in the process of social reproduction, notably by assisting in the processing and

transformation of locally available natural resources. The fact that men today are learning later,

usually at about eighteen, suggests that the prestige and economic importance attached to baskets

is no longer linked to local circuits of production and consumption, but rather, to their circulation

within larger, external, networks of exchange, and linked to new patterns of production and

consumption, in which money and industrialized goods provide a benchmark of value and social

prestige (see also, COICA 1996; Fisher 2000). Consequently, for example, wage-earning

professionals such as health agents or teachers, now get married more easily than expert basket-

makers.

The commercialization of basketry and textiles among the Kaiabi have enabled changes

in the social-economic structure of production, availability and circulation of raw materials, and

in the value associated with these objects, which are now mostly destined for the market. At the

same time, it has triggered innovation of weaving knowledge, with use of basketry designs in

new objects that were not traditionally produced for use. Thus, link to the market in this case has

an ambiguous effect on Kaiabi society, causing at the same time changes in social organization

and innovation and perpetuation of weaving knowledge.

Relationship between Weaving Knowledge, Language Proficiency and Socio-economic
Aspects

This section is closely related to the elements, arguments and statistical procedures used

in Chapter 7, in which I presented, compared and discussed the inter-relationships between

socio-economic variables and language proficiency in the four villages studied. Here, I extend

questions and discussions to the domain of weaving knowledge. Weaving knowledge in this case









is comparable to language proficiency as a dependent variable that indicates knowledge or

cultural vitality, and that might be used as a proxy to understand how indigenous knowledge is

created, transmitted and changed in face of displacement and interaction with western

institutions (Maffi, 2001; Zent, 1999; 2001; Athayde et al., 2009). For instance, Benz et al.

(2000) used loss of language proficiency and literacy level as proxy for modernization. As stated

before, even with the methodological limitations in measuring language proficiency, I want to

compare the statistical behavior of these two variables when in interaction with others, and also

to understand the interactions between weaving knowledge and the main independent socio-

economic variables considered in this study, namely age, sex or gender, formal schooling, and

market integration (proxied by income stability). I described these variables in detail in Chapter

7, so I recommend that the reader refer to it for any doubt or clarification about the variables and

their interaction.

I want to revisit the methodological question stated in the general introduction and in

Chapter 7, whether variables measured to indicate any given cultural domain (crop knowledge,

weaving knowledge, ethnobotanical knowledge) are comparable to language proficiency as a

measure of overall cultural vitality. While age, gender and language proficiency are variables

that tell us about the dynamics and functioning of knowledge systems in the population being

studied, formal schooling, market integration, political empowerment (explored in Chapter 6)

and participation in community-based projects indicate how indigenous knowledge systems are

interacting, influencing and being changed by growing participation in and assimilation of

western institutions by indigenous peoples (see Boster, 1986; Nabhan and St Antoine, 1993;

Zent, 1999; Godoy, 2001; Reyes-Garcia, 2001; Hill, 2001; Ross, 2002).









I will be testing five hypotheses (described in detail in Chapter 2): H1, regarding the

relationship between weaving knowledge and age. I have discussed this topic in previous

sections of this chapter, and here I present results of statistical analyses such as linear and

logistic regressions to further explore and thus synthesize the interaction of these variables; H2,

H3 and H4, related to effects of gender, market integration and schooling in weaving knowledge;

H5, related to the effect of community-based project in weaving knowledge. My hypotheses are

that first, older people are more knowledgeable in weaving designed baskets and textiles; second,

that there are marked differences between genders in relation to mechanisms of knowledge

creation, transmission and change; third and fourth, that greater levels of market integration and

formal schooling lead to erosion of weaving knowledge; and fifth, that the Kaiabi Araa project

enebled the revitalization of weaving knowledge among the participants. I have explored several

dimensions and interactions between these aspects throughout this document. Thus, here I

included them as independent variables in my logistic regression model, to test how they relate

specifically with weaving knowledge or the capacity to weave designed baskets and textiles. I

used the same statistical procedures applied to the variables in Chapter 7: a) chi-square and

correlation tests to explore interactions between the dependent variable (weaving knowledge)

and the independent socio-economic variables; b) bivariate linear regression, relating age and

knowledge on woven items, names and uses for basketry designs; and c) multivariate logistical

regression, testing the interaction of the multiple variables (see Chapter 2, methods).

The variables that showed significant statistical correlation with weaving knowledge

were village, age, schooling and language proficiency. Stability of income and gender scores

were not significant in the chi-square and correlations tests run against weaving knowledge.

Stability of income, used as a proxy for market integration, showed significant interaction with









age class and language proficiency (Chapter 7), but no significant relationship with weaving

knowledge here. Gender or sex is the variable that has shown less effect on the dependent

variables, and was also not significant in relation to language proficiency. Thus, the ability to

weave is not strongly affected by gender, analyzing all villages together. As shown in previous

sections, the number of men and women who weave designed baskets and textiles is consistent

among the villages; thus whereas mechanisms for sharing and transmitting knowledge may differ

among genders, these differences are rather subtle and could not be tested statistically. Therefore,

I reject hypothesis H2, but I believe that there is important qualitative information to let us

understand gender differences, summarized in the conclusion of this chapter and also in the

general conclusion. I have also to reject H3, relating market integration to erosion of weaving

knowledge, since they did not show a statistical correlation in the logistic regression model.

I expected that the existence of a significant number of people especially in Rio dos

Peixes and Teles Pires villages who are not proficient in the Kaiabi language and also have

greater integration with market economy through access to wages or other source of stable

income (such as retirement pension and other government programs), would result in a

significant negative correlation with weaving knowledge, just as happened with income and

language proficiency. Yet results showed important differences between weaving knowledge and

language proficiency, since they did not show the same behaviour regarding this specific

variable. As I have stated in the previous section about the complex relationships between

markets and indigenous knowledge, authors have shown the ambiguity of effects that the type

and the intensity of market integration might produce within indigenous societies and knowledge

systems (Reyes-Garcia, 2001; Godoy et al., 1998; Godoy, 2001). For instance, Reyes-Garcia

(2001) found no statistical significance between market integration and erosion of ethnobotanical









knowledge among the Tsimane' indians of Bolivia. However, she reported that market

integration erodes community agreement on ethnobotanical knowledge because of greater

specialization in economic activities and less use of plants. Godoy et al. (1998), working with the

Tawahka indians of Honduras, found that the effects of integration into market on folk

knowledge depended on the economic activity performed: greater integration though waged

labour and agricultural activities might erode ethnobotanical knowledge, whereas a forest-based

economy might enable the continuity of traditional ecological knowledge, especially of those

resources under commercialization. I suggest, concurring with the authors cited above, that the

different ways in which market exposure might affect indigenous social organization, knowledge

systems and thus cultural and environmental resilience should be understood in a multiple cause-

effect relationship, and not in a linear way. Thus, while in Xingu there are waged officers with

greater income stability, such as some teachers and health agents in Capivara and Tuiarare

villages, they do not necessarily show lower levels of competency in weaving knowledge. Good

examples are Aturi and Tare'i in Tuiarare village, who in spite of being waged officers, are

among the experts in basketry weaving. Furthermore, informal linkages to the market through

the sale of handicrafts in Xingu and elsewhere might enable innovation and persistence of

weaving knowledge, as discussed in the last section. While I did not include geographical

distance to markets in my analyses, it is an important factor to consider, since proximity to the

market might greatly affect the nature and degree of market integration by indigenous peoples.

When markets are closer, it is easier to commercialize all sorts of products (even fish, in the case

of Rio dos Peixes) and to get involved in off-village temporary jobs. Therefore, it is useful to

consider type, degree and distance to markets in any research attempting to understand the

effects of markets on indigenous knowledge.









Regarding the other variables, weaving knowledge showed a positive correlation with

language proficiency and age, and a negative correlation with schooling. It also showed

correlation with village, meaning that there is more probability to find weavers in some villages

(Capivara or Tuiarare) in contrast to others (Rio dos Peixes and Kururuzinho). The greatest

correlation was between weaving knowledge and language proficiency (r=0.436, p<0.001), in

which the more proficient a person is in the native language, the greater the probability that he or

she will know how to weave designed baskets and textiles. The correlation between weaving

knowledge and age classified in three age classes (same as used in Chapter 7: 15-25; 25-50;

>50 years of age) was also positive and relatively strong (r=0.245, p<0.001). Age was also

correlated with language proficiency in Rio dos Peixes and Kururuzinho villages (from Chapter

7, r=0.398, p<0.001). As expected, the older the person, the greater their probability of being

proficient in the language and of having weaving knowledge. As I mentioned in Chapter 7, other

studies have already shown this relationship, including Silva (2009), working with Kaiabi crop

knowledge on peanuts. Zent (1999), working with the Piaroa people from Venezuela, found that

age showed the strongest positive relationship with ethnobotanical knowledge (r2=0.539) when

compared to formal education and bilingual ability.

In order to better understand the relationship between age and weaving knowledge, I ran

linear regression analyses including the following correlations: a) age with number of items that

a man can weave; b) age with ability to name basketry designs (number of names given); and c)

age with the ability to weave or use the different basketry designs (number of designs woven).

This type of analysis informs us about which aspects of weaving knowledge are influenced by

the independent variable (age). Results of the linear regressions are presented in Figure 8-23. It is

interesting to note that the only strong and significant relationship was between age and number









of basketry items woven by men (Rsq=0.352). While the distribution between age and

knowledge of basketry items is more consistent through age classes, the other two measures

presented a lot of young people knowing fewer designs and few older people knowing more

designs. The relationship between age and uses (Rsq=0.086) or names (Rsq=0.068) for designs

was not coherent with the linear model, thus not significant. In 2002, I found no significant linear

relationship between age and use or ability to weave different designs (Athayde, 2003), but I did

find a significant relationship between age and ability to name basketry designs (Rsq=0.5446).

This inconsistency between 2002 and 2007 might be explained by methodological aspects and

also by changes in the knowledge of the population sampled. In terms of methodological aspects,

there were more people interviewed in 2007, in a slightly different sample than that of 2002.

Also, there is the problem of inconsistent testimonies regarding names for basketry designs.

Related to changes in the population, it seems that in 2007, younger people knew more names for

basketry designs than in 2002. Again, this might be consequence of their participation in the

Kaiabi Araa project and from their learning names from the book.

There is an important conclusion related to general (weaving baskets and textiles in

general) and specialized knowledge (weaving designed baskets and textiles). Weaving any type

of baskets might be considered general knowledge in contrast to weaving designed baskets. For

specialized knowledge, the relationship between variables might present variations in patterns

that are overall accepted for representing a given interaction. It seems that age is more important

in that general domain of basketry weaving than in that of expertise in designed baskets, in

which younger people might develop expertise in comparison to older people. In addition, it is

now impossible to isolate the effects of western institutions and cultural revitalisation projects on

the dynamics of weaving knowledge among the Kaiabi. I did not run this analysis for women,









given the fact that there are few designs and items woven by them to produce significant and

comparable results in this case.

Formal schooling showed a negative and relatively strong correlation with weaving

knowledge (r=-0.381, p<0.001). Interestingly, it was also negatively correlated with age (r=-

0.573, p<0.001) and language proficiency (r=-0.298, p<0.001). Therefore, the younger a person

is, the more formal schooling they are likely to have, and the greater the probability that he or

she won't be proficient in the language nor know how to weave designed baskets and textiles.

This type of interaction between schooling and indigenous knowledge has been shown by other

researchers working with native American and Amazonian indigenous peoples. Nabhan and St.

Antoine (1993:244) worked with O'odham and Yaqui indigenous communities in the Sonoran

desert, found that higher schooling levels among children led to loss of proficiency in native

language and its "encoded biological knowledge". Zent (1999) also found a negative correlation

(r2=0.220) between level of education or formal schooling and ethnobotanical competence

among the Piaroa indigenous people from Venezuela. Here, I validate my hypothesis H4,

concluding that formal schooling has the tendency to erode indigenous or weaving knowledge. I

have already discussed relevant aspects of formal education among indigenous peoples in

general and among the Kaiabi in particular. I just want to bring back the idea that, as happens

with markets and other western institutions, schooling might not always have perverse effects on

indigenous knowledge systems. Again, it depends on the type of school present in indigenous

villages, which might enable or constrain the transmission of native language and other

knowledge domains. In the case of the Xingu Kaiabi, the educational system present there has

attempted to combine the teaching of western and indigenous knowledge in a more integrated

way. In contrast, in Rio dos Peixes school the main language used and taught is Brazilian









Portuguese and the main knowledge transmitted is that of westernized Brazilian formal

education.

