Citation
Community Fire Management in the Maraba Region, Brazilian Amazonia

Material Information

Title:
Community Fire Management in the Maraba Region, Brazilian Amazonia
Creator:
CARVALHEIRO, KATIA ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Agriculture ( jstor )
Burning ( jstor )
Community associations ( jstor )
Fire breaks ( jstor )
Fire management ( jstor )
Fires ( jstor )
Forests ( jstor )
Peasant class ( jstor )
Recommendations ( jstor )
Wildland fire use ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Katia Carvalheiro. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
8/7/2004
Resource Identifier:
56813732 ( OCLC )

Downloads

This item has the following downloads:

carvalheiro_k ( .pdf )

carvalheiro_k_Page_123.txt

carvalheiro_k_Page_193.txt

carvalheiro_k_Page_021.txt

carvalheiro_k_Page_127.txt

carvalheiro_k_Page_118.txt

carvalheiro_k_Page_234.txt

carvalheiro_k_Page_012.txt

carvalheiro_k_Page_068.txt

carvalheiro_k_Page_146.txt

carvalheiro_k_Page_016.txt

carvalheiro_k_Page_204.txt

carvalheiro_k_Page_091.txt

carvalheiro_k_Page_105.txt

carvalheiro_k_Page_230.txt

carvalheiro_k_Page_043.txt

carvalheiro_k_Page_164.txt

carvalheiro_k_Page_061.txt

carvalheiro_k_Page_126.txt

carvalheiro_k_Page_077.txt

carvalheiro_k_Page_040.txt

carvalheiro_k_Page_103.txt

carvalheiro_k_Page_098.txt

carvalheiro_k_Page_049.txt

carvalheiro_k_Page_109.txt

carvalheiro_k_Page_205.txt

carvalheiro_k_Page_083.txt

carvalheiro_k_Page_059.txt

carvalheiro_k_Page_096.txt

carvalheiro_k_Page_240.txt

carvalheiro_k_Page_108.txt

carvalheiro_k_Page_060.txt

carvalheiro_k_Page_063.txt

carvalheiro_k_Page_133.txt

carvalheiro_k_Page_002.txt

carvalheiro_k_Page_132.txt

carvalheiro_k_Page_104.txt

carvalheiro_k_Page_009.txt

carvalheiro_k_Page_181.txt

carvalheiro_k_Page_198.txt

carvalheiro_k_Page_081.txt

carvalheiro_k_Page_220.txt

carvalheiro_k_Page_030.txt

carvalheiro_k_Page_191.txt

carvalheiro_k_Page_142.txt

carvalheiro_k_Page_044.txt

carvalheiro_k_Page_227.txt

carvalheiro_k_Page_211.txt

carvalheiro_k_Page_222.txt

carvalheiro_k_Page_163.txt

carvalheiro_k_Page_050.txt

carvalheiro_k_Page_032.txt

carvalheiro_k_Page_218.txt

carvalheiro_k_Page_092.txt

carvalheiro_k_Page_067.txt

carvalheiro_k_Page_214.txt

carvalheiro_k_Page_128.txt

carvalheiro_k_Page_034.txt

carvalheiro_k_Page_121.txt

carvalheiro_k_Page_008.txt

carvalheiro_k_Page_024.txt

carvalheiro_k_Page_006.txt

carvalheiro_k_Page_017.txt

carvalheiro_k_Page_212.txt

carvalheiro_k_Page_010.txt

carvalheiro_k_Page_093.txt

carvalheiro_k_Page_221.txt

carvalheiro_k_Page_170.txt

carvalheiro_k_Page_229.txt

carvalheiro_k_Page_228.txt

carvalheiro_k_Page_169.txt

carvalheiro_k_Page_075.txt

carvalheiro_k_Page_195.txt

carvalheiro_k_Page_139.txt

carvalheiro_k_Page_207.txt

carvalheiro_k_Page_129.txt

carvalheiro_k_Page_165.txt

carvalheiro_k_Page_037.txt

carvalheiro_k_Page_112.txt

carvalheiro_k_Page_111.txt

carvalheiro_k_Page_233.txt

carvalheiro_k_Page_217.txt

carvalheiro_k_Page_189.txt

carvalheiro_k_Page_209.txt

carvalheiro_k_Page_157.txt

carvalheiro_k_Page_114.txt

carvalheiro_k_Page_038.txt

carvalheiro_k_Page_066.txt

carvalheiro_k_Page_158.txt

carvalheiro_k_Page_182.txt

carvalheiro_k_Page_173.txt

carvalheiro_k_Page_134.txt

carvalheiro_k_Page_183.txt

carvalheiro_k_Page_147.txt

carvalheiro_k_Page_177.txt

carvalheiro_k_Page_238.txt

carvalheiro_k_Page_086.txt

carvalheiro_k_Page_115.txt

carvalheiro_k_Page_155.txt

carvalheiro_k_Page_179.txt

carvalheiro_k_Page_026.txt

carvalheiro_k_Page_046.txt

carvalheiro_k_Page_236.txt

carvalheiro_k_Page_087.txt

carvalheiro_k_Page_131.txt

carvalheiro_k_Page_054.txt

carvalheiro_k_Page_175.txt

carvalheiro_k_Page_036.txt

carvalheiro_k_Page_031.txt

carvalheiro_k_Page_149.txt

carvalheiro_k_Page_057.txt

carvalheiro_k_Page_042.txt

carvalheiro_k_Page_137.txt

carvalheiro_k_Page_224.txt

carvalheiro_k_Page_192.txt

carvalheiro_k_Page_130.txt

carvalheiro_k_Page_203.txt

carvalheiro_k_Page_085.txt

carvalheiro_k_Page_144.txt

carvalheiro_k_Page_184.txt

carvalheiro_k_Page_022.txt

carvalheiro_k_Page_023.txt

carvalheiro_k_Page_051.txt

carvalheiro_k_Page_094.txt

carvalheiro_k_Page_089.txt

carvalheiro_k_Page_150.txt

carvalheiro_k_Page_110.txt

carvalheiro_k_Page_194.txt

carvalheiro_k_Page_219.txt

carvalheiro_k_Page_053.txt

carvalheiro_k_Page_153.txt

carvalheiro_k_Page_185.txt

carvalheiro_k_Page_055.txt

carvalheiro_k_Page_215.txt

carvalheiro_k_Page_078.txt

carvalheiro_k_Page_208.txt

carvalheiro_k_Page_140.txt

carvalheiro_k_Page_011.txt

carvalheiro_k_Page_216.txt

carvalheiro_k_Page_047.txt

carvalheiro_k_Page_124.txt

carvalheiro_k_Page_197.txt

carvalheiro_k_Page_062.txt

carvalheiro_k_Page_226.txt

carvalheiro_k_Page_176.txt

carvalheiro_k_Page_180.txt

carvalheiro_k_Page_020.txt

carvalheiro_k_Page_088.txt

carvalheiro_k_Page_154.txt

carvalheiro_k_Page_225.txt

carvalheiro_k_Page_072.txt

carvalheiro_k_Page_007.txt

carvalheiro_k_Page_156.txt

carvalheiro_k_Page_107.txt

carvalheiro_k_Page_141.txt

carvalheiro_k_Page_045.txt

carvalheiro_k_Page_065.txt

carvalheiro_k_Page_188.txt

carvalheiro_k_Page_056.txt

carvalheiro_k_Page_013.txt

carvalheiro_k_Page_080.txt

carvalheiro_k_Page_125.txt

carvalheiro_k_Page_025.txt

carvalheiro_k_Page_073.txt

carvalheiro_k_Page_232.txt

carvalheiro_k_Page_186.txt

carvalheiro_k_Page_196.txt

carvalheiro_k_Page_005.txt

carvalheiro_k_Page_090.txt

carvalheiro_k_Page_200.txt

carvalheiro_k_Page_097.txt

carvalheiro_k_Page_135.txt

carvalheiro_k_Page_136.txt

carvalheiro_k_Page_168.txt

carvalheiro_k_Page_101.txt

carvalheiro_k_Page_152.txt

carvalheiro_k_Page_201.txt

carvalheiro_k_Page_187.txt

carvalheiro_k_Page_171.txt

carvalheiro_k_Page_100.txt

carvalheiro_k_Page_206.txt

carvalheiro_k_Page_001.txt

carvalheiro_k_Page_210.txt

carvalheiro_k_Page_070.txt

carvalheiro_k_Page_167.txt

carvalheiro_k_Page_172.txt

carvalheiro_k_Page_041.txt

carvalheiro_k_Page_119.txt

carvalheiro_k_Page_231.txt

carvalheiro_k_Page_102.txt

carvalheiro_k_Page_106.txt

carvalheiro_k_Page_099.txt

carvalheiro_k_Page_033.txt

carvalheiro_k_Page_138.txt

carvalheiro_k_Page_069.txt

carvalheiro_k_Page_213.txt

carvalheiro_k_Page_190.txt

carvalheiro_k_Page_116.txt

carvalheiro_k_Page_161.txt

carvalheiro_k_Page_120.txt

carvalheiro_k_pdf.txt

carvalheiro_k_Page_174.txt

carvalheiro_k_Page_076.txt

carvalheiro_k_Page_082.txt

carvalheiro_k_Page_084.txt

carvalheiro_k_Page_058.txt

carvalheiro_k_Page_235.txt

carvalheiro_k_Page_159.txt

carvalheiro_k_Page_239.txt

carvalheiro_k_Page_029.txt

carvalheiro_k_Page_064.txt

carvalheiro_k_Page_145.txt

carvalheiro_k_Page_071.txt

carvalheiro_k_Page_160.txt

carvalheiro_k_Page_148.txt

carvalheiro_k_Page_199.txt

carvalheiro_k_Page_027.txt

carvalheiro_k_Page_004.txt

carvalheiro_k_Page_019.txt

carvalheiro_k_Page_015.txt

carvalheiro_k_Page_162.txt

carvalheiro_k_Page_039.txt

carvalheiro_k_Page_052.txt

carvalheiro_k_Page_237.txt

carvalheiro_k_Page_014.txt

carvalheiro_k_Page_166.txt

carvalheiro_k_Page_117.txt

carvalheiro_k_Page_095.txt

carvalheiro_k_Page_018.txt

carvalheiro_k_Page_202.txt

carvalheiro_k_Page_003.txt

carvalheiro_k_Page_035.txt

carvalheiro_k_Page_151.txt

carvalheiro_k_Page_113.txt

carvalheiro_k_Page_079.txt

carvalheiro_k_Page_048.txt

carvalheiro_k_Page_178.txt

carvalheiro_k_Page_122.txt

carvalheiro_k_Page_028.txt

carvalheiro_k_Page_223.txt

carvalheiro_k_Page_143.txt

carvalheiro_k_Page_074.txt


Full Text












COMMUNITY FIRE MANAGEMENT IN THE MARABA REGION,
BRAZILIAN AMAZONIA
















By

KATIA CARVALHEIRO


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ARTS

UNIVERSITY OF FLORIDA


2004

































Copyright 2004

by

Katia Carvalheiro


































To Gustavo for our love















ACKNOWLEDGMENTS

This thesis was a long journey, which could only have been completed with the

help and support of the peasants of the Maraba region, my family, friends, and mentors. I

would like to express my gratitude to the peasants of the Sdo Francisco do Itacaiunas and

Cupu communities, that received me in their houses and communities, and patiently spent

their time in the Fire Action and teaching me about their culture. In particular, I am

grateful to the peasants Mrs. Bia and Mr. Placidino that, in1992 and 1993, received me

for months in their community, and deeply influenced my way of understanding

peasantry in Amazonia.

I sincerely thank Dr. Marianne Schmink for all her constructive criticism, patience,

and support which helped to strengthen this study. For their precious contributions,

suggestions, and guidance, I am also very thankful to the members of my committee:

Dr. Marianne Schmink, Dr. Anthony Oliver-Smith, and Dr. Karen Kainer. I especially

wish to thank Dr. Charles Wood for comments on my proposal for fellowships.

I gratefully acknowledge the support of the Inter- American Foundation (IAF), the

Natureza e Sociedade Program, and the Tropical Conservation and Development

Program (TCD) at the University of Florida for the fellowship that allowed my

enrollment in the graduate school. Moreover, I would like to thank the World Wildlife

Fund (WWF-BRAZIL) and the Tropical Conservation and Development Program (TCD)

for funding my research in Para state.









The support I received from Maraba's institutions FATA and LASAT was essential

for developing this study. I am especially grateful to Manu Wamberguer for his inspiring

life trajectory.

For their help at various points along the way, I extend my appreciation to the Latin

American Center staff Margarita Gaudia, Myrna Sulsona, Wanda Carter and Lydia

Gonzalez; the TCD staff, particularly Hannah Covert and Peter Polshek; the UF

International Center staff Maud Fraser; and the ETD staff especially John Fishman.

I also wish to express thanks to my friends Joelma, Paul and William Losh, Diana

Alvira, Juli Haradish, Carla Miller, Edviges loris, Olendina Cavalcante, Noemi Porro,

Neila Soares. In Brazil, I especially thank Marli Mattos, Paola, Guilhermina Cayres, and

Westphalen Lobato.

I express thanks as well to Denise and Cyro Negreiros, for the majority of English

corrections, Paul Losh for reviewing the introduction, John Dain for translating the Fire

Poem, and Nicolas Stahelin for reviewing this acknowledge.

My heartfelt gratitude goes to my husband Gustavo Negreiros, for all his support

since the very beginning of this study. His efforts in supporting the development and

conclusion of my study made the admiration that I feel for him grow even more. I thank

my two boys Francisco and Jose for supporting me in studying and working far away,

and for always demanding my return. I am also thankful to my friend and second mother,

Marina Oliveira, for taking care of my boys as her grandsons. Finally, I thank my parents

for inspiring me and always supporting my decisions, even when they were not very sure

what I was doing or where I was.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES ....................................................... ............ ....... ....... ix

LIST OF FIGURES ............................... ... ...... ... ................. .x

ABSTRACT ........ .............. ............. .. ...... .......... .......... xii

CHAPTER

1 IN TR O D U C T IO N ...................... .... .............. ..................... ..............

Current Governm ent Strategy ......................................................... .............. 2
Problem Statem ent .................. ......................................... ............ .. ..
R research Q u estion s........... .................................................................. ........ .. ... 13
Study Site ........................ ................................ ........ ......... ........... 14
Sdo Francisco do Itacaiunas Community .................................... ............... 16
Cupu Com m unity .................................. ... .. ..... ............ 20
R research M methods ................ .... ........ ..... ...... .. .. ...... ... 23
Selection of the Communities and Families.................................................26
D ata C collection ................................ ........................................ ..........2 8
Conceptual Discussion on Participatory Approach..................................................31
O organization of the Stu dy ................................................................ .....................36


2 COM M UNITY FIRE A CTION ........................................................ ............... 39

Introduction ......................... ... ..... ........ ....... ........................ 39
Peasantry Historical Construction in the Maraba Region .........................................40
P easant O rg anization s .............................................................. .....................4 5
FATA and LASAT ............................ ...... ... .. ..... ............ 48
Communities' Social Organization........... ............ ........ ................. 49
FATA/LASAT Community Fire Action ...... ............................................51
Process for defining Practices for Fire Management .......................................53
First A ction-Learning Cycle................................... ......... ............... 58
Second A ction-Learning Cycle ....................................... ............... 76
D iscu ssio n ...................... .. .. ......... .. .. .................................................. 8 2









3 COLON IST COM M UN ITIES.......................................................... ............... 83

In tro d u ctio n ........................................................................................................... 8 3
A ge of H eads of F am ilies .............. .............. .......................................................83
Migration and Length of Residence in the Community .................. ..................... 84
F am ily L ab or Sy stem .................................... .. ...... ...................... .......... .. .. ..86
Productive System ............................................ .. .. .... ........ ......... 88
F orm al E du cation ........... .................................................................... ........ .. ....... .. 92
C cultural Interpretations ...................................... .... ............... .. .............. .... 96
C om m unities and Tow ns ......................................... ................. ............... 98
Forest ................................. .......................... .... ..... ......... 104
F ir e ............................................................................................................... 1 0 7
D isc u ssio n ........................................................................................................... 1 1 3

4 COMMUNITY FIRE MANAGEMENT EVALUATION ........................................115

In tro d u ctio n ......................................................................................................... 1 1 5
Losses Caused by Fire ......................................... .................. ... ...... 117
Community Organization for Burnings ................................................................120
Sdo Francisco do Itacaiinas Community ........................................ ....122
C u pu C om m u n ity ........................................................................................ 12 9
Before Burning ...........................................................135
Responding to Official Fire Use Regulations ...................................................136
Communication between Neighbors. .................................. .. ......138
Period Chosen to Burn: M onth Choices ...........................................................142
Burning after the First R ain .................................................................. 147
Preventive Firebreaks ................................. .......................... ........ 154
O n the burning day ...........................................................................160
Number of People at the Burning Event ........................................................... 161
Time to Start Burnings ...........................................163
U se of B ackfire .................. .............................. ................... .. .. ...... .. 167
A after B turning .................... ....... ................................................ 170
Sum m ary of Practices Evaluated ......................................... ....................... ...172
F ire S trateg ie s ................. ....... .. ........... ... .... .............................. .......... .. .. .. 17 5
Peasants' Evaluation of Government and Community Fire programs ......................181
D isc u ssio n ................................... .................................................... ....1 8 4

5 C O N C L U SIO N S...........................................................................188

G overnm ent A approach ............................................................ ............... 188
Colonist Em pirical K now ledge........................................................................... 189
Participatory A approach ........................................................................ 191
General Conclusions ................................................ ........ ................. 191
Contributions of this Study ......................................................... .............. 192









APPENDIX

A GLOSSARY OF ACRONYM S........................................... ............................ 194

B PEA SA N T'S N A R R A TIV E S ........................................................... ....................196

C FATA/LASAT COMMUNITY FIRE ACTION BOOKLET.............................. 203

D IN T E R V IE W G U ID E ....................................................................... ....................2 14

E CODED BOOK .................................... ..... ....... .. ............215

L IST O F R E FE R EN C E S ........................................................................... ..............218

B IO G R A PH IC A L SK E T C H ........................................... ...........................................227








































viii
















LIST OF TABLES


Table page

2-1. Peasant critical assessment of the IBAMA/GTA booklet on fire management.....68

3-1. Most common kinds of workforce found in the Sao Francisco do Itacaiunas
(SFI) and Cupu com m unities ..................... ......... ........................ ............... 88

4-1. Distribution of kinds of losses caused by uncontrolled fires by year in A) Sao
Francisco do Itacaiunas community and B) Cupu community .........................119

4-2. Main practices evaluated, and possible changes due in part to the
FATA/LASAT Fire Action................................ ......... ................. 74
















LIST OF FIGURES


Figure pge

1-1. South America and Brazilian Amazonia region maps showing the study area in
the so-called 'arc of deforestation', the Maraba region, north of the Para state. ....17

1-2. Sao Francisco do Itacaiunas Community village................... ..... ............... 20

1-3. Cupu C om m unity village ............................................... ............................ 23

2-1. Representation of the FATA/LASAT Fire Action......................... ..........55

2-2. Codification of fire management actions into phases................ .............. ....61

2-3. Summary of the first systematization on Community Fire Management .............64

2-4. Some of the elements used in the "Burning Dynamics." ................................66

2-5. Two examples of Burning Dynamics. ................................. ....... ...... ............. 72

3-1. Origin distribution by state, by gender, for the heads of families of the Sao
Francisco do Itacaiunas and Cupu communities (n=72)................................85

3-2. Representation of land use distribution....... ........ .......... ..................... 90

3-3. Families' formal education grades attended. ................... ................. ......... 95

4-1. Distribution of families that reported any kind of losses ................................121

4-2. The FATA/LASAT Fire Action and Sao Francisco do Itacaiunas community
activities, from 1998 rainy season through 2000 dry season ............................123

4-3. Representation of the Sao Francisco do Itacaiunas community's strategy in
conducing fire management, from 1998 to 2000..........................................126

4-4. Related FATA/LASAT Fire Action and the Cupu community activities, from
1998 through 2000 dry season. ...................................................................... 132

4-5. Representation of Cupu community organization for fire management, from
1999 to 2000. ......................................................................135









4-6. Representation of a burning situation in the Cupu community, in 1999,in
which the lack of communication caused loss to one family............. ...............142

4-7. Rainfall distribution and months chosen for burning. .......................................145

4-8. A slashed primary forest burned after heavy rain, at the end of the dry season,
resulting in a bad quality burning for cropping. .............................................150

4-9. Firebreak bordering forests. ..................................................................... ....... 158

4-10. Distribution of ratio of people to size of area burned....................................162

4-11. Distribution of studied families' 1998 and 1999 choices of burning time for
their agricultural fields ........... .................................................. ............... 165

4-12. Conjoint burning situation of two neighbors. ......... ......................................178

4-13. Example of a successful use of control firebreak. ............................................180















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

COMMUNITY FIRE MANAGEMENT IN THE MARABA REGION,
BRAZILIAN AMAZONIA

By

Katia Carvalheiro

August, 2004

Chair: Marianne Schmink
Major Department: Latin American Studies

The Brazilian government has increased investment in education, technology, law

enforcement, and infrastructure to control fire use in Amazonia since 1998, but these

campaigns have not decreased fire use. Colonist peasants, the social actors focus of this

study, were the main target public of IBAMA fire campaigns in Amazonia. While

colonists make use of fire every year, and are directly affected by political decisions

concerning fire use, they have no voice influencing those decisions. The main groups

that could hear peasant's voices-- technicians, government, academics-- generally

disregard colonist's knowledge.

This study discusses government strategy for fire control, documents colonists'

knowledge on fire management, and evaluates a participatory approach based on Freirian

pedagogy of problem-posing, codification, and learning-cycles. Two colonist

communities involved with the FATA/LASAT Fire Action were evaluated in 1998, 1999

and partially in 2000 using a combination of participant observation and systematic









interviews of a purposeful sample of families from each community. The results

presented show colonists' unique history and set of values and beliefs, manifested in the

communities' forms of organization, communication, and adaptation of fire management

recommendations developed in the participatory Fire Action.

This study's general conclusions are that the success in fire control in the two

studied communities resulted from a combination of families' willingness to avoid fire-

related losses, the participatory approach adopted and coordinated by trusted leaders and

institutions, and experienced technician, and families' fear of government coercive

pressure for fire use. In addition, this study argues that if peasants are not seen as

partners with valuable knowledge to contribute with technicians for critical reflection on

better fire use, the current top-down government actions may lead to more social conflicts

in the region.















CHAPTER 1
INTRODUCTION

Images of tropical rain forest burnings have been spread worldwide during the last

decade, associated with forest destruction and mismanagement, raising awareness about

fire as a problem for tropical forests. In Brazil, the worst event of uncontrolled fire

happened in Roraima state, in 1998, when 33,000 kilometers were burned (BVRoraima

2004). The Roraima case was crucial for new Brazilian government policies on fire

monitoring and control in Amazonia.

Although international concern about fire use has been relatively recent, fire has

long been a part of the history of Amazonia. Charcoal from fires at least 2,000 years ago,

and even older events, can be found in soil pit layers all around the Amazonian region

(Pyne 1998; Meggers 1994; Negreiros personal communication). What has changed

lately, however, is the higher incidence of fire associated with agricultural frontier

expansion in Amazonian, particularly in regions with clear dry season1, logging

activities2, concentration of farmers whose main productive tool is fire, and in years

subject to El Nifio-Southern Oscillation.





1 The sharper the contrast between dry and wet seasons, the more vigorous the fire regime (Pyne 1998:65).

2 Intensive logging increases forest flammability: forest cuttings increase potential fuel loads and the
amount of solar radiation reaching the forest floor (Uhl et al. 1994:295).

3 "El Nifio" is part of a global configuration of atmospheric and oceanic changes resulting in droughts and
inundations around the world. Amazonia has seemed vulnerable to drought during El Nifio oscillations
(Meggers, 1994).






2


Because of losses caused by fire at local, regional, and global levels, there is a

common agreement that fire has to be better managed, and alternatives to its use have to

be adopted, a consensus which is reflected in the growing governmental and non-

governmental efforts in this direction. This common ground is a powerful step toward

integrated solutions. However, there is no agreement among the several different actors

on what 'better fire management' is, and how it can be reached.

Current Government Strategy

Brazilian government strategy, according to one of PREVFOGO4 director, is based

on the "tripod of education-enforcement-monitoring, under the slogan: to educate not to

burn" (Vargas 2003). Government educative actions consisted of media campaigns,

educative materials, and courses, whose main goals were to inform and persuade. The

government had assumed an anti-fire position, focusing on convincing farmers about the

terrible effects of fire (on soil, forest, animals, water sources, people's health,

transportation, carbon emissions, and the international image of Brazil), to show them

how to manage and control fire, and to warn of the penalties for those who do not comply

with the law. Courses were provided by PROARCO5 to firefighters and brigades (local

disseminator agents), and to peasant representatives, showing them how to do Controlled

Burnings.





4 PREVFOGO is the National System for the Prevention and Combat of Forest Fires, created in 1989 and
ratified in 1998, administrated by the IBAMA (Brazilian Institute for Renewable Natural Resources and the
Environment). For a complete list of acronyms see Appendix A.

5 At Amazonia level, the Brazilian government created the PROARCO (Programme for the Prevention and
Control of Burning and Forest Fires in the Deforestation Arc), set up also in 1998, after the Roraima case.
This program is jointly administered by IBAMA (Brazilian Institute for Renewable Natural Resources and
the Environment) and by MMA (Ministry for the Environment, Water Resources and Amazonia).









Important issues like fire control cannot be effectively answered with a top-down

approach, however. Such approaches assume a "domino or trickle-down effect," in

which information goes in a one-way direction, reaching increasing numbers of people,

without dialogue or feedback. Although local people are involved in PREVFOGO

actions, the approach adopted has been top-down, since decisions about the actions "are

made by a small set of powerful stakeholders, according to their own agendas, knowledge

and value systems" (Ingles et al. 1999:6). In an attempt to work with social movements,

in 1998-9, IBAMA (Brazilian Institute for Renewable Natural Resources and the

Environment) worked as a partner with GTA6 (Amazonia Work Group), with economic

support from PPG7 and USAID, forming an emergency program called PROTEGER,

which in Portuguese means 'to protect'. In its second phase (PROTEGER II), GTA was a

partner with MMA (Ministry for the Environment, Water Resources and Amazonia).

PROTEGER's purpose was to reach a large and well-distributed number of Amazonian

grassroots organizations in a short period. This program provided fire control and

prevention training given by firefighters to community leaders, who would pass the

information along to their communities. The strongest aspect of this program was

considered to be its intended multiplier effect, disseminating the information to a large

number of rural families. However, because of the lack of information about Amazonian

realities related to fire use, the technical information given in the training sessions was

based on urban/savanna firefighting knowledge and techniques (verified in the

IBAMA/GTA 1998 booklet, in which many of the recommendations did not fit

Amazonian conditions), and feedback from those indirectly reached by the courses.


6 GTA is an umbrella institution created in 1992 with support from the PPG7 program, and is composed of
more than 400 Amazonian NGO's.









Evaluation of the program's effectiveness was based mainly on the number of people

attending meetings.

Many questions remained unanswered. What practical fire management

applications had peasants made in their productive systems? Did they adapt learned

techniques? What worked well, and what did not work at all? Did peasants change their

fire practices? How? Virtually no information existed to answer these questions. My

study thus make a unique contribution to our understanding of the impacts of fire

management programs (in this case, using a different approach from the government's) on

peasants' use of fire.

The newest IBAMA/PROARCO action plan, called "Community Management of

Fire" (Gerenciamento Comunitdrio do Fogo), started in 2004 in Roraima state (IBAMA

2004c), and seems to be another example of a top-down attempt by the Brazilian

government to modify peasants' fire use. In this plan, 960 peasants (distributed in

brigades) appropriately trained and equipped by IBAMA will disseminate the right

practices for fire use and control (IBAMA op. cited). The italics were added by the

author to emphasize the uneven power of who owns the right knowledge (and therefore

decides about its content), and decides who is qualified to manage fire and, by

exclusion, who is not.

Beginning in 1998, IBAMA prohibited fire use in municipalities located in the "arc

of deforestation" for at least 2 months: those months were satellites showing high

numbers of hot pixels. In 2002, in Mato Grosso state (also located in the "arc of

deforestation") a "mega-action" involving the Brazilian Environment Ministry, State

Environmental Secretary, and IBAMA was organized to identify and penalize those using









fire during the prohibition. Given IBAMA's clear rationality on prohibiting fire during

the driest month of the year, one of the regional IBAMA coordinators gave an interview

in which he questioned farmers' and peasants' rationality: "The environment asks only

two months of armistice for not using fire. There are ten more months to bum their

fields. Why don't they burn when they are allowed to? We will not give them [burners]

an armistice. [IBAMA] action is to repress and combat those burnings." (Benedet 2002).

In addition to not understanding farmer and peasant logic on burning their fields during

the last half of the dry season (after they prepare the land, and before the rains), this

IBAMA coordinator's speech presented war terms in the agency's relationship with

farmers, not focused on education but on repression. Changing farmers' behaviors based

on this environment of war, repression, or unilateral view of right and wrong cannot be

expected to be very effective.

The second foot of the IBAMA tripod is monitoring. Indeed, large-scale remote

sensing has been shown to be a powerful tool for fire-occurrence monitoring (Setzer and

Pereira 1991; Skole and Tucker 1993; Sanderberg et al. 1998; Nepstad et al. 1999; Souza

Jr. et al. 2003; Eva and Fritz 2003). More recently, IBAMA inaugurated the CEMAN

(Environmental Monitoring Center), in partnership with the SIPAM (Amazonian

Protection System), equipped with the most advanced technologies to monitor

Amazonian environment conditions (IBAMA 2004a). According to the IBAMA

Director, CEMAN will be the "technological eye for Amazon security."

Sophisticated technological monitoring is a very important tool for controlling

Amazonian fires and deforestation. However, it may not be sufficient. Despite its

obvious advantages, Harwell (2000) warns about adoption of remote sensing and its









interpretations by different social actors during the Indonesia 1997-1998 fire disaster.

She points out that "in addition to the silencing of local voices, remote approaches to

disaster events also obscure these on-going linkages of humans with their environment"

(Harwell 2000:334). Cultural understanding of environment is historically constructed,

as well as social relations among different social actors, factors that have also to be

considered in remote sensing interpretations. If the "technological eye" is in hands of a

small set of powerful social actors only (government, international donors), participation

remains passive because local people, however included in the politics and policies of

conservation, remain peripheral to defining the ways in which conservation is perceived

and natural resources managed (Goldman 2003:834). This thesis explores ways to

involve local communities directly in monitoring.

The third foot and last of the IBAMA tripod is law enforcement. Since 1965, fire

use has been ruled by an imprecise Brazilian law, called the Forest Code (C6digo

Florestal). This federal law prohibited the use of fire "on forest areas and on other forms

of vegetation; however, it was allowed in specific farming or forestry situations under

governmental supervision" (IBAMA/GTA 1998:36). As explained above, in 1989, the

Brazilian government started the PREVFOGO program within IBAMA with the

objective of reducing the use of fire to "acceptable" levels (IBAMA 1998). This work

was coupled with monitoring by INPE (the Brazilian Space Agency). However,

Amazonia was not included among the areas to be served by this program at that time.

Only after early 1998, when fires in Roraima reached the global media, conveying

international pressure on the Brazilian government, were actions taken in Amazonia. The

governmental laws and actions concerning fire use then became more specific and









coercive. PREVFOGO was present in the region for the first time in 1998, supporting the

creation of another program specifically for the Amazon: PROARCO (Program for

Prevention and Control of Fires in the Arc of Deforestation). In 1998, IBAMA also

established that community and controlled burnings were required to follow a set of

procedures and rules, including obtaining an official burning permit form (authorization

was required since 1965, but no practical instrument was created until 1998) and payment

of fees per hectare to be deforested and burned. However, the government's

recommendations were out of reach for most peasants, who often did not even have the

means to prove their ownership of land (one of the documents required to obtain an

official burning permit).

In 1999 and 2000, the governmental program PROARCO established firefighter

groups based in the cities of the region where the fire problem is worst. The state of Para

received four new helicopters to be used by IBAMA in enforcement actions. Those

actions may include high fines and jail for those who use fire without an official permit or

who do not comply with the official authorization. Additionally, the official burning

permit can be cancelled in special situations, such as a very high number of burnings in a

municipality (based on INPE remote sensing data). When this happens, the government

prohibits fire use, even for those who have burning permits. This prohibition lasts in

general for two months (the driest months in the year), and it has taken place every dry

season7 since 1998. Parallel to the IBAMA program, INCRA (National Institute for

Colonization and Agrarian Reform) has organized other firefighter groups associated



7 The Amazonian region possesses two seasons: rainy (or winter) season and dry (summer) season, with its
length varying according to micro-regions.









with people from peasant settlements, but based in cities. However, in Amazonia, the dirt

roads and distant locations present serious limitations for mobilization.

Between the Roraima fires and Law of Environmental Crimes regulation in late

1999, the government's strategy was to create provisional measures to prohibit the use of

fire when it is judged to be highly risky, even if official permission has been given.

Although prohibitions have happened since 1998, the fees, tickets, and legal actions were

based on those provisional measures, easily contested through the legal system. Powerful

landowners and farmers had the resources and information to take those defensive

actions, but the majority of Amazonian peasants did not. Thus peasants ended up being

further punished, which created strong resentment against the government rules. The

Law of Environmental Crimes (MMA 2000) defined fees for forest illegally burned:

around US$ 515 per hectare8 and 2 to 4 years of prison. Those penalties were widely

disseminated through radio, TV, and educative material distributed around Amazonia,

always associated with the bad consequences of fire use. The Brazilian government

increased investment in education, technology, law enforcement, and infrastructure to

control fire use since 1998. The government approach has not proven efficient yet, given

that fire occurrence in Amazonia has increased9. In addition, higher fire occurrence has

been associated with weekends and with the IBAMA staff strike in 2003 (Werneck

2003), indicating that as long as enforcement is weaker, farmers burn more. The

government's top-down approach, supported by the media, is creating a scenario



8 R$ 1.500,00 per ha (R$ 1,00 = US$ 2.91).

9 According to IBAMA 2004b, since the number of hot spots has been systematically monitored in the
Amazon region, the higher rate was found in 2002 (160,329 hot spots). In 2003, the number of hot spots
was 139,000 (IBAMA 2004b).









unfavorable for partnerships with rural producers, by labeling burners without official

permits "transgressors", and depending on paid brigades to control fire in Amazonia.

This study argues that governmental actions implemented in a top-down manner, in

which peasant knowledge is disregarded, and technical knowledge is "deposited" in their

lives, has limitations. By contrast, studies have shown that in community-based fire

management there are less occurrences of fire-related losses1 because local people share

decision-making power, volunteer to maintain their natural resources (without depending

on external inputs), and their empirical knowledge is embodied into practices and policies

for fire management (practices culturally and environmentally better adapted). The study

sought to explore these ideas through an in-depth study of two communities involved in a

community-based fire management program.

Problem Statement

Colonist peasants, the social actors that are the focus of this study, were the main

target public of IBAMA campaigns in Amazonia. While ranchers are recognized as the

major social actors responsible for area deforested (and burned) in Brazilian Amazonia

(Margulis 2003:80), peasants are also major actors because of their quantity. In Para

state alone,11 169.273 landholdings (83% of the total farms) are smaller than 100 ha

(Trecanni 2000:430) (an indirect indicator of peasants representation)12, whose main

productive tool for land preparation is fire.



10 In China Lichang et al. 2001; in India--Darlong 2001; in Thailand Makarabhirom et al. 2001;
Rakyutidharm 2001; in Indonesia Abberger et. al 2001; in Gambia Sonko et al. 2001; for a review, see
Jackson and Moore 1998).

1 Brazilian Amazon or so-called Legal Amazonia comprises eight states, representing 61% of the Brazil
area.

12 The National Forest Code (Federal Law # 4771) defines "small rural property" or "family land", as any
land exploited mainly by its owner with personal or family labor, in which at least 80% of the income









While colonists make use of fire every year, and are directly affected by political

decisions concerning fire use, they have no voice influencing those decisions. The main

groups that could hear peasant's voices, technicians, government, academics, in general

disregard peasant knowledge. This study aims to contribute to this topic by documenting

colonists' knowledge on fire management, analyzing a participatory approach in which

colonists and grass-roots organizations are partners to develop and implement better

practices for fire management, and discussing how in those approaches in which peasants

are partners with technicians, sharing decision-power, supported by action-learning

cycles, fire will more likely be controlled.

Government technicians had shown that they did not believe that colonist peasants

possess knowledge for natural resource management, because of peasants' lack of

valuable modern "tools", such as formal education, land, infrastructure, entrepreneurial

views, or money. Appropriate practices are definedfor them but not i/ ih them. The

Brazilian Federal Agency for Agriculture and Livestock Research (EMBRAPA), for

example, explicitly identified the Amazonian peasants' three main problems, that explain

their "traditional and backwards technology": lack of education, economic organization,

and energy (EMBRAPA 2000:21). The rationale for not learning from colonist peasants,

thus, is that they are the ones "lacking".

On the other hand, academic ethnographies tend to exclude colonist peasant from

the category "traditional" because their culture is too mixed with modern societies.

While studies show indigenous tribes hold specific cultural values on fire use (Ruddle

1974; Peters and Neuenschwander 1988; Warner 1991), peasants, by comparison, are

comes from farming or extractivism, and with a maximum land size of 150 ha if located in the Legal
Amazonia (Presid6ncia da Repuiblica 2002).









identified as not "genuine" or "traditional" swiddeners, but "incipient swiddeners"

(Warner 1991) or "new swidden practitioners" (Peters and Neuenschwander 1988:74).

Because colonists do not practice the traditional form of swidden, they are assumed to

not be able to "understand the importance of fallow rotation" (Peters and

Neuenschwanderl988:74); or to be "there only for the purpose of a crop for a year or

two" (Warner 1991:9); or to be totally ignorant of the new land due to their absent

cultural ecology of the region, which explains their "search for short term and predatory

means of investments" (Lima and Pozzobon 2001:233).

One explanation given for this lack of "tradition" is their previous situation of being

landless. Brazil is one of the countries with the highest land distribution concentration

rate in the world (Trecanni 2001), and many peasants end up becoming sharecroppers.

Many migrate looking for their first piece of land. This research found out that the

majority of families studied were living in their first landholding. However, many had

significant farming and fire management experience in their regions of origin.

The general misconception of colonist ignorance has blinded academics and

technicians to their empirical knowledge. Few studies go beyond this preconception, and

document in detail different uses that colonist peasants have for their natural resources.

Muchagata, for example, studying five communities in the Maraba region, concluded that

colonists "recognize and use a wide range of plant and animal species (respectively 142

and 39 were listed) which provide food, fuel, raw materials, and medicinal plants, and

represent sources of income" (Muchagata 1996:76). Muchagata and Brown (2000) also

studied colonist perceptions of soil fertility.









Regarding community fire management, Mattos et al. (2000), working with a

colonist community in the Paragominas region (also located in the so-called arc of

deforestation), presents how a community-conceived participatory approach developed

and applied a Community Agreement on fire management, based on their practical

knowledge (AMPPDR 1996). The idea of this community work for fire management was

absorbed by IBAMA to officially recognize, in September of 1998, the Solidarity

Controlled Burning (Queimada Controlada Soliddria), in which one single permit would

include five or more peasant burnings. Although the collective permit was better adapted

to community reality, fire permits continued to require land titling, to charge for

deforested and burned areas, to require firebreaks beyond peasants' capabilities, and to

cancel permits during the high dry season. Therefore, some ideas from the community-

based experience were adopted by decision-makers technicians, but not the participatory

approach.

Why are educative campaigns ineffective in changing colonist peasants' fire

practices? Do colonist peasants have empirical knowledge of fire use? Could

participatory approach be more effective in finding solutions for fire management? What

do colonist peasants think about fire use law enforcement? Those questions are not only

academic, but also are key for those making decisions on fire management in Amazonia

today. Recognizing colonist empirical knowledge does not imply that they have all the

solutions for fire management. This study argues that if they are not seen as partners

with valuable knowledge to contribute i/ i/h technicians for critical reflection on better

fire use (better also for peasants) the actual top-down government actions may lead to

more social conflicts in the region. This study also describes a process for a participatory









approach, involving grass-roots organizations, colonist peasants, and in some stages

governmental representatives, in which the philosophy and methods can contribute to

show directions for collaboration between social actors for fire management.

In the Maraba region, Southeast Amazonia, the FATA (Tocantins Araguaia

Agrarian and Environment Foundation) and LASAT (Social-Agrarian Tocantins

Labotatory) Community Fire Action involved more than 20 colonist peasants and 3

indigenous communities. Recommendations for fire management were developed in a

participatory process, coordinated by grassroots organizations, and based on colonists'

empirical knowledge. The Fire Action produced positive impacts on the two

communities studied, as families developed agreements to use fire, and losses caused by

uncontrolled fires diminished drastically after the Fire Action. Some of the

stakeholders13 had discussed FATA/LASAT's proposal and they had supported it as a

regional substitute for the governmental strategies. Although the action is no longer

carried out as a specific activity of FATA or LASAT, systematic evaluation of this

experience can provide relevant insights to those interested in diminishing losses caused

by fire in Amazonia.

Research Questions

This research contributes to the relevant topic of fire management in Amazonia,

specifically involving colonist peasants and changes in their fire practices. This thesis

aims to discuss three main questions:

* Question 1: Why aren't government fire actions and laws more effective in
diminishing uncontrolled fires?

13 In this study, the terms "social actors" and "stakeholders" are used as synonymous, referring to any
"individual, social group or institution that possesses a stake (or interest) in the management of the natural
resource concerned" (Borrini-Feyerabend 1996, cited in Ingles et al. 1999).









* Question 2: Do colonist peasants have and use empirical knowledge to manage fire?

* Question 3: Is a participatory approach to fire management effective in diminishing
fire losses?

Question 1 was examined by analyzing government discourse on peasants and fire

use, families' discourse on government actions (historical and actual experiences), and

families' evaluation of the official fire actions (fire law, its enforcement, and educative

campaigns). Results are presented in Chapter 2 and 3.

Question 2 was developed through peasants' narratives about their practices and

perceptions of fire and natural resources management, where had they learned them,

changes over time, as well as during observation of burnings. Results are presented in

Chapter 3 and 4.

Question 3 was analyzed by evaluating the participatory approach applied to a

community fire management action (the FATA/LASAT Fire Action), and its impacts on

fire practices of colonists families in two communities. The process regarding the fire

action is described and analyzed in Chapter 2, and the evaluation of its impacts is

presented in Chapter 4.

Study Site

Para state is the second in the rank of fire occurrence in Amazonia in concentration

of hot spots.14 The Maraba region, located in an Amazonian frontier region of the so-

called "arc of deforestation" is one of the high concentration of hot spots (INPE 2004).

The Maraba region is also known as Bico do Papagaio15 region, and was originally


14 The expression "hot spot" is used for heat registered on the earth's surface by the AVHRR sensor on
board the NOAA satellites, which captures and records any temperature above 47 Celsius, and reads it as
"hot spots" (PROARCO 1998).
15 Bico do Papagaio region or "Parrot Beak" region has this nickname due the shape resulted from the
Tocantins and Araguaia rivers' confluence (see Figure 1-1)









occupied by indigenous groups. It comprises an area stretching across the state-lines of

Para, Maranhdo and Tocantins (Figure 1-1), and includes eight municipalities: Marabi,

Itupiranga, Sao Jodo do Araguaia, Sao Domingos do Araguaia, Sao Geraldo do Araguaia,

Nova Ipixuna, Brejo Grande do Araguaia, and Palestina do Pard (Forum das Entidades

pela Reforma Agrdria do Sul e Sudeste do Pard 2001:4). By the year 2000, the actions

conducted by FATA and LASAT included the first six municipalities on the list and later

that year extended to the remaining ones.

The importance of peasants in the state of Pard is illustrated by the high number of

small plots of lands, which account for nearly 82% of the total number of plots, or

169,273 plots (Treccani 2001:430), distributed in 33% of the area (DFID1998:16). In

addition, peasants in the state are estimated to produce as much as 84% of all the manioc

produced, as well as 63% of the corn, 59% of the beans, and 46% of the rice (Hurtienne

1999, cited by Veiga 2000:15).

This region is located in the so-called the "Brazil-nut Polygon", an area named after

the abundant Brazil-nut trees (Bertholletia excelsa). The average temperature is 260C,

and rainfall around 2000 mm/year (Muchagata 1996:13). It is characterized by defined

rainy and dry seasons. The rainy season is concentrated from November to March, and

the dry season from May to September. April and October are transition months. The

vegetation is classified as Ombrofila Densa Montana Forest (Sestini 2002:44).

The two communities studied, Sao Francisco do Itacaiunas and Cupu, were formed

in much the same way. In most cases, families were landless migrants originally from

Brazil's northeastern region occupying part of a large-scale farm. They faced landowners

in conflicts over access to the land, and struggled to make their living by means of









extractivism, agriculture, and cattle, for which they had very little government support.

They built community roads, churches, schools, and soccer fields; they harvested their

own crops and marketed their own products; they founded their associations and

partnered with NGOs; they obtained loans (and therefore debts) from the government for

their fields. The Catholic Church and grassroots organizations were cited as their main

supporters. Complaints about government absence or negligence were present in many

different moments of this research, pointing to a potential resistance to future partnership

agreements or collaboration with governmental guidelines laws. Some details concerning

the history of each community are given in Figure 1-1.

Sao Francisco do Itacaifinas Community

The Sao Francisco do Itacaiunas community, in the Municipality of Maraba, is

located approximately 55 km southwest of Maraba and connected to the main town

center by a dirt road. To the south of the village runs the Itacaiunas River. Nowadays,

the community is divided into 64 plots, plus a village center. According to long-time

resident Mr. Vendncio Dias, the community was started on July 14, 1983 when

representatives of 41 families settled on the site in a non-official colonization process.

Occupants were landless peasants living in local villages. More than 50% of them came

from Boa Esperanca, a village located 18 km from the occupied land. For safety reasons,

peasants occupied temporary houses in groups. They worked together to harvest a first

annual crop. Sixteen years after the village was established, the community was

officially recognized by the federal government as a Projeto de Assentament (Settlement

Project) in early 1998. The community was established as a 'new' land reform area

during former President Cardoso's administration. Official recognition represents a very

important step towards guaranteeing peasants' land ownership rights, and has long been










waited for. As of the time of this study, the long journey through the bureaucracy had not

yet allowed authorities officially to demarcate any border lines or issue any land tenure

documentation.


Figure 1-1. South America and Brazilian Amazonia region maps showing the study area
in the so-called "arc of deforestation," the Maraba region, north of the Para
state. The map below details community locations, the "Parrot Beak" shape
created by Tocantins and Araguaia Rivers, the Tucurui Dam and the Carajas
Mine, as well as highways and railroads (Bellow map adapted by author
from IBGE 1998).


Study Area
.... Mwaupdflte Pn




S MUnidpolny Cpitl
SbFatdciCodmmunity
A aplbuCMMity









The history behind Sao Francisco do Itacaiunas and how it was built is as typical as

that of any community in the Bico do Papagaio region. The area was a traditional

castanhal, or a concession of brazil-nut extractive area owned by the government.

Although brazil-nut concessions theoretically could not be deforested, a first contract

granted 3,500 ha to an oligarchic family which turned it into a 7,500 ha cattle ranching

farm or latiffndio16. After that, the land was sold twice before 1983, when the peasants

occupied it. Some of those interviewed said the farm owner was never a violent person,

and for this main reason the conflict with the settlers has never been very violent. Nor

did the farmer have land titles to prove land ownership.

During the conflict resolution with the landowner, mediated by Catholic Church

leaders, representatives of INCRA and the families agreed with the farmer that the 3,500

ha of the original brazil-nut concession would become a community (Nova Canaa), and

the remaining 4,000 ha would continue as a concession to the farmer. Nevertheless, by

the time negotiations were concluded, the number of families had grown and 56 families

were excluded from the distribution of land, among which some of those who had fought

from the days of the first occupations. The excluded group received the support of

"Manu" (Mr. Emmanuel Wanderberg, the Land Pastoral Commission (CPT) coordinator

and later one of the FATA founders), resisting and struggling for more land. The

families, which did not manage to settle in Nova Canaa, founded a new village and

named it "Sao Francisco do Itacaiunas". They chose Mr. Silva as their community

representative, representing the community at the Rural Workers Union (STR) too. The

owner eventually gave up the fight for the land. Some months later, the peasants


16 Latifindio is a word of Latin origin and means a large, privately owned landholding (Macedo 1985:viii).









concluded the construction of a school, which was the first landmark of the establishment

of the community. Mr. Silva volunteered to teach classes until a teacher was established,

with salary paid by the municipality. Mr. Silva was an important leader, and he has been

part of the Fire Action since it started.

The Sao Francisco do Itacaiunas village center (Figure 1-2) occupied an area of 25

ha. This village was composed of 49 plots, four of them serving also as small grocery

stores, one with a machine for processing rice, one Catholic Church, one plot for a

Protestant Church (with no construction), an Afro-Brazilian Center, one small wood

school built by the families just after the occupation, around 1985 (the material was

donated by a politician), the soccer field, and a new school built by the federal

government in 2001.

As shown in Figure 1-2, besides those occupied plots, families planned areas for

new plots, roads, and whatever else the community decides (like the area chosen for the

new school). From the time of the community's creation, the village was planned in such

a way that all the families that possess land had the right to a plot (10 x 30 m) so they

could build their houses in the village. In the Sao Francisco do Itacaiunas village, there

has been electric power since late 1998, as an indirect service from the electric power

extended to the Tain6polis train station (part of the Vale do Rio Doce Company railroad

for bauxite export) through one of the main roads that cross the village center. The

access to electric power brought some new appliances and changes such as TVs, freezers,

light, and rice processing machines, among others.




































Planned O -

S Dirt Road
SFamilies Living in the Community
SFamilies with Grocery Stores
9 c? Female and Male Bath


O 30m


Figure 1-2. Sdo Francisco do Itacaiunas Community village. Representation of the
individual plots for houses, school and churches, roads, rivers, and soccer
field. The upper left part of the drawing represents where the village is
planned to grow (Drawing by author).

Cupu Community

Located north of Maraba, Cupu is located in the municipality of Jacunda, which

can be reached through a 48-km dirt road and Highway PA-150. It is located 52 km

northwest of Nova Ipixuna, a city 109 km southwest of Jacunda (as shown in Figure 1-1).

On its west side, the Cupu community is bordered by the Tucurui Dam. The community

was established along the Tocantins River, before the dam.









Information on community formation is not very clear, since the occupation was

gradual. As in the Sdo Francisco case, the land also was a castanhal or brazil-nut

concession. The pioneer families got to know the area while collecting brazil-nuts. It

was originally a forested region with few anthropogenic activities, and gradually families

started moving to the area, building their houses, and producing subsistence crops. A key

referential year for the establishment of the community was 1986, when the manager of

the farm denounced the presence of peasants on the land to the owner, who lived in the

neighboring Maranhdo state. Despite the fact that the owner tried to expel the peasants

with police support, nobody was injured. It seems that because she could not prove her

rights to the land, and perhaps due to her weak political power, the owner gave up the

struggle. After some time, the manager's departure was seen as a sign of the peasants'

victory, and the families remained in the occupied land. They were also supported by the

Catholic Church, and one of the main community leaders since the occupation, Mr.

Soares, is still working as the Catholic Church coordinator17 in the community. Soares'

family were involved in the Fire Action since it began in the community in 1999.

Some time after the conflict, the community received the single official visit in the

history of its land tenure regulation. At that time, representatives from the GETAT, an

extinct federal institute for land reform created during the military dictatorship period,

visited the land once, made a list of families living there, but never returned. Since then,

some families have moved out, others have arrived, and some members have died.

Bureaucratically, they cannot have official permission to use fire, since they still cannot



1 Because of their distant location, peasant communities in Amazonia have priests, nuns, or ministers that
visit it periodically, every few months. In their absence, there are coordinators who celebrate mass weekly,
and keep the church 'alive'. These persons have a key role in influencing peasant opinions.









prove their land ownership. By 2000, the community still was not recognized as a

Projeto de Assentamento or Settlement Project, making it very difficult for them to

receive government support. The community was composed of 52 individual plots, and a

small village.

The Cupu village center (Figure 1-3), located at the edge of the main road, was

composed of two churches (Catholic and Protestant), one school, a soccer field, and three

small grocery stores, two of them with power generators and televisions. There was

another village, with land donated to the community by one of the families, free of

charge, but in practice that did not work, ending around 1993, due to absence of rivers or

other water sources, and its distance (around 6 km) from the lake of the Tucurui Dam.

The current village center is located about 2 km from the Tucurui Dam, where

many families fish and raise their cattle during the dry season when the water level

diminishes. Some families own land on the many artificial islands created by the Dam.

The present village was not planned, and plots for houses had to be bought by residents

from a peasant family, the Ribeiros. The land for the school, churches, and soccer field

was donated by the Ribeiro family. The Ribeiros are one of the families that buy the

peasants' milk production and sell it in the city. The head of this family was also director

of the regional Association, and was engaged in the Fire Action since 1999. As of 2000,

there was no electric power except for those families who owned generators. There were

27 houses in plots measuring 15 x 30 meters each, and on three of them there were small

grocery stores. In two of those stores, there was a TV set in 1999. In 2000, two more

families had bought TVs, the same two families that owned cars and used them to buy the

milk production and transport people.










































Figure 1-3. Cupu Community village. Representation of the individual plots for houses,
churches, school, and soccer field. The plots were sold to families by a
peasant, who donated land for school, churches, and soccer field (Drawing by
author).

Research Methods

The author of this thesis was the FATA/LASAT Fire Action proponent and its first

coordinator, from the 1998 dry season until the first regional seminar, held in July 1999.

The Maraba region choice was a combination of the researcher's previous experience

with community fire management in other Amazonian regions, her institutional

connection with grassroots institutions in the Maraba region, and those institutions'









interest in developing a community fire management action. The field research was

conducted between 1999 and 2000, and focused on two peasant communities in the

Maraba region. The Sao Francisco do Itacaiunas community, located in the municipality

of Maraba, was involved in the Fire Action since its very start, and the Cupu community,

located at the municipality of Jacunda, only engaged in 1999. This choice of

communities with different times and degrees of engagement in the Action was

intentional, to allow comparisons between them in terms of losses caused by fires and

community organization for fire management.

I had a twofold role of (a) implementing the Action during its first year and (b)

evaluating it during the following years. I tried to make it clear to the families that the

main goal of the study was to understand their strategy concerning fire management and

their social-productive system, as well as support them with information about fire law

and its impacts, other community fire experiences, and the importance of their

community actions to other communities. Many families wanted to express their

opinions in recorded tapes, hoping that I would help taking their voices to government

representatives. In the Sao Francisco do Itacaiunas community, I facilitated some of the

community meetings while Fire Action coordinator.

The researcher's double role-first as facilitator and then as researcher-surely biased

the research results because of people's likelihood of responding in ways favorable to the

Fire Action, and because of my own commitment to helping with community fire

management. On the other hand, with a subject as sensitive as fire-associated with

threats of fines and imprisonment-my previous trusted relationship with the community

was a key strategy in getting access to the details of their views and practices with respect









to fire management. My approach stressed listening and observing the words and actions

of community members, and conveying them in this thesis in their own words. This in-

depth qualitative understanding of peasant's perspectives on fire management emerged

from the researcher's close relationship with the communities. To offset the potential

subjective bias in the research, I combined this close familiarity with a sampling design

in the communities that would encompass a wide range of types of families, in order to

capture diverse perspectives in the interviews. This purposeful sample provided

systematic data on social characteristics, and burning practices during two dry seasons, to

complement oral histories, participant observation, and other qualitative techniques. The

combination of these methods was effective in providing a unique understanding of

peasant views on fire management.

The researcher's direct support received from key grassroots organizations (STR,

FATA, LASAT, COOCAT, and AGRAF) was crucial in obtaining the confidence of

families and their openness during the interviews conducted on fire management, on the

losses caused by fire, and on fire monitoring. This was very important because families

were aware of IBAMA's prohibition and penalties, and wanted to make sure that the

research data collected would not be used against them. The regular visits to the field

helped to build trust between researcher and families, and the academic nature of the

study became clear-it definitely had nothing to do with an IBAMA inspection. In Sdo

Francisco, members of the Agro-forest group supported by FATA were always ready to

lead me throughout the community for family's interviews and burnings. Their presence

introducing myself as someone to be trusted (connected to FATA and STR) certainly had

a strong influence on families' willingness to talk and show me their realities. Members









of the Fire Group in the Sao Francisco do Itacaiunas community, in turn, claimed that my

regular visits (from 1998 to 2000) supported the adoption of local rules for fire use for

those families not used to participating in community actions. The Cupu community,

visited in 1999 and 2000, was not as organized as Sao Francisco at the time of this

research, and a key leader (Catholic church coordinator) was the main peasant that helped

with this research, introducing me to key families. From this first contact, the majority of

the families were interviewed by myself, but always referring to those leaders and to the

grassroots that this study was connected with. The relatively lower level of general

community organization of Cupu families may be one of the explanations for their lower

degree of formal organization of fire management in particular.

As to its conceptual basis, the present study adopted the fire classification

developed during the FATA/LASAT Fire Action. Accordingly, 'uncontrolled fires' were

defined as fire events which break out from the planned burning area; 'accidental fires'

were defined as uncontrolled fire events regardless of the use of any measure to prevent

and/or combat uncontrolled fire. In this context, 'criminal fires' were defined as

uncontrolled fire events in which the use of measures to prevent and/or combat

uncontrolled fire was insufficient or absent (Carvalheiro & Aquino 1999:16-17).

Real names of communities and peasants were changed to protect them from any

eventual consequence of this study.

Selection of the Communities and Families

The Sao Francisco do Itacaiunas community had taken part in the Fire Action since

its very beginning in 1998 and was one of the communities where FATA had developed

its Perennial Crops Project. The second community studied, Cupu, was chosen because it









joined the Fire Action one year later, in 1999, and it is one of the communities where

LASAT developed its Community Forest Management Project.

The families studied were chosen as a purposeful sample with the help of

community leaders, and comprised more than 50% of the families of each community.

Leaders and researcher discussed the main general variations among families,

emphasizing that representatives of each "type" would be included in the study sample.

After evaluating the differences among families, a consensus was reached regarding final

criteria for selection of the families, which included variations in the following features:

(a) participation in different community activities (different religions, participation in

workers' associations, and mutirdo or collective works), (b) length of residence in the

community (from pioneers to newer residents), (c) location of residence (on the land, in

the community village, or in the city), and (d) main source of income (agriculture, cattle,

schoolteacher, grocery owner, middleman for milk market, with/without government

loan).

The selection process resulted in 39 families (61%) in the Sdo Francisco do

Itacaiunas community (the community with initial involvement in the Fire Action) and 29

families (56%) in the Cupu community (with later Fire Action influence). Sixty-seven

families were interviewed during the first set of interviews, plus 5 families that wanted to

participate later. Nine families from the first set of interviews were not interviewed the

second time because five of them had sold their lands, three were abcent during the

interview period (health problems), and one had moved away to the city. The families

that did not take part in both sets of interviews were not included in the sequential fire

analyses. The interviews focused on heads of families, both men and women. In this









context, the heads of families were the older members, but the answers many times

included other family members' opinions. Questions often generated discussions among

the family members about the issues considered. In addition, 12 oral stories were

narrated and recorded.

