“w 3,.
.1 . . L3

1.
.0

(I pt,
. 3‘4. . 3.
.u MWW« 70x

.u.
7.9%?
has...“ 4%.”...

 

35.0““...
. .31. at:

$ Fol..-

‘
ii
i

I
- z

.si

'2

T. .t
.. t.r.:)..;..
:«Kszio ¢ .
. W. -l Vll.‘ (15'
Eva... 96. 1
. .994.

 

 

 

LIBRARY
Michigan State
University

This is to certify that the
dissertation entitled

SETTLEMENT FORMATION AND LAND COVER AND LAND
USE CHANGE: A CASE STUDY IN THE BRAZILIAN
AMAZON

presented by

MARCELLUS MARQUES CALDAS

has been accepted towards fulfillment
of the requirements for the

PhD. degree in Geographx

 

 

/€at£a/ L041%V~'
Major Professor’s Signature

film/J 2Q b 5

 

 

Date

MSU is an affirmative-action, equal-opportunity employer

w..-I-O-U-I-O-I-O-I-I-I-O-C-n-o—u-n-oa—n---—--‘-—o-u-n-n—o-o-n-n-n.‘lI‘D-I-.-.-¢--.--O-O-.-O-I-o-I-C-O-l-O-O-o-I-O-l-

 

PLACE IN‘RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:/Prolecc&Pres/CIRC/DateDue.indd

SETTLEMENT FORMATION AND LAND COVER AND LAND USE
CHANGE: A CASE STUDY IN THE BRAZILIAN AMAZON

By

Marcellus Marques Caldas

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Geography

2008

ABSTRACT

SETTLEMENT FORMATION AND LAND USE AND LAND COVER
CHANGE: A CASE STUDY IN THE BRAZILIAN AMAZON

By

Marcellus Marques Caldas

Land reform in Brazil has moved into the reaches of the Amazon, affecting
the environment in colonization areas throughout the basin. Research
addressing the causes of Amazonian deforestation has implicated many factors,
ranging from the role of markets in the south of Brazil to the size of individual
farming households. Much remains to be learned, however, about the impact of
land reform on this massive process of environmental degradation. This
dissertation seeks to comprehend land cover and land use change in Amazonia
by direct reference to the underlying social and institutional circumstances that
have contributed to it. Specifically, poor farmers, who have grown disillusioned
with the Brazilian government’s promise of land redistribution, have taken it upon
themselves to form settlements, or assentamentos, in what has been referred to
as direct action land reform, or DALR.

The main objective of this research is to comprehend the social processes
leading to spontaneously-formed assentamentos, and to assess associated

impacts on Amazonian deforestation. Three hypotheses were tested about

spontaneous DALR and developed with the support of anecdotal field evidence.
Two were related to factors driving the appearance of these settlements, and one
related to the environmental impacts on the landscape. More specifically, the first
one states that spontaneous DALR occurs on unclaimed public lands beyond the
frontier, given the low land value of primary forest areas. The second
hypothesizes that spontaneous DALR is a consequence of early land reform
policies that stimulated in-migration to the region as well as from demographic
household life cycles. The last hypothesis states that spontaneous DALR
increases deforestation and fragmentation. Also, I argue that the
underlying/proximate causation (UPC) framework of land use and land cover
change to date has tended to provide considerable generalization across highly
aggregate factors, which complicates applications to specific processes of
possible interest. To achieve my goals, I conducted key informant interviews
with community leaders and government officials about the settlement location
selection process, and undertook a case study of five fully — or near fully - formed
settlements. In addition, I conducted a remote sensing analysis of deforestation
and fragmentation occurring in nine assentamentos found in the Transamazon
Highway region with the objective of evaluating their impacts on the landscape.
The results support my hypotheses, although fragmentation and
deforestation magnitudes vary among different settlements. Also, this
dissertation calls attention to the role of land law in facilitating settlement

formation.

Copyright by ©
Marcellus Marques Caldas
2008

All Rights Reserved

To Martha,

A Martha, que mais do que ninguem me deu amor, carinho e suporte na
elaboracao desse trabalho. Sem ela jamais teria realizado esse sonho.

Aos meus filhos, pais e irmaos pelo amor, apoio, incentivo e amizade.

ACKNOWLEDGMENTS

Acknowledgments are difficUlt to do, at least for me. I prefer to say them
than to write them. There are so many people that helped me to make this dream
come true that it would be impossible to cite everyone here. However, it would be
unfair from my part not to thank some of you.

I would say that l was blessed by God in 1996 when I got a call from Bob
Walker. At that time, I was a regular individual without much ambition or
inspiration. I had a simple but good life, with my wife and kids. His invitation to be
part of his team and work in the Amazon brought me a feeling that I had lost in
time, the desire to learn. Since that time I have been learning under his
supervision, and his guidance has been crucial to my development.

I would like to thank professor Dick Groop, Chairman of the department of
Geography at Michigan State University, Dr. Randy Schaetzl, Dr. Antoinette
WinklerPrins, Sharon Ruggles and Judy Reginek, who were accessible and
helped me on many occasions. I also would like to thank my committee
members, Dr. Joe Messina, Dr. Cynthia Simmons and Dr. Aaron McCright, for all
the support received during these four years as a graduate student.

To my friends Eugenio Arima, Eraldo Matricardi, Steve Aldrich, Steve
Perz, Pedro Mourao, Ritaumaria Pereira, Cindy Simmons and Bob Walker. All of

your deserve my respect and gratitude.

vi

To the people of the Amazon, specially the colonists of Uruaré, who came
to the Amazon in hope for a better life, but were forgotten in the middle of the
jungle by the government.

I would like to thank the National Science Foundation project “Socio
Spatial Processes of Road Extension and Forest Fragmentation in the Amazon”,
the NASA-LBA project “A Basin-Scale Econometric Model for Projecting Future
Amazonian Landscape”, the National Science Foundation, the College of Social
Sciences, The Center for Latin American and Caribbean Studies (Tinker
Research Grant), and the Graduate School for funding this research.

Finally, this would not be possible without the dedication of all my
fieldworkers (students). Also, I am in debt to my friends in East Lansing, specially
Bob and Cindy, Eraldo and Cleusa, Eugenio and Norma, Rita, Irvin and Romelia,

Steve and my soccer teammates.

vii

TABLE OF CONTENTS

TABLE OF CONTENTS ................................................................................... VIII
LIST OF TABLES ............................................................................................... XI
LIST OF FIGURES ............................................................................................ XII
LIST OF ABBREVIATIONS .............................................................................. XIV
CHAPTER 1: INTRODUCTION ............................................................................ 1
1.1 LAND COVER AND LAND USE RESEARCH IN THE AMAZON .................................. 4
1.1.1 Aggregate Studies ................................................................................. 6
1.1.2 Studies Addressing Individual Agents ................................................. 12

CHAPTER 2: THE HISTORIC EVOLUTION OF LAND LAW IN BRAZIL AND

ITS EFFECTS ON LAND CONCENTRATION .................................................... 25
2.1 THE SESMARIA SYSTEM IN BRAZIL ................................................................. 25
2.2 THE END OF THE SESMARIA SYSTEM AND THE REAFFIRMATION OF THE POSSE
SYSTEM ............................................................................................................ 31
2.3 AGRARIAN POLICES BETWEEN 1930 AND 1970 .............................................. 37
2.4 THE MODERN DETERMINANTS OF LAND CONCENTRATION IN BRAZIL ................ 43

CHAPTER 3: THEORIES OF SOCIAL MOVEMENTS AND THE DIRECT

ACTION LAND REFORM IN BRAZIL ................................................................ 56
3.1 THE CLASSICAL VIEw OF SOCIAL MOVEMENTS ............................................... 56
3.1.1 Mass Society Theory ........................................................................... 58
3.1.2 Status Inconsistency Theory ............................................................... 59
3.1.3 Collective Behavior Theory ................................................................. 59
3.1.4 Relative Deprivation Theory and the Davis J Curve ............................ 60
3.2 RESOURCE MOBILIZATION THEORY ................................................................ 65
3.3 POLITICAL PROCESS THEORY ....................................................................... 67
3.4 SOCIAL MOVEMENTS IN LATIN AMERICA ......................................................... 69
3.4.1 The Social Movement Organization of Rural Workers and Direct Action
Land Reform (DALR) in Brazil ...................................................................... 73

CHAPTER 4: SOCIAL CONTENTION, LAND OCCUPATION, MIGRATION AND

ASSENTAMENTO FORMATION ....................................................................... 80
4.1 SOCIAL CONTENTION IN THE AMAZON ............................................................ 82
4.2 THE CONCEPTUAL FRAMEWORK .................................................................... 84

viii

4.2.1 Macropolicies and Social Marginalization ........................................... 85
4.2.2 Migration and Network Theory ............................................................ 90
4.2.3 Settlement Formation and Land Cover Change .................................. 94

CHAPTER 5: RESEARCH AND SAMPLE DESIGN, STUDY AREA, AND DATA

COLLECTION ................................................................................................... 107
5.1 RESEARCH DESIGN .................................................................................... 107
5.2 THE STUDY AREA ...................................................................................... 116

5.2.1 A Brief Description of Southern Para ................................................ 117
5.2.2 The Transamazon Study Site ............................................................ 121
5.3 DATA COLLECTION ..................................................................................... 127
5.4 SAMPLE DESIGN ........................................................................................ 131

CHAPTER 6: RESULTS AND DISCUSSIONS ................................................. 133

6.1 DESCRIPTION OF SPONTANEOUS AND SMO SETTLEMENT CHARACTERISTICS .133
6.1.1 Settlement Farming Systems and Land Tenure Characteristics ....... 137
6.1.2 Settlement and Contention for Land Reform ..................................... 146

6.2 SPONTANEOUS SETTLEMENTS IN THE TRANSAMAZON REGION ....................... 147
6.2.1 Characteristics of Spontaneous Settlements .................................... 148
6.2.2 Land Tenure Characteristics ............................................................. 153
6.2.3 The Contentious Characteristics of Settlement Formation ................ 154

6.3 THE REGRESSION ANALYSIS ....................................................................... 159
6.3.1 Results and Discussion ..................................................................... 161

6.4 SETTLEMENT FORMATION AND ITS EFFECTS ON DEFORESTATION AND

FRAGMENTATION ............................................................................................. 166

CHAPTER 7: CONCLUSIONS ......................................................................... 181

APPENDICES ................................................................................................... 188

APPENDIX A .................................................................................................... 189

APPENDIX B .................................................................................................... 198

APPENDIX C .................................................................................................... 214

APPENDIX D .................................................................................................... 217

APPENDIX E .................................................................................................... 227

BIBLIOGRAPHY ............................................................................................... 228

 

 

LIST OF TABLES

TABLE 1. LAND REFORM SETTLEMENTS IN BRAZIL FROM 1960-2007 ......................... 23
TABLE 2. RESIDENT POPULATION, 1940 - 1970 ...................................................... 48
TABLE 3. SETTLEMENT DATES OF OFFICIAL CREATION BY THE BRAZILIAN GOVERNMENT.
................................................................................................................... 130
TABLE 4. EXISTENCE OF PREVIOUS ROADS BEFORE ARRIVAL BY TYPE OF SETTLEMENT.
................................................................................................................... 1 35
TABLE 5. SETTLER HOUSE TYPE AT ARRIVAL. ......................................................... 136

TABLE 6. SETTLERS HOUSEHOLD POSSESSION OF DURABLE GOODS AT TIME OF ARRIVAL.

................................................................................................................... 136
TABLE 7. LAND USE BY SETTLEMENT TYPE. ............................................................ 139
TABLE 8. LAND TITLE BY SETTLEMENT TYPE. .......................................................... 141
TABLE 9. AVERAGE LENGTH OF RESIDENCE IN THE SETTLEMENTS. ........................... 143
TABLE 10. How SETTLERS ACQUIRED LAND IN THE SPONTANEOUS SETTLEMENT. ...... 145
TABLE 11. How SETTLERS ACQUIRED LAND IN SMO SETTLEMENTS. ........................ 145
TABLE 12. LAND TYPE BEFORE SETTLEMENT FORMATION. ....................................... 145
TABLE 13. SETTLER INVOLVEMENT IN DALR ACTIVITY BY TYPE OF SETTLEMENT. ...... 147

TABLE 14. SPONTANEOUS SETTLEMENT CHARACTERISTICS IN THE TRANSAMAZON. ...15‘I

TABLE 15. LAND TITLE BY SETTLEMENT. ................................................................ 154
TABLE 16. SETTLER INVOLVEMENT IN DALR ACTIVITY BY SETTLEMENT. ................... 155
TABLE 17. EXISTENCE OF PREVIOUS LANDOWNER. ................................................. 156
TABLE 18. LAND TYPE BEFORE OCCUPATION .......................................................... 157
TABLE 19. LOGIT REGRESSION OF SETTLEMENT FORMATION. .................................. 165

xi

LIST OF FIGURES

FIGURE 1. BRAZIL LAND CONCENTRATION INDICES BY REGION, 1972. SOURCE: INCRA

(2007B) ......................................................................................................... 45
FIGURE 2. BRAZIL LAND CONCENTRATION INDICES BY REGION, 1992. SOURCE: INCRA
(2007B) ......................................................................................................... 46
FIGURE 3. BRAZIL LAND CONCENTRATION INDICES BY REGION, 2000. SOURCE: INCRA
(2007B) ......................................................................................................... 47
FIGURE 4. EVOLUTION OF LAND OCCUPATION IN BRAZIL, 1995 — 2005. SOURCE:
INCRA 2005 ................................................................................................. 78
FIGURE 5. THE CONCEPTUAL FRAMEWORK OF SETTLEMENT FORMATION. ................... 89
FIGURE 6. ROAD EXPANSION AND SETTLEMENT FORMATION IN TERRA DEVOLUTA. ..... 101
FIGURE 7. TWO REGIONS WITH ASSENTAMENTO FORMATION IN THE PARA STATE. ..... 111
FIGURE 8. RADIOMETRIC AND GEOMETRIC CORRECTIONS OF LANDSAT IMAGERY ....... 114

FIGURE 9. HYBRID CLASSIFICATION: DEFORESTATION AND FRAGMENTATION
CALCULATION. .............................................................................................. 1 20

FIGURE 10. LAND COVER ALONG THE TRANSAMAZON REGION, STATE OF PARA, BRAZIL

................................................................................................................... 169
FIGURE 11. PA TUTUI NORTE - NUMBER OF PATCHES, 1986 - 1999. DATE OF
CREATION: 10/29/1999 ................................................................................ 1 70
FIGURE 12. PA Turul NORTE - MEAN PATCH SIZE, 1986 - 1999. ......................... 171
FIGURE 13. PA TRAIRAO - NUMBER OF PATCHES, 1986 - 1999. DATE OF CREATION:
10/07/1997. ................................................................................................ 172
FIGURE 14. PA TRAIRAO - MEAN PATCH SIZE, 1986 - 1999. ................................. 173
FIGURE 15. PA SURUBIM — NUMBER OF PATCHES, 1986 — 1999. DATE OF CREATION:
5/18/1998. .................................................................................................. 174
FIGURE 16. PA UIRAPURU - NUMBER OF PATCHES, 1986 - 1999. DATE OF CREATION:
6/10/1997. .................................................................................................. 175
FIGURE 17. PA RIO Do PEIXE — NUMBER OF PATCHES, 1986 - 1999. DATE OF
CREATION: 10/25/1995. ............................................................................... 1 76
FIGURE 18. PA SURUBIM - MEAN PATCH SIZE, 1986 - 1999 .................................. 177

xii

FIGURE 19. PA RIO Do PEIXE — MEAN PATCH SIZE, 1986 - 1999. ......................... 178
FIGURE 20. PA UIRAPURU - MEAN PATCH SIZE, 1986 - 1999. ............................... 179

xiii

LIST OF ABBREVIATIONS

CPT: Comisséo Pastoral da Terra

DALR: Direct Action land reform

DN: Digital Number

ERDAS: Raster Graphic Editor

ETM+: Enhanced Thematic Mapper Plus

FAO: Food Agriculture Organization

FETAGRI: Federacao dos Trabalhadores da Agricultura
GPD: Gross Domestic Product

GIS: Geographic Information System

GOES: Global Observatory for Ecosystem Services

GPS: Global Position System

IBGE: lnstituto Brasileiro de Geografia e Estatistica

INCRA: Instituto Nacional de Colonizacao e Reforma Agréria
ITR: lmposto Territorial Rural

LCLUC: Land Cover and Land Use Change

MLPT: Movimento de Luta Pela Terra

MST: Movimento dos Sem-Terra

PA: Projeto de Assentamento

PCB: Partido Comunista do Brasil

PDA: Plano de Desenvolvimento do Assentamento

PIC: Programa lntegrado de Colonizacao

PIN: Programa de lntegracéo Nacional

POLOAMAZONIA: Programa de POIOS Agropecuarios e Agrominerais da
AmazOnia

PNRA: Plano Nacional de Reforma Agraria

PROTERRA: Programa de Redistnbuicéo de Terras e Estimulo a Agroindustria
do Norte e Nordeste

RMSE: Root-Mean Square Error

SMO: Social Movement Organization

TM: Thematic mapper

xiv

UPC: Underlying Proximate Causation
UDR: Uniéo Democratica Ruralista
ULTAB: Uniéo dos Lavradores e Trabalhores Rurais do Brasil

XV

CHAPTER 1: INTRODUCTION

Concern about loss of the Amazon forest has mounted over the past few
decades, given its significance as a reservoir of biodiversity and carbon, and
what appears to be an inexorableprocess of forest destruction. These concerns
have spawned a great deal of research and even international research
programs attempting to elucidate the role of AmazOnia in the global climate
system (e.g., NASA’s Large Biosphere Atmosphere program). Research
addressing the causes of forest loss (to which this dissertation seeks to
contribute) has implicated many factors, ranging from the role of markets in the
south of Brazil to the size of individual farming households. Much remains to be
learned, however, about the social and institutional conditions underlying this
massive process of environmental degradation.

Land cover and land use change has been pronounced in the Amazon
basin, and Brazil continues to experience the highest rates of forest loss in the
world.1 Although many factors have been identified as drivers of deforestation
(e.g., agricultural expansion, infrastructure expansion, technological factors,
policy and institutional factors), an increasingly important impact is associated
with colonization by smallholders and spontaneous appropriation of unclaimed
public land in the process of assentamento (settlement) formation. Between
2003 and 2005, colonist households organized 552 assentamentos in the

Amazon region, covering an area of about 15,984 ha (INCRA 2007a). These

 

1It is important to remember that recent studies have produced different estimates of
deforestation, raising questions about the accuracy of the rate of deforestation in the Brazilian
Amazon.

assentamentos represented one-third of all new settlements formed during 2003-
2005 throughout Brazil, which points to the significance of AmazOnia to Brazil’s
colonist-led direction action land reform, or DALR (INCRA 2007a). Brazilian Law
forbids assentamento formation in forestlands, but such areas are often selected
for occupation in the Amazon basin, especially the states of Para and RondOnia,
which have borne the brunt of spontaneous migration and colonization
(Fernandes 2000). Research on Amazonian deforestation — and development
— has addressed the farming systems of smallholder households (e.g., Walker
and Homma 1996; PichOn 1997; McCracken et al. 1999; Perz and Walker 2002;
Walker 2003; Pan et al. 2004; Aldrich et al. 2006; Caldas et al. 2007), the political
economy of opening the region to colonization (e.g., Schimink and Wood
1984,1992; Bunker 1985; Hecht 1985; Browder 1988,1994), and the motivations
driving colonist migration decisions (e.g., Skole et al. 1994). However, the
assentamento formation process is not well understood, given its relative
newness, as many of the assentamentos date from the middle 1990s and are still
in phases of active development. Although their landcover impacts are
necessarily linked to the farming decisions of small holders, and stem from
colonization and migration decisions, which together put potential deforesters
into the midst of forest, the specific factors leading to the spontaneous formation
of assentamentos remain unknown.

This dissertation seeks to comprehend land cover and land use change in
AmazOnia by direct reference to the underlying social and institutional

circumstances that have contributed to a source of deforestation in the Amazon

basin that has grown in significance in recent years and shows no signs of
abating. Specifically, poor farmers, who have grown disillusioned with the
Brazilian govemment’s promise of land redistribution, have taken it upon
themselves to form settlements, Or assentamentos, in what has been referred to
as direct action land reform, or DALR (Simmons 2006). This study addresses a
particular type of DALR, which I call spontaneous DALR. The goal of this
dissertation is to comprehend the social processes leading to such organically
formed assentamentos and to evaluate any associated impacts on Amazonian
deforestation.