The results of the logistic regression contrasting weaving knowledge with the other

independent variables are presented in Table 8-7. Considering weaving knowledge as the

dependent variable, the differences between Xingu villages and Rio dos Peixes and Kururuzinho

or Teles Pires were significant for two variables: language proficiency, with a positive

correlation, and schooling, with a negative correlation. In the first analysis, with all variables

included, only language proficiency was significant (p<0.001, see table 8-7). This happened

because of the high levels of proficiency in Capivara and Tuiarare, which are positively

correlated with a greater amount of weavers among men and women in comparison to the other

two villages. The odds ratio for language proficiency was 7.116, indicating that a person which is

proficient in the Kaiabi language has 7 times more chance of being a weaver. The second

variable in that model was schooling, which was significant at a 95% level of confidence

(p<0.009). Controlling for language proficiency, or taking this variable out of the analysis, the

second most important variable is schooling (p<0.001), in a negative correlation and not so

strong association. People with no schooling have 2.5 times more chance of weaving designs in

comparison with those who have some level of formal education. I also took Xingu out of the

model and ran the logistic regression with only Rio dos Peixes and Kururuzinho village, because

of the fact that everybody is proficient in Xingu. Considering only Rio dos Peixes and

Kururuzinho, language proficiency is by far the most important variable explaining weaving

knowledge (p<0.001), with 5 times more chance for people who are proficient also being able to

weave designed hammocks and textiles. Considering only women, language proficiency was still

the most important variable.









I ran the analysis including participation in the Kaiabi Araa project as one of the

independent variables, and it appeared in first place along with language proficiency, in a

positive relation with high significance (B=1.612, p<0.001). The odds ratio for participation in

the project was 5.013, meaning that there is 5 times more chance for a person who participated in

the project of being a weaver of designs. Based on this and on previous results related to the

project Kaiabi Araa, I validate H5, in which I state that there was an increase of weaving

knowledge in villages which participated in the project. Taking men out of the model, only

language proficiency was significant in explaining weaving knowledge among Kaiabi women.

This may be related to the fact that more men participated in the project or that more women who

weave designs did not participate (such as Capivara women for instance).

According to the results presented in Chapter 7, in which we had language proficiency as

the dependent variable, age was the most important variable explaining language proficiency

among people from Rio dos Peixes and Teles Pires. The second variable in that model was

stability of income, showing a negative relationship with proficiency. In that analysis, we did not

have weaving ability in the model, because we were trying to understand the effect of the other

socio-economic variables on language proficiency. I ran the analysis again, only for Rio dos

Peixes and Teles Pires villages, choosing language proficiency as the dependent variable and

including weaving ability together with the other independent variables. This time, weaving

ability was matched with age class as the most important variable explaining language

proficiency, at the same level of significance (B=1.593, p<0.001). Stability of income still comes

in second place, but with a lower level of significance (p<0.020). The chance of someone who

weaves also being proficient in the language is 49.17%, similar to what we had above, in which

weaving ability was the dependent variable.









The results of the logistic regression model show that proficiency in the native language

and weaving knowledge are strongly related. However, their statistical behaviors differ such that

for language proficiency, age and stability of income are more influential, whereas for weaving

knowledge, level of formal schooling is more influential. Therefore, I conclude that in the

multivariate logistic regression model, the weaving knowledge domain presented different

statistical behavior when compared with proficiency in indigenous language, and thus language

proficiency should be used cautiously as a proxy for indigenous knowledge, since it might not

capture the nuances and specificities of different cultural domains. There is still a strong

correlation between the two, in that it is expected that greater language proficiency also means

greater retention of indigenous knowledge. However, in the case of weaving knowledge, age is

not so relevant in explaining its distribution, since this is a specialized knowledge domain which

might be mastered by younger and older people in a comparable way. Even if older people are

more knowledgeable, the relationship is not that strong. However, for other knowledge domains,

such as crop knowledge or ethnobotanical knowledge, age may still be the most influential

variable and therefore more closely comparable to language proficiency than a specialized

knowledge domain (see for example Zent, 1999 and Silva, 2009, among others).

Observing Figure 8-24 which shows the score of each village relating to the different

variables it is visible that language proficiency and weaving ability relate to each other,

whereas stability of income and schooling also go together, especially in Rio dos Peixes. Thus,

less stability of income and less formal schooling mean greater language proficiency and

weaving knowledge and vice-versa. The behavior of these different variables can also be

visualized in Figure 8-25, in which I present some models for the interaction of variables

explored in this study. In the model, I include the panarchy curve from Gunderson and Holling









(2002) presented in Chapter 1, which I consider a better way to depict the effect of ambiguous

variables such as market integration on indigenous knowledge systems: while integration with

market might lead to disruptions in social organization, it may also lead to innovation and

resilience of knowledge systems. Disruption occurs when knowledge is not being renovated, and

thus cultural resilience is lost, such as in the case of Rio dos Peixes. Linear representations are

somehow too mechanical and bi-polar to capture the nuanced and systemic character of real

world dynamics of indigenous knowledge in the face of change and of integration with non-

indigenous society's knowledge systems, economic structure and development paradigms.

Conclusion

Results presented in different sections of this chapter support the argument that weaving

knowledge has been simultaneously innovated and eroded among the Kaiabi as a group.

Innovations have been developed with the possibility to commercialize baskets, textiles and

other objects carrying Kaiabi insignias or symbols, and through new institutions such as

indigenous organizations and projects. The complex and nuanced mechanisms through which

knowledge might be gained, transformed, transmitted, eroded and even regained among

indigenous societies were illustrated by exploring similarities and differences among villages and

genders on different weaving knowledge sub-domains or modules (form, function, materials,

meaning and motifs).

Whereas in Xingu villages basketry knowledge has been constrained by lack of aruma

and lack of interest to learn by the youth, in Rio dos Peixes the situation is more critical, since

there is no incentive for learning and younger men and women are not learning from their

relatives. Both Kaiabi language and weaving knowledge are seriously threatened with

disappearance in Rio dos Peixes.









In Xingu and in Teles Pires there are some younger people still learning, especially after

the development of the Kaiabi Araa project. This would be one major condition for knowledge or

cultural resilience. Any given indigenous society might be able to keep their knowledge

patrimony, as long as there are new and young people learning, even with all the innovations. I

consider innovations as assets and as a capacity to adapt and survive, thus another ingredient for

achieving cultural resilience. In Xingu, even with the lack of arumd, the capacity to innovate in

basketry and textiles knowledge or production has been developed by applying multiple

strategies such as using substitutes, applying basketry designs in other objects, developing a

textile "industry" using basketry designs, and even being able to carry out a community-based

project for knowledge revitalization.

Through this study, I confirm hypothesis H1, that older people retain more detailed

knowledge on basketry than youth. However, I have shown that cultural domains might differ in

their patterns of knowledge distribution and interaction with other variables. Thus, age is less

important as a determinant of weaving knowledge of designs among the Kaiabi in comparison to

language proficiency or schooling. Language proficiency and weaving knowledge, as expected,

are strongly correlated. Schooling also affects weaving knowledge in a negative way, taking into

account that in Rio dos Peixes schooling has also encouraged the dis-use of the native language,

which erodes along with weaving knowledge and probably with other cultural domains. I

confirm H4, that schooling erodes weaving knowledge. However, the effects of variables are

never linear or one-way, as exemplified by market integration, which might have positive and

negative impacts on indigenous knowledge. Since I found no statistical evidence linking markets

to erosion of weaving knowledge, I reject H3. However, the relationship between language

proficiency and income stability (proxy for market integration) was strong.









I found that men and women do present different mechanisms in knowledge creation and

transmission: whereas women's knowledge transmission is more embedded in the household and

in kinship dynamics, among men the learning process is more scattered and individualistic. Thus,

women's knowledge might be more resilient or less prone to disappearing, since it is more

embedded in kinship and social structure, and includes less sophisticated items. Also, the fact

that women can use industrialized cotton to produce textiles is an advantage in comparison to

men who need to harvest and process natural resources to produce different baskets. As I found

no statistical evidence to ground my findings on gender differences, I reject H2 that men and

women use different mechanisms in knowledge creation and transmission.

One of the interesting results of this research was that mechanisms of knowledge

transmission seem to have changed among the Xingu and Teles Pires Kaiabi after the

development of the Kaiabi Araa project, in which another mechanism for learning- that of a

community of practice in a workshop- was invented by them. For the specialized domain of

weaving knowledge, there are seven mechanisms for transmission in place among the Xingu

Kaiabi today. In Rio dos Peixes, the only mechanism used by men to learn new designs was the

book Kaiabi Araa that somebody from Xingu brought to them. Thus, Xingu is at the same time a

laboratory for innovative cultural experiments, and a repository and source of knowledge for the

other villages. In Kururuzinho, the project produced good results, since there are some younger

and older men and women who learned and are actively weaving baskets and textiles for sale to

tourists. I therefore accept hypothesis H5, that the Kaiabi Araa project was responsible for an

increased number of basketry and textiles weavers in the villages that participated in the project.

Finally, shamanism has played an important role in knowledge creation and transmission

related to basketry and textiles among the Kaiabi. The fact that there are fewer and not so









powerful shamans nowadays in comparison to past times might interfere in the resilience of the

deeper spiritual meanings and artistic innovations that are dependent on shamans and dreams.






































Figure 8-1. Graphic representation of domains or modules of knowledge involved in weaving
knowledge, considered in this research. Broad categories considered: 1) Form; 2)
Function; 3) Materials; 4) Meanings; 5) Designs. Modified from Ellen (2009).








Weapons/Traps


Jesi'a

Headfeathers
Kangytat
Kangytareta




Ba




Matt


Fans


Figure 8-2. Types of Kaiabi baskets grouped according to semantic categories and uses, with distinction of use by men, women or
both. Based on Ellen (2009).










Table 8-1. Basketry items and textile related items produced by Kaiabi men.
Basketry Name in Name in Raw materials used in Uses Used by Produced
items name Portuguese English manufacture man (M), for sale?
in Kaiabi woman Y/N
(W), both
(B)
1 Tapekwap Abanador Fan Young leaf of tucuma To fan household B Y


palm, vine for rim,
cotton to tie
Young leaf of tucuma
palm, vine for rim,
cotton to tie


Palm wood, woven
handle made from
aruma fiber, died with
tree bark red resin,
adorned with woven
cotton




Vines' fiber




Aerial roots of imbe
plant


fire, to cover food,
to preach (shaman)
To store tools and
other objects



In the past, used as
a weapon,
nowadays, as
symbol of identity
in meetings and
festivals


To transport food
from forest or
garden plot, to
transport objects


To store objects
and food


2 Tamakari






3 Muap






4 Myayta


5 Paneyrii


Cesto
Cilindrico





Borduna






Maiaco




Cesto
paneiro


Cylindrical
basket


Wooden
Club w
woven
handle






Tied vine
basket



Lattice
work
basket










Table 8-1. Continued
Basketry items Name in Name in Raw materials Uses Use by Produced
name in Kaiabi Portuguese English used in man (M), for sale?
manufacture woman Y/N
(W), both
(B)
6 Yriokote'em Bolsinha de Palm leaf Young leaf of To store small B N


7 Panakfiawet


8 Yripewai
00
oo


9 Tapekwajowai


Cesto


Peneira com
cabo


Abanador de
Dois Cabos




Armacio
para Cocar


little bag


Basket


Basket with
handle


Two-handle
fan




Headfeather
support


tucumd palm,
vine for rim,
cotton to tie


Inajc palm leaf,
vine for the rim,
cotton to tie
Arumd fiber,
vine for the rim,
cotton to tie

Young leaf of
tucumd palm,
vine for rim,
cotton to tie


Arumd fiber,
vine for the rim,
cotton to tie


10 Kangytaryta


objects


To store objects and
food


Used by the shaman
to store food and
objects

To fan household
fire, to cover food,
to preach (shaman)



To hold feathers
string, headfeather
is used in meetings
and festivals


Palha de
Tucum










Table 8-1. Continued
Basketry items
name in Kaiabi


Name in
Portuguese


Name in
English


Raw materials
used in
manufacture


11 Arapi




12 Pinosing




13 Yriipemeauu


14 Yriipemeai'i




15 Panakfi


Enfeite da
Flecha



Esteira


Peneira de
Malha
Grossa


Peneira de
Malha Fina



Cesto
cargueiro,
jamaxim


Peneira de
Tucum


Arrow
ornament



Woven
mat

Strainer
big sieve


Strainer
small
sieve


Backpack
twill-
plaited
basket
Palm leaf
basket


16 Yripejuap


Leaf petiole of
inajc palm, cotton
to tie


Inajc palm leaf


Arumd fiber, vine
for the rim, cotton
to tie


Aruma fiber, vine
for the rim, cotton
to tie


Aruma fiber, vine
for the rim, cotton
to tie


Young leaf of
tucumd palm, vine
for rim, cotton to
tie


Ornament for arrows,
used for hunting or in
festivals


To cover objects, to
seat

To prepare beverages
and porridges, to sieve
flour


To prepare beverages
and porridges, to sieve
flour


To transport hammocks
and clothing (in the
past)


To store objects, to spin
cotton


Uses


Used by
man (M),
woman
(W), both
(B)


Produced
for sale?
Y/N










Table 8-1. Continued
Basketry items Name in Name in Raw materials Uses Used by Produced
name in Kaiabi Portuguese English used in man (M), for sale?
manufacture woman Y/N
(W), both
(B)
Peneira Twill- Aruma fiber, vine To store objects, to W Y


18 Yrfifuku


19 Juyp Munuwi
Yrfi


20 Yrfipefuku


Armadilha
para Peixes


plaited
designed
basket



Long
basket



Peanut
storing
basket

Twill-
plaited
oblong
designed
basket


Fish trap


for the rim, cotton
to tie, red resin
from tree bark



Young leaf of
tucumd palm, vine
for rim, cotton to
tie
Leaf petiole of
inajc palm, cotton
to tie

Aruma fiber, vine
for the rim, cotton
to tie, red resin
from tree bark


Leaf petiole of
inajc palm, vine,
cotton


spin cotton, to
adorn the house, to
preach (shaman),
in the Maraka
festival
To store objects




To store peanuts



To store objects, to
spin cotton, to
adorn the house, to
preach (shaman),
in the Maraka
festival


To catch fish


Desenhada


17 Araa


Cesto
Comprido



Cesto para
guardar
amendoim

Peneira
oblonga
desenhada


21 Jesi'a










Table 8-1. Continued
Textile related items made by men

Textile items Name in Name in Raw materials used Uses Used by man Produced
name in Kaiabi Portuguese English in manufacture (M), woman for sale?
(W), both (B) Y/N
1, Fuso para Spindle Palm wood Used by W N


2 Taity Pypykap


3 Tupaam






4 Awanifu'am






5 Taity Retykap


Textile
comb


Fiber rope






Cotton
woven
feather wig



Wooden
loom


(pupunha-brava),
turtle shell disc


Wood


Fiber from a
bromeliad
cultivated in garden
plot (Ama 'yp)


Cotton, bird
feathers




Wood, vine for tie


women to spin
cotton


Used by
women to
weave textiles


To tie objects
or hammock




Used by men in
festivals and
political
meetings


Used by
women to
weave textiles


Sources of information: Ribeiro (1985); Athayde (1998);
common plants can be found in Table 5- 1 (Chapter 5).