Data Collection

Information was gathered through unstructured interviews (interview guide in

Appendix D), participant-observation, and oral histories (Lofland 1971; Holstein &

Gubrium 1995; Rubin and Rubin 1995). Two data collection methods were used in this

study: unstructured interviews sequentially conducted in two sets (at the beginning of the

1999 and 2000 dry seasons, before burning activities commenced) and participant

observation during burnings (during the dry season of 2000), which provided both

quantitative and qualitative information. Participant observation was also carried out in

the communities during visits to the households and to families' burned and planned-to-

burn fields, as well as during social events (religious celebrations, meetings held by the

Association, parties, TV viewing). Oral histories were carried out with the elders and

provided detailed material on their own life stories, their involvement in conquering the

land and its characteristics, how the community started, its changes through time

(families arriving and leaving, roads, schools, transportation, vegetation, weather), and

interpretations of causes and consequences of different factors (partnerships,

organization, challenges).

The information was documented in field notes (Emerson et al. 1995), recorded on

tapes, and photographed. Data gathered was coded (see Appendix E for Coded Book)

and organized in a database for quantitative analyses. Recorded narratives were

transcribed and translated by the author for qualitative analyses (see Appendix B for









original narratives in Portuguese). Analysis and results combine both quantitative and

qualitative data.

The first set of sequential interviews, conducted in 1999, covered three groups of

characteristics: social, productive and fire practices (fire topic related to the previous year

burnings and plans for the coming one). In order to evaluate FATA/LASAT Fire Action

impacts, the second round of interviews (which occurred in 2000) repeated the same

questions regarding fire practices that had been asked before, and provided the data to

make year-to-year comparisons, whose results are presented in Chapter 4.

For interviews with selected families, the social characteristics included

information on age, size and origin of family, number and places of migrations, length of

residence in the community, level of formal education, kinship, religion (Catholic,

Protestant, Afro-Brazilian etc.), association with grassroots organizations, families' plans

to remain in the community or to leave, their points of view concerning community

needs, and other topics considered relevant by the interviewees. Productive conditions

referred to information on size of the family plot, distribution of the main ecosystems

(percentage of primary and secondary forest, pasture, agricultural field, and perennial

crops), quantity of cattle, labor (family, community and paid labor), equipment (chainsaw

and backpack sprayer), and access to government loans. Results on social characteristics

and productive conditions are presented in Chapter 3.

Sequential fire practices interviews, referring to 1998 and 1999 dry seasons, were

based on FATA/LASAT Fire recommendations, and included information on accessing

official law for fire use, community organization for burnings, communication between

neighbors, period chosen to burn, burning after the rain, use of preventive and control









firebreaks, number of people on the burning day, use of backfires, losses caused by fire.

ecosystems chosen to burn (primary or secondary forest), and burning of pasture. Results

compare practices throughout the studied years. Fire topics also studied included

peasant's discourse on a rich variety of concepts, such as the meaning of'losses' in their

productive systems, the importance and effectiveness of fire use (successes and losses),

individual versus community labor, evaluation of the FATA/LASAT Fire Action and

governmental laws, implications involving changes in fire use and alternatives to fire, as

well as proposed solutions.

In addition to the sequential fire interviews and oral histories, evaluating the

FATA/LASAT Fire Action included observations of burning events in the Sao Francisco

do Itacaiunas community in September and October of 2000. Initially, the study planned

to include burnings in different situations occurring in the two communities. However,

due to the frequency of rains and the scarcity of the researcher's field time, these

observations were limited to the Sao Francisco do Itacaiunas community.

The researcher was particularly careful not to interfere with families' decisions

regarding the choice of the day on which to burn. Although it was no easy task, since the

families knew that I was there to observe their burnings, and they clearly expressed their

helpfulness, I managed to stress to them that my purpose was to understand what their

choices were so that I could minimize the impact my presence might have on their fire

management techniques. An effective way to make the families feel more comfortable

with the presence of researchers was to carry out complementary interviews on

qualitative information in between burning events. This way, families would not feel bad

for letting the researcher 'waste her time' while they were still not ready to burn. These









complementary interviews brought more information not only on the families' evaluation

of the Fire Action itself, but also on their understanding of governmental regulations and

cultural strategies, as well as community organization for fire management, and formal

association challenges, among other topics.

Conceptual Discussion on Participatory Approach

This study analyzes the FATA/LASAT Fire Action according to its development of

a participatory approach coordinated by grassroots organizations (FATA and LASAT)

with colonist peasants in the Maraba region. Due to different meanings that

'participation' can assume, this section will explore the concepts behind the Fire Action,

as well as its main methods. The analysis of the Fire Action is presented in Chapter 2.

'Participation' has been a common word to those who are involved with

development, and it is used by different actors (such as social groups, NGOs,

government, researchers, donors), in different forms, for very broad purposes. Because it

has became a vague label, different typologies were developed to specify different kinds

and levels of participation, and interests of the social actors involved (Bordenave 1992;

White 1996; Ingles 1999; Stone 2003). The common ground among those typologies is a

differentiation in kind (from passive to active), and level of participation (decision-

making power at the different activities). In general, the authors agree that the higher the

power in decision-making, the higher the participation. As Schmink points out, "the

degree of participation by different local groups in project decision-making and

implementation is a key factor in empowering local groups to defend their own interests

and to develop and adapt the institutions required to sustain natural resource management

strategies over the long term" (1999:6).









The methods developed under the participatory approach were based on Paulo

Freire's Pedagogy (1986), in which consciousness, the self guide, is a process "by which

human beings participate critically in a transforming act" (Freire 1985:106).

Consciousness, therefore, is an active process. "Knowledge that exists today was once

only a possibility, and it then became a new knowledge, relative and therefore successive

to yesterday's existing knowledge" (Freire 1986:115). What Freire argues is that new

knowledge, even proven scientifically to be the most suitable for a certain situation,

cannot be deposited in people's minds (known as "banking education"), but has to be

connected with previous knowledge, and previous understanding of the subject reality.

This connection between previous and new knowledge is made via a thematic 'problem-

posing' methodology, possible only through dialogue.

According to Freirian Pedagogy, in thematic problem-posing situations, the role of

the educator is to propose problem situations or themes. The starting point must be the

present, concrete situation of the learners, not only at the intellectual level, but also at the

level of action (Freire 1986:85). Learners' ways of explaining the world involve their

comprehension of their world and presence in it (Freire 1997:76). Educators should help

the learners to arrive at a more critical view of their reality by, through dialogue,

analyzing the dimensions of their reality, that ultimately are dimensions of total reality.

This analysis involves the exercise of abstraction, through which, by means of

representations of concrete realities, learners and educator seek knowledge of that reality.

The instrument proposed by Freire for this abstraction is codification, or representation of

the existential situations of the learners (Freire 1985:51). Codification may take the form

of a photography or sketch that represents reality, or a reality constructed of a proposed









theme by the learners-a dimension of the reality. Codification, thus transforms what was

a way of life in the real context into an "object" in the theoretical context. Therefore, the

learners, rather than receiving information, creatively analyze aspects of their own

existential experience, represented in the codification.

In the case of Fire Action, the thematic problem-posing proposed to the colonist

peasants at meetings was "how to use fire better", which became synonymous with "how

to reduce losses caused by escaped fires". The codification exercise was developed with

colorful drawings on cardboard tags ("Burning Dynamics") representing land use types

(such as primary and secondary forest, pasture, slashed vegetation, perennial crops) and

elements usually present in burning situations (people, houses, firebreaks, wind, sun,

buckets, fence). Those drawings were mobiles, combined to represent burning situations

(real or proposed), leading to peasants' re-codification of their knowledge through critical

analysis of their own situation (for more details see Chapter 2).

Freire states that one of the important points of the consciousness process is 'to

provoke reactions of the world, not as a "given" world, but as a world dynamically "in the

making" (1985:106). Humans are praxis-the unity of action and reflection-as reflection

and action, which truly transform reality, are the source of knowledge and creation

(Freire 1986:91). Neither pure action (not guided by reflection of the problematic

situation) nor reflection (theoretical abstraction of reality) alone can lead to critical

analysis. In this process, dialogue is the encounter in which reflection and action of the

dialoguers are addressed to the world which is to be transformed.

The Fire Action was planned in a way that each dry season would be a cycle of

learning by praxis (learning-cycles): reflection about the current reality would generate









actions, which in its turn would generate a new reality, again subject to reflection, in a

continuous knowledge process. This means that solutions were not fixed, but had to be

incessantly confronted with reality. This led to increased awareness on the part of

peasants: through reflection on a given practice, ingenuous curiosity became self-

conscious so as to advance to the critical stage (Freire 1998:43). Peasants, as active

actors through the dialogical method, incorporated and appropriated the solutions

developed by themselves. Many of the peasant narratives express that it was the first

time they were using certain practices proposed by the Fire Action. Many of the peasants

that were actively involved in the Action since its beginning, after only one year, referred

to the Fire Action as their own action, probably because of Fire Action incentives that

each community developed their own set of rules. Peasants involved in the Fire Action in

a more passive manner (absent during meetings) referred to it as an IBAMA's action (see

Chapter 4 for Fire Action evaluation). A key aspect of the approach was therefore to

develop and support forms of communication and social organization that would support,

and multiply, participation in fire management activities.

Thus, Freire's approach dictates that in order to overcome a problematic situation it

is necessary to go through it, and not to stay away from it, or deny it. In this context,

effective solutions involving peasant societies can be reached by a learning process, in

with basic elements are: respect for local knowledge, dialogue, critical reflection, and

collaboration.

The FATA and LASAT, through their history of support and collaboration with

local peasantry, represented key institutions for the development of a participatory

approach to fire management for peasants: trust in the institutions' intentions and









methods; open channels of communication and representation; and faith in peasants'

ability to overcome the problematic situation.

The FATA and LASAT supported a technician to coordinate the Fire Action, and

provided transport, food, and lodging for regional meetings, and educative materials. No

peasant received any payment for fire management. Institutions and peasant leaders

believed that peasants should get involved only if they believe it was worth it to them, but

not as a direct economic source. Payments create dependency on outside inputs, if

benefits are not linked to resource management and conservation (Schmink 1999:6).

They can discourage others who receive no payment to get involved, and weaken the

credibility of peasant leaders' real commitment. Especially in fire management cases, in

which fire can result in fines and jail for its users, trusting in leaders was a key factor in

conflict resolution situations. The Fire Action strategy was supporting each community

to develop their own set of rules for fire management, including mechanisms of

compensation in cases of losses caused by fire. Some peasants were not used to

participating in community activities for fear that leaders were working for IBAMA, a

governmental institution historically perceived as "police," acting only in enforcement,

and always against peasants. Leaders' volunteer work, supported by trusted grassroots

organizations, were strong arguments to convince outlying (non participant) peasants to

get involved in the Action.

In sum, in presenting learners' own objective reality (how and where they are),

using a problem-solving method, with a thematic investigation, peasants begin to revise

their previous views of their real world through codified situations. They achieve an

understanding of their previous knowledge. In so doing, they expand the limits of









knowledge. In this process, the educator exercises with the peasants a critical evaluation

of their world view, resulting in their clear involvement with the real world in

transformation (Freire 1985:33-34). This process promotes peasants' decision-making

participation, once they get involved in creating solutions.

This study evaluated changes in fire practices in 1998 and 1999 in two

communities involved with the Fire Action, and observed burning situations in 2000.

Despite the short time, this study found changes in fire practices at individual (family)

and collective (community) levels, and fire losses were low after the Fire Action started.

In the Sdo Francisco do Itacaiunas, involved in the Action since its beginning, changes

were stronger than in the Cupu community, involved one year later.

Organization of the Study

This introduction has discussed how the government strategy for controlling fire

use in Amazonia based on a top-down approach, remote monitoring, and enforcement

towards fire users is unlikely to be effective in bringing consciousness to colonist

peasants. I argue that colonists possess empirical knowledge on fire management, which

has been ignored by many government representatives, technicians, and academics.

Another point stressed so far is that a participatory approach in which peasants share

decision-making power, analyze their reality critically, are challenged using problem-

posing exercises through action-learning cycles, and encouraged to develop and

implement community solutions, will be more effective in managing fire. Instead of

defending or attacking fire use, this study assumes that fire use is the option for the

majority of colonist peasants, and they have to be seen as partners in the process of

constructing better fire management strategies-including participating in the definition of

"'better" fire management.









In Chapter 2, I present and analyze a participatory Fire Action, coordinated by two

grassroots organizations, FATA and LASAT, in the Maraba region, which involved

peasants as well as other grassroots organizations, and some indigenous groups, during

1998, 1999 and early 2000. I address the participatory approach adopted, based on Paulo

Freirian pedagogy, the process that originated in the proposed Fire Action, and the

recommended practices for fire management. This set of recommendations developed by

the peasants themselves provides the framework for the subsequent analysis of actual

practices used in fire management by two communities during two successive dry

seasons.

Chapter 3 presents descriptive information about the two communities and the

families studied. Families' cultural interpretations of living in communities, and the

importance of forest and fire are also discussed. This Chapter demonstrates some key

aspects of local knowledge and situation that are typically overlooked or misunderstood,

and which should provide the point of departure for fire management programs.

Chapter 4 evaluates the impacts of the FATA/LASAT Fire Action on the two

peasant communities for 1998, 1999, and 2000 dry seasons. The Sdo Francisco do

Itacaiunas community got involved in the Action since it started in 1998, while the Cupu

community in 1999. The history of losses suggests the positive impacts of the Fire

Action on peasants' changes regarding fire management. The assessment of

implementation of the Fire Action's recommendations focused on different phases: before

burning, on the burning day, after burning, and controlling accidental fires. Changing

practices within and between communities, are compared in order to identify impacts of






38


the Action on families' fire practices. This Chapter also interprets peasants' narratives on

the Fire Action government and fire Action evaluations.

The last chapter, Chapter 5, summarizes the main findings of the study, confronting

the three main research questions regarding government actions, colonist empirical

knowledge, and the participatory approach analyzed. Other experiences on community-

based fire management in Asia and Africa are compared with this study, and suggestions

are made for the management of fire in the Brazilian Amazonia.














CHAPTER 2
COMMUNITY FIRE ACTION

Introduction

An office is a small and compact setting, isn't it? And the Amazonian environment
is a large, vast one. Thus I would ask the [government] agencies not to limit their
actions to their own offices, but instead to make themselves present in the
communities too18

Mr. Valdir Silva, leader of the Sao Francisco do Itacaiunas community, during the
first regional workshop.

In the Maraba region, the history of colonists' relationships with both non-

government and government institutions reflects on their present partnerships, their trust

in them, as well as their willingness to engage in government policies, especially the laws

which directly affect their productive practices, as the case of fire use. Conditions

influencing the transformation of landless migrants into colonist peasants, specifically in

the Maraba region are embedded in a long political struggle for the creation and the re-

creation of their identity, a process shaped by peasants' forms of organization and

resistance, as well as their strategies of permanence on the conquered piece of land. The

above speech reflects the governmental history of absent support to peasants and their

grassroots organizations. Government is criticized as remote and bureaucrat, creating

office solutions distant from peasant realities.

In this context of fragile links between peasants and government, I argue that top-

down actions are even less likely to be adopted. Complementarily, this study supports

the idea that colonist communities have valuable empirical knowledge, and they should

18 Translation by the author. Original Portuguese texts are included in Appendix B.









participate actively in decision-making concerning fire use and management. It does not

defend the notion that peasants' actual fire use is perfect, but on the contrary assumes that

changes are needed and argues that only i/ ih peasants through participatory approaches

can better practices and strategies be developed. The participatory approach adopted

follows a problem-posing methodology, based on codification of the actual reality to

develop consciousness about the reality, and learning-action cycles. This methodology is

based on Paulo Freire Pedagogy (1986), and it is only possible through continuous

dialogue among partners.

This Chapter is organized in two parts: The first part describes the two grassroots

institutions coordinating the Fire Action: the FATA (The Tocantins Araguaia Agrarian

and Environment Foundation) and the LASAT (Tocantins Social-Agrarian Laboratory).

The second part presents the Community Fire Action conjointly coordinated by the

FATA and the LASAT, an innovative experience of community fire management which

involved peasants as well as their grassroots organizations, and some indigenous groups,

during 1998, 1999 and early 2000. The text addresses the participatory approach

adopted, and the process that originated the proposed Fire Action, as well as summarizing

the recommended practices for fire management. Chapter 3 presents the characterization

of the communities and their families. The impacts of this Action on two peasant

communities will be evaluated in Chapter 4.

Peasantry Historical Construction in the Maraba Region

The Maraba region has a long history, and migrants arriving, their strategies, and

their means of production, are part of this history. This region is connected to an

extensive process of expansion started in the 16th century, which ran from the coastland

into the inner continent. Velho (1982:29) described how this process relates to the









economic cycles of colonial Brazil. As a result of its geographical location (at the mouth

of the Itacaiunas river into the Tocantins River, and highways) and its richness in natural

resources, the colonization of the Maraba region was marked by several cycles of

exploitation (brazil-nuts, rubber, gold, and gems, cattle, timber, minerals and electric

power dams). Those economic cycles induced and still generate different migration

cycles which directly influenced the composition of its population. During the dictatorial

military regime (1964 to 1985), it was considered a national security region due to the

occurrence of the Araguaia Guerrilla19, when social movements were severely repressed.

The Church played a key role supporting peasants and other powerless groups in a

struggle for their rights. In this context, a brief review of the main economic cycles and

its protagonists should yield a general picture of the colonist peasant's formation.

Exactly when Maraba village was founded is still debated-June 7 of 1898

according to Velho (1972), and 1895 according to Emmi (1999). However, authors agree

that it started with the establishment of a commercial store during the rubber boom

period. The merchants used a system known as aviamento, in which they would

determine prices of rubber or nuts exchanged for goods in their store-mainly groceries

and working tools-at very high prices. This led the peasants to accrue never-ending

debts, carried from one season to the next one, making the worker a new type of slave

(Velho 1972:41). Due to the distorted relation with the extractors based on indebtedness,

merchants obtained increasing power and acted as "landlord", later becoming the

"owners" of rubber and brazil-nut trees. Around 1919, following the rubber bust, the


19 The Guerrilha do Araguaia was a resistance movement coordinated by the Partido Comunista do Brasil
or Communist Party of Brazil, sited in the area between Maraba and ConceiiAo do Araguaia, in direct
confrontation with the military dictatorship from 1972 to 1975 (for more information see Schmink and
Wood 1992: 72-74; Alves Filho 2000).









exploitation of brazil-nuts grew stronger and used the infrastructure provided by the

rubber industry, and for decades it remained the main product exported by the state of

Para (Velho 1972:45). Emmi points out that in the late 1950's the brazil-nut business

(from extraction to commercialization) in the region was concentrated in the hands of

brazil-nut tree 'owners' in a perpetual system of "taking" lands to themselves or sistema

de aforamento (Emmi 1999:70-71). Those two economic cycles (rubber and brazil-nuts)

consolidated two main social groups in the region: a mass of low-paid workers and an

oligarchic class, where a few local families held political positions in a monopoly, as well

as credit (capital), means of transportation and commercialization, natural resources, and

land. Peasants were completely excluded, as were other small brazil-nut entrepreneurs,

and the indigenous (Emmi 1999:151-152). Despite this system of unequal distribution of

wealth, there were no land conflicts officially registered in Para before the 1960's.

Several generations of peasants lived without any official document to give legal power

to their rights over their own land (Treccani 2001:307). This is explained by the fact that

"ownership" was associated with the products extracted above ground and their

accessibility, not with the land itself In addition, accessing land deeds was (and still is)

not a possibility for peasants that are, in practice, excluded from the legal system due to

its high costs (they are required to have a lawyer representing them and to follow up on

the legal suit), to a sluggish judicial structure, and to the police that generally connive

with thugs (Santos 1984:457-458).

It was particularly during the military regime that the Brazilian government started

to provide for the means of speeding up the occupation of the Amazonian area. The

government intended to assure national sovereignty using policies that included









investments in local infrastructure (such as federal and state roads), interregional

migration and economic development (Browder and Godfrey 1997:68). Those policies

were organized outside the region, in the National Integration Plans or Planos de

Integrag o Nacional and had as consequences deep changes in the productive structure.

Since the region is strongly connected with the history of extractivism in Amazonia, the

last few decades were marked by the decline of the traditional extractivism involving

brazil-nuts, rubber, fur and small-scale crystal- and diamond- mining in rivers, giving

place to new forms of extractivism: logging, large-scale mining (Hebette and Moreira

1997:12), and agro-cattle ranching. The so-called Large-Scale Projects carried out by

manganese, bauxite, nickel, cassiterite, gold deposits, wolframite iron, copper and caulim

mining companies were associated with hydroelectric plants, which supported them with

power (Schmink and Wood 1992:66-67). Logging activities grew as the frontier

advanced, along with agriculture and pasture implementation.

Through the National Plan of Integration, with government calls in the media, an

example of which was the famous "land without people to people without land", the

Amazonian frontier attracted people from all over Brazil and from diverse social classes.

Since Brazil was facing increasing inflation rates, transferring capital to Amazonia

represented an excellent option for investors, especially those investing in the cattle-

ranching industry (Santos 1984:452-453). This caused a big and chaotic "land rush",

where fake deeds and frauds were common, and multiple claims to a same piece of land

occurred (Santos 1984:453). One of the most serious consequences was the closing of

the traditionally "open" frontier to migrating landless peasants from Northeastern, who









started to face their "last frontier", "from where they have nowhere to go and nothing else

to lose" (Treccani 2001:308).

In Maraba, roads made the city a strategic center for the region and brought

significant changes to its social and economic structure. The brazil-nut concessions or

castanhais were turned into large farms, and official colonization brought millions of

colonists who no longer were dependent on traditional local oligarchies. Former brazil-

nut extractors became posseiros, or squatters, making the pressure for land ownership

even stronger (Treccani 2001:313). Most migrants were landless people expelled from

other regions, who left in search of land and of labor opportunities in road constructions

and other large construction projects, as well as small-scale mining positions (Hebette

1991:200). Frontier expansion in this scenario increased land conflicts between

newcomers and old forms of land occupation, and among newcomers themselves. This

increase provided the elements for the most violent land conflicts in Brazil.

The social consequences of this development model are clearly reflected in the

concentration of the means of production, such as the land itself, in the hands of

multinational entrepreneurs or big large-scale farmers. The ways in which these policies

reflect on the state of Para is the object of a deep analysis conducted by Girolamo

Treccani in his book Violkncia e Grilagem: Instrumentos de Aquisigdo da Propriedade

da Terra no Pard (2001). For example, Treccani shows that in 1996, 82% of the farms

with areas of 100 ha or less represented 169,273 farms, corresponding to only 19.2% of

the lands (totaling 4,328,158 ha). On the other hand, farms with 5,000 ha and larger

represented only 0.2% of the total numbers of farms (419) but occupied 31% of the entire

area of 7,138,104 ha (Treccani 2001:431). This difference in farm size is the









consequence of a historical process of political and economical favor for certain social

groups in detriment of others, fostering not only the widening of the economic gap, but

also and more importantly, a strong cultural distance.

In this context of land concentration and violence against peasants, the Land

Reform has being made by peasants themselves. They occupy latifindios or large farms,

and, supported by grass-roots organizations such as unions and pastoral commissions,

struggle to transform the occupied land into peasant areas. Politically, peasant leaders

call this land occupation process 'conquering land', while ranchers, elite, and media call it

"land invasion." In the Maraba region, the historical process of peasantry consolidation

is reflected in the amount of conquered lands:20: from 1% in the early 1980s to more than

30% of the properties in 1998, as result of their resistance, supported by a few allies

(Wambergue 2000:41). Maintaining conquered lands as peasant's communities is a

bigger struggle for grassroots organizations, and one of their strategies is to support

productive activities that help the family's permanence in their lands. Uncontrolled fires

have became a threat to peasants' means of production due to losses caused in forested

areas, crops, pasture, and goods. Therefore, managing fire in a way to avoid local losses

has became a concern to grass-root organizations.

Peasant Organizations

Especially during the years of military regime in Brazil, unions and associations not

connected to the government were forbidden and fought against with great violence.

Two of the tenets of the military government were undermining a movement called


20 Grassroots leaders and some studies (Trecanni 2001) affirm that in Brazil there was never a Land Reform
but only colonization projects. Colonist peasants' institutions use the political term "conquering" to refer to
land occupation actions and their struggle for pressuring governmental agencies to recognize their rights
over the occupied land. On the other side, media and elite use the term "occupation"'.









Peasants' Leagues or Ligas Camponesas, and the halt of the land reform program that

started with Brazil's previous democratic president, Jodo Goulart. Both led to the death

or "disappearance" of several community leaders (Oliveira 1989:31). In this scenario, the

Catholic Church experienced a constant evolution of its social concerns (Catholic Church

Archdiocese of Sdo Paulo 1998: 124), and grassroots organizations tied to the Catholic

Church were often the only means for peasants to voice their needs and their opposition

to events occurring in the frontier areas (Schmink and Wood 1992:180). The so-called

Journey Church or Igreja da Caminhada, built by supporters of the Liberation Theology

movement, opened its doors to peasants and, in 1975, established a committee called

Comissdo Pastoral da Terra or CPT (Land Pastoral Commission) whose goals were "to

interconnect, assist, and activate the support to rural social movements" (Almeida

1991:261). The Church, together with the CPT at a community level, created ecclesial

"base" communities called Comunidades Eclesidsticas de Base (CEB), in which local

leaders would conduct religious services and facilitate community organization. From

this point on, the military regime directed its violent repression not only to peasants and

indigenous people, but to priests, nuns, and union leaders, who were starting to get

reorganized.

After five years of military regime, a CEB was started in the region and a CPT

office was set up in the city ofMaraba. Several local leaders strengthened their

commitment with the CEB movement and later became union directors, regional officers

with the Worker's Party (PT), with the SDDH (Para State Society for the Defense of

Human Rights) with lawyers supporting peasants in judicial cases, the FETAGRI (Rural

Workers Federation), and the University of Para, among others. In these circumstances,









some of the FATA/LASAT founders (inside the CAT Program) played key roles, for

example, coordinating the CPT/Maraba, heading four peasant unions, and teaching at

universities. This will be discussed in the next section.

Although the Rural Workers Unions (STR's) were created under the military

regime to control social movements organized in rural areas, the STR board was

gradually passed on to those with a strong commitment to the rights of peasants, long

before the regime fell. Nevertheless, the fall of the military government in 1985 gave

place to greater violence against colonists, for the powerful landowners were responding

to the new presidential decision to promote a nationwide land reform, where the region

known as Bico do Papagaio would represent a priority. The "big" farmers set up an

office to start the Rural Democratic Union (UDR), and large ranchers in the region

started recruiting private militias to fight land invasion, which was likely to be carried out

by peasants but also supported by government agencies attempting to expropriate the

areas regarded as their own (Schmink and Wood 1992:188). Between 1989 and 1999 the

average number of murders resulting from land conflicts in the Maraba region was 120

times higher than the country-wide average (Forum das Entidades pela Reforma Agraria

do Sul e Sudeste do Para 2001:4). This means that a leader or peasant living in the

Maraba region and engaged in a land conflict between 1989 and 1999 had a much higher

chance to be killed than any other worker living in any other Brazilian state. This history

of violence against peasants and the impunity of farmers and loggers21, both supported by





21 From 1985 to 2003, the Land Pastoral Commission (CPT) registered 1,003 violent cases related to land
conflicts, with 1,349 dead victims. Only 75 cases went to judgment, with few considered guilty (CPT
2004).









the legal and the executive systems, represent the main obstacle for partnership

agreements to be implemented between these social groups in the Maraba region.

Peasants' history is marked by active resistance in order to guarantee their rights, in

spite of harsh conditions and violence. Due to the historical distrust between government

and peasants, top-down governmental actions imposed on peasants in order to control

their natural resources management, especially those related to their main source of

income, such as fire use, are likely to fail. This study argues that actions that see peasants

as partners, with decision-making power, through a methodology that respects and

supports their culture, with effective channels of communications and conflict resolution,

are more likely to succeed.

FATA and LAST

This section describes the two institutions coordinating the Fire Action, FATA and

LASAT. Their history, as well as their goals and achievements are closely connected

with peasantry development in the region that directly reflected on peasants' willingness

to adopt the productive practices promoted by them. FATA was composed by peasant

union leaders (STR's), and LASAT by researchers connected to the local university.

Both FATA and LASAT were non-governmental institutions created in 1988 under

a wider agricultural and environmental program called "Tocantins Environmental Center"

(CAT). At the time of its foundation, the CAT Program was a partnership agreement

celebrated between four Rural Workers' Unions (STRs) and the Federal University of

Para. The main goal established by the agreement was "to assert their [peasants']

identity, to achieve full citizenship and to express their determination to live and to take

their rightful place in the construction of a fair and fraternal society" (Hebette 2000:xxii).

In a moment when land was being conquered by peasants through struggle, peasants'









challenges were to build good conditions to establish themselves, as well as to produce

and guarantee social reproduction in their lands. From the goals set out by the CAT, it is

clear that justice and fraternity were principles linked to those defended by Liberation

Theology. This may be due to established partnership with the Land Pastoral

Commission, also reflected in the composition of the CAT founding board: CPT's first

coordinator in the Maraba region, Mr. Emmanuel "Manu" Wambergue; four peasant

union directors who grew as leaders inside the CEBs; and a professor from the local

university, who was assistant to the CPT, Mr. Jean Hebette. Also, STR Directors had

grown up inside the church groups. The identification of the CAT Program with the

peasants' struggle is well reflected in Manu Wambergue's evaluation of the role played by

the CAT Program on its 10th anniversary as "one of so many fruits that flourished from

the peasants' struggle in the region" (2000:41). Peasants' struggle was not violent; their

form of resistance was their resilience and active search for new solutions, despite harsh

conditions of exclusion from government incentives.

Communities' Social Organization

The two communities evaluated had been partners with FATA and LASAT. In the

Sdo Francisco do Itacaiunas community, in September 1998, FATA's perennial crop

group (for implementation of perennial crops) led the creation of the community

association, the AGRAF, in an attempt to work more closely with families' needs than did

the regional association, created in 1994 to facilitate access to federal government loans.

The Association's main goal was to "carry out sustainable development activities in their

community" (AGRAF 1998). They borrowed the necessary money for bureaucratic









expenses (around US$ 512)22 from FATA and paid it back later with their labor-one

week's work of ten men-at the FATA headquarters. Very often, the cost of an

organization's creation is supported by local politicians, creating a certain relation of

dependency on the part of the peasants and the organization. AGRAF's representatives

proudly explained how they had paid off the loan with their own labor. The recognition

of the Sdo Francisco do Itacaiunas community as a Projeto de Assentamento (Area for

Land Reform) was a result of its leaders' organization for solutions. The community's

leaders participated and supported families' participation in demonstrations organized

regionally by the STR and FETAGRI when families camped out at the INCRA

headquarters in Maraba, an event called Grito da Terra, or 'Cry of the Land'. In 1999,

77% of the families in this community had at least one member in the community

association, and 95% were members of the STR. In 2000, AGRAF's members elected a

new board (6 men and 1 woman), all of whom work as volunteers.

Families at the Cupu community had a close relation to the COOCAT through one

of its directors who lives in the community. COOCAT has supported the marketing of

agai fruits,23 among other products. Since 2000, they have been helping with the

preparation of a regional proposal for the planting of coffee, along with the Jacunda STR

and Jacunda municipality representatives. The peasants' regional association was

working with COOCAT and STR to achieve official recognition of their existence as a


22 In September 1998, the average conversion rate was US$ 1.00 = R$ 1,17 (Banco Central do Brasil
2004).

23 Acai (Euterpe oleracea Mart.) is a native Brazilian palm in Amazonian region whose fruit is much
appreciated in regional markets, which are growing in national interest. The palm apical bud is appreciated
nationally and internationally for palm heart, but the extraction of the Atlantic Forest species (Euterpe
edulis) requires felling the tree. This Amazonian species is very promising for sustainable harvest because
of its sprouting capacity.









community (Projeto de Assentamento), supporting them in acquiring land title and loans.

In the Cupu community, the majority of the families (62.1%) were members of a regional

association created to facilitate access to federal bank loans (FNO) in 1996. Seventy-one

percent were members of the STR. There was no specific community association. In the

past, the community had worked in the FATA's perennial crop project. Ten families took

part in LASAT's Community Timber Project. As an indicator of their hope in this

Project, by 2000 none of the participants had sold any trees to middlemen since the

Project started in 1997.

FATA/LASAT Community Fire Action

The FATA and LASAT direct interest in controlling fire use was to protect

alternative projects for peasant productive system: FATA perennial crops

implementation24 and LASAT forest management.25 Local peasant organizations

believed that uncontrolled fires threatened alternative production systems in slash-and-

burn agriculture and cattle ranching, consequently leading to new migration cycles,

conflicts and suffering. The proposed Fire Action did not aim to encourage families to

stop using fire in their productive systems, but rather to promote a more responsible use

of fire, specifically preventing and controlling undesirable losses.





24 Between 1991 and 2000, FATA implemented the PAF (Agricultural-Forest Project) to establish and to
market perennial crops such as fruits and lumber. Later, activities of horticulture, honey production, and
Acai palm (Euterpe oleracea) management have also been incorporated. In 1998, when the Fire Action
started, 44 communities, including three indigenous communities (a total of 76 groups) were working with
FATA in the different activities (Carvalho 2000:65).

25 LASAT started a Community Forest Management project in 1996, initially involving three peasant
communities from three different municipalities. The main goal of this project was to promote sustainable
lumber extraction and processing by peasants, which should result in less damage to the forest and in
higher profit.









Besides the grassroots institutions' and leaders' reasons to control fire, peasants in

general had many questions about the new official fire regulations. The dry season of

1998 was the first after the fire disaster occurred in Roraima, and the changes introduced

by IBAMA to the fire management laws were: the requirement to submit an official form

to obtain a burning permit, the obligation to provide details on the type and the size of

firebreaks, and for the first time, a total ban on fires in municipalities with high rates of

burning (measured by the INPE National Institute for Spatial Research). Although not

all rules were new, the latest change introduced by IBAMA was to massively divulge,

through the radio and TV, the requirement to obtain a license to burn, of the fire use ban,

and that IBAMA staff would be enforcing the fire law in the field. In the communities,

FATA/LASAT technicians were initially thought by many peasants to be IBAMA

members, who were disappointed to find that the technicians could not provide them with

any official burning permits, or even get them to be heard at governmental levels. They

wanted to explain to IBAMA not only their reasons but also their very need to burn,

which is clearly expressed in the extract below:

Many have burned, but I have not and many others have not either. We expected
someone to give us explanations, because we do not understand the law, and we
want to find a way to work legally. Because in a place like this, with no roads and
no assistance, if a poor peasant has to pay a fine for burning his field, it would be
better to bury him straight away. So, let us say a family cannot harvest their field,
how would they live? This is why I want to know what we have to do in order to
burn. Do we have to make firebreaks, to get a license? I want you to explain it to
me so that I will understand it. If you had come earlier, I believe nobody would
have burned without permission. Our group always works together to do things the
right way. But we never had anybody come here to explain things to us.

Mr. Jodo Pinheiro, Jabuti Community, Oct 14th 1998, community meeting
(Carvalheiro 1999:18).

Despite peasants' investment in other productive systems, fire was their main

productive tool for agriculture and pasture maintenance, and they wanted to know what to









do in order to produce their crops without becoming a "criminal". Peasants complained

that it was hard for them to have access to information about government regulations.

For example, in the whole Maraba region there was only one IBAMA agency bearing

responsibility for all environmental issues in the region (from illegal logging to river

pollution), whose technicians were not used to visiting colonist communities. The radio

was helpful to bring general awareness. However it did not explain the official laws and

regulations, and the necessary steps that should be taken in order to bum.

The Fire Action adopted a participatory approach, based on Paulo Freire's

pedagogy (Freire 1986), as presented in Chapter 1. The Action adopted the problem-

posing methodology, whose theme was fire use, more specifically "how to manage fire in

order to keep it only where it is planned" (agricultural fields and pasture maintenance).

The resulting recommendations made by the Fire Action were intended to be closely tied

to their social, economic and ecological realities, and given the urgency of the problem-

established perennial crops and forest areas were being burned-also to produce short-

term answers. Below is presented the application of the participatory approach adopted.

Process for defining Practices for Fire Management

The Fire Action stemmed from a study proposal made in 1997 to LASAT by the

author of this study, and the interest of LASAT, FATA and peasant leaders for better fire

management. The diversification of peasants' productive systems with community forest

management and investments in perennial crops depends on the ecosystems, which are

very susceptible to fire, due both to fuel accumulation and to their proximity to the

ecosystems where fire is present as an agronomic tool. The FATA/LASAT Fire Action26


26 For the purpose of this thesis, the Fire Action coordinated by FATA and LASAT will be referred to as
FATA/LASAT Fire Action, or just Fire Action.









started during the 1998 dry season and ended at the beginning of 2000, from lack of

funding. The Fire Action had economic support for the Action coming from outsiders

(distant donors), was implemented by technicians from the local grassroots institutions

(FATA and LASAT), and planned, developed, and evaluated with peasants at regional

and community levels. Documents produced (such as reports) were shared with

community leaders and unions. Educational material was developed by technicians and

peasants. The author of this thesis coordinated the Fire Action from 1998 to July 1999.

The proposal was based on peasants' active participation, because they were the

ones who had developed knowledge on using fire in their environment, and therefore

were skilled to evaluate what was not working well, and to be important partners to

improve fire management. Another reason to get peasants involved is that they are the

ones who ultimately take the final decisions on fire management. If they are actively

involved analyzing the situation, proposing and approving changes in their practices,

those changes will be closer to their realities (infrastructure, social, and productive) and

thus more likely to be incorporated into their activities.

The action-learning cycles support peasants' development of their own proposals,

and recognized that changing practices is a dynamic process, and therefore has to be

analyzed periodically. Peasants had a reading of their reality, and the starting point of the

action-learning cycles was their actual understanding of fire. Actions of fire management

were carried out at the community level, and learning (evaluation and planning) in

meetings at community and regional levels. Each dry season corresponded to a full

learning cycle (see Figure 2-1). Regional seminars took place in the beginning of a dry

season, for evaluation of the previous and planning of the following, and at the end of









that dry season, also for evaluating and planning, closing a cycle by incorporating lessons

learned. The Fire Action worked for two action-learning cycles, in 1998 and 1999.


Figure 2-1. Representation of the FATA/LASAT Fire Action showing key activities of
the participatory Fire Action. Action-learning cycles started in the beginning
of a dry season, with activities of evaluation, planning, and dissemination at
regional (workshops) and community level. Actions were taken at the
community level, and evaluated in the following workshop (By author).

The Fire Action did not establish any instance of centralized decision, nor did it

plan to follow any pre-defined agenda. Instead, it sought to work flexibly to reach and

involve the families' realities at the community level, at same time promoting community

organization. As form of incentive, grassroots institutions (FATA, LASAT and STR)

declared they would support organized communities in case of conflicts with IBAMA

regarding fire use, and the Fire Action's technician would visit those better organized









more communities regularly. Subsequently, regional workshops would draw general

lines of action, but it was at the community level that more specific agreements and

recommendations should first take place, since at this level those agreements are made

and monitored.

During the dry seasons, the Fire Action technician visited some of the communities

in order to monitor their actions. After the dry season was over, community

representatives presented their community experiences at regional workshops. More than

40 communities were invited, with support for transportation, food, and lodging. This

flexible strategy was chosen because it gave families an opportunity to play an active role

at the community level, where they could change their community, shaping it to their

reality, while at same time they would feel important for having played an active role and

for the motivation to participate (Bunch 1994:69).

The meetings held at the community level were key to the participatory approach

adopted because they created opportunities for conducting discussions, accessing

information, exchanging experiences, planning, as well as for dissemination and

evaluation of the proposals. In general, every family had heard something about fire

management and the penalties imposed by the government for its use. However, they had

always been uncertain about legal details. Many families expected that IBAMA's

burning guidelines would be presented to them in 1998. Given that all families use fire in

their productive systems, community meetings attracted many in search of information on

fire use and of an opportunity to complain about the government's proscription of fire

use. The government's prohibition of fire use27 resulted in a dilemma for peasants: they


27 Since 1998, the government has prohibited fire use during the dryer weeks of the dry season, in general
for eight weeks. Even those who got permits are prohibited to use fire during this period.









do not want to break the law on the one hand; on the other, they depend on using fire for

their social reproduction. They did not want to go against the law, but they did need to

burn. After receiving no information from IBAMA by the end of the dry season, they

started burning in a careless manner and without an effective plan (Carvalheiro, 1999:21-

22). This condition allowed for deep reflections about their own situation.

The visits paid by the Fire Action coordinator to the communities reinforced the

work done by community leaders because, in addition to promoting discussions, the

peasants who did not attend the regional workshop could access information by asking

questions directly to the technician and comparing answers to clarify some of their

uncertainties. Some families are less involved in community organization and do not

trust leaders or their intentions regarding fire management, suspecting that leaders would

receive some form of payment from IBAMA to control the use of fire by families. This

type of suspicion represented a very common conflict among communities in the Maraba

region. The ban imposed by IBAMA on the use of fire represented bigger losses to

peasants than uncontrolled fires usually did, because for them there is no production at all

without fire. Therefore, for those with poor access to information, the regular payment of

a "salary" would represent a logical justification for some peasants to go against their

own class, i.e., to support fire control. The technical visits strengthened the legitimacy of

the work done by the leaders, and emphasized the fact that none of the work done by

peasants was done for money, since they worked as volunteers, and FATA/LASAT only

assumed the cost of transportation, food and lodging for regional workshops held at the

FATA Center. The fact that the leaders involved in the Fire Action were generally the

same ones taking part in other actions promoted by FATA and LASAT helped to









encourage other peasants to believe in the proposals of the Fire Action, and to play an

active role in it.

Regional one and three-day workshops were held at the FATA Center, and brought

together peasants and technicians to discuss and evaluate their community actions and

government laws, to plan strategic actions, and to present and discuss their own proposals

with the representatives of both governmental and non-governmental organizations. The

Fire Action coordinator also held meetings at the STR's (Rural Workers Unions) and

visited governmental institutions (INCRA, IBAMA, SEMMA, EMATER, Public

Defensor Office).

First Action-Learning Cycle

The first action-learning cycle started in a regional meeting, aiming to analyze the

present fire use situation based on peasants' understandings, challenging their knowledge

with information on fire law and fire occurrence in the region, and drawing up with the

peasants a first proposal for fire management to be carried out in the coming dry season.

At the community level, during the dry season, actions consisted of meetings, to inform

and discuss proposals defined in the regional meeting. At the end of the dry season, the

closing cycle would be an evaluation of the action carried out, and planning for the

following year dry season. For the closing regional meeting of the first cycle, leaders

decided to invite other grassroots institutions related to peasants, as well as governmental

institutions-the presence of IBAMA would be crucial for peasants to present to them

their developed proposals. Details of each step of the first action-learning cycle are

presented as follows.

The first action-learning cycle started at the beginning of 1998 dry season, during

the workshop on "Perennial Crop Management" held at the FATA Center, when an entire









day was dedicated to discuss fire use with peasant participants. In this first meeting,

there were a total of 31 peasants (from 20 communities), and 4 indigenous (from 2

tribes), all connected to the Perennial Crops Project, and representatives of two

communities also involved with the Community Forest Management project. There were

also 13 technicians (8 from FATA, 4 from LASAT, and 1 from EMATER/Itupiranga).

Instead of presenting to the peasants the methodology of the coordinator's previous

experience in community fire management in Paragominas (Mattos et al. 2002), and then

discussing how it could be adapted to their conditions, the strategy used was exactly the

opposite: first the discussions were on what participants would do in order to prevent

losses caused by the fire, and only then was the other experience presented. This strategy

was very positive because participants were given the opportunity to see that the results

achieved by their discussions were similar to a consolidated experience carried out in

other Amazonian regions, which raised their self-esteem. As discussed in Chapter 1,

problem-posing methodology empowers participants by raising questions on their

situations and by leading to solutions based on their own understanding of their reality,

making it possible to change this reality. In addition, participants will be more likely to

appropriate practices they themselves developed.

In order to theorize on peasants' knowledge, fire use was codified into three

different phases: (a) before a planned burning event, (b) the burning day, and (c) after a

planned burning started. The exercise resulted in the creation of a fourth category, (d)

control of accidental fires. This codification constituted of discussions on each phase by

a brainstorming exercise, and ideas organized in flipcharts, facilitated by the Fire Action

coordinator. The problem-posing question was "What can we do before, during, and









after the use of fire in order to prevent uncontrolled fires?" The following

complementary question was "What can we do in order to control fire?" Every idea was

considered, even when some participants did not agree on its effectiveness. This first

exercise resulted in a set of general activities (listed below), with more or less agreement

on their efficiency or feasibility.

The division or codification of fire management actions into phases was key for

planning during meetings because it allowed the connection of the several fire prevention

and control techniques practices to the best period to use them, according to peasants'

usual slash-and-burn system steps (Figure 2-2). The upper part of Figure 2 shows the

approximate division in months of rainy and dry seasons; the middle part shows six basic

slash-and-burn steps influenced by fire, each followed by a bar indicating the period

when the step is taken; the lower part indicates the three Fire Action phases, before the

burning, on the burning day, and after the burning. Following this time division during

discussions, peasants were challenged to think in terms of responsible fire use during

each of the slash-and-burn steps based on their own previous experience or information.

Thus, this step of relating phases to local system was a major innovation that emerged

from the participatory process.

This emphasis on collective action was a second major innovation. In all four

phases, the overall recommendation for 'union' was consensual, meaning that the work

should be done in groups of family members, friends, and neighbors. It was said that fire

was "democratic," i.e., it would burn everybody's land without distinction of origin, kin,

religion, etc. Acting together, families would be better able to manage fire efficiently.

For instance, when choosing the burn location (sometimes during the rainy season) it is









important to analyze the fire danger to surrounding ecosystems, also taking into account

neighbors' ecosystems. Neighbors should unite to organize a period to slash the

vegetation, allowing fields to dry and be burned together at the same time. Burning in

groups would help in the use of prevention and control techniques, such as firebreak and

backfire.


Dry- Season
Rainy Season Dr Sao
DEC JUN ...'1 DEC



Choosing the burning place

Slashing the original vegetation
(2 to 4 weeks)
Cv,
E Waiting for egetation to dry
(2 to 4 \ eeks)
Burning day

Re-burning (if necessary)

Planting i

L BEFORE BURNING A AFTER
BURNING
DAY
CONTROLLING
ACCID. FIRES


Figure 2-2. Codification of fire management actions into phases (before the burning day,
at the burning day, and after burning) and its connection to the best period to
use them, according to peasants' usual slash-and-burn system steps (Drawing
by author).









The first systematization on Community Fire Management, developed in the first

regional meeting, June 1998 (Carvalheiro 1999:5-6), is shown in Figure 3-3. After this

brainstorming exercise, the coordinator presented an experience of fire management also

developed by colonist peasants in Paragominas, through a participatory approach from

1996 to 1998 (Mattos et al. 2002). Participants concluded that their recommendations

were very similar, differing only in the details about the size and type of firebreaks, and

in the form of community organization for the case of accidental or criminal fire. The

exercise of facilitating peasants to organize their knowledge empowered them to believe

in their own capacities. The similarity between strategies from the other case and their

own knowledge showed them that they already knew a lot, and that what was really

missing was their organization and motivation to implement practices in a more

consistent way. Another sign of their self-confidence was that they did not request copies

of the Paragominas Booklet, but proceeded with their discussion. Their decision on

developing their own knowledge rather than basing on other experience was a

demonstration of empowerment. Peasants attending this first meeting decided not to

define the details about techniques or community rules at that moment, but instead to

focus on more general recommendations. They claimed that details on each family's

agreement should be developed at the community level.

It was clear to the peasants attending the workshop that the recommendations

defined during the meeting bore no official power to replace the IBAMA law, currently

in effect. Still, peasants were motivated to develop their own rules in order to confront

IBAMA's top-down rules, because of their belief that if they could manage fire well

(causing no losses to the forest or to their neighbors), they would have the legitimacy to









face any eventual confrontation with IBAMA's representatives. At the end of this

meeting, some agreements were reached:

* Each participant should present and promote this discussion on community fire
management in their own communities;

* Each participant should support the creation of Fire Groups open to all members of
their communities;

* The FATA/LASAT consultant should visit all the communities where Fire Groups
were created, giving priority to those communities that could inform possible dates
of burning events;

* FATA/LASAT should prepare instructional materials in the form of posters and/or
flipcharts, which would include the results of the workshop; the material would be
distributed and discussed during the regular technical visits conducted by the
FATA;

* FATA/LASAT would support the coordinator's work for the 1998 dry season with
funds provided by the DFID.

The results of these agreements were positive. Jadiel Souza (personal information),

FATA director, said that more than 30 colonist communities with families involved with

the FATA and LASAT projects organized Fire Groups. The Fire Action produced a

poster and a serial flipchart summarizing Fire Management recommendations. The Fire

Action coordinator visited nine communities, facilitated community and Fire Group

meetings in five of them, and made subsequent visits to two of those five communities.

Burning situations (before, during and after) were also monitored.









Before a planned burning: Invite neighbors to be present and, After a planned burn starts:
* Talk to neighbors about the location if possible, to help with the burning. Continue observing the burning for
of agricultural fields and pasture as long as flames are high;
management; also talk about when
management; also talk about when Return to the site at the end of the
to slash the vegetation (in order to day and on the following day in
day and on the following day in
be able to bum together); On the planned burning day: order to observe any escaped fire;
order to observe any escaped fire;
* Talk to community members about The person responsible for the Extinguish and/or destroy any
Extinguish and/or destroy any
the best time to burn (in order to burning needs to be present; dangerous tree stump or trunk with
,,dangerous tree stump or trunk with
avoid burnings in the beginning of
o b o Do not bur a field alone, but in the potential to fling sparks at
dry seasons);
dry seasons); groups of at least 4 people; adjacent ecosystems.
* Use firebreaks; 1
SUse firebreaks; Use 'fire-against-the-wind' or
* Fell dead trees on the borders of contra-fogo (start burning on the
agricultural fields and forests or side opposite to the main wind);
other ecosystems; this technique is If possible, organize water Controlling fires:
If possible, organize water
more important at the border where .
Si a containers and tools for fire control; Make a control firebreak (varrida);
fire is usually stronger;
SWhen felg tr Insist on neighbor participation. Backfire to control the main fire;
* When felling trees, try to direct
them to the center of the Smother the fire with branches and
agricultural field in order to reduce soil;
the amount of fuel near the field
borders; Extinguish it with water.
borders;
* Inform neighbors in advance of the
day when burning is planned, and
confirm it one day before;

Figure 2-3. Summary of the first systematization on Community Fire Management, developed in the first regional meeting, June 1998
(Carvalheiro 1999:5-6).









Despite participants' initial frustration at not being able to communicate with

government representatives, the meetings held at the community level led to rich

discussions and to community planning. The results of the first regional workshop were

presented, and their local burning situations were discussed. Each peasant's burning

situation was presented to the community, which generated discussions on possible

dangers and how to minimize them. In order to codify the group's visualization of the

different burning situations, the FATA/LASAT technician developed an instructional kit

called "Burning Dynamics" (Carvalheiro and Mattos, in press). This kit contained

colorful drawings on cardboard tags representing the ecosystems (primary forest,

secondary forest, perennial crops, agricultural crops, pasture) and other significant

elements in a burning event, such as wind, sun, fire, rivers, firebreak, water buckets,

houses, fences, equipment, and people (Figure 2-4). There were several copies of each

drawing, and they could be positioned on a larger card according to different situations.

The peasants would build their situation and keep moving the elements around (fire,

people, water) to represent the burning dynamics, and incorporate suggestions. The kit

was very helpful in promoting discussions, especially helping group understanding of a

burning situation. Usually, when a peasant explains his burning situation, he or she

draws in the ground with a stick, showing the agriculture field, surrounding ecosystems,

wind direction, where did they started burning, etc. After a while, others in the group

cannot remember what the ecosystems were around the burning, or the wind direction, or

neighbors' land; the discussion does not go deeper because the person responsible for the

burning has to keep explaining the burning situation again. With the burning-dynamic

kit, all participants could visualize the situation, and make suggestions, adding or moving









objects, facilitating discussions of possible alternatives for the studied situation. Those

meetings held at the community level attracted many people, including those not used to

attending them.

As planned before, the second regional meeting called "Management Alternatives

for Small Scale Production" was held in December of 1998, at the end of the dry season,

during three days. It was composed of three parts, as follows:

* Community Fire Management (December 1 and December 2).
* Community Forest Management (December 3).
* Assembly open to invited institutions (December 4).



--" -- -, ---- -



,. ', W i.-,.

flmJ\ ?Perennial Crops

Secondary Forest T




Primary R..
M .,,.

Forest l'Zo g

Some Others Elements Slashed Vegetation

Figure 2-4. Some of the elements used in the "Burning Dynamics," which could be
rearranged according to each burning situation, helping the group to visualize
the burning situation and assisting in the consequent in-depth discussions
(Drawings by author).

As part of the action-learning cycle methodology, the goals of the workshop were

to (a) evaluate the fire Action for the 1998 dry season, (b) organize its results in a fire

management proposal for the region, and (c) present and discuss this proposal with









invited institutions (IBAMA was the most hoped to attend). A total of 47 peasants

representing nine communities attended the meeting (28 men, 14 women and 5 children).

A stronger presence of peasants was expected, but the workshop coincided with the

beginning of the rainy season, which forced families with related agricultural activities to

stay in their lands. Two FATA Directors, as well as FATA technicians, the Director of a

Honey Production Association, the Fire Action coordinator, and a researcher from the

University of Washington were also present during the entire workshop.

The first half-day was used for theorizing the interconnections of local fire use with

regional, national and international contexts. Participants discussed sustainability in

Amazonia and the growing susceptibility of the forest to fires, as well as data from INPE

on fire statistics, the insertion of Maraba in the Arc of Deforestation, international

interests in stopping fire use in Amazonia, the fire disaster that occurred in the state of

Roraima, the history of governmental fire laws and their current changes.

The most recent IBAMA booklet on fire management (IBAMA/GTA, 1998) was

analyzed. Table 1 shows a summary of this analysis. Each participant received a copy of

the booklet, which was read by the group, and time was allowed for discussion. As a

result of this study, participants first criticized the symbol chosen by IBAMA to represent

forest fire prevention and control: a humanized anteater. It was a consensus that it was

not a good symbol for peasants because the anteater is an appreciated dish in Amazonia,

and not at all related to fire management awareness, even to those who know that forests

are their habitat. It was agreed that the representation of an experienced technician

presenting practices in accordance with their reality would be more likely to be accepted









by them. During Fire Action evaluation, peasants referred to elder peasants as models for

good management, as discussed in Chapter 4.

Participants also concluded that official fire permits were too hard to obtain,

especially for peasants. Among other bureaucratic procedures, it required numerous

documents peasants simply did not have, and which were often incompletely listed in the

booklet. Although some of the participants knew that the acquisition of some of those

documents cost a lot in fees they usually could not afford, the booklet did not explain any

of that. Participants expressed their concern about what this money was used for.





Table 2-1. Peasant critical assessment of the IBAMA/GTA booklet on fire management
(IBAMA/GTA 1998), as discussed during the second workshop in December
1998 (Carvalheiro 1999:21).
Summary of the critical assessment of the IBAMA Booklet on Fire Management
WEAK POINTS STRONG POINTS
Burning permission requires land tenure Fosters community organization
documentation
Burning permission requires payment of Instructs participants of different fire
fee for inspection by deforestation agents groups to bum on different days
Burning permission requires payment of Fosters use of adequate equipment
fee
Firebreak guidelines do not fit the reality of
peasants or of the region-concerning
length, distribution and format
The Brazilian government's regulation of
fire use shown at the end of the booklet is
printed in small font, making it hard to
read, and uses technical jargon


Some strong points were that the booklet encouraged community organization,

regarded by participants as one of the most powerful tools for fire management. Also,

the instruction saying that participants of fire groups should not burn on the same day









was evaluated as positive, because the fire groups have more time to observe the burning,

and to be present if any accidental fire starts. Everyone emphasized the need for

adequate equipment to optimize fire control and minimize the risks involved in the

activity.

Participants held that the IBAMA guidelines for burning after the first rains could

not always be followed, depending on (i) the location of the agricultural field in relation

to its topography (places close to swampy areas do not burn well after rain), (ii) the

previously existing vegetation (burning pasture after rain events helps grass regrowth;

slashed primary forest will not burn well when it receives a strong rain), and (iii) the

severity of the dry season (some dry seasons are so harsh that burning without a rain

event becomes highly dangerous, while in others dry seasons rain is more frequent so that

it is hard to avoid burning before the rain).

The hardest criticism was placed on the length, distribution, and format of the

firebreak guidelines presented in the booklet. IBAMA requires a 3-meter-wide firebreak

in general, and a 6-meter-wide firebreak when the burning is adjacent to forest and

neighboring areas. Three meters is a difficult target for peasants to meet because they

have no access to machinery; six meters is out of the question, since this is generally a

wider lane than most roads accessing their communities. Peasants also questioned the

statement that the wider the firebreak, the smaller the risk of accidental fires

(IBAMA/GTA 1998:14), because sparks can fly very far. Firebreaks located far away

from the agriculture field were evaluated to be much more efficient (as discussed in the

Fire Action recommendations). In addition, the IBAMA booklet's recommendation to

make firebreaks in the middle of large agricultural fields in order to "bur in steps"









sounded like a joke, or rather like a clear sign that the people who wrote the instructions

had never visited a slashed vegetation in Amazonia prepared to burn. Slashed vegetation

(especially from primary forest but also from secondary forests) is composed of many

fallen trunks and branches superimposed one over the others, in many layers; walking

through a slashed field means walking over trunks, and in small areas over the ground.

Constructing firebreaks by hand in this condition is impracticable. All this, associated to

other guidelines such as building trenches inside the agricultural field when the land is

steep, carrying large tree trunks located close to firebreaks to the center of the field, and

the representation of slashed vegetation as pasture field, resulted in participants' total

disbelief in IBAMA's real expertise and consequently in its competence to provide advice

or to monitor fire management in Amazonia.

The last aspect participants criticized in the IBAMA booklet was the font size

chosen to print the text content of the booklet and the language used, particularly where it

presented governmental fire regulations. Besides being too small, which made it hard to

read, the language was too technical and therefore difficult to understand.

On the morning of the second day of the workshop, representatives of each

community presented an evaluation of the 1998 fire season in their community. They

described the severity of the dry season, examples of Fire Group work, and the lack of

losses in most of communities. Only one community recorded loss because of burnings,

when a peasant burned a field without asking for the community Fire Group's support.

Members of a community in which accidental fires were successfully prevented

complained about the difficulty and danger involved in fire prevention and control, for it

is done with no outside support or adequate equipment. This discussion was illustrated









using the example of a burning in which the Fire Group helped. After most of the fire

had been extinguished, the group left. Later that same day, when the responsible person

and his 12-year-old son returned to observe, they found that the fire had spread to the

neighboring forest. There was no way to call for help, and the two of them spent many

hours controlling the fire, without water or equipment. They had trouble breathing for

many days after the event.

All participants then agreed on the need for equipment and expressed their concern

about not being able to follow IBAMA's laws. This concern is shown in the following

speech:

When people work together for a common goal, nothing is difficult. And our
community is like that. We hope that from this point on, things [fire use] go right,
and that IBAMA does its part too. (...) If the government is pressing us to control
fire, then it [the government] has to provide us with the appropriate equipment.
Nobody can control it [fire] like it is done now, using only our hands.

Antonio Pintassilgo, Tracua Community, December 3rd 1998,second regional
workshop (Carvalheiro 1999:26)

Besides limitations with controlling fires because of absence of external support

and equipment, the peasants' evaluation at the workshop closing the first action-learning

cycle for the 1998 dry season showed positive reactions to the Fire Action campaign.

Without discontinuing their use of fire, the families involved claimed to be better

organized to prevent and control fire in their communities, and therefore reduce losses.