My dissertation is divided into six chapters. In chapter 1, I consider land
cover and land use research in the Amazon basin. I include this in order to
provide the necessary context for my subsequent focus on agrarian reform. In
chapter 2, I analyze the evolution of land law in Brazil and its effects on land
concentration. In chapter 3, I focus briefly on social movement theories (seeking
to identify the historical roots, main points, strengths, and weaknesses of the
major theoretical perspectives) and their influence in Latin America. In addition, I
approach the social movement organizations of rural workers and the DALR. In
chapter 4, I analyze social contention, land occupation, and assentamento
formation along the Transamazon region. In this chapter, I describe the
contentious process of assentamento formation and place this within a land
cover and land use change framework by describing the underlying processes
that lead to spontaneous assentamento formation. In chapter 5, I present a

discussion of the research and sample design, the study area, and data

collection. In chapter 6, I present the results and discussion. This includes
descriptive statistics and a regression model of assentamento formation. Also, I
analyze the effects of settlement formation in the landscape through land cover
changes and fragmentation metriCS. The final chapter is a presentation of my

concluding remarks.

1.1 Land Cover and Land Use Research in the Amazon

The causes and processes that drive deforestation in tropical countries
have attracted the attention of social scientists for a long time. In reality, since the
beginning of the 19805, many attempts have been made to explain tropical
deforestation throughout the world (e.g., Tucker and Richards 1983; Allen and
Barnes 1985; Walker 1987; Richards and Tucker 1988; Rudel 1989; Burns et al.
1994; Williams 1990, 1994; Kimsey 1991; Stem et al. 1992; Brown and Pearce
1994; Bilsborrow 1994; European Space Agency 1994; Lambin 1994, 1997;
Krutilla et al. 1995; Capistrano and Kiker1995; Painter and Durham 1995;
Bernard and Koninck 1996; Sponsel et al. 1996; Rudel and Roper 1996, 1997;
Kant and Redantz 1997; Murali and Hedge 1997; Bawa and Dayanandan 1997,
1998; Amsberg 1998; Ehrhardt-Martinez 1998; Kaimowitz and Angelsen 1998;
Angelsen and Kaimowitz 1999; Wunder 2000; Irwin and Geoghegan 2001;
Lambin et al. 2001; Geist and Lambin, 2001, 2002; Lambin et al. 2003; among
others). These studies represent an inexhaustible account of the subject and
have produced many arguments about what drives deforestation. In considering
the specific Case of the Brazilian Amazon, the topic of this Ph.D. dissertation,

previous studies have pointed out that government policies, logging, mining,

agricultural development, and migration (among other factors) are major drivers
of deforestation (Browder 1988; Mahar 1989; Serréo and Homma 1993; Serréo
et al. 1996; McCraken et al. 1999). In general, these studies have used different
conceptual frameworks, primarily from economics or political economy. If, on the
one hand, studies using economic frameworks are more concerned with the role
of market over deforestation and are centered in quantitative models; on the
other hand, political economy is more a commentary on the likelihood of social
welfare impacts of Amazon development (e.g., Forewaker 1981; Sawyer 1984;
Schimink and Wood 1992). Although it is possible to note that approaches based
on political economy are not very common in explaining the causes of
deforestation in the Brazilian Amazon, there are some exceptions (Schimink and
Wood 1984; Hecht 1985; Browder 1988). Consequently, whatever approach is
used, the causes of deforestation continue to be an intriguing question. For
instance, recent land occupations in the Amazon region by landless workers
have lead to the appearance of new land reform settlements; and little attention
has been paid to either the creation of these settlements or their possible impacts
on land cover. Taking this in consideration, I now undertake a review of the
deforestation literature in the Amazon, with the Specific goal of demonstrating
that land reform in the Amazon region is a new and important cause of
deforestation and, as such, has not yet received enough attention.2

I would like to start by highlighting that the first deforestation studies were

based on analyses at aggregate scale, and the data used were collected for

 

2 For general reviews addressing deforestation studies that have been undertaken, see
Kaimowitz and Angelsen (1998), Geist and Lambin (2001, 2002).

5

geographic units. In general, these first efforts to understand deforestation
sought to explain land cover change at aggregate scale, calculating deforestation
occurring over some time period as a function of socioeconomic and physical
variables. Also, the data used in this type of analysis were acquired for
jurisdictional units using government censuses on forest area and socioeconomic
indicators (Allen and Barnes 1985; Rudel 1989; Barbier 2001 ). Later, models
based on the behavioral theory of individual agents implemented analyses using
disaggregate data. Thus, in light of these two basic approaches, this literature
review is divided into two parts: the first considers studies of deforestation that
used aggregate data (Appendix A). The second part focuses on studies of
individual agents and shows in Appendix A the diversity of variables used in this

type of analysis.

1.1.1 Aggregate Studies

The literature often partitions the causes of deforestation into two types:
indirect or underlying causes and immediate or proximate causes (Kaimowitz and
Angelsen 1998; Geist and Lambin 2002). Proximate causes are defined as those
human activities that directly affect the environment (Geist and Lambin 2002).
The underlying causes can be conceptualized as broader economic, cultural,
political, technological, and demographic forces that underpin proximate causes
(Kaimowitz and Angelsen 1998). The empirical research on deforestation that
uses geographically aggregate data generally approaches the issue of land cover
change through proximate causation frameworks, where the variables used can

be seen as immediate (or direct) factors in a land use decision and,

consequently, on land cover change. In general, the data are grouped in different
categories. For instance, Geist and Lambin (2001) grouped proximate causes of
deforestation in three categories: agricultural expansion (e.g., shifting cultivation,
cattle ranching, permanent cultivation), logging (e.g., fuelwood extraction,
charcoal production, commercial wood extraction), and infrastructure expansion
(e.g., transportation infrastructure, settlement expansion, market infrastructure).
Despite the fact that no general rule exists about how to group the variables, in
this dissertation I group the causes of deforestation using five general categories:
market factors, socioeconomic and demographic factors, institutional factors,
environmental factors, and control variables (see Appendix A). The variables
grouped as market factors are described as the ones that affect production costs
directly or indirectly. In this context, transportation costs and land prices are, in
general, the most commonly used. The influence of market factors on
deforestation in the Amazon region associate the land use decision with
transportation costs or to a proxy (e.g., distance). Nelson and Hellerstein (1997)
discussed the development of new roads as intensifying the pace of
deforestation by making access to forested lands less costly. Empirical models
that use this approach Show that easy access to forest lands, and to the market,
increase deforestation (Reis and GUzman 1994; Andersen 1996; Pfaff 1999).
However, proximity to market is not the only opportunity that creates land cover
change. For instance, a given tract of land has many possible useful resources
linked with it; consequently, these resources will (assumably) be used if it is

lucrative (Nelson and Hellerstein 1997). According to this assumption, the choice

of a particular land use is made by comparing the net present value of the
profitability of all possible land uses. Chomitz and Gray (1996), Nelson et al.

(2001), and Walker (2004) hypothesized that a set of h land uses generates

rents, or net present values, Rh, Where h C h and rents are defined on prices for

inputs and outputs and production magnitudes. The observed land use on a tract
of land is given as h, when Rh > Rh; for all h at h’ (Walker 2004). Consequently,
these rents can be acknowledged as functions of distance. Thus, following

Walker (2004), Nelson et al. (2001), and Nelson and Hellerstein (1997), the net

present value for land use h at location iand time period T can be defined as

w 0
_ _ —zt
Rth " Xmax t—IO (Pth+tthT+t Zth+tXth+t)e dt
th+t

where P is the output price, Q is the quantity of output, Z is a vector of input
prices, X is a vector of inputs under the land manager control, and iis a

discount rate, all for each land use h at location iand time t. Walker (2004)
argues that time invariance in prices converts this static maximization problem

suchthat

00 _.t
Rhl = mat};(Ph1th ‘ Z th hl)t I5 I d‘
hl

Consequently, the land use chosen for the tract of land has the highest ha value.

However, access to a road is clearly not the only variable that increases

deforestation. The simple existence of roads has different effects on land cover,

8

L /_

facilitating or constraining the access to primary forest areas (Southgate et al.
1991 a; Reis and Guzman 1994; Andersen et al. 2002). The correlation between
distance to roads and deforestation assumes that roads cause deforestation.
However, deforestation can be an important factor in attracting the development
of a road system. In other words, deforestation and progress of local economies
can cause new infrastructure creation, such as roads, by stimulating a
socioeconomic and political momentum for advancing road creation (Chomitz
and Gray 1996). In addition, deforested areas and roads can simultaneously be
influenced by other factors, such as soil quality or population density (Southgate
et al. 1991b; Wood and Skole 1998; Pfaff 1999). Moreover, considering this
market perspective, other factors can influence land cover change; for instance,
land price. Besides the lack of research with aggregate data using this type of
variable, there is an exception (Andersen et al. 2002).

Market factors are not the only factors influencing land cover change.
Socioeconomic and demographic variables are also used in deforestation
models. The argument commonly used states that deforestation can also
increase because an increasing population needs more land and derivative
products. In this context, demographic factors make the list of variables much
larger; and these variables, in the majority, are related to population factors, such
as population size. These studies argue that a large population size has a
positive effect on deforested areas (Southgate et al. 1991a; Andersen 1996).
This result is not surprising as other authors found similar results for other Latin

American countries, such as Mexico (Barbier and Burgess 1996). However,

when this analyzes combines many countries simultaneously, and additional
independent variables are used, the correlation between deforestation and
population density disappears. In this context, the population variable could be
acting as a proxy for others (Harrison 1991; Deacon 1994). Moreover, in some
models, population growth has no effect on deforestation (Palo 1994). However,
demographic factors are not restricted to population Size. Many studies have
used population density and migration. According to these studies, the higher the
population density, the greater the amount of deforestation (Wood and Skole
1998; Pfaff 1999). In addition, these studies noted that rural migration has a
positive influence on deforestation (Wood and Skole 1998).

Another important group of variables used in land cover models is related
to environmental factors. In other words, considering the important role that
natural resources have on economic activity, many authors have already
described their effects on the forest. For instance, Southgate et al. (1991a) and
Pfaff (1999) Showed that soil type affects the size and proportion of area
deforested for agricultural use. Also, Wood and Skole (1998) argued that
deforestation is less in municipalities that have a large proportion of land in
savanna; and Chomitz and Thomas (2003) stated that precipitation has a
negative effect on the proportion of land used for agricultural purposes. In
addition to the variables discussed above, there is a group of variables that
attempts to capture the effect of institutional factors on land cover change. In
general, these variables seek to evaluate the influence that land security, credit,

and fiscal incentives have on land use decisions (Southgate et al. 1991a; Pfaff

1O

1999; Andersen et al. 2002). In the absence of well-defined property rights
deforestation becomes the most common way to acquire land rights. Besides,
property rights stimulate investments; and if the investment involves land cover
change, deforestation can increaSe as a result. Despite important analyses in
land cover change, these factors are difficult to evaluate with aggregate data as it
is very complicated to acquire precise information. Consequently, few studies
with aggregate data have analyzed the effect of property rights on land cover
change. However, institutional factors are not constrained to property rights.
Many government institutions have given credit and fiscal incentives in the
Amazon that have been pointed out as primary drivers of deforestation (Pfaff
1999; Andersen et al. 2002). Finally, a group of variables have been used as
control variables. These variables are used to control, or isolate, the effects of
development. Among these variables are the previous amount of deforestation,
property density, number of bank agencies, and land conflict. No doubt that In
the last century, land conflict was an important driver of sociopolitical and
environmental change (Homer and Dixon 1999); and considering that property
rights in the Amazon region are strongly related to conflicts, Wood and Skole
(1998) argued that there is a positive relationship between conflict and
deforestation.

Clearly, aggregate studies have identified many important factors driving
deforestation. Nevertheless, the research framework has changed over time and
now considers deforestation as a discrete event affecting individual parcels of

land where potential uses are evaluated economically (Walker 2004). Thus, the

11

decision unit involves individual parcels of land and discrete changes between
land cover change categories. In addition, the availability of remotely sensed
information, and the functionality of geographic information systems, has largely
facilitated these types of analyses because they allow an accurate gathering of
information on land cover Change and enable spatial projections at pixel levels
(Walker 2004). Consequently, the analysis of land cover change is adapted to
include the analysis of the behavior of individual agents (Bockstael 1996;

Chomitz and Gray 1996; Nelson et al. 2001; Walker et al. 2002; Walker 2004).

1.1.2 Studies Addressing Individual Agents

The types of individual agents involved in deforestation in tropical
countries varies from region to region and generally includes loggers, shifting
cultivators, farmers, ranchers, and colonists, among others. These agents act
interdependently, and deforestation is a result Of their complex interactions
(Walker 2004; Geist and Lambin 2001).

Similar to aggregate studies, the variables that affect the decisions of
individual agents can be divided into market factors, environmental factors,
socioeconomic and demographic factors, and institutional factors (see Appendix
A). The importance of market factors is observed in both approaches, and
transportation costs affect deforestation in both cases. According to individual
agent studies, distance to market has a negative impact on deforestation (Caldas
et al. 2007; Walker et al. 2002; PichOn 1997). Also, the use of variables that
measure the effects of natural resources on the landscape are not so common in

models of individual agents because this type of data is difficult to acquire.

12

However, a few authors that have used this type of information (Caldas et al.
2007; Pan et al. 2004; McCraken et al. 1999; PichOn 1997). PichOn (1997), for
instance, argued that a relationship exists between land quality and
deforestation. According to this author, better soil quality can lead to greater
deforestation. In addition, he showed that topography is also an important factor
in land use decisions, where properties with hilly land and steep slopes have
more area in forests. On the other hand, Pan et al. (2004) stated that flat land is
important to annual crops and pastures and, by extension, to deforestation.
McCraken et al. (1999) argued that previously deforested areas have a
significant impact on the amount of land deforested annually. Caldas et al. (2007)
showed that soils suitable for pasture (and with water) can positively affect the
total area deforested. However, their results are not sustained when spatial
autocorrelation analysis is implemented.

Considering the effects of demographic factors on individual agent
models, one can observe that migration is an important factor in the development
of the Amazon frontier. Consequently, many studies have used a set of
demographic variables to explain deforestation. Most of this research has shown
that family size (Rudel and Horowitz 1993; PichOn 1997; Coomes et al. 2000;
Coomes 2004), number of men in the household (Sydenstricker and Vosti 1993;
Walker et al. 2002; Pan and Bilsborrow 2005), and dependency (Walker et al.
2002) have an impact on agricultural systems or on the amount of land to be
deforested. However, other attributes are also important, such as level of wealth

(Alston et al. 1993; Jones et al. 1995; PichOn 1997; Murphy et al. 1997; Walker

13

et al. 2002) and length of residence on the property (PichOn 1997; Walker et al.
2002). Clearly, the analysis shows that variables that seek to measure the effects
of family structure (e.g., number of men, women, children, and elderly) are, in
general, associated with labor availability. Consequently, large deforested areas
appear on properties that have families with substantial family labor resources,
including hired labor (Rudel and Horowitz 1993; Godoy et al. 1997, 1998a,b;
PichOn 1997; Walker et al. 2000; Pan et al. 2004). In addition, deforestation also
relates to the length of residence on the property, education level, and age of the
household head (Alston et al. 1993; Godoy et al. 1997, PichOn 1997, Godoy et al.
1998b). Families with longer periods of residency have deforested larger areas.
However, Pichén (1997) called attention to the difficulty of isolating the effects of
length of residency from the overall effects of soil degradation.

Even though demographic and environmental factors are indicated as
drivers of deforestation, institutional factors (such as credit and land security) are
also important. In other words, forest conservation is positively associated with
land security (Alston et al. 2000; PichOn 1997; Muller et al. 1994). Nevertheless,
disagreements regarding this view emerged. Pan et al. (2004) argued that land
security is also associated with large areas of pasture and less area in perennial
crops. These contradictory views can be related to credit acquisition or land
speculation. Walker et al. (2000) agreed with the statement above and argued
that land security can be a facilitator in credit acquisition and, consequently, can

be used for pasture formation.

14

The above literature review has demonstrated that plentiful research and
systematical methods have focused human-environment studies on clearing
patterns, which have identified small and large producers as a primary proximate
cause of land cover change (Geist and Lambin 2001, Walker et al. 2000).
However, Pan and Bilsborrow (2005) argued that as the compilation of data to
predict changes in clearing have become complex, operating at multiple temporal
and spatial scales, the statistical models applied to test theoretical relationships
have not been developed. These authors argue that many theories of land use
follow a structural approach in which the conceptualization of land use by
households is seen to occur within a larger institutional context and that existing
models fail to adequately capture the effects of either household or community-
level socioeconomic factors that may influence land cover and land use change
(LCLUC). In addition, they argued that previous research neglected to
incorporate factors influencing land use at higher institutional levels or did not
account for the clustering of observations (e.g., farms) within higher level units
(e.g., communities) resulting in inefficient estimates of factors influencing land
use. Thus, to accomplish this task, they incorporated a theoretical model of land
use with applied statistical technology because theoretical land use models often
combine factors at multiple temporal and spatial scales to illustrate land use and
land cover change. Their results can be divided in two parts. The first part
approaches the traditional variables used in household models (e.g.,
demography variables) and in the second approach community variables (e.g.,

civil registration office) are used. Pan and Bilsborrow (2005) showed that higher

15

population pressures are associated with less forest and more land allocated to
all other uses. Hiring labor and holding title to land, which indicates higher
socioeconomic status of the household on the farm, appears to be strongly
related to less forest and more pasture. Also, better land quality is associated
with less land in forest. The presence of a civil registration office indicates
communities that are political centers (e.g., proxy to community development),
and the presence of a civil registrar was significantly associated with more
pasture and less fallow land because farms located near major cities were less
likely to have fallow land when there is a ready local market. Their results
demonstrate the need to more comprehensively explore various community and
structural effects on land use.

The above literature review calls attention to potential problems in using
survey data in land cover change analysis, leading some authors to suggest the
use of independent sources of information, such as the ones generated by
remotely sensed data (Walker et al. 2000). The existence of potential bias and
the difficulty in obtaining accurate information on land cover change creates a
greater interest in georeferenced data, such as remotely sensed data. However,
high-resolution satellite data provide an empirical basis for measuring the amount
and the spatial configuration of forest clearing, but satellite data do not explain
the causes of deforestation (Wood and Skole 1998). Thus, concern over the
limited ability to explain the social causes of deforestation have increased the
attention of social scientists. In this context, Rindfuss and Stern (1998) argued

that there is increased interest in making scientific progress through the use of

16

remotely sensed data in social science research, and Geoghegan et al. (1998)
stated that the impacts of Remote Sensing and Geographic lnforrnation System
(GIS) to date have been greatest within the environmental and policy arenas
because space-based and other imagery is used primarily to determine the
physical attributes of the biosphere and the earth’s surface (such as forest cover)
that is needed in spatially explicit form by various decision makers.
Consequently, many studies have used this approach (Walsh and Crews-Meyer
2002; Fox et al. 2003; Boucek and Moran 2004; Gutman et al. 2004; Entwisle
and Stern 2005). However, in the case of the Brazilian Amazon region, only a
few researchers have combined georeferenced information with data collected at
the household level (Moran and Brondizio 1998; McCracken et al. 1999; Walker
et al. 2000; Aldrich et al. 2006; Caldas et al. 2007).