ATIX (2006 a). Names of plants are in Portuguese, scientific names for most


tecelagem


Pente para
Tecelagem


Corda






Touca de
penas




Tear de
madeira


y-
























A B


D E


Figure 8-3. Stages of basket weaving. A. and B. Collecting and transporting arumm canes Pirapd,
Coroner and Pop6). C. Osmar depithing "arumm" fiber. D. Arumm strands drying in the
sun. E. Aturi preparing the dye. F. Eroit weaving the body of the basket. G. Pi'u
preparing the rhym. H. Tu'im and his grandson painting baskets during workshop. I.
Basket ready, collected by Georg Grunberg in Rio dos Peixes in 1966. Photographs
A-H by Simone Athayde, photograph I by Georg Grunberg.













2 3 3aS 3 2





_ I














3 466 78 9 t!i12!;3fMf5f6f7


Figure 8-4. Kaiabi ethnomathemathics: counting, grouping and design structure. The design
woven in the center of the basket square is the "I'yp" pattern (Mendes, 2001).










Table 8-2. Main natural resources used in Kaiabi basketry and textiles.
Kaiabi name Portuguese Species (Family)
name
Ama 'yp Caraud Neoglaziovia variegate
(Bromeliaceae)
Amyneju Algodao Gossipium barbadense (Malvaceae)


Ambewyt


Inata 'yp



Jemore 'yp


Yryp


Ywy ujupe


Yrfipepepoyta


Tukuma 'yp


Uruyp


Imbe


Inaja



Jequitibi


Siriva


Mirticea


Cip6


Tucuma


Aruma


Heteropsis sp (Araceae)


Maximilliana maripa
(Arecaceae)

Cariniana sp
(Lecythidaceae)
Bactris macana (Arecaceae)


Myrcia deflexa (Myrtaceae)


NI


Astrocaryum aculeatum
(Arecaceae)
Ischnosiphon spp
(Marantaceae)


Part used Uses /Items


Fiber Hand-made ropes


Fiber Hammocks, straps for carrying
babies, bags, belts, various
baskets
Aerial roots Paneiri basket


Leaf and leaf petiole Fish trap, arrow ornament, mats,
basket for storing peanuts,
wicker basket
Resin from bark Painted baskets, clubs
(dye)

Wood Spindles, clubs


Resin from bark Painted baskets, clubs (substitute
for jemore'yp)

Stem Rim of various baskets


Young leaf Fans, various baskets


Stems Strainers, twill-plaited designed
baskets









Table 8-3. Plants currently used as arumd substitutes by the Kaiabi people.
Kaiabi name Portuguese Species (Family) Habitat Availability Uses
name


Kwasingewi


Myricipe'yp


Panakfiwa,
Maraja'i
Pokop




Takwasing


Wywa


Taquarinha


Buriti


Jacitara


Banana-brava




Taquara



Cana-brava


NI- (Poaceae)


Mauritia flexuosa
(Arecaceae)


Desmoncus sp.
(Arecaceae)
Heliconia sp.
(Heliconiaceae)



NI (Poaceae)


Gynerium
sagittatum
(Poaceae)


Non-flooded
forests, in the
headings of small
river courses.
In dense
populations called
"buritizais" along
small river
courses.
Riverside forests.


Swampy forests,
non-flooded
forests and
riverside forests.
Non-flooded
forests and
riverside forests.
Planted in
agricultural plots.


Medium to low.


High, the species is well
represented within the Park's
boundary.


Medium to low.


High, ; found in patchy,
high-density distributions.



Medium to low.


High.


Stems used for smaller
baskets.


Petiole used for some baskets,
not usually painted.


Stems used for "panakii" and
for baskets.
Leaves used for the body of
the baskets and the "panaki"
(define).


Stems used for body of
baskets


Stems used for baskets,
(quality generally low).


Adapted from Athayde et al. (2006). NI = not identified.













Table 8-4. Basketry items and textile related items produced by Kaiabi men with respective frequencies in each village by weavers
and by total of men included in the sampling. CA Capivara; TU Tuiarare; RP Rio dos Peixes; TP Teles Pires.
Basketry items Name in % of % % of % % of % % of % % total %
name in English weavers total weavers total weavers total weavers total weavers total
Kaiabi men men men men men
CA CA TUTU R RP TP TP

1 Tapekwap Fan 30.43 28.00 66.67 60.87 13.79 11.11 53.33 30.77 37.50 30.00


2 Tamakari


3 Muap


4 Myayta


5 Paneyrui




6 Yruiokote'em


7 Panakuawet

8 Yrupewai





9
Tapekwajowai


10
Kangytaryta


Cylindric
al basket


Wooden

Club

Vine
basket

Basket


Palm leaf
little bag


Basket

Basket
with
handle


Two-
handle
fan

Headfeat
her
support


42.86


47.62


76.19




28.57


0.00


33.33


14.29





19.05



19.05


39.13


43.48


69.57




26.09


0.00


30.43


13.04





17.39



17.39


10.34


89.66


62.07




6.90


3.45


31.03


10.34





10.34



17.24


8.33


72.22


50.00




5.56


2.78


25.00


8.33





8.33



13.89


6.67


46.67


53.33




40.00


0.00


13.33


13.33





6.67



0.00


3.85


26.92


30.77




23.08


0.00


7.69


7.69





3.85



0.00


19.32


63.64


69.32




17.05


1.14


26.14


10.23





10.23



12.50


15.45


50.91


55.45




13.64


0.91


20.91


8.18





8.18



10.00













Table 8-4. Continued
Basketry items Name in % of % % of % % of % % of % % total %
name in Kaiabi English weavers total weavers total weavers total weavers total weavers total
men men men men men
CCA TU TU RP RP TP TP

11 Arapi Arrow 21.74 20.00 23.81 21.74 27.59 22.22 6.67 3.85 21.59 17.27
ornament


12 Pinosing


Woven
mat


13Yruipemeauu Twill-
plaited
basket
big sieve



Twill-
14 Yripemeai'i plaited
basket
small
sieve


15 Panakui


16Yruipejuap





17 Yrfipem
Araa


Backpack
basket



Palm leaf
twill-
plaited
basket

Twill-
plaited
basket
with
graphic
designs


8.70


52.17


8.00


48.00


40.00


4.35 4.00


26.09




26.09


24.00




24.00


33.33


47.62


71.43




0.00


38.10




42.86


30.43


43.48


65.22




0.00


34.78




39.13


20.69


34.48


27.59




0.00


20.69




17.24


16.67


27.78


22.22




0.00


15.38


15.38


46.67


26.92


21.59


40.91


17.27


32.73


45.45 36.36


0.00 0.00


16.67




13.89


7.69




30.77


25.00 20.00




31.82 25.45













Table 8-4. Continued
Basketry Name in % of % % of % % of % % of % % total %
items name English weavers total weavers total weavers total weavers total weavers total
in Kaiabi men men men men men
CA CA TUTU RP RP TP TP

18 Yruifuku Long basket 4.35 4.00 9.52 8.70 10.34 8.33 0.00 0.00 6.82 5.45


47.83


44.00


42.86


8.70 8.00 14.29


34.78


32.00


23.81


39.13



13.04





21.74


34.48



6.90





37.93


27.78



5.56





30.56


6.67



6.67





13.33


35.23 28.18


29.55 23.64


Textile Items Manufactured by Men


69.57




52.17


39.13





21.74


64.00




48.00


36.00





20.00


61.90




28.57


33.33





9.52


56.52




26.09


30.43





8.70


27.59




27.59


34.48





13.79


22.22




22.22


27.78





11.11


69 57 64 00 66 67 60 87 24 14 19 44


6.67




0.00


20.00





6.67


3.85




0.00


11.54





3.85


43.18 34.55




29.55 23.64


32.95 26.36





13.64 10.91


1333 769 4432 3545


19 Juyp
Munuwi
Yrin

20
Yruipefuku




21 Jesi'a


Peanut
storing
basket

Twill-plaited
oblong
basket with
graphic
designs

Fish trap


22 E'ym


23 Taity
Pypykap


24 Tupaam




25
Awanifu'am


26 Taity
Retykap


Spindle


Textile comb




Fiber rope


Cotton
woven
feather wig


Wooden
loom











Table 8-5. Textile items produced by Kaiabi women.
Textile Name in Name in English Materials used Uses Used by man Produced
items name Portuguese (M), woman for sale?
in Kaiabi (W), both (B)

1 Tupai Tip6ia Strap for carrying Cotton To carry babies W Y
baby


Bolsa padrio
entretecido
sarjado desenhado



Rede padrio
entretorcido



Rede tecida
padrio simples
xadrezado




Redes padrio
entretecido
sarjado desenhado


Woven bag,
twilled technique




Hammock,
twined technique



Plain woven
hammock without
designs




Hammock, twilled
technique with
graphic designs


Cotton






Cotton


Cotton







Cotton


To carry objects, for
sale




To sleep, traditional
Kaiabi hammock


To sleep







To sleep


Cinto, padrio
entretecido
sarjado desenhado


Woven belt,
twilled technique


Cotton, seeds,
beads, feathers,
tapir claws, etc


Used by men in
festivals, political
meetings and
ceremonies


2 Tepyril






3 Taity jepe/
Taity tare'a


4 Taity
pypyk





5 Taity
jewak


6 Ku'afaap









Table 8-6. Textile items produced by Kaiabi women with respective frequencies in each village by weavers and by total of women
included in the sampling. CA Capivara; TU Tuiarare; RP Rio dos Peixes; TP Teles Pires.



Textile items name in -
Kaiabi *



Strap for Woven bag, Hammock Plain woven Hammock, twilled Woven belt,
Name in English carrying twilled twined hammock technique with graphic twilled
baby technique technique without designs designs technique
% of weavers CA 90.91 4.55 45.45 90.91 77.27 54.55
% total women CA 83.33 4.17 41.67 83.33 70.83 50
% of weavers TU 95 65 25 65 65 5
% total women TU 86.36 59.09 22.73 59.09 59.09 4.55
% of weavers RP 81.25 6.25 43.75 87.5 18.75 18.75
% total women RP 32.5 2.5 17.5 35 7.5 7.5
% of weavers TP 100 62.5 0 0 37.5 62.5
% total women TP 28.57 17.86 0 0 10.71 17.86
% total weavers 90.91 30.3 33.33 71.21 54.55 31.82
% total women 52.63 17.54 19.3 41.23 31.58 18.42













Jesi'a / Fish trap
Yrupefuku / Oblong designed baslket
Juyp Munuwi Yru / Peanut storing basket
Yrufuku / Long basket
Araa / Twill-plaited designed basket
Yrupe Juap / Palm leaf strainer
Panaku/Backpack basket
Yrupemeai'i / Strainer small sieve
Yrupemeauu / Strainer big sieve
Pinosing / Woven mat
Arapi / Arrow ornament
Kangytaryta / Headfeather support
Tapekwajowai / Two-handle fan
Yrupewai / Strainer
Panakuawet / Wicker basket
Yruokote'em / Palm leaf little bag
Paneyru / Lattice work basket
Myayta / Vine tied basket
Muap / Club
Tamakari / Cylindrical basket
Tapekwap / fan


0% 20%


40% 60%

% of men each village


Figure 8-5. Distribution of men who weave each type of basketry item by village. CA-Capivara;
TU Tuiarare; RP Rio dos Peixes; TP Teles Pires.