However, they worried about uncontrolled fires that originated in large farms, and the

government's outlawing of the use of fire and punishing those who use it. Those

uncertainties about government actions regarding fire use are well expressed in the

speech by representative Augusto Ferreira, when evaluating the work in his community:

Many people that are really interested do not want to obstruct the law; no one wants
to go against it. However, we have to know how far we can go without breaking the









law, because if it [the law] does not conform to our reality, how do you follow the
recommendations in the [IBAMA] booklet, for example, that do not match our
reality?...

Mr. Augusto Ferreira, Nova Jerusalem community, December 2nd, 1998, regional
meeting (Carvalheiro 1999:23).

On the afternoon of the same day, the participants were divided into groups

according to their communities. Each group developed an exercise of codification using

the Burning-Dynamic kit. They were asked to represent a common burning situation in

their community, and to discuss in each group, fire practices for prevention and control of

accidental fires. Burning-Dynamic material was provided. After that, each group

presented the burning situation using cardboard visual aids, and described the practices

taking place before, during and after the burning. This discussion on fire management

practices was very rich, with different situations of vegetation to burn, surrounding

ecosystems, practices, etc. Two of the seven examples discussed in this exercise are

presented in Figure 2-5.









c71

74411

(5a) (5b)

Figure 2-5. Two examples of Burning Dynamics represented by community
representatives at the second Regional Seminar, held in Maraba city in
December 1998. Each example represented a possible burning situation, and
it was used to present and discuss this situation with seminar participants.
According to Freirian Pedagogy, the codification is a visual representation of









the problem-posing situation, promoting critical reflection on peasants' reality,
and their critical self-insertion into this reality (Photo by author).

Figure 2-5a shows the fire management example presented by the Nova Jerusalem

and Plano Dourado communities (Carvalheiro 1999:27-28). They displayed a

hypothetical situation of a burning where the crop field to be burned was surrounded by

pasture (on the right side), neighboring forest (on the left side), forest located in the

family's land on top, and secondary forest (at the bottom). The wind blows from right to

left, and the family house and perennial crops are located to the right. The family

organizes a group of five men to make the four firebreaks and to burn. On the day of the

burning, they prepare tools (machete and scythe) and water, and start the burning from

the left side, which is downwind. The woman stays close to the house and the boy

remains at the firebreak, close to the fastest way out. That way, he can go and get help if

necessary. The group stays and observes until the strong fire is out, checks the firebreaks

for accidental fires, and observes dry wood or dry standing trees located close to the

surrounding ecosystems. They return on the following days to check for any accidental

fires. In case of accidental fire, they try to smother the flames with green branches and

soil, also using water. If the fire gets out of control, one of them has to go for help.

The other example, Figure 2-5b, shows the situation presented by representatives

from the Tracua and Josin6polis communities (Carvalheiro 1999:28-29). They portrayed

a burning carried out by an individual family with the help of three neighbors. A

secondary forest is located to the right of the burning field, to its left there is a primary

forest, at the bottom there is a pasture with animals, and the perennial crops and the

family house are located at the bottom right-hand corner. The black line around the

burning field represents firebreaks. They also showed a small brook crossing the









ecosystems. Firebreak width and location varied according to the neighboring

vegetation: primary and secondary forest, and pasture. When the burning field was next

to a pasture, the firebreak was located at the border of the burning field, and was 2 to 3

meters wide (or wider if there were fences). When the burning field was next to primary

forest, the firebreak was located 8 meters from the field border, and was 1.5 meters wide.

When the burning field was next to secondary forest, the firebreak was located 5 to 6

meters from the burning field border, and was 1.5 meters wide. The reason for locating

the firebreak within the vegetation and not at the border of the burning field (as

prescribed by IBAMA) was explained by a peasant, when he pointed out their burning

situation:

(...) because when fire goes with the wind, it throws sparks, and when it goes into
flame, nobody can stop it, even with a wider firebreak, because it [fire] jumps,
throws sparks; when it goes inside these 8 meters of forest [between the border of
the burning field and the internal firebreak], the fire runs close to the forest floor,
where the forest hasn't been cut. It is easier to control the fire when the firebreak is
1.5 m wide.

Jose dos Santos, Tracua Community, December 2nd, 1998, Regional Workshop
(Carvalheiro 1999:28)

During the presentation by the representatives of the Tracua and Josin6polis

communities, they explained that this kind of firebreak works together with the contra-

fogo, or preventive firebreak. In this practice, peasants are positioned on two sides of the

burning field. The burning is started by those located downwind, and only then do the

others start the main fire (which goes with the wind). The fire started against the wind

runs slower than the main fire, but fast enough to create a strip of land without fuel. This

results in the main fire going out before it reaches the field border, and consequently

reduces the risk of fire reaching the neighboring vegetation. In addition, they talked

about remaining in the burning field until the main fire was extinguished and returning on









the same day and on following days to check for any possible accidental fires.

Equipment was also mentioned, mainly machetes, scythes, hoes, water pumps, and if

possible a chainsaw. The peasants emphasized that a chainsaw was faster and more

efficient in separating the burning from the green vegetation through varridas, or control

firebreaks (firebreaks which are made after the burning was started and are thinner than

preventive firebreaks).

After the seven examples were presented, the practices and recommendations listed

were classified in five categories: equipment, before the burning day, during the burning

day, after the burning day, and conflict resolution. Each of those topics was discussed by

the participants. The participants also defined some widely used key concepts which at

many times take different meanings, and which are basic for defining responsibilities

among fire users, such as "responsible", "accidental," and "criminal" fires. This

discussion on fire practices and recommendations developed into the first Community

Fire Management Recommendations for the Maraba Region, later organized into a

booklet. The results of the peasants' discussion on concepts are listed below (Carvalheiro

1999:35; Carvalheiro & Aquino 1999:16-17):

* Responsible Fire: when fire is used as an agronomic tool, and its user follows all
rules established in the Community Agreement.

* Accidental Fire: when fire is used as an agronomic tool, and although its user
follows all rules established in the Community Agreement, the fire exceeds the
limits of the planned burning field, causing losses.

* Criminal Fire: when fire is used as an agronomic tool or for any other reason and
its user does not follow the rules established in the Community Agreement.

On the last day of the workshop, the first Community Fire Management

Recommendations for the Maraba Region were read and approved by the peasants taking

part in the workshop. After their approval, the recommendations were presented to the









invited institutions. The representative from IBAMA was anxiously expected by all, but

in spite of a personal invitation two weeks earlier (IBAMA has an office in Maraba), no

one from IBAMA attended the meeting. There were representatives from four STRs

(Rural Workers' Union): Itupiranga, Jacunda, Maraba and So Domingos do Araguaia, as

well as the Director of Maraba SEMMA (Municipal Environmental Secretariat), the

Itupiranga SEMAGRI Director (Municipal Agricultural Secretariat), two technicians

from INCRA, a technician from the Itupiranga EMATER branch, and the Maraba Public

Defender for environmental affairs. Notwithstanding the frustration because of the

absence of representatives from IBAMA, the meeting went on as planned, and the key

agreements achieved were:

* To send the Community Fire Management Proposal, as well as the list of all
attending institutions, to IBAMA, along with a letter explaining that they were
missed and encouraging their attendance in subsequent workshops;

* Each peasant should convey the Community Fire Management discussions to
his/her communities, and promote (1) the creation of a community fire
management proposal, based on the discussions in the workshop (each community
received the approved proposals), and (2) the development of a conflict resolution
proposal;

* Institutions should work to create an Environmental Council, at the municipal level,
with power to make decisions. This must be done through Municipal
representatives, and be approved by the Mayor;

* To disseminate the Fire Community Management proposals in communities by
means of a booklet to be prepared by FATA/LASAT before the following dry
season,

* To organize another Fire Workshop at the beginning of the following dry season
(second action-learning cycle).

Second Action-Learning Cycle

During the 1998-1999 rainy season, the Fire Action coordinator documented the

1998 community fire management experience, raising funds for one more year with









DFID. The coordinator and other LASAT technicians organized the Fire Action proposal

into a booklet (with 750 printed copies),28 which was distributed after being approved at

the July 1999 regional workshop.

The third regional workshop on Community Fire Management, held at the FATA

Center on July of 1999, brought together during three days representatives from 15

communities (14 peasants and 1 indigenous), comprising a total of 36 men, 13 women

and 6 children. It was facilitated by the Fire Action coordinator, a LASAT technician,

and two FATA directors. This time, a representative from IBAMA was present. The

IBAMA state Director for Environmental Education attended the entire workshop, and

the IBAMA representative at the Maraba Office was invited to-and attended only the

last day of the workshop (which was open to invited institutions). Two other IBAMA

employees were also present on the last day. Additionally, representatives from the

Paragominas Community Fire Management experience, a key leader and a key

researcher, were present on all days.

The main goals of this workshop were to (a) review the 1998 experience, (b)

evaluate 1998 agreements and achievements, (c) present the Paragominas experience

represented by key social actors, (d) plan the 1999 dry season, (e) present and evaluate

the FATA/LASAT Fire Action booklet, and (f) disseminate the Fire Action among new

families and institutions.

The workshop was considered to be a positive experience, in which all goals were

partially or totally achieved. The previous regional workshop was evaluated, and

participants concluded that four of the five agreements from the second regional

28 Funds for the booklet came from a field fellowship from the Natureza e Sociedade Program for this
thesis.









workshop had been put into effect. The exception was the creation of the Environmental

Council that was never formed. More families from the communities involved and from

neighboring peasant communities were interested in taking part in the Fire Action, and

the presence of representatives from other communities at the workshop was considered a

positive sign of the Action's dissemination. The presence of the IBAMA director was the

workshop's greatest achievement. For the first time, participating peasants and the

representative from IBAMA were communicating face to face. Although the IBAMA

representative had no decision power, and could only inform the participants of IBAMA's

laws, he was respectful and listened to peasants' complaints and accounts, thus taking a

step towards reducing the distance between IBAMA and peasants. It was a learning

experience for all.

The Fire Action booklet was read in the assembly and after a few changes was

approved by participants. The practices recommended in this booklet are presented in

Appendix C.

The Paragominas leader and researcher and the peasants from the Maraba region

shared experiences, exchanging information, challenges, problems, and achievements.

This peasant-to-peasant information exchange tends to empower them by furnishing a

wider and more profound picture of their own situation, bringing consciousness of their

situation at a regional level. Horizontal communication among Amazonian peasants is

very important because in general they have limited access to any means of

communication and transportation.

In the previous months, the Fire Action coordinator had transferred her position to

another LASAT technician, to focus on her graduate studies evaluating two of the









communities attended by the FATA/LASAT Fire Action (the present thesis). During the

workshop, the participants had the opportunity to meet the new Fire Action coordinator,

and five communities were selected to be monitored in the 1999 dry season. It was

established that one of them should be an indigenous community, so as to involve other

actors besides peasants (the Surui tribe was chosen). Two communities were to be

monitored by the author of the current thesis (one in Maraba and the other in Nova

Ipixuna). The other two communities to be monitored were defined in the assembly,

according to spatial distribution (regions that had not been previously involved).

Several other institutions were invited to attend the last day of the workshop, and

representatives from five STRs, EMATER/Maraba, SEMMA/Maraba, a Fire Department

Lieutenant, two representatives from COOCAT, the EFA Director, a representative from

DFID, and an Honey Association Director were present. Several important NGOs (CPT,

MST, FETAGRI and CNS) were not able to attend because they were assisting 130

families in a land conflict with police and gunmen.

After all participants introduced themselves, they received the Fire Action booklet,

and it was read out loud by a peasant. The recommendations in the booklet generated

several discussions. For instance, the firefighters' main comments were that the

firebreaks were too thin and that the backpack sprayer proposed was not efficient to

control fire. The firefighter Lieutenant suggested that peasants should use more effective

equipment, such as asbestos fire swatters (abafador de amianto) and a large water pump

(with a 20 m range), and should always call the firefighters. The peasants replied by

explaining the reasons for the thin firebreaks, the lack of other kinds of equipment, the

usually long distance from agricultural fields and pastures to water sources, and the









absence of public telephones for several kilometers. This face-to-face discussion among

representatives from different institutions and peasants was a great learning experience.

In this process, different social actors had the chance to present their realities and contrast

them with government laws and peasant recommendations, making it clear to participants

that many of the fire laws were actually inapplicable. Based on this discussion, the Fire

Department Lieutenant proposed that a multi-institutional caravan be created, which

should be composed by FATA, the Fire Department, IBAMA, the Municipal

Environmental Secretariat, the army and the police. The caravan should not only inform

about fire use but also punish non-conforming farmers. The FATA director agreed with

the idea of a multi-institutional caravan that would educate peasants about fire use, but

without inflicting punishment. He defended his position by pointing out that changing

fire practices is a gradual process, and if the caravan had an educative role, it would be

seen by peasants as an ally; however, if the caravan punished peasants, it would be seen

by them as an enemy.

As a result of the discussions in the workshop, participants agreed that (a) they

would work towards the creation of a multi institutional caravan which should play an

important role in raising peasant consciousness regarding responsible fire use, but

without punishments, at least at this point; (b) the Fire Action booklet should be widely

distributed in the Maraba region, and its proposals should be acknowledged as official

recommendations for peasants; (c) participants would raise funds to support the Fire

Groups Forum, caravan expenses, and the Fire Groups; and (d) another regional

workshop was scheduled for the late 1999 dry season, at the FATA Center, with the

purpose of evaluating and planning the following dry season.









During the 1999 dry season, the new FATA/LASAT coordinator visited all five

communities. The caravan was not created, and the workshop planned for the end of the

1999 dry season was not held until the following rainy season (in the beginning of 2000),

with low attendance (many dirt roads are impassable during rainy season). In early 2000,

the financial support by DFID ended, FATA faced structural changes, and thus the Fire

Action was discontinued.

Nevertheless, two peasant communities continued to be monitored by the author of

this thesis, who also took part in burning events in 2000. The study explores the impact

of this approach on community organization, fire losses, and changes in actual practices

implemented among the Fire Action recommendations during two successive dry

seasons. During this period, the results show a clear positive impact in both communities

of changes in fire management practices. The combination of fear of IBAMA sanctions,

and the possibility of developing community-based alternative programs appropriate to

their situation, worked well to reduce losses and change the timing and organization of

burning events. This led to burners taking responsibility for their burns and for following

community rules and practices, with support from other people in the community. These

positive changes, and the practical experience of beginning to implement and evaluate

them in different situations, provide an excellent basis for future fire management in the

communities. The Fire Action depended upon outside technical assistance to facilitate

the process of development and dissemination of the rules and practices, and meetings to

discuss them at the regional level, with authorities from different agencies. How the

absence of this outside technical assistance will affect future fire management practices,









and the relationship with IBAMA, is an important question for future research on the

effectiveness of the Freirian approach used in the Fire Action.

Discussion

This Chapter presented the process for developing FATA/LASAT Fire Action

recommendations. Based on Paulo Freirian Pedagogy, the Fire Action approach was

participatory by assuming a dialogic approach in which a technician was facilitator of the

problem-posing situation: of fire management, codifying colonist actual knowledge of

their reality through 'burning dynamic' exercises, and challenging them to critically

analyze their own situation. Other innovations were (a) relating colonist productive steps

as a function of the burning day (before, on the burning day, and after) and practices for

fire prevention and control; (b) working with peasant empirical knowledge; (c) leading

peasants to consciousness of their own situation (codification) before presenting external

solutions; (d) emphasis on collective work as the basis for the action; (e) promoting

peasants to develop their own agreements at community level; and (f) technical support at

the community level.

The recommendations proposed by the Fire Action, developed by means of the

presented participatory process, included regional and community meetings, and

innovative practices applied by some community members, as discussed above. Such

recommendations were organized into a booklet (Carvalheiro and Aquino 1999), with the

purpose of promoting community fire management by sharing the recommendations

discussed and approved in the workshops. Copies of the Fire Action booklet were

distributed to at least 44 communities related to FATA and LASAT, during the 1999 dry

season. The impact of the Fire Action was monitored in two of these communities, one

of which was involved since 1998, and the other was involved since 1999.














CHAPTER 3
COLONIST COMMUNITIES

Introduction

This Chapter presents a summary of the social-productive characteristics of 72

families in the two communities studied, Sdo Francisco do Itacaiunas and Cupu, both

located in the Maraba region, emphasizing aspects related to fire management and social

organization. Those two communities were studied in order to evaluate impacts of the

Fire Action on their fire practices, and results of this evaluation are presented in Chapter

4. Knowing some of the families' characteristics is important to understand their reasons

behind practices, and their decision-making, and also furnish a frame for disseminating

this experience in other situations. As this Chapter will show, the history of these

colonist communities, and their associated views on production, education, rural life,

forest and fires constitute the important cultural context for developing fire management

programs.

Age of Heads of Families

The average age of heads of families was 40 yrs (SD=12.2) for women, and 47 yrs

(13.0) for men. There were, on average, 5.3 (SD=2.84) persons/house. Besides those

living in the communities, some families maintain sons, daughters, or other family

members living in cities, usually to continue their studies or because of health problems.

In the communities, twenty-two families (31%) of the forty interviewed families

maintained on average four persons (SD=3.2). This maintenance was done with products









(crops and animals), and money. Sending kids outside to study represents high economic

and social costs for families, as discussed below in the education section.

Migration and Length of Residence in the Community

Data on families' origins showed that the majority of heads of families migrated

from other Brazilian states, which characterizes the families as colonists. Most of the

heads of families living in Sao Francisco do Itacaiunas and Cupu communities (46%

women and 57% of men) were born in the neighboring state of Maranhao, located in the

northeast of Brazil (half of the Maranhao state is located in the Legal Amazonia region).

Only 8% of the female and 3% of the male heads of families interviewed were born in the

state of Para. The other female heads of families were born in Goias (18%), Bahia

(18%), Piaui (3%), Espirito Santo (3%), Minas Gerais (3%), and Ceara (1.5%).

Remaining male heads of families were born in Bahia (10%), Goias (7%), Espirito Santo

(7%), Ceara (6%), Minas Gerais (6%), Piaui (3%), and Pemambuco (1%). Figure 3-1

presents origin distributions, in percentages, for both male and female heads of families

of the communities. The Figure considers the states of Tocantins and Goias as one

because of their having only recently been politically divided29

Each head of family was asked to trace his/her trajectory, listing places, villages or

municipalities where he/she had lived. In order to differentiate places just visited from

places where the migrant spent a "significant" time period, this study only considered

places where the interviewee performed some productive activity, such as agriculture,

extractivism, and/or had a job, either in rural or urban areas. In both communities,


29 Goias was politically divided in two states in 1990, with its north region becoming Tocantins state.









women had migrated an average of 4 times (SD=2.1) and men migrated an average of 5

(SD=3.0) times.





8 ::* .......... .. 7
mRORAINA *tB D OE





PAWRfBA







"" .........lg / lSC /_,....--,
Peru 6 GROSSO ) I 9 10%
Bolivia/



DO S I_ "-

Study Sit Paraguay / IO
t Igal taaxonia Border | PARANA 'd "



---' l .o.. .. i
fl. CITARINA
GRANDE
e Ra oDO SUL0




Figure 3-1. Origin distribution by state, by gender, for the heads of families of the Sao
Francisco do Itacaiunas and Cupu communities (n=72). The majority of men
and nearly half of the women were born in Northeastern state of Maranhao,
showing that few were born in Para state (Tocantins and Goias were
considered as one because of its recent political division) (By author).

The data regarding length of families' residence and land ownership in the

communities studied indicated a direct relation between length of residence and first land

ownership. In 1999, the interviewed families' length of residence in both communities









averaged over 10 years, and most families were still living on their first plot of land. The

average length of residence was 11 years (SD=4.2). In Sdo Francisco do Itacaiunas,

77.5% of the interviewed families had been landless before living in the community.

Eleven families (27.5%) were living there since the creation of the community, and

twenty-nine families (72.5%) were living there for ten or more years. Of those living in

the community for more than 10 years, 82.8% had been landless before arriving in the

community, while 57.2% out of those living there for less than five years had been

landless. The same pattern was found in the Cupu community. Eleven of the thirty-one

families (35.5%) were living in the community since its creation, and of those living there

for ten or more years, 77.0% had been landless before arriving in the community. Of the

families living in the community for less than five years, the number of those that had

been landless before living in the community was only 66.7%.

For many colonist peasants, the right to cultivate land is a result of their struggle

for self-esteem, for better conditions for their family. Many interviewees emphasized the

harsh conditions of living as sharecroppers, or low paid employees in rural and urban

areas. These difficulties associated with the unaffordability of land led people to occupy

lands as squatters. Occupying land in the Maraba region was and still is very risky due to

conflicts, high rates of malaria, and lack of any basic infrastructure or outside support.

The importance of owning and managing their own land is a key incentive for fire

management practices described in the next chapter.

Family Labor System

Family labor is the main productive force in the Maraba peasant communities, and

their means to control fire. In general, they do not have access to machinery (to plow, to

pick, to harvest, etc.), and a minority (28% n=69) own chainsaws. Therefore, the amount









of labor has a direct influence on the family's ability to produce. Besides family labor,

the two other main ways to improve their labor force are family social organization for

collective work (workday exchanges and mutirdo, in which everyone works together),

and obtaining the economic means to hire employees. For this characterization, this

study considered only labor above 15 years old.

The amount of labor force, divided by gender, was similar in both communities.

The average number of female and male workers per household for both communities

was 1.5 (SD=1.01) and 1.9 (SD=1.32), respectively. The labor force involved in

workday exchanges does not involve money. It is a verbal agreement between two or

more producers to work together in each other's land for a similar number of days. The

mutirdo, on the other hand, is the collective work of a group of people on a specific

project, like building a house, harvesting rice, burning, etc. In general, when the work

takes several hours, the person or persons directly benefiting from the mutirdo supplies

most of the food for the meal. Hired workforce is that which implies payment of daily

wages or salary. For the purpose of this study, it was divided into three frequency

parameters: always hires, hires during certain times) of the year, and never hires.

Table 3-1 shows the results for the three types of work, showing that in Sao Francisco do

Itacaiunas there is a tendency for higher participation in workday exchange and collective

work.

Table 3-1 shows that around two thirds of families hire labor, especially during part

of the year. The most common jobs were during dry seasons, for cutting weeds in pasture

areas, slashing forest, and making firebreaks along fences. Data show that workday

exchanges and collective work are used more in the Sao Francisco do Itacaiunas




Full Text

PAGE 1

COMMUNITY FIRE MANAGEMENT IN THE MARAB REGION, BRAZILIAN AMAZONIA By KATIA CARVALHEIRO A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Katia Carvalheiro

PAGE 3

To Gustavo for our love

PAGE 4

ACKNOWLEDGMENTS This thesis was a long journey, which could only have been completed with the help and support of the peasants of the Marab region, my family, friends, and mentors. I would like to express my gratitude to the peasants of the So Francisco do Itacainas and Cupu communities, that received me in their houses and communities, and patiently spent their time in the Fire Action and teaching me about their culture. In particular, I am grateful to the peasants Mrs. Bia and Mr. Placidino that, in1992 and 1993, received me for months in their community, and deeply influenced my way of understanding peasantry in Amazonia. I sincerely thank Dr. Marianne Schmink for all her constructive criticism, patience, and support which helped to strengthen this study. For their precious contributions, suggestions, and guidance, I am also very thankful to the members of my committee: Dr. Marianne Schmink, Dr. Anthony Oliver-Smith, and Dr. Karen Kainer. I especially wish to thank Dr. Charles Wood for comments on my proposal for fellowships. I gratefully acknowledge the support of the InterAmerican Foundation (IAF), the Natureza e Sociedade Program, and the Tropical Conservation and Development Program (TCD) at the University of Florida for the fellowship that allowed my enrollment in the graduate school. Moreover, I would like to thank the World Wildlife Fund (WWF-BRAZIL) and the Tropical Conservation and Development Program (TCD) for funding my research in Par state. iv

PAGE 5

The support I received from Marab's institutions FATA and LASAT was essential for developing this study. I am especially grateful to Manu Wamberguer for his inspiring life trajectory. For their help at various points along the way, I extend my appreciation to the Latin American Center staff Margarita Gaudia, Myrna Sulsona, Wanda Carter and Lydia Gonzalez; the TCD staff, particularly Hannah Covert and Peter Polshek; the UF International Center staff Maud Fraser; and the ETD staff especially John Fishman. I also wish to express thanks to my friends Joelma, Paul and William Losh, Diana Alvira, Juli Haradish, Carla Miller, Edviges Ioris, Olendina Cavalcante, Noemi Porro, Neila Soares. In Brazil, I especially thank Marli Mattos, Paola, Guilhermina Cayres, and Westphalen Lobato. I express thanks as well to Denise and Cyro Negreiros, for the majority of English corrections, Paul Losh for reviewing the introduction, John Dain for translating the Fire Poem, and Nicolas Stahelin for reviewing this acknowledge. My heartfelt gratitude goes to my husband Gustavo Negreiros, for all his support since the very beginning of this study. His efforts in supporting the development and conclusion of my study made the admiration that I feel for him grow even more. I thank my two boys Francisco and Jos for supporting me in studying and working far away, and for always demanding my return. I am also thankful to my friend and second mother, Marina Oliveira, for taking care of my boys as her grandsons. Finally, I thank my parents for inspiring me and always supporting my decisions, even when they were not very sure what I was doing or where I was. v

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................ix LIST OF FIGURES.............................................................................................................x ABSTRACT......................................................................................................................xii CHAPTER 1 INTRODUCTION.........................................................................................................1 Current Government Strategy.......................................................................................2 Problem Statement........................................................................................................9 Research Questions.....................................................................................................13 Study Site....................................................................................................................14 So Francisco do Itacainas Community............................................................16 Cupu Community................................................................................................20 Research Methods.......................................................................................................23 Selection of the Communities and Families........................................................26 Data Collection....................................................................................................28 Conceptual Discussion on Participatory Approach....................................................31 Organization of the Study...........................................................................................36 2 COMMUNITY FIRE ACTION...................................................................................39 Introduction.................................................................................................................39 Peasantry Historical Construction in the Marab Region...........................................40 Peasant Organizations.........................................................................................45 FATA and LASAT..............................................................................................48 Communities' Social Organization......................................................................49 FATA/LASAT Community Fire Action.............................................................51 Process for defining Practices for Fire Management..........................................53 First Action-Learning Cycle.........................................................................58 Second Action-Learning Cycle....................................................................76 Discussion...................................................................................................................82 vi

PAGE 7

3 COLONIST COMMUNITIES.....................................................................................83 Introduction.................................................................................................................83 Age of Heads of Families...........................................................................................83 Migration and Length of Residence in the Community.............................................84 Family Labor System..................................................................................................86 Productive System......................................................................................................88 Formal Education........................................................................................................92 Cultural Interpretations...............................................................................................96 Communities and Towns.....................................................................................98 Forest.................................................................................................................104 Fire.....................................................................................................................107 Discussion.................................................................................................................113 4 COMMUNITY FIRE MANAGEMENT EVALUATION........................................115 Introduction...............................................................................................................115 Losses Caused by Fire..............................................................................................117 Community Organization for Burnings....................................................................120 So Francisco do Itacainas Community..........................................................122 Cupu Community..............................................................................................129 Before Burning.........................................................................................................135 Responding to Official Fire Use Regulations....................................................136 Communication between Neighbors.................................................................138 Period Chosen to Burn: Month Choices............................................................142 Burning after the First Rain...............................................................................147 Preventive Firebreaks........................................................................................154 On the burning day...................................................................................................160 Number of People at the Burning Event............................................................161 Time to Start Burnings......................................................................................163 Use of Backfire..................................................................................................167 After Burning............................................................................................................170 Summary of Practices Evaluated..............................................................................172 Fire Strategies...........................................................................................................175 Peasants' Evaluation of Government and Community Fire programs......................181 Discussion.................................................................................................................184 5 CONCLUSIONS........................................................................................................188 Government Approach..............................................................................................188 Colonist Empirical Knowledge.................................................................................189 Participatory Approach.............................................................................................191 General Conclusions.................................................................................................191 Contributions of this Study.......................................................................................192 vii

PAGE 8

APPENDIX A GLOSSARY OF ACRONYMS.................................................................................194 B PEASANT'S NARRATIVES....................................................................................196 C FATA/LASAT COMMUNITY FIRE ACTION BOOKLET....................................203 D INTERVIEW GUIDE................................................................................................214 E CODED BOOK.........................................................................................................215 LIST OF REFERENCES.................................................................................................218 BIOGRAPHICAL SKETCH...........................................................................................227 viii

PAGE 9

LIST OF TABLES Table page 2-1. Peasant critical assessment of the IBAMA/GTA booklet on fire management.....68 3-1. Most common kinds of workforce found in the So Francisco do Itacainas (SFI) and Cupu communities.................................................................................88 4-1. Distribution of kinds of losses caused by uncontrolled fires by year in A) So Francisco do Itacainas community and B) Cupu community............................119 4-2. Main practices evaluated, and possible changes due in part to the FATA/LASAT Fire Action..................................................................................174 ix

PAGE 10

LIST OF FIGURES Figure page 1. South America and Brazilian Amazonia region maps showing the study area in the so-called 'arc of deforestation', the Marab region, north of the Par state.....17 1-2. So Francisco do Itacainas Community village...................................................20 1-3. Cupu Community village.......................................................................................23 2-1. Representation of the FATA/LASAT Fire Action.................................................55 2-2. Codification of fire management actions into phases............................................61 2-3. Summary of the first systematization on Community Fire Management..............64 2-4. Some of the elements used in the "Burning Dynamics.".......................................66 2-5. Two examples of Burning Dynamics....................................................................72 3-1. Origin distribution by state, by gender, for the heads of families of the So Francisco do Itacainas and Cupu communities (n=72)........................................85 3-2. Representation of land use distribution..................................................................90 3-3. Families' formal education grades attended...........................................................95 4-1. Distribution of families that reported any kind of losses.....................................121 4-2. The FATA/LASAT Fire Action and So Francisco do Itacainas community activities, from 1998 rainy season through 2000 dry season...............................123 4-3. Representation of the So Francisco do Itacainas community's strategy in conducing fire management, from 1998 to 2000.................................................126 4-4. Related FATA/LASAT Fire Action and the Cupu community activities, from 1998 through 2000 dry season.............................................................................132 4-5. Representation of Cupu community organization for fire management, from 1999 to 2000........................................................................................................135 x

PAGE 11

4-6. Representation of a burning situation in the Cupu community, in 1999,in which the lack of communication caused loss to one family...............................142 4-7. Rainfall distribution and months chosen for burning..........................................145 4-8. A slashed primary forest burned after heavy rain, at the end of the dry season, resulting in a bad quality burning for cropping...................................................150 4-9. Firebreak bordering forests..................................................................................158 4-10. Distribution of ratio of people to size of area burned..........................................162 4-11. Distribution of studied families' 1998 and 1999 choices of burning time for their agricultural fields.........................................................................................165 4-12. Conjoint burning situation of two neighbors.......................................................178 4-13. Example of a successiful use of control firebreak...............................................180 xi

PAGE 12

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts COMMUNITY FIRE MANAGEMENT IN THE MARAB REGION, BRAZILIAN AMAZONIA By Katia Carvalheiro August, 2004 Chair: Marianne Schmink Major Department: Latin American Studies The Brazilian government has increased investment in education, technology, law enforcement, and infrastructure to control fire use in Amazonia since 1998, but these campaigns have not decreased fire use. Colonist peasants, the social actors focus of this study, were the main target public of IBAMA fire campaigns in Amazonia. While colonists make use of fire every year, and are directly affected by political decisions concerning fire use, they have no voice influencing those decisions. The main groups that could hear peasant's voices-technicians, government, academics-generally disregard colonist's knowledge. This study discusses government strategy for fire control, documents colonists' knowledge on fire management, and evaluates a participatory approach based on Freirian pedagogy of problem-posing, codification, and learning-cycles. Two colonist communities involved with the FATA/LASAT Fire Action were evaluated in 1998, 1999 and partially in 2000 using a combination of participant observation and systematic xii

PAGE 13

interviews of a purposeful sample of families from each community. The results presented show colonists' unique history and set of values and beliefs, manifested in the communities' forms of organization, communication, and adaptation of fire management recommendations developed in the participatory Fire Action. This study's general conclusions are that the success in fire control in the two studied communities resulted from a combination of families' willingness to avoid fire-related losses, the participatory approach adopted and coordinated by trusted leaders and institutions, and experienced technician, and families' fear of government coercive pressure for fire use. In addition, this study argues that if peasants are not seen as partners with valuable knowledge to contribute with technicians for critical reflection on better fire use, the current top-down government actions may lead to more social conflicts in the region. xiii

PAGE 14

CHAPTER 1 INTRODUCTION Images of tropical rain forest burnings have been spread worldwide during the last decade, associated with forest destruction and mismanagement, raising awareness about fire as a problem for tropical forests. In Brazil, the worst event of uncontrolled fire happened in Roraima state, in 1998, when 33,000 kilometers were burned (BVRoraima 2004). The Roraima case was crucial for new Brazilian government policies on fire monitoring and control in Amazonia. Although international concern about fire use has been relatively recent, fire has long been a part of the history of Amazonia. Charcoal from fires at least 2,000 years ago, and even older events, can be found in soil pit layers all around the Amazonian region (Pyne 1998; Meggers 1994; Negreiros personal communication). What has changed lately, however, is the higher incidence of fire associated with agricultural frontier expansion in Amazonian, particularly in regions with clear dry season1, logging activities2, concentration of farmers whose main productive tool is fire, and in years subject to El Nio-Southern Oscillation3. 1 The sharper the contrast between dry and wet seasons, the more vigorous the fire regime (Pyne 1998:65). 2 Intensive logging increases forest flammability: forest cuttings increase potential fuel loads and the amount of solar radiation reaching the forest floor (Uhl et al. 1994:295). 3 "El Nio" is part of a global configuration of atmospheric and oceanic changes resulting in droughts and inundations around the world. Amazonia has seemed vulnerable to drought during El Nio oscillations (Meggers, 1994). 1

PAGE 15

2 Because of losses caused by fire at local, regional, and global levels, there is a common agreement that fire has to be better managed, and alternatives to its use have to be adopted, a consensus which is reflected in the growing governmental and non-governmental efforts in this direction. This common ground is a powerful step toward integrated solutions. However, there is no agreement among the several different actors on what 'better fire management' is, and how it can be reached. Current Government Strategy Brazilian government strategy, according to one of PREVFOGO4 director, is based on the "tripod of education-enforcement-monitoring, under the slogan: to educate not to burn" (Vargas 2003). Government educative actions consisted of media campaigns, educative materials, and courses, whose main goals were to inform and persuade. The government had assumed an anti-fire position, focusing on convincing farmers about the terrible effects of fire (on soil, forest, animals, water sources, people's health, transportation, carbon emissions, and the international image of Brazil), to show them how to manage and control fire, and to warn of the penalties for those who do not comply with the law. Courses were provided by PROARCO5 to firefighters and brigades (local disseminator agents), and to peasant representatives, showing them how to do Controlled Burnings. 4 PREVFOGO is the National System for the Prevention and Combat of Forest Fires, created in 1989 and ratified in 1998, administrated by the IBAMA (Brazilian Institute for Renewable Natural Resources and the Environment). For a complete list of acronyms see Appendix A. 5 At Amazonia level, the Brazilian government created the PROARCO (Programme for the Prevention and Control of Burning and Forest Fires in the Deforestation Arc), set up also in 1998, after the Roraima case. This program is jointly administered by IBAMA (Brazilian Institute for Renewable Natural Resources and the Environment) and by MMA (Ministry for the Environment, Water Resources and Amazonia).

PAGE 16

3 Important issues like fire control cannot be effectively answered with a top-down approach, however. Such approaches assume a "domino or trickle-down effect," in which information goes in a one-way direction, reaching increasing numbers of people, without dialogue or feedback. Although local people are involved in PREVFOGO actions, the approach adopted has been top-down, since decisions about the actions "are made by a small set of powerful stakeholders, according to their own agendas, knowledge and value systems" (Ingles et al. 1999:6). In an attempt to work with social movements, in 1998-9, IBAMA (Brazilian Institute for Renewable Natural Resources and the Environment) worked as a partner with GTA6 (Amazonia Work Group), with economic support from PPG7 and USAID, forming an emergency program called PROTEGER, which in Portuguese means 'to protect'. In its second phase (PROTEGER II), GTA was a partner with MMA (Ministry for the Environment, Water Resources and Amazonia). PROTEGER's purpose was to reach a large and well-distributed number of Amazonian grassroots organizations in a short period. This program provided fire control and prevention training given by firefighters to community leaders, who would pass the information along to their communities. The strongest aspect of this program was considered to be its intended multiplier effect, disseminating the information to a large number of rural families. However, because of the lack of information about Amazonian realities related to fire use, the technical information given in the training sessions was based on urban/savanna firefighting knowledge and techniques (verified in the IBAMA/GTA 1998 booklet, in which many of the recommendations did not fit Amazonian conditions), and feedback from those indirectly reached by the courses. 6 GTA is an umbrella institution created in 1992 with support from the PPG7 program, and is composed of more than 400 Amazonian NGOs.

PAGE 17

4 Evaluation of the program's effectiveness was based mainly on the number of people attending meetings. Many questions remained unanswered. What practical fire management applications had peasants made in their productive systems? Did they adapt learned techniques? What worked well, and what did not work at all? Did peasants change their fire practices? How? Virtually no information existed to answer these questions. My study thus make a unique contribution to our understanding of the impacts of fire management programs (in this case, using a different approach from the government's) on peasants' use of fire. The newest IBAMA/PROARCO action plan, called "Community Management of Fire" (Gerenciamento Comunitrio do Fogo), started in 2004 in Roraima state (IBAMA 2004c), and seems to be another example of a top-down attempt by the Brazilian government to modify peasants' fire use. In this plan, 960 peasants (distributed in brigades) appropriately trained and equipped by IBAMA will disseminate the right practices for fire use and control (IBAMA op. cited). The italics were added by the author to emphasize the uneven power of who owns the right knowledge (and therefore decides about its content), and decides who is qualified to manage fire and, by exclusion, who is not. Beginning in 1998, IBAMA prohibited fire use in municipalities located in the "arc of deforestation" for at least 2 months: those months were satellites showing high numbers of hot pixels. In 2002, in Mato Grosso state (also located in the "arc of deforestation'') a "mega-action" involving the Brazilian Environment Ministry, State Environmental Secretary, and IBAMA was organized to identify and penalize those using

PAGE 18

5 fire during the prohibition. Given IBAMA's clear rationality on prohibiting fire during the driest month of the year, one of the regional IBAMA coordinators gave an interview in which he questioned farmers' and peasants' rationality: "The environment asks only two months of armistice for not using fire. There are ten more months to burn their fields. Why don't they burn when they are allowed to? We will not give them [burners] an armistice. [IBAMA] action is to repress and combat those burnings." (Benedet 2002). In addition to not understanding farmer and peasant logic on burning their fields during the last half of the dry season (after they prepare the land, and before the rains), this IBAMA coordinator's speech presented war terms in the agency's relationship with farmers, not focused on education but on repression. Changing farmers' behaviors based on this environment of war, repression, or unilateral view of right and wrong cannot be expected to be very effective. The second foot of the IBAMA tripod is monitoring. Indeed, large-scale remote sensing has been shown to be a powerful tool for fire-occurrence monitoring (Setzer and Pereira 1991; Skole and Tucker 1993; Sanderberg et al. 1998; Nepstad et al. 1999; Souza Jr. et al. 2003; Eva and Fritz 2003). More recently, IBAMA inaugurated the CEMAN (Environmental Monitoring Center), in partnership with the SIPAM (Amazonian Protection System), equipped with the most advanced technologies to monitor Amazonian environment conditions (IBAMA 2004a). According to the IBAMA Director, CEMAN will be the "technological eye for Amazon security." Sophisticated technological monitoring is a very important tool for controlling Amazonian fires and deforestation. However, it may not be sufficient. Despite its obvious advantages, Harwell (2000) warns about adoption of remote sensing and its

PAGE 19

6 interpretations by different social actors during the Indonesia 1997-1998 fire disaster. She points out that "in addition to the silencing of local voices, remote approaches to disaster events also obscure these on-going linkages of humans with their environment" (Harwell 2000:334). Cultural understanding of environment is historically constructed, as well as social relations among different social actors, factors that have also to be considered in remote sensing interpretations. If the "technological eye" is in hands of a small set of powerful social actors only (government, international donors), participation remains passive because local people, however included in the politics and policies of conservation, remain peripheral to defining the ways in which conservation is perceived and natural resources managed (Goldman 2003:834). This thesis explores ways to involve local communities directly in monitoring. The third foot and last of the IBAMA tripod is law enforcement. Since 1965, fire use has been ruled by an imprecise Brazilian law, called the Forest Code (Cdigo Florestal). This federal law prohibited the use of fire "on forest areas and on other forms of vegetation; however, it was allowed in specific farming or forestry situations under governmental supervision" (IBAMA/GTA 1998:36). As explained above, in 1989, the Brazilian government started the PREVFOGO program within IBAMA with the objective of reducing the use of fire to "acceptable" levels (IBAMA 1998). This work was coupled with monitoring by INPE (the Brazilian Space Agency). However, Amazonia was not included among the areas to be served by this program at that time. Only after early 1998, when fires in Roraima reached the global media, conveying international pressure on the Brazilian government, were actions taken in Amazonia. The governmental laws and actions concerning fire use then became more specific and

PAGE 20

7 coercive. PREVFOGO was present in the region for the first time in 1998, supporting the creation of another program specifically for the Amazon: PROARCO (Program for Prevention and Control of Fires in the Arc of Deforestation). In 1998, IBAMA also established that community and controlled burnings were required to follow a set of procedures and rules, including obtaining an official burning permit form (authorization was required since 1965, but no practical instrument was created until 1998) and payment of fees per hectare to be deforested and burned. However, the government's recommendations were out of reach for most peasants, who often did not even have the means to prove their ownership of land (one of the documents required to obtain an official burning permit). In 1999 and 2000, the governmental program PROARCO established firefighter groups based in the cities of the region where the fire problem is worst. The state of Par received four new helicopters to be used by IBAMA in enforcement actions. Those actions may include high fines and jail for those who use fire without an official permit or who do not comply with the official authorization. Additionally, the official burning permit can be cancelled in special situations, such as a very high number of burnings in a municipality (based on INPE remote sensing data). When this happens, the government prohibits fire use, even for those who have burning permits. This prohibition lasts in general for two months (the driest months in the year), and it has taken place every dry season7 since 1998. Parallel to the IBAMA program, INCRA (National Institute for Colonization and Agrarian Reform) has organized other firefighter groups associated 7 The Amazonian region possesses two seasons: rainy (or winter) season and dry (summer) season, with its length varying according to micro-regions.

PAGE 21

8 with people from peasant settlements, but based in cities. However, in Amazonia, the dirt roads and distant locations present serious limitations for mobilization. Between the Roraima fires and Law of Environmental Crimes regulation in late 1999, the government's strategy was to create provisional measures to prohibit the use of fire when it is judged to be highly risky, even if official permission has been given. Although prohibitions have happened since 1998, the fees, tickets, and legal actions were based on those provisional measures, easily contested through the legal system. Powerful landowners and farmers had the resources and information to take those defensive actions, but the majority of Amazonian peasants did not. Thus peasants ended up being further punished, which created strong resentment against the government rules. The Law of Environmental Crimes (MMA 2000) defined fees for forest illegally burned: around US$ 515 per hectare8 and 2 to 4 years of prison. Those penalties were widely disseminated through radio, TV, and educative material distributed around Amazonia, always associated with the bad consequences of fire use. The Brazilian government increased investment in education, technology, law enforcement, and infrastructure to control fire use since 1998. The government approach has not proven efficient yet, given that fire occurrence in Amazonia has increased9. In addition, higher fire occurrence has been associated with weekends and with the IBAMA staff strike in 2003 (Werneck 2003), indicating that as long as enforcement is weaker, farmers burn more. The government's top-down approach, supported by the media, is creating a scenario 8 R$ 1.500,00 per ha (R$ 1,00 = US$ 2.91). 9 According to IBAMA 2004b, since the number of hot spots has been systematically monitored in the Amazon region, the higher rate was found in 2002 (160,329 hot spots). In 2003, the number of hot spots was 139,000 (IBAMA 2004b).

PAGE 22

9 unfavorable for partnerships with rural producers, by labeling burners without official permits "transgressors", and depending on paid brigades to control fire in Amazonia. This study argues that governmental actions implemented in a top-down manner, in which peasant knowledge is disregarded, and technical knowledge is "deposited" in their lives, has limitations. By contrast, studies have shown that in community-based fire management there are less occurrences of fire-related losses10 because local people share decision-making power, volunteer to maintain their natural resources (without depending on external inputs), and their empirical knowledge is embodied into practices and policies for fire management (practices culturally and environmentally better adapted). The study sought to explore these ideas through an in-depth study of two communities involved in a community-based fire management program. Problem Statement Colonist peasants, the social actors that are the focus of this study, were the main target public of IBAMA campaigns in Amazonia. While ranchers are recognized as the major social actors responsible for area deforested (and burned) in Brazilian Amazonia (Margulis 2003:80), peasants are also major actors because of their quantity. In Par state alone,11 169.273 landholdings (83% of the total farms) are smaller than 100 ha (Trecanni 2000:430) (an indirect indicator of peasants representation)12, whose main productive tool for land preparation is fire. 10 In China Lichang et al. 2001; in India--Darlong 2001; in Thailand Makarabhirom et al. 2001; Rakyutidharm 2001; in Indonesia Abberger et. al 2001; in Gambia Sonko et al. 2001; for a review, see Jackson and Moore 1998). 11 Brazilian Amazon or so-called Legal Amazonia comprises eight states, representing 61% of the Brazil area. 12 The National Forest Code (Federal Law # 4771) defines "small rural property" or "family land", as any land exploited mainly by its owner with personal or family labor, in which at least 80% of the income

PAGE 23

10 While colonists make use of fire every year, and are directly affected by political decisions concerning fire use, they have no voice influencing those decisions. The main groups that could hear peasant's voices, technicians, government, academics, in general disregard peasant knowledge. This study aims to contribute to this topic by documenting colonists' knowledge on fire management, analyzing a participatory approach in which colonists and grass-roots organizations are partners to develop and implement better practices for fire management, and discussing how in those approaches in which peasants are partners with technicians, sharing decision-power, supported by action-learning cycles, fire will more likely be controlled. Government technicians had shown that they did not believe that colonist peasants possess knowledge for natural resource management, because of peasants' lack of valuable modern "tools", such as formal education, land, infrastructure, entrepreneurial views, or money. Appropriate practices are defined for them but not with them. The Brazilian Federal Agency for Agriculture and Livestock Research (EMBRAPA), for example, explicitly identified the Amazonian peasants' three main problems, that explain their "traditional and backwards technology": lack of education, economic organization, and energy (EMBRAPA 2000:21). The rationale for not learning from colonist peasants, thus, is that they are the ones "lacking". On the other hand, academic ethnographies tend to exclude colonist peasant from the category "traditional" because their culture is too mixed with modern societies. While studies show indigenous tribes hold specific cultural values on fire use (Ruddle 1974; Peters and Neuenschwander 1988; Warner 1991), peasants, by comparison, are comes from farming or extractivism, and with a maximum land size of 150 ha if located in the Legal Amazonia (Presidncia da Repblica 2002).

PAGE 24

11 identified as not "genuine" or "traditional" swiddeners, but "incipient swiddeners" (Warner 1991) or "new swidden practitioners" (Peters and Neuenschwander 1988:74). Because colonists do not practice the traditional form of swidden, they are assumed to not be able to "understand the importance of fallow rotation" (Peters and Neuenschwander1988:74); or to be "there only for the purpose of a crop for a year or two" (Warner 1991:9); or to be totally ignorant of the new land due to their absent cultural ecology of the region, which explains their "search for short term and predatory means of investments" (Lima and Pozzobon 2001:233). One explanation given for this lack of "tradition" is their previous situation of being landless. Brazil is one of the countries with the highest land distribution concentration rate in the world (Trecanni 2001), and many peasants end up becoming sharecroppers. Many migrate looking for their first piece of land. This research found out that the majority of families studied were living in their first landholding. However, many had significant farming and fire management experience in their regions of origin. The general misconception of colonist ignorance has blinded academics and technicians to their empirical knowledge. Few studies go beyond this preconception, and document in detail different uses that colonist peasants have for their natural resources. Muchagata, for example, studying five communities in the Marab region, concluded that colonists "recognize and use a wide range of plant and animal species (respectively 142 and 39 were listed) which provide food, fuel, raw materials, and medicinal plants, and represent sources of income" (Muchagata 1996:76). Muchagata and Brown (2000) also studied colonist perceptions of soil fertility.

PAGE 25

12 Regarding community fire management, Mattos et al. (2000), working with a colonist community in the Paragominas region (also located in the so-called arc of deforestation), presents how a community-conceived participatory approach developed and applied a Community Agreement on fire management, based on their practical knowledge (AMPPDR 1996). The idea of this community work for fire management was absorbed by IBAMA to officially recognize, in September of 1998, the Solidarity Controlled Burning (Queimada Controlada Solidria), in which one single permit would include five or more peasant burnings. Although the collective permit was better adapted to community reality, fire permits continued to require land titling, to charge for deforested and burned areas, to require firebreaks beyond peasants' capabilities, and to cancel permits during the high dry season. Therefore, some ideas from the community-based experience were adopted by decision-makers technicians, but not the participatory approach. Why are educative campaigns ineffective in changing colonist peasants' fire practices? Do colonist peasants have empirical knowledge of fire use? Could participatory approach be more effective in finding solutions for fire management? What do colonist peasants think about fire use law enforcement? Those questions are not only academic, but also are key for those making decisions on fire management in Amazonia today. Recognizing colonist empirical knowledge does not imply that they have all the solutions for fire management. This study argues that if they are not seen as partners with valuable knowledge to contribute with technicians for critical reflection on better fire use (better also for peasants) the actual top-down government actions may lead to more social conflicts in the region. This study also describes a process for a participatory

PAGE 26

13 approach, involving grass-roots organizations, colonist peasants, and in some stages governmental representatives, in which the philosophy and methods can contribute to show directions for collaboration between social actors for fire management. In the Marab region, Southeast Amazonia, the FATA (Tocantins Araguaia Agrarian and Environment Foundation) and LASAT (Social-Agrarian Tocantins Labotatory) Community Fire Action involved more than 20 colonist peasants and 3 indigenous communities. Recommendations for fire management were developed in a participatory process, coordinated by grassroots organizations, and based on colonists' empirical knowledge. The Fire Action produced positive impacts on the two communities studied, as families developed agreements to use fire, and losses caused by uncontrolled fires diminished drastically after the Fire Action. Some of the stakeholders13 had discussed FATA/LASAT's proposal and they had supported it as a regional substitute for the governmental strategies. Although the action is no longer carried out as a specific activity of FATA or LASAT, systematic evaluation of this experience can provide relevant insights to those interested in diminishing losses caused by fire in Amazonia. Research Questions This research contributes to the relevant topic of fire management in Amazonia, specifically involving colonist peasants and changes in their fire practices. This thesis aims to discuss three main questions: Question 1: Why aren't government fire actions and laws more effective in diminishing uncontrolled fires? 13 In this study, the terms "social actors" and "stakeholders" are used as synonymous, referring to any "individual, social group or institution that possesses a stake (or interest) in the management of the natural resource concerned" (Borrini-Feyerabend 1996, cited in Ingles et al. 1999).

PAGE 27

14 Question 2: Do colonist peasants have and use empirical knowledge to manage fire? Question 3: Is a participatory approach to fire management effective in diminishing fire losses? Question 1 was examined by analyzing government discourse on peasants and fire use, families' discourse on government actions (historical and actual experiences), and families' evaluation of the official fire actions (fire law, its enforcement, and educative campaigns). Results are presented in Chapter 2 and 3. Question 2 was developed through peasants' narratives about their practices and perceptions of fire and natural resources management, where had they learned them, changes over time, as well as during observation of burnings. Results are presented in Chapter 3 and 4. Question 3 was analyzed by evaluating the participatory approach applied to a community fire management action (the FATA/LASAT Fire Action), and its impacts on fire practices of colonists families in two communities. The process regarding the fire action is described and analyzed in Chapter 2, and the evaluation of its impacts is presented in Chapter 4. Study Site Par state is the second in the rank of fire occurrence in Amazonia in concentration of hot spots.14 The Marab region, located in an Amazonian frontier region of the so-called "arc of deforestation" is one of the high concentration of hot spots (INPE 2004). The Marab region is also known as Bico do Papagaio15 region, and was originally 14 The expression "hot spot" is used for heat registered on the earth's surface by the AVHRR sensor on board the NOAA satellites, which captures and records any temperature above 47 Celsius, and reads it as "hot spots" (PROARCO 1998). 15 Bico do Papagaio region or "Parrot Beak" region has this nickname due the shape resulted from the Tocantins and Araguaia rivers confluence (see Figure 1-1)

PAGE 28

15 occupied by indigenous groups. It comprises an area stretching across the state-lines of Par, Maranho and Tocantins (Figure 1-1), and includes eight municipalities: Marab, Itupiranga, So Joo do Araguaia, So Domingos do Araguaia, So Geraldo do Araguaia, Nova Ipixuna, Brejo Grande do Araguaia, and Palestina do Par (Forum das Entidades pela Reforma Agrria do Sul e Sudeste do Par 2001:4). By the year 2000, the actions conducted by FATA and LASAT included the first six municipalities on the list and later that year extended to the remaining ones. The importance of peasants in the state of Par is illustrated by the high number of small plots of lands, which account for nearly 82% of the total number of plots, or 169,273 plots (Treccani 2001:430), distributed in 33% of the area (DFID1998:16). In addition, peasants in the state are estimated to produce as much as 84% of all the manioc produced, as well as 63% of the corn, 59% of the beans, and 46% of the rice (Hurtienne 1999, cited by Veiga 2000:15). This region is located in the so-called the "Brazil-nut Polygon", an area named after the abundant Brazil-nut trees (Bertholletia excelsa). The average temperature is 26C, and rainfall around 2000 mm/year (Muchagata 1996:13). It is characterized by defined rainy and dry seasons. The rainy season is concentrated from November to March, and the dry season from May to September. April and October are transition months. The vegetation is classified as Ombrofila Densa Montana Forest (Sestini 2002:44). The two communities studied, So Francisco do Itacainas and Cupu, were formed in much the same way. In most cases, families were landless migrants originally from Brazil's northeastern region occupying part of a large-scale farm. They faced landowners in conflicts over access to the land, and struggled to make their living by means of

PAGE 29

16 extractivism, agriculture, and cattle, for which they had very little government support. They built community roads, churches, schools, and soccer fields; they harvested their own crops and marketed their own products; they founded their associations and partnered with NGOs; they obtained loans (and therefore debts) from the government for their fields. The Catholic Church and grassroots organizations were cited as their main supporters. Complaints about government absence or negligence were present in many different moments of this research, pointing to a potential resistance to future partnership agreements or collaboration with governmental guidelines laws. Some details concerning the history of each community are given in Figure 1-1. So Francisco do Itacainas Community The So Francisco do Itacainas community, in the Municipality of Marab, is located approximately 55 km southwest of Marab and connected to the main town center by a dirt road. To the south of the village runs the Itacainas River. Nowadays, the community is divided into 64 plots, plus a village center. According to long-time resident Mr. Venncio Dias, the community was started on July 14, 1983 when representatives of 41 families settled on the site in a non-official colonization process. Occupants were landless peasants living in local villages. More than 50% of them came from Boa Esperana, a village located 18 km from the occupied land. For safety reasons, peasants occupied temporary houses in groups. They worked together to harvest a first annual crop. Sixteen years after the village was established, the community was officially recognized by the federal government as a Projeto de Assentament (Settlement Project) in early 1998. The community was established as a 'new' land reform area during former President Cardoso's administration. Official recognition represents a very important step towards guaranteeing peasants' land ownership rights, and has long been

PAGE 30

17 waited for. As of the time of this study, the long journey through the bureaucracy had not yet allowed authorities officially to demarcate any border lines or issue any land tenure documentation. Figure 1-1. South America and Brazilian Amazonia region maps showing the study area in the so-called "arc of deforestation," the Marab region, north of the Par state. The map below details community locations, the "Parrot Beak" shape created by Tocantins and Araguaia Rivers, the Tucuru Dam and the Carajs Mine, as well as highways and railroads (Bellow map addapted by author from IBGE 1998).

PAGE 31

18 The history behind So Francisco do Itacainas and how it was built is as typical as that of any community in the Bico do Papagaio region. The area was a traditional castanhal, or a concession of brazil-nut extractive area owned by the government. Although brazil-nut concessions theoretically could not be deforested, a first contract granted 3,500 ha to an oligarchic family which turned it into a 7,500 ha cattle ranching farm or latifndio16. After that, the land was sold twice before 1983, when the peasants occupied it. Some of those interviewed said the farm owner was never a violent person, and for this main reason the conflict with the settlers has never been very violent. Nor did the farmer have land titles to prove land ownership. During the conflict resolution with the landowner, mediated by Catholic Church leaders, representatives of INCRA and the families agreed with the farmer that the 3,500 ha of the original brazil-nut concession would become a community (Nova Cana), and the remaining 4,000 ha would continue as a concession to the farmer. Nevertheless, by the time negotiations were concluded, the number of families had grown and 56 families were excluded from the distribution of land, among which some of those who had fought from the days of the first occupations. The excluded group received the support of "Manu" (Mr. Emmanuel Wanderberg, the Land Pastoral Commission (CPT) coordinator and later one of the FATA founders), resisting and struggling for more land. The families, which did not manage to settle in Nova Cana, founded a new village and named it "So Francisco do Itacainas". They chose Mr. Silva as their community representative, representing the community at the Rural Workers Union (STR) too. The owner eventually gave up the fight for the land. Some months later, the peasants 16 Latifndio is a word of Latin origin and means a large, privately owned landholding (Macedo 1985:viii).

PAGE 32

19 concluded the construction of a school, which was the first landmark of the establishment of the community. Mr. Silva volunteered to teach classes until a teacher was established, with salary paid by the municipality. Mr. Silva was an important leader, and he has been part of the Fire Action since it started. The So Francisco do Itacainas village center (Figure 1-2) occupied an area of 25 ha. This village was composed of 49 plots, four of them serving also as small grocery stores, one with a machine for processing rice, one Catholic Church, one plot for a Protestant Church (with no construction), an Afro-Brazilian Center, one small wood school built by the families just after the occupation, around 1985 (the material was donated by a politician), the soccer field, and a new school built by the federal government in 2001. As shown in Figure 1-2, besides those occupied plots, families planned areas for new plots, roads, and whatever else the community decides (like the area chosen for the new school). From the time of the community's creation, the village was planned in such a way that all the families that possess land had the right to a plot (10 x 30 m) so they could build their houses in the village. In the So Francisco do Itacainas village, there has been electric power since late 1998, as an indirect service from the electric power extended to the Tainpolis train station (part of the Vale do Rio Doce Company railroad for bauxite export) through one of the main roads that cross the village center. The access to electric power brought some new appliances and changes such as TVs, freezers, light, and rice processing machines, among others.

PAGE 33

20 Figure 1-2. So Francisco do Itacainas Community village. Representation of the individual plots for houses, school and churches, roads, rivers, and soccer field. The upper left part of the drawing represents where the village is planned to grow (Drawing by author). Cupu Community Located north of Marab, Cupu is located in the municipality of Jacund, which can be reached through a 48-km dirt road and Highway PA-150. It is located 52 km northwest of Nova Ipixuna, a city 109 km southwest of Jacund (as shown in Figure 1-1). On its west side, the Cupu community is bordered by the Tucuru Dam. The community was established along the Tocantins River, before the dam.