Moran and Brondizio (1998), for instance, link social science and
biological field methods with remotely sensed data to advance understanding of
how human decisions about land use have influenced both rates of deforestation
and secondary regrowth in AmazOnia. McCracken et al. (1999) looked at the
influence Of environmental and market factors on deforestation at the property
level and demonstrated that distance has a negative effect on land cover change.
Their results also affinTIed that a given property with less pasture has more areas
in secondary and primary forest. Walker et al. (2000) used satellite imagery to
measure deforestation to discuss the relationship between ranching and
deforestation in the Amazon, and they argued that large ranchers can be

responsible for a greater share. But they also reinforced the idea that small

17

farmers, through proximate causation, are also responsible for a substantial
portion of deforestation. Aldrich et al. (2006) presented analyses based on
cadastral information, a panel of smallholders collected in 1996 and 2002, and
satellite image analysis of land cOver change of the Transamazon region from
1986 to 1999. These authors sought to analyze the structural processes of land
aggregation by ranchers along the Transamazon highway. The panel data
analyses demonstrated that property aggregation is largely compensated for by
disaggregation in the study area; and when it occurs, it does not appear to result
in the formation of large, well-capitalized ranches. Instead, the aggregation
process that does exist arises from the acquisitive interests of the local
smallholders themselves and may be regarded as endogenous to the
colonization process; In other words, wealthy outsiders do not appear to be
penetrating the region and appropriating the land of subsistence farmers (Aldrich
et al. 2006). Colonization is also approached by Caldas et al. (2007). These
authors seek to understand the land cover impacts of an important agent in the
Amazonian landscape, the household colonist. The main objective of their
research was to develop a peasant economy perspective on deforestation in
tropical forest frontiers and to apply it in a study of colonist households living in
the Brazilian Amazon along or near the Transamazon highway. The information
used in their study was generated by a GIS linking household survey data and
classified satellite images. They paid particular attention to statistical issues
affecting regression-based, land cover change studies, including endogenous

relations between key variables and spatial autocorrelation. Their results

18

demonstrate that deforestation is a spatially auto-correlated process. In addition,
their statistical findings point to the importance of household structure and
expected land values (reflecting market access) on deforestation. In this regard,
they drew two conclusions. First, household labor endowment is the most
important demographic variable in the deforestation process. Their results
showed that among the household life-cycle variables age of household head
had no impact on the magnitude of deforestation (Walker et al. 2002). Also, they
stated that length of residence was statistically insignificant, thus, confirming the
results of previous studies (Alston et al. 1993; Almeida and Campari 1995). Their
second conclusion was that the simultaneous significance of household and
market variables illustrates the duality of the peasant household, reliant as it is on
family resources but hopeful of market success (Turner and Brush 1987; Walker
2003). Besides the fact that satellite imagery has been used to measure
deforestation in the Amazon region, new studies are analyzing changes in the
landscape through the use of pattern metrics, such as patch, density, contagion,
and landscape shape index. For instance, Pan et al. (2004) demonstrated the
integration of spatial information (i.e., remote sensed images) with household
survey data to statistically explore different models of landscape complexity in
two different time periods. More specifically, they explored the relevance of
spatial patterns of LCLUC at the farm level and how socioeconomic,
demographic, biophysical, and geographical vaIiables explain the observed
variation in land cover and land use patterns using a set of landscape metrics.

Their results showed that rapid population growth is causing substantial

19

subdivision of plots, which in turn has developed a more intricate and fragmented
landscape (e.g., patch density increase). Key factors forecasting landscape
complexity are population size and composition, plot fragmentation through
subdivision, expansion of roads, and topography.

As indicated, many researchers have studied LCLUC in the Amazon
basin. These studies have included econometric estimation, spatially explicit
models, and household level research; they have ranged from basin-scale
analyses to highly circumscribed locality studies. The factors that have been
used in a statistical modeling framework cover a very wide range and include
variables that can be grouped under those reflecting market factors (e.g.,
distance), environmental factors (e.g., soil quality, vegetation), socioeconomic
and demographic factors (e.g., GDP, poverty, population density), and
institutional factors (e.g., credit, land security). However, much of the research to
date overlooks underlying social and institutional factors affecting land cover
change. This can be a serious omission, particularly in Brazil where the federal
government's role in Amazonian deforestation processes has involved
geopolitical considerations, the provision of sweetheart subsidy deals for wealthy
landowners, and efforts to mitigate rising populism directed at poverty and land
reform.

In sum, the research on deforestation in the Amazon basin is helping to
explain the effects of anthropogenic change in the landscape. However, the
literature shows a lack in an important social phenomenon that today affects the

Amazon landscape: the formation of land reform settlements. Brazil is well known

20

as a relatively rich country, and its GDP is one of the largest in the world.
However, today most Brazilians are living in poverty, and Brazil is among the
countries with the highest income inequality worldwide (Rangel et al. 2007). No
doubt that there are many causes for poverty in Brazil, and land concentration
has been pointed out as a very important one (Cline 1970; Roumani and Coirolo
2005). Brazil has a landless population of 5 million families that are looking for
land in every part of Brazil, and the Amazon region is no exception. The
importance of this phenomenon can be demonstrated by the number of new
settlements in the northern region of Brazil where it is estimated that 1,916 new
settlements occupy an area of 5.4 million ha (INCRA 2007a). These new
settlements represent almost one fourth of the total in the whole country, which is
estimated at more than 7,694 settlements ( Table 1).

The creation of land reform settlements is an important form of land
redistribution, and it has already benefited many poor people in Brazil. In the
case of the Legal Amazon, in just eight years (from 1994 to 2002), the number of
families living in land reform settlements went from 161,000 to 528,000 families
(INCRA 2003). However, land use and its consequences on land cover can
generate serious problems for the environment. Despite the importance of the
problem, few authors have addressed the subject; and those that have
addressed it have done so only superficially. Van de Steeg et al. (2006), Brandéo
and Souza (2006), Simmons et al. (2002), and Feamside (2001) have brought
attention to the effects of social processes on the landscape and the potential

impacts that land reform settlements can have on the forest. Brandéo and Souza

21

(2006) combining deforestation data with the map of settlements created from
1970 to 2002 by the Brazilian Institute of Colonization and Land Reform (INCRA
— Instituto Nacional de Colonizacao e Reforma Agréria) affirrn that around
106,000 km2 (49% of the area of Settlements mapped) were deforested up to
2004, representing 15% of all Amazon deforestation. In addition, these authors
argued that from 1997 to 2004, the rate of deforestation in these settlements was
1.8% per year. Besides that, these authors stated that the rate of deforestation in
land reform settlements is four times greater when compared to the mean rate of

deforestation in the Amazon as a whole.

22

Table 1. Land Reform Settlements in Brazil from 1960-2007

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Settlement . .
States (Alrleae)I Capacity # £23255 Settlefnents
(13 Families)
Para 18,532,161.49 255,534 191,954 893
Mato Grosso 5,669,978.80 91,174 75,151 542
Acre 5,645,299.07 30,709 35,590 131
Amazonas 17,934,105.58 41,518 34,661 90
RondOnia 5,679,956.96 73,879 40,275 151
Amapa 2,095,394.61 12,190 10,598 37
Roraima 1,543,063.00 19,790 16,755 48
Tocantins 1 ,153,576.06 22,619 26,444 338
Sum 58,253,535.57 547,413 431,428 2,230
Rest of the
13,051 ,295.43 443,772 390,856 5,464
Country
Brazil 71 ,304,831.00 991,185 822,284 7,694

 

Source: INCRA (2007a)
Note: These numbers include colonization settlements created between 1960 and 1970

Feamside (2001) is more dramatic and has suggested that 30% of

deforestation in the Amazon can be attributed to land reform programs. In

addition, he argued that the continuing migration of the landless population to the

Amazon region makes unviable any plan to support the region’s present rural

population through agrarian reform. Simmons et al. (2002) expressed that it was

surprising that little attention has been given to the role of social and political

movements as agents of land cover change and the impact of land reform on the

23

 

landscape. Van de Steeg et al. (2006) argued that the cultivation of legal
reserves (and deforestation) are the main issues related to environmental
impacts of settlement formation. However, none of them explained how
settlements appear, which factors influenced their formation, and what impacts
they had on the landscape.

The above review demonstrates that land reform is a potentially significant
driver of Amazon deforestation. However, land reform is a consequence of land
concentration in Brazil. From the very beginning of Brazil’s colonization, wealth
has been largely associated with land; and as such, to understand settlement
formation it is first necessary to have an historical perspective on land

concentration and land law in Brazil, a tOpic that I will develop in the next chapter.

24

CHAPTER 2: THE HISTORIC EVOLUTION OF LAND LAW
IN BRAZIL AND ITS EFFECTS ON LAND
CONCENTRATION

Land concentration in Brazil has many reasons for its existence. Since the
start of the Brazilian colonization, many land laws were created with specific
objectives. In the beginning, the land laws were created to protect the
Portuguese land claims in the New World. Later, the land laws were created to
explore the colony. As the World economy developed, SO did the interests Of the
Portuguese Crown. Thus, the necessity to expand the colony economy
stimulated a disorganized territorial expansion and land acquisition in Brazil. This
problem of unauthorized land occupation generated a land concentration that
was never fully resolved. On the contrary, the land laws created to solve this
problem just aggravated land concentration.

This chapter reviews the historical evolution of land law in Brazil and its
effects on land concentration. The main objective of this chapter is to
demonstrate how land laws elaborated since the colonial times favored a

property system that is based on land concentration.

2.1 The Sesmaria System in Brazil

Historically, land concentration in Brazil has many reasons. Land law in
Brazil is derived from the system of Sesmarias, created in Portugal at the end of
the fourteenth century. During this period, land that was part of the dominion of
the Portuguese Crown was governed by the feudal system, leading much Of it to

remain uncultivated (Rau 1982). The basic objective of the Sesmaria system was

25

to do away with unused lands by making it obligatory to cultivate the land or risk
losing control over it. As part of this system, landlords who did not cultivate or
sublease their land would lose their land rights, and the land would be returned to
the Crown to be redistributed to Someone who would make it productive and
respect the collective interest.

The need to explore and defend Portuguese land claims in the New World
led the Portuguese to apply the Sesmaria system in Brazil, beginning in 1530. To
do so, the King of Portugal granted land to Portuguese nobles who were deemed

capable of developing and defending these lands with their own resources.

These land grants were called Capitanias Hereditérias.3 Further, because of the
importance of trade to the Portuguese, the king authorized these nobles to
subdivide their granted lands among other nobles. The new grants were called
Sesmarias. However, the Sesmaria system did not operate as planned.

In Portugal, land structure of the sixteenth century allowed the Sesmaria
system to create smaller properties than was possible under the older feudal
system. In Brazil, the Sesmaria system was the primary impetus for the
appearance of latifdndio (latifundia)4 (Costa Porto 1978). However, the Sesmaria
system cannot be solely blamed for the growth of latiftindio in Brazil. Silva (1996)
argued that the historical conditions were different in colonial Brazil than in

Portugal. According to Silva, the primary characteristic that differentiated colonial

 

3 The Crown divided Brazil into 14 sections every 50 leagues (a league was a measure of
distance that varied from country to country) along the Atlantic coast with boundaries extending
westward. Each section (capitanias) was given to a captain who had complete authority over his
land and who responded only to the Crown.

4 Latift’rndio means a very large property. It comes from the Latin Latifundium where Létum
means large, spacious, and fundus means farm, estate.

26

lands from lands in Portugal was that colonial land was, more or less, empty. In
other words, these lands were without a landlord, inhabited only by indigenous
peoples with no concept of Portuguese land rights.

The lack of settlement in the New World was responsible for an
unintended interpretation of one of the central terms of the Sesmaria system: the
expression terras devolutas (Silva 1996). The original meaning of the term
devoluto was “to be returned to the original owner [devolvido ao senhor originalj"
or, more simply, to be returned to the Crown. Land granted or appropriated that
was not productive would be returned to the Portuguese King. According to this
notion, even those sesmarias in the colonies that were unproductive would be
returned to the Crown. However, over time, terra devoluta came to be used to
represent all unoccupied or unused land, or land considered vacant (Silva 1996).
The consequence was that lands termed devoluta were redistributed among the
subjects of the capitanias instead of being returned to the dominion of the
Portuguese Crown. The Crown, wishing the productive use and occupation of the
colony, sought to grant lands to nobles with resources. However, the financial
difficulties of the Crown in the middle of the seventeenth century and the
possibility of the commercial exploitation of the colony through sugarcane
plantations (which demanded large extensions of land) led the Crown to “close its
eyes" to the way that land was granted in the colony.

The redistribution of unused lands in the middle of the seventeenth
century among the lesser nobility in Brazil (e.g., friends of the Crown or anyone

who claimed to have the means and desire to make use of the land) gradually

27

caused land concentration. In the beginning, besides the Crown
recommendations, the Sesmaria system did not have any set of property size
limits, at least not until the middle of the seventeenth century. Also, the methods
used to demarcate land were rudimentary. The delineation and description of
properties was largely based on natural features (and remained used until the
nineteenth century), and the practice to petition many sesmarias by the same
noble (or friend of the Crown) was not uncommon. Thus, there were cases of
many sesmarias granted to the same individual (called sesmeiro), and cases
where the individual’s petitioned sesmarias in the name of their sons or
daughters (Mendonca 1976). The sesmarias were clearly not homestead
properties, and a labor force would have to be introduced. The solution to this
labor need was found first through the use of indigenous people and, later, in the
introduction of slave labor from Africa. However, by the end of the seventeenth
century and the beginning of the eighteenth century, the colony of Brazil
experienced some modification that had repercussions over the way land had
been granted. One factor that contributed to the modification of the colony was its
own growth. This growth was not restricted to population Size, but was also an
economic and territorial growth. In other words, migration flux increased from
Portugal to the colony, the importance of the colony increased in the world
market, and the frontier area that was formerly restricted to the coastal area of
the colony expanded. The expansion of the new frontier led to the discovery of
gold in the south-central area of the colony, thus redesigning the initial

occupation. The combined increase in the slave population coupled with colonial

28

growth (population, economy, and territorial) stimulated and forced a large part of
the population of the colony to migrate to regions of the country where land could
be had and mines could be established. The territorial expansion of the colony in
the eighteenth century highlighted the most important modification that occurred
since the beginning of colonization: the dissemination of another form of land
appropriation, the posse system.

In the first two centuries of colonization, posse represented a form of
occupation by small farmers without the necessary conditions to maintain a
sesmaria. In other words, the posse system involved informal rights to use land
by squatting on it and enforcing land claims through social norms (e.g., by
showing the use of land). In a supposed latifundiario regime, poor colonists were
so economically constrained that migration to distant areas with available land
(often with absent landlords) began to occur. As described before, with the
evolution of worid commerce, the Portuguese Crown developed a keen interest
in the cultivation of sugarcane, which demanded grand expanses of colonial land.
This led the king of Portugal to ignore the irregularities of the sesmaria system
and, in particular, the extensive land grants under the system that remained
unused. These unused areas were the overwhelming recipients of those forced
to migrate because of land concentration in coastal areas. This occupation of
unused sesmaria land by poor colonists created a problem for colonial authorities
who sought to use this land for the cultivation of sugarcane. Thus, the poor
colonists who had occupied the previously unused portions of the sesmarias

were considered to be in the way of the Portuguese Crown.

29

This problem of unauthorized land occupation persists today and is related
to the imprecision of property boundaries. The delineation and description of
properties was largely based on natural features, a process that was common in
the colonial era. Complicating the problem, the registration of property at that
time was imprecise, creating many conflicts between smallholders and landlords.
The Alvaré of 5 January 1785 states that the Sesmaria system should have to
conform to three explicit basic conditions: agricultural production, measurement,
and confirmation (Alvarenga 1985). This means that the sesmeiro had to
cultivate, demarcate, and obtain from the authorities recognition of the
measurement of the land. However, people specialized in this type of work were
very difficult to come by and the land was very hard to penetrate. Thus,
measurement was not effective, and the use of natural features was used to

demarcate the boundaries of a property.

By the end of the seventeenth and the beginning of the eighteenth
century, the posse system increased in its impact, acting outside of the control of
the Portuguese Crown. One important characteristic of the posse system was
that, like the sesmarias themselves, the boundaries of a posseiro (the “owner” of
a posse) were defined by the posseiros themselves. Following the posse system,
large and small areas were demarcated throughout Brazil. Although in principle
posseiros began as small producers, they frequently became large landholders
(Rios 1950).

AS a result of these developments, the posse system (which had always

existed in the colony) began to cause severe problems. The colony already had

30

issues related to irregularly defined land grants (sesmarias), and the posse
system added to these problems. Under the sesmaria system, the colony needed
to grant new lands for sugar production. Thus, a significant risk that the
boundaries Of newly granted lands would overlap with the boundaries of the
lands already claimed by a posseiro, or even areas of sesmaria started to
appean

In sum, this history suggests that the distance of the colony from Portugal
led to a lack of enforcement by the Crown regarding the size of sesmarias. In
addition, the economic development of world commerce led the Portuguese
Crown to “close its eyes” to the way that land was being accumulated in the
colony in the hope that it would stimulate sugarcane plantations. This land
accumulation, either under the sesmaria or posse system, contributed to the
formation of latifUndio in the colonial period. However, by the time of Brazil’s
independence in 1822, the difficulty in monitoring sesmarias and the conflicts
between posseiros and landlords led to the suspension of the sesmaria system

(Vasconcellos 1860).

2.2 The End of the Sesmaria System and the Reaffirmation of the
Posse System

With the abolition of the sesmaria system, the posse system became the
most common means of acquiring land in colonial Brazil (Lima 2002). On the one
hand, the sesmaria was IatifUndio, a large landholding inaccessible to colonists
without reso-urces. On the other hand, posse (at least in the beginning) was a

small landholding created by necessity. Once posseiros marked their lands with

31

the “naked eye" to match their desires, the lands under their control became as
extensive as the sesmarias (Lima 2002). Because of their extensive
landholdings, the colony sought to remedy land acquisition and regularization by
changing the agrarian policy.

Brazil’s newly achieved independence in 1822, the accompanying
changes in political power, and an economic expansion largely driven by coffee
cultivation created conditions favorable to address the issues surrounding
landholdings that Brazil had inherited. The expansion of the coffee industry in the
first half of the nineteenth century had immediate repercussions for land
appropriation. The most important issues were the spatial location of the new
economic cycle and the availability of land in the south-central area of Brazil that
benefited from the decadence of the mining regions (Furtado 1959). Thus, the
new economic cycle (the coffee cycle) was a result of the reorientation of unused
resources and the use of available land.

Nevertheless, the amount of land that could be economically exploited
was limited. In addition, depending on the location, the characteristics of some
land resources made their use uneconomical. Further, other areas were located
too far from transport or roads, making the cultivation of coffee unprofitable. In
addition, the land in the coastal regions of Rio de Janeiro and 850 Paulo, as well
as farther inland in Minas Gerais was concentrated. The prosperity brought about
by the coffee industry in these regions also increased population density. Thus,
the concentration of agricultural interests and population would generate conflict

over the agrarian question, creating land insecurity that only land regularization

32

could minimize. The development of the coffee economy that began to alter
Brazil did not change the use of slave labor or the continuing incorporation of
new lands (Silva 1996). As I have tried to demonstrate, landownership in Brazil
was problematic. The landowner that had developed in the colony could not be
called a landowner in its strict sense because the large majority of landowners
(either sesmeiro or posseiro) did not have a land title. But this was not important;
the most important goal was to keep the economic system based in the latiftindio
and in slavery. In other words, the interpretation of the resolution ending the
sesmaria system was that it was not applicable to the posse system (Silva 1996).
Consequently, by keeping the possibility of land acquisition and the use of slave
labor, landowners did not see any reason for land regularization. Finally, in 1850
after 28 years without land titles this situation changed. The Land Law of 1850
prohibited the acquisition of terras devolutas in any way, except by purchase,
and established a new definition of the concept of terras devolutas. Specifically,
the Land Law of 1850 determined that sesmarias, or other government land

concessions that were cultivated or used under the principles of “culture e

morada habital” (i.e., the basic principles of usocapia’o5), would be recognized. In
addition, this law stated that in the case of conflicts between posseiros and
sesmeiros, the most important criterion for determining who would control the
land was the present cultivation of it. Finally, the law allowed the government to

reserve terras devolutas for colonization and military use.

 

5 Usocapia'o is a form of land acquisition of a property by using the land for many years without
interruption.

33

The Land Law of 1850, although recognizing the acquisition of land by
posse through cultivation, allowed for the correction of some land problems but
avoided massive expropriations. Eventually, the law brought an end to the posse
system in Brazil by establishing a system of private property that recognized the
holdings of posseiros under the concept of effective cultivation (article 8 of the
1850 Land Law). In addition, the Land Law of 1850 defined terras devolutas as:
(a) lands that were not private; (b) lands that were not in sesmarias or other
government concessions; (C) lands that were not occupied by posseiros and
were not regularized under the law (see above); and (d) lands that were not
reserved for public use at the municipal or state level (article 3 of the 1850 Land
Law).