I


100%


* CA
* TU
0 RP
OTP












6 Ku'afaap/Belt


5 Taityjewak / Designed
Hammock

4 Taity pypyk/Hammock
(Plain)

3 Taity Jepe/Hammock
(traditional)

2 Tepyru / Bag


1 Tupai / Strap for baby


0% 10% 20% 30% 40% 50% 60% 70%
% of women each village


80% 90% 100%


Figure 8-6. Distribution of women who weave each type of textile item by village.
TU Tuiarare; RP Rio dos Peixes; TP Teles Pires.


CA-Capivara;


* CA
* TU
O RP
O TP


II I


II I


I I


I I


II I


I I














Myayta / Vine tied basket
Muap / Club
Yrupemeai'i / Strainer small sieve
Yrupemeauu / Strainer big sieve
Tapekwap / fan
Juyp Munuwi Yru / Peanut storing basket
Araa / Twill-plaited designed basket
Jesi'a / Fish trap
Panakuawet / Wicker basket
Yrupe Juap / Palmleaf strainer
Pinosing / Woven mat
Arapi / Arrow ornament
Tamakari / Cylindrical basket
Paneyru / Lattice workbasket
Kangytaryta / Headfeather support
Yrupewai/ Strainer
Tapekwajowai/ Two-handle fan
Yrupefuku / Oblong designed basket
Yrufuku / Long basket
Yruokote'em/ Palm leaf little bag
Panaku/Backpack basket


55.45

50.91
36.36
32.73
30.00
28.18
25.45
23.64
20.91
20.00
17.27
17.27
15.45
13.64
-10.00
8.18
- 8.18
S7.27
S5.45
10.91
10.91


0 10 20 30 40 50 60

% of men all villages




Figure 8-7. Percentage of men in all villages, who weave each type of basketry item.













Tupai / Strap for baby

Taity pypyk/Hammock
(Plain)

Taityjewak / Designed
Hammock

Taity Jepe/Hammock
(traditional)

Ku'afaap/Belt


Tepyru / Bag


52.63


141.23


131.58


19.30


18.42


S17.54


10.00


20.00 30.00 40.00
% of women all villages


Figure 8-8. Percentage of women in all villages, who weave each type of textile item.


H


50.00


60.00












79.17

71 68.00


Capivara


81.82

65.22


32.14 34.62


E Women
* Men


90.00
80.00
70.00
60.00
50.00
40.00
30.00
20.00
10.00
0.00


Villages

Figure 8-9. Percentage of women and men who weave designed baskets and textiles in four
Kaiabi villages. CA- Capivara; TU Tuiarare; RP Rio dos Peixes; TP Teles Pires.


Tuiarare Rio dos Peixes Teles Pires















70

60
S10-15
50
S16-21
4 40 0 22-30


4o 40
20

10


F M F M F M F M

CA TU RP TP
Villages/gender


Figure 8-10. Age of learning designed baskets and textiles among female (F) and male (M)
weavers in four Kaiabi villages. CA- Capivara; TU Tuiarar6; RP Rio dos Peixes;
TP Teles Pires.












Modes of cultural transmission
Verticalor Horizontal or On-to-many Concertcd or
pare-nl -rhijd "contagious maany-to. ne


IV V A w

Transminer Parent(s) Unrelared Teacher/ Older members
Icadrr./mcdia of social gruup
Transnmicree Child Unrelated Pupils/ Younger
citizens/ members of
audience social group
Acceptance of Intermediatc Easy Easy Very dlilcult
innovation difficult
Variation between
individuals within Highi Low LoweiC
ipop,,la ior
Variation beLweenl gh Can be Can be Smals
groups high high
Cultural evolution Can be
SI-.... rtad t rapid Most conervaiv


Figure 8-11. Typology of modes of cultural transmission developed by Cavalli-Sforza and
Feldman (1981) and Hewlett and Cavalli-Sforza (1986).















35 30 30
3030 30
f 25 20
20 -
15 -10
o 10 5 5
o 5









Figure 8-12. Ways of learning to weave designed baskets among Kaiabi men in two villages in
Xingu in 2002.













35.00
3077
30.00 89
2615 2667
2500 22 2222
20 0
S20.00 Women
153
15.00 1333Men

10.00 8 89
462 444 444
5.00 3- 8 308 3082

I0 A --r 0001
0.00







Ways of learning


Figure 8-13. Ways of learning to weave designed baskets and textiles among Kaiabi women and
man in all villages in 2007.















Kinsip


Non-Kimship


S Veraial
Motr9/Fathbr




\ blique
X Between genetions
inkin rop


Harimntal
Between idiviu4d in
kin grap


Many-to-ane
Older member.
d fkingroip
teaching younger


Figure 8-14. A model for the study of indigenous knowledge domains and mechanisms of

transmission based on the Kaiabi case. Adapted and expanded from Cavalli-Sforza

and Feldman (1981) and Hewlett and Cavalli-Sforza (1986).


O\ bliqne
Between
gnierahions



---- Hrintal
Between any
tvo individuals


O(e-to-many

Indigenous
a (boo ks


fn etc)



Many-to- me
VbWksbops
dd riteaching
)nungw-





mr
Many-to-miany

Callaborative
leaning

,k sbap, field
tkpE> p-qeled


Alone

By copyr observation



Presence of

Tra1-&ror

Bookds
Otbects















PE29

PE28

PE27

PE26

PE25

PE24

PE23

PE22









g PE 2I-- | evr
PE21

PE20

PE19

PE18

PE17

PE16

PE15

PE14

SPE13 I % total men
SPE12 % weavers

SPE11

PE10

PE9

PE8

PE7

PE6

PE5

PE4

PE3

PE2

PE1

PA7

PA6

PA5

PA4

PA3

PA2

PAl

000 10 00 20 00 30 00 40 00 50 00 60 00 70 00 80 00 90 00

% of men


Figure 8-15. Distribution of knowledge on different basketry designs among Kaiabi men,

comparing total men interviewed and only basketry weavers.











421













PE27

PE26

PE25

PE24

PE23

PE22

PE21

PE20

PE19

PE18

PE17

PE16

PE15

PE14

PE13

PE12

PEll

PEO10

PE9

PE8

PE7

PE6

PE5

PE4

PE3

PE2

PE1

PA6

PA5

PA4

PA3

PA2

PAl

0.00 10.00 20.00 30.00 40.00 50.00 60.00

Percentage of men who weave the design CA/TU, same informants


S2007
O 2002


70.00 80.00


Figure 8-16. Distribution of knowledge on designs among men from Capivara and Tuiarare
villages interviewed in 2002 and 2007.













* >25


* 21 to 25
E 16 to 20
l 11 to 15
E 6 to 10
S1 to 5


0.00


5.00


10.00


15.00


20.00


25.00


% of design weavers


Figure 8-17. Percentage of men who weave different designs in each village, grouped by classes.


I


I













Tewit
Makupa
Myau'i
Tare'i
Tare'a
Jepyk
Aturi
Jywapan
Towaju'i
Valdir
Kupeap
Moacir
Joao
Elenildo
Sirawejup
Simao
Tamakari
Arlindo
Juru
Jemy
Eroit
Perun
Corone
Tafut
Masi'a
Canisio
Xupe
Raimund
Chico


S0


10.464
0.381
0.381
0.376
0.371
0.344
0.319
0.273
I 0.262
0.222
I 0.147
0.14
0.09
0.046


0.:


'0.57
0.572
.556
.556
541


0.886
0.879


0.847


0. 27
0.807
0.775
0728


S0.657
0.655
0.625
8


SCapivara

STuiarard


W Rio dos Peixes


Teles Pires


0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Competence scores
basketry designs names



Figure 8-18. Competence scores on consensus analysis regarding agreement on naming the
different basketry designs among men in four Kaiabi villages. CA- Capivara (red);
TU-Tuiarare (blue); RP Rio dos Peixes (orange); and KU Kururuzinho (green-
Teles Pires).


]


2


0






















# Pemn


* Jun


K05 upeap '
-05 Towau
Towaju'wi '' i
1.


* Chlic


SMasia -1


Tamaan
Srawejup

* T .pan
05




Jemy


* Xupe


Figure 8-19. Similarity between men from Capivara (red) and Tuiarare (blue) villages regarding

names given to basketry designs. Non-metric multidimensional scaling (MDS)

analysis of similarity.


I I












z.-.


* Canisio


2


1.5


1


0.5


Elenildo *
rn


Raimundo







* Eroit

Arlindo


Joao Valdir


S* Moacir
.5 -1 -0.5 0 0.5 1 1
-0.5 Tafut
Corone Simao
-1


Figure 8-20. Similarity between men from Rio dos Peixes (orange) and Kururuzinho (green)
villages regarding names given to basketry designs. Non-metric multidimensional
scaling (MDS) analysis of similarity.










w
r
r
r


* % of weavers
0 % of women interviewed


TE9
TE8
TE7
TE6
TE5
TE4
TE3
TE2
TEl


0.00 20.00 40.00 60.00 80.00 100.00
% of women

Figure 8-21. Distribution of knowledge on textile designs among Kaiabi women in four villages.


I~I











Kunharop 0.953
Aruti 0.953
Reaju'i 0.953
Jaupi 0.953
Moreru 0.953
Jakap 0.953
Zulmira 0.774
Katue'i 0.774
Jacira 0.73
Juwi 0.73
Diana 0.72
Mytang 0.715
Rywapo 0.715
Moete 0.715
Kunha'em 0.715
Cunhaete 0.715
Rywujan 0.715Capivara
Mytanip 0.715
reajup 0.661 Tuiarar
Morea'i 0.661 Tuiarar
Poit 0.647
Kyryma 0.647 Rio dos Peixes
Ryw Tru'i 0.647
Katu 0.647
Rywetejup 0.647 Teles Pires
Kwasi 0.647
Pefuku 0.647
Reaju 0.478
Jemo 0.478
Juwytang 0.478
Iru 0.478
Katuryp ,0.415
Josiane 0.267
Kujareajup 0.024

0 0.2 0.4 0.6 0.8 1

Competence scores names textile designs

Figure 8-22. Competence scores on consensus analysis regarding agreement on naming the
different textile designs among women in four Kaiabi villages. CA- Capivara (red);
TU-Tuiarar6 (blue); RP Rio dos Peixes (orange); and KU Kururuzinho (green-
Teles Pires).















N nft Iniuarytun rnenw&svu


a



o 0
O

Ca
a '












A


Kr iwldp a? ram ror ukhiby de2NF
C C











,C 0 0 D ,0 .--






B
I '*j s ar r larl









,T jE tJ it!


SB
UD ct Wnlgr
CQUQ n








P I -

*L


U^- C rln


Rsq=0.352


Ou


Rsq=0.086


-Lnr


Rsq=0.068


Figure 8-23. Linear regressions showing relationship between age and basketry knowledge

among Kaiabi men: A) Relationship between age and number of basketry items

woven by Kaiabi men; B) Knowledge of names for basketry designs; C) Ability to

weave (or use) of basketry designs.


. & ,
@1 a o D
a3 4a M M
X^ T^ _
s *] Li j '









Table 8-7. Results of logistical regression analysis for selected socio-economic variables, having
weaving knowledge as the dependent variable.
Variables/parameters B S.E. Wald df Sig. Exp(B)
Step 1 Schooling -0.964 0.372 6.735 1 0.009 0.381
Language Proficiency 1.962 0.427 21.092 1 0.001 7.116
Stability of income -0.193 0.400 0.233 1 0.629 0.824
Age class -0.236 0.265 0.793 1 0.373 0.790
Gender -0.766 0.322 5.664 1 0.017 0.465
Constant 0.977 0.836 1.365 1 0.243 2.656


























* Participation in political meetings
O Interethnic mixing
* Formal Schooling (elementary)
O Ability to weave (yes)
* Language proficiency (proficient)
* Stability of income (stable)


20.00 40.00 60.00 80.00 100.00 120.00


Figure 8-24. A comparison of selected socioeconomic variables, including language proficiency
and weaving knowledge (ability to weave) between four Kaiabi villages. CA-
Capivara; TU- Tuiarare; RP- Rio dos Peixes; TP- Teles Pires.


I













PDsitive correlations


We -avng aw]I dgt(gEen ) cB


Lanmga pearfleicy


AmbiguousCombined
Relatiorihips


Negative correlations





















Leanpage profic ency






















Weaving Inowledgg (dPi3 s)


Figure 8-25. A model for understanding interactions among variables related to Kaiabi language

proficiency and weaving knowledge.


~~ClaHertonmlute it
W]wd

CubiadbJ,
m'dtD











Huda~l]nnn3.uiwt

(tan4 tulh ;an 4he atrU
Imovericed win
Ub~iVS \.

iniit;* g^\\^^d/r^ _, wem


I~~F 4Mlda eii^(^


t aatOijicil ufiea
| llbL~tlfli~ti~tll / l-^-
Iai t^ '~^f


]reilin
9VItrm









CHAPTER 9
CONCLUSIONS: A SYSTEMS APPROACH TO THE STUDY OF CULTURAL CHANGE

This research represents a testimony to the incredible capacity that indigenous peoples

have of recovering from shock and adapting to novel and unexpected situations. It brings

contributions to the interdisciplinary theoretical fields of socio-ecological systems, displacement

and place making, political ecology, development studies and cognitive anthropology (dynamics

of indigenous knowledge systems). It also contributes to the development of public policies and

frameworks for implementation of development and conservation programs in indigenous lands

in the Amazon.