PAGE 34

21 Information on community formation is not very clear, since the occupation was gradual. As in the So Francisco case, the land also was a castanhal or brazil-nut concession. The pioneer families got to know the area while collecting brazil-nuts. It was originally a forested region with few anthropogenic activities, and gradually families started moving to the area, building their houses, and producing subsistence crops. A key referential year for the establishment of the community was 1986, when the manager of the farm denounced the presence of peasants on the land to the owner, who lived in the neighboring Maranho state. Despite the fact that the owner tried to expel the peasants with police support, nobody was injured. It seems that because she could not prove her rights to the land, and perhaps due to her weak political power, the owner gave up the struggle. After some time, the manager's departure was seen as a sign of the peasants' victory, and the families remained in the occupied land. They were also supported by the Catholic Church, and one of the main community leaders since the occupation, Mr. Soares, is still working as the Catholic Church coordinator17 in the community. Soares' family were involved in the Fire Action since it began in the community in 1999. Some time after the conflict, the community received the single official visit in the history of its land tenure regulation. At that time, representatives from the GETAT, an extinct federal institute for land reform created during the military dictatorship period, visited the land once, made a list of families living there, but never returned. Since then, some families have moved out, others have arrived, and some members have died. Bureaucratically, they cannot have official permission to use fire, since they still cannot 17 Because of their distant location, peasant communities in Amazonia have priests, nuns, or ministers that visit it periodically, every few months. In their absence, there are coordinators who celebrate mass weekly, and keep the church 'alive'. These persons have a key role in influencing peasant opinions.

PAGE 35

22 prove their land ownership. By 2000, the community still was not recognized as a Projeto de Assentamento or Settlement Project, making it very difficult for them to receive government support. The community was composed of 52 individual plots, and a small village. The Cupu village center (Figure 1-3), located at the edge of the main road, was composed of two churches (Catholic and Protestant), one school, a soccer field, and three small grocery stores, two of them with power generators and televisions. There was another village, with land donated to the community by one of the families, free of charge, but in practice that did not work, ending around 1993, due to absence of rivers or other water sources, and its distance (around 6 km) from the lake of the Tucuru Dam. The current village center is located about 2 km from the Tucuru Dam, where many families fish and raise their cattle during the dry season when the water level diminishes. Some families own land on the many artificial islands created by the Dam. The present village was not planned, and plots for houses had to be bought by residents from a peasant family, the Ribeiros. The land for the school, churches, and soccer field was donated by the Ribeiro family. The Ribeiros are one of the families that buy the peasants' milk production and sell it in the city. The head of this family was also director of the regional Association, and was engaged in the Fire Action since 1999. As of 2000, there was no electric power except for those families who owned generators. There were 27 houses in plots measuring 15 x 30 meters each, and on three of them there were small grocery stores. In two of those stores, there was a TV set in 1999. In 2000, two more families had bought TVs, the same two families that owned cars and used them to buy the milk production and transport people.

PAGE 36

23 Figure 1-3. Cupu Community village. Representation of the individual plots for houses, churches, school, and soccer field. The plots were sold to families by a peasant, who donated land for school, churches, and soccer field (Drawing by author). Research Methods The author of this thesis was the FATA/LASAT Fire Action proponent and its first coordinator, from the 1998 dry season until the first regional seminar, held in July 1999. The Marab region choice was a combination of the researcher's previous experience with community fire management in other Amazonian regions, her institutional connection with grassroots institutions in the Marab region, and those institutions'

PAGE 37

24 interest in developing a community fire management action. The field research was conducted between 1999 and 2000, and focused on two peasant communities in the Marab region. The So Francisco do Itacainas community, located in the municipality of Marab, was involved in the Fire Action since its very start, and the Cupu community, located at the municipality of Jacund, only engaged in 1999. This choice of communities with different times and degrees of engagement in the Action was intentional, to allow comparisons between them in terms of losses caused by fires and community organization for fire management. I had a twofold role of (a) implementing the Action during its first year and (b) evaluating it during the following years. I tried to make it clear to the families that the main goal of the study was to understand their strategy concerning fire management and their social-productive system, as well as support them with information about fire law and its impacts, other community fire experiences, and the importance of their community actions to other communities. Many families wanted to express their opinions in recorded tapes, hoping that I would help taking their voices to government representatives. In the So Francisco do Itacainas community, I facilitated some of the community meetings while Fire Action coordinator. The researcher's double rolefirst as facilitator and then as researchersurely biased the research results because of people's likelihood of responding in ways favorable to the Fire Action, and because of my own commitment to helping with community fire management. On the other hand, with a subject as sensitive as fireassociated with threats of fines and imprisonmentmy previous trusted relationship with the community was a key strategy in getting access to the details of their views and practices with respect

PAGE 38

25 to fire management. My approach stressed listening and observing the words and actions of community members, and conveying them in this thesis in their own words. This in-depth qualitative understanding of peasant's perspectives on fire management emerged from the researcher's close relationship with the communities. To offset the potential subjective bias in the research, I combined this close familiarity with a sampling design in the communities that would encompass a wide range of types of families, in order to capture diverse perspectives in the interviews. This purposeful sample provided systematic data on social characteristics, and burning practices during two dry seasons, to complement oral histories, participant observation, and other qualitative techniques. The combination of these methods was effective in providing a unique understanding of peasant views on fire management. The researcher's direct support received from key grassroots organizations (STR, FATA, LASAT, COOCAT, and AGRAF) was crucial in obtaining the confidence of families and their openness during the interviews conducted on fire management, on the losses caused by fire, and on fire monitoring. This was very important because families were aware of IBAMA's prohibition and penalties, and wanted to make sure that the research data collected would not be used against them. The regular visits to the field helped to build trust between researcher and families, and the academic nature of the study became clearit definitely had nothing to do with an IBAMA inspection. In So Francisco, members of the Agro-forest group supported by FATA were always ready to lead me throughout the community for family's interviews and burnings. Their presence introducing myself as someone to be trusted (connected to FATA and STR) certainly had a strong influence on families' willingness to talk and show me their realities. Members

PAGE 39

26 of the Fire Group in the So Francisco do Itacainas community, in turn, claimed that my regular visits (from 1998 to 2000) supported the adoption of local rules for fire use for those families not used to participating in community actions. The Cupu community, visited in 1999 and 2000, was not as organized as So Francisco at the time of this research, and a key leader (Catholic church coordinator) was the main peasant that helped with this research, introducing me to key families. From this first contact, the majority of the families were interviewed by myself, but always referring to those leaders and to the grassroots that this study was connected with. The relatively lower level of general community organization of Cupu families may be one of the explanations for their lower degree of formal organization of fire management in particular. As to its conceptual basis, the present study adopted the fire classification developed during the FATA/LASAT Fire Action. Accordingly, 'uncontrolled fires' were defined as fire events which break out from the planned burning area; 'accidental fires' were defined as uncontrolled fire events regardless of the use of any measure to prevent and/or combat uncontrolled fire. In this context, 'criminal fires' were defined as uncontrolled fire events in which the use of measures to prevent and/or combat uncontrolled fire was insufficient or absent (Carvalheiro & Aquino 1999:16-17). Real names of communities and peasants were changed to protect them from any eventual consequence of this study. Selection of the Communities and Families The So Francisco do Itacainas community had taken part in the Fire Action since its very beginning in 1998 and was one of the communities where FATA had developed its Perennial Crops Project. The second community studied, Cupu, was chosen because it

PAGE 40

27 joined the Fire Action one year later, in 1999, and it is one of the communities where LASAT developed its Community Forest Management Project. The families studied were chosen as a purposeful sample with the help of community leaders, and comprised more than 50% of the families of each community. Leaders and researcher discussed the main general variations among families, emphasizing that representatives of each "type" would be included in the study sample. After evaluating the differences among families, a consensus was reached regarding final criteria for selection of the families, which included variations in the following features: (a) participation in different community activities (different religions, participation in workers' associations, and mutiro or collective works), (b) length of residence in the community (from pioneers to newer residents), (c) location of residence (on the land, in the community village, or in the city), and (d) main source of income (agriculture, cattle, schoolteacher, grocery owner, middleman for milk market, with/without government loan). The selection process resulted in 39 families (61%) in the So Francisco do Itacainas community (the community with initial involvement in the Fire Action) and 29 families (56%) in the Cupu community (with later Fire Action influence). Sixty-seven families were interviewed during the first set of interviews, plus 5 families that wanted to participate later. Nine families from the first set of interviews were not interviewed the second time because five of them had sold their lands, three were abcent during the interview period (health problems), and one had moved away to the city. The families that did not take part in both sets of interviews were not included in the sequential fire analyses. The interviews focused on heads of families, both men and women. In this

PAGE 41

28 context, the heads of families were the older members, but the answers many times included other family members' opinions. Questions often generated discussions among the family members about the issues considered. In addition, 12 oral stories were narrated and recorded. Data Collection Information was gathered through unstructured interviews (interview guide in Appendix D), participant-observation, and oral histories (Lofland 1971; Holstein & Gubrium 1995; Rubin and Rubin 1995). Two data collection methods were used in this study: unstructured interviews sequentially conducted in two sets (at the beginning of the 1999 and 2000 dry seasons, before burning activities commenced) and participant observation during burnings (during the dry season of 2000), which provided both quantitative and qualitative information. Participant observation was also carried out in the communities during visits to the households and to families' burned and planned-to-burn fields, as well as during social events (religious celebrations, meetings held by the Association, parties, TV viewing). Oral histories were carried out with the elders and provided detailed material on their own life stories, their involvement in conquering the land and its characteristics, how the community started, its changes through time (families arriving and leaving, roads, schools, transportation, vegetation, weather), and interpretations of causes and consequences of different factors (partnerships, organization, challenges). The information was documented in field notes (Emerson et al. 1995), recorded on tapes, and photographed. Data gathered was coded (see Appendix E for Coded Book) and organized in a database for quantitative analyses. Recorded narratives were transcribed and translated by the author for qualitative analyses (see Appendix B for

PAGE 42

29 original narratives in Portuguese). Analysis and results combine both quantitative and qualitative data. The first set of sequential interviews, conducted in 1999, covered three groups of characteristics: social, productive and fire practices (fire topic related to the previous year burnings and plans for the coming one). In order to evaluate FATA/LASAT Fire Action impacts, the second round of interviews (which occurred in 2000) repeated the same questions regarding fire practices that had been asked before, and provided the data to make year-to-year comparisons, whose results are presented in Chapter 4. For interviews with selected families, the social characteristics included information on age, size and origin of family, number and places of migrations, length of residence in the community, level of formal education, kinship, religion (Catholic, Protestant, Afro-Brazilian etc.), association with grassroots organizations, families' plans to remain in the community or to leave, their points of view concerning community needs, and other topics considered relevant by the interviewees. Productive conditions referred to information on size of the family plot, distribution of the main ecosystems (percentage of primary and secondary forest, pasture, agricultural field, and perennial crops), quantity of cattle, labor (family, community and paid labor), equipment (chainsaw and backpack sprayer), and access to government loans. Results on social characteristics and productive conditions are presented in Chapter 3. Sequential fire practices interviews, referring to 1998 and 1999 dry seasons, were based on FATA/LASAT Fire recommendations, and included information on accessing official law for fire use, community organization for burnings, communication between neighbors, period chosen to burn, burning after the rain, use of preventive and control

PAGE 43

30 firebreaks, number of people on the burning day, use of backfires, losses caused by fire. ecosystems chosen to burn (primary or secondary forest), and burning of pasture. Results compare practices throughout the studied years. Fire topics also studied included peasant's discourse on a rich variety of concepts, such as the meaning of 'losses' in their productive systems, the importance and effectiveness of fire use (successes and losses), individual versus community labor, evaluation of the FATA/LASAT Fire Action and governmental laws, implications involving changes in fire use and alternatives to fire, as well as proposed solutions. In addition to the sequential fire interviews and oral histories, evaluating the FATA/LASAT Fire Action included observations of burning events in the So Francisco do Itacainas community in September and October of 2000. Initially, the study planned to include burnings in different situations occurring in the two communities. However, due to the frequency of rains and the scarcity of the researcher's field time, these observations were limited to the So Francisco do Itacainas community. The researcher was particularly careful not to interfere with families' decisions regarding the choice of the day on which to burn. Although it was no easy task, since the families knew that I was there to observe their burnings, and they clearly expressed their helpfulness, I managed to stress to them that my purpose was to understand what their choices were so that I could minimize the impact my presence might have on their fire management techniques. An effective way to make the families feel more comfortable with the presence of researchers was to carry out complementary interviews on qualitative information in between burning events. This way, families would not feel bad for letting the researcher 'waste her time' while they were still not ready to burn. These

PAGE 44

31 complementary interviews brought more information not only on the families' evaluation of the Fire Action itself, but also on their understanding of governmental regulations and cultural strategies, as well as community organization for fire management, and formal association challenges, among other topics. Conceptual Discussion on Participatory Approach This study analyzes the FATA/LASAT Fire Action according to its development of a participatory approach coordinated by grassroots organizations (FATA and LASAT) with colonist peasants in the Marab region. Due to different meanings that 'participation' can assume, this section will explore the concepts behind the Fire Action, as well as its main methods. The analysis of the Fire Action is presented in Chapter 2. 'Participation' has been a common word to those who are involved with development, and it is used by different actors (such as social groups, NGOs, government, researchers, donors), in different forms, for very broad purposes. Because it has became a vague label, different typologies were developed to specify different kinds and levels of participation, and interests of the social actors involved (Bordenave 1992; White 1996; Ingles 1999; Stone 2003). The common ground among those typologies is a differentiation in kind (from passive to active), and level of participation (decision-making power at the different activities). In general, the authors agree that the higher the power in decision-making, the higher the participation. As Schmink points out, "the degree of participation by different local groups in project decision-making and implementation is a key factor in empowering local groups to defend their own interests and to develop and adapt the institutions required to sustain natural resource management strategies over the long term" (1999:6).

PAGE 45

32 The methods developed under the participatory approach were based on Paulo Freire's Pedagogy (1986), in which consciousness, the self guide, is a process "by which human beings participate critically in a transforming act" (Freire 1985:106). Consciousness, therefore, is an active process. "Knowledge that exists today was once only a possibility, and it then became a new knowledge, relative and therefore successive to yesterday's existing knowledge" (Freire 1986:115). What Freire argues is that new knowledge, even proven scientifically to be the most suitable for a certain situation, cannot be deposited in people's minds (known as "banking education"), but has to be connected with previous knowledge, and previous understanding of the subject reality. This connection between previous and new knowledge is made via a thematic 'problem-posing' methodology, possible only through dialogue. According to Freirian Pedagogy, in thematic problem-posing situations, the role of the educator is to propose problem situations or themes. The starting point must be the present, concrete situation of the learners, not only at the intellectual level, but also at the level of action (Freire 1986:85). Learners' ways of explaining the world involve their comprehension of their world and presence in it (Freire 1997:76). Educators should help the learners to arrive at a more critical view of their reality by, through dialogue, analyzing the dimensions of their reality, that ultimately are dimensions of total reality. This analysis involves the exercise of abstraction, through which, by means of representations of concrete realities, learners and educator seek knowledge of that reality. The instrument proposed by Freire for this abstraction is codification, or representation of the existential situations of the learners (Freire 1985:51). Codification may take the form of a photography or sketch that represents reality, or a reality constructed of a proposed

PAGE 46

33 theme by the learnersa dimension of the reality. Codification, thus transforms what was a way of life in the real context into an "object" in the theoretical context. Therefore, the learners, rather than receiving information, creatively analyze aspects of their own existential experience, represented in the codification. In the case of Fire Action, the thematic problem-posing proposed to the colonist peasants at meetings was "how to use fire better", which became synonymous with "how to reduce losses caused by escaped fires". The codification exercise was developed with colorful drawings on cardboard tags ("Burning Dynamics") representing land use types (such as primary and secondary forest, pasture, slashed vegetation, perennial crops) and elements usually present in burning situations (people, houses, firebreaks, wind, sun, buckets, fence). Those drawings were mobiles, combined to represent burning situations (real or proposed), leading to peasants' re-codification of their knowledge through critical analysis of their own situation (for more details see Chapter 2). Freire states that one of the important points of the consciousness process is 'to provoke reactions of the world, not as a "given" world, but as a world dynamically "in the making" (1985:106). Humans are praxisthe unity of action and reflectionas reflection and action, which truly transform reality, are the source of knowledge and creation (Freire 1986:91). Neither pure action (not guided by reflection of the problematic situation) nor reflection (theoretical abstraction of reality) alone can lead to critical analysis. In this process, dialogue is the encounter in which reflection and action of the dialoguers are addressed to the world which is to be transformed. The Fire Action was planned in a way that each dry season would be a cycle of learning by praxis (learning-cycles): reflection about the current reality would generate

PAGE 47

34 actions, which in its turn would generate a new reality, again subject to reflection, in a continuous knowledge process. This means that solutions were not fixed, but had to be incessantly confronted with reality. This led to increased awareness on the part of peasants: through reflection on a given practice, ingenuous curiosity became self-conscious so as to advance to the critical stage (Freire 1998:43). Peasants, as active actors through the dialogical method, incorporated and appropriated the solutions developed by themselves. Many of the peasant narratives express that it was the first time they were using certain practices proposed by the Fire Action. Many of the peasants that were actively involved in the Action since its beginning, after only one year, referred to the Fire Action as their own action, probably because of Fire Action incentives that each community developed their own set of rules. Peasants involved in the Fire Action in a more passive manner (absent during meetings) referred to it as an IBAMA's action (see Chapter 4 for Fire Action evaluation). A key aspect of the approach was therefore to develop and support forms of communication and social organization that would support, and multiply, participation in fire management activities. Thus, Freire's approach dictates that in order to overcome a problematic situation it is necessary to go through it, and not to stay away from it, or deny it. In this context, effective solutions involving peasant societies can be reached by a learning process, in with basic elements are: respect for local knowledge, dialogue, critical reflection, and collaboration. The FATA and LASAT, through their history of support and collaboration with local peasantry, represented key institutions for the development of a participatory approach to fire management for peasants: trust in the institutions' intentions and

PAGE 48

35 methods; open channels of communication and representation; and faith in peasants' ability to overcome the problematic situation. The FATA and LASAT supported a technician to coordinate the Fire Action, and provided transport, food, and lodging for regional meetings, and educative materials. No peasant received any payment for fire management. Institutions and peasant leaders believed that peasants should get involved only if they believe it was worth it to them, but not as a direct economic source. Payments create dependency on outside inputs, if benefits are not linked to resource management and conservation (Schmink 1999:6). They can discourage others who receive no payment to get involved, and weaken the credibility of peasant leaders' real commitment. Especially in fire management cases, in which fire can result in fines and jail for its users, trusting in leaders was a key factor in conflict resolution situations. The Fire Action strategy was supporting each community to develop their own set of rules for fire management, including mechanisms of compensation in cases of losses caused by fire. Some peasants were not used to participating in community activities for fear that leaders were working for IBAMA, a governmental institution historically perceived as "police," acting only in enforcement, and always against peasants. Leaders' volunteer work, supported by trusted grassroots organizations, were strong arguments to convince outlying (non participant) peasants to get involved in the Action. In sum, in presenting learners' own objective reality (how and where they are), using a problem-solving method, with a thematic investigation, peasants begin to revise their previous views of their real world through codified situations. They achieve an understanding of their previous knowledge. In so doing, they expand the limits of

PAGE 49

36 knowledge. In this process, the educator exercises with the peasants a critical evaluation of their world view, resulting in their clear involvement with the real world in transformation (Freire 1985:33-34). This process promotes peasants' decision-making participation, once they get involved in creating solutions. This study evaluated changes in fire practices in 1998 and 1999 in two communities involved with the Fire Action, and observed burning situations in 2000. Despite the short time, this study found changes in fire practices at individual (family) and collective (community) levels, and fire losses were low after the Fire Action started. In the So Francisco do Itacainas, involved in the Action since its beginning, changes were stronger than in the Cupu community, involved one year later. Organization of the Study This introduction has discussed how the government strategy for controlling fire use in Amazonia based on a top-down approach, remote monitoring, and enforcement towards fire users is unlikely to be effective in bringing consciousness to colonist peasants. I argue that colonists possess empirical knowledge on fire management, which has been ignored by many government representatives, technicians, and academics. Another point stressed so far is that a participatory approach in which peasants share decision-making power, analyze their reality critically, are challenged using problem-posing exercises through action-learning cycles, and encouraged to develop and implement community solutions, will be more effective in managing fire. Instead of defending or attacking fire use, this study assumes that fire use is the option for the majority of colonist peasants, and they have to be seen as partners in the process of constructing better fire management strategiesincluding participating in the definition of "'better" fire management.

PAGE 50

37 In Chapter 2, I present and analyze a participatory Fire Action, coordinated by two grassroots organizations, FATA and LASAT, in the Marab region, which involved peasants as well as other grassroots organizations, and some indigenous groups, during 1998, 1999 and early 2000. I address the participatory approach adopted, based on Paulo Freirian pedagogy, the process that originated in the proposed Fire Action, and the recommended practices for fire management. This set of recommendations developed by the peasants themselves provides the framework for the subsequent analysis of actual practices used in fire management by two communities during two successive dry seasons. Chapter 3 presents descriptive information about the two communities and the families studied. Families' cultural interpretations of living in communities, and the importance of forest and fire are also discussed. This Chapter demonstrates some key aspects of local knowledge and situation that are typically overlooked or misunderstood, and which should provide the point of departure for fire management programs. Chapter 4 evaluates the impacts of the FATA/LASAT Fire Action on the two peasant communities for 1998, 1999, and 2000 dry seasons. The So Francisco do Itacainas community got involved in the Action since it started in 1998, while the Cupu community in 1999. The history of losses suggests the positive impacts of the Fire Action on peasants' changes regarding fire management. The assessment of implementation of the Fire Action's recommendations focused on different phases: before burning, on the burning day, after burning, and controlling accidental fires. Changing practices within and between communities, are compared in order to identify impacts of

PAGE 51

38 the Action on families' fire practices. This Chapter also interprets peasants' narratives on the Fire Action government and fire Action evaluations. The last chapter, Chapter 5, summarizes the main findings of the study, confronting the three main research questions regarding government actions, colonist empirical knowledge, and the participatory approach analyzed. Other experiences on community-based fire management in Asia and Africa are compared with this study, and suggestions are made for the management of fire in the Brazilian Amazonia.

PAGE 52

CHAPTER 2 COMMUNITY FIRE ACTION Introduction An office is a small and compact setting, isn't it? And the Amazonian environment is a large, vast one. Thus I would ask the [government] agencies not to limit their actions to their own offices, but instead to make themselves present in the communities too18. Mr. Valdir Silva, leader of the So Francisco do Itacainas community, during the first regional workshop. In the Marab region, the history of colonists' relationships with both non-government and government institutions reflects on their present partnerships, their trust in them, as well as their willingness to engage in government policies, especially the laws which directly affect their productive practices, as the case of fire use. Conditions influencing the transformation of landless migrants into colonist peasants, specifically in the Marab region are embedded in a long political struggle for the creation and the re-creation of their identity, a process shaped by peasants' forms of organization and resistance, as well as their strategies of permanence on the conquered piece of land. The above speech reflects the governmental history of absent support to peasants and their grassroots organizations. Government is criticized as remote and bureaucrat, creating office solutions distant from peasant realities. In this context of fragile links between peasants and government, I argue that top-down actions are even less likely to be adopted. Complementarily, this study supports the idea that colonist communities have valuable empirical knowledge, and they should 18 Translation by the author. Original Portuguese texts are included in Appendix B. 39

PAGE 53

40 participate actively in decision-making concerning fire use and management. It does not defend the notion that peasants' actual fire use is perfect, but on the contrary assumes that changes are needed and argues that only with peasants through participatory approaches can better practices and strategies be developed. The participatory approach adopted follows a problem-posing methodology, based on codification of the actual reality to develop consciousness about the reality, and learning-action cycles. This methodology is based on Paulo Freire Pedagogy (1986), and it is only possible through continuous dialogue among partners. This Chapter is organized in two parts: The first part describes the two grassroots institutions coordinating the Fire Action: the FATA (The Tocantins Araguaia Agrarian and Environment Foundation) and the LASAT (Tocantins Social-Agrarian Laboratory). The second part presents the Community Fire Action conjointly coordinated by the FATA and the LASAT, an innovative experience of community fire management which involved peasants as well as their grassroots organizations, and some indigenous groups, during 1998, 1999 and early 2000. The text addresses the participatory approach adopted, and the process that originated the proposed Fire Action, as well as summarizing the recommended practices for fire management. Chapter 3 presents the characterization of the communities and their families. The impacts of this Action on two peasant communities will be evaluated in Chapter 4. Peasantry Historical Construction in the Marab Region The Marab region has a long history, and migrants arriving, their strategies, and their means of production, are part of this history. This region is connected to an extensive process of expansion started in the 16th century, which ran from the coastland into the inner continent. Velho (1982:29) described how this process relates to the

PAGE 54

41 economic cycles of colonial Brazil. As a result of its geographical location (at the mouth of the Itacainas river into the Tocantins River, and highways) and its richness in natural resources, the colonization of the Marab region was marked by several cycles of exploitation (brazil-nuts, rubber, gold, and gems, cattle, timber, minerals and electric power dams). Those economic cycles induced and still generate different migration cycles which directly influenced the composition of its population. During the dictatorial military regime (1964 to 1985), it was considered a national security region due to the occurrence of the Araguaia Guerrilla19, when social movements were severely repressed. The Church played a key role supporting peasants and other powerless groups in a struggle for their rights. In this context, a brief review of the main economic cycles and its protagonists should yield a general picture of the colonist peasant's formation. Exactly when Marab village was founded is still debatedJune 7 of 1898 according to Velho (1972), and 1895 according to Emmi (1999). However, authors agree that it started with the establishment of a commercial store during the rubber boom period. The merchants used a system known as aviamento, in which they would determine prices of rubber or nuts exchanged for goods in their storemainly groceries and working toolsat very high prices. This led the peasants to accrue never-ending debts, carried from one season to the next one, making the worker a new type of slave (Velho 1972:41). Due to the distorted relation with the extractors based on indebtedness, merchants obtained increasing power and acted as "landlord", later becoming the "owners" of rubber and brazil-nut trees. Around 1919, following the rubber bust, the 19 The Guerrilha do Araguaia was a resistance movement coordinated by the Partido Comunista do Brasil or Communist Party of Brazil, sited in the area between Marab and Conceio do Araguaia, in direct confrontation with the military dictatorship from 1972 to 1975 (for more information see Schmink and Wood 1992: 72-74; Alves Filho 2000).

PAGE 55

42 exploitation of brazil-nuts grew stronger and used the infrastructure provided by the rubber industry, and for decades it remained the main product exported by the state of Par (Velho 1972:45). Emmi points out that in the late 1950's the brazil-nut business (from extraction to commercialization) in the region was concentrated in the hands of brazil-nut tree 'owners' in a perpetual system of "taking" lands to themselves or sistema de aforamento (Emmi 1999:70-71). Those two economic cycles (rubber and brazil-nuts) consolidated two main social groups in the region: a mass of low-paid workers and an oligarchic class, where a few local families held political positions in a monopoly, as well as credit (capital), means of transportation and commercialization, natural resources, and land. Peasants were completely excluded, as were other small brazil-nut entrepreneurs, and the indigenous (Emmi 1999:151-152). Despite this system of unequal distribution of wealth, there were no land conflicts officially registered in Par before the 1960's. Several generations of peasants lived without any official document to give legal power to their rights over their own land (Treccani 2001:307). This is explained by the fact that "ownership" was associated with the products extracted above ground and their accessibility, not with the land itself. In addition, accessing land deeds was (and still is) not a possibility for peasants that are, in practice, excluded from the legal system due to its high costs (they are required to have a lawyer representing them and to follow up on the legal suit), to a sluggish judicial structure, and to the police that generally connive with thugs (Santos 1984:457-458). It was particularly during the military regime that the Brazilian government started to provide for the means of speeding up the occupation of the Amazonian area. The government intended to assure national sovereignty using policies that included

PAGE 56

43 investments in local infrastructure (such as federal and state roads), interregional migration and economic development (Browder and Godfrey 1997:68). Those policies were organized outside the region, in the National Integration Plans or Planos de Integrao Nacional and had as consequences deep changes in the productive structure. Since the region is strongly connected with the history of extractivism in Amazonia, the last few decades were marked by the decline of the traditional extractivism involving brazil-nuts, rubber, fur and small-scale crystaland diamondmining in rivers, giving place to new forms of extractivism: logging, large-scale mining (Hbette and Moreira 1997:12), and agro-cattle ranching. The so-called Large-Scale Projects carried out by manganese, bauxite, nickel, cassiterite, gold deposits, wolframite iron, copper and caulim mining companies were associated with hydroelectric plants, which supported them with power (Schmink and Wood 1992:66-67). Logging activities grew as the frontier advanced, along with agriculture and pasture implementation. Through the National Plan of Integration, with government calls in the media, an example of which was the famous "land without people to people without land", the Amazonian frontier attracted people from all over Brazil and from diverse social classes. Since Brazil was facing increasing inflation rates, transferring capital to Amazonia represented an excellent option for investors, especially those investing in the cattle-ranching industry (Santos 1984:452-453). This caused a big and chaotic "land rush", where fake deeds and frauds were common, and multiple claims to a same piece of land occurred (Santos 1984:453). One of the most serious consequences was the closing of the traditionally "open" frontier to migrating landless peasants from Northeastern, who

PAGE 57

44 started to face their "last frontier", "from where they have nowhere to go and nothing else to lose" (Treccani 2001:308). In Marab, roads made the city a strategic center for the region and brought significant changes to its social and economic structure. The brazil-nut concessions or castanhais were turned into large farms, and official colonization brought millions of colonists who no longer were dependent on traditional local oligarchies. Former brazil-nut extractors became posseiros, or squatters, making the pressure for land ownership even stronger (Treccani 2001:313). Most migrants were landless people expelled from other regions, who left in search of land and of labor opportunities in road constructions and other large construction projects, as well as small-scale mining positions (Hbette 1991:200). Frontier expansion in this scenario increased land conflicts between newcomers and old forms of land occupation, and among newcomers themselves. This increase provided the elements for the most violent land conflicts in Brazil. The social consequences of this development model are clearly reflected in the concentration of the means of production, such as the land itself, in the hands of multinational entrepreneurs or big large-scale farmers. The ways in which these policies reflect on the state of Par is the object of a deep analysis conducted by Girolamo Treccani in his book Violncia e Grilagem: Instrumentos de Aquisio da Propriedade da Terra no Par (2001). For example, Treccani shows that in 1996, 82% of the farms with areas of 100 ha or less represented 169,273 farms, corresponding to only 19.2% of the lands (totaling 4,328,158 ha). On the other hand, farms with 5,000 ha and larger represented only 0.2% of the total numbers of farms (419) but occupied 31% of the entire area of 7,138,104 ha (Treccani 2001:431). This difference in farm size is the

PAGE 58

45 consequence of a historical process of political and economical favor for certain social groups in detriment of others, fostering not only the widening of the economic gap, but also and more importantly, a strong cultural distance. In this context of land concentration and violence against peasants, the Land Reform has being made by peasants themselves. They occupy latifndios or large farms, and, supported by grass-roots organizations such as unions and pastoral commissions, struggle to transform the occupied land into peasant areas. Politically, peasant leaders call this land occupation process 'conquering land', while ranchers, elite, and media call it "land invasion." In the Marab region, the historical process of peasantry consolidation is reflected in the amount of conquered lands:20: from 1% in the early 1980s to more than 30% of the properties in 1998, as result of their resistance, supported by a few allies (Wambergue 2000:41). Maintaining conquered lands as peasant's communities is a bigger struggle for grassroots organizations, and one of their strategies is to support productive activities that help the family's permanence in their lands. Uncontrolled fires have became a threat to peasants' means of production due to losses caused in forested areas, crops, pasture, and goods. Therefore, managing fire in a way to avoid local losses has became a concern to grass-root organizations. Peasant Organizations Especially during the years of military regime in Brazil, unions and associations not connected to the government were forbidden and fought against with great violence. Two of the tenets of the military government were undermining a movement called 20 Grassroots leaders and some studies (Trecanni 2001) affirm that in Brazil there was never a Land Reform but only colonization projects. Colonist peasants' institutions use the political term "conquering" to refer to land occupation actions and their struggle for pressuring governmental agencies to recognize their rights over the occupied land. On the other side, media and elite use the term "occupation"'.

PAGE 59

46 Peasants' Leagues or Ligas Camponesas, and the halt of the land reform program that started with Brazil's previous democratic president, Joo Goulart. Both led to the death or "disappearance" of several community leaders (Oliveira 1989:31). In this scenario, the Catholic Church experienced a constant evolution of its social concerns (Catholic Church Archdiocese of So Paulo 1998: 124), and grassroots organizations tied to the Catholic Church were often the only means for peasants to voice their needs and their opposition to events occurring in the frontier areas (Schmink and Wood 1992:180). The so-called Journey Church or Igreja da Caminhada, built by supporters of the Liberation Theology movement, opened its doors to peasants and, in 1975, established a committee called Comisso Pastoral da Terra or CPT (Land Pastoral Commission) whose goals were "to interconnect, assist, and activate the support to rural social movements" (Almeida 1991:261). The Church, together with the CPT at a community level, created ecclesial "base" communities called Comunidades Eclesisticas de Base (CEB), in which local leaders would conduct religious services and facilitate community organization. From this point on, the military regime directed its violent repression not only to peasants and indigenous people, but to priests, nuns, and union leaders, who were starting to get reorganized. After five years of military regime, a CEB was started in the region and a CPT office was set up in the city of Marab. Several local leaders strengthened their commitment with the CEB movement and later became union directors, regional officers with the Worker's Party (PT), with the SDDH (Par State Society for the Defense of Human Rights) with lawyers supporting peasants in judicial cases, the FETAGRI (Rural Workers Federation), and the University of Par, among others. In these circumstances,

PAGE 60

47 some of the FATA/LASAT founders (inside the CAT Program) played key roles, for example, coordinating the CPT/Marab, heading four peasant unions, and teaching at universities. This will be discussed in the next section. Although the Rural Workers Unions (STR's) were created under the military regime to control social movements organized in rural areas, the STR board was gradually passed on to those with a strong commitment to the rights of peasants, long before the regime fell. Nevertheless, the fall of the military government in 1985 gave place to greater violence against colonists, for the powerful landowners were responding to the new presidential decision to promote a nationwide land reform, where the region known as Bico do Papagaio would represent a priority. The "big" farmers set up an office to start the Rural Democratic Union (UDR), and large ranchers in the region started recruiting private militias to fight land invasion, which was likely to be carried out by peasants but also supported by government agencies attempting to expropriate the areas regarded as their own (Schmink and Wood 1992:188). Between 1989 and 1999 the average number of murders resulting from land conflicts in the Marab region was 120 times higher than the country-wide average (Forum das Entidades pela Reforma Agrria do Sul e Sudeste do Par 2001:4). This means that a leader or peasant living in the Marab region and engaged in a land conflict between 1989 and 1999 had a much higher chance to be killed than any other worker living in any other Brazilian state. This history of violence against peasants and the impunity of farmers and loggers21, both supported by 21 From 1985 to 2003, the Land Pastoral Commission (CPT) registered 1,003 violent cases related to land conflicts, with 1,349 dead victims. Only 75 cases went to judgment, with few considered guilty (CPT 2004).

PAGE 61

48 the legal and the executive systems, represent the main obstacle for partnership agreements to be implemented between these social groups in the Marab region. Peasants' history is marked by active resistance in order to guarantee their rights, in spite of harsh conditions and violence. Due to the historical distrust between government and peasants, top-down governmental actions imposed on peasants in order to control their natural resources management, especially those related to their main source of income, such as fire use, are likely to fail. This study argues that actions that see peasants as partners, with decision-making power, through a methodology that respects and supports their culture, with effective channels of communications and conflict resolution, are more likely to succeed. FATA and LASAT This section describes the two institutions coordinating the Fire Action, FATA and LASAT. Their history, as well as their goals and achievements are closely connected with peasantry development in the region that directly reflected on peasants' willingness to adopt the productive practices promoted by them. FATA was composed by peasant union leaders (STR's), and LASAT by researchers connected to the local university. Both FATA and LASAT were non-governmental institutions created in 1988 under a wider agricultural and environmental program called "Tocantins Environmental Center" (CAT). At the time of its foundation, the CAT Program was a partnership agreement celebrated between four Rural Workers' Unions (STRs) and the Federal University of Par. The main goal established by the agreement was "to assert their [peasants'] identity, to achieve full citizenship and to express their determination to live and to take their rightful place in the construction of a fair and fraternal society" (Hbette 2000:xxii). In a moment when land was being conquered by peasants through struggle, peasants'

PAGE 62

49 challenges were to build good conditions to establish themselves, as well as to produce and guarantee social reproduction in their lands. From the goals set out by the CAT, it is clear that justice and fraternity were principles linked to those defended by Liberation Theology. This may be due to established partnership with the Land Pastoral Commission, also reflected in the composition of the CAT founding board: CPT's first coordinator in the Marab region, Mr. Emmanuel "Manu" Wambergue; four peasant union directors who grew as leaders inside the CEBs; and a professor from the local university, who was assistant to the CPT, Mr. Jean Hbette. Also, STR Directors had grown up inside the church groups. The identification of the CAT Program with the peasants' struggle is well reflected in Manu Wambergue's evaluation of the role played by the CAT Program on its 10th anniversary as "one of so many fruits that flourished from the peasants' struggle in the region" (2000:41). Peasants' struggle was not violent; their form of resistance was their resilience and active search for new solutions, despite harsh conditions of exclusion from government incentives. Communities' Social Organization The two communities evaluated had been partners with FATA and LASAT. In the So Francisco do Itacainas community, in September 1998, FATA's perennial crop group (for implementation of perennial crops) led the creation of the community association, the AGRAF, in an attempt to work more closely with families' needs than did the regional association, created in 1994 to facilitate access to federal government loans. The Association's main goal was to "carry out sustainable development activities in their community" (AGRAF 1998). They borrowed the necessary money for bureaucratic

PAGE 63

50 expenses (around US$ 512)22 from FATA and paid it back later with their laborone week's work of ten menat the FATA headquarters. Very often, the cost of an organization's creation is supported by local politicians, creating a certain relation of dependency on the part of the peasants and the organization. AGRAF's representatives proudly explained how they had paid off the loan with their own labor. The recognition of the So Francisco do Itacainas community as a Projeto de Assentamento (Area for Land Reform) was a result of its leaders' organization for solutions. The community's leaders participated and supported families' participation in demonstrations organized regionally by the STR and FETAGRI when families camped out at the INCRA headquarters in Marab, an event called Grito da Terra, or 'Cry of the Land'. In 1999, 77% of the families in this community had at least one member in the community association, and 95% were members of the STR. In 2000, AGRAF's members elected a new board (6 men and 1 woman), all of whom work as volunteers. Families at the Cupu community had a close relation to the COOCAT through one of its directors who lives in the community. COOCAT has supported the marketing of aa fruits,23 among other products. Since 2000, they have been helping with the preparation of a regional proposal for the planting of coffee, along with the Jacund STR and Jacund municipality representatives. The peasants' regional association was working with COOCAT and STR to achieve official recognition of their existence as a 22 In September 1998, the average conversion rate was US$ 1.00 = R$ 1,17 (Banco Central do Brasil 2004). 23 Aa (Euterpe oleracea Mart.) is a native Brazilian palm in Amazonian region whose fruit is much appreciated in regional markets, which are growing in national interest. The palm apical bud is appreciated nationally and internationally for palm heart, but the extraction of the Atlantic Forest species (Euterpe edulis) requires felling the tree. This Amazonian species is very promising for sustainable harvest because of its sprouting capacity.

PAGE 64

51 community (Projeto de Assentamento), supporting them in acquiring land title and loans. In the Cupu community, the majority of the families (62.1%) were members of a regional association created to facilitate access to federal bank loans (FNO) in 1996. Seventy-one percent were members of the STR. There was no specific community association. In the past, the community had worked in the FATA's perennial crop project. Ten families took part in LASAT's Community Timber Project. As an indicator of their hope in this Project, by 2000 none of the participants had sold any trees to middlemen since the Project started in 1997. FATA/LASAT Community Fire Action The FATA and LASAT direct interest in controlling fire use was to protect alternative projects for peasant productive system: FATA perennial crops implementation24 and LASAT forest management.25 Local peasant organizations believed that uncontrolled fires threatened alternative production systems in slash-and-burn agriculture and cattle ranching, consequently leading to new migration cycles, conflicts and suffering. The proposed Fire Action did not aim to encourage families to stop using fire in their productive systems, but rather to promote a more responsible use of fire, specifically preventing and controlling undesirable losses. 24 Between 1991 and 2000, FATA implemented the PAF (Agricultural-Forest Project) to establish and to market perennial crops such as fruits and lumber. Later, activities of horticulture, honey production, and Aa palm (Euterpe oleracea) management have also been incorporated. In 1998, when the Fire Action started, 44 communities, including three indigenous communities (a total of 76 groups) were working with FATA in the different activities (Carvalho 2000:65). 25 LASAT started a Community Forest Management project in 1996, initially involving three peasant communities from three different municipalities. The main goal of this project was to promote sustainable lumber extraction and processing by peasants, which should result in less damage to the forest and in higher profit.

PAGE 65

52 Besides the grassroots institutions' and leaders' reasons to control fire, peasants in general had many questions about the new official fire regulations. The dry season of 1998 was the first after the fire disaster occurred in Roraima, and the changes introduced by IBAMA to the fire management laws were: the requirement to submit an official form to obtain a burning permit, the obligation to provide details on the type and the size of firebreaks, and for the first time, a total ban on fires in municipalities with high rates of burning (measured by the INPE National Institute for Spatial Research). Although not all rules were new, the latest change introduced by IBAMA was to massively divulge, through the radio and TV, the requirement to obtain a license to burn, of the fire use ban, and that IBAMA staff would be enforcing the fire law in the field. In the communities, FATA/LASAT technicians were initially thought by many peasants to be IBAMA members, who were disappointed to find that the technicians could not provide them with any official burning permits, or even get them to be heard at governmental levels. They wanted to explain to IBAMA not only their reasons but also their very need to burn, which is clearly expressed in the extract below: Many have burned, but I have not and many others have not either. We expected someone to give us explanations, because we do not understand the law, and we want to find a way to work legally. Because in a place like this, with no roads and no assistance, if a poor peasant has to pay a fine for burning his field, it would be better to bury him straight away. So, let us say a family cannot harvest their field, how would they live? This is why I want to know what we have to do in order to burn. Do we have to make firebreaks, to get a license? I want you to explain it to me so that I will understand it. If you had come earlier, I believe nobody would have burned without permission. Our group always works together to do things the right way. But we never had anybody come here to explain things to us. Mr. Joo Pinheiro, Jabuti Community, Oct 14th 1998, community meeting (Carvalheiro 1999:18). Despite peasants' investment in other productive systems, fire was their main productive tool for agriculture and pasture maintenance, and they wanted to know what to

PAGE 66

53 do in order to produce their crops without becoming a "criminal". Peasants complained that it was hard for them to have access to information about government regulations. For example, in the whole Marab region there was only one IBAMA agency bearing responsibility for all environmental issues in the region (from illegal logging to river pollution), whose technicians were not used to visiting colonist communities. The radio was helpful to bring general awareness. However it did not explain the official laws and regulations, and the necessary steps that should be taken in order to burn. The Fire Action adopted a participatory approach, based on Paulo Freire's pedagogy (Freire 1986), as presented in Chapter 1. The Action adopted the problem-posing methodology, whose theme was fire use, more specifically "how to manage fire in order to keep it only where it is planned" (agricultural fields and pasture maintenance). The resulting recommendations made by the Fire Action were intended to be closely tied to their social, economic and ecological realities, and given the urgency of the problemestablished perennial crops and forest areas were being burnedalso to produce short-term answers. Below is presented the application of the participatory approach adopted. Process for defining Practices for Fire Management The Fire Action stemmed from a study proposal made in 1997 to LASAT by the author of this study, and the interest of LASAT, FATA and peasant leaders for better fire management. The diversification of peasants' productive systems with community forest management and investments in perennial crops depends on the ecosystems, which are very susceptible to fire, due both to fuel accumulation and to their proximity to the ecosystems where fire is present as an agronomic tool. The FATA/LASAT Fire Action26 26 For the purpose of this thesis, the Fire Action coordinated by FATA and LASAT will be referred to as FATA/LASAT Fire Action, or just Fire Action.

PAGE 67

54 started during the 1998 dry season and ended at the beginning of 2000, from lack of funding. The Fire Action had economic support for the Action coming from outsiders (distant donors), was implemented by technicians from the local grassroots institutions (FATA and LASAT), and planned, developed, and evaluated with peasants at regional and community levels. Documents produced (such as reports) were shared with community leaders and unions. Educational material was developed by technicians and peasants. The author of this thesis coordinated the Fire Action from 1998 to July 1999. The proposal was based on peasants' active participation, because they were the ones who had developed knowledge on using fire in their environment, and therefore were skilled to evaluate what was not working well, and to be important partners to improve fire management. Another reason to get peasants involved is that they are the ones who ultimately take the final decisions on fire management. If they are actively involved analyzing the situation, proposing and approving changes in their practices, those changes will be closer to their realities (infrastructure, social, and productive) and thus more likely to be incorporated into their activities. The action-learning cycles support peasants' development of their own proposals, and recognized that changing practices is a dynamic process, and therefore has to be analyzed periodically. Peasants had a reading of their reality, and the starting point of the action-learning cycles was their actual understanding of fire. Actions of fire management were carried out at the community level, and learning (evaluation and planning) in meetings at community and regional levels. Each dry season corresponded to a full learning cycle (see Figure 2-1). Regional seminars took place in the beginning of a dry season, for evaluation of the previous and planning of the following, and at the end of

PAGE 68

55 that dry season, also for evaluating and planning, closing a cycle by incorporating lessons learned. The Fire Action worked for two action-learning cycles, in 1998 and 1999. Figure 2-1. Representation of the FATA/LASAT Fire Action showing key activities of the participatory Fire Action. Action-learning cycles started in the beginning of a dry season, with activities of evaluation, planning, and dissemination at regional (workshops) and community level. Actions were taken at the community level, and evaluated in the following workshop (By author). The Fire Action did not establish any instance of centralized decision, nor did it plan to follow any pre-defined agenda. Instead, it sought to work flexibly to reach and involve the families' realities at the community level, at same time promoting community organization. As form of incentive, grassroots institutions (FATA, LASAT and STR) declared they would support organized communities in case of conflicts with IBAMA regarding fire use, and the Fire Action's technician would visit those better organized

PAGE 69

56 more communities regularly. Subsequently, regional workshops would draw general lines of action, but it was at the community level that more specific agreements and recommendations should first take place, since at this level those agreements are made and monitored. During the dry seasons, the Fire Action technician visited some of the communities in order to monitor their actions. After the dry season was over, community representatives presented their community experiences at regional workshops. More than 40 communities were invited, with support for transportation, food, and lodging. This flexible strategy was chosen because it gave families an opportunity to play an active role at the community level, where they could change their community, shaping it to their reality, while at same time they would feel important for having played an active role and for the motivation to participate (Bunch 1994:69). The meetings held at the community level were key to the participatory approach adopted because they created opportunities for conducting discussions, accessing information, exchanging experiences, planning, as well as for dissemination and evaluation of the proposals. In general, every family had heard something about fire management and the penalties imposed by the government for its use. However, they had always been uncertain about legal details. Many families expected that IBAMA's burning guidelines would be presented to them in 1998. Given that all families use fire in their productive systems, community meetings attracted many in search of information on fire use and of an opportunity to complain about the government's proscription of fire use. The government's prohibition of fire use27 resulted in a dilemma for peasants: they 27 Since 1998, the government has prohibited fire use during the dryer weeks of the dry season, in general for eight weeks. Even those who got permits are prohibited to use fire during this period.

PAGE 70

57 do not want to break the law on the one hand; on the other, they depend on using fire for their social reproduction. They did not want to go against the law, but they did need to burn. After receiving no information from IBAMA by the end of the dry season, they started burning in a careless manner and without an effective plan (Carvalheiro, 1999:21-22). This condition allowed for deep reflections about their own situation. The visits paid by the Fire Action coordinator to the communities reinforced the work done by community leaders because, in addition to promoting discussions, the peasants who did not attend the regional workshop could access information by asking questions directly to the technician and comparing answers to clarify some of their uncertainties. Some families are less involved in community organization and do not trust leaders or their intentions regarding fire management, suspecting that leaders would receive some form of payment from IBAMA to control the use of fire by families. This type of suspicion represented a very common conflict among communities in the Marab region. The ban imposed by IBAMA on the use of fire represented bigger losses to peasants than uncontrolled fires usually did, because for them there is no production at all without fire. Therefore, for those with poor access to information, the regular payment of a "salary" would represent a logical justification for some peasants to go against their own class, i.e., to support fire control. The technical visits strengthened the legitimacy of the work done by the leaders, and emphasized the fact that none of the work done by peasants was done for money, since they worked as volunteers, and FATA/LASAT only assumed the cost of transportation, food and lodging for regional workshops held at the FATA Center. The fact that the leaders involved in the Fire Action were generally the same ones taking part in other actions promoted by FATA and LASAT helped to

PAGE 71

58 encourage other peasants to believe in the proposals of the Fire Action, and to play an active role in it. Regional one and three-day workshops were held at the FATA Center, and brought together peasants and technicians to discuss and evaluate their community actions and government laws, to plan strategic actions, and to present and discuss their own proposals with the representatives of both governmental and non-governmental organizations. The Fire Action coordinator also held meetings at the STR's (Rural Workers Unions) and visited governmental institutions (INCRA, IBAMA, SEMMA, EMATER, Public Defensor Office). First Action-Learning Cycle The first action-learning cycle started in a regional meeting, aiming to analyze the present fire use situation based on peasants' understandings, challenging their knowledge with information on fire law and fire occurrence in the region, and drawing up with the peasants a first proposal for fire management to be carried out in the coming dry season. At the community level, during the dry season, actions consisted of meetings, to inform and discuss proposals defined in the regional meeting. At the end of the dry season, the closing cycle would be an evaluation of the action carried out, and planning for the following year dry season. For the closing regional meeting of the first cycle, leaders decided to invite other grassroots institutions related to peasants, as well as governmental institutionsthe presence of IBAMA would be crucial for peasants to present to them their developed proposals. Details of each step of the first action-learning cycle are presented as follows. The first action-learning cycle started at the beginning of 1998 dry season, during the workshop on "Perennial Crop Management" held at the FATA Center, when an entire

PAGE 72

59 day was dedicated to discuss fire use with peasant participants. In this first meeting, there were a total of 31 peasants (from 20 communities), and 4 indigenous (from 2 tribes), all connected to the Perennial Crops Project, and representatives of two communities also involved with the Community Forest Management project. There were also 13 technicians (8 from FATA, 4 from LASAT, and 1 from EMATER/Itupiranga). Instead of presenting to the peasants the methodology of the coordinator's previous experience in community fire management in Paragominas (Mattos et al. 2002), and then discussing how it could be adapted to their conditions, the strategy used was exactly the opposite: first the discussions were on what participants would do in order to prevent losses caused by the fire, and only then was the other experience presented. This strategy was very positive because participants were given the opportunity to see that the results achieved by their discussions were similar to a consolidated experience carried out in other Amazonian regions, which raised their self-esteem. As discussed in Chapter 1, problem-posing methodology empowers participants by raising questions on their situations and by leading to solutions based on their own understanding of their reality, making it possible to change this reality. In addition, participants will be more likely to appropriate practices they themselves developed. In order to theorize on peasants' knowledge, fire use was codified into three different phases: (a) before a planned burning event, (b) the burning day, and (c) after a planned burning started. The exercise resulted in the creation of a fourth category, (d) control of accidental fires. This codification constituted of discussions on each phase by a brainstorming exercise, and ideas organized in flipcharts, facilitated by the Fire Action coordinator. The problem-posing question was "What can we do before, during, and

PAGE 73

60 after the use of fire in order to prevent uncontrolled fires?" The following complementary question was "What can we do in order to control fire?" Every idea was considered, even when some participants did not agree on its effectiveness. This first exercise resulted in a set of general activities (listed below), with more or less agreement on their efficiency or feasibility. The division or codification of fire management actions into phases was key for planning during meetings because it allowed the connection of the several fire prevention and control techniques practices to the best period to use them, according to peasants' usual slash-and-burn system steps (Figure 2-2). The upper part of Figure 2 shows the approximate division in months of rainy and dry seasons; the middle part shows six basic slash-and-burn steps influenced by fire, each followed by a bar indicating the period when the step is taken; the lower part indicates the three Fire Action phases, before the burning, on the burning day, and after the burning. Following this time division during discussions, peasants were challenged to think in terms of responsible fire use during each of the slash-and-burn steps based on their own previous experience or information. Thus, this step of relating phases to local system was a major innovation that emerged from the participatory process. This emphasis on collective action was a second major innovation. In all four phases, the overall recommendation for 'union' was consensual, meaning that the work should be done in groups of family members, friends, and neighbors. It was said that fire was "democratic," i.e., it would burn everybody's land without distinction of origin, kin, religion, etc. Acting together, families would be better able to manage fire efficiently. For instance, when choosing the burn location (sometimes during the rainy season) it is

PAGE 74

61 important to analyze the fire danger to surrounding ecosystems, also taking into account neighbors' ecosystems. Neighbors should unite to organize a period to slash the vegetation, allowing fields to dry and be burned together at the same time. Burning in groups would help in the use of prevention and control techniques, such as firebreak and backfire. Figure 2-2. Codification of fire management actions into phases (before the burning day, at the burning day, and after burning) and its connection to the best period to use them, according to peasants' usual slash-and-burn system steps (Drawing by author).

PAGE 75

62 The first systematization on Community Fire Management, developed in the first regional meeting, June 1998 (Carvalheiro 1999:5-6), is shown in Figure 3-3. After this brainstorming exercise, the coordinator presented an experience of fire management also developed by colonist peasants in Paragominas, through a participatory approach from 1996 to 1998 (Mattos et al. 2002). Participants concluded that their recommendations were very similar, differing only in the details about the size and type of firebreaks, and in the form of community organization for the case of accidental or criminal fire. The exercise of facilitating peasants to organize their knowledge empowered them to believe in their own capacities. The similarity between strategies from the other case and their own knowledge showed them that they already knew a lot, and that what was really missing was their organization and motivation to implement practices in a more consistent way. Another sign of their self-confidence was that they did not request copies of the Paragominas Booklet, but proceeded with their discussion. Their decision on developing their own knowledge rather than basing on other experience was a demonstration of empowerment. Peasants attending this first meeting decided not to define the details about techniques or community rules at that moment, but instead to focus on more general recommendations. They claimed that details on each family's agreement should be developed at the community level. It was clear to the peasants attending the workshop that the recommendations defined during the meeting bore no official power to replace the IBAMA law, currently in effect. Still, peasants were motivated to develop their own rules in order to confront IBAMA's top-down rules, because of their belief that if they could manage fire well (causing no losses to the forest or to their neighbors), they would have the legitimacy to

PAGE 76

63 face any eventual confrontation with IBAMA's representatives. At the end of this meeting, some agreements were reached: Each participant should present and promote this discussion on community fire management in their own communities; Each participant should support the creation of Fire Groups open to all members of their communities; The FATA/LASAT consultant should visit all the communities where Fire Groups were created, giving priority to those communities that could inform possible dates of burning events; FATA/LASAT should prepare instructional materials in the form of posters and/or flipcharts, which would include the results of the workshop; the material would be distributed and discussed during the regular technical visits conducted by the FATA; FATA/LASAT would support the coordinator's work for the 1998 dry season with funds provided by the DFID. The results of these agreements were positive. Jadiel Souza (personal information), FATA director, said that more than 30 colonist communities with families involved with the FATA and LASAT projects organized Fire Groups. The Fire Action produced a poster and a serial flipchart summarizing Fire Management recommendations. The Fire Action coordinator visited nine communities, facilitated community and Fire Group meetings in five of them, and made subsequent visits to two of those five communities. Burning situations (before, during and after) were also monitored.