Silva (1996) argued that this redefinition of terra devoluta changed the
legal meaning of the term devoluto, which came to be applied (after the colonial
period) to signify vacant land. The original meaning of devoluto was land in a
concession that, if the landholder did not put in production, would be returned to
the Portuguese Crown. Eventually, the new meaning of devoluto became
commonly used in Brazil. However, the Land Law waited 4 years to be
promulgated, and the land competence was attributed to the Repartiga'o Geral de
Terras Pablicas (General Public Land Agency) through the Imperial Decree
1.318 of January 1854. According to this decree, this agency was supposed to
drive and organize the measurement, description, and division of terras
devolutas and to recommend to the government the lands that should be

reserved for colonization, put up for sale, or put to military use. However, the

34

government did not achieve its objective of demarcating land for sale. Thus, the
government concentrated its action in demarcating terras devolutas which could
be used for colonization by migrants. But these actions were not enough to
regulate and stop land acquisition. The Brazilian government was unable to

implement the Law of 1850, and consequently, terras devolutas continued to be

illegitimately oc'cupied.6 The failure of the 1850 Land Law would lead to a new
legislative change. The 1891 constitution would remove the recognition of land
concessions in areas labeled as terras devolutas, would reaffirm the concept of
private property, and perhaps most importantly, transfer all unoccupied lands to
the states, except in the case of frontier areas. In other words, the Brazilian
government put the organization and administration of terra devoluta in the
hands of the regional oligarchy. Thus, the history of the territorial appropriation
would be linked from this point on to the history of each old province (now
states). Consequently, with this legislation, the states began to develop their own
laws governing the occupation of lands under state control. Among the larger
states was Para State in the Amazon region. This state, after Decree 410 of
1891, defined its own rules for mediation, buying, selling, obtaining title, and
validation of land holdings. This law would recognize all landholders up until the
date of its enactment, including the possessions of small landholders as well as
those of sesmeiros and their descendents. However, the sesmarias were
subjected to revalidation and existing posses to regularization. Because of this, in

both cases, it was necessary to prove that the land was effectively being

 

6 This illegal occupation of land would come to be recognized as grilagem (Silva 1996). Grilagem
is a complex form of fraudulent land titling by which lands are seized with forged documents.

35

cultivated or that the landholder was a permanent resident of the property
(Lamarao 1977).

In this period, rubber extraction was not a small landholder activity. The
owners of the seringais (extensiVe rubber extraction areas) and rubber buyers
involved in the activity of extraction were already members of the elite in
AmazOnia (Weinstein 1980). The areas of the seringais, as well as natural
pastures used for cattle ranching, were not required to be cultivated. Therefore,
Decree 410 defined forest conservation for extractive industries as a form of
cultivation (Monteiro 1980). This definition was yet another step whose purpose
was the preservation of Iatifiindio and the concentration of land. During this
period it would have been ideal to regularize all existing properties, thereby
demarcating all unoccupied public lands that could then be sold or reserved for
specific uses such as colonization projects.

The collapse of the rubber boom would change this situation. To avoid the
migration of small landholders who were associated with rubber extraction and
food production, the State of Para established public land concessions for
everyone who needed land, with a maximum concession of 100 ha (Law 1.432 of
1914). In addition, provisional titles could be acquired, permitting the occupant to
obtain definitive title after two years (State Law 1.584 of 1914). Despite the
decline of the rubber economy, other activity expanded in the region, such as
Brazil nut extraction, which was organized in the same way as rubber and
therefore a legitimate extractive activity. Extractive activities became the most

important factor in agrarian politics before 1930. Local political elites interested in

36

Brazil nut extraction were able to establish long-term leases (e.g., aforamento
perpe’tuo) that allowed them to control enormous expanses of land in the region
(Simmons et al. 2007). These leases (made possible to obtain complete control
over Brazil nut reserves) were inexpensive for the lessees (Santos 1984; Emmi
1999). Ultimately, the state government approved a law that would allow the
lease terms to be of undetermined length and would allow the acquisition of
expansive landholdings for agro-industrial groups (Coimbra 1981 ), a fact that

would cause substantial public protest in Para State.

2.3 Agrarian Polices between 1930 and 1970

Public protest at the end of the 1920s and the Great Depression of the

early 1930s aggravated the political and economic situation of the “old republic."7
This crisis would lead to the increasing Visibility of labor mobilization and would
threaten the traditional political alliances between states in southeastern Brazil,
principally between $50 Paulo and Minas Gerais [often called “Coffee and
Cream” politics (Politica do Café com Leite)]. This political and economic
instability would lead to the 1930 revolution. To avoid possible popular revolt to
the detriment of the economic gains of the oligarchy at the time, the leaders of
the 1930 revolution did not look upon state government decisions favorably.
Revolutionary Decree ll of 7 November 1930 required the immediate return to the
Union of all areas granted by the state, including long-term leases for Brazil nut

extraction. In addition, the revolutionary government started to review every land

 

7 The First Republic, or “Old Republic,” was a period of Brazil's history that begins with the
proclamation of the republic in 1889 and ends with the 1930 revolution.

37

title granted by the state through concession. However, the new state (Estado
Novo) did not express its opinion over an issue of great importance for the
agrarian situation in Brazil: the issue of usucapiéo, a principal according to which
property could be obtained by anyone who effectively cultivated it for a minimum
period of time. To the contrary, the Brazilian constitution of 1934 (article 125)
officially instituted usucapiéo (called usucapia'o pro-labore) in which a property of
10 ha could be obtained if cultivated for a minimum period of 10 years. This
decision was reaffirmed in the Brazilian constitution of 1937 (article 148) which

maintained the minimum period of 10 continuous years of cultivation for an area

of 10 ha.8 Once again, the law was not enforced and the aforamento perpétuo
did not come to an end; although this decree allowed, during a set period of time,
the Union control of the conflict between large and small landholders.

By the end of 1947, a communist wage was striking many countries in the
world. Brazilian communist sympathizers found supporters in the countryside and
helped the peasantry work toward the organization of Ligas Camponesas. For
theorists helping to organize the Ligas Camponesas, Brazil was being used by
imperialist countries, such as the United States. According to these theorists,
imperialist countries and the Brazilian collaborators (e.g., big coffee planters,
beef exporters) were interested only in the production of export commodities; and
in order to protect their interests, they conspired to hold the agricultural sector in

a backward state of development (Welch 2006). To avoid popular adherence to

 

8 The Brazilian constitution of 1946 (article 156, § 3) kept the minimum period but changed the
size to 25 ha, and the Brazilian constitutional amendment number 10 of 1964 (article 156)
changed the size again to limit it to 100 ha and a minimum period of 10 years.

38

the communists’ ideas, the Brazilian government decided to ban the Brazilian
Communist Party (PCB). Going underground, communists started to guide their
militants through the beginning of the 1950s with ideas that it was necessary for
the peasantry to rise up and fight for land seizures without compensation to
latifdndios. Also, the communists felt it was necessary to confiscate the
latifL'Indios, which then should be given freely to those who work and live on the
properties (Welch 2006). With elections approaching, the populist candidate
GetUlio Vargas decided to stand up against the “imperialists” in 1951. As part of
his economic plan, Vargas exploited coffee export earnings through taxation and
the manipulation of foreign exchange rates (Welch 2006). Consequently, this
policy irritated coffee producers; but they were able to protect their profits by
reducing labor costs through increased exploitation and the withdrawal of
privileges such as free worker housing (Welch 2006). Overwhelmed by scandal,
Vargas committed suicide in 1954. The communists took advantage of Vargas’
death to host a congress of rural worker and peasant delegates from
associations throughout the country. In this congress, the Farmers and
Agricultural Laborers’ Union of Brazil (ULTAB, Uniéo dos Lavradores e
Trabalhadores Rurais do Brasil) was founded, which shortly became the main
tool used to advance the rural labor struggle in Brazil.

The pressure for land reform in many parts of the country would lead the
Brazilian president to construct the Belem-Brasilia Highway in 1960 as a way to
release the pressure for land reform in southern Brazil. This action would attract

new migrants to the Amazon region. Small and large producers would be

39

attracted by the perspective of potential benefits offered by the presence of
mineral deposits, such as gold and diamonds, as well as soils fit for agriculture.
This migration opened the way for land speculation and intensified the demand
for land in the region (Santos 1979). To provide a picture of the volume of land
transactions between 1959 and 1963, the height of this rampant speculation, the
State of Para sold more than five million hectares of land, the majority granted to
large capitalist interests headquartered in southern Brazil (Santos 1979).
Between the construction of the Belem-Brasilia highway and the revolution of
1964, the economic growth of Brazil was attributed to import substitution through
increases in tax and favorable exchange rates. This growth continued through
the 1960s with increased foreign, government, and private investment. However,
by 1964, the country had a record budget deficit and an annual inflation rate of
52% (Wynia 1990). The aggravation of this new economic crisis and increasing
benefits to capitalists brought a spike in the number of small producer and rural
worker associations, nearly all of which favored agrarian reform. This fight for
agrarian reform would become part of the daily life of the Brazilian population and
would stimulate increasing pressure for change in the countryside. Thus, much
like examples in other countries (e.g., Cuba, Mexico, Bolivia), the Ligas
Camponesas started to organize again; and a movement in favor of social justice
and agrarian reform in the countryside spread throughout the nation, assuming
grand proportions in the mid 1960s.

Popular dissatisfaction throughout the Brazilian countryside would lead to

the military coup of 1964 and the establishment of the Land Statute of April 1964.

40

For the first time in Brazil’s history land reform had been defined as a set of
measures that could promote a better land distribution by modifying the land use
regime and landownership based upon the idea that the Land Statute was to
promote social justice and prodUctivity. This statute would drive agrarian politics
and agricultural policies in Brazil. The Land Statute had two principal objectives
for the military government: (a) the implementation of a land reform program and
(b) the development of agriculture. In addition, the statute would create
opportunities for access to proper titles for private land, with the condition that it
would be used to serve a “social function”. In article 2, the Land Statute of 1964
established that the “social function” of land would involve the well-being of the
landholder and the rural workers who worked on it, coupled with satisfactory
levels of productivity and the conservation of natural resources. However, should
lands not meet their “social function” as elaborated by the Land Statute, they
would be subject to expropriation as if they were unused (terras devolutas). In
addition, the Land Statute of 1964 would establish that (a) any rural worker that

had occupied terra devoluta for one year would have preference in acquiring that
land up to the size of one rural module9 in the region (article 97, § II); and (b)
everyone that had occupied a land in terra devoluta for 10 uninterrupted years

would have rights to that land (article 98).

The main contrast of the Land Statute of 1964 with earlier laws was the
clear wish to terminate latifdndio through the use Of expropriation. However, on

the one hand, the Land Statute provided a way to decrease land concentration;

 

9 A rural module is defined regionally and considers the quality of the land and the climatic
conditions of each region.

41

on the other hand, it emphasized the need to develop agriculture through
agricultural credits and subsidies. At that time, an increased public debate on
agricultural policies was arguing that the agricultural sector was backward,
traditional, and generally exhibited nonmarket behavior in its alleged lack of
supply response to price incentives (Silva 1981). Thus, Law 4.829 of 5 November
1965, following the Land Statute, would allow the modernization of Brazilian
agriculture through mechanization. The principal idea of this policy was to lower
the cost of capital for the agricultural sector. However, the combination of these
two policies (land reform and agricultural credit) distorted the underlying Objective
of each on their own. In other words, this combination of policies made unskilled
labor much more expensive and lessened agricultural capital (Rezende 2006).
As a result of this price factor distortion, 3 rapid change in agricultural technology
with the mechanization of production on a grand scale occurred but it came
without a solution to the problem of land concentration in Brazil (Figures 1, 2, &

3).

This was a period of the technological modernization of Brazilian
agriculture. However, this modernization was only partial and affected only some
products (e.g., sugarcane, cotton, cocoa, coffee, soybeans) and regions (south-
central and northeast). Another impact was an increased dependency of
agriculture on other economic sectors, principally industry and finance.
Consequently, the technical progress of Brazilian agriculture was not uniformly

distributed and would contribute to the maintenance Of land concentration.

42

2.4 The Modern Determinants of Land Concentration in Brazil

With the modernization10 of agriculture, agricultural specialization
occurred on a national scale. In some regions, crops outside of the family
economy were considered commercial and, hence, were developed and
modernized. This was the case with sugarcane, cotton, tobacco, and cocoa in
the northeast; coffee, cotton, and sugarcane in the southeast; and rice, wheat,
soybeans, and grapes in the south (Soares 2000). In southern Brazil, the
substitution of extensive cattle-raising occurred, and agriculture concentrated in
the growth of export products (i.e., coffee and soybeans). In contrast, west-
central Brazil became an area of agricultural occupation centered on large-scale
soybean, corn, rice, and beef-cattle operations, largely supported by special
government programs and financial incentives (Silva and Kohl 1984). Different
from other regions of the country, where intensive modernization occurred
through the use of fertilizer, the northern region saw an increase in cultivated
area by bringing land into production. The use of agricultural machinery and
modern inputs shows practically no change from 1960 to 1975 (Silva and Kohl
1984). This agrarian structure evolved to be more concentrated and exclusive,
making access to land for rural workers in Brazil difficult. Not only did land
become more concentrated, but access to the land that remained was practically

eliminated (Rezende 2006).

 

10 For some authors, such as Silva (1981), modernization was a transformation of the natural
form of the rural production process to industrialized farming.

43

This agricultural modernization also had strong effects on the rural
exodus, increasing the number of workers who left rural areas looking for jobs in
urban centers. Brazil shifted from an agrarian rural society to an urban society in
only few decades. In 1940, 28 million people (68.8%) of the population lived in
rural areas. By 1950, 18.7 million people were classified as urban inhabitants,
which was 36.2% of the total population. The 1960 census recorded an urban
population of 31 million; this was 44.7% of the total population of Brazil. The
estimated urban population of Brazil in 1970 was 52 million, which was 55.9% of
the country’s total population ( Table 2). This migration involved both
movements of people from rural areas to the more urbanized areas of the same

state as well as a shift from more rural states to the more urban states.

44

 

GINI Index

- 0.71
- 0.75
- 0.7a
- 0.79
- 0.80
- 0.84
- 0.88
- 0.39

 

      

Central/West

  

0 1.000 2.000 Kilometers
I

 

Figure 1. Brazil Land Concentration Indices by Region, 1972. Source:

INCRA (2007b)

45

 

 

      

Central/West

Southeast

GINI Index

- 0.71
- 0.75
- 0.78
- 0.79
- 0.80
- 0-84
- 0.88
- 0.89

  

O 1.000 2,000 Kilometers
I

 

 

 

Figure 2. Brazil Land Concentration Indices by Region, 1992. Source:
INCRA (2007b)

46

 

GINI Index

- 0.71
- 0.75
- 0.78
- 0.79
- 0.8

 

   

Central/West

 
 
   

Southeast

0 1,000 2.000 KIIOI'DONTI
l I

 

Figure 3. Brazil Land Concentration Indices by Region, 2000. Source:

INCRA (2007b)

47

 

Table 2. Resident Population, 1940 — 1970

 

 

 

 

 

 

Total
Years .
Urban Rural
1940 12,880,182 (31.2%) 28,356,133 (68.8%)
1950 18,782,891 (36.2%) 33,161,506 (63.8%)
1960 31,303,034 (44.7%) 38,767,423 (55.3%)
1970 52,084,984 (55.9%) 41,054,053 (44.1%)

 

 

 

 

 

Source: Séries HistOricas do Brasil. IBGE 2007a.
(http://www.ibge.gov.br/home/estatistica/populacao/defaulttab_historicas.Shtm)

This urbanization process began in the more developed regions of the
country where capitalization and mechanization of agriculture first occurred and
became most intensive. The result of the rural exodus was burgeoning urban
zones and the creation of a marginalized population without jobs and living in
subhuman conditions. This marginalized population would continue to apply
pressure for land reform. Thus, the military government, in the search for a plan
of action to resolve this “new" pressure for agrarian reform, decided to stimulate
the opening of new agricultural frontiers. This strategy focused on the occupation
of the Amazon, a region with immense areas of terra devoluta with low

population densities.

The Revolution of 1930, through revolutionary Decree II, had already

returned all unclaimed land to the Union. Thus, to achieve its plan, the military

48

regime created two important laws: Law 5.174 of October 1966 and Law 5.374 of
October 1967, which would emphasize the importance of private enterprise in
regional development. According to these laws, the government would provide
the infrastructure and general funding for development; and in addition, it would
also create many programs. For instance, the federal government created the
National Institute of Land Reform (INCRA - Instituto Nacional de Colonizacéo e
Reforma Agréria) in 1970, as well as various special programs for regional
development. Among these special programs were the National Integration
Program (PIN - Programa de lntegraca'o Nacional, or PIN -1970), the
PROTERRA program (Programa de Redistribuicéo de Terras e Estimulo a
Agroindustria do Norte e Nordeste), and the POLOAMAZONIA program
(Programa de POIOS Agropecuérios e Agrominerais da AmazOnia). Of these,
POLOAMAZONIA was designed to attract national and international companies
to the Amazon region through fiscal incentives and credit programs. ln addition,
the military regime also had concerns with the security of the region. To protect
the region, the military regime would declare the terras devolutas located in 100-

km strips along both sides of all federal roads in the Legal Amazon (Decree

number 1.164 of April 1971 )11 and Faixas de Fronteiras (Frontier Strips)12 as

 

11 Decree number 2.375 of November 1987 removed the clause that made the terras devolutas
located in 100 strips along both sides of all federal roads indispensable to national security.

12 The utilization of property in areas close to national borders was subject to special conditions
since the Old Republic. Law 601 of September 1890 had reserved a strip of 66 km (10 leagues)
along the national border that could be freely given. The constitution of 1891 established in its
article 64 that the terras devolutas would belong to the States and the Union would stay with a
portion of the territory that was indispensable for defending national borders, military fortification,
and federal railroads construction. Article 166 of the constitution of 1934 advanced the limits of
Faixa de Fronteira to 100 km, and the constitution of 1937 changed the limits to 150 km, the
extent that remains to today.

49

indispensable for national security. However, aside from these efforts, not many
changes in the agrarian structure of the country occurred. To the contrary, in the
Amazon region, this period was characterized by the development of rural
capitalist enterprises, as well as the continuation of traditional latifrindio (Hoffman

1998).

In terms of land issues, the 19803 were no different than the decades that
preceded them. In other words, struggles for land reform and increased
mobilization for democratization occurred in the whole country, principally in the
more industrialized cities. This increased mobilization was again responsible for
changes to land law in Brazil. Congressional law 6,969 of December 1981
(article 1) would reduce the usucapiéo time period to 5 years for anyone

continuously using an area of no more than 25 ha.

The democratization of Brazil after 1985 opened the door for the Brazilian
population to reaffirm the necessity of agrarian reform because of the great
masses of landless workers throughout the country. Many social movements,
which were under the oppression of the military regime and functioning only
clandestinely, began to operate with more liberty. Then, in 1985, the government
of José Samey implemented the first National Plan for Agrarian Reform (PNRA,
Primeiro Plano National de Reforma Agréria). The objective of this plan was the
rapid application of the Land Statute of 1964 and the enabling of agrarian reform

until the end of Samey‘s term in office.

The first National Plan for Agrarian Reform (PNRA) stated that the

principal method to obtain land for agrarian reform purposes was through the

50

expropriation of land for the social interest. In addition, the first PNRA also stated
that the owners of expropriated lands must be compensated in the amount of the
land’s declared value according to the Rural Territory Tax (ITR). However, the
declared value of lands on ITR paperwork was generally much lower than the
land's true market value. Consequently, this proposal assumed that
expropriations would penalize landowners for not using their land (Medeiros

2002)

As a result, the PNRA generated a conflict of interest between the rural
elites and emerging social movements. For instance, the UDR (Uniao
Democratica Ruralista) was created in 1985 to resist agrarian reform efforts
undertaken through land occupations that were occurring throughout the country
(Fernandes 1999). The UDR would defend the rights of landholders in
negotiating the sale of land to the state as an alternative to land expropriation. In
addition, the political power of rural elites led to the inclusion of a clause in the
final version of the first PNRA that prohibited the expropriation of productive
properties, regardless of their size or use. In its final version, the first PNRA took
as its objective the settlement of 1.4 million families in the whole country.
However, once more meaningful land reform only existed on paper. The Samey
government, at the end of its term, settled only 90,000 families, representing just

6% of the original goal of the PNRA (INCRA 1995).