I start this chapter with a general comment on the systems approach applied in this

research, describing factors that might enhance or constrain cultural and environmental resilience

among the three Kaiabi groups and which might be applied to other indigenous societies. Then, I

summarize main conclusions, theoretical contributions, answers to specific research questions,

and testing of hypotheses in selected sub-sections. Following, I include a section on policy

recommendations, finalizing with suggesting future steps that might be taken as continuity or

deepening of this research.

Collapse and Renewal: Displacement, Socio-Ecological Systems and the Kaiabi

A systems approach to the study of the dynamics of indigenous knowledge systems

provides elements that allow a better understanding of mechanisms involved in knowledge

creation, transmission, erosion and change. Indigenous knowledge exists within a socio-

ecological system which may be a negative force, but also may provide opportunities for its

simultaneous innovation and continuity. Resilience is thus an appropriate concept to refer to the

capacity of indigenous peoples to adapt to change, innovate and re-create their identity, territory

and knowledge without losing their essence as a group. It has been like this since the world came









to be, since our human ancestors started to inhabit the land. The most adaptable organisms have a

better chance to survive. However, favorable environmental conditions might enhance the

adaptation process. On the contrary, harsh external conditions might lead to the extinction even

of the most resilient species.

In the introductory chapter of this document, I presented a systems approach to the study

of indigenous knowledge systems, considering historic, ethnographic, environmental, political,

socio-economic and cultural aspects. I presented some graphic representations showing

components of the system, including an example of an adaptive cycle (Figure 1-3). After

analyzing each of the selected components of the system and some of their numerous possible

interactions, I developed a model of adaptive cycles applied to the Kaiabi history (Figure 9-1).

The displacement event triggered a big collapse in the adaptive cycle curve, leading to different

reorganization, exploitation and conservation phases that each Kaiabi group has gone through in

their adaptation after the crisis or collapse event.

According to socio-ecological systems theory (Gunderson and Holling, 2002), in

moments of crises or collapse uncertainty and unpredictability are high, and control is weak and

confused, but space is created for reorganization and innovation. Oliver-Smith (2002) comments

that resistance after displacement might give the resisting local community valuable experience

in dealing with political structures, enabling them to acquire useful allies. Resistance might

generate a sense of cohesion and identity among the society, enabling leadership and formation

of new institutions which might be used for social mobilization and future action.

Each of the three Kaiabi groups followed different reorganization and exploitation

phases, in a constant interaction between structural context and agency or decision-making

processes which led to various outcomes. One important conclusion based on the Kaiabi case,









but which can be extended to other indigenous peoples, is that changes in knowledge and broadly

in culture are not unidirectional or permanent or irreversible, as long as the loss of cultural

resilience is not severe enough to change the system permanently. For example, the initiative

taken by the Kaiabi of teaching weaving to the Teles Pires relatives reversed the trend of

knowledge erosion among them, transforming a tendency of collapse and loss of resilience into a

phase of reorganization and exploitation, since new knowledge and innovation is being

generated, opening possibilities for this knowledge to be spread within the population, reaching a

conservation stage. Making a parallel with ecological systems, as I mentioned in the

introduction, when any ecosystem faces a change or a collapse event (induced or not), such as a

fire, a flood or a deforestation, if it is given the opportunity and stimulus to recuperate, it might

recover its appearance and function, but it would never be exactly the way it was before, just as

even the "before" was already in a dynamic process of change. A social system presents a

comparable behavior in which knowledge that is being lost might be recovered, but under new

circumstances, thus never reaching an "optimal" stage because there is no such thing. It prevails,

but it changes. There is a Kaiabi expression which I loved to hear when asking about plant

species during fieldwork that is perfect to express this apparent contradiction: "E igual, mas e

diferente." It's the same, but it's different.

Indigenous knowledge systems are immersed in dynamic adaptive cycles interacting with

several other factors that might lead to their innovation, transformation and prevalence or to their

non-renewal and erosion. The process of erosion or loss can be slow enough to allow for

reversion and transformation by an interaction of internal and external factors before the

resilience is lost and the system is transformed. Under this perspective, the Rio dos Peixes group

is in a conservation phase where the native language and traditional knowledge are being lost.









This trend could be changed, but the extent to which language has been lost among them makes

its recovery a difficult task, that would require a lot of effort. It would be even more difficult to

reverse the trend of weaving knowledge erosion in Rio dos Peixes, where few youth are willing

to learn their parents' culture, than in Teles Pires, where there is more capacity installed and

interest for it. Just like an ecosystem that suffers a severe impact would need a lot of work and

incentives (labor, technological, financial) to be recovered, the extent to which knowledge is lost,

and the lack of will or interest, can turn prospects for its renovation into a complicated mission.

How long it might take for the knowledge to be permanently lost is unknown.

Policies and development projects might contribute to change social systems' adaptive

cycles, reversing trends leading to loss of knowledge, power and territorial control. The

environmental movement (allied with other factors described below) came to affect the Xingu

Kaiabi during an exploitative phase of their cycle, bringing incentives and opportunities for their

empowerment and achievement of self-determination in a new pluricultural, highly politicized

and contested territory. In Teles Pires, struggles over land and alliances with the Xingu relatives

and other agencies (such as the PPTAL program, see Chapter 4) have led to greater

empowerment and cultural resilience in comparison to Rio dos Peixes. Below, I summarize the

main factors that have enabled or constrained cultural resilience among the three Kaiabi groups.

More than forty years after the collapse and relocation to Xingu, Kaiabi society there has

flourished demographically, culturally and politically. The social skills learned by Kaiabi men in

the extractive frontier of Tapaj6s subsequently proved to be a good political asset in the social

environment of Xingu, allowing them to position themselves as key mediators with the outside

and as agents for indigenous governance (Athayde et al, 2009). The conditions found by the

displaced Kaiabi at Xingu Indigenous Park were in many ways exceptional: indigenous groups









there enjoy a privileged combination of a large and secure territory, an unusual degree of relief

from the pressures of colonization and modernization and, on the other hand, sizeable

interventions in the form of health services, financial and institutional assistance from state and

non-state agents. In effect, the modernization process in Xingu is filtered through the institutions

and social dynamics of indigenous and environmental politics (Menezes 2000; Senra et al. 2004;

Athayde et al., 2009). The condition of the Kaiabi as 'outsiders' in a social and political

environment where 'tradition' and indigeneity have come to constitute valued social and political

assets, has encouraged a process of cultural revitalization and affirmation of ethnic identity

through social institutions such as basketry and crop stewardship. The funding opportunities and

support coming from NGOs and governmental environmental agencies contributed to the

cultural revitalization processes, projects and initiatives in which they have been engaged in the

last ten years. As a result, the social outcomes faced by Xingu Kaiabi contrast markedly with

those who remained in the Tapaj6s, who have not been as successful in rebuilding indigenous

institutions amidst the more intense and disruptive intrusions from the state and the extractive

frontier. At the same time, the agency and capacity of the Xingu Kaiabi to access and use the

resources in their favour has been another important asset in the process of their empowerment

and capacity of self-determination.

Looking at the Xingu Kaiabi as an example of successful adaptation even with all the

threats and contradictions that have accompanied their trajectory I conclude that geographical

isolation, adequate health assistance and availability of financial and technical support, combined

with leadership development as well as with the capacity of innovation and of assimilating

western institutions, are the factors that led to greater cultural resilience in Xingu in contrast with

the other two areas. I consider that the capacity to successfully intertwine indigenous and









western knowledge and institutions is crucial to ensure the continuity of indigenous societies

nowadays (Gray, 1997).

The process of weaving together both indigenous and western cultures requires

innovation, but does not happen without conflict and contradictions. The isolation of the

displaced group in Xingu Park did not happen by choice, but by the policies implemented by the

Brazilian government regarding the creation of Xingu Park as a perceived vision of a paradise

where Indians and nature should coexist in a timeless and unchanged manner. Thus the Xingu

Kaibi became trapped in a romanticized view of what indigenous peoples should be and, more

critically, in a paternalistic relationship with the state. This isolation has been both beneficial and

constraining to the Xingu group. On one hand it has allowed the resilience of knowledge systems

and natural resource management practices; on the other it has caused increased dependence on

western (national and international) institutions and politics. The creation and development of

political organizations such as ATIX enabled freedom from the State, becoming instruments for

governance, territorial control, cultural resilience and self-determination. However, these

organizations have also brought internal conflicts, social change and financial dependence on

international funding (Albert, 2000, see Chapter 6). To what extent government institutions and

NGOs represent the needs and aspirations of Amazonian indigenous peoples is a question that

bears further examination (Gallois, 1998; Nugent, 2006).

In Teles Pires and Rio dos Peixes, the initial shock caused by the displacement was

aggravated in subsequent stages by the lack of access to qualified financial and technical support

and by the easier access to both labor opportunities and industrialized food and products.

Interestingly enough, it is in Rio dos Peixes which is closer to a city in comparison to Xingu or

Teles Pires that the process of loss of cultural resilience is moving at a faster pace. The









proximity to urban centers means both geographical and cultural proximity to markets, schools,

churches, non-indigenous values, language, and lifestyle. Thus, geographical and cultural

proximity to urban centers is a factor that causes loss of cultural resilience and thus erosion of

indigenous knowledge systems. The initial lack of options caused the Rio dos Peixes group to

sell valuable natural resources (mahogany wood, see Chapter 5) compromising the

environmental resilience of their territory. The amalgamation with western institutions such as

church, school and markets meant a replacement of cultural values and traditions among the Rio

dos Peixes group, instead of a co-existence or intertwining of both indigenous and western

cultures such as in Xingu. Therefore, in Rio dos Peixes, the Kaiabi culture has been replaced by

the cultural system of the non-indigenous Brazilian society. Native language and weaving

knowledge are seriously compromised in Rio dos Peixes, even with the current and constant

exchanges between Xingu (mainly Capivara village) and Rio dos Peixes residents through

kinship.

In Teles Pires, the relative geographical isolation from Alta Floresta and Jacareacanga

(closest municipalities) has helped to keep Kaiabi traditions alive, as well as the absense of a

western school in the village and lesser interaction with the market in comparison to Rio dos

Peixes. Considering that this group was severely impacted by the displacement event, even more

than the Rio dos Peixes group, it is curious to observe how resilient they have become. Other

factors that have helped to prevent higher degrees of knowledge loss during the exploitation and

current conservation phases were land struggles and the closer and constant interaction with the

Xingu group through kinship alliances. On the other hand, interethnic mixing has led to cultural

change and loss of proficiency in the Kaiabi language both in Teles Pires and Rio dos Peixes, but

more so in Teles Pires, due to frequent marriages of Kaiabi women with Apiaka and Munduruku









men. This contributes to the use of Portuguese as the main language to the detriment of the

Kaiabi language, and to weaken cultural traditions. Nevertheless, the possibility to show their

traditions and sell handicrafts to visitors and tourists that come to visit the village has helped to

conserve artistic knowledge among the Teles Pires group.

In the next sections, I briefly summarize the main contributions of this research to

specific theoretical fields, as well as review arguments explored and hypotheses tested.

Indigenous Knowledge, Displacement and Place

Comparable to other Amazonian indigenous peoples, the Kaiabi used to be very mobile

before and after the contact with westerners (Alexiades, 2009). They have gone through several

processes of re-territorialization through contact with diverse agents, from warfare with other

indigenous groups, conflicts with rubber tappers, trappers, loggers, miners, and ranchers, to

interaction with governmental agents and policies (Oakdale, 1996). These contacts affected the

Kaiabi in different ways through conflicts and exchange of knowledge, skills and labor. The

process of sedentarization of villages that is occurring in the three Kaiabi lands as well as in

many other indigenous lands in the Amazon brings important questions and challenges related

to the sustainability of natural resources and consequently to the resilience of indigenous lifestyle

and management practices (Athayde et al, 2009).

None of the Kaiabi groups live today in what was their traditional territory in Tapaj6s

watershed, whose boundaries were already fluid and contested by friction between them and

other indigenous groups even before the contact with rubber tappers and other westerners. The

Rio dos Peixes group is the one that lives closer to the "heart" of Kaiabi territory in Bateldo

River, currently under claim and contestation by ranchers (see Chapter 4). Unfortunately, roads,

ranches and towns were placed in what used to be traditional Kaiabi lands and where, as they

say, "nossos av6s estdo enterrados our grandparents are buried."









The displacement meant the loss of access to important resources for the Kaiabi, such as

the Castanha-do-Pard (Brazil nut), agai palm, arumd used in weaving, siriva (palm which wood

is used for bows) and others. These resources are still used and procured by the Kaiabi, who

frequently travel to their ancestral territory to gather them and bring to Xingu. In this respect, the

loss of access to strategic resources due to displacement, land demarcation or deforestation is

another factor that might interfere in the resilience of indigenous knowledge and management

practices. The loss of access to arumd, the main fiber used in basketry weaving, is one important

constraint in the continuity of weaving designed baskets among the Kaiabi men, even with the

use of substitutes.