PAGE 77

64 Before a planned burning: Talk to neighbors about the location of agricultural fields and pasture management; also talk about when to slash the vegetation (in order to be able to burn together); Talk to community members about the best time to burn (in order to avoid burnings in the beginning of dry seasons); Use firebreaks; Fell dead trees on the borders of agricultural fields and forests or other ecosystems; th is technique is more important at the border where fire is usually stronger; When felling trees, try to direct them to the center of the agricultural field in order to reduce the amount of fuel near the field borders; Inform neighbors in advance of the day when burning is planned, and confirm it one day before; Invite neighbors to be present and, if possible, to help with the burning. On the planned burning day: The person responsible for the burning needs to be present; Do not burn a field alone, but in groups of at least 4 people; Use 'fire-against-the-wind' or contra-fogo (start burning on the side opposite to the main wind); If possible, organize water containers and tools for fire control; Insist on neighbor participation. After a planned burn starts: rning for Continue observing the bu as long as flames are high; Return to the site at the end of the day and on the following day in order to observe any escaped fire; Extinguish and/or destroy any dangerous tree stump or trunk with the potential to fli ng sparks at adjacent ecosystems. Controlling fires: Make a control firebreak ( varrida ); Backfire to control the main fire; her the fire with branches and Smot soil; Extinguish it with water. Figure 2-3. Summary of the firs t systematizatio gement, developed in the first regional meeting, June 1 998 (Carvalheiro 1999:5-6). n on Community Fire Mana

PAGE 78

65 Despite participants' initial frustration at not being able to communicate with government representatives, the meetings held at the community level led to rich discussions and to community planning. The results of the first regional workshop were presented, and their local burning situations were discussed. Each peasant's burning situation was presented to the community, which generated discussions on possible dangers and how to minimize them. In order to codify the group's visualization of the different burning situations, the FATA/LASAT technician developed an instructional kit called "Burning Dynamics" (Carvalheiro and Mattos, in press). This kit contained colorful drawings on cardboard tags representing the ecosystems (primary forest, secondary forest, perennial crops, agricultural crops, pasture) and other significant elements in a burning event, such as wind, sun, fire, rivers, firebreak, water buckets, houses, fences, equipment, and people (Figure 2-4). There were several copies of each drawing, and they could be positioned on a larger card according to different situations. The peasants would build their situation and keep moving the elements around (fire, people, water) to represent the burning dynamics, and incorporate suggestions. The kit was very helpful in promoting discussions, especially helping group understanding of a burning situation. Usually, when a peasant explains his burning situation, he or she draws in the ground with a stick, showing the agriculture field, surrounding ecosystems, wind direction, where did they started burning, etc. After a while, others in the group cannot remember what the ecosystems were around the burning, or the wind direction, or neighbors' land; the discussion does not go deeper because the person responsible for the burning has to keep explaining the burning situation again. With the burning-dynamic kit, all participants could visualize the situation, and make suggestions, adding or moving

PAGE 79

66 objec ts, facilitating discussions of possible alternatives for the studied situation. Thmeetings held at the community level attracted many people, including those not uattending them. As planned before, the second regional meeting called "Management Alternativfor Small Scale Production" was held in December of 1998, at the end of the dry seaduring three days. It was composed of three parts, as follows: Community Fire Management (December 1 and December 2). Assembly open to invited institutions (December 4). rearranged according to each burning situation, helping the group to visualize (Drawings by author). As part of the action-learning cycle methodology, the goals of the workshop wereto (a) evaluate the fire Action for the 1998 dry season, (b) organize its results in a fire ose sed to es son, Community Forest Management (December 3). Figure 2-4. Some of the elements used in the "Burning Dynamics," which could be the burning situation and assisting in the consequent in-depth discussions management proposal for the region, and (c) present and discuss this proposal with

PAGE 80

67 invited institutions (IBAMA was the most hoped to attend). A total of 47 peasants representing nine communities attended the meeting (28 men, 14 women and 5 children). A stronger prese nce of peasants was expected, but the workshop coincided with the begin a researcher from the p. nections of local fire use with regional, national and international contexts. Participants discussed sustainability in of not a good symbol for peasants because the anteater is an appreciated dish in Amazonia, and not at are their hat the representation of an experienced technician prese ning of the rainy season, which forced families with related agricultural activities tostay in their lands. Two FATA Directors, as well as FATA technicians, the Director of a Honey Production Association, the Fire Action coordinator, and University of Washington were also present during the entire worksho The first half-day was used for theorizing the intercon Amazonia and the growing susceptibility of the forest to fires, as well as data from INPE on fire statistics, the insertion of Marab in the Arc of Deforestation, international interests in stopping fire use in Amazonia, the fire disaster that occurred in the state of Roraima, the history of governmental fire laws and their current changes. The most recent IBAMA booklet on fire management (IBAMA/GTA, 1998) was analyzed. Table 1 shows a summary of this analysis. Each participant received a copythe booklet, which was read by the group, and time was allowed for discussion. As a result of this study, participants first criticized the symbol chosen by IBAMA to represent forest fire prevention and control: a humanized anteater. It was a consensus that it was all related to fire management awareness, even to those who know that forests bitat. It was agreed tha nting practices in accordance with their reality would be more likely to be accepted

PAGE 81

68 by them. During Fire Action evaluation, peasants referred to elder peasants as modegood management, as discussed in Chapter 4. Participants also concluded that official fire permits were too hard to obtain, especially for peasants. Among other bureaucratic procedures, it required numerous documents peasants simply did not have, and which were often incompletely listed in the booklet. Although some of the participants knew that the acquisition of some of those documents cost a lot in fees they usually could not afford, the booklet did ls for not explain any of thament (IBAMA/GTA 1998), as discussed during the second workshop in December Summary of the critical assessment of the IBAMA Booklet on Fire Management t. Participants expressed their concern about what this money was used for. Table 2-1. Peasant critical assessment of the IBAMA/GTA booklet on fire manage1998 (Carvalheiro 1999:21). WEAK POINTS STRONG POINTS Burning permission requires land tenure documFosters community organization entation fee for inspection by deforestation agents groups to burn on different days fee peasants or of the regionconcerning The Brazilian government's regulation of printed in small font, making it hard to fire use shown at the end of the booklet is read, and uses technical jargon Some strong points were that the booklet encouraged community organizationregarded by participants as one of the most powerful tools for fire management. Also, the instruction saying that participants of fire groups should not burn on the same day Burning permission requires payment of Instructs participants of different fire Burning permission requires payment of Fosters use of adequate equipment Firebreak guidelines do not fit the reality of length, distribution and format

PAGE 82

69 was evaluated as positive, because the fire groups have more time to observe the burning, and to be present if any accidental fire starts. E veryone emphasized the need for adequth; slashed primary forest will not burn well when it receives a strong rain), and (iii) the severity of the dry season (some dry seasons are so harsh that burning without a rain it is hard to). criticism was placed on the nd format of the ines presented in the booklet. IBAMA requires a 3-meter-wide firebreak hennd hboring areas. Three meters is a difficult target for peasants to meet because they rs is out of the question, since this is generally a communities. Peasants also questioned the e smaller the risk of accidental fires (IBAMA/GTA 1998:14), because sparks can fly very far. Firebreaks located far away from e ate equipment to optimize fire control and minimize the risks involved in the activity. Participants held that the IBAMA guidelines for burning after the first rains could not always be followed, depending on (i) the location of the agricultural field in relation to its topography (places close to swampy areas do not burn well after rain), (ii) the previously existing vegetation (burning pasture after rain events helps grass regrow event becomes highly dangerous, while in others dry seasons rain is more frequent so that avoid burning before the rain The hardest length, distribution, a firebreak guidel in general, and a 6-meter-wide firebreak w the burning is adjacent to forest a neig have no access to machinery; six mete wider lane than most roads accessing theirstatement that the wider the firebreak, th the agriculture field were evaluated to be much more efficient (as discussed in thFire Action recommendations). In addition, the IBAMA booklet's recommendation to make firebreaks in the middle of large agricultural fields in order to "burn in steps"

PAGE 83

70 sounded like a joke, or rather like a clear sign that the people who wrote the instructions had never visited a slashed vegetation in Amazonia prepared to burn. Slashed veg(especially from primary forest but also from secondary forests) is composed of many fallen trun etation ks and branches superimposed one over the others, in many layers; walking throuis vice choseere it f burnings, t ted gh a slashed field means walking over trunks, and in small areas over the ground. Constructing firebreaks by hand in this condition is impracticable. All this, associated to other guidelines such as building trenches inside the agricultural field when the land steep, carrying large tree trunks located close to firebreaks to the center of the field, andthe representation of slashed vegetation as pasture field, resulted in participants' total disbelief in IBAMA's real expertise and consequently in its competence to provide ador to monitor fire management in Amazonia. The last aspect participants criticized in the IBAMA booklet was the font size n to print the text content of the booklet and the language used, particularly whpresented governmental fire regulations. Besides being too small, which made it hard to read, the language was too technical and therefore difficult to understand. On the morning of the second day of the workshop, representatives of each community presented an evaluation of the 1998 fire season in their community. They described the severity of the dry season, examples of Fire Group work, and the lack of losses in most of communities. Only one community recorded loss because owhen a peasant burned a field without asking for the community Fire Group's support. Members of a community in which accidental fires were successfully prevented complained about the difficulty and danger involved in fire prevention and control, for iis done with no outside support or adequate equipment. This discussion was illustra

PAGE 84

71 using the example of a burning in which the Fire Group helped. After most of the fire had been extinguished, the group left. Later that same day, when the responsible person and his 12-year-old son returned to observe, they found that the fire had spread to the neighboring forest. There was no way to call for help, and the two of them spent manyhours controlling the fire, without water or equipment. They had trouble breathing for many days after the event. All participants then agreed on the need for equipment and expressed their concerabout not being able to follow IBAMA's laws. This concern is shown in the following speech: When people work together for a common goal, nothing is difficult. And our and that IBAMA does its part too. (...) I n community is like that. We hope that from this point on, things [fire use] go right, f the government is pressing us to control fire, then it [the government] has to provide us with the appropriate equipment. workshop (Carvalheiro 1999:26) Besides limitations with controlling fires because of absence of external support and equipment, the peasants' evaluation at the workshop closing the first action-learning cycle for the 1998 dry season showed positive reactions to the Fire Action campaign. Without discontinuing their use of fire, the families involved claimed to be better organized to prevent and control fire in their communities, and therefore reduce losses. However, they worried about uncontrolled fires that originated in large farms, and the government's outlawing of the use of fire and punishing those who use it. Those uncertainties about government actions regarding fire use are well expressed in the speech by representative Augusto Ferreira, when evaluating the work in his community: Many people that are really interested do not want to obstruct the law; no one wants to go against it. However, we have to know how far we can go without breaking the Nobody can control it [fire] like it is done now, using only our hands. Antonio Pintassilgo, Tracu Community, December 3rd 1998,second regional

PAGE 85

72 law, because if it [the law] does not conform to our reality, how do you follow thereality?... Mr. Augusto Ferreira, Nova Jerusalm community, December 2nd, 1998, regiona recommendations in the [IBAMA] booklet, for example, that do not match our l meeting (Carvalheiro 1999:23). On the afternoon of the same day, the participants were divided into groups according to their communities. Each group developed an exercise of codification using the Burning-Dynamic kit. They were asked to represent a common burning situation in their community, and to discuss in each group, fire practices for prevention and control of accidental fires. Burning-Dynamic material was provided. After that, each group presented the burning situation using cardboard visual aids, and described the practices takingpractiecosystems, practices, etc. Two of the seven examples discussed in this exercise are prese place before, during and after the burning. This discussion on fire management ces was very rich, with different situations of vegetation to burn, surrounding nted in Figure 2-5. Figure 2-5. Two examples of Burning Dynamics represented by community December 1998. Each example represented a possible burning situation, and representatives at the second Regional Seminar, held in Marab city in it was used to present and discuss this situation with seminar participants. According to Freirian Pedagogy, the codification is a visual representation of

PAGE 86

73 the problem-posing situation, promoting critical reflection on peasants' realityand their critical self-insertion into this reality (Photo by author). Figure 2-5a shows the fire management example presented by the Nova Jerusalm and Plano Dourado communities (Carvalheiro 1999:27-28). They displayed a hypotd by of the lp if aks bserves dry wood or dry standing trees located close to the surrounding ecosystems. They return on the following days to check for any accidental anches and fromary d the family houburning fie hetical situation of a burning where the crop field to be burned was surroundepasture (on the right side), neighboring forest (on the left side), forest located in the family's land on top, and secondary forest (at the bottom). The wind blows from right toleft, and the family house and perennial crops are located to the right. The family organizes a group of five men to make the four firebreaks and to burn. On the day burning, they prepare tools (machete and scythe) and water, and start the burning from the left side, which is downwind. The woman stays close to the house and the boy remains at the firebreak, close to the fastest way out. That way, he can go and get henecessary. The group stays and observes until the strong fire is out, checks the firebrefor accidental fires, and o fires. In case of accidental fire, they try to smother the flames with green brsoil, also using water. If the fire gets out of control, one of them has to go for help. The other example, Figure 2-5b, shows the situation presented by representatives the Tracu and Josinpolis communities (Carvalheiro 1999:28-29). They portrayed a burning carried out by an individual family with the help of three neighbors. A secondary forest is located to the right of the burning field, to its left there is a prim forest, at the bottom there is a pasture with animals, and the perennial crops anse are located at the bottom right-hand corner. The black line around the ld represents firebreaks. They also showed a small brook crossing the

PAGE 87

74 ecosystems Firebreak width and location varied according to the neighboring vegetation: primary and secondary forest, and pasture. When the burning field was next to a pasture, the firebreak was located at the border of the burning field, and was 2 to 3 meters wide (or wider if there were fences). When the burning field was next to primary forest, the firebreak was located 8 meters from the field border, and was 1.5 meters wide. When the burning field was next to secondary forest, the firebreak was located 5 to 6 meters from the burning field border, and was 1.5 meters wide. The reason for locating the firebreak within the vegetation and not at the border of the burning field (as prescribed by IBAMA) was explained by a peasant, when he pointed out their burning situation: (...) because when fire goes with the wind, it throws sparks, and when it goes into throws sparks; when it goes inside these 8 meters of forest [between the border of where the forest hasn't been cut. It is easier to control the fire when the firebreak is (Carvalheiro 1999:28) During the presentation by the representatives of the Tracu and Josinpolis communities, they explained that this kind of firebreak works together with the contra-fogo, or preventive firebreak. In this practice, peasants are positioned on two sides of the burning field. The burning is started by those located downwind, and only then do the others start the main fire (which goes with the wind). The fire started against the wind runs slower than the main fire, but fast enough to create a strip of land without fuel. This results in the main fire going out before it reaches the field border, and consequently reduces the risk of fire reaching the neighboring vegetation. In addition, they talked about remaining in the burning field until the main fire was extinguished and returning on flame, nobody can stop it, even with a wider firebreak, because it [fire] jumps, the burning field and the internal firebreak], the fire runs close to the forest floor, 1.5 m wide. Jos dos Santos, Tracu Community, December 2nd, 1998, Regional Workshop

PAGE 88

75 the same day and on following days to check for any possible accidental fires. Equipment was also mentioned, mainly machetes, scythes, hoes, water pumps, and if possible a chainsaw. The peasants emphasized that a chainsaw was faster and more efficient in separating the burning from the green vegetation through varridas, or control firebreaks (firebreaks which are made after the burning was started and are thinner than preventive firebreaks). After the seven examples were presented, the practices and recommendations listedwere classified in five categories: equipment, before the burning day, during the day, after the burning day, and conflict resolution. Each of those topics was discussed bthe particip burning y ants. The participants also defined some widely used key concepts which at manyamondiscuFire Management Recommendations for the Marab Region, later organized into a booklet. The results of the peasants' discussion on concepts are listed below (Carvalheiro 1999: rules established in the Community Agreement. Accidental Fire: when fire is used as an agronomic tool, and although its user limits of the planned burning field, causing losses. Criminal Fire: when fire is used as an agronomic tool or for any other reason and ing times take different meanings, and which are basic for defining responsibilities g fire users, such as "responsible", "accidental," and "criminal" fires. This ssion on fire practices and recommendations developed into the first Community 35; Carvalheiro & Aquino 1999:16-17): Responsible Fire: when fire is used as an agronomic tool, and its user follows allfollows all rules established in the Community Agreement, the fire exceeds the its user does not follow the rules established in the Community Agreement. On the last day of the workshop, the first Community Fire Management Recommendations for the Marab Region were read and approved by the peasants takpart in the workshop. After their approval, the recommendations were presented to the

PAGE 89

76 invited institutions. The representative from IBAMA was anxiously expected bin spite of a personal invitation two weeks earlier (IBAMA has an office in Marab), none from IBAMA attended the meeting. There were representatives from four STRs (Rural Workers' Union): Itupiranga, Jacund, Marab and So Domingos do Araguaia, as well as the Director of Marab SEMMA (Municipal Environmental Secretariat), the Itupiranga SEMAGRI D y all, but o irector (Municipal Agricultural Secretariat), two technicians from attending institutions, to IBAMA, along with a letter explaining that they were his/her communities, and promote (1) the creation of a community fire nity received the approved proposals), and (2) the development of a conflict resolution ironmental Council, at the municipal level, with power to make decisions. This must be done through Municipal To disseminate the Fire Community Management proposals in communities by To organize another Fire Workshop at the beginning of the following dry season Second Action-Learning Cycle During the 1998-1999 rainy season, the Fire Action coordinator documented the 1998 community fire management experience, raising funds for one more year with INCRA, a technician from the Itupiranga EMATER branch, and the Marab Public Defender for environmental affairs. Notwithstanding the frustration because of the absence of representatives from IBAMA, the meeting went on as planned, and the key agreements achieved were: To send the Community Fire Management Proposal, as well as the list of all missed and encouraging their attendance in subsequent workshops; Each peasant should convey the Community Fire Management discussions to management proposal, based on the discussions in the workshop (each commuproposal; Institutions should work to create an Env representatives, and be approved by the Mayor; means of a booklet to be prepared by FATA/LASAT before the following dry season, (second action-learning cycle).

PAGE 90

77 DFID. The coordinator and other LASAT technicians organized the Fire Action proposalinto a booklet (with 750 printed copies),28 which was distributed after being approved at the July 1999 regional workshop. The third regional workshop on Community Fire Management, held at the FATA Center on July of 1999, brought together during three days representatives from 15 communities (14 peasants and 1 indigenous), comprising a total of 36 men, 13 womenand 6 children. It was facilitated by the Fire Action coordinator, a LASAT technician, and two FATA directors. This time, a representative from IBAMA was present. ThIBAMA state Director for Environmental Education attended the entire workshop, and the IBAMA representative a e t the Marab Office was invited to-and attended only the A emplorom the researoals of this workshop were to (a) review the 1998 experience, (b) represseason, (e) present and evaluate familstitutions. partially or totally achieved. The previous regional workshop was evaluated, and partic last day of the workshop (which was open to invited institutions). Two other IBAMyees were also present on the last day. Additionally, representatives f Paragominas Community Fire Management experience, a key leader and a key cher, were present on all days. The main g evaluate 1998 agreements and achievements, (c) present the Paragominas experience ented by key social actors, (d) plan the 1999 dry the FATA/LASAT Fire Action booklet, and (f) disseminate the Fire Action among newies and in The workshop was considered to be a positive experience, in which all goals were ipants concluded that four of the five agreements from the second regional 28 Funds for the booklet came from a field fellowship from the Natureza e Sociedade Program for this thesis.

PAGE 91

78 workshop had been put into effect. The exception was the creation of the Environmental Council that was never formed. More families from the communities involved and from neighboring peasant communities were interested in taking part in the Fire Action, and the pra s the A's ted in ion l. Horizontal communication among Amazonian peasants is very iition to esence of representatives from other communities at the workshop was considered positive sign of the Action's dissemination. The presence of the IBAMA director waworkshop's greatest achievement. For the first time, participating peasants and the representative from IBAMA were communicating face to face. Although the IBAMA representative had no decision power, and could only inform the participants of IBAMlaws, he was respectful and listened to peasants' complaints and accounts, thus taking a step towards reducing the distance between IBAMA and peasants. It was a learning experience for all. The Fire Action booklet was read in the assembly and after a few changes was approved by participants. The practices recommended in this booklet are presenAppendix C. The Paragominas leader and researcher and the peasants from the Marab regshared experiences, exchanging information, challenges, problems, and achievements.This peasant-to-peasant information exchange tends to empower them by furnishing a wider and more profound picture of their own situation, bringing consciousness of their situation at a regional leve mportant because in general they have limited access to any means of communication and transportation. In the previous months, the Fire Action coordinator had transferred her pos another LASAT technician, to focus on her graduate studies evaluating two of the

PAGE 92

79 communities attended by the FATA/LASAT Fire Action (the present thesis). During the workshop, the participants had the opportunity to meet the new Fire Action coordinator, and five communities were selected to be monitored in the 1999 dry season. It was established that one of them should be an indigenous community, so as to involve other actors besides peasants (the Suru tribe was chosen). Two communities were to be monitored by the author of the current thesis (one in Marab and the other in Nova Ipixuna). The other two communities to be monitored were defined in the assembly, according to spatial distribution (regions that had not been previously involved). Several other institutions were invited to attend the last day of the workshop, and representatives from five STRs, EMATER/Marab, SEMMA/Marab, a Fire DepartmLieutenant, two rep ent resentatives from COOCAT, the EFA Director, a representative from DFIDPT, nd conflict with police and gunmen. e water pump ays call the firefighters. The peasants replied by expla and an Honey Association Director were present. Several important NGOs (CMST, FETAGRI and CNS) were not able to attend because they were assisting 130 families in a la After all participants introduced themselves, they received the Fire Action bookletand it was read out loud by a peasant. The recommendations in the booklet generated several discussions. For instance, the firefighters' main comments were that the firebreaks were too thin and that the backpack sprayer proposed was not efficient to control fire. The firefighter Lieutenant suggested that peasants should use more effectivequipment, such as asbestos fire swatters (abafador de amianto) and a large (with a 20 m range), and should alw ining the reasons for the thin firebreaks, the lack of other kinds of equipment, the usually long distance from agricultural fields and pastures to water sources, and the

PAGE 93

80 absence of public telephones for several kilometers. This face-to-face discussion among representatives from different institutions and peasants was a great learning experience. In this process, different social actors had the chance to present their realities and cothem with government laws and peasant recommendations, making it clear to participantthat many of the fire laws were actually inapplicable. Based on this discussion, the Department Lieutenant proposed that a multi-institutional caravan be created, whichshould be composed by FATA, the Fire Department, IBAMA, the Municipal Environmental Secretariat, the army and the police. The caravan should not only ntrast s Fire inform about en t ly fire use but also punish non-conforming farmers. The FATA director agreed withthe idea of a multi-institutional caravan that would educate peasants about fire use, but without inflicting punishment. He defended his position by pointing out that changing fire practices is a gradual process, and if the caravan had an educative role, it would be seen by peasants as an ally; however, if the caravan punished peasants, it would be seby them as an enemy. As a result of the discussions in the workshop, participants agreed that (a) they would work towards the creation of a multi institutional caravan which should play an important role in raising peasant consciousness regarding responsible fire use, buwithout punishments, at least at this point; (b) the Fire Action booklet should be widedistributed in the Marab region, and its proposals should be acknowledged as official recommendations for peasants; (c) participants would raise funds to support the Fire Groups Forum, caravan expenses, and the Fire Groups; and (d) another regional workshop was scheduled for the late 1999 dry season, at the FATA Center, with the purpose of evaluating and planning the following dry season.

PAGE 94

81 During the 1999 dry season, the new FATA/LASAT coordinator visited all five communities. The caravan was not created, and the workshop planned for the end of the 1999 dry season was not held until the following rainy season (in the beginning of 2000), with low attendance (many dirt roads are impassable during rainy season). In early 2000, the financial support by DFID ended, FATA faced structural changes, and thus the Fire Action was discontinued. Nevertheless, two peasant communities continued to be monitored by ththis thesis, who also took part in burning events in 2000. The study explores the impact of this approach on community organization, fire losses, and changes in actual practices implemented among the Fire Action recommendations during two successive dry seasons. During this period, the results show a clear positive impact in both communities of changes in fire management practices. The combination of fear of IBAMA sanctionsand the possibility of developing community-based alternative programs appropriate to their situation, worked e author of well to reduce losses and change the timing and organization of burniing e ate s to the es, ng events. This led to burners taking responsibility for their burns and for followcommunity rules and practices, with support from other people in the community. Thespositive changes, and the practical experience of beginning to implement and evaluthem in different situations, provide an excellent basis for future fire management in the communities. The Fire Action depended upon outside technical assistance to facilitate the process of development and dissemination of the rules and practices, and meetingdiscuss them at the regional level, with authorities from different agencies. How absence of this outside technical assistance will affect future fire management practic

PAGE 95

82 and th e relationship with IBAMA, is an important question for future research on the effectiveness of the Freirian approach used in the Fire Action. Discussion This Chapter presented the process for developing FATA/LASAT Fire Action recommendations. Based on Paulo Freirian Pedagogy, the Fire Action approach was participatory by assuming a dialogic approach in which a technician was facilitator of the probl s for at e season. The impact of the Fire Action was monitored in two of these communities, one of which was involved since 1998, and the other was involved since 1999. em-posing situation: of fire management, codifying colonist actual knowledge of their reality through 'burning dynamic' exercises, and challenging them to critically analyze their own situation. Other innovations were (a) relating colonist productive stepsas a function of the burning day (before, on the burning day, and after) and practicefire prevention and control; (b) working with peasant empirical knowledge; (c) leading peasants to consciousness of their own situation (codification) before presenting external solutions; (d) emphasis on collective work as the basis for the action; (e) promoting peasants to develop their own agreements at community level; and (f) technical support the community level. The recommendations proposed by the Fire Action, developed by means of the presented participatory process, included regional and community meetings, and innovative practices applied by some community members, as discussed above. Such recommendations were organized into a booklet (Carvalheiro and Aquino 1999), with thpurpose of promoting community fire management by sharing the recommendations discussed and approved in the workshops. Copies of the Fire Action booklet were distributed to at least 44 communities related to FATA and LASAT, during the 1999 dry

PAGE 96

CHAPTER 3 cial s ute the important cultural context for developing fire management progrs ems. s COLONIST COMMUNITIES Introduction This Chapter presents a summary of the social-productive characteristics of 72 families in the two communities studied, So Francisco do Itacainas and Cupu, both located in the Marab region, emphasizing aspects related to fire management and soorganization. Those two communities were studied in order to evaluate impacts of the Fire Action on their fire practices, and results of this evaluation are presented in Chapter 4. Knowing some of the families' characteristics is important to understand their reasonbehind practices, and their decision-making, and also furnish a frame for disseminating this experience in other situations. As this Chapter will show, the history of these colonist communities, and their associated views on production, education, rural life, forest and fires constit ams. Age of Heads of Families The average age of heads of families was 40 yrs (SD=12.2) for women, and 47 yr(13.0) for men. There were, on average, 5.3 (SD=2.84) persons/house. Besides those living in the communities, some families maintain sons, daughters, or other family members living in cities, usually to continue their studies or because of health problIn the communities, twenty-two families (31%) of the forty interviewed families maintained on average four persons (SD=3.2). This maintenance was done with product 83

PAGE 97

84 (crops and animals), and money. Sending kids outside to study represents high economic and social costs for families, as discussed below in the education section. Migration and Length of Residence in the Community Data on families' originads of families migrated from other Brazilian states, which chaamilies as colonists. Most of the headse he -1 s their having only recently been politically divided29. Each head of family was ory, listing places, villages or municre, s showed that the majority of he racterizes the f of families living in So Francisco do Itacainas and Cupu communities (46% women and 57% of men) were born in the neighboring state of Maranho, located in thnortheast of Brazil (half of the Maranho state is located in the Legal Amazonia region). Only 8% of the female and 3% of the male heads of families interviewed were born in tstate of Par. The other female heads of families were born in Gois (18%), Bahia (18%), Piau (3%), Esprito Santo (3%), Minas Gerais (3%), and Cear (1.5%). Remaining male heads of families were born in Bahia (10%), Gois (7%), Esprito Santo(7%), Cear (6%), Minas Gerais (6%), Piau (3%), and Pernambuco (1%). Figure 3presents origin distributions, in percentages, for both male and female heads of familieof the communities. The Figure considers the states of Tocantins and Gois as one because of asked to trace his/her traject ipalities where he/she had lived. In order to differentiate places just visited from places where the migrant spent a "significant" time period, this study only considered places where the interviewee performed some productive activity, such as agricultuextractivism, and/or had a job, either in rural or urban areas. In both communities, 29 Gois was politically divided in two states in 1990, with its north region becoming Tocantins state.

PAGE 98

85 women had migrated an average of 4 times (SD=2.1) and men migrated an average of 5 (SD=3.0) times. n communities studied indicated a direct relation between length of residence and first land ownership. In 1999, the interviewed families' length of residence in both communities Figure 3-1. Origin distribution by state, by gender, for the heads of families of the So Francisco do Itacainas and Cupu communities (n=72). The majority of meand nearly half of the women were born in Northeastern state of Maranho,showing that few were born in Par state (Tocantins and Gois were considered as one because of its recent political division) (By author). The data regarding length of families' residence and land ownership in the

PAGE 99

86 averaged over 10 years, and most families were still living on their first plot of land. Theaverage length of residence was 11 years (SD=4.2). In So Francisco do Itacainas, 77.5% of the interviewed families had been landless before living in the community. twenty-nthe commfaming there famd cupy The impormanageme pick, to harvest, etc.), and a minority (28% n=69) own chainsaws. Therefore, the amount Eleven families (27.5%) were living there since the creation of the community, and ine families (72.5%) were living there for ten or more years. Of those living in unity for more than 10 years, 82.8% had been landless before arriving in the community, while 57.2% out of those living there for less than five years had been landless. The same pattern was found in the Cupu community. Eleven of the thirty-one ilies (35.5%) were living in the community since its creation, and of those livfor ten or more years, 77.0% had been landless before arriving in the community. Of the ilies living in the community for less than five years, the number of those that habeen landless before living in the community was only 66.7%. For many colonist peasants, the right to cultivate land is a result of their struggle for self-esteem, for better conditions for their family. Many interviewees emphasized the harsh conditions of living as sharecroppers, or low paid employees in rural and urban areas. These difficulties associated with the unaffordability of land led people to oclands as squatters. Occupying land in the Marab region was and still is very risky due to conflicts, high rates of malaria, and lack of any basic infrastructure or outside support. tance of owning and managing their own land is a key incentive for fire nt practices described in the next chapter. Family Labor System Family labor is the main productive force in the Marab peasant communities, and their means to control fire. In general, they do not have access to machinery (to plow, to

PAGE 100

87 of labor has a direct influence on the family's ability to produce. Besides family labor, the two other main ways to improve their labor force are family social organization focollective work (workday exchanges and mutiro, in which everyone works together),and obtaining the economic means to hire employees. For this characterization, this study considered only labor above 15 years old. The amount of labor force, divided by gender, was similar in both communities. The average number of female and male workers per household for both communitiwas 1.5 (SD=1.01) and 1.9 (SD=1.32), respectively. The labor force involved in workday exchanges does not involve money. It is a verbal agreement between two or more producers to work together in each other's land for a similar number of days. The mutiro, on the other hand, is the collective work of a group of people on a specific project, like building a house, harvesting rice, burning, etc. In g r es eneral, when the work takes e ies hire labor, especially during part of the year. The most common jons, for cutting weeds in pasture areas, several hours, the person or persons directly benefiting from the mutiro supplies most of the food for the meal. Hired workforce is that which implies payment of daily wages or salary. For the purpose of this study, it was divided into three frequency parameters: always hires, hires during certain time(s) of the year, and never hires. Table 3-1 shows the results for the three types of work, showing that in So Francisco do Itacainas there is a tendency for higher participation in workday exchange and collectivwork. Table 3-1 shows that around two thirds of famil bs were during dry seaso slashing forest, and making firebreaks along fences. Data show that workday exchanges and collective work are used more in the So Francisco do Itacainas

PAGE 101

88 community than in the Cupu community. This may indicate more cohesive social relationships among families in the So Francisco do Itacainas community. The basic proposal for fire management discussed in the Fire Action was collective work, and if iSo Francisco do Itacainas there is a higher willingness to work collectively, it is expected that it will reflect in their collective acti n on for fire management. Table(SFI) and Cupu communities, and its average distribution, in percent, for the collective work rates are lower than in the SFI community, which may Hiring Workforce Workday Exchange Collective 3-1. Most common kinds of workforce found in the So Francisco do Itacainas studied families. Note that in the Cupu community, workday exchange and indicate weaker community cohesion. (%) (%) Work (%) Times Cupu 12.9 58.0 29.0 35.5 64.5 13.3 86.7 Comm. Most of SometimesNever Yes No Yes No SFI 12.5 50.0 37.5 59.0 41.0 55.0 45.0 Productive System The families' productive system is mainly based on their land, and can be broadly classified into three categories: agriculture (including annual and perennial crops), animal raising (small, medium-sized, and large), and natural resource management (forest and rivers). Fire is mainly used for annual agricultural production, as well as for pasture maintenance. A few families have other sources of income, such as selling groceries, transporting people and production, teaching, working on a daily basis for large-scale farmers, or a pension. Those that have other economic activities also work on the land (cropping, raising animals) because income from the alternative sources available is usually very meager. In addition, many affirm that they like to work the land, as a statement of their identity as peasants.

PAGE 102

89 In both communities, the main activities related to the market are annual crop(rice, manioc, corn and bean), cattle (for milk production), extractive products (brazil-nu-Bertholletia excelsa, H.B.K.; aa fruitEuterpe oleracea, Mart.; babassu coconut oilOrbignya speciosa, Mart., Barb.Rodr.; cupuau fruit-Theobroma grandiflorum, WilEx. Spreng., Schum), and perennial crops (mainly cupuau, banana, orang s t ld. e, lemon, Basistrategy, wrops, or perennial crops. Independent of the strategies adopted, and the year of community foon (So Fra83 and Cupu in 1986), the proportion of primary forest in both communities was similar (r half h plotreas dtionsture,dary anperennial crops tended to show more variation (Figure 3-2). In the So Francisco do Itacainas community (Figure 3-2A), the average size of family plots was 45.7 ha (SD=21.7). Primary forest occupied an average of 49% of the plot, followed by secondary forest (23%), pasture (20%), agricultural field (4%), and perennial crops (2%). In the Cupu community, the average size of family plots was larger: 77.6 ha (SD=49.7). Primary forest occupied an average of 46% of the plot, followed by larger areas of pasture (41%), smaller areas of secondary forest (8%), and similar sizes of agricultural field (4%) and perennial crops (1%) (Figure 3-2B). Although in So Francisco the number of persons per house was, on average, 1.2 times higher than in Cupu, the average size of plots in the Cupu community was 1.6 times larger than the average plot in So Francisco do Itacainas. A probable explanation for this is Cupu's focus on cattle, which, when raised extensively, require large pasture areas. coffee and coconut). cally, the distribution of land uses is a consequence of families' productive hich may be focused on cattle, annual c rmati ncisco in 19 oughly of eac ), whe istribu s of pa secon forest d

PAGE 103

90 nas 1.2 1.6 times the ce the supply of secondary forest. Grass Figure 3-2. Representation of land use distribution in A) the So Francisco do Itacaicommunity and B) Cupu community (Drawing by author). Although in So Francisco the number of persons per house was, on average,times higher than in Cupu, the average size of plots in the Cupu community was larger than the average plot in So Francisco do Itacainas. A probable explanation for this is Cupu's focus on cattle, which, when raised extensively, require large pasture areas. Another indication of the community's trend towards focusing on livestock is type of ecosystem used for the crops: primary forest or secondary forest, since the continuous implementation of pastures tends to redu y species are not compatible with the locally called 'white' cultures (rice, corn, manioc) because grasses grow faster and spread profusely, and their roots pierce the manioc roots, compromising their quality. Thus, once grassy species are planted, that soil has to be plowed before it can hold any crops. So, if the supply of secondary forest

PAGE 104

91 runs out, this could lead to higher pressure over primary forests. In the So Francisco community, 27% and 38% placed their crops in primary forest area in 1998 and 1999, respectively. The So Francisco community showed greater preference towards perennial crops than the Cupu community. In So Francisco, 100% of the interviewed families grew perennial crops, while in the Cupu community, that rate was only 77%. Comparatively, in the Cupu community, in 1998 and in 1999, 67% of the interviewed families placed their crops where there once was primary forest, against 32% who placed their crops where there once was secondary forest. Confirming Cupu's families tendency to invest in livestock, in 1999 approximately 80% of the families (double the families in So Francisco) planted grassy species in their agriculture field area. As to livestock, in Cupu 93% of the families own milk cattle, while in So Francisco do Itacainas only 64% do. The average number of animals per family for Cupu. obserg upu y e is 33.4 animals (SD=22.5), and in So Francisco, it is much smaller: 6.4 (SD=9.4)Data indicated that the Cupu community had relatively bigger areas of pasture, and livestock is leading to higher implementation of agriculture in primary forest. Although the percent of forest in both communities is similar, plots in Cupu are bigger, and vation in the field showed that some families tend to buy neighbors' plots containinpasture, and concentrating more land. At same time that pasture may represent to Cfamilies a means of securing land tenure rights (Muchagata and Brown 2003:799), it malead to weaker social organization if migration out of the community is high. Another explanation for investing in cattle is its better economic return, since thfinancial returns obtained from cattle are higher when compared with agriculture and perennial crops. Cattle raising seems to supply some degree of security that family

PAGE 105

92 agricultural production cannot offer and neither can the government (credit, health, education). The agricultural products sold achieve low market prices, compared to theiproduction costs, and families do not have the necessary means to transport them, or even to store them to sell during the period in between harvests when prices are higher. Moreover, agriculture implies heavy and hard work, which requires lots of manual laboCattle, on the other hand, presents high productivity in Amazonia, allowing flexible anlower investment in labor (Muchagata and Brown 2003:800). Cattle production is not seasonal and has a year-long market. Additionally, its sub-product, milk, can also be sold, either directly or processed as cheese, as well as contribute to the nourishment of the family and of small animals. r r. d rrived that much for too fethings. Mr. Gomes (G): They don't understand. Who is crazy around here? Nobody. Few people around here are really [intellectually] limited. But the others, if they can't Formal Education This section starts with an elder peasant, illiterate, who during his whole life hasworked in agriculture. Mr. Gomes was talking while he was making a fishing net, a skill learned from his father in Maranho state, where he and his wife were born. They ain the Cupu community in 1986, with four of their eleven kids, always cropping and fishing. Mrs. Gomes helps as a midwife when necessary. Besides the importance families give to education, Mr. Gomes was shown to be very upset regarding outsiders' views on the relationship between illiteracy and stupidity, or incapacity, as expressed below. He, in his class awareness, attributed this discourse on peasant "illiteracy" to mean "incapacity" to powerful people who do not share their power: "it is too w." Question (Q): Some people say that farmers are illiterate and don't understand

PAGE 106

93 read, they can at least sign their name. Everybody has understanding. This is people you are talking about are a community, a strong one! People who say that We live under despicable conditions, but we are not aliens, we are all people and but they are not crazy. They have other very important forms of culture. And to be appreciated. But this contempt, contempt, contempt,(...) only God knows... too few. Mr. Gomes, 65 yrs old, Cupu community. When explaining their leve a movement, Mrs. Katia, from people who work under despicable conditions. These are from this [elite] community. () That's why we are suspicious of everyone. we need love and respect. Q: People can read the stars, can read the forest, the wind, the earth Mr. Gomes (G): They [illiterate people] don't have that culture [formal schooling], people [outsiders] offend them like that. That's not how it really is. They deserve Lots of suffering. What is lacking, Mrs. Katia, is responsibility, it's too much for l of education, heads of families were ashamed when it was low, but proud to explain a relo school for their daughters and sons. e e their experience. (Freire 1985:7-8) ls were built and maintained by families, staff salaries were paid by theoks) and b atively better access t To them, higher educational levels mean better work opportunities than for those without studies. Mr. Gomes speech finds echoes in Paulo Freire literature about illiteracy: [Illiteracy] indices, statistically compiled by international organizations, distort thlevel of 'civilization' of certain societies (...). This distortion fails to acknowledgtheir real-life experience and all the past and ongoing knowledge acquired throughThis study discusses this topic by first providing information on the education system in the communities, and years of schooling for studied heads of families. The following section addresses peasants' cultural understanding on land, forest and fire. In both communities, schools were implemented a short time after the creation of the communities. Schoo local government, and some basic student materials (such as books and notebo reakfast were supplied by the government. Both communities' schools, at the time

PAGE 107

94 of thi s study, went only up to 4th grade, and in every classroom, a single teacher cted more than one grade at a time. An average of one third of families maintain family members living outside the instrucommunity, as presented above. The majority of those living outside were students that completed the 4th grade offered at the community and went elsewhere to continue their studieden for faacquire new needs, and renting or buying a house, as well as maintaining it, entails monthrvision teenagers may stray from their origins of families, both female and male, were asked if they had attended school, and fothe sthouses of famil s. Sending kids to study in other places implies a high economic and social burmilies. It is expensive because in cities every good has to be bought, teenagers ly costs. In addition, without parental supe al goals and become involved with drugs or in promiscuous behavior. In many cases, the women heads of families move to the city to take care of the kids; the families end up divided between the community and the city, which brings down peasants' production, raises expenses, and contributes to migration to cities. Head r how long. In order to determine families' ability to access written information, udy inquired what was the highest education level of any family member in the hold, in general daughters and sons. The educational level of family members living outside the house was not considered. The main consequence of communities' precarious education system is the head ies' low formal educational level. In both communities, more than one third of the interviewed female (41%) and male (39%) heads of families had no formal schooling (n=61) (Figure 3-3a-b). The proportion of men and women whose studies were interrupted between the 1st and 4th grades is also similar, 39% of the female and 41% of

PAGE 108

95 the male heads of families. Data showed that only 20% of female and male headsfamilies studied further than the 5th grade of (for female and male, respectively, 16% and 13% i nterrupted their schooling between the 5th and the 8th grade, and 4% and 7% between the 9th and the 11th grade). Figure 3-3. Families' formal education grades attended in the So Francisco do Itacainas and Cupu communities. A) Women and B) minterviewed families in both communities who had no sc en heads of hooling, ranged from g s l s and daughters) in the household that was living n the 1st 31% to 46%. C) The proportion with no schooling was considerably lower when the research considered education level of any household member livinin the community, indicating that younger generations have been much morelikely to go to school. Because heads of families in general do not live alone, and share their skills, thistudy wanted to know the overall family schooling level. However the level of formaeducation for any family member (usually son in the So Francisco do Itacainas community was quite different when compared to heads of family only (Figure 3-3c). The rate of total absence of formal education was 3%, corresponding to two families of older people who lived alone. The percentage ofhouseholds whose highest level of schooling in the two communities was betweeand the 4th grade was 55 % of the interviewed families. The proportion of households

PAGE 109

96 whose highest level of schooling was between the 5th and the 8th grade was 31%. Theto 11th grade' category was to 11%. The higher level of formal education per family, compared to data on heads offamily only, indicates that the new ge '9th nerations are having more access to schooling, nd/or their parents tend to invest more in education than did the previous generation. Durinf topics, exp topicso e place to another. In the studied communities, there was a strong consensus among a The increase in the proportion of families whose educational level is above the 5th grade, was more than 20% in the So Francisco do Itacainas community, and more than 10% in the Cupu community. This improved level of education indicates that some family members return to live in the community after going away to study. Differences in levels of schooling may also indicate that new generations in colonist communities in the Marab region have more tools to interact with the outside world than did the previous generation. Cultural Interpretations g the development of this research, peasants discussed a very wide variety oressing their history, struggles, beliefs, challenges, and dreams. Some of these have been presented before. This section aims to deepen the discussion on peasants' knowledge through their cultural interpretations of some of the topics related tthe studied theme. It is necessary to emphasize, however, that culture is a dynamic process (Freire 1988:54), and the interpretations presented refer exclusively to a specific time period. This section will first discuss families' observations when comparing life in the communities (rural life) with life in the city (urban life). This topic is important because colonist peasants are seen by some outsiders as leading a nomadic life, moving from on

PAGE 110

97 interviewed families that living in rural areas was better than living in cities, and that they wanted to continue living there. Fam ilies' level of attachment to their land is important becau will st. search of forested areas, and most of them seeking their first piece of land. They bring with them their prevto contact with other cultures and w r local knowledge acquired with observations), as if it were nonsense. A wide rangey. se it can be an indicator of their interest in managing their land in a more sustainable way, and in taking into account the use of natural resources by future generations. Another topic discussed in this section is what the forest means to the peasants. There is a widespread pre-conception that colonist peasants do not maintain a deep relationship with the forest, mainly because they come from other regions, and soonbe moving to another place. Some argue that the forest is a new environment to them, a strange environment, and in order to produce they have no other option but to deforeThis study found that most families do come from other regions. However, many of them come in ious culture which comes in ith the environment, creating new knowledge. As colonists, families try to improvetheir knowledge to overcome the new environment. For instance, Mr. Pereira, from Cupu community, born in Maranho state, stated that they talk about these experiences oknowledge with others, "trying to congregate others' experience, share ours, try other person's experience. Each person has a kind of experience, a system, for our own survival." People in general are shy to talk about this topic to technicians, what they call experiences ( of empirical knowledge was registered, providing material for a whole new studThis section presents a brief review of this knowledge.

PAGE 111

98 The other topic addressed was the meaning of fire and its uses to the peasant families. Some researchers believe that families would change their fire practices if onlythe negative impacts caused by fire were explained to them. However, this is anextremely naive assumption, because families have lived with the local impact of from losses o fire, f goods to respiratory problems, and they unanimously affirm that if there were os) the munity infrastructure improvements. In early 2000, for the first time, families in the So Francisco do another way to work the land as productively as the slash-and-burn system, they would adopt it. Yet, fire is positively connected to their lives in many ways besidesproduction, and this study describes some of them. Communities and Towns For the majority of the families in the communities, this was their first piece of land, since most had previously lived in other people's land as sharecroppers (agregador employees. In the So Francisco Community and in the Cupu Community 77% and 70% of the families, respectively, were living in the first plot of land they had ever owned. All interviewed families in the So Francisco do Itacainas community declared that they plan to continue living in the community, whereas in the Cupu community, 83% of the interviewed families plan to continue living there. This difference can be explained by the advanced stage of land title recognition in the So Francisco do Itacainas community. While this community was officially recognized as a Settlement Project, the Cupu community was not. Recognition as a Settlement Project is merely first step required for families to receive land rights documents, but it makes it possible for families to receive some government support for com

PAGE 112

99 Itacai nas community received a government grant of US$ 1,10030 to be invested inhousing and agricultural tools. Leaders in this community were planning new roads the government promised to open. Families in the Cupu community had never receivsupport from the government, and some families mentioned they felt left aside and undervalued by the government. The lack of recognition by the government of the rightof the families living in Cupu seemed to be the main reason why some families consideleaving the community. When asked what would be necessary to impr ed any s r ove life in the community, nearly all mentioned a number of items and several said 'there are so many e of hunger in the country, followed by the possibility of living in a community, their f to, has that there's nothing to bother us. Especially at my age. We have very few bugs. I right? those interviewed promptly things that could be done I wouldn't know where to start'. However, when askedwhether they would prefer to reside in the city or in the countryside, 100% of the interviewees opted for the country, including those families who proclaimed they had plans of moving out of the Cupu community. The main reason stated was there is thabsence reedom, the better climate, the fewer mosquitoes and the low crime rate. Ms. Gomes, was born in 1946 in Maranho and has lived with her family in So Francisco do Itacainas since 1984. Blind since age 20, Gomes helps as a midwife when askedten children of her own and in the following words compares city and country: Question (Q): Do you ever think of living in town? Mrs. Gomes (Mrs G.): I never do. I like it here so much, the silence and the fact hope God lets me stay. Q: They say many people want to move from town into the forest, is that 30 In March 2000, the average conversion rate was US$ 1 = R$ 1,82 (Banco Central do Brasil 2004).

PAGE 113

100 Mrs G.: But there are also many who do not want to move to the country. They[which could not be harvested] and all, but they refuse to come. They're of a women begging for change in the city! Q.: Even women with babies, isn't it ? Mrs G.: Yes, 'spare me some change for my child' or 'some medicine', or 'buy this', have never begged for coins. I have always lived in the country, where no one Q.: Do you think th die in need and do not have the things we have, such as beans, broad beans, lost rice different nature, aren't they? People's natures are not alike. We can even see young 'buy that'. We see a lot of this but thank God during all my years in blindness I refuses to give away anything they have. It is very different in the city. ere's more barter in the countryside? While Mrs. Gomes talked about quality of life (silence, less mosquitoes, plenty of food) and dignity (never had to beg despite to her special condition), a similar explanation was given by Mr. Reis, an important leader in the So Francisco do Itacainas community, for whom living in the community also has to do with dignity: the male's ability to feed his family. Below is the statement made by Mr. Reis, who also lives in the So Francisco do Itacainas community and is a Catholic Church coordinator, a director in the Local Association, member of the Marab Rural Workers' Union and of the community Fire Group since July 2000: Mr. Reis: I always say it to many peoplea good place today is a place where we m one place to another, hunting for a nice spot. (...) If you have nothing and have many the city or somewhere else to ld stay there. r after its occupation, Catholic Church coordinator, Association Director, Fire Group member. Mrs G.: Sure, we raise a pig, kill it and share it. We hunt to eat, we share it and never sell a piece of it. can make our daily bread. So, we have to stay. We cannot always run fro children and you sell your little property to move to hunt for a better spot, chances are you won't even be able to make your bread anymore. You'd have to get help from others, pitching in to pay what is yours. Nowadays when someone has a family and a piece of land, and he is able to easily make his bread, he shou Mr. Reis, born in Maranho, living in So Francisco do Itacainas since one yea

PAGE 114

101 Another example of living in communities and thus having access to better living tions was presented by Mrs. Lima, 56. She was born in the state of Maranho in the So Francisco do Itacainas comm condiand lives unity with seven of her eight children. r for your kids, the country or the city? es in Cupu associate living in a cexpressed their concern with youths who prefer to live in cities, and with improvements rations. ed by Mr. Moraes, born in the state of Bahia and a resident in the community since 1989. Mr.: Moraes: (...) Going to the city will mean more suffering, so a person has to salary I was on made it impossible. I came back to the country expecting to make should provide help to those living in the country, since most things consumed in ey? Listen to this: in a short time very few of these young fellows will want to live in the country because things are not easy e eas, jobs outside the community are hard work g forests to pulling out weeds, from taking care of cattle to felling timber and processing Question: Which would you say is bette Ms. Lima: I wish they'd move to the city but they won't go without me. I'd like them to receive education so they would not have to work as hard as I did. But they say they won't go and I really don't know what's best, I just guess for those born to poor parents it might be better to stay in the country." As in the So Francisco do Itacainas community, famili ommunity with better access to food and to better living conditions. Some also that would have to be made in the communities in order to attract younger geneThe contrast between rural and urban life, as well as youth interests were expressQuestion: What if peasants go to the city? hold on tight. (...) I went to the city and I realized it would not work because the things better. (...) The city can't offer anyone jobs, and this is why the government the city come from the country don't th here, it is not for everyone. And these people born these days will only stay in thecountry if there's something there for them, otherwise they will certainly stay in thcity. I think the situation is getting worse and worse, people leave in search of education and forget the country, where most things we eat come from, don't they?A man works in a bank, or he can do anything, but his food comes from the country. For landless peasants living in rural ar and low salaries, on a day wage basis. Job activities on ranches range from slashin

PAGE 115

102 wood As many peasants have mentioned, these kinds of work are done almost on a slavery basis because is it so time-consuming and the working conditions are nearly inhuman. In addition, salaries are so low workers cannot save any money. Since they do not haey are ills familse someone 'showed up out of the blue' saying that it was his land after they had bought it, and they had to move to Par, arriving in the So Francisco community in 1984. At that time, his father and a brother worked in the gold mines during the dry season. They made some money and invested in their current lands. Today Mr. Reis is 34 years old and married to the daughter of another colonist. They have three kids; he was one of the coordinators with the Catholic Church. He was also a director at the local association (AGRAF), an active member of the Rural Workers' Union (STR), and a member of the Community Fire Group: political class awareness: they mention the peasants' union, mutual help, and critical analy ve any type of health benefit, they have no alternative source of income when th The difficult situation of the landless is described in the words of Mr. Reis. Hiy came from the state of Maranho; they lost their first piece of land in Par becau Mr. Reis: (...) We see lots of people working for day wages. If someone doesn't work you can tell he's really, really sick. Here in our community there's John, who kept having more kids and I told him: 'listen man, we have to work for ourselves!' If you have a plot of land, work for yourself. It is hard in the first year but it only gets better after that, even when you don't have any money, because you will have rice and beans, the corn, the cassava, you will have it all: you have what you need to eat. Now working for others is much harder. You're making someone else rich; you're working for him. As we can see from the statement made by Mr. Reis, many peasants show a ses of their situation. Selling their labor for little money seems awful to many ofthem and they prefer to continue living as peasants. Mr. Cruz, president of the local

PAGE 116

103 association, summarizes their awareness as "Here we have nothing, but at least we haour freedom." Leaders' concern about who buys land in the community is associated with their concern about the peasants' identity and with their struggle for better living conditions in their community. They have a clear idea of what a peasant is as opposed to an average farmer or a rancher. Once road access to a rural region improves, the tendency will be that more and more ranchers will buy lands, transforming the existing peasant communitinto a group of ranchers. From the peasants' point of view, this transformation would represent a big loss, since ranchers do not generally reside in the lands they own and not communicate with peasants. As a consequence, these ranchers do not get involved community organization for improving education, health, infrastructure in the rural areas, nor practices to protect their environment such as fire management. In general, they decide about land use without consulting or informing their neighbors, and see their landas a capital investment, es ve y do in s pecially for timber extraction and extensive cattle ranching. move livingthe raremaining forest (around 80% of the plot). A large burning such as this one is very difficult to control, and may cause a lot of damage to the surrounding ecosystems. This farmer probably would hire outside workers because everybody in the community is busy preparing for burning too. Outside workers usually do not get involved in community agreements for land uses. For example, in the So Francisco do Itacainas community, a peasant leader d to the city in 1999 because of health problems, and sold his land to a businessman in Marab. His neighbors were worried because they knew from the manager that ncher's plan was to change the entire plot (150 ha) into pasture after burning the

PAGE 117

104 The entire community expressed concerns regarding the sale of land to large farmers, who b uy so many plots of land that eventually the village may become a single large as long ind ed that one of the reasons for their migration to Amazonia was the presence of the forest, a resource that they have seen be depleted in farm. Peasants rely on the government's recognition of the area as a Settlement Area, in which the average plot size is 50 ha. On October of 2000, peasants decided consensually in a general meeting held by the Community Association (AGRAF), attended by members of the Marab Rural Workers Union (STR), that the Association would prepare an official declaration to INCRA concerning the commerce of land. In this document, the community would state that they were against the sale of plots to ranchers or large landholders, and declared their support for the policy that families that sell their plots cannot receive another piece of land from colonization projects during the following 10 years. In spite of this concern, in both communities studied, more than 70% of the families interviewed have resided in the community for over ten years. Forest There is a general misconception that colonist peasants only care about the forest t is useful to their "destructive" productive system (slash-and-burn agriculture apasture). As discussed in Chapter 1, many academics and technicians assume that because colonists have no "environmental tradition," no historical and cultural ties to the area, they are unable to have knowledge about natural resource management, or worry about resource preservation for their descendents. Colonists in the Marab region, however, in general have a keen understanding of the renewal limitations of natural resources. As presented before, in both communities, which were formed more than 15 years ago, an average of almost 50% of the primary forest was left. Some families claim

PAGE 118

105 their homelands, such as Bahia and Maranho. This awareness that the health of theforest can be strongly influenced by human actions and that the forest's regeneration cycle can be broken by humans, leads to colonists' openness to more sustainable techniques. They have observed and lived with the consequences of deforested areas,dried rivers, absence of game, uncontrolled fires. Therefore, these people are the bones to teach about the impacts of unsustainable management of natural resources on a local level, and about the high value of forested areas. The association of Amazonia withe forest was identified in statements by many families. One example is presented bMrs. Souza, 36 years old, from the So Francisco do Itacainas community. She was born in Bahia, in northeastern Brazil, and has been living in the region since 1976: Mrs. Souza: () My first memories are from that farm. My father was a cowboy, 'small' [non valuable] fragments of woods. When it was time to plant ric est th y manager for my uncle. There, you would look and see no green forest, just those e, we did it in a meadow, like those over there with the aa trees. But in Bahia, aa palm isn't ed, nobody eats it. We learned to eat aa after we got here. () When I arrived at Serra Norte [Par state], in 1976, everything was beautiful: a lavish green forest. only green trees we saw were the aa trees, because that's where the cattle drank. I water in the places where there was only pasture, without trees. Question: And when someone talks about "caring for nature," what comes into your destroy it. Because God left it like that, to be left alone, and people come and mess s a work of nature. Do you know why? There are crops people plant which are attacked by a curicas don't come. The places that attract the birds are the ones where there was a you don't work with that kind of thing. In the place where there's a periquiteira y red berries which they eat. In the following year they [the birds] come and they forest is pretty. Back there where we used to live there was no forest, it was ugly, us And in Bahia I didn't see that kind of thing. That's why I think it's pretty. () The always say the aa trees are the ones holding the water there, because there was nomind? Mrs. Souza: I think it's beautiful to leave a forest like that, all green, and not it up. The aa trees harbor the toucans and several kinds of birds. It' plague of curica [a species of bird]. And there are other people who plant and the tree with fruit for them to feed on. Just study that, pay attention to it. It's because [parakeet tree] where they eat every yearit's a tree like the pitanga tree, with tindon't find the tree, so they attack the crop because they have to feed. () Thisnobody could see the forest, it was just those slopes fading away

PAGE 119

106 As expressed by Mrs. Souza, God is constantly associated with Nature: if an evehappens, it was due to God's will. Humans possess their physical abilities and knowleto manage nature, but in the last instance,"only God knows." Fire acts mediate naand culture. While in forested areas some products can be extracted, who 'manages' is nature itself. In agricultural fields, the human, below God, is the manager; the higher tproductivity of a field, the better peasants they are, blessed by God and socially admiredby others. Fire, therefore, is part of the peasant culture that makes it possible to transform nature into agriculture, to feed God's sons and daughters through their work. This explains peasants' exasperation when the government requires intangible rules for fire use or prohibits its use, without assuring if there are alternatives to fire use. Thi nt dge ture he s topic actionlives,gamerelated the primary forest to beauty and to the coolness of the climate, as Mrs. Souza expressed in her speech. Mr. Lopes, 51 years old, is from So Francisco do Itacainas and w Groupr. sed to rain a lot will be developed more in Chapter 4, on peasants' evaluation of government s. When peasants were questioned about the uses and meaning of the forest in their answers included as a reserve for future agriculture field; water cycle maintenance; ; wood for construction; tool handles; and medicinal plants. Many peasants also as born in Maranho. He is a member of the local association and of the Fire, and also described the connection between the forest and the availability of wateQuestion (Q.): What is the forest good for? Mr. Lopes (Mr. L.): The cold [primary] forest? Holy Mary, it's very good! Q.: And why is it good? Mr. L.: Why is it good? Because if the forest isn't there everything goes dry. Without the forest everything turns into dry grasslands, the water dies out, everything is dry, it even attracts the summer [dry season]. () It u

PAGE 120

107 here in Par because of the forests. Now that everything is bare [deforested], they are protected by the forest, they remain really beautiful (). The concern expressed by Mr. Lopes is that Par state is getting drier, the dry season becoming longer and more severe than before. This verification had made peasants change their previous knowledge that primary forests do not burn, recognizing that they no longer functioned as a natural firebreak, and therefore prevention of uncontrolled fires also had to include protecting forests, as discussed in the followsection. Fire Peasants were asked about the importance of fire in their lives. Because the valuand usefulness of fire is so obvious to them, and also because they feared not givin"right" an everything is lacking. And the water [sources], the dells, everything dries out. If ing e g the swer, many were reluctant to talk. The question was then asked differently, somed familluded that fire directly affects peasants' qualitk, as obserts' accounts. what like this: "if you were explaining what fire means to you, to someone who hanever visited a community before, how would you explain it?" Observation and notes taken during the unstructured interviews were also considered in this discussion. Fire was basically explained by the interviewed peasants as a special 'tool' that directly and indirectly makes it possible for them to produce most of their products. Peasants' production is consumed by the extended family31, and also sold. Since agriculture and cattle raising were shown to be the main sources of income in most ies in the communities studied, it can be conc y of life and dignity of being able to maintain their family through their wor ved in many of the peasan nded family includes the family nucleus (mother, father, daughters and sons) and other famrs, as well as godmothers and godfathers. 31 Exteily membe

PAGE 121

108 In the words of Mr. Duarte, 49, born in the state of Gois and living in the Soisco do Itacainas community since 1983: Francle. It is important to use fire in many places. I don't know how to explain it, I only know to use fire on all of those. We use fire on a pan [laughs], on [manioc] flour under lain [pause]. Now, about fire, the worst is when it comes in our direction. When h as this house comes, when there's some wind and this flame bends in our direction, t's when it's tough, we all risk getting roasted. The conflictive relationship with fire presented above was found in many other statem good if we could live without it. Older son: But [fire] makes it easier for us. When we burn, less weeds come out, line of land you have to take out all the weeds to work, kill yourself working, throwing You have to take out all those weeds and sow the raw land, without plowing... Mrs. Almeida: But if we could have a plow, to plow the land, we wouldn't be machinery is the most common alternative to fire presented by families, probably because machinery is regularly used by medium uently shown in TV programs. Almost none of Question: Why is fire important? What do you use it for? Mr. Duarte: I think fire is important because we use it in burnings, for exampwe use it in many ways, on tree trunks, on pastures, on crop fields. It is important the oven, to burn a brick, a tile, we use it for everything. That's all I can exppeople come and say "let's control" the fire, then it's tough. When a flame as higtha ents. If, on the one hand fire is their main agricultural tool, on the other it is also very dangerous and hard to control. The following quote of the Almeida family, who were born in Maranho and had lived in the Cupu community since 1992, suggests theuse of machinery as a possible substitute to fire use. Question: What is the importance of fire to you? Mr. Almeida: Fire is very good, but it devours a lot, ends lives. It would be the land is cleaner and better to work on. Without fire, if you're going to crop athose weeds out. Then you have to carry the stumps. Fire burns the stumpsdealing with fire. As did many other peasants, the Almeidas' account contrasted the use of fire withthe use of machinery. Working the land with and large-scale farmers in Brazil, and freq

PAGE 122

109 the fa milies interviewed have ever tried any substitute to fire in their productive systeOnly two of the interviewed families have produced the ms. ir main crops using the slash-and-mulchwo families, one learned the tein the are acproduity of upcoming wet and dry seaso dry enough for most leaves and branc the period of time elapsed betwewhethbeforfive wregrohe vegetation is burned long before it rains, planted seeds will take longer to sprout, birds and other animals have more time to eat the seeds, and weeds will take up the area before the planted crops have a chance to grow. In addition, if slashed vegetation is left to be burned later in the dry season, and if there is more than one strong rain event, the system and fertilizing with leguminous plants. Of those t chnique from FATA workshops; the second learned from the first family (both live So Francisco do Itacainas community). Although peasants in the Marab regionquainted with the slash-and-mulch system, it is used only in small areas for bean ction. Cultural weather forecasts on the length and sever ns are very important in families' planning of their productive activities. In the Marab region, as in Amazonia as a whole, only two seasons are considered: the dry season (known as summer) and the rainy season (known as winter). The end of one implies the beginning of the other. The use of fire is also associated with the rainy season. Slashed vegetation must be burned when it is hes to burn, and the humidity level will depend on en the moment the vegetation is cut and the moment it is burned, as well as on er the slashed vegetation is exposed to any rain. If the vegetation is slashed long e it rains (on average more than three weeks for secondary forests, and more than eeks for primary forests), many green leaves will sprout due to vegetation wth, and consequently the vegetation will not burn well. On the other hand, if t

PAGE 123

110 vegetation may not burn well anymore. All these situations present risks for peasants, as the quality of the burning is strongly correlated to crop production. Because fire and rain have such a strong influence on their productive systems, peasants have developed several cultural weather forecasting methods, based on which they define their productive calendar. Some of these weather forecasting methods include observing stars and the behavior of certain animals, as if they were 'reading nature'. The following piece presents examples of 'nature reading' using the stars: er. u? is re, another one there. The more spots covering it, the stronger the winter is. Mr. Oliveira, born in Maranho, living in the Cupu community since 1985. Another example of weather forecasting based on reading stars was mentioned by a peasant in the So Francisco do Itacainas community. Mr. Cavalcante, who was born in Cear, predicts how dry the dry season will be by observing the Milky Way too. However, his observation is done during Holy Week. The signs of rain are the same observed by Mr. Oliveira, born in the state of Maranho and living in another community: dark spots on the Milky Way indicate rain, and the more often it happens, the rainier will be the dry season, and vice versa. Mr. Oliveira (Mr. O.): [Talking about production] You have to look at the weathQuestion (Q.): If you know how to, right? Mr. O.: It's very simple. You know the Saint Thiago way [Milky way], don't yoQ.: Yes, I do. Mr. O.: Well, it is usually bright. If the night is dark, it is even brighter, right?And when it is dark and spots can be seen, it is showing next month's rains. Sometimes, if you are looking at the beginning of the month, the rains come in thsame month. Q.: Now it is not very bright; it has some spots... Mr. O.: It has some spots on, not all over it but some spots in the middle, one he

PAGE 124

111 Peasants in the Marab region have special folklore involving the use of fire in their productive system, for choosing the burning day. Moon phases interfered with the choic it n, claimquarteill grow. numbilies declare their preferences for days that end in 7. Some of the familoo. OtherNascid as a peasant. Mrs. Nascimento was bfor 11 years in this community, where she met her husband. Mr. Nascimento explained his numerological preferences, stating "I try to burn on Aug 27, or Sep 7th, or Sep 17th; those days are foggy in the morning, and from noon on through the afternoon it is hot and windy." Mr. Fonteles, born in Maranho and living in the So Francisco do Itacainas, also chooses the days ending in '7', and explained that "if in those days it continues to be cold and foggy, and does not rain, day 9 will be a great day to burn". Besides using moon phases and numerology, there was a general tradition among many families, that when the summer [dry season] is strong, you can choose the day; but if it is too rainy, the e of burning date. Mr. Lopes, from So Francisco do Itacainas, born in Cear state, relies on moon phases to chose the burning day: "I burn the field when it is a full moon to a waning moon: trunks and branches are dryer, and the field will have lessweeds." He said that "some people make fun of me, but I don't care, because I knowworks." Other families from Maranho state also prefer to burn during the full moo ing that the field will have less weeds. They said that burning during the firstr moon is not good, because many weeds w Other folklore about the burning day involves choosing dates that end in the er 7. Many fam ies say that "the older people" have this belief, without affirming that they use it ts admit it, such as Nascimento's family, living in So Francisco do Itacainas. Mr. mento was born in Marab city, but always worke orn in Tocantins state. Her family migrated to Par, and her family has been living

PAGE 125

112 most important factor is the vegetation dryness, considering how many days without rand the possibility of more rain or not. Mr. Oliveira, from the Cupu community, talked about observing animals to predthe weather. He and his family live by the lake of the Tucuru Dam, whose water level presents seasonal variations throughout the year. He observes the behavior of ants aalligators to foresee what the next rainy season will be like, as in his own words, below:Mr. Oliveira: Listen, Katia, not calling us animals...Are they rational or are we? Question: It is said that we are. Mr. O.: We are the rational ones? And they are irratio ain, ict nd nal? Well, not comparing us to irrational animals, I'll tell you something: irrational animals have more e intelligent, but we have no knowledge of what happens, only God does. And m high areas to lower areas, they're announcing the start of summer [dry season]. expect rain on that day or on the following one, it's certain. Here we have an alligator wants to get wet'. But if we are here listening to the alligator [makes a alligator croaks and there is no rain. Sometimes we have the pretense of telling attention, others don't. Two families in the So Francisco do Itacainas community described a ritual to coordinated by a pioneer resident, Mr. Fontes. Although Mr. Fontes was a member of to take something special from someone, and hide it until the "robbed" person gets so one of three rocks employed for cooking can be used; the idea is to take something that comprehension than we do. Did you know that? On the one side we are morsometimes you see it in a little animal. When you see a bunch of ants moving froWhen you see a bunch of ants moving all their young ones to higher areas, you can understanding, we who live by the lake, that others make fun of us. I say 'today thesound like an alligator croaking], then it's rain for sure. Rare is the day when the what is going to happen, because an animal is showing us. Now, some pay "call rain," referred by them as a brincadeira or "play." They said it was generally one of the studied families, he had not mentioned the ritual when interviewed earlier andthe description presented was given by other participants in the ritual. The main idea is angry that she or he will beg for God's help. An image of a saint, for instance, or even

PAGE 126

113 the "robbed" person will really miss. Generally, Mr. Fontes is the one who coordinates the play and chooses the person to be 'robbe d' and the object to be taken. All other famil peopl then sritual was not performed in 2000, probably because it was the humaknowshowand thas to be al education. This Chapter described the migratory and land use histories of families in the two study communities, as well as their views on education, rural living, forests and fire. The families in the communities studied have a history and current conditions of active effort to improve their living conditions, marked by strong partnerships with the Catholic Church and grass-roots institutions, and weak governmental support. God is frequently related to nature, and fire plays an important role in colonist peasants' culture by turning possible nature transformation into agriculture, to feed God's sons and ies in the community know that it is part of the play, except the "robbed" person. After some days, when the "robbed" person is very angry, begging for God's help, and cursing her/his bad luck, the whole community goes to her/his house in a procession, chanting, carrying food, and carrying the "stolen" object. The 'robbed' person's and other e's happiness, as well as the party, would make God happy as well, and He would end the much needed rain. The not a severe dry season. The previous speech again refers to God as the one who has the knowledge, andns those with intelligence. Humans must use their intelligence to observe God's ledge manifested in animals, or stars, to better manage their land. The ritual ed how peasants, through their faith, actively attempt to improve their production, erefore social reproduction. Those peasants' knowledge, culturally constructed, hrespected as an expression of their society, independent of their level of form Discussion

PAGE 127

114 daughters through their work. This explains peasants' exasperation when the government imposes intangible rules for fire use or prohibits its use, without assuring if there arealternatives to fire use. The Chapter also demonstrates that, despite their characteristics of migrants, and previously landless people, families in these colonist communities can build on their strong history of collective and individual struggle for land and living conditions upon which to build fire management efforts. They possess a significant commitment to "place" and to long-term management of forests and fires, that provide an important basis for fire management programs. The following chapter will explore, in detail, how this uniqu e history and set of values and beliefs is manifested in the communities' forms of organization, communication, and compliance with fire management recommendationsdeveloped in the participatory Fire Action.