The convocation of the 1987 Constitutional Assembly was the next phase
in the struggle against land concentration. A strong popular interest in favor of

increasing agrarian reform efforts emerged, but rural elites represented by

51

traditional rural organizations (e.g., Confederaca’io Nacional da Agricultura,
Sociedade Rural Brasileira) once again impeded the implementation of
constitutional articles that benefited agrarian reform efforts (Fernandes 1999).
Even so, the new constitution advanced, principally with respect to social rights.
The Constitution of 1988, as in the Land Statute of 1964, established that land
must serve a social function, where social function was defined as (a) the well-
being of landowners and those working for them; (b) the maintenance of
satisfactory levels of productivity; and (c) the rational use and conservation of
natural resources. However, introduced by the Land Act, the land expropriation
legislation was significantly changed by Brazil’s 1988 Constitution. Since then,
only unproductive land was under the risk of expropriation for which the state
needed to pay a price. In other words, the constitution defined how the
compensation for the expropriation of land for reasons of social interest would be
awarded. Consequently, landowners would be paid in agrarian bonds over a
period of 20 years. Ultimately, the constitution exempted the expropriation of
lands < 15 “rural modules” (mOdulos rurais), as well as those properties that were
productive. In addition, article 189 made any rural land acquired through land
reform programs non-negotiable for a period Of 10 years. This legislation took 5
years to be ratified in the Brazilian Congress — a delay that only served to
increase tensions between landless people and large farmers. In 1992, the
National Institute of Colonization and Land Reform was authorized to acquire
land for land reform purposes in areas of social tension to satisfy the

requirements of the National Program of Land Reform. But even with all these

52

arrangements, the tension between landless people and large farmers did not
end. The federal government, concerned with putting an end to land invasions,
established in 1993 the agrarian Law 8.629, which would simultaneously allow
deeper agrarian reform legislation but would eliminate many other opportunities.
For instance, two impacts of this law were that public lands could be used for
land reform purposes, but that the definitions of “social function” and “productive”
land remained unclear. This created opportunities for the use of both terms to
conform to the needs of interested parties (i.e., rural elites and the landless). In
addition, the law stated that any rural property that was invaded would not be
evaluated or disappropriated for a period of time of 2 years following its
disoccupation. Also, the period of time would double if the same property were
invaded again. Finally, the law would penalize with exclusion any individual that
had been enrolled to receive a parcel of land (or had already received land in a
settlement) if it was proved that he/she had participated in an invasion of public
or private properties.

Despite these efforts, the land reform problem did not change much in the
1990s compared to the 19803. To the contrary, land invasions did not cease, and
the concentration and the underuse of land increased in the 1990s. Between
1992 and 1998, the area devoted to large properties (more than 2,000 ha)
increased by 56.2 million hectares, an area the size of France (INCRA 1996).
Further, the Atlas Fundiério do Instituto Nacional de Colonizacao of 1996 shows

that IatifLindios (2.8% of rural establishments) occupied 56.7% of the country,

53

while miniftindios13 (62.2% of rural establishments) occupied only 7.9% of the
total area of the country. In addition, the Atlas Fundiério provides an account of
the amount of unused land in Brazil — 185 million hectares, or 40% of all usable
area, had a landholder but was not considered productive (i.e., was latifcindio).
Although many elected presidents had developed various agrarian reform plans,
the concentration of land did not diminish as expected in Brazil (Figures 1, 2, &
3). Consequently, throughout Brazil, social movements of rural landless workers
emerged, tired of waiting for concrete action by the federal government. These
people initiated their own land reform movement by increasing the number of

occupations of private land and terras devolutas.

The main objective of this chapter was to provide an account of the
evolution of agrarian law in Brazil, from the initial domination of the Portuguese
through the present day. Throughout this complicated 500-year history, the
changes in agrarian law did not bring about a deconcentration of land in Brazil.
To the contrary, these changes maintained the same historical patterns of land
concentration. However, sociopolitical shifts that occurred in Brazil made
possible the mobilization and collective action of the population at large, through
the emergence of social movements favoring substantial changes in land reform
policies. Thus, considering the appearance of social movements and the

importance of these movements to agrarian reform, in the next chapter I will

 

13 Miniftindios were properties of less than one rural module. In general, Latifdndios could be
classified according to utilizaca'o e tamanho. Latifcindios de Utilizagao (use) were properties
between 1 and 600 rural modules in size in which the level of productivity was less than the
regional average. Latiffindios de tamanho (size) were properties with areas larger than 600 rural
modules without any consideration of the type or characteristics of production (Campanhole
1969).

54

outline the trajectory of social movements and identify their historical origins and

their role in direct action land reform (DALR) in Brazil.

55

CHAPTER 3: THEORIES OF SOCIAL MOVEMENTS AND
THE DIRECT ACTION LAND REFORM IN BRAZIL

The 1960s had a profound influence on societies around the world.
Among those affected, American societies perhaps saw the greatest number of
protests and collective actions. This period was marked by substantial public
dissatisfaction, organized in protests, social movements, “political” organizations
independent of political parties (e.g., workers' unions), among other forms of
public manifestations. These collective actions became very important in the
US, as in Europe. However, the explanations of the causes of the rise of

collective action and social movements diverged in both places.

In this chapter, I present the trajectory of the study of social movements,
with emphasis on a number of principle theories explaining their emergence.
First, I will review the theories that dominate the initial studies of social
movements in the 19605, often referred to as the “classical model of social
movements.” Following this review, I will approach the two theories that
challenge the classical view: resource mobilization and collective process. Next, I
discuss how those who studied social movements in Latin America viewed these
theories. To conclude this chapter, I present the role of social movements in the

struggle for agrarian reform in Brazil.

3.1 The Classical View of Social Movements

The classical approach dominated initial studies of social movements until

the middle of the 1970s. Led by the “Chicago School,” these studies viewed

56

social movements as a result of a behavior crisis or, in other words, as a
consequence of “social strains” (Merton 1938, 1968). According to this theory,
social movements emerge from general and profound dissatisfaction, which
could be a result of social change that modifies existing social relations, which
are considered inadequate (Jenkins and Perrow 1977). Consequently, this
“social strain” would create dissatisfaction, and collective action rises with the
intent of alleviating these “strains.” The existence of social frustration and crisis
automatically produces revolt. This type of collective behavior was described as
irrational or nonconventional. In other words, the classical theory viewed
collective behavior as “deviant” (Smelser 1962). Following this View, it was
argued that, in democratic societies, it would be rational to use conventional
means to enact social change through institutionalized channels instead of

rebellion, strike, or other forms of collective action.

The classical theory is not a theory comprised of a single argument. Some
variations of the theory could be classified as (a) mass theory, (b) status
inconsistency theory, (c) collective behavior theory, and (d) relative deprivation

and its variation (Davis’ so-called “J curve”).

McAdam (1982) argued that every variation of the classical model has
unique characteristics. However, the most important aspect of these variations is
the consistency with which each name identifies the same general cause as
responsible for the emergence of social movements. In other words, all the
variations identify a structural problem in society that leads to a psychological

effect, which disrupts the existing status quo, resulting in the production of social

57

movements (McAdams 1982). Every version of the classical model has as its
central point the identification of a structural fault of a problem and differs only
with respect to the conceptualization of the structural problem itself. With the
object of comparing the similarities of each model, I will briefly describe each of

them in the following sections.

3.1.1 Mass Society Theory

Mass society theory refers to the lack of an intermediate structure through
which individuals can be integrated with society (Komhauser 1959). Through this
View, social isolation is a fundamental prerequisite for social protest, and feelings
of isolation and anxiety cause a disposition to become involved in collective
action. Komhauser (1959) argued that one of the principal causes of “structural
strain” is the disappearance of a sense of community. According to this author,
the collapse of a stable community creates a feeling of isolation and anxiety
among individuals. Hamilton (2001) argued that in mass society individuals are
atomized, leading them to isolation and solitude. According to this theory, social
movements emerge when society's intermediary organizations are lost (e.g.,
workers’ unions, churches, associations) causing dissatisfied individuals to join
these movements (Komhauser 1959). These organizations function as
intermediaries between individuals and the new social structure, aggregating

individual preferences and serving to relieve the pressures of social life.

58

3.1.2 Status Inconsistency Theory

Another variation of the classical model is ”status inconsistency.” This

theory argues that an individual with an inconsistent social status can become

more frustrated than an individual with a more consistent social status14 (Lenski
1954). Nevertheless, this type of approach only proliferated between the 19603

and 19703, with the intent to explain the relationship between inconsistency and
political responses. The theory of status inconsistency argues that there exists a

set of conflictive expectations that produce anxiety and social tension.

McAdam (1982) stated that status inconsistency refers to the discrepancy
between the ranking of an individual according to a number of status dimensions
(e.g., education, income, occupation). According to McAdam, if the discrepancy
is severe, an individual can begin to feel tension similar to that of individuals
afflicted with feelings of atomization as described by mass society theory. Thus,
status inconsistency is another variation of the classical model, in that severe
discrepancies in social status produce a “strain,” a dissatisfaction that generates

collective protest and results in the appearance of social movements.

3.1.3 Collective Behavior Theory

Differing from the two theories presented above, the theory of collective
behavior does not highlight a specific condition, such as status inconsistency or
atomization, as a primary cause of the appearance of social movements. To the

contrary, this model, similar to the one proposed by Lang and Lang (1961) and

 

14 Social status can be interpreted as the social position of an individual in society.

59

Smelser (1962), argues that any type of severe “strain” may contain the
necessary components to stimulate the emergence of social movements.

Consequently, this approach became one of the more general classical models.

Following Smelser (1962), a form of “strain” must be present for an
episode of collective action to occur. Therefore, collective behavior assumes that
individuals enter into episodes of collective action because something is wrong
with the social environment in which they exist. For example, a rapid economic
crisis that may lead to significant economic depression and unemployment may
generate a “strain” that brings about social insurgency. Gusfield (1970), for
instance, argued that social movements are a response to social change that
follows new ideas that break down existing social structures. This line of thought
sees social change as a source of structural “strain.” In this context, social
change has been seen as a stressor because it causes a breakdown in the

normative order to which individuals are accustomed.

3.1.4 Relative Deprivation Theory and the Davis J Curve

Relative deprivation theory is associated with the existing links between
depravation, frustration, and aggression. Following this theory, the dispossession
of goods and services may lead to frustration, which may produce aggression
when frustration is released. This theory is based on the hypothesis that social
movements are made up of individuals who feel deprived of the ability to Obtain
resources or goods (Morrison 1978). This approach considers the fact that
individuals who lack the conditions to obtain these types of goods (or services)

are the most favorable individuals to organize social movements with intent to

60

improve their conditions. In other words, these individuals feel frustrated with
their conditions in relation to others. However, relative deprivation theory
presents a problem. The majority of people feel deprived at some moment in
their life of some goods or services (Morrison 1978). Consequently, this theory is
unable to explain why individuals join together to form a social movement while

others similarly deprived of goods and services do not.

The Davis J Curve is a variation of relative depravation theory. This theory

results from an attempt by James C. Davis15 to reconcile two explanations for
revolutionary phenomena. According to Davis (1969), Karl Marx and Alexis de
Tocqueville (two important scholars of revolution) arrived at opposite conclusions
about it. Davis (1969) argued that for Marx revolutions probably occurred when
the situation became critical. In other words, revolution occurs when proletarian
misery increases in relation to the economic power of the bourgeoisie. In
contrast, Davis (1969) saw in Tocqueville’s work that revolutions were the result

of the liberation of oppressive conditions.

Davis (1969) attempted to reconcile these two contrary explanations
through the argument that both were incomplete. He argued that revolutions
occur when a period of progress is follow by a period of profound decline (Davis

1969:86y

 

15 James C. Davis (1969). “Toward a Theory of Revolution" in Studies in Social Movements, ed.
Barry McLaughlin.

61

Revolutions are most likely to occur when a prolonged period of
objective economic and social development is followed by short

period of sharp reversal. The all-important effect on the minds of
people in a particular society is to produce, during the former

period, an expectation of continued ability to satisfy needs...

The brief explanation above demonstrates that the various versiOns of the
classical model, despite their differences, have a number of points in common. In
all of the models presented here, structural “strains” create tensions and
frustrations that, when they reach a certain threshold, cause social insurgency to
appear. In addition, though taking as its goal the description of the rise of
collective action, classical models are in reality more preoccupied with the
psychological effects of “strain” on individuals. In other words, individual
dissatisfaction, anxiety, and frustration are the principle drivers of the emergence
of collective action. Ultimately, in all the classical models discussed above, the
appearance of collective action is associated with a basic preoccupation with the
search for a solution to the problems of frustration and anxiety — not a search for

political objectives.

During the 19603 and 19703, as with many other aspects of society, social
movements underwent a number of grand transformations. In the 19603, the
majority of scholars avoided addressing social movements. In their estimation
social movements were viewed as groups of individuals that were the irrational

product of “social strain.” This presupposition was subsequently considered

62

problematic during the 19703 because it assumed that the emergence of social
movements was a consequence of a certain level of “strain” in a society.
Consequently, this approach indirectly assumed that society was always in
equilibrium and without “social strain.” However, for Wilson (1973) this perception
was problematic. He argued that society was rarely in equilibrium and was
always showing social “strain” because it was always mutating. Therefore, social

movements always had stimulus to emerge.

The idea that social movements is a consequence of frustration,
deprivation, and social strain started to change in the end of the 19703 and the
beginning of the 19803 with the inclusion of the political context argument. For
instance, McAdam (1982) argued that the classical model lacked a discussion of
the political context in which social movements appear. For this author, social
movements do not appear instantaneously. Much to the contrary, social
movements are modeled by a large variety of social-environmental factors. In this
context, the appearance and development of social movements occur as a
product of individuals involved in organized and dynamic interactions within an

environment constantly undergoing political and socioeconomic mutation.

Jenkins and Perrow (1977) argued that the success of rural works in the
United States during the final years of the 19603 can be explained by changes in
the political environment of the period, as well as the internal organizational
characteristics of the movement. Cohen (1985) demonstrated that the
appearance of well-organized social movements was not a symptom of

irrationality. To the contrary, the individuals involved in protests and mobilizations

63

have a clear objective and strategy. Therefore, McAdam (1982) noted that
“strain” could be necessary to generate social movements, but is not a sufficient

cause on its own.

Another criticism of the classical theory of social movements is centered
on the hypothesis that social insurgency is a function of individual discontent and
that it is necessary to arrive at a maximum limit of discontent or “social strain” to
produce a social movement. However, according to McAdam (1982) the classical
theorist has not stipulated a threshold at which this level of discontent or “strain”
is able to produce a social movement. In addition, classical theories ignore the
fact that social movements are a collective phenomenon. Thus, these scholars
are not able to explain how discontent or unsatisfied individuals are transformed
into an organized movement. In the end, the argument that the appearance of
social movements is the consequence of the psychological irrationality of
individuals seeking an escape valve does not consider the possibility that these
same individuals may be involved in rational collective action. In other words, the
classical theory does not consider that social movements may have well-defined

political objectives and power.

Bachrach and Baratz (1973) and Prewitt and Stone (1973) argued that
social groups differ based upon the political power that they have. Those groups
with sufficient political power are able to bargain and advance their interests. As
a result, these groups control the political environment, excluding those groups
that are without political power. Keeping in mind this perspective, the above

authors argued that it is possible to observe that social movements are not a

64

form of irrational behavior but are instead a planned response against a

repressive political system.

In sum, the classical approach assumed that the increase of individual
complaints generates “social strain” that enables the appearance of social
movements. While the specific hypotheses of the classical approach vary
between different currents of thinking, each of these versions agrees that the
individuals involved in social movements act irrationally as a consequence of
“social strain.” The diversity of social movements in this period (19603 and
19703) helped change this way of thinking and contributed to the development of
alternative hypotheses. The fact that different social groups with varying levels of
political power exist helped the development of the hypothesis that social
movements act rationally and in an organized manner. This new idea gave rise to

the approach of resource mobilization theory.

3.2 Resource Mobilization Theory

The social movements that appeared between the 19603 and 19703
stimulated new hypotheses about the rise of organized social protest. This new
way of thinking would be known as resource mobilization theory (McCarthy and
Zald 1973, 1977; Oberschall 1973; Gamson 1975; Tilly 1978, among others).
The central hypothesis of this theory states that social movements generally
consist of formal organizations, thus leading the above authors to argue that (a)
any organization requires a certain level of resources, especially money, to
sustain it; and (b) in society there are always dissatisfied individuals who are

ready to create a movement of protest. However, according to these authors, the

65

change that has occurred over time has been that social movements use the
resources available to them, such as financial support and the individual skills of
movement participants. Following this realization, studies of social movement
changed their focus from addressing the type of individual that joins social

movements to studies of the infrastructure necessary to sustain a movement.

In general, resource mobilization theory could be seen as a reaction to the
deficiencies Of the classical model. These theories viewed social movements
through two distinct conceptualizations. On the one hand, the traditional form
viewed social movements as extensions of elementary forms of collective
behavior (Jenkins 1983). On the other hand, resource mobilization theory viewed
social movements as extensions of institutional actions and constrained the focus
on movements that sought institutional change (McCarthy and Zald 1977). In
other words, the view of resource mobilization theory argued that social
movements were composed of political actors dedicated to achieving their
objectives, which in some cases could be political. This vision of clearly
delineated objectives attributed to the participants within social movements a

perspective of rationality.

Eyerrnan and Jamison (1991) argued that resource mobilization theory
studied the mechanisms through which social movements recruit their members
as well as the organizational forms through which resources and movement
participants are mobilized. Contrary to the approach of collective behavior, the
approach of resource mobilization centered its analysis on organizations instead

of individuals. For some authors, the grievance used by theories of collective

66

behavior to justify the rise of social movements always remains relatively
constant, and social movements form from long-term changes in the availability
of existing resources, the organization of movements, and the opportunities for
collective action (Jerkins and Perrow 1977; Tilly 1978). However, changes in
society also could be caused by changes in interests and new opportunities that
favor the mobilization of individuals. In other words, changes in society could be
occurring by the opening of political opportunity that facilitates for individuals to
organize and mobilize. This View was the starting point for the political process

theory.

3.3 Political Process Theory

The concept of political opportunity also has its origins in the 19603.
During this period, changes in society were stimulating collective actions. The
perspectives of Europeans and North Americans aided the understanding of the
reasons for mobilization, but none of these ways of thinking were able to explain
why individuals supported some movements during a period of time and why
some Western countries experienced more collective actions during the 19603
than others. In reality, the importance of political context in the formation of
collective actions did not rise instantly. According to Tarrow (1998), a brief
analysis of modern history demonstrated that episodes of collective action cannot
be explained only through the deprivation or disorganization of society. For this
author, what varied in history were the opportunities that involved individuals had
for experimentation. Agnew (1977) argued that opportunities for collective actions

were sometimes greater in some places than in others. Consequently, individuals

67

saw in these opportunities a way to advance their causes. However, for political
process theory, what was important were the reasons for the appearance of
collective actions and not how they mobilized. Within this context, for the political
process the most important task was to analyze the cause of structural change
that created political opportunities in society. Following this, Tarrow ( 1998)
identified five dimensions in political process theory (i.e., political opening,
political alignment, appearance of allies, appearance of divisions, and decline in
the capacity of repression). The process of political opening allows individuals
excluded from the decision-making processes of movements to gain access to
them. In nondemocratic systems, the individuals of less-favored classes do not
possess rights and do not have representatives in the decision-making process.
The process of political opening allows the opportunity for these individuals to
organize and mobilize. The organization and mobilization allow new alliances to
be formed, often creating conflicts within the dominant class, and thus generating
divisions that serve as an incentive for collective actions. Political opportunity
also favors the appearance of alliances within society that consequently act as
another stimulus to collective action. Taking Brazil as an example, the division
between hard-line militarists and moderates allowed the mobilization for
collective action in favor of direct presidential elections and the support of society

for social movements struggling for land reform.