In Xingu, the resettled Kaiabi have developed at least seven different adaptation

mechanisms to cope with the environmental constraints faced at Xingu after the transfer (Chapter

5): 1) knowledge innovation or creation, through new names given to ecological zones and/or

resources; 2) increase in diversity of resources used for different purposes (e.g. to build canoes)

due to village sedentarization and scarcity of important forest resources; 3) conscious protection,

multiplication and development of new crop varieties, enhanced by community-based initiatives

such as the Munuwi project (see Chapter 6); 4) substitution of strategic resources by others of

similar quality, such as what happens with the "arumd" plant used in basketry (Chapter 8); 5)

travelling to ancestral land or to other areas to collect strategic resources; 6) exchange of crop

varieties or saplings of strategic resources among the three groups; and 7) development of

management practices for specific resources, through institutional support and collaboration

between western and traditional knowledge (such as the "aruma" and some fruit species present

in the ancestral land). Interestingly, two of these practices include intertwining and/or

collaboration between indigenous and western science ecological knowledge.









The process of adaptation and re-adaptation of the different Kaiabi groups after the

displacement can be compared to the writing of a new chapter in the history of the group, which

resulted in the concomitant making of new territories, identities and places, but not without

conflict and resistance (Gupta and Fergusson, 1992). For instance, the Kaiabi inhabited and used

for many years an area close to the Arraias River, located in the North-western portion of Xingu

Park, in the process of being claimed (see Chapter 4). In spite of not being considered an

ancestral land, it is a land where the Kaiabi inscribed history (Santos-Granero, 2005), and which

has many resources of strategic importance for the group, such as the Brazil nut and the aruma.

Thus, the traditional use of the three areas being claimed by the Kaiabi, as well as the need for

strategic resources in them, would provide reason enough for their recognition as indigenous

lands according to the Brazilian Constitution of 1988 (Ramos, 1998, see Chapter 4).

The study of movements of resistance, adaptation, innovation and empowerment of

indigenous peoples towards regaining rights over their lands might be a fertile field for exploring

theories behind the responses of people to development-forced displacement and resettlement

(Oliver-Smith, 2009). The study of the Kaiabi case brings some implications for DFRD theory.

First of all, it illustrates the innovative responses people might develop after a displacement

event, such as in terms of resistance to assimilation, development of social movements, renewed

identities, and innovations in indigenous knowledge systems.

The Kaiabi after-displacement history fit well in the four-stage model presented by

Scudder (1981, 2009): the Kaiabi experienced the collapse event, coped with the initial impact,

reorganized socially, politically and economically and were able to hand over a sustainable

resettlement process to the second generation of resettlers. Scudder (2009:31) stated that he

developed his theory based on the hypothesis that "regardless of cultural or environmental









setting, the majority of resettlers will take advantage of appropriate opportunities for improving

their livelihoods, provided that appropriate assistance is available to enable them to benefit from

the opportunities." That's exactly what happened with the displaced Xingu Kaiabi, whose second

generation already embraces the "xinguano identity." The other two groups can also be

considered displaced, since they were pushed to settle in areas that are not located in the heart of

their traditional territory, both in the case of Rio dos Peixes and Kururuzinho. In Rio dos Peixes,

the lack of support and options led them to embark on an assimilation process. In Kururuzinho,

the struggle for land rights has helped to keep their culture and identity alive.

As discussed in the introduction, the Kaiabi case shows the relativity of the "rootedness"

of indigenous knowledge, also named "local knowledge" by some authors (Hunn, 1999; Ellen

and Harris, 2000). Adaptive mechanisms entailed the genesis of new knowledge and

management practices, in the domains of both ecological and weaving knowledge. In

consonance with previous authors, this analysis calls for the need to expand the

conceptualisation of indigenous knowledge as adaptive and fluid (Berkes et al., 2000; Ellen and

Harris, 2000). Place and identity making are thus intertwined and in constant movement. I

conclude that indigenous knowledge might be mobile if resilient and more rooted in ethnicity

than in land.

Empowerment, Western Institutions and Territorial Control

In the Amazon, the concomitant development, around the 1970s, of the indigenous

grassroots movement, the environmentalist movement and the writing of the new Brazilian

constitution of 1988 provided an array of opportunities for indigenous peoples to access financial

and technical resources available for conservation and development projects (Conklin and

Graham, 1995; Little, 2001; Albert, 2000; 2005). However, these opportunities were not

available to everyone, but restricted to groups who have showed the ability to, first, frame their









discourses according to expectations of national and international political instances, institutions

and funding agencies; and second, provide social capital and leadership needed to access and

appropriate of these resources (Turner, 1991; Fisher, 1994; Jackson, 1994; Sahlins, 1997;

Oakdale, 2004). In the case of the Kaiabi, leadership formation occurred mostly through previous

experience with rubber tappers, involvement in Xingu Park's administration, interaction with

other Xingu groups, and through participation in political struggles at national levels, in which

they interacted with other political leaders, notably with the Kayap6. Interestingly enough,

leadership formation among the other two Kaiabi groups has been inspired by interaction with

Xingu political leaders such as Mairaw6 and others. The formation of national and international

networks constituted by indigenous and forest peoples has been an important outcome of

reconciling social and environmental movements in the Amazon and elsewhere.

The three Kaiabi groups have created political associations to represent communities'

interests, develop projects, and manage their territories. With the retraction of the paternalistic

role played by the State, indigenous organizations have played an increasingly important role in

managing indigenous lands in the Amazon (Albert, 2005). It was through the leadership of the

Xingu Kaiabi, with support from international (Norwegian Rainforest Foundation) and national

(Instituto Socioambiental) environmental agencies, that the first program for patrolling

indigenous lands in the Amazon was developed and administered by a local indigenous

organization, in this case, ATIX. The role of controlling their own territory conferred visibility to

ATIX among Kaiabi communities and other ethnic groups that inhabit Xingu Park, as well as

enabled the development of a sense of responsibility and agency among the Indians, who used to

depend on the paternalistic hand of the State.









This research showed that women have had poor participation and access to information

regarding the work and role performed by indigenous associations across three Kaiabi lands.

Overall, political participation by women is greater in Teles Pires in comparison to Xingu and

Rio dos Peixes. This is the result of tighter involvement with land struggles in Kururuzinho

village. In Rio dos Peixes, people are less aware of the role of the local association due to its

precarious condition and lack of both community and outside support. In general, people

mentioned the role of indigenous associations in developing projects. This relates to the "projects

market" that has sustained indigenous territorial control and management in the Amazon since

the 1980s (Albert, 2000). In Xingu, people mentioned the role of territorial patrolling that ATIX

plays in Xingu Park. People cited four main projects that they would like to be developed in their

villages: 1) handicrafts; 2) traditional crops; 3) organization of Brazil nut commercialization (Rio

dos Peixes); and 4) beekeeping (Teles Pires and Rio dos Peixes). I will comment on these aspects

in the following section on policy recommendations.

I hereby reiterate my argument that political empowerment, agency and self-

determination in a multi-cultural environment have led to a better territorial control among the

Xingu Kaiabi. In contrast, the lack of access to opportunities and resources, combined with

external pressures, weak leadership and control, have led to environmental degradation in the

two other areas. As much as this is an inspiring case, we should be aware that empowerment and

assimilation of new institutions among indigenous peoples often result in conflict, vulnerability

and mis-representation of communities' interests. Internal conflicts arise from clashes between

traditional and new forms of power: among leaders, chiefs and communities; among different

groups that share a common territory (such as in Xingu); and between old and new generations

(Oakdale, 1996; 2005; Rosengren, 2003; Keckenberger, 2005). The new forms of representation,









governance and political empowerment that indigenous peoples have developed within the

context of social movements and national and international politics for conservation and

development provide a fertile ground for further research.

Processual Studies on Indigenous Knowledge Systems

This research involved two main interrelated fields of inquiry: a study of the dynamics of

indigenous knowledge among women and men in four villages (mechanisms of distribution,

transmission, erosion and innovation of IK); and an exploratory analysis of the relationships and

interactions between socio-economic factors and indigenous knowledge. Quantitative and

qualitative data collection and analysis were used in the development of a processual study of

weaving knowledge among men and women in a time interval of five years in Xingu Park and

three years in Kururuzinho village in Teles Pires.

The main contributions of this research to theoretical fields of inquiry regarding

indigenous knowledge systems are:

1. Review and adaptation of models exploring mechanisms of knowledge transmission in
indigenous societies;

2. Evaluation of the role and effects of western institutions and community-based projects
on indigenous knowledge systems;

3. Better understanding of patterns of knowledge transmission, retention and erosion when
considering specialized knowledge domains;

4. Exploration of gender relationships in mechanisms of knowledge transmission and
distribution;

5. Adoption of a systems approach to the study of indigenous knowledge as embedded in a
bigger complex socio-ecological system.

Weaving knowledge of designed basketry and textiles can be considered specialized

knowledge, one which is not necessarily shared by all individuals in a population. This domain

was approached as a sub-system embedded in a broader socio-ecological system in which









historic, territorial, environmental and political factors interact at different scales and speeds.

Weaving knowledge was divided into five sub-domains or modules, namely form (techniques),

function (uses, including production for the market), materials (natural resources used in

basketry and textile weaving), meaning (social meaning, spiritual meaning), and designs

(knowledge of different graphic designs woven in baskets and textiles (Ellen, 2009, my

adaptation).

Complementarily, I collected socio-economic data used to build a profile of each of the

four villages and to explore relationships and test hypotheses involving selected socio-economic

variables and weaving knowledge. The variables were selected in order to study variations in the

distribution and transmission of weaving knowledge among the villages according to factors

such as age, gender, proficiency in the Kaiabi language, schooling and market integration.

Language proficiency has been often used as a proxy for cultural diversity and vitality of

indigenous knowledge systems (Maffi, 2001, Zent and Maffi, 2007). In this study, I also wanted

to compare the statistical behaviour of language proficiency to that of weaving knowledge in

relation to the other variables, to verify to what extent one could be used as proxy or as an

indicator about the status of the other.

Through this processual study, it was possible to register the changes that have occurred

in a cultural domain as a result of political empowerment and thus greater access to financial and

technical resources by the Xingu Kaiabi. The development of a community-based project (Kaiabi

Araa) for cultural revitalization of basketry and textiles weaving among men and women in

Xingu and in Kururuzinho village in Teles Pires has triggered changes in knowledge distribution

and transmission among the communities involved.









For this research, I established five hypotheses related to the patterned intracultural

distribution of IK and to the effect of western institutions (school, market) on IK (Chapter 2), as

follows:

* HI: Elders retain deeper and different knowledge when compared to younger people;

* H2: Men and women use different mechanisms in knowledge creation and transmission;

* H3: Greater levels of market integration lead to erosion of weaving knowledge across three
Kaiabi groups;

* H4: Formal schooling erodes indigenous knowledge: higher levels of formal schooling
leads to lower levels of weaving knowledge among men and women;

* H5: Role of community-based projects: the Kaiabi Araa project is responsible for an
increased number of basketry and textiles weavers among the younger generations.

I registered that weaving knowledge has been both innovated and eroded among the

Kaiabi as a group. While in Xingu and in Teles Pires knowledge has been at the same time

maintained and re-created, in Rio dos Peixes it has been eroded and can be considered threatened

with extinction, along with Kaiabi native language. Interestingly enough, communication and

interactions between the villages, mostly through kinship, have enabled the transfer of

knowledge between residents in the three areas in unexpected ways. For instance, some men in

Rio dos Peixes have learned from the book Kaiabi Araa (which contains a photographic

catalogue of designs) brought there by a Xingu relative.

The testing of hypotheses resulted to acceptance of H1, H4 and H5 and rejection of H2

and H3. I accepted H1, that older men and women are more proficient in Kaiabi language and do

hold greater knowledge on weaving designs when compared to youth. However, there seem to be

important differences between general and specialized knowledge domains among the Kaiabi.

Age is more important as a determinant of language proficiency than as influencing degree of

knowledge on designed baskets and textiles. Furthermore, general knowledge on different types









of baskets is more influenced by age than specialized knowledge on uses or names for designs.

This result is also indicative of changes in mechanisms of learning and transmitting brought by

the Kaiabi Araa project and the renewed notion of cultural pride and status attached to basketry

weaving in Xingu and Teles Pires. Nowadays, younger men might also become knowledgeable

in the art of weaving, in contrast with the decades immediately following the diaspora.

The statistical tests run for language proficiency and weaving knowledge revealed no

significant influence of gender on the variables that indicate robustness of indigenous

knowledge. In spite of rejecting H2 due to lack of statistical significance, I found important

differences among men and women in different villages, regarding socio-economic profile and

mechanisms of knowledge transmission. Women in Rio dos Peixes show less proficiency in

Kaiabi language and higher attendance at formal schooling in comparison to men. Weaving

knowledge among Rio dos Peixes women is kept in the hands of few elders and among women

who have lived and learned in Xingu. The lack of young people learning in Rio dos Peixes is

another indicator of knowledge erosion and loss of cultural resilience in contrast to both Xingu

and Teles Pires.