PAGE 128

CHAPTER 4 bers of the two communities, in order to address the research questions presented in Chapter 1: Question 1: Why aren't government fire actions and laws more effective in diminishing uncontrolled fires? Question 2: Do colonist peasants have and use empirical knowledge to manage fire? Question 3: Is a participatory approach to fire management effective in diminishing fire losses? The framework used to address these questions was through in-depth interviews focused on each of the recommendations developed through the Fire Action for practices to be observed before, during, and after burnings. The interviews collected empirical information on practices used, as well as colonist statements about their motivations, and evaluations of their experiences. More details on the methods used are given in Chapter 1. COMMUNITY FIRE MANAGEMENT EVALUATION Introduction This chapter aims to analyze how families managed fire in their productive systems, and the potential impacts of the FATA/LASAT Fire Action in the two communities. As explained in Chapter 2, and presented in detail in Appendix C, the FATA/LASAT Fire Action's recommendations were divided into four phases, accordingto peasants' usual slash-and-burn system steps: before burning, on the burning day, afterburning, and controlling accidental fires. For this research, systematic data were collected from a purposeful sample of mem 115

PAGE 129

116 The results, presented in the next section, showed clear decreases in loss from fire in both communities from 1998-1999. The dynamics of internal community organization for fire management, the differences between the two communities, and their implications forr presents descriptions of changes in sp, divided into the three phases of beforr of t of the r the tion. Given their historical disttams, peate Fire Action. Because of their mixed feelings about some of the results of their experiethe long-term prospects for effective fire management will dy ld be better rule fire use, therefore avoiding losses caused by uncontrolled fires. fire management are discussed below. The majority of the chapte ecific practices e, during, and after burning. The discussion shows how a combination of feaIBAMA sanctions, families' desired to control fire-related losses, and the supporFire Action led to a variety of positive changes in fire management practices. Anothesection of the Chapter conveys the complexity of putting these practices together in concrete fire strategies for specific burning experiences that require adaptation to varied conditions. The Chapter concludes with a review of peasants' interview responses related togovernment fire management programs and the Fire Ac rus of government agencies, and the inadequacy of many features of their progr sans were found to be more comfortable with the grass-roots learning approach of th nces with new practices, epend on the evolution of community and government actions in the future. Results showed some similarities between communities concerning fire management. In both communities, families were aware of fire restrictions enforced bthe government, and requested more information to support the continuity of their productive systems. Moreover, families claimed that community agreements wounecessary to

PAGE 130

117 The m ain differences between communities concerned families' strategies for communityactions, such as whether or not a Fire Group was formed, and whether the discussion of fire agreements was held in community assemblies or exclusively among neighfamily. Those differences had a key influence on the organization of the communitythe family with respect to fire use, as the discussion below will show, although in bothcommunities the losses caused by fire dropped after the Fire Action. Losses Caused by Fire How can we live in the country and not be allowed to crop? We will control the 1984. Fire m bors and and fire in order to cause no losses to each other. Mr. Ferreira, 62 yrs old, born in Maranho, living in the Cupu community since akes agriculture possible for peasants in Amazonia, and they do not perceive fire ue is valuable is for sa difference goods. se per se as a problem. The problem understood by them is associated to escapee fires, due to losses that an uncontrolled fire can cause. Defining losses caused by firnot a simple issue. It means expressing the value of something negatively affected by fire.The complexity lies in the variations of the social construction of 'value', which varies within and among peasant families. What is valuable for one person may not befor another. For example, the burning of an area of aa palm (Euterpe oleracea Mart.)a loss of value for those who extract its fruits or palm heart, for their own consumption or le in a market. For those who do not use it, its burning is not a loss. Thisin conceptualizing value is even stronger between urban outsiders and peasants. For an urban outsider, such as a researcher, there is a tendency to relate loss with material Other cultural values are less visible. In order to capture a better picture of those losses,the term 'loss' was not used directly during the interviews, unless the interviewees chose

PAGE 131

118 to do it. Instead, I talked directly about occurrences of any uncontrolled fire in their lands, and their views of these fires' consequences. For the purpose of Fire Action evaluation, this study focused on change in measurable losses throughout the years that the families had been living in the communities, up to the moment of the research. Ten different measurable kinds of losscaused by fire were considered (Table 4-1), with each community pre es senting eight of each. The number of losses wasf families because a family couldIn the So Francisco community (Table 4-1A), all of the households talked about fire lof the 40 families studied, 95% (38 families) had had losses caused by uncone bigger than the number o suffer more than one kind of loss in the same year. sses. O trolled fires. Seventeen of those families (42.5%) had problems in two different years, and one family (2.5%) had problems in three different years. Primary forest represented more than 44% of the total reported losses. It was followed by pasture (23% of the losses), perennial crops (14% of the losses), crop fields (6% of the losses), and secondary forest (5% of the losses). The years 1996 and 1997 were those in which morfamilies suffered losses caused by fire, 18 and 21 families respectively. In 1998, when the Fire Action started, the number of families experiencing losses dropped to five, and continued low in 1999. Another change observed after the Fire Action started was that uncontrolled fires in forest, before the most common cause of loss, had the lowest rate in four years, and no longer were the most common type of loss. This result suggests that Fire Action's emphasis on working collectively to protect forest had positive results. Peasants also reported that large farms located near by the community diminished their

PAGE 132

119 fire use, which contributed to less outside-originated fires in the forests located by thecommunity's border. Table 4-1. Distribution of kinds of losses caused by uncontrolled fires by year in A) So So Francisco Francisco do Itacainas community and B) Cupu community. Community (A) Years Primary Forest1 1313142235 Secondary Forest 12 14 Perennial Crops 1 322 311 Agricultural Fields 1 2 2 Kinds of Losses 89 94 95 96 97 98 99 Total 2 5 Aa Palm Area 11 2 Pastu re 2752218 Fence 0 Barn 1 12 no. of Families 11518215556 Kinds of Losses 89 94 95 96 97 98 99 Total Primary Forest 1 135111 1 Secondary Forest 2114 Perennial Crops 121 4 Crop Fields2 3 3 Aa Palm Area 11 Cattle 1 1 House 0 no. Total of losses 01061217339 and includes timber-exploited forest and pre-burned forest. 2 'Agricultural Fields' refers to slashed crop field that was burned before the planned Cattle 0 House 1 12 no. Total of losses 218282461079 Cupu Community (B) Years Pasture 329 14 Fence 1 1 Barn 0 no. of Families 0104814229 1 'Primary Forest' refers to all kinds of forest that had never been clear-cut, time, and one case of an experimental annual crop field with leguminous plants in a 4-year-old rotation system.

PAGE 133

120 In the Cupu community, as shown in Table 4-1B, men and women of 29 familiestalked about their loss es. Among those, 79% (23 families) had had negative onsequences with uncontrolled fires. Six of these families (21%) had problems in two different ye area as averagof 41% of the land, as shown at Chapter 2), pasture is the kind rted, represeing6%f totosst isllowed by the loss of 8% of the loes),ecoaryres10%of lossed perennial ses). Inis cmmity, the Fire Action started in 1999, and this ented the lowest numer o famiesat suffered losses in fohe distribution of the nmb fmilis that repted y kind osses caused ontrolled fire per year theo cmmnitie stued number of filies that sufdsesuse by uncontrolled fire assoginno FAS Fn i positive ction. Thiresu is vry iportant for the evaluation of Fire Action in 1998 thesetwoommwere located in the samegion, subject er condition, aninflunce. Community Organiztionor Brnigs e Actionas bed comtle of the Fire Action booklet: "Community Fire Management: Stronger Bonds, Fewer s Prejuo). This union sarvalheiro and c ears. In this community, where pasture occupies almost the sam primary forest (an e of loss most repo nt 3 o he t al l I fo primary forest (2 ss s nd fo t ( the s), a n crops (10% of th e los th o un year pres b f il th ur y ears. T u er of a e or an of l by unc in tw o u s di is presented in Figure 4-1. The reduction in am fere los ca d ciated with the be ing f the AT /LA AT ire A ctio ndicate s a influence of Fire A s lt e m impacts becau se c unities e r to the same weath s d e d by IBAMA and other initiatives to control fire use a f u n The whole Fir w as on m unity organization, as observed in the ti Losses". (Manejo Comunitrio de Queimadas: Mais Uni, Meo o n z hould include families living in communities, peasants, and technicians C Aquino 1999:1):

PAGE 134

121 We believe that only with the evenly balanced association of technical knowledge readings of nature), will we together reach concrete and applicable solu (theories, research) and practical knowledge (accumulated across generations, tions. Figure 4-1. Distribution of families that reported any kind of losses caused by Cupu communities. Note the decrease in number of losses after the 1999 for Cupu community). As uncontrolled fire, in numbers per year, in So Francisco do Itacainas and FATA/LASAT Fire Action started (1998 for So Francisco community and well as pointing out the intrinsic power of union, the Fire Action also emphasized that isolatedature of fire propas, frastructure, and external support to fight fires. firefighting is inefficient, because of the n gation. Once fire gets out of control, it spreads throughout surrounding ecosystemregardless of families' willingness to work together or not. The Fire Action also stressedthat preventing uncontrolled fires is more effective than fighting fire after it gets out of control. This last statement comes closer to peasants' reality, due to their scarcity of adequate tools, in

PAGE 135

122 So F rancisco do Itacainas Community Representatives of the So Francisco do Itacainas community have been invo lved in the Fire Acollectivs to theirecomms shared their experg families working wn to reporting, they studied the IBAMA booklet and to 1999, h), and ct community leaders. ction since it began. Figure 4-2 (A-I) shows the FATA/LASAT Fire Action and So Francisco do Itacainas community's activities from 1998 through 2000. Until July 1998, the Agro-forestry Group was composed of 10 families working ely in a local nursery located at the village center. They worked as volunteers on a weekly basis, producing perennial plants. In July 1998, two representatives of FATA'Agro-forestry Project Group attended the first FATA/LASAT Fire Workshop, and discussed the proposal for community fire management. These representatives returned r communities and discussed the topic of fire and the FATA/LASAT's endations within the Agro-forestry group. The group then decided to work as a Fire Group too, helping all interested families. The same representatives of the Agro-forestry group led the creation of a local association, the AGRAF. At the second Fire Action workshop, held in Marab in December 1998, community representativeience in community fire management. They reported no losses amonith the Fire Group. In additio ok part in the development of the Fire Action's recommendations. In Maythe first set of Sequential Fire Evaluation activities was carried out (field researcthe community received a copy of the complete FATA/LASAT Fire Action report. The report was received by peasants in the nine communities taking part in the second workshop, and they worked on the development of the recommendations. The act of providing a copy of the report to the communities proved to have a very positive impaon the process of building trust among the FATA/LASAT and the

PAGE 136

123 Figure 4-2. The FATA/LASAT Fire Action and So Francisco do Itacainas community were shared among participants, and plans for the dry season were outlined. The greatest activities, from 1998 rainy season through 2000 dry season. Representatives of the So Francisco do Itacainas Community, many of whom are members of the Fire Group, attended a third workshop, held in Marab in July 1999. TheFire Action booklet was presented and 750 copies started to be distributed; experiences

PAGE 137

124 achievem ent of this workshop was the presence of an IBAMA representative, with whomants were able, for the first time, to directly discuss their reality, their main challenges and their most significant needs. Upon returning to the community, the AGRAF association held an assembly, where the community defined rules for fire ment, distributed Fire Action booklets, and decided to divide the Fire Group into five Fire Groups. Each group would have a coordinator, who would meet regularly with coordinators. The community was successful in fire prevention and control, and presented low rates of fire-related losses. The fourth workshop was held during the rainy season, and no representatives from So Francisco or Cupu communities attended. After the workshop, the FATA/LASAT Fire Action was discontinued because of the lack of funding. In May 2000, the second field research for sequential fire management took place (field research), and in October of the same year, the field research involving thobservation of fire strategies was carried out. In October, the Fire Groups were still working, even with no support from FATA or LASAT, and no losses were reported by Figure 4-3 describes the relationship of the So Francisco do Itacainas commwith the Fire Action, focusing specifically on community organization and communication for fire management. The So Francisco do Itacainas community the peasmanagee them. unity organization for fire management can be broken up into four phases. 'Phase 1' was the phase befoestry group in the co re the Fire Action started. At this time, there was an Agro-for mmunity, supported by FATA, and limited to the 10 participating families, as indicated by the unidirectional arrows in the chart (Figure 4-3A). Although the group was open to community members, participants were required to follow strict rules, and

PAGE 138

125 the production was divided only among those who worked in the group. An example of such strict rules is the requirement regarding attendance at weekly activities: if the member cannot attend, he/she should send someone to replace him/her at work. Absences are only excused in cases of health problems or other really exceptional occasions. If a member does not attend, his/her absence would imply a reduction in the number of seedlings received. Thus, although the group was open to community members, its location and internal rules limited the participation of families. Furthermore, technicians give their support exclusively to those families engaged in the project, and according to a FATA requirement, the seedlings cannot be sold because both seeds and basic materials are received as donations. Those restrictions limited the impacts of the Agro-forestry project on the community. Before the Fire Action startedaround 50% of the families studied had suffered some kind of loss caused by uncontrolled fires during previous years (46% of families in1996, and 54% in 1997). The Fire Action, however, had a different dynamic from that of the Agro-forest project Although the same members of the Agro-forestry Group composed the Fire Groupeturning so, its proposal was to help burnings carried out by any family in the community, regardless of their social commitments (Figure 4-3, Phase 2). Shortly after rfrom the FATA/LASAT workshop, they decided to work as a Fire group, under the called collective system (mutiro), helping not only each other, but also any family asking for help. This is indicated in Figure 4-3 by the bi-directional arrows (Figure 4-3B). The strong interest demonstrated by other families was in part explained by IBAMA's coercive fire laws. IBAMA began by requiring burning permits in the 1998

PAGE 139

126 dry season, and prohibited burnings during the high dry season. Families felt lost and were not sure of how they should burn their fields. ts created Fire Groups, with rules for fire use. Figure 4-3. Representation of the So Francisco do Itacainas community's strategy in conducing fire management, from 1998 to 2000. In this community, peasan

PAGE 140

127 In September 1998, the group called a community assembly, whose main topics were fire management and the founding of the local association. Although many people were interested in fire management, only the families from the Agro-forestry Group mfim committed themselves to the work, and to taking part in the Fire Group and in the new association, the AGRAF32 (Agro-Environmental Association for Family Agriculture Activities of the So Francisco Community). The Fire Action coordinator facilitated a eeting to discuss fire management in the community with all those who were interested, providing families with more information and reinforcing the Fire Group's importance. The Fire Group's main points were: any interested family should inform the group about the burning day in order to receive the Fire Group's help in the burning (required always burning in groups); families should make firebreaks; avoid burning in the beginning of the dry season; observe the burning; and use control practices in case of uncontrolled res. The greatest benefit for the families was that the AGRAF, STR and FATA would support them in any fire conflict between a family and IBAMA if the burning had been carried out according to the Fire Group's guidance. Thus, the fear of IBAMA sanctions provided a strong motivation for peasants to change their burning practices, but little accessible and appropriate information or resources to support these changes. The Fire Action empowered them to develop and implement their own rules and practices for fire anagement, tailored to their own culture and resources, and use these to support their actions in any potential conflict with IBAMA. At the end of the season, the numbers of ly, the Group was in the process of founding their local 32 Concurrent association, for which purpose FATA lent the necessary money, and the families paid it back working collectively at the FATAs Center (10 days of 10 men, weeding).

PAGE 141

128 losses caused by fire dropped to 13% among the interviewed families, and none of the families working in the Fire Group suffered fire-related losses. In December 1998, the Fire Group's representatives presented the results of theactions at the 2nd Regional Fire Workshop, where participants analyzed IBAMA's fire booklet, exchanged experiences with other peasants, and took part in the developmenthe first Marab regional community fire management proposal. After the burning season was over, families worked as usual in the preparation of their crops, while the members of the Fire Group also continued their activities as directors of the Agro-forestry group and the AGRAF. During this time, the number of members of the AGRAF grew to reach most of the families in the So Francisco do Itacainas community. In June 1999, the first set of interviews for this study took place. In July of that same year, community representatives attended the 3rd Fire Workshop, in which they received the previously approved regional recommendations organized in the form of a booklet. They also discussed plans for the upcoming dry season and talked to an IBAMArepresentative for the first time without intermediation, asking questions and listeningexplanations (Figure 4-3, Phase 3). Members of the So Francisco do Itacainas Fire Group brought a new proposal to the workshop: working in several groups instead of in a single fire group, according to geographic distribution and to the number of families in the area. Each group should choose a coordinator, and all coordinators would meet ir t of to regularly to discuss, to plan, and to share information. Fire Group members had evaluated their activities in the past dry season, and concluded that having a single group working with all the families in the community represented too much work for the

PAGE 142

129 group's members, since they also have their family activities to perform. A communityassembly coordinated by the AGRAF's Directors was held in th e four days after the worksn t the Fire ting ifference was that each coordinator would be responsible for his/her region, but nor ond n community. hop. The minutes of the assembly specify that the FATA/LASAT booklet was read, and that IBAMA's laws were discussed. The minutes also recorded discussions othe fact that burning in groups was an IBAMA requirement. They emphasized that it wasnot fair that the Fire Group had only a few members, and stressed that "union is the foundation;" this union meaning that all the community should be actively involved. Athe end of the assembly, eight Fire Group coordinators were appointed according tocommunity's geographical regions, and Fire Action booklets were distributed. TheGroups' general strategies were similar to those of 1998, with coordinators meeweekly; the d t limited to it. This responsibility included raising families' fire awareness by means of door-to-door work. The number of losses caused by fire was low again. In 2000, AGRAF called another assembly at the beginning of the dry season, and families agreed to continue working as regional fire groups, even after the Fire Action had ended. Rules regarding fire management were stressed again. No losses were registered by October of that year (Figure 4-3, Phase 4). Cupu Community The Cupu community engaged in the FATA/LASAT Action in 1999, one year latethan did the So Francisco Itacainas community. Methodologically, choosing a seccommunity, located in the same region that did not participate since the Fire Actiostarted, was designed to compare the two communities in different stages of the Action's development. Figure 4-4 shows related activities of the Fire Action and the Cupu

PAGE 143

130 In terms of formal social organization before the FATA/LASAT Action started (Figure 4-4A), the Cupu community had had a FATA Agro-forest group in 1995, whnursery was located on the land of group member Mr. Souza, near the place where thefirst Village Center was implemented. The Agro-forestry group, formed by members of 10 families, ended about one year later. Mr. Souza Jr. (son to Mr. Souza) became a Director at the COOCAT and organized courses, such as the course on aa management, as an attempt to draw the community to the Coop. The LASAT Community Forest Project started in 1997 and involved representatives of 10 families. By 2000, this group no longer held regular meetings in the community but met for the LASAT technician's visits. There was also a group of 9 or 10 families which called itself the Cattle Association. In addition to the organizations described above, there was the Regional Association, created by reason of a bank loan and whose members occasionally met iorder to solve specific issues related to the loan. Almost half of the families intervie ose n wed repor, egarding governmental laws on July 1999, the Cupu community was invited to the 3rd FATA ted losses caused by uncontrolled fires. The first set of Sequential Fire Evaluation activities were carried out in June 1999aiming to evaluate the previous dry season, to collect data on the social-productive conditions of the families, and to understand their views r fire management. In /LASAT Fire workshop, held in Marab. Three families attended, including Mr. Souza Jr. representing the COOCAT (Figure 4-4B). They received the Fire Action booklet, discussed fire proposals, exchanged experiences, and planned to support thecreation of the Fire Group in their community.

PAGE 144

131 After the regional meeting, community representatives discussed the fire proposwith the other members of the community. Because the leaders involved with LASAT and COOCAT were Catholics, they were used to having weekly meetings in church. Tlocal Protestant Pastor was interviewed, and he expressly said that fire management was not discussed during their meetings. The Protestant church was established in the previous year, and its members (10% of the male heads of families interviewed, and 38% of the female heads of families) had been Catholic in the past. Due to families' historelationships, this study assumes that differences concerning religions was not a barrier. Booklets were handed out to most of the families. In the community, the Fire Action coordinator facilitated a meeting in which she emphasized the importance of bugroups, discussed the laws by IBAMA, and distributed more booklets. There was no official formation of "Fire Groups", but they organized burnings with their relatives andneighbors. Losses caused by fire dropped to less than 10% in 1999, for the families studied. al he rical rning in ing. ial Fire to r's limited time in the field. Therefore, In the 2000 dry season, the FATA/LASAT Fire Action ended as a result of financial limitations. Its last activities were the 4th workshop and a community meetThe workshop was held during the rainy season, and very few peasants (from communities located closer to the city) attended. The second set of SequentEvaluation activities was carried out in June 2000. Families were planning to continue with fire management among kin and neighbors (Figure 4-4C). This study intended observe fire management during this season, but this was made impossible due to the unusually long rainy season and to the researche

PAGE 145

132 fire m anagement data for the Cupu community solely considered their plans, not their actions. 1998 through 2000 dry season. period before the Fire Action started. As described in Figure 4-4, there had been a Figure 4-4. Related FATA/LASAT Fire Action and the Cupu community activities, fromFigure 4-5 relates the Fire Action and the Cupu community, focusing on internal community organization for fire management. The Cupu community's organization regarding the Fire Action can be divided into three phases. Phase 1 corresponds to the

PAGE 146

133 previous experience in community organization with the FATA Agro-forestry project; regional a association devoted especially to a governmental loan administration; a COOCAT Director; a Cattle Group for meat commercialization among members who mefams situamamanahe t regularly; and the LASAT Community-forest Project, whose technician provided support only to group members (Figure 4-5, Phase 1). The Catholic Church coordinator was a member of the LASAT Community-forest group. The closed circle in the drawing (Figure 4-5A) points to the fact that the LASAT Community-forest Group was a closed group, and the unidirectional arrows indicate that the action focused on group members only. In the 1998 dry season, with no interference from the Fire Action, 45% of the ilies studied had suffered losses caused by uncontrolled fires (against 26% in 1997). In July 1999, community representatives attended the Fire Action's regional workshop (Figure 4-5, Phase 2). In this meeting, they were informed of the Fire Action'goals, received booklets with the Fire Action's main recommendations, discussed their tion with the IBAMA representative, and exchanged experiences with other peasants. Returning to the Cupu community, the Fire Action was presented and discussed mainly during the Catholic Church meetings, and Fire Action booklets were handed out. The Fire Action coordinator facilitated a community meeting on fire nagement, open to all interested persons. Although no specific Fire Group was created in the Cupu community, families and neighbors worked together for fire gement, as represented in Figure 4-5B. The house in the center represents the Catholic Church coordinator, the main Fire Action disseminator in the community. Tunidirectional arrows indicate him or other peasants talking about fire management, but

PAGE 147

134 with no cooperation from the listener. The bi-directional arrows indicate teamwork among peasants. The number of losses due to uncontrolled fires dropped to 7%. The Fire Action ended in 2000. The second set of Sequential Fire Evaluation activities was carried out in June of the same year. Families evaluated the previous dry season as successful in preventing fire losses. They were planning to work again togetherwith kin and neighbors, claiming that many families that once were not involved wanted to participate because of the success achieved by their fire management practices, and also because they feared IBAMA's repressive actions. The analysis of changes in patterns of communication and organization within two communities shows how the Fire Action was able to build and expand outward, the basedho tion ation on existing community groups, especially among those of the Catholic faith wwere used to church-sponsored collective actions and supported by church leadership. Improved communication and mobilization of peasants was a key strategy in the Fire Action approach. People in both communities were concerned about IBAMA regulations, and motivated to seek alternatives for compliance, which the Fire Achelped to organize. In the So Francisco de Itacainas community, stronger initial organization helped to support a broader organizational structure and more participin community definition of rules, which suggests greater potential for long-term firemanagement in that community.

PAGE 148

135 Figure 4-5. Representation of Cupu community organization for fire management, from n, encouraged by local leaders. Before Burning The "before" burning' recommendations were organized into four sections in the FATA/LASAT booklet: official burning license, planning, preventive firebreaks, and communication between neighbors. Those recommendations were analyzed by 1999 to 2000. In this community, peasants worked among families and ki

PAGE 149

136 evaluating questionnaire responses regarding the interviewee's knowledge of official fire use regulations, community organization for burnings, the period chosen to burn, burning after rain or not, use of preventive firebreaks, and communication between neighbors. Responding to Official Fire Use Regulations This study wanted to evaluate how government rules were reaching families at the community level. Information about governmental regulations is of difficult access by peasants. For example, in the entire Marab region there is one single office representing IBAMA, which is responsible for all environmental issues pertaining the region, from illegal logging to river pollution. The best means for farmers to obtain information from outside the rural areas is the radio, where programs are an effective way to explain dangers associated with fire and to suggest the use of preventive techniques. However, programs do not explain government laws and regulations, or present the steps required to carry out legal burnings. The studied communities had not been visited a single time by any IBAMA representative. When asked in 1999 about the IBAMA Fire booklet33, only 16% of the milies in So Francisco and 13% and of the families in the Cupu Community, reported fathey had ever seen a copy. The families which were familiar with the booklet published by IBAMA the STRs, the FATA or the LASAT. But only those who had attended the Fire Action's regional workshops could offer straightforward explanations of the issues addressed in the booklet. Possible reasons for that might lie in the in-depth analyses of the IBAMA Fire booklet conducted in the second regional workshop (described in Chapter 2). were those in which one or more members had a more active participation in or 33 IBAMA's Fire booklet was one of the most important informative sources of official fire law fpeasants.

PAGE 150

137 Despite having access to the booklet, most of the interviewees were not able to explain details of official regulations concerning fire use, such as the steps that are necessary to follow in order to obtain a permit from IBAMA, or the need for firebreaks and their specific sizes, or the criteria for fire prohibition. ision official information regarding fire use was poorly received by peasa nd Despite not being able to explain the fire legislation in detail, all the families knew that IBAMA has been applying more restrictive rules for fire use, including requirementof official permits, prohibition, fines and even physical arrest for those who do not comply with such fire laws. Also, families were aware of the fee charged per area planned to burn in order to receive IBAMA permit. Families learned about IBAMA's new rules on the radio (100% of the total for So Francisco; 97% for Cupu), on telev(49% of the total for So Francisco; 33% for Cupu), and/or through the STR or Association representatives (100% for So Francisco; 70 % for Cupu). This study found that nts at community level, raising general topics such as "fire is bad for the environment" or "do not burn without IBAMA permit." From the 72 families interviewed, only two leaders from the So Francisco were able give deeper explanations on government rules. Peasant's fears of IBAMA sanctions were a strong motivating factor in their willingness to change fire management practices, especially for those peasants not used to participating in community organizations. However, the information on the regulations was inaccessible, hard to understand, and not appropriate to their situation. Many recommendations were simply beyond their means. The Fire Action was successful, in part, because of the fear of IBAMA sanctions, which resonated with a

PAGE 151

138 reinforced a history of government neglect and coercion. This motivated people to participate in a program designed to develop their own community-based program for firmanagement, and set their own rules for avoiding fire-related losses. Communication between Neighbors. e ed ng al and commurning, by d invited e the meaning of "inform." The way in which one was to inform his ne stated neighboring lands. In 1999, five families did not inform their neighbors, arguing that the The Fire Action stressed, at several different moments, the importance of communication between neighbors, as part of the Fire Action's bottom-line recommendation on collective work. The FATA/LASAT booklet stated that "Goodcommunication is the foundation for union." The emphasis on union can be explainnot only by peasants' history of struggle, but by their social organization as a "community" (the term used by people in the two communities, mainly those sharing the same principles of the Catholic church), and by the empowering aspect of workitogether for a common goal. The same message was discussed in region unity meetings. This study attempted to evaluate the practice of informing neighbors of basking the family if they had informed their neighbors about the burning day antheir participation. This turned out to be difficult because was no formal invitation, and it was hard to standardiz ighbors was too subjective, since no specific rules were set. Therefore, it wasdifficult to consider the validity of the 'informing neighbors' practice. An ethnographic study would bring valuable information on this topic. Still, the results show a very high incidence of informing neighbors, especially those close by. In 1998, in So Francisco do Itacainas (n=36), only one family they had not informed neighbors of a burning, because they were located far from

PAGE 152

139 burning did not present any danger to them, either because it was far from their landbecause only a small plot was to be burned, after it rained. In the same year, all the families interviewed in the Cupu community (n=27) declared that the s, or y had informed their ng year, after the Fire Action started, two famile n Fire d lled ould also help families in case of conflicts with IBAMA due to fire u eeking compensation for losses causeto neighbors of the burning day. In the followi ies declared that they had not informed their neighbors because the area to bburned was distant from the neighbors' land. The practice of "inviting neighbors" proved to be more measurable than was 'informing neighbors', although the invitation was also informal (which meant no writteproof of the invitation). Burning collectively was the main point made by the Action, and families were informed that the Fire Action (represented by the local association, FATA, LASAT and STR) would only support those families that burnecollectively. This support was not restricted to preventing and fighting uncontrofires; the Fire Action w se. Families were aware, through radio and TV broadcasts, of IBAMA's new procedures on fire use prohibition, fines, or even jail for those who did not comply. As seen in Chapter 1, since Roraima's large uncontrolled fires in early 1998, burnings could only be performed with a permit issued by IBAMA, and during the weeks of the high dryseason, even those that had licenses were forbidden to burn their fields. If neighborswere invited to participate in a burning event and did not come, they would have no support from the Fire Group or community leaders in s d by uncontrolled fire reaching their land. The methodology used in order to determine neighbors' presence at burnings was ask interviewees to describe their burning situations, including proximity to lands

PAGE 153

140 belonging to neighbors who were present on the burning day, and their connections with those persons (kin, neighbor, friend, and/or member of Fire Group). If the neighborswere not present at the burning, the interviewee was asked if they had been invited. Crossing data from interviews of different neighbors also helped to check people's participation during burnings. In the So Francisco community, in whi ch the Fire Action started in 1998, the numbways unity e ing a family, was ready to be burned, while the cor it to er of families inviting neighbors was high and stable in 1998 and 1999, and alhigher than in the Cupu community: 81% and 80%, respectively for the two years. The number of families that invited their neighbors to burning events in the Cupu comm(n=29) increased from 34% in 1998 to 64% of the families in 1999. The differencregarding this practice in the two communities might be due to the Fire Action. In theCupu community, the percentage of families inviting neighbors almost doubled from one year to the other, while in So Francisco, the percentage of families inviting neighbors continued to be high. The higher number of families inviting neighbors in the So Francisco do Itacainas community was probably due to their commitment to fire management since 1998, represented by the Fire Groups' work and by the local association support, which also discussed fire management in assemblies. The importance of communication between neighbors can be illustrated by the following example, in which losses were caused by lack of communication and plannamong neighbors. In 1999, in the Cupu community, a slashed 1.5-hectare plot of old secondary forest (juquiro), owned by the Pereir ntiguous neighbor's pasture was not (Figure 4-6). To avoid causing the neighbor any losses, the field's owner waited for either his neighbor to burn the pasture or fo

PAGE 154

141 rain, so as to reduce the risk of uncontrolled fire. However, it rained twice and the field, which was located by the dam, was too wet to catch fire when the Pereiras tried to burlater. Mr. Pereira claimed that his loss would have been even greater if an uncontrollfire had escaped from his field and burned his neighbor's pasture. He was afraid theneighbor would protest and he n it ed would have had to feed the neighbor's animals, but his pastuvery er of ch ost al foundation to pre re was not large enough for that. Mr. Pereira and his neighbor do not get along well. Nevertheless, he said that in the coming year it would not happen again, because the neighbor had sold his land to his godfather, and they will plan burning events together. It was the first time that Mr. Pereira did not burn his slashed field, and this may reflect a higher awareness that losses caused by fire are the responsibility of the ownthe burning event. Mr. Pereira's family ended up cropping 0.15 ha in the slash and mulsystem. They harvested 200 kg of beans, but did not produce rice and had to buy it from neighbors. Thus, this study found a surprisingly high frequency of informing and invitingothers regarding a planned burning, albeit much of it informal and unrecorded. Communication and cooperation among community members is probably the mimportant factor in fire management, and the Fire Action build on this soci scribe adequate numbers of people required, and specific practices they should follow, and to promote and support collective action.

PAGE 155

142 Figure 4-6. Representation of a burning situation in the Cupu community, in 1999,in slashed a plot of 1.5 ha in an old secondary forest (juquiro), and when it was pasture, and decided to wait for a rain which the lack of communication caused loss to one family. Pereira's family ready to be burned, they feared that fire would reach the contiguous neighbor's to wet the pasture and, consequently, diminish the risk of fire to spread. Their neighbor did not plan to burn the it rained twice, and the Pereira field, located by the dam, did not catch fire when that year, losing all the investment in labor and other services that the old g of as that an uncontrolled fire starting in the beginning of a dry season is much more dangerous than a fire starting later in the dry season, since when it gets out of control, it may take weeks or even months before a rain can turn ecosystems wetter. Changes in this practice (burning later) were measured though interviewees, asking peasants burning dates and number of days before or after the rain that they burned their fields. pasture that year. The two families do not get along very well. However, he tried to burn it later. The Pereira family did not plant on that fieldsecondary forest would provide, and had to buy rice. Period Chosen to Burn: Month Choices One of the Fire Action's recommendations was to avoid burning in the beginninthe dry season. The reason for this recommendation w

PAGE 156

143 Results indicated that peasants in both communities were burning slighary practice, after the Fire Action (Figure 4-7). Evaluation of this practice wasse only average precipitation data were available for the Marab region as a whole, and no inter-annual variations were specified on the frequency of rains in the Figure 4-7A shows the monthly precipitation in the Marab region (average of 20 rding to SECTAM 2004), against the total of hot pixels obtained for the state of Par (average hot pixels obtained by NOAA-12 satellite from 1998 to 2003, according to INPE 2004). Although the Marab region is located in an equatorial super-wet climregion (IBGE 2002), with average annual precipitation of 2043 mm, the region presents a tly later than custom not easy becauliterature. yrs accoatic precipitatiostate of ParSeptember the Marab strong rainder these circumstances, and likely to be ineffe clear four-month-long dry season (May through September) during which the average n rate stays below 100 mm. However, the distribution of hot pixels in the shows clear concentrations of burnings events occurring in August and : 53% of all the hot pixels. Nonetheless, is still a risky month to start a fire in region, because ecosystems have been drying since late April or May, ands are expected to arrive only in October. Therefore, controlling fires is hard un ctive and pose risks to the lives of those taking part. As local peasants say: "August-dry vegetation feeds fire like gasoline." Although peasants have reached a consensus on the fact that burning early in the dry season increases the chances of fire losses, families in both communities stated they usually conducted burnings in August, before the rains. Some of the peasants justified this practice by explaining that fire did not normally get out of control 10 or 15 years ago as it does nowadays. In the words of

PAGE 157

144 Mr. A d y mbrosio, a Fire Group member in the So Francisco do Itacainas community, "back in 1983 when we arrived here in the community, burning an agricultural field well was difficult because of the constant rain. It is different now: the dry season has been very strong in the last few years." In those days, he explained, "the forest worked as afirebreak." The data from interviews indicated changes in the choice of month for burning in both communities studied: families tended to burn later in the year, and to concentrate burnings in certain months. Although the communities engaged in the Fire Action at different moments, data showed no differences between them in terms of choice of monthfor burning. Changes were noticed, however, in the reasons given by families for burning late in the season, in 1998 and 1999. The families in So Francisco relate changes to the need to prevent uncontrolled burnings, while in Cupu, families explainethat in 1998 they burned late because of IBAMA changes in rules for fire use, and thewere waiting for more explanations on this topic. They claimed they had burned in a hurry, after receiving no information, and many families suffered fire-related losses in this year (see section on fire losses).

PAGE 158

145 Figure 4-7. Rainfall distribution and months chosen for burning. The upper graph A) shows the average precipitation in the Marab region (SECTAM 2004) in relation to the average number of hot pixels in the Par state (INPE 2004). It shows a clear dry season, and high concentration of burnings on August and September. The following two graphs relate the Marab rainfall distribution with months chosen for burning by families studied in So Francisco do Itacainas community and B) Cupu community C), for 1998 and 1999. Figure 4-7 shows the average precipitation distribution in Marab (SECTAM 2004) associating it with families' choice of month for burning in the two communities studied, (B) So Francisco and (C) Cupu, for 1998 and 1999. The average rainfall distribution

PAGE 159

146 explains families' rationality for burning in August and September. As rainf are ready to start preparing their agricultural fields, which means slashinthe vegetation in June (2 to 4 weeks), waiting for it to dry in the mweeks), and burning before the rain starts, in August or early Septemthe FATA/LASAT Action started in 1998, Figure 4-7B 1998 to 1999, there was a small decrease in the num 12% to 9%), and an increase in the number of burning events in 50% to 68%). No changes were seen in October, and a decrease was ber (from 15% to 0). This result shows a higher c 4 to 3 months), which may reflect the collective burning events supported ally, the higher concentration of burning events in y be related to families' decision to wait for the first raight correlate to families' insecurity concerning how to burn in ce with IBAMA's laws, since it was the first year of IBAMA's mnation of fire-use prohibitions. pattern was observed in the Cupu community (Figur (from four to three months), reduced number of burning all decreases in May, theyg onth of July (3 to 4 ber. In the So Francisco community, where shows that, fromber of burning events in August (fromSeptember (fromnoted in Novemoncentration of burning events (fromby the Fire Action. AdditionSeptember main. Burning events late in November 1998 maccordanassive media dissemiThe samee 4-7C): a higher concentration of burning eventser (from 48%November formal fireith neigh events in August (from 11% to 0), increase in the number of burning events in Septemb to 64%), unchanged figures for October (30% and 32%), and a decrease in (from 14% to 5%) and December (from 4% to zero). Despite the fact that no groups were created in the Cupu community, families claimed to burn w bors and family more often than they used to before the Fire Action started.

PAGE 160

147 Given the limitation on local microclimate data for a more accurate evaluation on families choice of month for burning, it appears that peasants in the two communities studied in the Marab region concentrated their burning in September and October. Thwas a shift of several weeks later than their traditional time of burning in August and September, before the rains began. This shift to later burning reflected a combination fear of IBAMA sanctions, and compliance to Fire Action rules. Burning after the First Rain The practice of burning after the first rains is indirectly connected with the choiceof certain months for burning, since the rainy season generally starts in late September in the Marab region. The Fire Action stressed the importance of burning after the first rain, both in its booklet and in meetings. Unlike the evaluation of impact on famchoice of month to burn, which was limited due to lack of local climatologic data, the 'burning after rain' practice allowed more precise measuring. Families were very precise when remembering rain events and burning date and answering if the burning event wasbefore or after the rain. If the burning e is of ilies' vent was carried out after rain, they were able to say ht ning the plot, grass seeds are sown where weeds had prevailed prior to the burning. Families claim that in addition to renewing pasture fields, burning is important to reduce the ow many rainfalls there had been before burning, and how many days after the lasrain they burned their agricultural fields. Evaluation of this practice focused on burning before or after rain, and on understanding the reasons for each choice. The general Fire Action recommendation was to burn plots for agricultural fields after a first rain whenever possible, and to always burn after rain when burning pasture. Pasture fields are divided by fences, and burned every 3 to 4 years following a rotation system, aiming to reduce the amount of weeds and favor pasture sprout. After bur

PAGE 161

148 amou nt of weeds which are not suitable for cattle grazing, to scare away snakes and also to exterminate ticks. Results showed that all of the peasants who burned pasture fields after the rains were unanimous in affirming that it was a very good practice for pasture maintenance.They claimed that only a few days of sun were enough to dry the pasture and for its slashed weed to burn effectively. The experience also showed th at because the soil was he vegetation, and the soil temperature was not as high as when t they had not alway wet, the fire spread faster over t the pasture is burned before the rains. Therefore, burning after the rain rains did not burn grass roots. In addition, grasses sprouted faster when compared to earlier burnings. If a pasture is burned long before a rain, other dryness-tolerant species startgrowing and therefore competing with the grass. Another great advantage, emphasizedby the Fire Action, is that the rain wets surrounding vegetations, therefore reducing therisk of fire spreading. In So Francisco do Itacainas, all of the families that burned pasture fields in 1998(24% of families) and in 1999 (40% of the families) claimed that they had waited for it torain before they burned. In the Cupu community, 66% of the families in 1998 and 39% in 1999 burned their pasture fields, respectively. Also, all the families claimed they had waited for rain before burning their pasture fields, in both years, but tha s done this before 1998. As mentioned before, some families explained that they burned late in the dry season of 1998 because they were waiting for IBAMA's guidance regarding fire use. This guidance did not take place. Families' lack of information on IBAMA's fire laws and the fear of not being allowed to burn may explain the high

PAGE 162

149 number of families (66%) burning their pasture fields in 1998, since in usual years families burn their fie ld each 3 or 4 years. the does e re a single rain might leave the slashed vegetation too wet to dry prope nd ld. ex Spreng] Schum.) and brazil-nut trees (Berthollethia excelsia H.B.K.), among others. He Burning pasture fields after the rain was not a controversial topic to peasants because pasture fields burn well after only a few days of sun following a rainfall. Scientists had reached the same conclusion in the Paragominas region, also located inAmazonian 'arc of deforestation' (Uhl and Kauffman 1990). With regard to agricultural fields, however, waiting for it to rain before burning is a more complex topic. Too much rain results in the sprouting of weeds, and the slashed field may not be dry enough again for a good burning. Moreover, burning after rainnot work on slashed palhada, or one-year-old fallow fields: there is too much straw leftfrom the previous harvested plants, and if it rains, the straw 'sticks' to the ground, and firwill not spread to it. Another critical situation is when slashed fields are located on lowland wet soils, whe rly, favoring weed sprouting, or, even worse, not allowing it to become dry enough to burn. Figure 4-8 exemplifies a slashed primary forest burned after heavy rain, at the end of the dry season (Oct 10, 2000). The peasant was afraid that it would rain again, and burned the field before it was dry enough. However, the large number and size of trunks left unburned made it impossible to reburn, and rendered the plot unsuitable forplanting. Living in the So Francisco do Itacainas community since 1994, he had always cropped secondary forest and for the first time he slashed a primary forest. His concern with the primary forest can be seen by the uncut plot left between the pasture athe slashed field, in which there were cupuau plants (Teobroma grandiflorum [Wil

PAGE 163

150 expressed his sorrow about the situation, stating "Agriculture (roa) is a lot of workalready; if the field does not burn, we are fa cing a huge loss". Figure 4-8. A slashed primary forest burned after heavy rain, at the end of the dry The reburning technique, locally called encoivara, is a practice in which the unburned vegetal material, resulted from a not very good burning, is manually piledusing scythes and machetes for bigger trunks, and then ignited. Smaller pieces are piledby the bigger ones. It is a difficult and time-consuming activity. In spite of such difficulties, peasants generally claimed it was better to burn after a rainfall because it did not affect their production, and it really helped in the prevention and control of season, resulting in a bad quality burning for cropping (Photo by author). up,

PAGE 164

151 undesirable fires. Some peasants, however, stated that burning agricultural fields aftethe rain was good for preventing uncontrolled fires, but bad fo r r production because the lashed field did not burn well, which negatively affected their crops. out of control if it ants to control it. In 1997, for instance, the Duarte famunity, lost 90 coconut trees he had nt loans, plus forested and pasture areas that they did not wre born in the north state of 1983. Below is one of their commat uncontrolled fires started before the rain: here, almost past the field. When I arrived homgh crackling noise in the coconut fields. The fire had jumeld with pasture and rice. There was fire it. Ninety coconut trees were burned. because sit escape from the e leaves in the forest were wet 'by life' do not wait for the rain, fire spreads re. In the past, however, we could burn one alqueire [5 ha] or 2 [10 ha] of slashed vegetation, and tation. Nowt the fire would not spread into the forest, but now is does? Mr. D.: I believe that it changed due to rain [frequency]. Before [20 yrs ago], it Then, the forest was wet, you would walk inside it and it was cold. And nowadays en the forest gets dry. When fire reaches the forest, it gets stronger and spreads all over, leaves on the floor, and they were wet. s When the vegetation is too dry, before the rains, fire quickly gets escapes from the planned burning area, and it is very difficult for peasily from the So Francisco do Itacainas commacquired with governmeant to burn. Mr. and Mrs. Duarte weGois (now Tocantins), and have been living in the community since its occupation in ents on trying to combMr. Duarte (D.): I was at the barn, around 3 km frome, we could hear this hiped to a year-old fallow fieverywhere; and there was no way to control This happened prior the rains in 1997. Back then we used to burn before the rain nce we arrived [in the state of Par], the fire did noplanned burning field to the forest. I think that at that tim, you know? But nowadays if weinto the forest just as if we had put dry branches under the fi the fire would surround the neighboring forest, burning only the slashed vege it is not like that anymore. Question: What do you believe has changed from that time, so that in the paswould rain almost every month in the state of Par. It used to rain very often. you see how it is, sometimes we have 2, 3, or even 4 months without rain; thbecause everything is dry! In the past it was not like that. You would scratch the

PAGE 165

152 Mr. Duarte lives in the So Francisco do Itacainas community. He was born in th e state of Gois, and has been living in the community since its occupation (1983). Fifty-seven percent of the families studied in the So Francisco do Itacainas community, where the Fire Action started in 1998, waited for the rain to come before they set out to burn their agricultural fields in 1998. This percentage went up to 86% in 1999 ("n" total for this variable was 35 families). In 1998, peasant families were more resistant to change their behavior because they feared that their agricultural fields would not burn well after the rain, perhaps reducing their production. Some of those families who burned after the rain admitted that they burned late in the season fearing IBAMA reprimands, but not because they believed it was a good practice. In 1998, three of the ten Fire Group members burned before the rain, one did not crop, and the other six burned after the rain. All of the peasants that burned after the rain said it was the first time tby acrain. positihis agand his family would have to work hard doing the reburthe rain' to social pressure towards not causing losses to anyone, and not because he believ in the Spractiwhole year's production of the agricultural field depended on the burning quality. In June hey planned to do so. Before the Fire Action, they would only burn after the rain cident. In 1999, there was an increase in the number of families burning after the Some families and most Fire Group members associated this increase with the ve results obtained in the prevention and control of fires. In spite of the apparent success of this practice, one Fire Group member stated that ricultural field did not burn well, n (encoivara). Another peasant credited his adoption of the practice 'burning after ed it was the best practice. Changing behavior was a process, as the Fire Groupo Francisco do Itacainas community explained. Even for those who lead new fire ces, such as the Fire Group members, changing behavior was challenging. The

PAGE 166

153 of 19 99, for instance, the Fire Group was evaluating the past dry season, and planning the next one. Mr. Carvalho, then Director of the AGRAF, explained how he understood the reluctance on the part of some families to change their behavior, especially those who did not participate in the meetings held to discuss the fire issue. He stressed the role of Fire Group members in bring consciousness to those that usually do not get involved in community organization. In the following segment there is a discussion of this reluctance: Mr. Carvalho: I think that, in order to bring consciousness to the people, we should regional seminar] (...). I can say this for me: when it was August, I used to burn my plantation. And the worse is that: people are used to burning early. He [peasant] is is going on, and he gets worried, I agree with him. When a person is used to burning I am one of those. When it rained [in 1998] and the slashed vegetation was getting burning their fields before the rains. There are many people with this habit (...). agriculture.' I don't know, it is our habit. But sometimes it works like that: we burn Then, besides burning their forests, one can even go to jail...(...) If we are conscious others. If one is resistant to change, we have to understand his behavior, and try to communicate to ing ent organize a [community] meeting, as soon as we arrive [from the Fire Action agricultural field; the agriculture field was then clean, just ready for the rain and apart from the 'movement' [organization for fire use], he does not know whatearly, when it rains, and his agricultural field is not burned yet, he gets worried. ready to sprout, I said: 'this year I will not have agriculture!' People are used to When it rains at night, he thinks: 'my God, it seems that this year I won't have early, and suddenly a uncontrolled fire starts, burning forest, or neighbors' landthat we have to change, we have to bring consciousness to him what we know. Mr. Carvalho, So Francisco community, then the AGRAF Director. The discussion presented in the speech above touches the very essence of the Freirian pedagogy that was the basis for the Fire Action: the peasants critically analyztheir situation, challenged by the complex situation, in a process of consciousness development. In this process of how to disseminate awareness inside the community, fire group members, mediated by dialogue, plan their next steps. Their solutions are differfrom the coercive government strategy; they have empathy with other peasants in their

PAGE 167

154 reluctance to change their behavior (burning after the rain) because they 'have been there', i.e., before their present awareness they used to behave like that too. And they conclude that only through communication will they continue as partners, likely to develop a more critical understanding of their reality. In 1998, 64% of the studied families in the Cupu community reported that thewaited for rain before burning their agricultural fields (n=28). Families stated thnever used t y had at they o intentionally wait for the rain to burn. All of the families interviewed in this cinformactionthe Fiafter r, 41% sy had uin thethis practice, is probably due to the higher level of community organization for fire management in the community. This organization was reflected in local rules discussed in assemblies and in the implementation of fire groups, resulting in more appropriation of the Fire Action discourse. Preventive Firebreaks Question: People used to say that the forest itself was the firebreak. Did it really make one, the forest will not prevent the fire from spreading because the fire will ommunity credited the high number of families adopting this practice to the lack of ation about IBAMA's laws and fear that IBAMA might engage in repressive None said they had changed because of their belief in it as a better practice. After re Action started in 1999, the number of families saying they had waited to burn ain increased to 74% (n=23). Among those saying they waited for the rain to burnaid they planned to burn after the rain from them on. The remaining 59% said thenintentionally burned their fields after the rain. The overall higher number of families burning their agriculture fields after the rain So Francisco community, as well as the higher level of consciousness in adopting work as a barrier to fire? Mr. Duarte: Yes, the forest used to be a real firebreak. Nowadays, if you do not

PAGE 168

because you kill the fire here, a small spark starts a fire again over there. No one in the community since 1983. There was a cultural belief, based on experience, that the forest itself was a natural firebreak, but peasants have found out that this is no longer true, as shown in the narrative above. This critical evaluation of their knowledge-weather had changed, and peasants' practices should then change too was key for peasants changing their practices. The Fire Action's recommendation about preventive firebreaks (called locally aceiro) was clear, especially concerning agricultural fields, as discussed in Chapter 2. The Action stresses the fact that preventive firebreaks, i.e., firebreaks made before the burning day, were mandatory (despite not being official). Several possible definitions of "firebreak" were discussed during the meetings, from "border between the slashed vegetation and the contiguous ecosystems" to "clean and contiguous line on the ground surrounding the area to be burned." After a great deal of debating, the latter definition was always the final choice. Every head of family interviewed said they knew about firebreaks, usually made to protect material goods such as houses, silos, and fences, and also to protect other direct investments such as perennial crops, slashed fields to be burned later, and pasture fields not planned for burning. Firebreaks made along fences in pasture fields were not new to peasants. They usually make firebreaks to protect what represents direct high investments to them. When a fence was located at the border between two properties, families would take on conjoing the firebreak, with each family responsible for the firebr burn it, there's no doubt about that. Putting it out will be a very difficult task alone can control a forest fire, it will take a lot of men. Mr. Duarte, So Francisco do Itacainas community, born in Tocantins state, living t responsibility for makin eak side of their land. All the families said they protected their fences from fires, 155

PAGE 169

both bund that thnd 2 meters wide. Peasants said thalong pasture fields, regardless of any recommendation proposed by any authority or NGO. Therefore, the present evaluation of the Fire Action was based mainly on the firebreaks used for implementation of agricultural fields. The new aspect concerning firebreak practices in both communities was the use of firebreaks as a means to protect forests, as recommended by the Fire Action. In the two communities studied, preventive firebreaks located far away from the burning border and in the forest are usually familiar to the elderly in Northeastern states of Maranho and Cear. Although they had never used this kind of preventive firebreak in Par before, they claimed that their families used it back in their home states when they were children. Figure 4-9 (above) shows a photograph of a typical borderline between the slashed vegetation and the remaining forest, where Brazilian laws require 6-meter wide firebreaks. Figure 4-9 (below) shows the Fire Action's representation of a firebreak located far away from the burning border. The "far-away" firebreak proposed by the Fire Action is located inside the forest (8 to 15 meters away from and along the area to be burned). In this way, the standing trees work as a barrier, reducing the fire speed (wind barrier) and intensity of the fire (due to the colder and wetter microclimate), and the fire goes down, spreading on the ground. After a few meters a firebreak, which is only 1.5 to 2 meters wide, may hold back this ground fire that has became too weak to cross the firebreak. Even when a tree in flames falls, crossing the firebreak, this accident is easily verified and controlled because efore and after the Fire Action was implemented. Observations in the field foe firebreaks protecting fences are normally between 1 a at all families have always used this kind of preventive firebreaks in fences located 156

PAGE 170

157 peasants can walk through the firebreak faster and safer, and combat it while uncontrolledfire is still beginning. This technique was observed to be considerably more efficient and safer than a firebreak located by the burning border, as recommended by IBAMA. Similar practices using firebreaks inside the forest were observed in another colonist community located in Paragominas, in the state of Par (Mattos et al. 2002). Brazilian law's minimal requirement for firebreaks i s 3 meters wide, and twice this width nd the one can observe the burning after it starts becau when the burning field is adjacent to a forested area or a neighboring property. In the case of protecting forested areas, peasants claim that firebreaks proposed by the law have two main limitations. First, they own neither machinery nor the resources to rent any to make a firebreak this wide. In their communities, 6-meters-wide firebreaks are only comparable to their main roads, such as the one crossing the community. Second,firebreaks located near the border of the burning area are not very effective in preventing fires from escaping, for three main reasons: (a) the fire often crosses it anyway. Sixmeters is not enough to separate the standing forest and the burning vegetation, aheat will dry the green leaves and thin branches, which is easily seen after a burning (brown leaves) along the forest line; (b) no se it is too dangerous (heat, smoke, falling trees and branches); and (c) it is comparatively harder to control escaped fires through firebreaks located near the border of the burning area, since one cannot walk through it, and people and equipment transit from inside the forest is less effective (takes longer; is more dangerous).