McAdam (1982) argued that political arrangements change over time,
allowing certain segments of society to benefit from these arrangements as well

as advance their interests. This author also argues that isolated factors related to

68

the sociopolitical environment and internal factors within social movements are
not sufficient to generate and develop collective action. In reality, the theory of
political process is centered on the hypothesis that social movements are a
product of the linkage between two connected factors. Thus, social movements
vary in characteristics and organization according to the type of political authority
that they challenge. In other words, the changes in social movement strategy are
the result of changes in power structure. Consequently, the collective identity and
the strategy of actions are determined by interactions with the political
environment on the ground and, particularly, with state institutions (McAdams

1982).

Besides all of these theories of social movements, Meyer and
Staggenborg (1996) argued that what is important is the recognition that the
development of social movements, their tactics, and impacts are profoundly
affected by changes in factors external to movements themselves. These
external factors can be linked to the presence of the state as a facilitator or
constrainer of political opportunities. Therefore, the strong presence of the state
in Latin American politics has created specific political conditions for the rise of

social movements in that part of the world.

3.4 Social Movements in Latin America

In the same way that social movements influenced social science in
Europe and the United States, they also influenced Latin America. The European
theory 0f social movements had been built in a social and historical context that

included social-democratic consensus, the growth of state welfare, and a strong

69

corporatist tradition (Foweraker 1995). Consequently, the theory of social
movements sought to explain social movements as a result of social changes. In
the United States, to the contrary, there was not a strong corporatist tradition and
the worker's movement was not as strong as in Europe. Therefore, the
emergence of social movements was explained not by social changes but as a
consequence of the ability of movements to organize and gain representation

within the pluralist political system (Forewaker 1995).

In the beginning, Latin American scholars used the same type of
frameworks developed by scholars in the United States and Europe, and little
effort was put toward the development of alternative models sensitive to the
political and social context of Latin America (Foweraker 1995). Nevertheless, at
the time, scholars observed that resource mobilization theory, as applied in the
United States, was not completely applicable to the reality of Latin America. The
rejection of resource mobilization theory was based on the argument that there
had been a crisis of the State in Latin America with the existence of
nondemocratic regimes. Consequently, for Latin American scholars, the theory of

a pluralist state did not apply to the Latin American context, where the state was

authoritarian.16 For Foweraker (1995), resource mobilization theory should not be
completely discarded once the appearance of social movements with specific
and well-elaborated strategies occurred in countries with nondemocratic regimes.

As an alternative, Latin American scholars searched for a greater focus on the

 

16 There are strong differences among Latin American countries; however, for the Latin American
scholar, what was important to justify the rejection of the resource mobilization theory was the
lack of nondemocratic regimes and a strong dictatorialship in many Latin American countries in
the 1970s and 1980s.

70

theory of political process (Calderén et al. 1992; Escobar 1992; Escobar and

Alvarez 1992). Ultimately, this approach also suffered criticism once the majority
of Latin American scholars perceived that the political opportunity model was too
centered on state actions and consequently had little utility in a region where the

state was considered an opponent (Davis 1999).

In this context, for Latin American scholars, the political opportunity
approach was over-preoccupied with the elite and the resources of individuals
(Foweraker 1995). In other words, this approach was not seen to have any
concern with society. Traditional approaches demonstrated inadequacies in the
characterization of protests that grew in Latin America in the 1980s, principally in
urban areas. These protests were characterized in a manner that differentiated
them from those occurring in the 1960s and 1970s in the United States and
Europe, leading scholars to call them “New Social Movements” (NSM) once
scholars believed that these movements expressed new popular interests

(Foweraker 1995).

In reality, the approach of New Social Movements stemmed from post-
structuralist and post-Marxist frameworks coming from Europe in the 19708 and
becoming popular among Latin American scholars. However, New Social
Movements in Europe appeared as a response to post-industrial contradictions,
while in Latin America these movements appeared as a response to material
demands that sought to satisfy basic needs (Hellman 1992). Within this context,
scholars believed that social movements had the power to transform civil society

(Davis 1999). Thus, the theory of New Social Movements would be perceived as

71

focusing on civil society (Oxhom 1995; Escobar and Alvarez 1992), as well as
allowing questions relating to the meanings and identity of the movements
themselves instead of resource mobilization strategies (Cohen 1985).
Consequently, what made this theory applicable to the region was its autonomy
and distance from the state and formal political process (Melucci 1980; Scott

1990)

Evers (1985) argued that social movements in Latin American rose where
the “left” was suppressed. For example, the development of grassroots
movements in Chile, Brazil, and Mexico are frequently seen as evidence that
these social movements are the result of authoritarian regimes. Thus, these
movements grew and spread in locales with a lack of democratic institutions or in
areas with few opportunities for political expression. The existence of repression
had a strong impact on the rise of social movements in some Latin American
countries, principally in urban regions or areas with few opportunities for political
expression. However, on the one hand, there is no dearth of studies focused on
the role of the state in the strategies and formation of social movements in Latin
America (e.g., Foweraker 1995, Alvarez et al. 1998; Eckstein 2001); on the other
hand, there is a gap in the literature about the mobilization of rural workers (Starn
1992). It is well known that a rural social movement organization in Latin America
is associated with the presence of unequal agrarian structures, lack of
institutionalization, and poor legal and political channels through which rural

workers can pursue their claims. However, the focus of studies about land reform

72

in many Latin American countries has typically been on impacts of land reform

programs.

For this reason, the rise of social movements focused on agrarian reform
in Latin America in general, and Brazil in particular, has recently called attention
to their mobilization tactics (which include the occupation of private and public
properties through encampments) and settlement formation, but never was the
problem of land reform settlement formation explored. Thus, in the next section I
will explore the role of social movement organizations of rural workers in Brazil

and their role in direct action land reform.

3.4.1 The Social Movement Organization of Rural Workers and Direct

Action Land Reform (DALR) in Brazil
Among Latin American countries, Brazil is without doubt one of the

countries with the highest level of land concentration,17 and remains a place
where the struggle for the redistribution of land is not new. This struggle has
existed since the 1940s and 1950s and was initially led by the Ligas
Camponesas (Peasant Leagues) and the Unia”o dos Lavradores e Trabalhadores
Agricolas do Brasil (ULTAB, Union of Rural Workers). The post-war era was
marked by significant conflicts between capitalist and communist nations, and the
cold war dominated the international political scene. As a result, social
movements were frequently associated with communist uprisings, and in the

case of Brazil, social movements were associated with the Cuban revolution.

 

17 Paraguay has the highest land concentration index (0.93) among Latin American countries,
followed by Peru (0.86), and Brazil (0.85) (FAO 2007).

73

The example of communist revolution in many Latin American nations
served as a model to strengthen social movements and increase popular
protests that led the Brazilian government to act to constrain popular revolt. One
of the first steps was the creation of the Superintendéncia de Reforma Agréria
(SUPRA, Superintendence for Agrarian Reform) in 1962. This agency, among
other activities, facilitated the unionization of rural workers that would serve as
the base for the creation of the Conferéncia Nacional dos Trabalhadores Rurais
(National Conference of Rural Workers) and the creation of the Rural Worker
Statute in 1963. However, with the military coup in 1964, social movements and
workers’ unions became clandestine. Despite these conditions, these
associations and movements did not cease their activities meant to stimulate the
struggle for structural changes in the agrarian sector of the country. Preoccupied
with potential popular revolt and attempting to ameliorate a tense situation in the
countryside, the military government created the previously discussed Land

Statute (Estatuto da Terra) in April 1964.

If, in the beginning, the main objective of the Land Statute was to calm
and satisfy the anxieties of the general public, this law would also orient the
future of agrarian politics during the following decades, persisting in its influence
until today. In addition, this law allowed the possibility of land expropriation, with
monetary compensation, in cases where social interest dictated the need.
However, the definition of social interest was associated primarily with the
question of land conflict. The legal benefits that the capitalist class received since

the colonial period facilitated the concentration of land and had been the source

74

of conflicts in various regions of Brazil. The goal of the Land Statute was to
eliminate a property that was a source of conflicts. In other words, the Land
Statute of 1964 had as one of its objectives the elimination of properties of less

than one rural module (minifundio) and the end of latiftindio.

According to the Land Statute, minifundios were incapable of generating
adequate levels of production to meet subsistence needs. In addition, the Land
Statute sought to end IatifLindio that over centuries was seen as the principal
factor leading to the underdevelopment of Brazilian agriculture. To accomplish
this, the Land Statute sought to modernize agriculture. By addressing Brazil’s
problems of agricultural development while simultaneously attempting to resolve
the existing conflicts between small and large producers, the Land Statute would
create further conflicts. in other words, the modernization of agriculture would be
accompanied by a massive mechanization that would bring about the dislocation
of rural workers. The Brazilian literature covering these shifts is rich in examples
of the mass of rural workers that had been dislocated to cities, joining the urban
society already marginalized by a lack of formal employment (Chase 1999;
Browder 1994; Silva 1982). The result of migration to these urban areas was an

increase in protests for agrarian reform.

This period of agricultural modernization was also marked by struggles for
the democratization of Brazil, principally in most industrialized cities. Tired of
violence, economic crises, and repression by the military government, the
Brazilian elite began to indirectly support social movements focused on the

struggle for the democratization of Brazil. Taking advantage of the struggle for

75

democratization, many social movements, unions, and community associations
began to emerge. Among these was the movement of rural landless workers
(Movimento dos Trabalhadores Rurais Sem-Terra, or MST), which is one of the
most important social movements for land reform in Latin American history. Tired
of waiting for ample land reform, this movement was founded in 1984 and it
recognized the necessity to use the occupation of land as an instrument in their
struggle.

Thus, the democratization of Brazil in 1985 opened the doors for the
Brazilian population and social movement organizations to reaffirm the necessity
of land reform. Taking advantage of the end of the military regime, in 1985 the
MST realized its first national congress under the slogan “occupation is the only
solution (“Ocupacéo é a (mica solucéo"). The principal argument of the Landless
Rural Movement, sustained by many rural social movements (e.g., FETAGRI -
Federacéo dos Trabalhadores da Agricultura, CPT - Comisséo Pastoral da
Terra, MLPT - Movimento de Luta Pela Ten‘a), is that land occupations are
necessary in the venture to redistribute land and thus accomplish a more

democratic agrarian structure.

Consequently, the pressure of social movements and the support of civil
society led President José Samey to approve the Primeiro Plano National de
Reforma Agraria in 1985 (I PNRA, First National Agrarian Reform Plan), which
had as its objective the rapid application of the Land Statute and the
implementation of land reform. Once more, land reform remained on paper.

Although various elected presidents have developed their own agrarian reform

76

plans, the concentration of land in Brazil has yet to decline. As a result, leading
social movements (e.g., MST, FETAGRI, MLPT, CPT) rely on direct action land
reform (DALR) (Simmons 2006), which has led to an explosion of encampments
(acampamentos), land occupations, and the creation of settlements

(assentamentos) throughout Brazil (Figure 4).

Between 1992 and 2000, the number of families participating in direct

action land reform (DALR), with the support of the Movimento do Trabalhadores

Sem-Terra (MST), increased from 100,000 to 560,000 families18 (MST 2006).
Solely in the Amazon region, nearly 705 assentamentos were created,
encompassing 185,000 families and 12 million hectares between 1979 and 1999

(Fernandes 2000).

 

18 . .
Land reform numbers vary among government agencres and nongovernmental organizations,
such as The Landless Movement of Rural Workers.

77

 

Number of Occupations

 

 

83 c» 82 m 8% o 8 8 8 8 o 8
V- v- ‘- 1- v- N N N N N N N
Years
-l- Number of Occupations

 

 

Figure 4. Evolution of Land Occupation in Brazil, 1995 - 2005. Source:
INCRA 2005

Thus, the explosion of rural assentamentos in Brazil was called to the
attention of researchers19 (Simmons 2006; Brandéo and Souza 2006; Feamside
2001; Femandes 2000). However, many of these studies have been focused
more on the characteristics of assentamentos and the causes of their success or
failure than on the factors that led to their appearance (Simmons 2006).
Consequently, a significant gap in the literature exists, and the role of land reform
assentamentos in the modification of land cover is only beginning to be
questioned. For instance, Simmons (2006) suggested that new land reform

assentamentos in the Amazon could be responsible for significant changes in

 

19 Brazil has 7,694 settlements involving an area of 74 million hectares and 822,284 families
(INCRA 2007a).

78

vegetative cover. However, her study focuses on the settlements led by social
movement organizations (SMOs). These SMOs typically focus their struggle for
land reform through strong discipline, popular mobilization, political struggle, and
strategies of occupation of private property. Nevertheless, little is known about
the mechanisms that promote the appearance of land reform assentamentos by
settlers who, by and large, are not part of social movements — a topic that I will

address in the following chapter.

79

CHAPTER 4: SOCIAL CONTENTION, LAND
OCCUPATION, MIGRATION AND ASSENTAMENTO
FORMATION

Up to this point in this dissertation l have argued that assentamento
formation has important environmental impacts on the tropical forest. However,
the spontaneous appropriation of unclaimed public land through assentamento
formation is not well understood. The assentamento formation schema is a result
of social processes that can be addressed through a conceptual framework
called contentious politics. This conceptual framework is very common in the
political sciences, though it has not been frequently used in geography with some
exceptions (Simmons et al. 2002, 2007; Wolford 20033, 2004; Simmons 2005).
Wolford (2003a), for instance, applied contentious political concepts in describing
why landless participants decide to join the landless movement in Brazil. She
argued that Brazil has an unequal land distribution and that a few landowners
control the vast majority of land. Thus, the potential benefits of joining social
movement organizations (such as the Movement of Rural Landless Workers) are
clear (see also Wolford 2004). Simmons et al. (2007), on the other hand,
detailed the contentious mechanisms (environmental, cognitive, and relational)
that have led to violent land conflict in southern Para State. Their approach
considered the link between violence, land distribution, and the institution of
property as promulgated in the Brazilian constitution. However, violence in
assentamento formation is not new or uncommon in the Amazon. Simmons et al.

(2002) developed a hypothetical two-stage model of land conflict evolution in the

80

Amazon that acknowledges rural and urban dynamics. The first stage of their

model involves a purely agricultural frontier20 with individualized conflicts
between small farmers and large holders clashing in a competition for land
resources. The second stage of their model of land conflict approaches
organized resistance emerging as a result of urbanization. In other words,
urbanization in this stage facilitates the large-scale political mobilization of
different groups on both sides of the land struggle. Nonetheless, I adopted the
notion that assentamento formation can be nonviolent, and as such, it is different
from the approach used by Simmons et al. (2002, 2007). Thus, the type of
settlement I refer to in this dissertation is not social movement oriented. On the
contrary, I argue that existing direct action land reform (DALR) appeared as a
consequence of the first colonization policies that attracted colonists to the
region, and continue to attract migrants, as early agrovilas grow into towns. In
addition, I argue that as the sons and daughters of early colonists begin to
establish their own properties, it consequently leads to new settlement formation.
Thus, throughout this dissertation l will refer to this type of settlement as
spontaneous.

This chapter seeks to describe the contentious process of assentamento
formation, and places it within a land cover and land use change framework by
describing the underlying processes that lead to assentamento formation, and it

also details the resulting proximate land cover impacts.

 

20 “Frontier” is defined as a region of abundant resources and weak institutions (for more details
Simmons et al. 2002; Mueller et al. 1994; Alston et al. 2000).

81

4.1 Social Contention in the Amazon

In the previous sections l highlighted the ambiguity of Brazil’s land laws
and the institutional policies that favor land concentration. In addition, I
suggested that direct action land reform (DALR) uses this ambiguity as an
institutional cover for targeting large holdings that have not fulfilled their social
function in accordance with the Land Statute of 1964 (Estatuto da Terra) and the
Brazilian Constitution of 1988 (Treccani 2001; Simmons et al. 2007). Also, I
argued that assentamento formation can be addressed by a conceptual
framework called contentious politics. But what is contentious politics?
Contentious politics does not have a specific definition. For instance, McAdam et

al. (2001, 5) referred to contentious politics as:

episodic, public, collective interaction among makers of claim and
their objectives when (a) at least one government is a claimant, an
object of claims, or a party to the claims and (b) the claims would, if

realized, affect the interests of at least one of the claimants.

These authors stated that this definition can be roughly translated as
meaning collective political struggle (McAdam et al. 2001) because contentious
politics, in their point of view, is episodic rather than routine. Also, they used the
term public to eliminate claim making that happens within well-defined
organizations, such as firms. Thus, they concentrate on claims that have political

implications. Therefore, contentious politics in this dissertation will differ from

82

politics “as usual” by virtue of its innovative, often conflictive, tactics deployed in
episodic efforts to redress a social wrong (McAdam et al. 2001; Sewell 2001).
Thus, DALR may be regarded as contentious because it seeks to redress the
maldistribution of land and because its premier tactic, land occupation, falls
outside conventional societal grievance channels.

Contention in the struggle for land in Amazénia results from a
convergence of political interests, social needs, and institutional opportunities.
Political interests have been well described by research on Brazil’s social
movements; and these studies have demonstrated that political interest, such as
the democratization of Brazil or the struggle for land reform, had put together
different social movement organizations with different political agendas (Cardoso
1983; Mainwaring 1989; Assies 1994; Femandes 2000). The social needs for
land reform in Brazil verges on the obvious and has been the discussion of
extensive commentary and much research (Muller et al. 1994; Alston et al. 2000,
2005; Feamside 2001; Simmons et al. 2002; Simmons 2004; Wolford 2003a). On
the other hand, institutional opportunities (a focus of the present discussion) are
less apparent but no less important to the current situation in Amazonia and
throughout Brazil more generally. I refer here to the structure of land law as well
as to the status of certain lands in federal holdings, terra devoluta. The particular
nature of the institutional opportunity results from a definition of property
stemming from the earliest days of the colony and from a relatively novel
statement of land rights found in the new (1988) republican constitution. Under

this set of laws, individuals wanting land have clearly stated rights and may

83

possess land under constitutional stipulations. Private land can be expropriated
if it does not meet its social use function (article 184 of the 1988 Brazilian
Constitution), or it can be expropriated if occupied without contestation for an
uninterrupted period of 5 years (article 191 of the 1988 Brazilian Constitution).
Terra devoluta, which is vulnerable to actions of physical possession by land-
desirous individuals, cannot be expropriated (article 191, §1 of the 1988 Brazilian
Constitution). However, terra devoluta in the Brazilian Amazon is presently the
land most vulnerable to occupation because of its historical origins in the system
of sesmarias.

Private holdings have their defenders, as do public lands dedicated to
distinct social purposes, such as national parks. But no one champions the
sanctity of terra devoluta. In fact, a sense of “free for all” often prevails, with
people taking what they want and paying little or no attention to the legal niceties.
Thus, this dissertation hypothesizes that current property institutions in Brazil,
formal and informal, legal and extra-legal, constitute a primary underlying cause
of land cover and land use change in the Amazon basin via the proximate

causation of assentamento formation.

4.2 The Conceptual Framework

This dissertation deploys contentious politics theory within the land cover
and land use change discourse, and specifically within the underlying/proximate
causation (UPC) framework. I argue here that the UPC framework to date has
tended to provide considerable generalizations across highly aggregate factors

(e.g., demographic factors, economic factors, technological factors,

84

infrastructure, wood extraction, among others; see Geist and Lambin 2002),
which complicates applications to specific processes of possible interest. Thus, I
adapt the UPC conceptual framework to the purposes of the present dissertation
by drawing attention to assentamento formation processes and land cover
change in the Amazon basin, as depicted in Figure 5.

The conceptual framework is not a definitive model, and l base it in part on
anecdotal information collected over the past few years of field visits to the
region. Thus, the conceptual model developed here can be interpreted as a set
of hypothetical statements regarding both assentamento creation and impact on
the landscape. The framework in Figure 5 shows several “underlying” factors
grouped in the upper panel. These underlying causes acting at macro- and
microscales affect population. Thus, to understand how underlying causes affect
population, it is first necessary to understand how macropolicies were developed

and how they affected the Brazilian population, leading to social marginalization.