Qualitative data indicate that weaving knowledge among women is more embedded in

the household and kinship dynamics in comparison to basketry knowledge among men. In

addition, the development of new marketable objects such as woven bags by the women, along

with the use of industrialized cotton and the transfer of new designs from basketry to textiles

confer greater resilience to women's weaving knowledge in contrast to men's. Industrialized

cotton is easily bought in towns, in contrast to the scarce arumd preferred by men for weaving

designed baskets. Thus, I conclude that women's weaving knowledge may be more resilient than









men's, and that women play an important (and sometimes overlooked) role in keeping and

innovating knowledge in indigenous societies.

I expected that greater levels of market integration would lead to erosion of IK among the

Kaiabi. Related to this, I found, comparable to what other authors have found, that market

integration can produce ambiguous outcomes on IK among indigenous peoples (Godoy, 2001;

Godoy et al, 2005; Reyes-Garcia, 2001; 2005). Market integration (measured through level of

income stability among the interviewees) showed statistical significance in the logistic regression

model in which I had language proficiency as the dependent variable. Thus, greater levels of

income stability lead to less proficiency in the native language among the Kaiabi. This explicitly

refers to the situation in Rio dos Peixes, where there is more socioeconomic heterogeneity

through different access to waged income or to governmental programs for low income rural

families.

Despite showing significant influence in language proficiency, I found no significance for

market integration related to weaving knowledge, an thus I rejected H3 that greater levels of

market integration lead to erosion of weaving knowledge. The attempt to quantitatively measure

market integration and transform the diversity of possibilities of interaction with various market

instances among indigenous peoples in one single variable may also be a methodological

constraint. I found for the Kaiabi case that interaction with markets might bring both innovation

and erosion of indigenous knowledge, depending on the type of relationship and geographic

proximity to market opportunities. In predominantly subsistence-oriented economies such as in

Xingu, market interaction may lead to innovation of IK and, at the same time, to deepening of

knowledge related to marketable items (or plants) in detriment of others, just like it happens with

basketry and textiles (Godoy, 2001).









Level of formal schooling showed an opposite statistical behaviour to that of income

stability in the logistic regression models developed to explore the interactions between distinct

variables on language proficiency and weaving knowledge, considered dependent variables

indicative of cultural resilience. There was no significance for schooling in relation to language

proficiency. However, it was statistically correlated with weaving knowledge, in which greater

levels of formal schooling are linked to lower levels of weaving knowledge. I accepted H4, that

formal schooling erodes indigenous knowledge. However, just like market integration, we should

not consider the influence of western institutions as a one-way directional process. The type of

educational system developed in schools placed in indigenous lands might encourage or

constrain the continuity of indigenous language and traditions among the communities. In

Chapter 8, I presented some alternative models to represent the possible interactions between the

variables considered in this research. The situation in all villages is that people who have greater

access to formal schooling might also have other sources of income and thus less interest in

weaving.

Comparing the situation of weaving knowledge in Xingu and Teles Pires between 2002,

2004 and 2007, I found that there was a significant increase in designs among both men and

women in Tuiarare (Xingu) and Kururuzinho (Teles Pires), the participant villages. The Kaiabi

Araa project brought both revitalization and innovation of weaving knowledge among the

participants. I thus accepted H5, that there are more young people knowing how to weave

designed baskets and textiles after the development of the project in both villages. The outcomes

of this project bring important theoretical questions for the study of indigenous knowledge

systems. First of all, the communities developed a new way of transmitting knowledge in the

workshops of the project, through a mechanism which I named "many-to-many" or collaborative









learning. In this new mechanism, learning can happen between any two or more individuals

independent of their age or kinship tie (see Chapter 8 and Figure 8-14). Additionally, the Kaiabi

have copied and learned basketry designs from a book (Kaiabi Araa) containing photographs of

baskets taken in Kaiabi villages and national and international museums. The book serves as a

mnemonic instrument for the Kaiabi, in which more complex designs threatened with

disappearing have been produced again by more experienced weavers. Thus, the book is another

innovation that came from my partnership with the Kaiabi, resultant from the political

empowerment and alliances with the environmentalist movement, which enabled my long term

work and commitment with the group. It's interesting to note that similarly to innovations

produced solely for the market, transmission within western institutions (such as projects) may

permit innovative forms of transmission that would be otherwise unacceptable within the

traditional system of social organization. In Chapter 8, I reviewed models of cultural

transmission developed by Cavalli-Sforza and Feldman (1981), and proposed a model for the

study of knowledge transmission in a given cultural domain among indigenous peoples, based on

the Kaiabi case.

Exploring the sub-domain linked to the spiritual meaning of basketry weaving, I found

that shamanism, indigenous knowledge and diversity of plants and objects (in this case basketry

designs) are intrinsically related, and that many acts of creation, discovery and naming plants and

objects among indigenous peoples are developed through shamanism and dreaming (Guss, 1989;

Athayde et al, 2009; Silva, 2009). The main mechanism used for learning basketry designs that

of observation of ready made baskets (and lately of photographs contained in a book) -

reproduces a myth in which the ancestral hero Tuiarare first copied a basketry design from a skin

of a snake he had stolen from a snake's house. The mechanism of learning from others, often









enemies, or transforming others' knowledge and skills into something of their own is an

important feature of Tupi societies that is very present in Kaiabi social organization and learning

style until today (Viveiros de Castro, 1992). There are indicators that the most complex group of

designs of Kaiabi basketry, those depicting a mythical figure named ta'agap, were learned with

the Apiaka people, fierce enemies of the Kaiabi in the past.

Proficiency in the Kaiabi language and weaving knowledge are strongly correlated.

Language proficiency was the most important variable influencing weaving knowledge in the

logistic regression model including all the variables. A person who is proficient in the Kaiabi

language has 71.16% more chance of knowing how to weave designed baskets and textiles in

comparison to someone who is not proficient. I compared the statistical behaviour of these two

variables, to verify to which extent they could be used as proxies to indicate the status of IK

among a given indigenous group. I found that there are some differences concerning the

relationship between these and the other selected variables. In this case, given the fact that

weaving designs is specialized knowledge, I suggest that the differences found may not follow

the same trend if we consider a different knowledge domain. This is especially true when

considering the relationship between age and weaving knowledge in the linear regression

analysis, which was only significant for the number of different basketry items known by men in

all villages. I conclude that language proficiency may be used as a good indicator of cultural

resilience, but it should not be applied as a proxy to every knowledge domain, since those may

present their own specificities concerning interactions with other factors.

The results support the argument that there may be differences in patterns of distribution

and transmission of knowledge between specialized knowledge domains (e.g. designed baskets)

in contrast to other more general domains, such as basketry weaving in general and crop









knowledge, for instance. Regarding knowledge related to crop plants such as cassava and peanut,

other authors have found that age, kinship and gender are important determinants and follow

certain patterns: older women in the same kin group are the ones that show more agreement and

deeper knowledge on cassava and peanut varieties (Boster, 1986; Silva, 2009). In this study, I

found that gender and kinship are not expressively influential in the distribution of weaving

knowledge, and that young and middle-aged people might also retain comprehensive knowledge

of designs. I suggest that studies on indigenous knowledge systems cannot ignore the role that

western institutions, including markets, schools, projects and associations might be playing in

changing patterns of knowledge distribution and transmission.

The main factors that have enabled innovation and resilience of weaving knowledge in

Xingu are: 1) type of market integration existent there, in which handicrafts are still the main

source of family income among Kaiabi communities; 2) the process of political empowerment

and consequent development of community-based projects targeted to cultural revitalization; 3)

the leadership and persistence of Aturi Kaiabi, who coordinated the project; and 4) the adoption

of basketry designs as symbol of Kaiabi identity, including their transfer to other objects such as

textiles, wooden benches, books and even body painting; and 5) the use of substitutes for the

arumd, main plant fibre used in basketry weaving. I argue that innovation is a key factor in

enabling the resilience of indigenous knowledge systems in times of globalization and increased

participation in and assimilation of western institutions. Basketry and the complex and diverse

designs associated with the Kaiabi, has been transformed in the context of twentieth century

politics of identity from a symbol of individual prestige specifically attached to the weaver and

his family in the context of community social dynamics- to a collective one, attached to a larger

collectivity and in the context of social, political and market dynamics (Athayde et al, 2009).









Applications and Policy Recommendations

This research showed that the most relevant factors enabling adaptation, innovation and

cultural resilience among the Kaiabi are leadership, agency and self-determination combined

with long term access to financial and technical resources brought by the alliance between

indigenous peoples and environmentalism. Thus, an important role to be played by practitioners,

researchers and policy-makers would be to support and facilitate the process of training and

political empowerment of indigenous peoples, so that they are able to run their own associations

and projects according to their agenda and interests. A sense of ownership is critical in the

success of any initiative among indigenous peoples: what comes from the outside is someone

else's project; what is developed from within is their project. This does not mean that they won't

depend on external support in the form of projects, financial and technical resources, but that we

should work under the premise of transfer of power to the hands of indigenous peoples, as much

as possible. The construction of platforms that enable indigenous peoples to learn from each

other's experiences in programs of territorial and project management should also be supported

and enhanced by funding agencies and other institutions.

This study presents practical implications for development and conservation programs to

be implemented in the Amazon. I suggest that more flexibility could be given to community-

based initiatives, who might want to prioritize cultural revitalization processes over strict

environmental management and conservation actions (Pilgrim et al, 2009). In the long term and

at the end of the road, funds are better applied if they are relevant to indigenous people's

objectives, in which environmental conservation may be a consequence of empowerment,

territorial control and cultural pride (Colchester, 2000). The process of empowerment that comes

from planning, collaboration and execution of a community-based project, often overlooked, may

be more valuable that its practical outcomes.









This research has potential applications in processes of land claims and struggles in

which the three Kaiabi groups are involved, as well as in the development of partnerships

between the three groups in running their associations and projects and in the process of

establishing control over their territories. Efforts could be directed to support a greater

connection between Xingu leaders and the Rio dos Peixes community, maybe through the

development of projects of common interest, such as the involvement of both Xingu and Teles

Pires groups around the Kaiabi Araa project. The closer interaction between these groups could

help the Rio dos Peixes group to improve the organization and capability of Itaok association to

run projects of the communities' interest, such as commercialization of Brazil nut, handicrafts or

beekeeping. Partnerships between the Xingu group, ATIX and Kawaip should also be

encouraged, and a project on beekeeping for the Teles Pires area could be developed in

partnership or under the supervision of some Kaiabi leaders.

Another recommendation is that more attention and opportunities should be given to

women to access information concerning the role of political organizations, as well as on

decision-making processes regarding project development or territorial management. There

should be more efforts to have women as leaders in political arenas.

Efforts could be directed to promote the revitalization of Kaiabi language and knowledge

in indigenous schools, as is being done in Xingu. This is especially true for Rio dos Peixes, in

which the school could play a role in valuing the Kaiabi language among the youth. Educational

books produced for Xingu indigenous schools could be used in Rio dos Peixes, and the teachers

could have the opportunity to interact more with Xingu teachers. Joint events such as handicraft

workshops and fairs could be promoted linking the three groups, enhancing platforms for joint

learning and action.









Future Steps

In this research, I developed some models and approaches to indigenous knowledge

systems that could be applied to the study of other knowledge domains in order to compare

similarities and differences that might exist among domains in relation to socio-economic,

political and environmental factors. This would provide a better understanding of contrasts and

tendencies among more general or specialized knowledge domains within a given cultural group

or among groups. I suggest that a more in-depth comparison between the domains of weaving

knowledge to that of crop knowledge (Silva, 2009) should be conducted among the Kaiabi. I also

suggest that research on crop knowledge could be extended to the Rio dos Peixes and Teles Pires

groups.

Besides providing the return of the results of this research to all three Kaiabi

communities, I suggest that finalizing the Kaiabi Araa educational book, getting it published and

distributing it to all the Kaiabi villagesin the three areas will be a great contribution to the

resilience of weaving knowledge across the three Kaiabi groups.

As this research focused on a comparison among the three Kaiabi groups, it raises the

compelling need for a more careful analysis on the effects of Kaiabi expansion in the new

configuration of Xingu Park, for instance how they have occupied the space inside the Park and

how their political alliances and actions have affected the Xingu landscape and other Xinguano

groups in the last forty years (Oakdale, 1996). The same is true for Rio dos Peixes and Teles

Pires areas, in which the territory is shared with Apiaka and Munduruku groups. Understanding

the relationships between the Kaiabi and other indigenous groups with which they share the

territory, as well as between them and non-indigenous social actors that inhabit the surrounding

of their current or claimed lands is another important step to be taken to support processes of

land claiming and prospects for an improved management and control of their territories. Even









with all the challenges ahead, the Kaiabi history is an inspiring example of indigenous

adaptiveness and innovation to forced displacement and concomitant social and environmental

change.