PAGE 171

158 between agricultural field and forested area, where fire law recommend a 6 mforest near by the burning field, located away from the burning Figure 4-9. Firebreak bordering forests. The above figure shows a typical border wide firebreak. The below figure shows the firebreak recommendation for border, developed by peasants. In 1998, all of the interviewed families in the So Francisco community reported having used some kind of firebreak for agriculture fields. Five families (12.5%) reported fire-related losses. These families had not worked with the Fire Group. In 1999, the one

PAGE 172

159 fam ily which did not use a firebreak was the family of a Fire Group member, and they had no losses. He said the firebreak was not necessary because it was "just a smof slashed secondary forest" (0.9 ha); besides, "it had rained well and there were five people working on the burning day"he burned in early October, and they were able to prevent uncontrolled fires. This adaptation of practices was seen in other situations, when Fire Group opted by not to use important practices due to their understanding that the other practices used for that specific condition would be effective in preventing an uncontrolled fire occurrence. In 1998, 75% of the families interviewed in the Cupu community who had ed burnings in their agriculture fields (21 families) reported having used s. Only one of them incurred losses due to their fire. Of the 7 families that hnot used firebreaks, one suffered fire-related losses from their burning. In 1999, fifteen ilies said that they had used firebreaks (65% of the families burning in that yemily reported losses. Of the eight families which had not used a firebreak, one reported losses. In the Cupu community, in which the Fire Action started in 1999, there ecrease between the years in the percent of families that used firebreakdifference may, in part, result from a discrepancy regarding the firebreak concept. As we have discussed above, many peasants refer to a simple border between the slashed all field performfirebreakad famar), and one fais a slight ds. This vegetationvegetation Cupu community in 1998 were made inside the forest (far-away firebreak); all were located by the burning border. On the other side, in 1998, more families may have used firebreaks due to fear of IBAMA's new rules. The high number of losses reported in this and the standing forest as a "firebreak," not considering whether or not the spread along this area has been removed. None of such firebreaks in the

PAGE 173

160 community in 1998 (48% of the 27 families reported losses that year) indicated that firebreaks used were not efficient in preventing uncontrolled fires. In 1999, only two families reported fire-related losses, against fourteen in the year before. The data above do not point to a positive correlation between the use of firebreaksand prevention of losses, nor to an increased rate of firebreak use from one year to theother. Despite discrepancies concerning their different types, firebreaks per se represent no guarantee against escaped fires. As will be presented in the section on Fire Strategiedifferent sets of practices appl s, ied in each situation determinate the success of a burning event (considering both cropping quality, and fire control). On the burning day The Fire Action booklet was composed of ten recommendations for the burning day, as described in Appendix C. The first four recommendations discussed the importance of the presence of the families responsible for the burning, of communication among neighbors, and an adequate number of people participating on the burning. Thefifth recommendation concerned the use of backfires. The sixth and seventh recommendations covered water and fire control equipment. The eighth recommendation concerned the time to start a burning event. The ninth provided more details on avoiding the driest days, as well as windy periods. The last recommendation was about the importance of monitoring the burning. This study evaluated these practices by asking about the number of people participating in the burning, the use of backfires, and the time the burnings are started. In addition, control practices such as water and equipment, and monitoring of the burnings, were evaluated. Results for control practices are presented inthe section After the Burning.

PAGE 174

161 Number of People at the Burning Event Using fire in groups was the main issue emphasized by the Fire Action, and it is reflected on the booklet title "Stronger Bonds, Fewer Losses." On the bu rning day, the Fire A e ire Action started in 1998, none of the interviewed family meing events only on their own. In this c1; the o do annual unities regarding the Number of People by Area Burned ratio (people/ha), So Francisco do Itacainas presented a wider distribution, while Cupu ction recommendation was that fields should not to be burned by one single personalone, but rather by a group formed by at least four people. Interviewees were asked tolist the names of the persons participating during the burning. The following evaluation focused on number of people during burnings of agricultural fields in 1998 and 1999. The results showed clear changes in the pattern of "number of people present on thburning day". In the So Francisco community, where the F mbers conducted burn ommunity, the average number of people present at the burnings was 6.0 (SD=4.0n=36) in 1998, and 8.3 (SD=4.90; n=33) in 1999. In the Cupu community, whereFire Action started in 1999, the average number of people present at the burnings was 3.3 (SD=1.56; n=26) in 1998, and 4.5 (SD=2.75; n=24) in 1999. The maximum number of people present also increased in 1999: up to 20 from 14 in 1998 (So FranciscItacainas), and up to 15 from 7 in 1998 (Cupu). In order to obtain a clear understanding of the changes in the number of people attending the burnings, while simultaneously allowing a comparison between inter-and inter-community patterns, a ratio was calculated dividing the number of people present at the burning by the area burned (people/ha). Figure 4-10 shows how this ratio was distributed against the percentage of families interviewed, in (A)So Francisco do Itacainas community and (B) Cupu community. Comparing the two comm

PAGE 175

162 presented a higher concentration of lower ra tios. One other common pattern found betwe r en the two communities was the displacement of Frequency of Families from thelower to the higher ratios between 1998 and 1999, which points to an increasingly highenumber of people attending the burnings from one year to the next. Figure 4-10. Distribution of ratio of people to size of area burned, for percentage of 3 in 1999), and (B) Cupu (n= 26 in 1998; n=24 in 1999). The ratio was eople present at the burning by the Area Burned. families interviewed, in (A) So Francisco do Itacainas (n=35 in 1998; n=3calculated by dividing the Number of P

PAGE 176

163 In So Francisco do Itacainas, the ratio was higher than 2 people/ha in 20 burevents (57%) in 1998. In 1999, this ratio was found in 25 burning situations (76%). Theratio above 4 people/ha rose from 8 burning situations (23%) to 15 burning situations (46%). Comparatively, in the Cupu community, only three burning ning situations (12%) (50%). sed Time to Stawho discuss a series of factors that influence the time chosen for burnings. The FATA/LASfire: after 3ns on this practice were observed. During the Fire Action meetings, technicians presented the official recommendations for burnings events, both at the beginning and at the end of reached a people/ha ratio above 2 in 1998; this ratio was found in 7 burning situations (29%) in 1999. The ratio above 4 people/ha only was found in one burning situation in 1999. In the Cupu community, the ratio was one or less person/ha in 17 burning situations (65%) in 1998, and in 5 burning situations (21%) in 1999. In 1999, burnings were concentrated at around 1.1 to 2.0 persons/ha, with a frequency of 12 burnings The Cupu community's low patterns regarding people/ha in 1998, together with the changes to a higher number of people/ha point to the positive impacts of the Fire Action in this community. Many families in So Francisco do Itacainas reported that they uto conduct burnings in their fields by themselves before the Fire Action, as this study observed in the Cupu community in 1998. This shift toward a higher number of people attending the burnings is directly connected with the lower number of losses caused byfire, as we discussed in the section Losses Caused by Fire. rt Burnings This study founded that the time to start a burning is a complex topic for peasants, AT booklet provides specific recommendations regarding the time to start a p.m. in agricultural fields, and after 4 p.m. in pasture fields. Many variatio

PAGE 177

164 the dad y here ultural field is located: wind is faster during a burning on a hill or highland plot, comprt a ted in agricultural y families in the two communities studied. For both commd f y. However, the vegetation is generally cold in the early morning, and the wind is too light. At the end of the day, wind is normally light too. Peasants claimed that this factor results in bad burnings because fire does not spread easily. Slow fires represent a problem because they burn the soil for too long, which peasants say is not good for agriculture. In fact, one study of the effects of swidden fires on soil temperature showethat heating the ground to 200C increases soil nutrient availability, but temperatures above 400C lead to a complete destruction of organic matter, to the fusion of claparticles into sand-sized aggregates, and to a reduction in the cation exchange capacity (Sertsu and Sanchez 1976). Peasants argued that the best time to burn depends on wthe agric ared to fires in lowland plots. Furthermore, plot size also has its influence: the bigger the plot, the faster the winds. The type of vegetation to be burned plays a role, too: winds are faster in slashed primary forests than in secondary forests. The number of days elapsed after the last rain is also a factor to consider when defining the time to stafire, since it influences the humidity of slashed materials. For pasture management, peasants do not want the fire to spread slowly, because slow fires kill the grasses. Figure 11 shows the distribution of times in which fires were star fields in 1998 and 1999, b unities, a common pattern of later burning times was observed in the two years studied. In 1998, in the So Francisco community, the burnings were concentratemainly at 1 pm (41% of the families) and at 2:00 p.m. (37%) (n=32). Nineteen percent othe families burned their fields at 3:00 p.m., and 3% at 4:00 p.m. In that year, the Fire

PAGE 178

165 Group worked mainly helping each other and those asking for support, which may explain the minority of families burning after 2:00 p.m. (22%). In 1999, after an assembly meeting held in the community voted to maintain eight Fire Groups (instead of only one) and reinforced community rules for burnings afterp.m. only, the result on this practice became more effective. 3 Figure 4-11. Distribution of studied families' 1998 and 1999 choices of burning time for their agricultural fields, for A) So Francisco do Itacainas community and B) Cupu community.

PAGE 179

166 After Fire Groups were distributed by region and a defined rule for later burniwas defined in assembly, 78% of the families studied in this com ngs munity followed this recom Fire Group members: Mr. orest, well becaufallow seconassembchurch'matime was somewhere around noon, when winds are stronger. In 1998 (n=26) and prior to the Fire Ac mendation. Only 6% of the burnings started at 1 p.m., followed by 15% starting at2 p.m., 48% at 3 p.m., 21% at 4 p.m., 6% at 5 p.m., and 3% at 6 p.m. (n=34). The threeground-breaking individuals who burned after 4:00 p.m. were Valdinei Cruz, (major Fire Group coordinator and director of local association AGRAF) burned at 6 p.m. Cruz burned a 1 ha. of uphill agricultural field from the primary fand he said his field burned well. Mr. Dionisio burned late (5 p.m.) because he had helped with two other burnings earlier that day. He complained that his field did not burn se it was flat land and he had waited for the rain. If this happens again in the future, he plans to burn earlier in the day. Mr. Raimundo burned a 1.5 ha, one-year dary forest (palhada) at 5:00 p.m. and his field burned well. In the Cupu community, due to the absence of a local community association, no ly was held in 1999 to set the community rules for fire management. The Catholic s local leader seem to be the main coordinator of the Fire Action, by reading the Fire Action booklet and discussing fire management during weekly meetings at the Church. Others organized among families and neighbors. Some peasants explained that ny burned their fields late in 1998 because they had heard about changes introduced by IBAMA and were waiting for guidance on fire use. But once the dry season was nearly over and no guidance had been provided, they burned later in the dry season and later in the day, fearing repression from IBAMA. They explained that the usual burning tion, 46% of the studied families concentrated their burnings between noon

PAGE 180

167 and 1e ned imately an hour later t explained before, slow fires stay too long in the soil, harming the soil and consequently p.m., 27% of the families burned at 2:00 p.m. and 23% of families burned their fields at 3:00 p.m. A single family (4%) burning at 5 p.m. reported they had burned latfor fear of IBAMA, but not because they believed it was a good practice. In 1999 (n=22),when the Fire Action had been implemented but not the Fire Group, burnings were concentrated at 2:00 p.m. in 68% of the cases. One family burned at 11 p.m. in September, therefore before the rains came. They said it was the first time they were burning so early in the day, and that they had liked it because it caused no losses to anyone involved and, in addition, the field had burned well. Two of the families burat 1 p.m. (9%) and 19% burned at 3 p.m. Whereas traditionally peasants had burned their fields at midday when the sun washottest and the wind the strongest, results of the interviews in 1998 and 1999 showedpeople in both communities tending to burn later in the day, some approx han before and others in the late afternoon, which was previously unheard of. Thischange in practice was partly due to fear of discovery by IBAMA, and partly through the recommendations of the Fire Action. Use of Backfire Backfires, locally called contra-fogo, are a practice that can be used in preventing and controlling fires. Its use as a control technique will be discussed in After Burning section. This practice when used for prevention consists in starting the burning at theborder in opposition to the main wind direction. For example, if the main wind directionis from east to west, the burning should be started on the west side. After the fire has spread slowly some meters against the wind, the main fire (pro wind) is lighted too. It is not desirable that fire should spread against the wind for a long distance because, as

PAGE 181

168 the agricultural production. Since fires have to be started in different positions, at leastwo persons are needed, one starting the backfire and the other starting the main fire. This backfire works, at the end, as a firebreak, eliminating fuel near the border where the main fire comes stronger, as peasants say, na batida do fogo, or the side where main"hits." Peasants in the two studied communities were used to using preventive backfireswhen some border of the field to be burned was located by something that they wantprotect from fires, such as houses, barns, perennial crops, or pasture fields. In general, preventive firebreaks were not used to prot t fire ed to ect forests. The Fire Action recommended that p ding ecosystem is also important: if the adjacent flammable or valuable, backfires need to be wider. The humidity of slashe in e, ke that. My father was poor as I am too, and every of his agricultural fields was reventive firebreaks would be required for any burning (agriculture or pasture). Peasants argued that backfires' width (i.e. how long the fire has to go against thewind before starting the main fire) varies in accordance with the level of danger of each situation. Backfire width, therefore, varies according to the vegetal material to be burned. For example, slashed primary forest needs wider backfires than a young secondary forest. The kind of surroun ecosystem is very d material is also a factor: when burning after rain, or fields located in lowlands or close to rivers, backfires can be shorter. Peasants said they had learned about backfirestheir homelands, from their parents, as explained by Mr. Fontes, born in Maranho statNortheast Brazil. As shown by the Mr. Fontes' speech below, firebreaks were known by him. The newness was applying it in Par state: Question (Q.): Where had you seen or learned it [fire management]? Mr. Fontes (Mr. F.): There where I grown up; there it [fire management] was lisurrounded by firebreaks.

PAGE 182

169 Q.: Where did it happen? Mr. F.: In Maranho. Every agricultural field was surrounded by firebreaks (...).border with the forest: the firebreak. Then when the main fire [pro-wind] was side. They [firebreak and main fire] would reach each other far from the border. On the burning day, he would start the burning in the firebreak located by the coming from there, before it could reach this side, the firebreak had burned this Thus, this I learned when I was a boy, working with my father. Mr. Fontes, So Francisco do Itacainas community, born in Maranho state, living that u ing ate in re management in 1998 was the high rate oes Caused by Fire' s in the community since 1983. In the So Francisco community, in 1998, 79% (n=34) of the families studied used backfires. All the eight families that did not use backfires were not involved with the Fire Group, living far from the village. Among them, one had losses caused by his fireand half of them used backfires in the following year. In 1999, the number of families sed backfires increased to 88% (n=34). Of the four families that did not use firebreaks, one had losses caused by their own fire. This family did not invite neighbors,and only two persons burned the field. Those examples of losses due to not followcommunity rules were used by Fire Group members to convince others to participthe community fire management, as discussed on section of Peasants' Evaluation. In theCupu community, in 1998, 82% (n=28) of the families interviewed said that they used firebreaks. In 1999, this percent dropped to 61% (n=23). As in the firebreak case, it seems that the high rate of families that reported they had used backfire in 1998 might be connected with their fear of the Fire Action's connection with IBAMA. After one year working building trust in the community, the rate reported might be closer to the families'reality. Another indication of not very efficient fi f losses caused by fire in this community, as discussed before in 'Loss ection.

PAGE 183

170 After Burning returnescape same day, and each of the three following days. The recommendation was to call for support (if needed), and to use control practices. Evaluation of this practice was related to que and s. t s. isco do Itacainas and Cupu communities, respectively. A chainsaw is another very useful type of equipment to control fires, when quick actions are needed, The Fire Action recommendation for fire management after the burning was to ing to the burned field to observe if any fire had escaped or had the potential to e (like burning trunks, stumps, or palm). It was suggested to return at the end of th stions about control of fires, since people were more likely to remember when an accidental fire happened, and control techniques used. The main control techniques measured through interviews involved the use of water, access to equipment, and varrida or backfire for control. Families have many limitations on equipment for fire prevention and control, no safety protection at all,no external support on fire fighting actions. Peasants usually control a fire with their agricultural tools (terado or long knife, scythe and ax), backpack-sprayers, and bucketThe absence of external help was due to communities' distant location, bad quality of roads and bridges, no means of communication, and limited firefighter actions. In these conditions, the Fire Action stressed, in its booklet and meetings, controlling fire when iis beginning, since small fires control is more feasible and less dangerous. Regarding equipment, no family possessed any safety equipment for those who go to burnings. Nor did they have equipment to control fires, besides their agricultural toolThere were some backpack-sprayers used to spray chemicals on plants, which could be used with water (mainly on localized fire in trunks). Families involved with loans fromthe FNO (Federal loan) usually possess this equipment: 47% and 42% of the families in the So Franc

PAGE 184

171 helping to make control firebreaks, a trees and trunks. In the So Francy, and fin In rted the fire in an agricultural fieldwe were around ten peopleand the fire jumped to the pasture. we beat on the fire with branches. And I ran to the other side, and the fire where I We gave up because we were almost being burned. I lost my eyebrow; it burned nd to cut burning isco, nine families (24%) reported having chainsaws, and in the Cupu communit10 families (32%). Families' first control technique used was control backfires, used in all controlling situations. A detailed description of this technique is presented in the section on Fire Strategies. Water is harder to use due to distance from water sources, lack of adequate containers and sources of transport. In the So Francisco community, seven families (29%) used water to control fires in 1998: two of them had losses caused by those fires, ve controlled fires efficiently (two of those five were from the Fire Group). In 1999, eleven families (32%) used water to control fires, and two of them had losses. Ithe Cupu community, four of the interviewed families (15%) used water to control fires in 1998, and all of them suffered losses because they were not able to control the fire. 1999, only two families (9%) needed to use water and none of them suffered losses. Mr. Duarte, from the So Francisco do Itacainas community, described an experience trying to control a fire during the early dry season, and how fast fire spreads and gets strong and dangerous. Mr. Duarte: Once we went to extinguish a fire at Jose's. They had sta It happened before the rain. The grass was high, very dry, and the fire grew fast; was had grown again. Using water, extinguishing, but our control did not work. due to the high heat. (...) But if the fire is small, it is easy to control. Mr. Duarte, So Francisco do Itacainas community. Before the Fire Action, there was a general belief that nobody could control fire, that fire was uncontrollable. The awareness that the burner is also responsible for its consequences (that it was not "fire's fault") brought a new scenario for community

PAGE 185

172 organization. Families were encouraged to help each other, especially if a Fire Growas created. Examples of success in controlling fire, or at least trying to, brought awareness of fire pr up evention and control and new behavior to families: those that became used or ve. rs. s for agricure during, and after burning. The second column presents the prActio, and fi for firmmunication for fire use" was k nings. to being more careful with fire use felt an incentive to do so, and those not used to caring about their fire consequences felt pressure to do so. For the latter, usually not involved with community organizations, the possibility of being charged by neighbors flosses caused by their fire, and even to be reported to IBAMA, were their main incentiThose engaged in community organization linked responsible fire management with living in community: helping each other, and being careful not to cause losses to otheSummary of Practices Evaluated Table 4-2 summarizes some of the main practices discussed, and points out changerelated to the FATA/LASAT Fire Action. Table 4-2 shows that seven of the practices used during fire management were familiar from use in peasants' productive system, lture implementation as well as for pasture maintenance. Those practices wediscussed during meetings, workshops, and Fire Action booklet. The first column categorizes the practices before, actices, the third explains how the practice was generally used before the Fire n, then the fourth column summarizes changes in that practice after the Fire Actionnally the last column discusses if the practice was known by peasants as a concepte management, or if it was an innovation. In the phase "before the burning day," the practice "co nown as a concept, but mainly associated to kin and to neighbors (if they got alongwell). The Fire Action promoted communication and collective work for fire management, and its results were observed in a higher number of people during bur

PAGE 186

173 In So Francisco do Itacainas, the number of people during burnings was higher than iCupu, probably due to earlier engagement to the Fire Action, organized Fire Groupdiscussion of fire management in community assemblies, and definition of local rules. The second practice presented in the Table, 'use of preventive firebreak', prior to the FireAction was used to protect material goods and direct investments mainly. After the FiAction, forested areas were also protected by this practice, resulting in less forested areas burned. Another newness relating to this practice was the use of far-away firebreaks (located inside forests), known by some families but never applied in Par state. In the second burning phase (on the burning day), the practice "burning in groups" n s, re was promoted as the grounted by the STR (Rural Worker get ive ir acceptance over the long term will depen d-line by the Fire Action, and suppor ers Union) in case of conflicts with IBAMA regarding fire use. Number of people during burnings, fire group's organization, and success in fire control were signs of bettcommunity organization for fire management. Peasants worked collectively for differentactivities, but the Fire Action's motivation, associated with the fear of IBAMA's repressive actions, encouraged people that were not used to working collectively toinvolved with community fire management. The practices "burning after the rain, later inthe dry season, on less windy times", were more sensitive than others. While collectwork would imply changes in social behavior, and use of firebreaks in extra work, they do not affect directly the burning quality, as much as rain, and time affect it. Therefore, changes in those practices were observed, but the d on burning qualities obtained and consequent final crop production. If familiesexperience bad production due to bad burning quality, they are less likely to incorporate those practices in their management.

PAGE 187

174 2. inct eA/LASATretion twud mmtin abgiBriamazonia. PRTIoreT/LASATngfter Fire Action ason fore 4Ma praicesvaluated, and possible changes due in part to the FTAcounies iMar reon, aziln AACCE Bef FAA Fire Action Chaes A mmcatifor Fire Use Geral cusn ovfire andwear foastiamokin d neigors. Mrity famis comuniire pns; Comonvitatn to ghbire oup. e ofevene ebre d tootecater gooand rect invementsuch hous, sil, fens, pereial cps, a paste witanim) Ud alws, to protect materooddirecnvesentfoareas; Ineaseiscuon oindsffecenesf firreaknnoirebk foorest rninn Grps tirdo Fo) Bningrganed ang kmai; Cmo onenglersonurnifielInasedumpele inurns; Ine Sranco comunieasts orgizedire Gups, encamiand ghbo to bn tog After the r in n leses ter iurniis d berait the ginn of y seaso, cloe to noon, in or toxime burng qity ( bettcropoduon prea) gr thein toumidurrodingosymsan(exceion fcropn wowlan oyeard falw, oralhaBuing r 2 3 p. evene ogo) trol, to proct mterial gooand ect iestmts. Us for venn, ate bef a ning ntro foontrl, to proect teriagooand ect iestmts. Us as bore, t also pro g theurinnd Ually ithtind to serve fire til mn slaed vtatioburnnorneurinthe fowidaysobse possible escapedres. Atast bning wnerntinbse until maine finhes; eturs late on the dand ollong d. Fi Acn, ihe to stiedKnown a Cceptr FiManagement? ing and n uunicate in specsiationsuch protinguseother ods. ted d ive parThewness was using itorresnd fst fbrealoted f frohe border of the bing;is ki of fbreawas kwn bsomamilibut ver aied Par e ade ageer. parThee was kishipcectivworkr esificsiationthe wness was always onizi gro (kiip anhbo, and STR suport bers ganin gups. se s an; n part. Befo the tione gowas burn bforee r; burnen o in uannsituns. e nness w planng twait fothe n bre bning ning parThewness was using ifprevtion. ct parThe veltyas ug it tprotect forestedeas,d orpple a mooftenan bore. s to o, itas anovon. bodebse firbefo Counion endissioer use therecng ng anhbajoof liemcatflam inioneiors/orFGr Ipart. Kin and close friends sed to commial tus, astec hos r ogo B E F O R E F I R E Us PrtivFirakUse prt mialds dists ( asseoscennrondurh alsseayialgs, t itms and rescrd dssin k anetivs oebs; ivatfrear f protection.Int. net potect fot, aoreirek caarm turn thndirek noy e fes nepplin Bug iou(Muo ogurs oizmoin nlyomnly si pe bng ds. cre nber of op thbing tho Fiscmtypanan Fromclear rules for burnings, andourd fly neirsurethInt. rn olle fopec tus; nergangupsnshnd eigrs)p tournorzed iro BurningRain, latedryseason, os windy timBetf bng onefore n, abeingdrnsrde maizniualforer prctier a. Waitin fo ra hifysun ecste and ld uptor s iet lndd one ollo pda)rnafteor m. IreAc, thmainal toe thainnig after th rainlynpled atio Thewasnio r raiefour. BURNING DAY Use of PrtivBackfires(Contra-f Usedfor conteads dirnvenedpretio thgin obur; Int. net or en Use of Col Backfires(Varrida) Usedr cotmal ds dirnvenedefbuo tteforested areas. Int. no wsino ar anwith me eondre thef A F T E R Observin Burning dg aafter Main Fire suneer conueobe thunaishegen ed, r retud dg ollng to erv fi leuro coueorve firisowner rne samay fwiaysN wn inatiNoy claimd to oerves re. Table

PAGE 188

The "use of preventive backfire" also affects crop production, but on a limited strip nd. Before the Fire Action, it was used to efter the main fire" was the only practice that all families studied said that did noc be ge, alt to contrSo far, practices for fire minant for succel fire prevention and control. Peasants' use of practices varied tiscussion among participuncontrolled fires-a prshscussion is, primarily, based ethod that allowcher to witness all the practices used and ner, six bituatio conc strategies for fire management were monitored during the 2000 nned to mhe tcnitiin fe in the field, only the So Francisco community was ted.nsed observing the fire group's ning, mainly for control, aimwas using backfires for prevencontrol backfire" remfor forest protection too, besides goods and diduring and atheyafter they had caused damimproved initial escaped fire detectioare key for fire mdetermaccordbalancing burning quality and prevention of called "fire strategy." This diburning events, a madapted in pDobserved. Tdry season. As explained before, it was plato ravisi of the burned field that is burned against the wi ing to protect mtion, and to protect forested ained very sim aterial goods and direct investmilar to before the Fire Action, but it began to be used rect investm nts. The newness areas. The practice "use of ents. "Observing the burning t pra ticeanagem foreaent, peasant adaptive the Fire Action. nd their conditions were difficuanage Before, escaped fires were detected only tegies nt had been analyzed. Although practices used variations on practices used seem ol. Observa tion ed to be n, and efficiency in its control. Fire Strame ssfu of 2000 ing uri o each situation, after analysis, and dractice. g Octob ants, always whictions w ocestrol a here is ere on participant observation of ns and two ed the researurn ing s he peasants'ency and lim onitor t wo om mu es but, due requ ited tim M on ito ri ng bur ing im pli plan an d d oi ng the 175

PAGE 189

176 burniis ty ices on burning quality, previous experience, social rbors, cultural variations for practi ation (Figure 4-12), a good example, two neighbors planned to burn togethgure ng, at same time recording, taking pictures and notes. The researcher's previous experience monitoring burnings helped in following the peasants' velocity, trying not to delay or perturb their practices. Peasants' comments about how fast the researcher operated indicated minimal impacts of researcher presence. The results presented in thsection are based on this monitoring but also on some described burning situations, whereobservations were done at posteriori, and on interviews. Each burning situation was analyzed, varying as a function of physical characteristics: field size, dryness, kind of slashed vegetation, relief, weather, availabiliof labor and equipment, access to water, surrounding ecosystems, distance from houses, and others. In addition, cultural values were considered, such as: cho elation with neigh ces used (such as kind of firebreaks), and level of family awareness. The final decision on strategy chosen was taken by the fire responsible (or fire 'owner'). Preventivefirebreak was also the duty of the owner. In order to present more information on fire strategies, this section presents threeexamples of fire management. The first situation describes a planned burning, with FireGroup coordination (monitored), the second without the Fire Group planning it (from interviews), and the third a fire control situation (monitored). In the first situ er, Mr. Machado, one of the Fire Group coordinators, and Mr. Fontes. The Fi4-12A (upper figure) represents the burning situation when it was starting. From Machado's side, the field was composed of slashed secondary forest, surrounded by secondary forest on two sides, pasture another, and his neighbor's land on the other. He

PAGE 190

177 decided firebreaks were not needed, because the slashed vegetation was secondary foreit had rained, and the main wind power was in the opposite direction. From the Fontes' side, they prepared two far-away firebreaks in the primary forest, around 10 meters the field border, around 1 m width. One firebreak was made by the field border and the secondary forest, with around 2 m width. They invited two other neighbors to help, plus a member of Mr. Fontes' family, putting together five me st, from n. They prepared some contade, and d that r all Figure 4-12B). While starting the burning, and obserd ber one hour later, and detec iners with water. They divided the group in three: two from the Fonts' sithree from Machado's side. They started burning at 2 p.m., from the opposite side from the main wind direction (preventive backfire), two men taking different directions. Whenthe other three saw the smoke, they started the fire from the two extremes of the field, starting a flame each 1 to 2 meters, until they reached the other fire. They explainethe fire has to go together, as a big flame, so it can spread faster, burning better. Aftethe borders were in flames, three stayed observing from the Machado's side, and two walking through the far-away firebreaks ( ving the main fire, peasants keep yelling. They explained that yelling was their custom for calling the wind. I stayed observing by the border between the field and pasture until the main fire was set, and walked through the far-away firebreaks. Aroun4 p.m., they dispersed. Mr. Fontes returned with a family mem ted the beginning of an accidental fire. The two prepared backfires, and were able to control the fire. They returned to observe the following day, but no more accidental fire happened. They believe that waiting for the rain was key for controlling the accidental fire. The far-away firebreak also prevented accidental fires in one place.

PAGE 191

178 Figure 4-12. Conjoint burning situation of two neighbors, Mr. Machado regional Fire y were able to control it, and the burning was successiful (Drawing by author). The second example was Onorio's family, one of the two who suffered losses in 1999, and was cited as a bad example for the Fire Group, and a lesson for families about how planning and collective work is needed for fire management. He was the first to Group's coordinator, and Mr. Fontes. Although an accidental fire started, the

PAGE 192

179 burn in 1999, in September 5th, together with his neighbor Mr. Leao. Mr. Leao lives in the city, and hired a landless family to take care of his lands. The two neighbors agreed to burn together, and invited a third neighbor, but did not invite the Fire Group. Although not living in the community, Mr. Leao oriented his employee, Mrs. Joao, to follow community rules for fire management. Both fields were composed of slashed secondary forest. Four people started the burning, after a small rain only, because part of the slashed vegetation was located on lowland. They started the burning at 2 p.m. The area to burn was relatively big, 3.5 ha of Onorio's, contiguous to 4.5 ha of Mr. Leao. They did not use preventive backfire, starting with the main wind power, and the fire escaped to another neighbor's land, burning banana trees, and a slashed secondary forest not ready to burn. Mr. Onorio used control backfire, but the fire spread too fast, getting out of control. When fire started to get uncontrolled, someone called the Fire Group, and at the end a total of 15 people helped to combat the fire, using control backfires, water, branches, and a chainsaw. Although they tried to control the fire, it burned Onorio's house, barn, and some fruit trees located by the house. This burning was used by Fire Group members as a bad example to alert other peasants about better planning for fire management. The third example referred to a fire control situation. A Fire Group member, Mr. Ambrosio and his neighbor, Mr. Celestino were crossing Mr. Celestino's land to talk with another nedirection of his slashed secondary forest, not ready to burn yet. Using their scythes, in 30 minutes they prepared a control firebreak (varrida), and set a backfire. They stayed 30 minutes monitoring, but no more fire spread (Figure 4-13). ighbor, when they saw an uncontrolled fire spreading in his pasture, in the

PAGE 193

180 xample of a successiful use of control firebreak. The above left picture ws the accidental fire coming from lefto to right, in direction to a slashed Observing burnings was a very rich experience. There are things that cannot be detected by interviews, and experiencing their reality is key for technicians that want to be partners with peasants, so peasants' rationality can became clearer to them. The burning process requires a series of judgments and choices depending on the biophysical Figure 4-13. Esho field not ready to burn, and the two peasants making a firebreak. The middleleft picture shows the peasant 'brooming' the firebreak with branches, in orderto take off any fuel from the firebreak. The below left picture and the above left show the peasants starting the backfire. The below right shows how the fire stopped at the firebreak, marking the success of the control technique.

PAGE 194

181 and social conditions of the moment, personnel involved and the burning process itself. Experiencing it from the peasant's point of view would help technical people to appreciate their knowledge, experience, and ability to adapt and innovate in fire management. Besides rationality, there are customs, folklore, yelling "to call" the wind, or moon phases, numerology, among others, that should be respected. Soares, an anthropologist who worked with Nambiquara indigenous in Central Brazil (Mato Grosso) found the recorded colonists 'calling the wind' to be very similar to the way indigenous people yell during burnings, also to call the wind (Neila Soares 2002, personal communication). Peasants' Evaluation of Government and Community Fire programs IT IS FOGO (means "fire" but also "awful" or "out of control") Poetry written by Mr. Adaildo during one of the FATA/LASAT Fire Action workshops, in December 1998. If the citizens do not participate in decisions, it is fogo. When the poor live under oppression, it is fogo. To live together in a community nowadays is fogo. If it is hard to achieve a more dignified life, that is also fogo. BankwronThe ncommWhe credit for planting often never comes, and when it does come, the timing is g. That is fogo. egligence of those that govern in terms of health care, education and unity infrastructure, this is fogo. n over and over again peasants work to produce but cannot sell their crops because the roads are poor, it is also fogo. When IBAMA applies the law only to peasants, and only they are punished, it is fogo. To be an active participant in popular movements, that is SO fogo. If burnings are disorganized and uncontrolled and lead to losses, it is fogo.

PAGE 195

182 When the workshop is for discussing fire, yet other things are discussed instead, that is fogo. If politicians lie, mislead, and cheat those with less information, this is realWhen ex ly fogo. perts do not deal with the practical and instead focus on theory, When a burning is started and it gets out of control, we have to find a way to also be extinguished with a little bit of love, or who knows, perhaps when the poor or all. Thi heard. This srnment and FATAeen as a positive thing, becaurom accidental fires, whose perpes considered to be the "fire's fault.e management practibehavior. The disagreement with the government was the way actions had been taken, especs on fire use, difficulties in getting a fire permi), the fe peasants, the absence of government representatives in the communities, the big emphasis on enforcement, and so little meaningful education. oh how this is fogo. extinguish this fire, and it will need lots of water or a huge extinguisher, but it can are valued. The poetry above summarized many of the peasants' evaluations reported in this research. It was written by a peasant during a Fire Action seminar, and read out loud f s poetry is another example of peasants' creativity, when their voice can be ection aims to present the peasants' main evaluations of the gove /LASAT actions. Government enforcement of fire control was generally s se of the losses peasants had suffered in the past f trators had not previously been blamed because it wa The study found peasants to be motivated to change their fir ces due to fear of IBAMA sanctions, a negative incentive that clearly affected ially: government periodical prohibition t, recommendations out of touch with peasants' realities (firebreaks, time to burn e charged to deforest and burn, the absence of adequate equipment and support for

PAGE 196

183 Those government actions end up causing disgust, like in the Cupu community, in 2000, when a peasant said "Because of my labor nobody takes me into jail! I already have me foistrust the gotions, seeing laws as class discrimination: "those governmental fire laWho really follows the law are the peasants, because they are afraid of punishments" (Mr. Rs, y would mean work against peasants. Peasants were more positiilies tions. ing it was not true was an more than 50% deforested, and I make my living from agriculture. Who will give od? It is the agriculture!" (Mr. Gomes 2000). In addition, some peasants d vernment's real inten ws are only for peasants; larger farmers start their fires and say it was accidental. eis 2000). This mistrust is promoted by the behavior of IBAMA's representativerarely present to dialogue, and to clarify doubts. Helicopters flying away are the closer presence, from which people hide fearing repression. Fire is a sensitive topic since it directly affects families' production, and workagainst fire use ultimatel ve about the Fire Action with the FATA, LASAT and STRs, grass-roots organizations that have historical relations of trust with peasants in the region. Famsaw the meetings at the community level to be essential, in order to directly confront thetechnician with things, helping to build trust about the Fire Action and leaders' intenAs a peasant at the Cupu community expressed it, when asked about the difference between technicians' presence in the communities, and messages disseminated through the radio: "For sure meetings are better than radio; in a meeting, messages are clear, challenging, stronger. I respect you. I do not respect my radio, if it says what I do not want to hear, I turn it off." (Mr. Ribeiro 1999) Another pertinent issue was receiving a salary for working with fire management. Many thought that leaders were receivpayment for working 'against' them, and convincing them that

PAGE 197

184 aware d ilies that suffered losses decret e ata e ere ness raising process. The interest was not in finding a job with the government, butcontrolling fire use in order to prevent losses caused by uncontrolled fires. Regrets about the Fire Action were expressed by some families whose burning fieldid not burn well due to rains. In addition, many said that they wished the Action had started ten years ago, so their forested areas would be in much better shape. Good consequences listed were that more people working together made it easier to control fire(higher efficiency and safety), and to protect forests, with less losses to all. Discussion The Fire Action introduced important changes in Marab colonist peasants' strategies on fire management. Still, the number of fam ased considerably in both communities after the Fire Action. These data suggesthat the Fire Action was a positive influence on fire management, although longer timstudies are necessary to capture families' lasting appropriation of Fire Action recommendations and approach. The main practices developed with the Fire Action and adopted by the families were community organization for burning, a higher number of people working together to prevent and control fires, waiting for the rain before burning, and timing to burn. Dshowed that those practices were key for prevention and control, and improved from oneyear to the next, within both communities. Other key practices adapted and used werfirebreaks, backfires, observing the burning, returning to the burning after the main fire was over, and control techniques. Beyond these isolated practices, peasants showed themselves to be capable of implementing complex fire management strategies that wadapted to the particulars of each situation.

PAGE 198

185 Many families reported that before the Fire Action, burnings usually took place in August and around 12 o'clock noon. The weather then is usually hot and w indy, and by burnit later s. Many families wished the Action had started so their forested areas would be in muchys. The ly involved the action too. This methodology supported peasants to approore ng during that period, families maximized their production and ignored any recommendations aimed at the prevention of uncontrolled fires. After the Fire Action, they were more likely to burn in September or October, after the rain, and somewhain the day when the wind might be lower. Precautions were aimed at preventing uncontrolled fires to protect forests, as well as agricultural fields and pasture ten years ago better shape. The practices, in general, were known by peasants, but used in different waFire Action's big achievement was to give confidence to peasants for them to organize and adapt their known practices, and discuss among them ways to improve fire management in the community. By doing this, peasants that were progressivein the Fire Action become part of priate the Action, and empowered local organizations. Data revealed a generally higher level of consciousness favoring change in practices of fire management in the So Francisco community. This was reflected in local rules discussed in assemblies, fire groups implementation, and appropriation of the Fire Action discourse when referring to local leaders as the main coordinators of the Action. Due to the higher level of community organization for fire management observed in So Francisco do Itacainas community (Fire Groups, local rules), it is mlikely that the fire management strategies adopted by families will persist.

PAGE 199

186 Local leaders played a key role in the Fire Action in both communities. In the So Francisco do Itacainas, Mr. Valdir Silva, a Fire Group member, was one of the key community members present in all FATA's workshops, and was also the communassociation's director. He was one of the most cited among interviewees' families as a reference for information on community fire management. He had been a representative at the STR (Rural Worker Union), one of the local association founders, and had aof commitment, struggling for community well-being (such as schools, teachers' wageroads, electric power, health). In Cupu, Mr. Soares, also had a history of commitment to the community, was ity history the church coordinator, and was a key peasant leading the Fire Actiocontext ed the community unite for a comrcive e in fear tinue and to improve their fire strategies. n in the community. Fire management was connected with individual family strategies of natural resources management. What and how to produce takes place inside the family of individual choice. One should not go to a family's house and tell them how they should work. It is offensive to family dignity, insinuating that they are not smart enough to do their job. At this point, however, government laws help mon goal: to burn their field without causing losses to others. IBAMA's coelaws were sometimes used as a threat to convince those who usually did not participatcommunity work. Local associations (So Francisco community), FATA and STRs would only support peasants against IBAMA penalties if the family were organized intheir communities, burning according to community rules. At the beginning, theirpromoted a higher degree of community organization, involving all the community. The positive results of community fire management encouraged families to con

PAGE 200

187 In order for the government to be seen as a partner, some barriers constructed throughout peasants' history should be broken down. As discussed throughout this stpeasants complained at different times about government absence in peasant communities: "we are forgotten here", said a peasant. Colonists, through their labor, have made their land reform, built their houses, cropped their lands, built roads, schools, churches, soccer fields and formal associations, and faced health difficulties. Without changing this image of neglect, the same government implemented top-down actions to control and prohibit fire use, and charged fees for burning. This evaluation showed that colonist families are ready to change their fire practices, if supported by trusted organizations, through an a udy, pproach that respects and promotes their local knowledge.

PAGE 201

CHAPTER 5 CONCLUSIONS This study discussed fire use by colonist peasants in the Marab region, Par state, Brazilian Amazonia. It aimed to answer three main questions: 1) Why aren't governmefire actions and laws more effective in diminishing uncontrolled fires? 2) Do colonist peasants have and use empirical knowledge to manage fire? and 3) Is a participatory approach to fire management effective in diminishing fire losses? Those questiondeveloped by analyzing government fire laws and narratives, discussing a participatory nt s were fire action coordinated by grass-roots institutions, and evaluating impacts of this action in two colonist peasants communities. This Chapter presents this study's main conclusions directly related to the above questions, discusses contributions and limitations of the study, and presents some suggestions for future fire actions in Amazonia. Government Approach The study found that the governments top-down approach to controlling fire use in Amazonia is not likely to succeed, and potentially might lead to peasants' revolt against government measures, weakening possibilities for partnerships. This is because: a) The government has a long history of neglect and mistrust with colonist peasants, especially at community level; b) Government educative programs, supported by the media, are based on anti-fire messages, in a unilateral approach ('banking' education) developed for peasants, but not with them, in general resulting on fire use recommendations not appropriate for 188

PAGE 202

189 Amazonian peasants' conditions -ecological (e.g., amount of forest biomass), economic (e.g., absence of adequate safety and combat equipment), and cultural (fire use not perceived as a crime); and c) Government bureaucratic r permits, recommendation of inappk ), they have always made decisions on their natural resous-r their ist peasants have a high level of awareness of their current resouhe implications for future geners that the wt as wet consciousness for better fire use. Better fire mana equirements for fire ropriated practices, and periodical fire use prohibition, based on enforcement, exclude peasants from being able to burn in accordance with fire laws. Peasants therefore "became" transgressors by virtue of working on their productive systems, worwhich is a strong basis for cultural pride. Colonist Empirical Knowledge This study showed that colonist peasants in the Marab region want better natural resource management in their lands. In their historical struggle for land and living conditions (individual and collective rces management, and assumed responsibility for those choices, supported by grasroots institutions. Due to their previous experience as landless, and the struggle foland and for social reproduction, in general colon rces' depletion, as well as t ations. At the community level, families suffer from fire-related losses caused by uncontrolled fires, and they want to manage fire better. Some peasants' perception eather is getting hotter, dry seasons getting stronger, and that the forest is noas it was ten years ago, contribute to their gement in the peasants' context means using fire only where it is necessary in their productive system, and preventing and controlling escaped fires.

PAGE 203

190 Colonist peasants have valuable empirical knowledge for fire management, developed through its use as their main agricultural tool for crop production and pasmaintenance. They learned ture how to use fire in their homelands, and adapted their knowhe re tion and control of escaped fires. This study found that ociated with Nature. Peasaefs for d their ritual for calling the rain. ledge to their actual conditions. For peasants, fire management includes both assuring good burning quality and avoiding losses caused by uncontrolled fires. At tsame time that fire is useful, originated by peasants, they are also susceptible to fihazards. Therefore, peasants empirical knowledge on fire management intrinsically associates fire for production, with preven in peasant culture, God is often ass nts believed that 'only God knows', but that humans possess their physical abilities and knowledge to manage nature, and they are expected to use their knowledge to manage natural resources. Fire acts mediate nature and culture, making it possible to transform nature into agriculture, to feed God's sons and daughters through their work. Colonist peasants' knowledge is expressed by their cultural understandings on weather foreseeing, where families interpret stars and observe animals behavior; their belichoosing the burning day, based on moon phases, numerology, and the daily weatherconditions; an Colonists' knowledge is also present in their fire practices and strategies forprevention and combat of uncontrolled fires, as described in detail in Chapter 4. Adaptedfirebreaks when a burning field neighbors forest (firebreak located far away, inside the forest), burning after the rain, and use of control firebreaks, are some of their practices, that became more effective with community organization.

PAGE 204

191 Participatory Approach The community fire management action presented was coordinated by trusted grass-roots institutions (FATA, STR, LASAT, COOCAT, and local association) aexperienced technician, with regional (several communities) and local (community) meetings and burning monitoring. The emphasis was on learning-cycles at the regional level, and collective work and local agreement at the local level, and peasants got involved voluntarily. The number of families that suffered losses decreased considerain both communities after the Fire Action. The Fire Action adopted the Frei nd an bly rian Pedagch, d l conflict with IBAMA. fire use. This thesis discussed that families' local willingness to avoid fire-related losses, associated with their need for fire in their productive systems, motivated families to ogy, based on a dialogic approach in which a technician was facilitator of the problem-posing situation of fire management. Peasants' empirical knowledge was codified through relating colonist productive steps as a function of the burning day (before, on the burning day, and after), and 'burning dynamic' exercises. In this approapeasants were challenged to critically analyze their own situation, and to find their own solutions to identify problems, and set their local rules and strategies. The Fire Action empowered colonist peasants to develop and implement their own rules and practices for fire management, tailored to their own culture and resources, anuse these to support their actions in any potentia General Conclusions Peasants' motivation for better fire management was mainly a combination of three factors: a) local willingness to avoid fire-related losses, b) influence of a participatory Fire Action coordinated by trusted leaders and institutions, and an experienced technician, and c) government coercive pressure for control

PAGE 205

192 engage to the Fire Action, led bworked on a voluntary basis, suppore ve used e eek more efficient fire mThis study aimed to presents history and set of values, beliefnatural ive actions (impl y local trusted leaders who rted by grass-roots institutions. The Fire Action encouraged and supported moactive peasants to develop their own rules, empowering them. The success of fire management motivated more passive peasants to engage in the action too, taking actiroles and appropriating the changes in fire management into their practices. The government pressure for fire use control and prohibition induced those families notto engage in collective actions to get involved, since in cases of conflicts with IBAMA, the Fire Action institutions would support only families involved with the ActionWhile the fear of IBAMA repressive actions provided a strong motivation for peasants to change their burning practices, there was little accessible and appropriatinformation or resources to support changes. People in both communities wanted toknow what to do in order to use fire in the official way, and the lack of support generated strong critiques to government actions. The Fire Action brought information and motivated peasants to s anagement strategies, encouraging families to seek their own solutions, resulting in fewer losses caused by uncontrolled fire in the two communities evaluated. Contributions of this Study the unique and preciou s, and practices of colonist families in Amazonia, by analyzing their voices, through incorporated narratives, and by showing that they do have a clear culture that deserves respect. In addition, they are the ones who ultimately take the final decision on resources management, and have to be actively involved in collaborat ying decision-making power).

PAGE 206

193 By evaluating families' forms of organization, communication, and compliance with fire management recommendations developed in the participatory Fire Action, thisthesis also makes a unique contribution to our understanding of the impacts of fire management programs on colonist peasants use of fire. Despite the richness of unique information on colonist peasants' knowledgeevaluation on fire management, this study is based on short-term evaluation, carried out in two communities only. There is no information on how dependent the Fire Action results were on external technical support, since no evaluation was done after the F and ire Actio on ship management should include other key social actorsunicate with them, in a dialogic approach, ieory with practice, and techn n was over. This study can formulate stronger conclusions if future monitoringthe communities impacted by the FATA/LASAT Fire Action is carried out. The Fire Action focused on the community level, and was not successful in developing partnerwith government agencies, or policy makers, which limited the actions impacts. Future actions for community fire such as environmental agencies, in order to reach regional changes on fire policies, contributing to more adapted and effective recommendations. Technicians aiming to work with peasants should be prepared to comm n order to combine together th ical with empirical knowledge.

PAGE 207

APPENDIX A GLOSSARY OF ACRONY MS AGRe EMAzilian Federal FNO aia nment Foundation) lhadores na Agricultura (Rural Workers Federation) FNO Fundo Constitucional do Norte (Constitutional Fund of the North) GETAT Grupo Executivo de Terras do Araguaia Tocantins (Executive Group for the Araguaia Tocantins Lands) GTA Grupo de Trabalho da Amaznia (Amazonia Work Group) IBAMA Instituto Brasileiro de Recursos Naturais Renovveis e do Meio Ambiente (Brazilian Institute for Renewable Natural Resources and the Environment) AF Associao Agro-Ambiental de Atividades de Agricultura Familiar da Vila So Francisco do Itacainas (Agronomic-Environmental Association for Peasant Activities at the So Francisco do Itacainas Village) CEB Comunidade Eclesial de Base (Christian Base Community) CEMAN Centro de Monitoramento Ambiental (Environmental Monitoring Center) COOCAT Cooperativa Camponesa do Araguaia Tocantins (Peasants Co-operativof Tocantins Araguaia) CPT Comisso pastoral da Terra (Land Pastoral Commission) DFID UK Department for International Development TER Empresa de Assistncia Tcnica e Extenso Rural (Agency for Technical Support and Rural Extension) EMBRAPA Empresa Brasileira de Pesquisa Agropecuria (The BraAgency for Agriculture and Livestock Research) Fundo Nacional do Norte (National Fund for the North) FATA Fundao Agro-Ambiental do Tocantins Araguaia (Tocantins AraguAgrarian and EnviroFETAGRI Federao dos Traba 194

PAGE 208

195 IBGE Instituto Brasileiro de Geografia e Estatstica (the Government Agency for Geography and Statistics) INCRA Instituto Nacional de Colonizao e Reforma Agrria (National Institute for Colonization and Agrarian Reform) INPE Instituto Na cional de Pesquisas Espaciais (Br azilian Space Agency) s MA o do Meio Ambiente, Recursos Hdricos e Amaznia (Ministry I ary) EMMA a Municipal de Meio Ambiente (Environment Municipal n Rights) o Brasil (Pilot sts) O tais ) o LASAT Laboratrio Scio-Agronmico do Tocantins (Social-Agrarian TocantinLaboratory) Ministri M for the Environment, Water Resources and Amazonia) SEMAGRSecretaria Municipal de Agricultura (Agriculture Municipal Secret Secretari S Secretary) SDDH Sociedade de Defesa dos Direitos Humanos do Par (Par State Society for the Defense of Huma PG7 Programa Piloto para a Proteo das Florestas Tropicais d P Program for Protection of the Brazilian Rainfore PREVFOGSistema Nacional de Preveno e Combate aos Incndios Flores(National System for the Prevention and Combat of Forest Fires PROARCO Programa de Preveno e Combate aos Incndios Florestais no Arco dDesflorestamento (Program for Prevention and Control of Fires in theArc of Deforestation) T Partido dos Trabalhadores (Workers Party) P PAF Projeto Agro-Florestal (Perennial Crop Project) IPAM Sistema de Proteo da Amaznia (Amazonian Protection System) S TR Sindicato dos Trabalhadores Rurais (Rural Workers Union) S USAID Unites States Agency for International Development

PAGE 209

APPENDIX B PEASANT'S NARRATIVES Translations were made by the author. arrative from 1 Translated n Chapter 2, page 39: "O meio ambiente de um gabinete xtendam s nos nas ." as eu mesmo q abalhar certo, ni, que no tem estrada, no tem um lta, meotar uma rocinhero saber o q vesse, por exemm. Por Translated n 2, page 71: "Quando as pessoas trabalham tudo isso. A gente espera que A tambm faa a papel que a combater o fogo, e trabalhar. Da maneira do vai o daqui dotem quem m que tra nessa ponta de i de 8 mata sem erruda, a j num metro e meio de largura j fica fcil para a gente controlar." 5. Translated narrative from Chapter 3, page 92: Pergunta: "Algumas pessoas falam que os agricultores so analfabetos, e que no entendem as coisas." Sr. Gomes (G.): "No entendem. Quem que doido por aqui neste interior? No tem nenhum. Povo deficiente que tem aqui pouco. Alguns. Mas os outros, se no tm o muito pequeno, muito reduzido, n? E o meio ambiente da Amaznia, ele grande, ele extenso. Ento eu pediria a essas instncias, a estes rgos [do governo] que no se seus gabinetes, mas que faam as suas presena tambm e comunidades 2. Translated narrative from Chapter 2, page 52: "Hoje muitos j queimaram, mpelo menos no queimei e outros por a no queimaram ainda. Mas a gente estava sempre esperando, por exemplo, uma pessoa assim que viesse para explicar, por que a gente ue no entende da lei, mas a gente quer ver o assunto que a gente possa. Por que, numa regio dessa daqu tr socorro de nada, um pobre dum colono desse aqui se queimar uma roa e tiver que pagar uma mulhor pegar ele e enterrar logo de vez. () Ento, vamos supor, se no a para fazer, como que se pode passar? O que eu qu b seguinte: o queue ns tem que fazer pra queimar? Se ns tem que fazer aceiro, tirar licena? A eu quero que voc passe isso pra mim, pra eu ficar sabendo, n, por que seplo, vindo antes, acho que ningum aqui teria queimado sem orde ti que o grupo da gente aqui sempre combina um com o outro, n, pra fazer as coisas mais ou menos certa. S que ns no tivemos essa pessoa que viesse aqui antes". arrative from Chapter 3 num s sentido, no tem coisa difcil. E na nossa comunidadea pra frente as coisas se encaminhem desde quando o IBAM d ele tem que fazer. (...) Se o governo est cobrando de ns que prnto ele tem de mandar um equipamento bem adequado pra gent e que vai indo no vai poder defender l s com as mos." 4. Translated narrative from Chapter 2, page 74: "...porque esse fogo quanesm vento que ele joga fasca l, ou quando ele vai labareda, no m atalhe, neseja um aceiro largo, mas ele pula, joga fasca, e ele enata aquetros, ele vai pegando aquela folhinha j no cho, na m md ba 196

PAGE 210

197 saber da leitura, mas ele assina o nome. Todos tm entendimento. Isso j movimento, dona Katia, de quem j de dentro daquela comunidade que s trabalha no desprezo. Esse povo que a senhora falou uma comunidade, e forte! Esse povo que diz isso so dessa comunidade (...). Por isso que a gente desconfia de todos. A gente vive no desprezo. A gente no coisa do outro mundo no, ns somos tudo gente. Precisamos de amor e de respeito." P.: "O pessoal sabe ler as estrento, a terra..." Sr. G [educao formal], mas doido ele no Outra. E a com essas desfeitas. No assim nsou : 'No, nunca pensei no. Acho to bom pr c, silncio, no tem ou pr comer, que aparece. A voc pega sua ma dade?" o las, sabem ler as florestas, o ve .: "Ele [analfabeto] no tem essa culturas [culturas] de muita importncia ele tem no. Merece ser acolhido. brincadeira essa coisa, desprezo, desprezo (...). S Deus... sofrido. Pois dona Katia, falta muito responsabilidade, muito para poucos." 6. Translated narrative from Chapter 3, page 100: Pergunta (P): "A senhora j peem morar na rua ?" ra. Gomes (G.) S tormento de nada. Ainda mais na idade que a gente j est. Aqui em casa no tem murioca. Sair, Deus o livre, nem pensar." P.: "Tem muito que querem sair da rua e vir pr roa, no ?" Sra. G.: Mas tem muitos que no querem vir para a roa. Morrem de preciso, e no tem como a gente, feijo, fava, arroz perdido [por no conseguir colher] e tudo, mas no vm. tm muitas naturezas, no ? As naturezas no so iguais. A gente v mulher nova pedindo [na cidade]." P.: "Mulher at com nenm, no ?" Sra. G.: "No ? 'D uma ajudinha aqui, pr dar comida pr criana', ou 'um remdio', 'compre isso'. A gente v muito. Graas a Deus, durante todo esse tempo que sou ceganum pedi esmola. Deus tem me ajudado. Sempre s moro na roa mesmo. Na roa ningum no tem pena de dar nada a ningum, no ? Na rua no." P.: "Na roa uma troca, no ?" ra. G.: ", porque a gente cria um porco, mata, divide. Pega [caa], mata s S divide, no vende nenhum pedao." 7. Translated narrative from Chapter 3, page 101: Sr. Reis: Ento hoje o lugar sempre falo a para um bocado [de gente] o lugar bom hoje aquele onde voc est arrumando seu pozinho de cada dia. Ento voc tem que ficar. Voc no pode mais ficar correndo de um lugar para outro, caando lugar bom no (...). A pessoa no tem ada, no ? pobrezinho, com muito menino que o n terra, vende, vai para a cidade, ou vai caar lugar bom. A muitas vezes voc no arrumais nem um po. obrigado os outros estarem ajudando, fazendo vaquinha para a pessoa. Ento hoje onde a pessoa tem a famlia, e terrinha, e est arrumando seu po direto, com facilidade." Translated narrative from Chapter 3, page 101: Pergunta: "E se vai para a ci 8 Sr.: Moraes: "(...) Se ele partir para a capital, ele vai sofrer mais, ento o cara tem que segurar (...). Sai para a cidade e vi que no dava. Porque o salrio no dava para continuar. A eu voltei para a roa, para melhorar a situao(...). A cidade no est tendemprego para ningum, por isso que o governo tem que dar ajuda ao pessoal que mora

PAGE 211

198 na roa, que da roa onde sai de tudo para todo mundo, no ? Olha, eu vou falar: daqui mais uns tempos dessa mocidade sero poucos os que vo querer roa, porque aroa duro, no fcil, no todo mundo. E esse povinho que de hoje em diante vai nascer, para ir para a roa s se tiver uma coisa muito boa para eles, se no tiver evo. Eu acho que a situao vai ficando pior toda vida. Porque cada um quer procurar o estudo, a vai esquece les no ndo da roa, e onde sai de tudo pro cara comer, no ? O cara o de ia por ue ele j est ruim demais. Aqui mesmo na nossa comunidade tem o Joo. Trabalhava trabalha fica mal [sem alimento]. o que eu digo 'rapaz, a pessoa o frente, mesmo que voc no tenha dinheiro, a fava, temmandioca, tem ive from Chapter 3, page 105: Sra. Souza: "(...) Eu me entendi vaqueiro, gerente do meu tio. L tu olhavas assim, a ito: uma mata verde sso no, por isso acho bonito. (...) S o mato verde que com a natureza', o que vocs pensam?" ra. Souza: por que eu acho bonito deixar uma mata assim, to verdinha, no destruir. be a to em, ntar na roa, por que um jeito de eles se alimentarem. (...) Fica onito essa mata. Logo para l onde ns morvamos no tinha [mata], era feio, ningum ata trabalha no banco, ele trabalha em qualquer servio, mas tem que vir da roa comer." 9. Translated narrative from Chapter 3, page 102: Sr. Reis: "(...) A gente v muita gente trabalhando a mais em diria, em diria. O dia que ele no trabalha em dir q mais s de diria. E menino aumentando cada vez mais. A a pessoa chega trabalhando assim, o dia que no tem que trabalhar para a gente mesmo!' Se a pessoa tem uma terrinha, ento vamostrabalhar para a gente. Quando a pessoa trabalha um ano [primeiro ano], aperread[difcil], passando mal. Mas de um ano para mas voc tem muito arroz, tem muito feijo, tem o milho, tem tudo; o que comer tem. Agora trabalhando para os outros, a fica difcil. Voc est enricando s o outro, direto, est trabalhando para ele." 10. Translated narrat por gente nessa fazenda. Meu pai era tu nem vias mata verde, s aquelas matinhas velhas. O arroz, quando era no tempo de plantar, a gente plantava num varjo, tipo esse a de aaizal. Mas s que o aa l na Bahia no tem serventia para nada, ningum come. Ns viemos aprender a comer o aaqui. (...) Quando cheguei na Serra Norte, em 76 [Par], tudo era bon fechada. E l na Bahia eu no via i a gente via era um aaizal, por que l era bebedouro do Pergunta: "E quando se fala 'cuidado S Porque Deus deixou foi tudinho, para viver sossegado, a a pessoa vai e desmancha. Oaaizal cria os tucanos, cria vrios tipos de passarinhos. J uma obra da natureza. Saporqu? Tem roa que o pessoal bota que d uma praga de curica [tipo de pssaro], umpraga. J tem outros que botam [a roa], e no d. Aquele lugar que a curica atenta mui porque ali tinha uma rvore que dava o fruto para eles se alimentarem. Pode estudar, pode prestar ateno. por que tu no trabalhas com este tipo de coisa. Aonde tem um p de piriquiteira [rvore], que todo ano eles comem ali uma rvore tipo pitanga, das frutinha vermelhas bem miudinhas, a eles comem. No outro ano eles [os pssaros] vno a acham, eles vo ate b via mata. Era s aquelas rampas, sumindo...." 11. Translated narrative from Chapter 3, page 107: Pergunta (P.): "No caso, a mbom para qu?" Sr. Lopes (L.): "A mata fria [primria]? Ave Maria, bom demais!"