4.2.1 Macropolicies and Social Marginalization

Following the military takeover of 1964, the federal government pursued
import-substitution industrialization by constraining imports of consumer goods
and stimulating investments in industrialization. To accomplish their goal, the
state invested in infrastructure in the country’s largest urban centers (Howe and
Goodman 1992). Brazilian agriculture in the 19703 was extremely
underdeveloped, with very low yields for world standards. Policies to modernize
agriculture were almost nonexistent, with some exceptions to improve coffee,

cotton, and sugarcane production (Pastore et al. 1976). The expansion of

85

Brazilian agriculture in this period was due mainly to road building programs. In
other words, the military regime enabled the construction of new roads that
facilitated the efforts of farmers to bring more land into cultivation. Thus, the
increase in production was from an increase in cultivated area and not from
increases in productivity. Nonetheless, Brazil still had plenty of unused land in
savanna (Cerrado) areas, but there were no technologies for their utilization
(Mueller and Mueller 2006).

Facing pressures for agriculture development, the military regime took
action to construct conditions for intensive exploitation of the lands. Thus, the
state embarked on policies of “conservative modernization” (Martine 1993).
Seeking to build a modern agriculture, the military regime created a rural credit
system to provide sufficient amounts of financing to agribusiness at low interest
rates (Mahar 1979; Hecht 1985; Browder 1988). Toward this goal, the state
offered low-interest credit for purchasing land and modern inputs. The objective
was to induce farmers to improve their production methods by purchasing
modern technologies, such as tractors (Silva 1981). However, low interest rates
favored large farmers or more capitalized landowners, permitting land
concentration. Consequently, this macropolicy favored the creation of a
minifLinfio-latifundio complex that did not make sufficient land available for
subsistence farming, thus forcing peasants to find work outside their farms in
order to supplement their livelihood (Foweraker 1981). This type of
macroeconomic policy favored agricultural expansion until the beginning of the

19805; but in the mid 19803, inflation was again more than 200% a year, and

86

Brazil suffered its worse recession in the twentieth century. Consequently, a
range of orthodox restrictions was introduced. For instance, wages were
controlled and credit was constrained. The real interest rates on farm loans
became positive; and the incentives and subsidies of the credit policy ceased,
increasing unemployment rates. Nevertheless, during this entire period land
concentration did not decrease. Consequently, the 1980s were known as “the
lost decade” (Weinmann 2007; Cohen and Portes 2003). The search for
economic growth led the Brazilian government to develop and try many
stabilization plans (e.g., Plano Cruzado land ll, Plano Collor I and II), but all
failed. Between 1993 and 1994, the Real Stabilization Plan was successfully
implemented. However, despite the changes meant to modernize agriculture and
the Brazilian economy as a whole, the country still showed one of the worst
patterns of income distribution in the world, with rural poverty levels still high
(Barros et al. 2001). Consequently, rural poverty, unemployment, and land
concentration continued to stimulate migration to urban settings. As a result,
masses of rural workers were led to social marginalization. This process of social
marginalization has streamed two distinct social processes of settlement
formation in the Amazon basin that are described in the conceptual framework
(Figure 5). On the left, the social marginalization lead to the formation of
assentamentos with Social Movement Organization (SMO) leadership, such as
the one lead by the Movement of Rural Landless Workers; while on the right, the
social marginalization lead to the formation of Spontaneous Direct Action Land

Reform settlement. Although the populations at risk are quite similar in both

87

cases, and step forward from the ranks of the rural landless and the urban
unemployed, I argue that a sharp distinction exists between these two types of
settlements. Here, I use the statement developed by Wolford (2003b) to support
the link between contentious political and social movements and my hypothesis
in my framework. In other words, the appearance of SMO-led settlements are
politically motivated, with landless people being the ones with the weakest family
ties and with the least to lose from violating social norms (Wolford 2003b). On the
other hand, I argue that spontaneous settlements appear as a result of family
networks that were developed as a consequence of early land reform policies
that attracted thousands of people to the region and stimulated the appearance
of towns that act as pull factors attracting migrants to the region. Clearly,
whatever the type of settlement, the land law in Brazil favors both the
assentamento formation by isolated individuals and assentamento formation by

social movement organizations.

88

Underlying Causes

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Macro t l p Micro
. Political and Institutional . Migration
. Population Growth . . Rural Poverty
. Technological Change . Household Demography
. Land Concentration . lnforrnation Network
. Poverty
. Migration
t
Affects Population
and Leads to Social
Marginalizatiorj l
SMO DALR DALR
Strong Spontaneous
Political Motivation
Motivation
l d' 'd l
Collective n M ua S
Mobilization 1
Land
Occupation
Contentious Land .. .
Invasions Mobilization
I
Contentious
New SMO-oriented Proximate causes New
Settlement Spontaneous
. Road Extension Settlements
.Accessible Sites v .Agriculture 4
. Within Frontier Expanslon . lnaccessible
. High Land Value - L099|09 _ . Sites
. Private Land - Shlfl Cultivation . Endogenous
. Cattle Expansion Process
. Beyond Frontier
. Low Land Value

 

 

 

Figure 5. The conceptual framework of settlement formation.

89

4.2.2 Migration and Network Theory

Several studies around the world have shown that personal networks of
friends and relatives are crucial supporters of migration (Mahmood 1991; Nair
1991, 1998; among others). No attempt is made here to fully review the body of
migration and network chain theory. Rather than attempt to describe all research
about migration and factors affecting migration, this section is restricted to
highlighting the importance of urbanization and social networks in attracting

people to settlement frontiers in the Brazilian Amazon.

4.2.2.1 Migration and Social Networks in Settlement Areas

Human migration is generally the movement of people across a specified
boundary for the purpose of establishing a new or semiperrnanent residence, and
migration has become an important livelihood strategy for many poor people
across the world (Deshingkar and Grimm 2004). However, the decision to move
can be attached to many factors. An individual's kinship ties, financial welfare,
personal values, and survival itself may be important in deciding to migrate.
Scholars have researched migration in Brazil from many different perspectives.
Some have described interregional movement by analyzing data from
government agencies (Dawsey 1983; Goodman 1989; Merrick 1989), while
others have focused more on communities and based their findings on interviews
with local residents (Muller 1986; Lisansky 1990). Many of these studies have
elaborated hypotheses regarding mechanisms causing migration and the impact

on destination and source area.

90

In general, the migration process can be a result of “pull and push” factors.
The “push” factors are those related to life experience that gives an individual
reasons to be dissatisfied with his present place (Shaw 1976; Dorigo and Tobler
1983). The “pull” factors are those characteristics of distant places that make
them appear attractive. Economic crisis, low social welfare (poverty), and
unemployment all generate dissatisfaction with an individual’s present locale.
Consequently, migration occurs if an individual’s basic needs at a potential
destination are perceived to be greater than at his present location. Investigation
of migration in Brazil has focused on rural-urban migration, a characteristic of
much of the underdeveloped countries. However, important differences exist
between Brazil and other countries. For instance, a high level of industrial
development in Brazil differentiates this country from African nations where urban
population is also expanding (Berry 1986; Gugler 1986). In evaluating rural-to-
urban migration in Brazil, the dominant motivator seems to be “push" conditions
in rural areas. The increasing scarcity of rural labor opportunities, agricultural
technological change, and land concentration (Shaw 1976; Saint and Goldsmith
1980; Goodman 1989; Michaels 1990) have forced many people off their lands.
In addition, a better transportation system facilitates migration and provides farm
workers with the option of living in urban centers rather than rural areas.
However, the same “push” factors of land concentration, agricultural
mechanization, industrialization, and unemployment in urban centers have also
driven people back to rural areas (Lisansky 1990). The Brazilian development

policy of the 1970s has arguably been the most important factor causing

91

migration to rural areas, mainly to the Amazon region (Schimink and Wood 1984;
Moran 1981; Browder and Godfrey 1997). The military government of 1964
promoted programs to stimulate migration and integrate the Amazon region into
the national economy. The efforts of the military government were heavily
concentrated in road building and colonization settlements, mainly along the
Transamazon highway; and many studies have focused on the characteristics of
migrants to this area and on the impacts on the land (Walker et al. 2002;
McCraken et al. 1999; Caldas et al. 2007). Undoubtedly, the Amazon region
experienced rapid rural in-migration and state-led colonization schemes
flourished in the basin. The wish of a better life motivated people to migrate to
the frontier. However, the long-standing rural idea of agrovillages is part of the
past, and urban centers started to proliferate in precarious conditions to become
municipalities, attracting migrants to work in agriculture, cattle ranching, timber
extraction, and other natural resource areas (Godfrey 1992). Browder and
Godfrey (1997) stated that many new towns in the Amazon emerged
spontaneously, especially at the intersection of national and state highways and
at local feeder roads. These authors explained that many of these towns provide
a limited range of functions associated with their strategic location, such as auto
repair, restaurants, hotels, and other services; and they often support only a
small resident population with a limited potential for growth or structural
development. They also noted that other cities at network junctions have “taken
off” and become important locations for extractive forest industries, regional

commerce, and services. Thus, the urban growth idea feeds the imagination of

92

the colonists that passes this idea on through an informal social network.
However, the role of informal social channels, consisting of friends and relatives,
from colonization areas in the Brazilian Amazon is not well understood; but my
anecdotal evidence recognizes them as an important factor for encouraging
additional migration to the forest frontier. Families and friends with migration
experience provide initial contacts and information and, in some cases, may
sponsor the move by a subsequent kin or friend. MacDonald and McDonald
(1964) describe this type of migration as chain migration, and it has alternatively
been labeled “kin-mediated migration" (Lomnitz 1977) and “network migration,”
where the latter term comprehends a variety of social relationships in addition to
kinship (Massey 1987; Mines and de Janvry 1982).

Network migration through friends and relatives in a settlement frontier
can result in several favorable consequences for migrants because the social
network can lower substantially the costs and risks associated with the move to
an unknown area. Kin and friends may aid migrants in a forest frontier in a variety
of ways: they inform migrants about land availability in terra devoluta, they can
assist migrants in finding land, they orient migrants to life in the town, and often
constitute the primary source of continuing social relationships and moral support
once migrants have established themselves at their parcel of land. My anecdotal
observations suggest that expansion of networks can lead to more migration,
which can lead to the expansion of the network. Thus, network chain migration
can be a very important mechanism for supporting migration and settlement

formation in the Amazon region.

93

4.2.3 Settlement Formation and Land Cover Change

Since 1985, with the approval of the slogan “Ocupaca'o é a (mica solucéo"
(“Occupation is the only solution”) by the Movement of the Rural Landless
Workers (MST), the fundamental tactic of any SMO struggling for land reform in
Brazil is land occupation and mobilization (Branford and Rocha 2002).21 Small
towns in the urban settings of the Amazon have limited urban opportunities for
employment. This leads to migrant and urban worker discontent and harbors
feelings of social injustice, thereby creating a group of people desiring a return to
the countryside, as well as one that may be easily mobilized for land occupation
(Wolford 2003a). However, not everyone affected by agricultural restructuring
and political transition join SMOs, such as the MST. Wolford (2003b) argued that
social relations are embedded in a particular spatial context and they are critically
important to the decision to join an SMO. She argued that people who chose to
join the MST during an economic crisis are the ones who have the weakest ties
to family and community and, as a consequence, had the least to lose from
violating social norms. Also, she stated that they were the ones where land
represented tradition, subsistence, social reproduction, citizenship, and
community (Wolford 2003a). Thus, in a country facing continuous economic
deprivation as a consequence of constant unemployment, economic crises, and
land concentration, the cities are transformed into places that provide inspiration

for involvement in the landless movement (Simmons et al. 2002).

 

21 A crucial distinction between occupation and invasion exists. In the discourse of the landless
militants, they occupy land that is not in fulfillment of a social function. However, for landowners,
the landless movement is invading their properties. Thus, in this dissertation, occupation and
invasion will be used as the same.

94

The concept of place has long been discussed by many authors as
important in the understanding of political, economic, and cultural processes (see
Massey 1984; Massey and Allen 1984; Cox 1997). In addition, this concept has
facilitated the understanding of place as areas of resistance (Miller 1994;
Staeheli 1994; McAdam et al. 2001; Wolford 2003b; Simmons 2005; Simmons et
al. 2007). Simmons et al. (2007), for instance, argued that place is a site where
economic, social, and political processes interact across spatial scale,
consequently creating the conditions of daily life where meanings and values
shape up, creating community identity and resistance (see also Agnew 1987;
Woldford 2004). Thus, SMOs, taking advantage of places as source resistance
and inspiration against social injustice, instruct their landless militants (Frente de
Massa) to spend weeks or even months recruiting families in both urban and
rural places. The objective of the landless militant is to target different groups of

individuals, including small farmers and sharecroppers who have lost the right to

use the land, rural wageworkers, posseiros”, and unemployed urban people. All
these impoverished recruits have in common a strong sense of social injustice
that facilitates the identity formation that is critical to the understanding of
membership in social movements, and especially those dedicated to land reform
in Brazil, such as the MST (Wolford 2003a; Simmons et al. 2007). In fact, some
SMOs such as the MST argued that the urban place is important for mobilization
and is a fundamental objective of its national political struggle (Petras 1998).

Consequently, identity formation among militants and the poor creates a

 

22 Posseiro means squatter. However, posseiro in Portuguese does not have the negative
meaning that squatter has in English.

95

conscious framework for contentious performances (Simmons et al. 2007). Thus,
collective mobilization occurs in urban places, and land invasions happen

simultaneously with contentious pressuring for land expropriation.

In general, the process of“ land occupation requires careful investigation,
starting with the identification of a large property believed to be unproductive or
eligible for agrarian reform for a failure to fulfill its social function. Here, the term
social function can assume different forms. It can mean the use of slave labor in
a farm, environmental problems, and so on. Thus, SMOs employ any kind of
information that can be used to justify occupation. However, the identification of a
property involves a complex network of rural workers, militants, lawyers, and
even sympathizers of land reform in public offices. With the information of a
property that is not fulfilling its social function, social movements start the
logistical process of planning the occupation. In general, these SMOs often move
in on targeted lands with military quickness and discipline, looking for areas with
easy access (close to main roads and within the agricultural frontier), usually
private land with high land value. In most cases, the process of occupation starts
late at night when landless militants, wives and children, and their recruits (using
cars, trucks, and buses) occupy the selected property with an encampment
engaging several hundred families (Branford and Rocha 2002). However, this
process is not an easy one. The decision to invade and settle a private property
requires maturity, cohesion, discipline, optimism, and organization (Martins

2006)

96

The organization process for an occupation starts very early, with some
basic precautions. For instance, movement militants contact the media with two
purposes: (a) to make sure that the occupation will be broadcast to the public,
and (b) to assure that they will have testimony and a public record in case of
violence from the landowner (Branford and Rocha 2002). Once on the land, the
families set up a large encampment made up of wooden huts covered with black
plastic sheeting or palm fronds. The process of land expropriation can take
months or years to occur. Nomially, landowners ask for property reinstatement,
and if that occurs, the landless workers are forced to leave the occupied land by
the military police. One common approach used by SMOs to call attention to their
problems is to shift their encampment to the margin of a public road. The main
idea of moving to the margin is to be exposed to the public and media. They stay
encamped and united until land is granted and to make use of the maximum
amount of pressure on the system. Consequently, negotiation takes place; and
the federal agency INCRA helps the encampment with food supplies (cesta
basica). The tactic of occupation can lead to the occupation of the same property
as many times as necessary to gain the land. While in the encampment, the
landless workers develop an internal camp organization responsible for food
production, education, health care, politics, and security, among other tasks. The
objective of the organizational scheme is to develop a process of political-
ideological formation through study groups of radical theory. In addition, landless
workers learn that planning means democratic decision-making. In other words,

landless workers participate in the organization and discussions by making

97

proposals (Martins 2006). The internal camp organization also plans and
employs a variety of other pressure tactics, such as marches, road blockages,
and the occupation of government buildings. The objective of all these tactics is
typically to obligate the government to implement the constitutional clause
mandating the expropriation of properties for agrarian reform. When triumphant,
the landless workers receive parcels and provisionary titles in govemment-
sponsored settlements, as well as investment in infrastructure such as schools,
electricity, and roads. In addition, families can also benefit from credit through the
program Agricultura Familiar. Last but not least, SMOs struggling for land reform
try to keep the communities who have already received land politically mobilized
to support further occupations by the still landless, as well as to keep fighting for
social change (Martins 2006). Thus, SMOs struggling for land reform have, in
their approach of land occupation, a complex combination of tactics and
confidence on constitutional clauses that call for the Federal Government to
expropriate uncultivated land and redistribute it to landless people, consequently

creating new settlements.

This narrative stands in sharp contrast to my own anecdotal observations
regarding spontaneous direct action land reform settlement, where
assentamentos appear to be formed ex-post. In other words, land reform occurs
after individuals acting in isolation, or disarticulated from each other, have
occupied their lands. I argue here that spontaneous settlement formation is not
only a consequence of democratization, but it also appears as a combination of a

social process that started with social marginalization from macropolicies in

98

addition to early land reform policies that stimulated in-migration to the region
(Moran 1981; Smith 1982; Simmons et al. 2002). It is important to remember that
the first colonization policies of the military regime opened the Amazon frontier
and attracted thousands of migrants to the region. These first wave migrants
arrived with their families to live in agrovilas (agrovillages) along the
Transamazon highway. After almost four decades, these agrovilas have grown
into small-sized towns and, as such, act as a “pull” factor, attracting people from
different regions through a network of information. In addition to this, through
the demographic life cycle, old colonist families mature and their growing children
start to search for land in order to begin a family (Chayanov 1966; Walker and
Homma 1996; Wolford 2003a), generating new land demands outside old sites
and consequently stimulating land occupation. Caldas et al. (2007), for instance,
described how the household's agricultural economy shifted from subsistence to
market production in colonization sites and consequently increased degradation
through deforestation. These authors argued that, given initially low levels of
capital and little farm experience, the first colonist families depended on shifting
cultivation and annual crops and that this continued while the children were
young and household dependency was high. With time, the children aged and
their parents acquired agricultural experience, which mitigated risk aversion. With
new family workers, the colonist household was now positioned to invest, through
deferred consumption, in a farming system of greater commercial potential,
typically a ranch, a perennials plantation, or some combination. Therefore,

agricultural activities evolved in concert with deforestation. Thus, deforestation

99

increased in old lotes, increasing the amount of land for cultivation; but,
depending on the agricultural system, environmental degradation can also rise.
Consequently, the new land demand is also the result of worsening
environmental conditions in the old colonization sites. As a result, forestlands in
terra devoluta around old colonization sites are seen as unproductive and are
targeted for occupation and settlement formation. Hence, the same legal
structures, based in the Brazilian land law, that justify land invasion of private

lands also are used to justify the occupation of public lands.

Nevertheless, I argue that the process of spontaneous settlement
formation is different from the ones led by SMOs. lts dynamics start with the
arrival of a young family in a parcel of land that contains primary forest. As the
family matures along the household life cycle, many land use types are used
(Walker and Homma 1996). As young children become adults, they begin their
own demands for land. In other words, with the arrival of the adult phase, the
new adults feel the necessity to start their own properties. Obviously, one way to
solve this problem is through new land occupations. Different from earlier
colonization when colonists did not have any type of family support, in this new
process of spontaneous settlement formation family support is very important.
Using the family’s property in the old colonization site as a base for the new task,
the new posseiro open trails (picadas) in the forest to choose a piece of land to
be occupied (Perz et al. 2007). Two goals can be observed in this process. The
first goal of the new posseiro is to make sure he/she is following the same road's

trajectory (direction) of the old colonization sites. In other words, the posseiro

100

knows where the last colonist is living in the settlement road. Thus, he/she looks

for the end of the colonization road and starts a new trail in the forest (Figure 6).

 

  

PA Surubim

     

PA Alto Para

 

 

 

 

 

Figure 6. Road expansion and settlement formation in terra devoluta.