KAIAHl] ANCESTMIAL
TERRnrrV

EIcdi=c :.tb LarTsri G BfTuTperf
braPmiep(B1.9'rno l /I De iimaIl
/- 1S9B0]3




JOE/)


bL*M bblnItttr

LJ, .I' A ..2

pen, bar~i:


OM TimESnlnS
WIVRAIWGY


I Tfiarbl F 9 X -I p-I Ia)-JI



PARR Ldrw





ab pr wo fd nlangicua
1 3I'1 Shiebi ium Osti Mkra lerlan~on =,jta.





iaa /ai -.eq t





l..... lelaltulo




1.L1d L**a/ 1r *a"



I'

SriP fmilike
Dtqi
7w L~pbuwlw c


Figure 9-1. A model based on the adaptive cycle from the socio-ecological systems theory,

showing collapse, reorganization, exploitation and conservation events applied to the

history of Kaiabi people.


RGI DOS PUXES
TAIW'T


P
EMiBMatWtWan
______ 1980 1*I *) ___









APPENDIX A
CODEBOOK USED IN THIS RESEARCH

SOCIO-ECONOMIC DATA QUESTIONNAIRE 1

1. Codes of interviewed people
CA (1) Capivara village. Example: CA011 Capivara village, household 1, head of the
household (person number 1).
TU (2) Tuiarare village
RP (3) Rio dos Peixes area (Tatuy Post and Novo Horizonte village)
KU (4) Kururuzinho village and Post Teles Pires River


2. Gender: 0- F female; 1


M male


3. Place of Birth:
1. PX Xingu Park
2. RP Rio dos Peixes
3. TP Teles Pires
4. OT -Other
999 missing data

4. Interviewed in 2002? (Kururuzinho-2004):
0 no
1 -yes


5. Relationship with the head
1. CHF chief (the head)
2. FIO son
3. FIA daughter
4. ESP- wife
5. SOB nephew
6. IRM brother
7. TIO uncle
8. TIA aunt
9. GEN son in law
10. PRO cousin (male)
11. PRA cousin (female)
12. IRA sister
13. Other
999 missing data


of the household:


6. Level of proficiency in Kaiabi language:
0 none (does not speak or understand);
1 low (understands but does not speak);
2 moderate (understands and speaks, but not fluent)
3 fluent
999 missing data











7. Level of formal schooling (complete or incomplete)
0 No instruction
1 Elementary
2 Secondary
3 University

8. Functions in the village:
1. Indigenous Teacher
2. Indigenous health agent
3. Indigenous dentist agent
4. Beekeeper
5. Agent for management of natural resources
6. Political leader
7. Village's chief
8. Village's vice-chief
9. Shaman
10. No specific function
11. Boat driver
12. Radio operator
13. FUNASA sanitation agent
999 missing data

9. Stability of income:
1. Waged officer
2. Retirement pension
3. Family pension
4. Irregular (no stability)
999 missing data

10. Sources of income:
1. Wage
2. Retirement pension
3. Family pension
4. Handicrafts
5. Sale of agricultural products
6. Temporary work
7. Family support
8. Others
999 missing data

11. Political participation (in meetings of the local Association):
0. No
1. Yes









12. Participation in political councils:
0. No
1. Yes

13. What is the work of the local indigenous association (ATIX/ITAOK/KAWAIP)?
0=no
1=yes
1. Patrolling/monitoring indigenous land's borders
2. Projects
3. Cultural revitalisation
4. Political articulation
5. Commercialisation of products
6. Land struggles
7. Other
8. Don't know
999 missing data

14. Preference for project/activity to be developed in the future:
1. Agricultural production
2. Natural resource management
3. Handicrafts
4. Musical revitalization
5. Beekeeping
6. Brazil nut
7. Cattle raising
8. Festivals
9. Political
10. Indigenous salt production
11. Indigenous pepper production
12. Education
13. Fish raising
555 don't know
999 Missing data

15. Travels between the three areas, two-way (number of trips)
Capivara Rio dos Peixes
Capivara Teles Pires
Tuiarare- Rio dos Peixes
Tuiarare-Teles Pires
Rio dos Peixes Teles Pires

15.1 Reasons for the travels:
1. Visit relatives
2. Fetch natural resources
3. Political (meetings, land claim processes)
4. Work









5. Marriage
6. Participation in training courses
7. Participation in projects
8. Other
999 missing data

BASKETRY KNOWLEDGE MEN QUESTIONNAIRE 2.


1. Do you weave baskets? A) Do you weave simple baskets? B) Designed?
1 for simple non-painted; 2 for designed baskets; 3 for both.

2. How did you learn?

1. SO alone;
2. PA father;
3. TI- uncle;
4. AV grandfather;
5. IR brother;
6. SG father-in-law;
7. OT other men;
8. OF workshop
999 missing data

3. Approximate age that you learn to weave designed baskets: number.

4. Have you sold? Yes/no.

5. Have you gave or exchanged? Yes/no to/with whom, qualitative.

6. Have you paid shaman services with baskets? Yes/no.

7. Did you learn any design from the book? Code.

8. Interview showing catalogue of basketry designs. A) Do you weave this design?
Yes/no. b) Can you name this design? Code for name of design reported.

TEXTILES KNOWLEDGE WOMEN QUESTIONNAIRE 3


1. Do you weave designed textiles? 1 traditional hammocks; 2 designed hammocks; 3
both; 4- other.

2. How did you learn?
1. SO-alone;
2. MA mother;









3. TI- aunt;
4. AV grandmother;
5. IR- sister;
6. SG mother-in-law;
7. OT other women;
8. OF-workshop;
999 missing data

3. Approximate age that you learn to weave designed baskets: number.

4. Have you sold? Yes/no.

5. Have you gave or exchanged? Yes/no to/with whom, qualitative.

6. Have you paid shaman services with baskets? Yes/no.

7. Did you learn any design from the book? Code.

8. Do you have cotton planted in your farm plot? Yes/no.

9. Types: 1-Sing, 2-Pytan, etc.

10. A little or a lot? 1- little; 2 lot.

11. Did you learn any design from the book? Code.

12. Interview showing catalogue of textile designs. A) Do you weave this design? Yes/no.
b) Can you name this design? Code for name of design reported.









APPENDIX B
NAMES AND MEANINGS OF BASKETRY AND TEXTILE DESIGNS

Table B-1. List of names for basketry and textile designs with respective tentative
translations, organized by alphabetical order.
Names of basketry designs Translation/Meaning of name
Apiywo'ok Larvae from Api tree
Araa Basketry design
Araa tayt Basketry design with child
Awara pypot Fox or wild dog footprint
Awara'i pypot fuku'i Little fox or wild dog long footprint
Awasiayj Maize or corn seed, grain
Awasiayj iru Corn seed inside a frame
Awasiayj iru e'em Corn seed without a frame
Inimo eta Many cotton threads
Ipirien Path
Iruj erap Back of basket
Irupe'yp Tree or path of basket
Iwirafu'a Rolling vine
Iwirapyj Vine

I'yp My tree, path to follow
Janipap wuu Fruit of genipapo tree
Jarukang Rib bones
Jarukangi Small rib bones
Jeywyu Straight drawing
Jowiterian Drawing that changes direction
Jowosiape Tortoise shell
Jywa pekangerowat Twisted arms
Kangytat pit Zig-zag, design of headfeahter support
Kupekang jyrowak Twisted back
Kururu'i Little frog
Kwasiarapare'a Drawing with arms and opened eyes









Table B-1. Continued
Names of basketry designs
Kwasiaraparuu
Kwasiarapat
Kwasiarapiayj
Kwasiarapi'wa
Kwasiari
Kwasiaruu
Kwasiat
Moiafu'a
Moiarangap
Panakukang
Panakukupe
Parasiesiewi
Pirapeku
Pyapyain
Ta'aga fwa era jywyri
Ta'aga fwa iru e'em
Ta'agafu'a
Ta'agafu'a e'a e'em
Ta'agafu'a fuku tayt
Ta'agafu'a j arukang
Ta'agafu'a tayre'em
Ta'agafu'a tayt
Ta'agap fu'a fwa eok
Ta'agap fu'a iru
Ta'agap fu'ajopep
Ta'agap fwa eok
Ta'agap j akunaap
Ta'agap jopyy
Ta'agap tayt


Translation/Meaning of name
Big drawing with arms
Drawing with arms
Crooked design
Drawing with arms
Small drawing
Big drawing
Drawing or writing
Rolling snake
Snake's head
Panaku bone
Drawing of panaku back side
Elbow
Fish throat or fish tongue
Zig-zag
Mythical person with a finger that goes around
Mythical person with fingers out of frame
Mythical person spinning or rollling
Mythical person spinning with eyes out of frame
Tall mythical person spinning with child
Mythical person spinning with ribs
Mythical person spinning with child out of a frame
Mythical person spinning with child
Mythical person spinning with bent fingers
Mythical person spinning inside a frame
Mythical person spinning
Mythical person with bent fingers
Mythical person in cross
Mythical person
Mythical people with child









Table B-1. Continued
Names of basketry designs
Ta'agawoku
Ta'agawoku j akunaap
Ta'agawoku j opep
Ta'agawoku tayt
Yogajurat
Yogii
Yok
Yokjopep
Yowawat


Translation/Meaning of name
Long or tall mythical person
Long or tall mythical person in cross
Long or tall mythical person divided in two (twins)
Long or tall mythical person with child
Bent larvae
Little larvae
Larvae
Larvae divided in two parts (twins)
Larvae that lives on tree









APPENDIX C
CODES AND POSSIBLE NAMES FOR BASKETRY AND TEXTILE DESIGNS

Table C-1. Codes, names and percentage of mentions for each basketry and textile design.


% of Number of
Code of design Names mentions names
for the cited
name


PAl
PAl
PAl
PA2
PA2
PA2
PA2
PA3
PA3
PA3
PA3
PA4
PA4
PA4
PA4
PA5
PA5
PA5
PA6
PA6
PA7
PE1
PE1
PE1
PE1
PE2
PE2
PE2
PE2
PE2
PE2


Kwasiarapat
Ta'aga fwa era jywyri
Kwasiamu
Ipirien and or Jarukang
Ta'agafu'a j arukang
Jeywyu
Pirapeku
Kwasiarapat
Ta'aga fwa era jywyri
Kwasiamu
Kwasiarapi'wa
Kwasiarapat
Ta'agawoku
Tangap fwa eok
Iwirapyj
Panakukupe
Irujerap
Jarukang
Awara pypot
Awara'i pypot fuku'i
Kwasiamu, kwasiarapat
Ta'agap tayt
Araa
Ta'agafu'a
Ta'agawoku tayt
Taangap fwd eok
Ta'agap fu'a fwd eok
Jywa pekangerowat
Kupekang Jyrowak
Ta'agap tayt
Tangafu'a


8.16
2.04
2.04
8.16
2.04
2.04
2.04
6.12
2.04
2.04
2.04
2.04
2.04
2.04
2.04
2.04
2.04
2.04
12.24
6.12
2.04
8.16
2.04
4.08
2.04
2.04
2.04
2.04
2.04
4.08
2.04









Table C-1. Continued
% of
mentions Number of
Code Names for the names
name cited

PE3_A Taangap fwd eok 4.08 4
PE3_B Kwasiaruujopep 2.04
PE3_C Ta'agap fu'a jopep 2.04
PE3_D Ta'agap jopep 2.04
PE4_A Taangap 4.08 3
PE4_B Ta'agawoku jopep 2.04
PE4_C Ta'agap jopep ifwd fuku 2.04
PE5_A Taangap jakunaap 2.04 7
PE5_B Ta'agafu'a tayt 4.08
PE5_C Ta'agawoku jakunaap 2.04
PE5_D Ta'agafu'a fuku tayt 4.08
PE5_E Ta'agap fu'a 2.04
PE5_F Taangap tayt 2.04
PE5 G Kururu'i 2.04
PE6_A Ta'agafu'a 8.16 6
PE6_B Ta'agafu'a tayre'em 2.04
PE6_C Taangap e'a e'em 2.04
PE6 D Araa 4.08
PE6 E Kururu'i 4.08
PE7 A Kwasiat 2.04 5
PE7_B Ta'agafu'a ea'em 2.04
PE7_C Ta'agawoku 2.04
PE7_D Panakukang 2.04
PE7_E Ta'agap ajaupit 2.04
PE8_A Kwasiarapat 6.12 5
PE8_B Iwirapyj 2.04
PE8 C Kwasiaruu 2.04
PE8_D Kwasiarapiayj 2.04
PE8 E Kwasiari 2.04
PE9_A Awasiayj 46.94 2
PE9_B Awasiayj iru e'em 2.04
PE10 A Inimo eta 8.16 9
PE10 B Yok 4.08
PE10 C Jowiterian 2.04









Table C-1. Continued
% of
mentions Number of
Code Names for the names
name cited

PE10_D Ta'agafu'a jarukang 2.04
PE10_E Yogii 2.04
PE10_F Apiywo'ok 2.04
PE10_G Ta'agap jopewi 2.04
PE10_H Yokjopep 2.04
PE10I Ta'agap fu'a 2.04
PE 11A Iwirapyj 16.33 2
PE 1_B Kwasiarapiayj 4.08
PE12_A Iwirapyj,Inimo eta 4.08 5
PE12_B Iwirapyj, yogajurat 4.08
PE12_C Iwirapyj e Yog'ii 4.08
PE12_D Yogii, iwirapyj 2.04
PE12_E Yokjopep 2.04
PE13 A Inimo eta 4.08 6