PAGE 212

199 P.: "Para que bom?" Sr. L.: "Por que Bom? Por que a gente no tendo a mata, seca tudo, sem a mata vira so agreste de capim, a gua se acaba, fica tudo seco, chama [atrai] at o vero. (...) Aquno Par chovia demais, devido s matas. Agora tudo pelado [desmatado], fica faltando tudo. E assim as [fontes de] guas, as grotas, secam tudo. se a vida toda guarnecida [protegida] de mata, ela fica a grota de um jeito s, bonita (...)." 12. Tran i slated narrative from Chapter 3, page 108: Pergunta: "Para que o fogo portante? Para que usa o fogo?" da] usa. 3. Translated narrative from Chapter 3, page 108: Pergunta: "Qual a importncia do ima a, sem ser arada..." no mexia s naquele dia ou no outro, que certeira. s aqui temos uma experincia, ns que moramos na beira da gua, que o baiano s utando o jacar [faz o som do jacar] esturrando dentro da gua, chuva chove, difcil. s im Sr. Duarte: "Acho que o fogo importante por que a gente usa ele em tocao [queimade roa, por exemplo. Em muitas partes o fogo importante a gente o usar. Ai, no sei nem como explicar, s sei que a gente o usa de muitos jeitos, num munturo [toco], num pasto, numa roa. Tudo isso importante a gente usar o fogo. A gente toca fogo numa panela [risos], na farinha [de mandioca] debaixo do forno; para caudar um tijolo, uma telha, em tudo a gente o usa. O que sei explicar s isso. De todos os jeitos a gente o[pausa] Agora [sobre] o fogo, o mais ruim quando ele vem do lado da gente. Quando chegam [dizendo] 'vamos atalhar [controlar] o fogo', a dose. Quando vem uma linguona de fogo da altura dessa casa, quando d um vento que ele deita rumo da gente, a que dose; est arriscado a se assar todinho." 1 fogo para vocs?" Sr. Almeida: "O fogo muito bom, s que ele devora muito, acaba com as vidas. Se a gente pudesse viver sem ele era bom." Filho mais velho: "Mas [o fogo] d mais facilidade para a gente. Quando a gente quevem menos mato, a terra fica mais limpa, melhor para trabalhar. Sem o fogo, s se for fazer uma linha de terra, tem que jogar o mato todinho fora para trabalhar, se arrebentar de jogar aquele mato fora. Depois tem que carregar os tocos. O toco, o fogo queima. Tem que tirar aquele mato todinho, e plantar naquela terra cru Sra. Almeida: "Mas se a gente pudesse ter o arado, para aradar a terra, a gente com ele [fogo] no." 14. Translated narrative from Chapter 3, page 112: Sr. Oliveira: "Olha, Ktia, no chamando a ns de animalzinho... Ele racional, ou racional somos ns?" Pergunta: "Diz-se que os racionais somos ns." Sr. O.: "O racional somos ns? E o outro o irracional? Pois bem, no comparando a ncom o animal irracional, mas eu lhe digo: os animais irracionais tm mais compreensso do que ns. Voc sabia disso? Por um lado! Ns temos muita inteligncia, mas conhecimento do que se passa, ns no temos nenhum, s Deus tem. E s vezes se v num animalzinho. Quando voc v um bocado de formiga se mudando do alto para o baixo, vero. Quando voc v um bocado de formiga no baixo se mudando com todos os seus filhotes para o alto, pode esperar chuva N vezes pega corda com a gente. Eu digo 'hoje o jacar est querendo se molhar. Mas se nsestamos aqui, esc na certa. raro demais o dia que o jacar esturra dentro da gua e no

PAGE 213

200 vezes ns temos a prete nso de dizer o que vai acontecer, por que o animalzinho est s." no para l da roa. Quando eu cheguei em casa estava a dear, ele tempo para c, que antes o fogo no ndava [na mata]e agora anda?" chover; era 15 dias, no mximo um ms. A chuva aqui era direto. A o eses sem chover; a resseca o mato. Na hora que nidade] de agosto, eu tocava fogo; deixava a roa limpinha, s esperando o mpo de chuva para eu plantar. E o pior o seguinte: muitas pessoas esto acostumadas e a pessoa que a roa sem o mostrando para a gente. Agora, uns prestam ateno, outros no." 15. Translated narrative from Chapter 4, page 117: "Como a gente pode viver na roa, e no poder fazer uma roa? Ns vamos controlar o fogo, para no dar prejuzos uns aos outro 16. Translated narrative from Chapter 4, page 152: Sr. Duarte (D.): "Eu estava l paiol, l para acol, uns 3 km, quase o maior estalo nos cocos, pois o fogo pegou numa palhada [roa do ano anterior], que tinha plantado arroz e grama. Era meio mundo de fogo, no dava conta de apagar, queimou uns 90 cocos em [19]97. Ainda no tinha chovido. Nesse tempo o povo estavanaquele esquema de queimar antes de chuva. Por que no Par, de primeiro [quando a famlia chegou], a gente tocava fogo numa roa bem aqui, o fogo no entrava uma brana mata. Por que acho que as folhas parecem que eram molhadas por vida, no ? Mas agora, se voc no deixar chover, o fogo entra na mata igual botando galho por debaixodo fogo. Mas de primeiro botava 1 alqueire, 2 alqueires de roa, o fogo s fazia roqueimar s o que estava brocado. Hoje em dia no est mais assim no." Question: "O que o senhor acha que mudou daqu a Sr. D.: "Mas eu acho que mudou por causa das chuvas. O Par, de primeiro, era difcil passar um ms sem a mata toda era molhada, voc andava [dentro dela] era frio. E hoje em dia voc v com, passa s vezes de 2, 3 meses, at 4 m chegar qualquer foguinho ali, [o fogo se] altera, pega fogo em tudo, por que est tudo seco! De primeiro no era assim no, voc ciscava a folha, era molhado." 17. Translated narrative from Chapter 4, page 154: Sr. Carvalho:"Eu acho que, para poder estar conscientizando mais o pessoal, tem que fazer a reunio [na comu assim que a gente chegar [do seminrio da Ao do Fogo] (...). Eu vou dizer por mim: quando dava um tanto te a queimar cedo. Ele est fora do movimento [organizao para o uso do fogo], no sabo que est se passando, e ele fica preocupado, eu concordo com ele. Porque acostumada a queimar cedo, quando d a primeira chuva e ele est com queimar, fica preocupado. Eu mesmo sou um. Quando deu as primeiras chuvas [in 1998] que a roa estava querendo ficar em tempo de nascer o broque para todo lado, eu disse: 'esse ano a roa eu no tenho!' Camarada est acostumado a quando dar a primeira chuva a roa j estar queimada. Ento muita gente que tem esse hbito (...) Quando d a chuva de noite, ele pensa 'meu Deus do cu, parece que este ano no vou ter roa.' Nsei, um hbito nosso. Mas s vezes funciona assim: a gente queimar cedo, de repente pode ter um fogo desabado, entrar na mata, ou ento passar para o lote do vizinho e destruir. A quando o camarada, alm de destruir a mata prpria, ele ainda pode ser preso... (...) Se a gente tem conscincia que tem que se conformar, conscientizar os outros. Mesmo que o camarada esteja meio duro, a gente compreender por que ele est meio duro, e tentar repassar aquilo que a gente sabe."

PAGE 214

201 18. Translated narrative from Chapter 4, page 156: Pergunta: "O pessoal falava quantes a mata era o aceiro, no era?" e r. Duarte: Era, a mata era um aceiro. E hoje em dia, se voc no fizer o aceiro, a mata oc apaga bem aqui, pensa que no, ficou um olho a turma; e no for, difcil de atalhar." sou, e hapter 4, page 173: "Uma vez fomos apagar um fogo a O il apagar." ogo. S no serve mais de aceiro no, que o fogo invade mesmo, no tem jeito. E olha l para voc apagar, atalhar ele... Por que v bem acol, uma faisquinha. Sozinho a pessoa no d conta no, [tem que ser] um s 19. Translated narrative from Chapter 4, page 170: Pergunta (P.): "E onde o senhor via fazer ou aprendeu isso [manejo do fogo]?" Sr. Fontes: "L onde eu me criei, l era assim. Meu pai era pobre, como eu tambmtoda roa dele era aceirada." P.: "E isso era aonde?" Sr. F.: Era no Maranho, todas as roas eram aceiradinhas (...). No dia do fogo, ele ia botar [o fogo] logo naquele aceiro na beira do mato: o contra-fogo. Que a quando o fogoda roa vinha sapecando del, antes de chegar at aqui no aceiro, aquele contra-fogo j iabem por l. J se encontravam [contra-fogo com o fogo principal] fora do aceiro. Entoisso eu aprendi de quando eu era menino, trabalhando com o meu pai." 20. Translated narrative from C no Jos, num capim. Tocaram fogo na roa, ns amos bem uns dez, e o fogo saltou paro pasto. Ainda no tinha chovido. O capim estava alto, bem seco, logo o fogo vinha e crescia; ns batamos o galho no fogo, p! E ia acol, corria, e esse [fogo] daqui j tinha ganho o mundo. Junto com gua, apagando, mas no deu jeito; ns largamos que ns estvamos para se queimar. Fiquei sem sobrancelha, caiu, queimou tudo, da quentura.fogo invadiu o pasto dele, queimou tudo. (...) Mas sendo um fogo leve fc 21. Translated narrative from Chapter 4, page 184: Fogo Se o povo no participa das decises, fogo. Quando a classe mais baixa vive oprimida, fogo. Para viver hoje em comunidade, fogo. Se est difcil para conseguir uma vida mais digna, isso tambm fogo. Muitas vezes o financiamento no vem para aumentar a produo, e quando vem fora de hora, isso que fogo. O descaso dos governantes com assistncia mdica, educao e infraestrutura das comunidades, isso fogo. Quando muitas vezes os agricultores produzem e no conseguem escoar a produo por falta de estrada, isso tambm fogo. Quando o IBAMA muitas vezes aplica as leis e s os pequenos que so punidos, fSer um participante ativo dos movimentos populares, como fogo. Se as queimadas muitas vezes so desorganizadas e descontroladas e causam muitos prejuzos, fogo. Quando o seminrio para falar de fogo, e se fala de outros assuntos, fogo. Se os polticos mentem, enganam e iludem os menos desinformados, isso que fogoQuando os tcnicos no fazem na prtica e ficam s na teoria, ah, como isso fogo.

PAGE 215

202 Quando se bota o fogo, e no se pode controlar, temos que achar um jeito para esse fogoapagar, e vai precisar muita gua, ou um grande extintor, mas tambm pode apagar com um pouco de amor, ou quem sabe quando o povo tiver valor.

PAGE 216

APPENDIX C FATA/LASAT COMMUNITY FIRE ACTION BOOKLET The FATA/LASAT Fire Action's recommendations were organized in a booklet (Carvalheiro and Aquino 1999), distributed to at least 44 communities related to FATA and LASAT, during the 1999 dry season. Below are the fire use recommendations, and their main illustrations as presented in the FATA/LASAT Fire Action booklet (Figure C-1). Figure C-1. Cover of the FATA/LASAT Fire Action booklet. 1. General Guidance The booklet starts by explaining that it was developed as part of a process of seeking more responsible fire management practices, and that it contains proposals by peasants and has no official legitimacy. It is composed of three parts, (a) Proposed Recommendations by the Fire Action, (b) Community Agreement regarding Fire Use, and (c) Brazilian Laws concerning Fire Use. As general guidance, it was recommended that an agricultural field (based on the slash-and-burn system) should not be larger than 5 ha per family. The size of a pasture to be burned, on the other hand, was said to be a family decision because fire management for pastures varies greatly (fire for pasture management is used for diminishing weeds). 2 Before Burning 2.1 Official Burning Permit 203

PAGE 217

204 The STR agreed to support community organization for fire management. Because of the INCRA's limitations in recognizing land rights for peasant communities, the STR would also issue a document acknowledging that the family is a peasant family, in much the same way as it already does for peasants applying for government loans. In addition, the STR would give families a written authorization to burn, provided that the families were organized into community grouwould charge no fees to issue the permit (Figure ps, and IBAMA C-2). s (Rural Worker Union) support regarding burning the FATA/LASAT Fire Action boophasized the importance of plt than to control." When Figure C-2. The STR'activities, inklet. 2.2 Planning The booklet emanning, based on the premise that "it is always easier to preven confronting fire prevention versus control live in commort, with no means difficult access. Moreover, the peasants in the Marab region and nst ng water. ocation of agricultural fields, keep fire management in e might be very helpful in preventing accidents related to fire." Decisions on field location are sometimes made during the rainy season, and prevention presents great advantages to peasants that generallynities far away from any Fire Department or other outside supp u of communication, andgenera lly have no access to fire control equipment, such as machinery, water pump appropriate asbestos fire swatter, and few possess chainsaws. The peasants also lack adequate safety equipment. Consequently, fire control is a very dangerous activity for them and every effort should be made to prevent instead of controlling escaped fires. Because of its importance, this section of the booklet includes ten recommendations, as follows: (a) "Discuss in your community what would be the best way to manage fires." It emphasizes community organization, because working collectively enhances peasantsability to implement fire preventive practices and control, such as firebreaks, fire againtrol firebreaks, and carryi fire, varridas, or co (b) "When choosing the lmind. Many times a buffer zon

PAGE 218

205 peasa t areas, and requires more caution. Fire propagates faster uphill because (1) the fuel located on top of a hill or mound is usually drier and is more intensively heated by fire, aintense fire diffusion, and (2) hot air currents move uphill, dbustion zone (Soares 1985:36). (d) "The quality ofor the quality of the burning, and better slashet rain, at a time of day when winds are slower." This recommes 'e' and 'f') are related to slash-and-burn activhich the upper part of plants are tota practice allows vegetation to dry better. If the vegetation is not tot's sap will keep its leaves and trunk green, and the fiel(e) "Weir fall to center of the slashed area." As the previous onand-burn activities, and aims to diminish the am the agricultural field, so as to minimis. (f) "Try to ferders of the slashed area" (Figure C-3). This recommendation calls attention to the need to felling dry trees and palms before the burningequent high risk of s, twigs or leaves. Pea of dead branch fall over peasant during felling), and economic (fire in general knocks down dry trees) s and Carvalheiro 2002:290). trol e support. d again that the rainy ut with some early rain events alternating with sunny days. Peasa ting uncontrolled fires from pastu nts do not always take fire management into account then. Thinking ahead about fire management can help in the implementation of buffer zones. It is also important to consult neighbors about ecosystems or goods to be protected from fire, and to be prepared to change field location, if necessary. (c) "When burning up or downhill, it is necessary to be more careful, due to fasterwinds and longer flames." Peasants and firefighters agree that burning on hills is more dangerous than on fla nd burns faster causing more rawing oxygen-rich fresh air to the com the slashing has direct consequences fd vegetation will allow burnings after a firsndation (as well as itemities. Well-slashed vegetation is that in wlly separated from their base. Thistally separated from its base, the pland will not burn well after rain. hen felling the bigger trees, try to direct the, this recommendation is related to slash-ount of fuel around the border ofze the risk of fire spreading to the neighboring ecosystemll dry big trees and palms located at the bo day, because of their high flammability and cons fire spreading to neighboring ecosystems through sparks or falling branchesants tend to not fell the dry trees, for safety (high risk reasons (Matto (g) "Avoid burnings during the first half of the dry season; each community shoulddefine the less dangerous months for burning." Controlling fire once it gets out of conis very difficult, especially for peasants without adequate equipment and outsidTherefore, this recommendation gives emphasis to the appropriate periods for burning, in the late dry season, for if uncontrolled fire achieves large proportions, rain will help control it. (h) "Burn agricultural fields after a first rain. "This recommendation is connectewith the previous ones (preparing the vegetation to be burned, burning late in the seasonetc). If the vegetation is dry enough, it will be able to receive rain and be dry enough after a few days of sun. This suggestion takes into account the factseason does not start at once, b nts agreed, however, that this recommendation could not be applied when the fieldis located in a swamp, nor if the vegetation is a palhada, the remains of crop, and other regrowth vegetation in a field that is going through a one-year fallow period. (i) "Burn pasture only after the first rains. "Preven re burnings is usually harder than preventing uncontrolled fires from agricultural field burnings. This is caused by the pastures' larger extensions, the difficulty involved in

PAGE 219

206 making manual firebreaks in pasture areas, and the inefficiency of firebreaks due to the long distance reached by sparks and the high speed of fire during the burning. The rain helps by wetting surrounding ecosystems (making it easier to control fires), and byimproving grass regrowth. to ins a (j) "The use of preventive firebreaks is mandatory. "The use of firebreaks was the only obligatory recommendation for the 'before the burning day' phase. Besides preventing fire from spreading to neighboring ecosystems, firebreaks also allow peoplemove about in neighboring ecosystems to spot and control fire. The booklet contaspecial section on firebreak use, as follows. Figure C-3. One of the Fire Action booklet's recommendations, before the burning day, on avoiding sources of sparks: 'try to fell the large dry trees (pau-seco) and palms located at the borders of the slashed area' (Carvalheiro & Aquino 1999:6). 2.3 Us e of Preventive Firebreaks Due to the many different ways in which peasants understand and use firebreaks, the two main kinds of firebreaks were addressed: preventive firebreaks and control firebreaks. Preventive firebreaks are made before the burning, and are either located nexto the border of the burning, or far away from the border of the burning. The firebreak located by the border of the burning is generally used when the vegetation to be burned is a pasture, when it is an agricultural field located right next to a pasture, a one-year-old agricultural field, or a lower secondary forest (capoeira fina). The booklet recommendsthat w t hen the firebreak is located next to the border of the burning its width be 2.00 meters and, if there is a fence, it has to be protected on both sides" (Figure C-4).

PAGE 220

207 Figure C-4. Recommendatiet regarding mandatory the burning. The left agricultuals, meaning that this situation, the firebreak is 2.0 m a slashed field. If there is a fence, The other type of preventive the border of the burning. "This kest (either primarythe border of the different reasons for the first one is its high efficiency. During the burning, the hot air dries leaves and thin trunks close to the border of the burning, and urrounding tree crowns. Firebreaks located next to the bo ning e the forest microclimate is wetter and the wind is slower when compared to the microclimate near the border of the burning. Better fire control, therefore, is the second advantage of on in the Fire Action booklpreventive firebreak use, located by the border ofside of the fence represents slashed vegetation being prepared for re; on the right side there is a pasture with animit has to be protected from fire. In wide and is located next to the border ofboth of its sides have to be protected. firebreak is far away from ind of firebreak is used when the border of the burning is contiguous to a for or old secondary forest) It is located approximately 8 to 15 meters from e burning, and is 1.5 to 2.0 meters wide (Figure C-5)." There ar use of this kind of firebreak. The fire flames and sparks easily reach s rder of the burning are not efficient, because sparks often fly more than 50 m away(personal observation). When the firebreak is located inside the forest, as shown in Figure C-4, the forest works as a green barrier for the fire. The fire starts burning the upper dry leaves by the border of the burning, and as it goes forward into the forest, finding more green leaves, the fire moves to the forest ground, where dry fuel can be found. Inside the forest, the wind loses its power, and fire moves slowly, tending to be extinguished. If the fire reaches the far-away firebreak, it usually does not have enough energy to cross it, due to weak winds and absence of fuel for 1.5 to 2 meters. If a burtree falls across the far-away firebreak, peasants can control it more easily becaus

PAGE 221

208 this kind of firebreak. Tants to stay at the firebreak during thtal fires. Water and available equipmoke, the danger is minimized. he third reason, also connected with the two previous ones, is peasant safety. Firebreaks located far from the burning area allow pease burning, walking through it and looking for accidenent can be transported easily, if necessary. Controlling fire is a dangerous activity, and by being relatively far from heat and sm y zone The fourreak is its feasibility. Peasants that in general do no posses machinery and even chainsaws can build far-away firebreaks inside d ir ts at Figure C-5. FATA/LASAT Fire Action booklet's recommendation on obligatorpreventive firebreak use, located far from the border of the burning. Tothe left is the slashed vegetation to be burned, then there is a bufferof standing forest, the firebreak and, on the right, the forest to be protected. th reason that justifies this kind of fireb forests without felling any large trees. Instead of being made in a straight line, thiskind of firebreak can be made around big trees, and be located further or closer to the burning border, depending on the composition of the standing vegetation (lianas, palms, dead trees). The fifth reason to use a far-away firebreak is the smaller amount of damage caused to the forest. Less adult trees are felled, compared to a 6.0 meter-wide firebreak, and if fire burns the standing forest between the border of the burning and the far-away firebreak, its recovery is faster when compared to a firebreak in a soil that has been cleared due to re-sprouting (85 % of forest recovery comes from re-sprouting: Moira Adams). Some of the families studied used this kind of firebreak during the 1998, 1999 an2000 dry seasons, others have just heard about it, and others learned about it from thecolleagues. Firebreaks located far from the burning border are also known by colonisthe Paragominas region (Mattos and Carvalheiro 2002). This kind of firebreak is not officially recommended by IBAMA, which instead requires firebreaks in neighboring forests to be located by the burning border, and to be about 6.0 meters wide. A firebreak located far away from the area of the burning has to be associated with the fire against fire technique, as well as with the practice of burning after the first rain, and should be done by a group of people, so that fire can be observed and controlled if necessary.

PAGE 222

209 2.4 Communication Between Neighbors The booklet emphasizes the importance of communication, "Good communication is the basis for unity", and focuses on four points: (a) always try to work collectively, (b)always try to agree among neighbors to burn at the same period, (c) anyone who will burn a field must officially communicate the planned burning date, at least 8 days iadvance to c n ontiguous neighbors; the STR may help with an official information form, (d) the more people in the burning event, the better; specifically invite your contiguous ong neighbors. In coloniss, religions, life ore, le task. Fire, on t of control in communities, the fire spssions or land ownership. This indisake families work together. The mther they can use better practices, sure people using the co 3 On thThe book(a) "The Any loss cause'owner' of the burning. "This recommendation calls attention to peasants' responsibility regarding fire use and its consequences. ted, o accidental fire" (Figure C-6). Here, the booklet also emphasizes group work in plann neighbors and insist on their participation. This section concentrates its recommendations on collective work amt communities, where families have different originhistories, etc., the newer the community, the fewer the family ties. Therefcommunication and willingness to work collectively may not be a simpthe other hand, can draw people to work together because once it gets oureads regardless of family, kinship, religion, possecriminate behavior at the community level tends to minimum recommendation is communicating to neighbors the chosen burning day, so neighbors can plan to observe or participate at the burning. Working collectively is advantageous for peasants because togech as preventive fire against fire and firebreaks, and having montrol practices increases the possibility of success. e Day of the Burning let contains ten recommendations in this section: family responsible for the burning must be present at the burning event. d by an uncontrolled/escaped fire will be the responsibility of the (b) "It is necessary to confirm the date of the burning again with the contiguous neighbors. "Before the burning day, it is recommended that the peasants notify their neighbors about the day in which the family plans to burn. The burning date can change due to several reasons, or neighbors may have forgotten about the burning date, and notifying them again reinforces the importance of neighbors' presence. (c) "Never burn a field alone, even if it is not considered a dangerous situation; always burn with a Fire Group, family, and friends. "This recommendation again emphasizes the importance of group work for prevention and control of accidental fires. (d) "Organize a group of at least 4 people, and discuss how the fire will be starthe role of each participant, codes of communication during the burning, and what to din case of ing, in discussing prevention and in control practices. (e) "Required use of preventive fire against fire, or contra-fogo". As discussed in the section on firebreak, fire against fire is a fire started from the side opposite the wind, which makes the fire spread against the wind. Almost at the same time, the main fire runs with the wind. Before the main fire reaches the other side where the contra-fogo started, a strip of land without fuel will be formed, and the main fire will finish before it reaches the burning field border, minimizing risks of accidental fires in the neighboringecosystem.

PAGE 223

210 Figure C-6. FATA/LASAT Fire Action's booklet Recommendation on community organization for burnings. (f ) "If possible, arrange containers with water near firebreaks". Many times from roads or trails, and from water sources. That is the reasoto fire ral field The advantage is that threasof the day. This will allow you tosk for help in controlling any accidental fires." This recommendation has the same purpose as the recommendation to remain at the burning field until the end of the agricultural fields are located far n for the conditional recommendation, using "if possible." (g) "Equipment (boots, gloves, helmet, scythe, ax, fire swatter, water pump and chainsaw) is essential but not always available; when accessible, leave them within reach." This section calls attention to the necessary safety equipment, as well as control. (h) "Start a burning later in the day, varying according to its being in an agricultufield (recommended after 3 p.m.), or in a pasture (recommended after 4 p.m.)". This recommendation tries to change a regional practice of burning field around midday, whenthe wind is stronger and the sun is hotter. Peasants argued that a few hours later, the to be burned is still hot and the wind strong enough for a good burning. e microclimate of surrounded ecosystems will not be as hot as it is at midday, which consequently makes it easier to control accidental fires. (i) "Avoid the season's drier days, and periods in which wind is too strong". The n for this recommendation is directly related to the previous one. Burning on days which are not so hot and dry minimizes the risk of accidental fires and, in the event of any accidental fire, makes it easier to control. (j) "Keep on observing the burned field until the end o control any accidental fires more easily." Continuous observation of the burning allows early detection of accidental fires, and it is easier to control an accidental fire in the beginning, when its flames are small and its sparks are few. 4 After Burning "Return to the burned field during the following three days, to monitor and, if necessary, a

PAGE 224

211 day, which is to allow early detection and control of any accidental fires. Many trunks any days after the burning day, and may fall over surrounding s or hurl sparks at them. The time during which trees can continue burning rding to tree species and age. For instance, a copaba tree was observed to go ore than two weeks after the burning day, even after rain. Older trees burn longer due their to amount of fuel, some burning even their roots, as was observed angelim vermelho. 5 Controlling Accidental Fires Control practices are used on the burning day or after, depending on causes of uncontrolled fires. The Fire Action booklet divided this section into four parts: ire Group must always be alert, in order to help in case of any accidenendation emphasizes Fire Group organization and responsibilities for (b) "Pay special attention to burning stumps, trunks or roots due to the risk of ended to extinguish or destroy them." This recommendation continue to burn for mecosystemvaries accoon burning for mwith an (a) "The Ftal fires." This recommthe entire dry season. sparks; it is recommcompments the one about observing the burning field until late on the burning day and for thre(c) "Combreaks, and backfires so as to protect forests, perennial crops or houses from any uncontrolled fires. But be careful: neverious!" m the one that will not be bu ady, Thus the name of this practice: contra-fogo, or fire against fire. The goal of this fihe can kill a person. In gener e wearing rubber sandals, and without any safetyThe Fire Action booklet puts emphasis on community organization as an essential sses caused by fire. In order to promote discussions on the importance of firen the le e days after that, in order to prevent accidental fires. bine the use of varridas, or control fire try to control fire alone, because it is faster than you, and your life is too precPeasants of the Marab region in general control fire with fire, preparing a varrida, or thin control firebreak, to separate the vegetation to be burned fro rned. This control firebreak creates a strip of vegetation between the uncontrolledfire and the ecosystem or goods to be protected from the fire. When the varrida is rethe peasant sets fire to the vegetation strip that tends to propagate to the uncontrolled firedirection re is to remove any fuel that may be close to the area to be protected, so that when the uncontrolled fire reaches it, it will be extinguished due to lack of fuel to continue tcombustion. The booklet also calls attention to the fact that this practice should not be used when you are alone, because fire is very fast, and building the varrida is very time consuming. The Fire Action booklet in general emphasizes that fire management should never to be done by a single person, but by a group (Figure C-7). (d) "Pay attention to smoke and hot air: smoke and hot air al, smoke is composed of toxic gases that act slowly, while the heat can kill a person in few minutes if she/he inhales hot air, burning her/his lung (...)." This part givessome safety and first aid for information people working close to a burning. It is common for peasants to try to control fire whil or fire control equipment. Therefore, items 'c' and 'd' call attention to safety risks while controlling fires. 6 Conflict Resolution key to minimize lo management, it addresses topics such as 'why did burnings increase so much ilast few years?', because of global warming and changes in forest flammability due to

PAGE 225

212 logging activities. It also emphasizes what peasants have noticed: 'nowadays fire use is more dangerous than years ago'. This section discusses the Brazilian government's feof fire in the Amazon, and IBAMA's coercive laws as an attempt at fire control solution ars s. inforcing their local organplies n fire sult from community discussions. It is important to discuss those agreements in depth, in order to establish the recommendations that have to be the community. If rules vary according to individual interests, recom Figure C-7. FATA/LASAT Fire Action's recommendation regarding the use of control firebreaks and backfire, in order to control the fire. The booklet presents concepts discussed during the regional workshop for responsible, accidental, and criminal fires. These concepts, based on community agreements, can be used to evaluate losses caused by uncontrolled fires, contrasting accidental fires with criminal ones. In case of criminal fires, the concepts help in attributing responsibility to the fire 'owner'. The booklet's proposed steps for community agreements regarding fire were: (a) "Community members should gather information at the STRs, and key NGOs and GOs." This recommendation aims to empower community representatives by suggesting that they be responsible for actively seeking information on fire use. When the representatives have relevant information, they are recognized by other peasants at the community level as credible resources for fire management, re ization. (b) "Organize the community for fire management." This recommendation imthat, besides getting information, the whole community should be motivated to discuss and organize fire use. This suggestion aims to stimulate community agreements ouse, involving most families in the community. (c) "Discuss community fire use recommendations." It is expected that recommendations will re followed by everyone in mendations will lose their reliability. (d) "Share responsibilities, and support the creation of a fire group." This suggestion aims to support the decentralization of community fire management, which is important to motivate families' responsibilities for their own acts. In order to be

PAGE 226

213 responsible, families have to be active in obtaining information, to communicate with neighbors, to plan fire use, to get organized and check their neighbors' activities during fire use, and to take responsibility for any losses caused by their fires. Besides family responsibility, the booklet encourages the creation of Fire Groups, established in assemlosses. mbers would receg on justice and fthat onions, will thcommunity recommendations, can they be responsible for their use of fire, and therefore take r on fire laws, and information that is not presented in IBAMA's booklet, such as deforestation and burning costs. It was includasants tion to e bly. Depending on community decisions, Fire Groups can motivate family planning, disseminate information, and act as a third party in conflicts involving fire (e) "Approve community agreements in assembly or official meetings." It is essential that the Community Assembly (the highest decision forum at the community level) approve the discussed recommendations, in order to render the recommendations and the Fire Group's work legitimate. Some peasants suspected that Fire Group meive payment from IBAMA for monitoring and punishing families, workinbehalf of individual interests, and this generated conflicts among peasants. Official recognition of Fire Groups' work reduces this kind of conflict, and generates a sense of airness for all. (f) "Disseminate agreements inside the community." This suggestion reinforces ly when all families in the community have access to the agreed recommendatey able to use fire consciously. In addition, only if all families know the esponsibility for consequences in case of losses resulting from fire. The last part of the booklet presents some basic points ed in the Fire Action booklet to complement the IBAMA booklet. Some pewere confused by this section because they understood the FATA/LASAT Fire Acbe defending IBAMA's fees. The FATA/LASAT booklet would have had a better reception if this section had not been included, but furnished as an appendix for thIBAMA booklet.

PAGE 227

APPENDIX D INTERVIEW GUIDE Losses caused by fire -Accessing IBAMA fire laws Ecosystem slashed for agriculture People during the burning Inviting neighbors to the burning event Informing neighbors of the burning event date Month and hour chosen for burning Firebreak Backfire Control practices (water, control backfire) Burning of pasture Participation in Fire Groups Interview guide for unstructured interviews carried out in 1998. The fire questions were repeated in 1999. Age (male and female heads of families) Length of residence in the community State of Origin Migration of heads of families Plans of continue living in the community Num ber of people that live in the community Number of family members living outside that depend on the family's production Formal education Possess land before this one Work force (male and female) Hirin g workers Collective activities Possess equipments Plan of continuing in the community Participation in formal social activities Size of the land Land use distribution Livestock 214

PAGE 228

APPENDIX E CODED BOOK R Tempo de Residncia na comunidade, em anos. es r na comunidade (1=sim; 2=no) nos de Continuar na Comunidade (1=sim; 2=no) =no) em no lote s dependem economicamente de, em anos, em er 2=GO+TO; 3=PA; 4=PI; 5=CE; 6=MG; 7=ES; A; 9=PE) de migraes at chegar na comunidade (onde exerceu atividades produtivas) am de obra (1= maior parte do ano; 2= parte do ano; 3=nunca) o) Participao em mutiro (1=sim; 2=no) =sim; 2=no) o de ManejoFlorestal (1=sim; 2=no) STR Scio STR (1=sim; 2=no) Ass Scio Associao (local ou regional) (1=sim; 2=no) Reli Religio frequentada (1=catlica; 2= evenglica; 3= Batista; 4=no participa) Pos Terra Possuiu terra antes de chega Cont Com Pla Vila Possue casa na Vila (1=sim; 2 Pessoas Nmero de pessoas que vivem no lote e: Res. Resid Dep. No residem ma Idade Ida H do hom M da mulh Orig Estado de origem (1=MA; 8=B Migrao Nmero M.O.adult Quantos trabalh M.O.$ Contrata mo M.O.troca Realiza troca de dia (1=sim; 2=n Muti CAT Participa Grupo do Viveiro (1 PDA Participa Projet 215

PAGE 229

216 H Homem M Mulher Tam Tamanho (em hectares) Lote do lote Mat da rea de mata Cap da rea de capoeira Pda rea de Pastagem Rda rea de roa as oc oc98 cnica de Controle na queima da roa (1=sim; 2=no) Ecos o na roa (1=mata; 2=capoeira; 3=roa anterior) j particparam no dia da queimada 5=nov; 6=dez) roa Fruda rea de fruteiras e perenes Mat sape Quantidade da mata sapecada Gad Quantidade de Gado Ani Quantidade de Animais de Carga Equipam Equipamentos 2=no) Mot Motosserra (1=sim; Bom Bomba Costal (1=sim; 2=no) Contro RNecessidade de T Ecossistema us ad Prej Ocorrncia de Prejuzo neste ano (1=sim; 2=no) Tipo de PreTipo de ecossistema/bem perdido (1=mata; 2=roa; 3=cap; 4=past; 5=fruteiras; 6=casa; 7=cerca; 8=aaizal; 9=gado) No. Pess Nmero de pessoas que Viz AvisAvisou Vizinho (1=sim; 2=no) Viz Con Convidou Vizinho (1=sim; 2=no) Acei Aceiro (1=sim; 2=no) Cont Contra-fogo (1=sim; 2=no) Mes Ms do ano que queimou (1=jul; 2=ago; 3=set; 4=out; Hor Hora do dia para queima de Chuva Queima aps um evento chuvoso (1=sim; 2=no) gua Uso de gua (1=sim; 2=no)

PAGE 230

217 Past 98sto em 98 (1=sim; 2=no) rej Past Ocorrncia de Prejuzo desta queimada (1=sim; 2=no) Past 99eimar Pasto 99 (1=sim; 2=no) IBAMA IBAMA (1=sim; 2=no) e normas do Ibama via Radio (1=sim; 2=no) bama via TV (1=sim; 2=no) a via STR/Assoc. oc99 Tamanho da Roa em 1999 cos99 Ecossistema que se planeja usar para roa (1=mata; 2=capoeira; 3=roa Prej: Hist Se teve histria de prejuzo (1=sim; 2=nao) Tipo Tipo de ecossistema/bem perdido (1=mata; 2=roca; 3=cap; 4=past; Queima de Pa P Planeja qu : Cart Leitura da Cartilha do Rad Conhecimento d TV Conhecimento de normas do I STR Conhecimento de normas do Ibam R E anterior) Prejuzo Ano Ano de ocorrncia do Prejuzo 5=fruteiras; 6=casa; 7=cerca; 8=aaizal; 9=gado)

PAGE 231

LIST OF REFERENCES Abberg er, Hut M., Bradford M. Sanders, and Helmut Dotzauer e Fo a tm2001 Tvrest Fire Mmat Cnference on mmungement. Balikpapan, Indonesia. Eletronic umenEP/006/AD348E/ad348e07.htm, acessed GRAF utoo Agro-Ambiental de Atividades de Agricultura Familiar da oocument, Marab, Jun 1998. 3Almeidre1991 O Intranos a Violncia na maznia (1965-1989). In Amaznia: A Fronteira Agrcola 20 anos Depois, edited by Philippe Lna and Adlia Engrcia de Oliveira, pp. 259-290. Museu Paraense Emlio Goeldi MPEG, Belm. Alves, Filho. Armando dos Santos 2000 A Guerrilha do Araguaia. In Pontos da Histria na Amaznia Vol 2, edited by Filho. Armando Alves, Jos Alves Jr., and Jos Maia Neto, pp. 71-99. Editora Paka-Tatu, Belm. AMPPDR 1996 Regulamento de Queimadas na Colnia Del Rey. Associao dos Mini e Pequenos Produtores da Colnia Del Rey/Commisso do Fogo/STR de Paragominas/IPAM. IPAM, Paragominas. 20 p. Banco Central do Brasil 2004 Cotaes e Boletins. BCB Ministrio da Fazenda. Electronic document, http://www.bcb.gov.br/?TXCOTACAO, accessed May 2004. Benedet, Maria Helena 2002 Ibama Realiza Mega Operao no MT para Combater as Queimadas. Electronic document, http://www.amazonia.org.br/noticias/noticia.cfm?id=20347, acessed August 2002. Bordenave, Juan E. Diaz 1985 O Que Participao. Editora Brasiliense, So Paulo. 85 p. he Delopment of a Community-based Approach for an Integratedsented anageent System in East Kalimantan, Indonesia. Paper preommuities in Flames: Proceedings of an International Con Coity Involvement in Fire Manadoct, http://www.fao.org/DOCR April 2004. A 1998 Estat da AssociaV ila SJoo do Itacainas. AGRAF. Unpublished d p. a, Alfdo Wagner Berno de sitivo do Transitivo: O Estado, os Conflitos Agrri e A 218

PAGE 232

219 Browder, J. O., and B. J. Godfrey 1997 Geopolitics, Regional Development, and Urbanization: Historical Dinamics of Amazon Frontier Expansion. In Rainforest Cities, pp. 55-82. Columbia university Press, New York. Bunch, Roland 1994 Duas Espigas de Milho: uimento Agrcola Participativo. AS-PTA/PESACRE, Rio de Janeiro. 221 p. 2004 tronic April Carvalheiro, Katia 1999 Manejo Comunitrio de Queimadas. LASAT/FATA, unpublished report, Marab, ary 1999. 49 p. Carva1999 jo Comunitrio de Queimadas:Mais Unio, Menos Prejuzos. LASAT/FATA/SUNY/DFID, Marab. 20 p. 2000 lm. 4. 19 p. urch Archdiocese of So Paulo CPT anece. Comisso Pastoral da Terra. Electronic document, accessed dings of an International Conference on Community Involvement in Fire Management, Balikpapan, Indonesia. Eletronic document, DOCREP/006/AD348E/ad348e07.htm, acessed April 2004. ma Proposta de Desenvolv BVRoraima O Fogo que Entrou de Vez para a Histria de Roraima e do Brasil. Elecdocument, http://www.bvroraima.com.br/roraima_chamas.htm, accessed 2004. Janu lheiro, Katia, and Helosa Aquino Mane Carvalho, Antnio de Souza O CAT na Perspectiva do Movimento Campons. In CAT-Ano Dcimo: Etnografia de uma Utopia, edited by Jean Hbette and Raul da Silva Navegantes, pp. 65-76. Editora Universitria UFPA, Be CAT 1994 Documentao sobre a Sistematizao de Dados e Anlise Preliminar das Atividades do Projeto Agroflorestal (PAF). CAT/FATA, unpublished report, Marab, August 199 Catholic Ch 1998 Torture in Brazil:a Shoking Report on the Pervasive Use of Torture by Brazilian Military Governments, 1964-1979. University of Texas Press, Austin. 238 p. 2004 A Impunidade Perm http://www.cptnac.com.br/?system=news&action=read&id=1159&eid=8April 2004. Darlong, V. T. 2001 Traditional Community-based Fire Management Among the Mizo Shifting Cultivators of Mizoram in Northeast India. Paper presented at Communities in Flames:Procee http://www.fao.org/

PAGE 233

220 DFID 1998 Projeto de Desenvolvimento Rural e Conservao Florestal no Tocentins-Fase II:Avaliao, Resultados/Objetivos. Projeto DFID/FATA, unpublished report, Marab, December 1998. 49 p. PA, Braslia. 42 p. Electronic document, ww.cptnac.com.br/?system=news&action=read&id=1159&eid=8, accessed Emerson, Robert M., Rachel I. Fretz, and Linda L. Shaw 1995 Writing Ethnographic Fieldnotes. The University of Chicago Press, Chicago. Emm1999 A Oligarquia do Tocantins e o Domnio dos Castanhais. UFPA/NAEA, Belm. Eva, H2003 Examining the Potential of Using Remotely Sensed Fire Data to Predict Areas of South America. Applied Geography 23:189-204. Forum2001 rais no Estado do Par. FETAGRI MST/CPT/SPDDH/STR's Sul e Sudeste do Par/FASE/CEPASP. 18 p. 1985 9 p. inuum Publishing Corporation, New York. 1998 Pedagogy of Freedom: Ethics, Democracy, and Civil Courage. Rowman & Littlefield Publishers, Boston. 138 p. Goldm2003 ature, Privileged Knowledge: Community-based Conservation in Tanzania. Development and Change 35:833-862. 2000 EMBRAPA 2000 Alternativas para a Prtica de Queimadas na Agricultura. Ministrio de Agricultura e Abastecimento MAA/EMBRA http://w January 2004. 254 p. i, Marlia 174 p. ugh, and Steffen Fritz Rapid Forest Change in das Entidades pela Reforma Agrria do Sul e Sudeste do Par A Violncia Contra Trabalhadores Ru Freire, Paulo The Politics of Education: Culture, Power and Liberation. Bergin & Garvey Publishers, South Hadley. 20 1986 Pedagogy of the Oppressed. The Cont 186 p. an, Mara Partitioned N Harwell, Emily E. Remote Sensibilities: Discourses of Technology and the Making of Indonesia's Natural Disaster. Development and Change 31:307-340.

PAGE 234

221 Hbette Jean by and Edna Silva Moreira 1997 IBAMA reas de Atuao. IBAMA/PREVFOGO. Electronic document, 2004a Roraima Ter Plano de Preveno a Incndios Florestais.Electronic document, ssed January 2004. 2004bc document, http://ibama.gov.br, accessed April 2004. _legal.htm, accessed April 2004. 1998 Cor do Brasil. IBAMA, Braslia. 36 p. 1998 s. IBGE-Instituto Brasileiro de Geografia e Estatstica, Rio de Janeiro. Map scale 1:1,000,000. 2002 o Planejamento, Oramento e Gesto/ IBGE/ Diretoria de Geocincias, Rio de Janeiro.Map scale 1:5,000,000. 1999 rative Management of Natural Resources: An Overview. FAO, Rome. 84 p. 2004 /CPTEC/DPI. Electronic document, http://www.doi.inpe.br/proarco/bdqueimadas/, accessed March 2004 2000 O CAT em Cronologia. In CAT-Ano Dcimo: Etnografia de uma Utopia, edited Jean Hbette and Raul da Silva Navegante, pp. 287-299. Editora Universitria UFPA, Belm. Hbette, Jean 1991 Situao Social das reas Rurais Amaznicas. :CAT unpublished report, Belm1991. 22 p. Situao Social das reas Rurais Amaznicas. CAT, unpublished report, Belm, 1997. 22 p. Holstein, James A., and Jaber F. Gubrium 1995 The Active Interview. Sage Publications, Thousand Oaks. 85 p. 1998 Prevfogo: http://www.ibama.gov.br/~prevfogo/atua.html, acessed September 2000. http://ibama.gov.br, acce Inaugurado o Centro de Monitoramento Ambiental do IBAMA. Electroni 2004c Focos de Calor na Amaznia. MMA/IBAMA/PROARCO MCT/INPE. Electronic document, http;//www2.ibama.gov.br/proarco/relatorio/Amaz IBAMA/GTA Queimada Controlada: Verde a IBGE So Lu Mapa de Clima do Brasil. Ministrio d Ingles, Andrew W., Arne Musch, and Helle Qwist-Hoffmann The Participatory Process for Supporting Collabo INPE Histograma dos Focos de Calor-Noaa-12. INPE/IBAMA

PAGE 235

222 Jackson, W. T. and P. F. Moore actices and Techniques for Community Forestry. Bangkok, Thailand, September1998. 2001 -based Forest Fire Management in Wenyime Village, Sancheahe Township, Dayao County, Chuxiong Yi Autonomous Prefecture, Yunnan Province, cument, http://www.fao.org/DOCREP/006/AD348E/ad348e07.htm, acessed April 2004. 2001 Amaznia Socioambiental Sustentabilidade Ecolgica e Diversidade Social. In ersidade Biolgica e Cultural da Amaznia, edited by Ima Clia Guimares lo 1971 tive Observation and Analysis. Maka2001 Community Involviment in Fire Management: Cases and Recommendations for unities alikpapan, Indonesia. Eletronic document, w.fao.org/DOCREP/006/AD348E/ad348e07.htm, acessed April 2004. ausas do Desmatamento da Amaznia Brasileira. Banco Mundial, Braslia. 100 p. 2002 madas como Resposta aos Fogos ed by Csar IFOR/EMBRAPA Meggpact of Mega-Nio Events on Amazonia During MMA000 Lei da Vida:a Lei dos Crimes Ambientais. Ministrio do Meio Ambiente, Braslia. 39 p. 1998 The Role of Indigenous Use of Fire in Forest Management and Conservation. Paper presented at International Seminar on Cultivating Forests:Alternative Forest Management Pr Lichang, Zhang, Wu Long, Zhao Ya qiao, and lu Caizhen Community China. In Communities in Flammes. FAO/Project FireFight South East Asia. Eletronic do Lima, Deborah, and Jorge Pozzobon Div Vieira, Jos Maria Cardoso da Silva, David Conway Oren, and Maria ngeD'Incao, pp. 196-251. Museu Paraense Emlio Goeldi, Belm. Lofland, John Analysing Social Settings: a Guide to Qualita Wadsworth Publishing Company, Belmont. 136 p. rabhirom, Pearmsak, David Ganz, and Surin Onprom Community-based Fire Management in Thailand. Paper presented at Commin Flames:Proceedings of an International Conference on Community Involvement in Fire Management, B http://ww Margulis, Sergio 2003 C Mattos, Marli M., Katia O. Carvalheiro, and Renata A. Alves Regulamento Comunitrio de Quei Descontrolados. In Simpsio Internacional IUFRO Manejo Integrado de Florestas midas Neotropicais por Indstrias e Comunidades:Aplicando Resultados de Pesquisa, Envolvendo Atores e Definindo Polticas Pblicas, edit Sabogal and Jos Natalino Macedo Silva, pp. 281-297. C Amaznia Oriental, Belm. ers, Betty J. 1994 Archeological Evidence for the Im the Past Two Millenia. Climate Change 28:321-338. 2

PAGE 236

223 Muchagata, Marcia, and Katrina B rown Muchagata, Marcia, and Katrina Brown 2003 Cows, Colonists and Trees: Rethinkinh Cattle and Environmental Degradation in 6. Much1996 s in Negreiros, Gustavo Hees de, Daniel Nepstad, David Sandberg, Ernesto Alvarado, Tom Nelson, Bruce W. El Nio: Florestas Alteradas pelo Fogo Natural na Amaznia. Paper Nepstad, Daniel, Adalberto Verssimo, Ane Alencar, Carlos Nobre, Eirivelthon Lima, aulo Moutinho, Elsa 1999 Olive1989 A Geografia das Lutas no Campo. Editora Contexto, So Paulo. 101 p. 1988 Slash and Burn: Farming in the Third World Forest. University of Idaho Press, Presid2002 2002. PROA Preveno e Controle s Queimadas e aos Incndios Florestais no ber 2003. 2000 Colonist Farmers' Perceptions of Fertility and the Frontier Environment in Eastern Amazonia. Agriculture and Human Values 17:371-384. Brazilian Amazonia. Agricultural Systems 76:798-81 agata, Marcia, and R. Gonalves Forest and People: The Role of Forest Production in Frontier Farming SystemEastern Amazonia. Master of Science, University of East Anglia. 89 p. Hinckley, and Marcos Pereira 1996 Fire Along the Transition Between the Amazon Forest and the Cerrado Ecossystems. Paper presented at Proceedings of the 13th Conference on Fire andForest Meteorology, Lorne Australia. 1997 O ndio e presented at Congresso Nacional de Botnica, Crato, Cear. Paul Lefebvre, Peter Schlesinger, Christopher Potter, P Mendona, Mark Cochrane, and Vanessa Brooks Large-Scale Impoverishment of Amazonian Forests by Logging and Fire. Nature 398:505-508. ira, Ariovaldo U. de Peters, William J., and Leon F. Neuenschwander Idaho.113 p. ncia da Repblica Lei no. 4.771, de 15 de Setembro de 1965: Institui o Novo Cdigo Florestal. Medida Provisria no. 2.166-67, de 24 de Agosto de 2001. Presidncia da Repblica. Eletronic document, http://www.planalto.gov.br, acessed OctoberRCO 1998 Programa de Arco do Desflorestamento. IBAMA/PROARCO, Braslia. Eletronic document, http://www2.ibama.gov.br/proarco/apresentacao.htm, acessed Septem

PAGE 237

224 Pyne, Stephen J. 1998 Forged in Fire: History, Land, and Anthropogenis Fire. In Advances on Historical Ecology, edited by William Bale, pp. 64-103. Columbia University Press, New York. 2001 rn Thailand. Paper presented at Communities in Flames: Proceedings of an International Conference re Management, Balikpapan, Indonesia. Eletronic 1995 rt of Hearing Data. Sage Publications, Thousand Rudd1974 The Yukpa Autosubsistence System: a Study of Shifting Cultivation and Ancillary olumbia and Venezuela. Ph.D., University of California, Berkeley. Sanderberg, David V., Ernesto Alvarado, Gustavo H. de Negreiros, Joo A. de Carvalho, 1998 and Use and Climate Scenarios. Santos, Roberto ge:The Problem of Land in the Brazilian Amazon. In Frontier ood, pp. 439-462. University Presses of Florida, Gainesville. 1999 amework for Gender and Community-Based Conservation. Gender, Community Participation and Natural Resource Management :1-14. 1992 7 p. SECTAM ologia Marab. SECTAM Secretaria Executiva de Cincia, Tecnologia e Sertsu, S. M., and P. A. Sanchez 978 Effects of Heating on Some Changes in Soil Properties in Relation to an Ethiopian Land Management Practice. Soil Science Society of America Journal :940-944. Rakyutidharm, Atchara Forest Fire in the Context of Territotial Rights in Northe on Community Involvement in Fi document, http://www.fao.org/DOCREP/006/AD348E/ad348e07.htm, acessed April 2004. Rubin, Herbert J., and Irene S. Rubin Qualitative Interviewing:The A Oaks. 302 p. le, Kenneth Activities in C 368 p. Sue A. Ferguson, and Guido Ranzani Modeling the Role of Fire in the Forest-Savanna Ecotone in the Brazilian Amazon:Fire Severity Under Present and Future L USDA Forest Service. 53 p. 1984 Law and Social Chan Expansion in Amazonia, edited by Marianne Schmink and Charles H. W Schmink, Marianne Conceptual Fr Schmink, Marianne, and Charles Wood Contested Frontier in Amazonia. Columbia University Press, New York. 38 2004 Meteor Meio Ambiente. Eletronic document, http://www.para30graus.pa.gov.br/meteorolo3c.htm, acessed April 2003. 1

PAGE 238

225 Sestini, Marcelo F rancisco, Regina Clia dos Santos Alvala, Eliana Maria Kalil Mello, 2002 ao de Mapas de Vegetao para Utilizao em modelos Meteorolgicos e Hidrolgicos. MCT/INPE/CPTEC. 72 p. 1991 Skole, David, and Compton Tucker 1993 Tropical Deforestation and Habitat Fragmentation in the Amazon: Satellite Data 1905-1910. Soares, Ronaldo Viana 1985 Preveno e Controle de Incndios Florestais. MEC/CAPES/ABEAS, Braslia. Sonko2001 on't They Come and Discuss Together? Community-initiated Stakeholder Co-ordination on Forest Fire Management in Rural Gambia. Paper presented at gement, Balikpapan, Indonesia. Eletronic sed Souza Jr, Carlos, Laurel Firestone, Luciano Moreira Silva, and Dar Roberts orest Degradation in the Eastern Amazon from SPOT 4 Through Stone2003 From Tapping to Cutting Trees: Participation and Agency in two Community-anagement Projects in Acre, Brazil. Doctor of Philosophy, Sunderlin, Willima D. tion in Indonesia:Steps Toward Overcoming irolamo Domenico Uhl, Christopher, and J. Boone Kauffman bility, and Potential Tree Responses to Fire in the Dalton de Morisson Valeriano, Chou Sin Chan, Carlos Afonso Nobre, Joo Argemiro de Carvalho Paiva, and Erica da Silva Reimer Elabor Setzer, Alberto W., and Marcos C. Pereira Amazonian Biomass Burning in 1987 and an Estimate of Their Tropospheric Emissions. Ambio 20:19-22. from 1978 to 1988. Science 260: 148 p. Kebba, Saikou Samateh, Kanimang Camara, and Clemens Beck Why D Communities in Flames: Proceedings of an International Conference on Community Involvement in Fire Mana document, http://www.fao.org/DOCREP/006/AD348E/ad348e07.htm, acesApril 2004. 2003 Mapping F Expectral Mixture Models. Remote Sensing of Environment 87:494-506. Samantha Sara Based Timber M University of Florida, Gainesville. 414 p. 1997 Shifting Cultivation and Deforesta Confusion in the Debate. Overseas Development Institute, London. Treccani, G 2001 Violncia e Grilagem: Instrumentos de Aquisio da Propriedade da Terra no Par. UFPA/ITERPA, Belm. 600 p. 1990 Deforestation, Fire Suscepti Eastern Amazon. Ecology 71:437-449.

PAGE 239

226 Uhl, Christopher, Adalberto Verssimo, Paulo Barreto, Marli Maria Mattos, and RicardoTarifa Lessons from the Aging Amazon Frontier: Opportunities 1994 for Genuine 287-303. Cambridge University Press. nic ecuiaba.com.br/, acessed June 2003. rio da UFPA:2000-2004. NEAF/CA/UFPA. Unpublished report, Belm. 1972 de Expanso e Estrutura Agrria:Estudo do Processo de Penetrao numa rea da Transamaznica. Zahar Editores, Rio de Janeiro. 178 p. 1982 Wanb2000 fia ia, edited by Jean Hbette and Raul da Silva Navegante, pp. 37-41. Editora Universitria UFPA, Belm. 1991 mid Tropics. Forestry Note 8. FAO, Rome. 2003 ticias/print.cfm?id=75749, acessed July 2003. 1996 Practice 6:6-15. Development. In Biodiversity and Landscapes:a Paradox of Humanity, edited by Ke Chung Kim and Robert D. Weaver, pp Vargas, Rodrigo 2003 Queimadas:Cronograma de Combate Est Atrasado. Dirio de Cuiab. Eletrodocument, http://www.diariod Veiga, Jr. Iran P. 2000 Plano de Desenvolvimento Integrado do Programa de Pesquisa-Formao-Desenvolvimento do NEAF Centro Agropecu Velho, Otvio Guilherme Frentes Velho, Otvio Guilherme Sociedade e Agricultura. Zahar Editores, Rio de Janeiro. 145 p. ergue, Emmanuel O Programa CAT: Memrias de um Protagonista. In CAT-Ano Dcimo:Etnograde uma Utop Warner, Katherine Local Technical Knowledge and Natural Resources Management in the Hu Werneck, Keka Proibio no Impede o Aumento das Queimadas. Gazeta Digital. Eletronic document, http://www.amazonia.org.br/no White, Sarah C. Depoliticising Development: the Uses and Abuses of Participation. Development in

PAGE 240

BIOGRAPHICAL SKETCH Katia Carvalheiro was born in September 1st, 1966, in So Paulo, Br azil. In 1986, ergraduate program in Forestry at the Escola Superior de Agricultura 'Luiz r program, she developed her first work in Amazonia, studying deep-roots distribution in the moved to Par state. In 1992, she took a one-year graduate program in Peasantry and state. From 1994 community fire management project with Ms. Marli Mattos, in Paragt in the Marab region, Par, with the LASAT. In 1998 she was awarded a Inter-American ith Tropical Conservation and Development, at the University of Florida. she started her und de Queiroz', University of So Paulo (ESALQ/USP). Before finishing he Paragominas region, Par state. In 1991, after finishing her undergraduate program, she Amazonian Natural Resources Management, in the University of Par to 1996, she developed a ominas, Par. In 1996, she coordinated a Community Forest Management projec Foundation and Programa Nature e Sociedade fellowships for masters-level studies, w concentration in After graduating, she returned to Par state, Brazil, to continue working with peasants. 227