In sequence, he/she chooses a piece of land to make a marcacéo of the
parcel that, in most cases, has the same size as the parcels in the old
colonization zones. Here, the objective is to show to the land reform agency that
he/she is just following the same pattern created by the federal government in
the old colonization sites. In the beginning of the 19705 the land reform federal
agency was responsible for building the official roads [also called primary roads
or development roads (Perz et al. 2007), such as the Transamazon highway] and

with demarcating 100-ha lots for colonists. To accomplish its task, INCRA also

101

built feeder roads running north-south, thus intersecting the Transamazon
highway perpendicularly at 5-km intervals and extending 5 to 10 km from the
Transamazon, called travessées or vicinais (Perz et al. 2007). In other words,
posseiros just follow this same pattern developed in the road construction; also,
they do not claim more land than the 100 ha that the government used to give in
colonization sites. The second goal takes into consideration the lack of money
and infrastructure necessary to develop and claim the land. Considering that the
area is pristine forest and contains valuable trees, the posseiro starts to count

and mark these trees with the goal of exchanging them for road construction

services in future negotiations.23 However, road building in the Amazon may
involve multiple actors who must negotiate their respective interests (Perz et al.
2007). In the Amazon, land price, good agricultural soils, and other natural
resources (such as timber) are important factors in the opening of picadas (trails)
and roads by landless people seeking land for agricultural production. Equally,
loggers are aware of the national and international timber market, paying close
attention to prices of various timber species, and will build roads to acquire
valuable trees. These two factors make terra devoluta in the Amazon very
attractive. However, it is necessary to negotiate road building to reach this type of

land, extract valuable trees, and gain land title to it by working it. As a result, the

 

23 Arima et al. (2005) describing road building along the Transamazon highway stated that
original federal road infrastructure, composed of the main highway and secondary roads, was
extended in the area. These authors identified two main types of road extension. The first one
comprises simple extension of the original secondary road following the government spur that
they call destination-indeterminate, with this road just replicating the initial settlement geometry of
the old colonization site. The second type of road they referred to as destination-determinate, as
it achieves a discernible spatial objective. Both destination-indeterminate and destination-
determinate roads are typically built by well-capitalized loggers with help from colonists and local
government.

102

social actors involved visualize their preferred route for the road, or (in other
words) the participating interest groups will act according to their own particular
motivations, depending on the natural resources they seek to control. Therefore,
roads can help to demonstrate investments toward productive use before the
state (Perz et al. 2007). Consequently, the next step is the deforestation process,
generally necessary to demonstrate productive use of the land. The initial
deforestation will serve to pressure the Federal Government to demarcate the
land as the land is in fulfillment of its social function, or in other words, it is in
production. Also, land occupation close to the family’s property has some
advantages because forest frontier is, in general, an inhospitable area without
roads or any kind of services, and proximity to the family’s property facilitates the
mobility to work in the new area. Thus, proximity facilitates the exchange of
family labor necessary to consolidate the new property. Also, proximity can help
the new posseiro to secure the new area. The Land Statute clause of morada
habitual (article 11, chapter Ill) almost obligates the posseiro to show that he/she
is using and living on the land, and his/her absence for a long period of time can
lead another posseiro to claim the land. In addition, the location of the new parcel
can facilitate the appearance of a family network that can be used in the future as
a source of resources for mobilization. Obviously, the network can also appear
among posseiros through time by solidarity and confidence. In this case, it can
lead to creation of associations, communities, or another form of grouping, formal

or informal.

103

In general, the mobilization and the appearance of an association in
spontaneous settlements are led by the oldest posseiros in the area and assume

two important roles. First, it is through mobilization that the new posseiros will

negotiate with loggers the creatiOn of roads24 necessary for land consolidation.
Normally, the posseiro opens trails in the forest and uses them for a long period
of time with two main goals: to demarcate the land and to demarcate the valuable
trees. In the process of land occupation, the posseiro demarcates valuable trees
to exchange for road construction, making the area more attractive and valuable.
Simmons (2005) argues that, on the frontier, land values are low, property rights
are ill-defined, and investments are minimal. However, with mobilization of
posseiros, enough trees can be joined to exchange for road construction, which

can stimulate production and raise property values.

One should also note that mobilization is not only important for road
construction. Brazil is going through an important political and socioeconomic
process with pressures for land reform occurring in every part of the country.
Obviously, posseiros have a perception that they can benefit from this process.
But to achieve their goals they need to organize and to mobilize to struggle for
their rights. Simmons (2006) stated that a cognitive mechanism that includes the
processes or events that shape people’s perception exists and sets the stage for
actions and reactions to contention. Thus, the posseiros, using the same type of

contentious politics that occurs in other parts of Brazil, bring people into collective

 

24 For more details about roads creation in the Amazon, see Arima et al. (2005).

104

action through alliance. Clearly, the practice of building solidarity among them
reinforces the collective commitments to pressure INCRA for land reform.
However, alliances are not only present among themselves. Support from
influential allies, such as local syndicates and local politicians, are also key
contributions to the process. Local politicians can benefit from settlement
consolidation through future votes in local elections, and local syndicates can
strengthen their political power with new members (Toni 1999).

The process described above suggests that there exists an endogenous
process that is occurring around old colonization sites. Unofficial roads (Perz et
al. 2007) built by the interaction of landless people and loggers involve a
sociospatial process that results in settlement formation. However, this process is
not the only one affecting spontaneous settlement creation. Migration seems to
be one of the main factors. Through family networks and social ties, the old
colonization schemes work as an engine that attracts people to the region. That
is, a family’s network and social ties inside the old colonization schemes
(agrovillages) serve as a source of information about new areas for land
occupation. Information spreads out rapidly through community networks
(Wolford 2003b) and many posseiros hear about a new area by talking to their
friends or relatives.

Differing from settlements led by SMOs, spontaneous direct action land
reform apparently seeks out unclaimed lands, beyond the agricultural frontier,
where land is of low value, thus minimizing the personal risk of individuals who

occupy land without organizational support or assistance from the state. On the

105

one hand, SMO-led efforts targeted properties pursuant to movement objectives
(such as obtaining productive lands with some infrastructure for their
membership or raising the organizational profile through confrontations with the
state or select individuals); on the other hand, in the spontaneous settlement
formation process, mobilization occurs after the land has been occupied. In
addition, the contention that occurs is for land legalization and assentamento
creation. In such a situation, assentamento creation evidently provides an
institutional mechanism for obtaining land title, and perhaps the liberation of
resources for road building and other social services useful to agricultural activity
in remote areas (Laurance et al. 2001). Consequently, with time and the
consolidation of the settlement, farming increases (or proceeds) and
deforestation increases. In other words, the formation of the assentamento acts
as a proximate cause of land cover change through the farming practices of
project participants. Thus, given the additional resources available for
assentamento formation and the security of land occupation in settlement areas,

deforestation and forest fragmentation may be intensified in the settlements.

In sum, assentamento formation results from the interaction between
broad national processes that operate across spatial scales and the specific local
institutional opportunity afforded by the land law that creates the conditions for

settlement formation.

106

CHAPTER 5: RESEARCH AND SAMPLE DESIGN, STUDY
AREA, AND DATA COLLECTION

5.1 Research Design

My overarching goal is to study the institutional framework and social
processes leading to spontaneous assentamento formation and to assess
associated impacts on the Amazonian forest. This entailed three practical
hypotheses and objectives. The first hypothesis, H1, argues that spontaneous
DALR seeks out unclaimed pubic lands (terra devoluta) beyond the frontier,
given the low land value of primary forest, thereby minimizing the risk of
individuals who occupy land without organizational support or assistance from
the state. The second hypothesis, H2, argues that spontaneous DALR is a
consequence of early land reform policies that stimulated in-migration to the
region. In addition, early migrants have created a local population pursuing
DALR via a “demographic” life cycle, whereby the second generation of the
original settlers now seeks its own land. The final hypothesis, H3, argues that
land cover change and forest fragmentation increase after DALR settlement
officialization by the federal government because of federal resources for
infrastructure creation. To test my hypotheses, three objectives were developed.
The objectives were to (1) interview key informants from the community, rural
syndicates, and government agencies; (2) undertake a case study of newly
formed assentamentos; and (3) conduct a remote sensing analysis of

deforestation occurring in the assentamentos and others like it in the region. I

107

now consider each of the research activities in relation to stated hypotheses and
design.

To address hypothesis H1, l undertook in the summers of 2004 and 2006
key informant interviews of INCRA officials and of community and syndicate
leaders. In these unstructured interviews, I identified seven spontaneous
settlements (four in the municipality of Uruara', one in the municipality of
Mediciladia, one in the municipality of Maraba, and one in the municipality of $50
Joao do Araguaia), and I discussed at length the history of the settlements. I
asked explicit questions in this regard, particularly about the legal issues involved
in land occupation and any possible reactions by municipal, state, or federal
government (as well as by other interested parties, such as loggers). In addition,
a survey questionnaire was applied among posseiros. I argue that H1 is upheld if
my key informants describe the site as easy to occupy regarding legal issues and
other possible risks associated with competition for land.

Objective two involves a case study of the settlements identified in the key
informant interviews, and it is addressed through survey data. Since the idea is to
test H2 where spontaneous DALR is a consequence of early land reform policies
and demographic life cycle, I identified and selected two different regions in the

State of Para where settlement formation is occurring (Figure 7). Five

assentamentos were along the Transamazon highway: PA25 Traira'o, PA
Uirapuru, PA Rio do Peixe, PA Tutui Norte, and PA Surubim; and eight

assentamentos were located in the southern Para State: PA 17 de Abril, PA

 

25 PA is the designation for Projeto de Assentamento (settlement project).

108

Canudos, PA Cabanas, PA Santa Maria do Pontal, PA 1° de Marco, PA Araras,
PA Palmares II, and PA Alegria.

The survey focused on household demography and, in particular, on
familial linkages between assentamento residents and earlier colonists in the
region. Questions addressed household demography (age cohorts, location of
immediate family member, kinship ties in the assentamento and elsewhere in
Uruara), migration history (place of birth, prior residence(s), year of all municipal
moves), and social capital (organization memberships, church, syndicates, etc.).
I also probed respondents about the land occupation process. Specifically, I
wanted to know if the state or other agents (e.g., loggers, ranchers) offered
resistance, whether land titles had been secured, and if not, how long they
expected to wait. Finally, I asked whether assentamento creation had liberated
any federal resources, especially for road building.

Although the survey has developed a great deal of contextual information
and shed light on hypothesis H1, hypothesis H2 is directly tested by a Logit
model. Because the main idea of H2 is to test if spontaneous DALR settlements
are a consequence of early land reform policies that attracted migrants to terra
devoluta. Thus, a dichotomous dependent variable was set equal to 1 (one) if an
individual chose to create a spontaneous settlement, or 0 (zero) if he/she chose
to create an SMO settlement. I expect the predicted value of the dependent
variable to fall mainly within the interval between 0 and 1.

ln specifying the Logit model, it assumed that the response probability is

linear in a set of parameters, flj, such that:

109

P(y =1 | x) =H(,BO+,8X1+,BX2+ ........ +kaK) =H(flO+X,B)
where (,60+Xfl) is equal to f , H being a function of f ; thus, taking on values

strictly between 0 and 1: 0(H (f )(1 , for all real numbers f . This ensures that the

estimated response probabilities are strictly between 0 and 1. Various nonlinear
functions have been suggested for the function H in order to make sure that the

probabilities are between 0 and 1. In the Logit model, H is assumed to be the
logistic function, such that H (f) = exp( f) / [1 + exp( f )] which is between 0 and
1 for all real numbers, f.

This Logit model suggests that the predicted value of the dependent
variable could be interpreted as the probability that some arbitrary individual will
choose to create (and consequently live) in a spontaneous assentamento rather
than to create (and live) in an SMO-led assentamento given a set of explanatory
variables, such as individual length of residence in the area, type of land (terra
devoluta or private land), settler political experience in contention activities,

migration pattern, distance from cities, among other explanatory variables.

110

Parfi State

 

     

,

500 Kilometnn [( f
x l I J

Figure 7. Two regions with assentamento formation in the Para State.

 

 

 

 

Finally, the last objective seeks to analyze land cover changes occurring
on assentamentos found around the municipality of Uruaré on the Transamazon
highway. The remote sensing analysis was undertaken using Landsat imagery
(TM and ETM", paths 227, 226 and rows 62, 63), covering a 3-year period (1986,
1991, and 1999). The satellite imagery used in this dissertation was obtained
from the archives maintained by the Tropical Rain Forest Information Center
(http://www.trfic.msu.edu) at the Global Observatory for Ecosystem Services
(GOES), Michigan State University. Prior to classification, the images were
geometrically registered, radiometrically and atmospherically corrected, and

mosaiced (Markham 2003).

111

Images acquired by satellites can be affected by atmospheric particles
through absorption and scattering of the radiation from the earth surface.
Examination of any image depends upon detection of differences in brightness of
objects and their features. Radiometric and atmospheric correction of remotely
sensed data normally involves the processing of digital images to improve the
fidelity of the brightness value. The main purpose for applying radiometric and
atmospheric corrections is to reduce the influence of errors or inconsistencies in
image brightness values that may limit one's ability to interpret or quantitatively
process and analyze digital remotely sensed images. Consequently, radiometric
correction was carried out to normalize satellite images for factors such as
sensor sensitivity, earth-sun distance variation, sun angle, topography, and time
of data collection. To solve this problem it was necessary to convert a digital
number (DN) into radiance and radiance into reflectance. Coefficients developed
by Chamder and Markham (2003) were used for converting digital numbers to
radiance, and to reflectance when metadata files that present calibration
coefficients were not provided for Landsat TM images. Consequently, the

following equation was employed:
L1 = (LMAXA — LMINA / QcaIMAX)'Qca1 + LMINA
where L1: spectral radiance at the sensor’s aperture in w/(m2.sr.pm)units,26

and QM]: quantized calibrated pixel value in units of digital numbers (DNs)

 

26 Chamder and Markham (2003) used specific radiance units of watts per square meter per

steradian per micrometer w/(m2 .sr.,um).

112

(Chamder and Markham 2003). In the next step, I converted the radiance values

to top of atmosphere reflectance using the following equation:

2
PP = ”'L/z ’d /ESUN,1°COS¢S
where pP = unitless planetary reflectance, L ,1 = spectral radiance at the sensor's

aperture, d = earth-sun distance in astronomical units, ESUN ,1 = mean solar

exoatmospheric irradiances, and ¢S = solar zenith angle in degrees. The resulting

satellite reflectance was then corrected for atmospheric effects producing surface
reflectance. Using ERDAS image software, a model was created in modelmaker
for the radiometric and atmospheric correction. Finally, geometric correction
accuracy was calculated by matching common points on corrected images to
points collected by GPS units from several locations within the Uruaré area.
Image rectification using 8— to 10-point coordinates were applied. Individual
geometric correction was considered acceptable only if the root-mean square
(RMS) was < 0.5 pixels. Figure 8 shows the procedures for performing these
corrections. The last step was image classification. Landsat Thematic Mapper
(1986 and 1991) and Enhanced Thematic Mapper (1999) images were
individually classified. To carry out this task, I used a hybrid
unsupervised/supervised classification method in each individual image (Figure

9).

113

 

Digital \

Number
(DN)

 

 

 

 

V
Radiance

l > Radiometric
Correction

TOA

Reflectance

1

Surface
Reflectance Z

l

Geometric
Registration

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

> Geometric
Correction

 

 

 

 

 

 

 

Figure 8. Radiometric and geometric corrections of Landsat imagery.
Through this method, four thematic classes were identified: forest,

deforestation, clouds and shadows, and water. This approach starts with an
unsupervised classification. ERDAS IMAGINE uses ISODA TA classifier
algorithm to perform this task. The ISODA TA clustering method uses the
minimum spectral distance formula to form clusters. The ISODA TA algorithm
repeats the clustering until a maximum number of interactions have been
performed or a convergence threshold has been reached. In this process I

identified 90 classes, and l recoded the classes to evaluate and test the accuracy

114

of the classification. I used the Raster Attribute Editor to compare original image
data with the individual classes of the thematic raster layer that was created from
the unsupervised classification. The final unsupervised image was than used in a
supervised classification. A filter majority 5 X 5 window was than created in
modelmaker and applied to the supervised images to remove any possible noise.
The last step involved the mosaic of the supervised images. In doing
classification analysis, notably, a significant portion of the land surface is
obscured by clouds. Consequently, this complicates the observations of surface
processes from space. The determination of cloudiness is essential to
deforestation evaluation. To conclude my image preparation, a cloud mask was
produced because my interest was to map the increment in deforestation from
1986 to 1999. The classified images were then overlaid on digital assentamento
maps, thereby enabling the creation of a subset that was transformed into a grid.
The calculation of deforestation associated with assentamento creation was
possible because my deforestation maps showed forest cover before and after
the assentamentos were formed given (a) the time period covered by our image
archive, which predates the assentamento process, and (b) the availability of
information on assentamentos, including dates of creation via public records
called portarias. Also, spatial analysis was undertaken using the landscape
pattern recognition software FRAGSTA TS (McGarigal et al. 2002). To
accomplish this task, I calculated the number of patch and the mean patch size

to characterize the patterns of land cover changes occurring in the

115

assentamentos. I argue that H3 is upheld if deforestation -— and fragmentation —-

shows sharp increases after assentamento formation.

5.2 The Study Area

The dataset used in this dissertation was developed through two National
Science Foundation research proposals titled “Brazil’s Direct Action Land Reform
Movement: Environmental Impacts and Socio-Spatial Dynamics” and “Settlement
Formation and Land Cover and Land Use Change: A Case Study in the Brazilian
Amazon.” The second research proposal is my dissertation proposal that
complements the first one, and both involve research in the State of Para in the
Legal Amazon region. The “Brazil’s Direct Action Land Reform Movement:
Environmental Impacts and Socio-spatial Dynamics" research has the main
objective of understanding direct action land reform (DALR) and to assess its
implications on land cover and land use change in the Brazilian Amazon. To
achieve its objective, this proposal identified several key counties where social
movement organizations (SMOs) are active. My dissertation complements this
proposal by adding the case of spontaneous DALR, in which SMOs do not
support settlement formation. Thus, to achieve the objectives, two locations
where DALR is particularly prevalent in the State of Para were identified and
selected. The first location was in southern Para (Figure 7), specifically, the
Poligono dos Castanhais (Brazil Nut Polygon), an area that has experienced
longstanding conflict overland and where settlements have been formed with
(and withOut) SMO support (Simmons et al. 2007). The second location is the

Transamazon region where much state-led colonization occurred in the 1970s.

116

However, in some areas along this highway, DALR is a relatively new
phenomenon (Figure 7), and it is persisting around some of the federally initiated
colonization projects (Projectos Integrados de Colonizacéo-Ple), such as the
municipality of Uruaré where I identified relatively large numbers of recent

settlements and new migrants.

5.2.1 A Brief Description of Southern Para

Southern Para is a place where boundaries are not clear and many
authors have defined it in many different ways. Kotscho (1981), for instance,
referred to this region as an area that lies along the border with Tocantins and
Maranhao States, the so-called Bico do Papagaio, or Parrot's Beak. Schmink
and Wood (1992) and Campos (2002) described it as the southeastern part of
the State. Simmons et al. (2007) defined this region in a broader way, as those
municipalities that lie within the geographic expanse below the Transamazon
highway in the State of Para and east of the Xingu River. This encompasses the
southeastern comer of Para and also areas west of the Araguaia River beyond
State Highway PA-150, where land ownership has long been in contention.

This place has experienced significant political and socioeconomic
changes since the early 1960s, served as a portal to the “opening" of the
Amazon, and enjoyed a great deal of government interest (Simmons et al. 2007).
Consequently, infrastructure development paved the way for migration and
resource exploitation. As a result, land conflicts between private landowners
(generally members of old oligarchies), posseiros, and landless people flourished

in the region; and many SMOs joined the struggle for land reform in this area.

117

5.2.1.1 A brief history of the two spontaneous settlements in southern Para'

Two spontaneous settlements were identified in southern Para, and these
settlements have similar formation trajectories. Key informant interviews have
similar opinions of how the settlements were created.

The PA Araras was created in 1987 by the federal government. It is
considered a spontaneous settlement in this analysis because it started with 22
posseiros that demarked land inside indigenous areas without an occupation
plan. In the beginning, posseiros were encamped for almost two years without a
problem. The posseiros did not know that the lands were indigenous land. It was
thought the land was terra devoluta. However, the history of free land started to
spread and many new posseiros started to move to the area. The increase in the
number of families started to create conflicts between posseiros and indigenous
people, and the federal agency INCRA had to intervene. The federal agency
decided to remove the posseiros from the indigenous area and negotiated new
land in private property that had primary forest. However, to justify the
settlement, the agency had to combine posseiros and landless people. Each
settler received 10 alqueires (about 50 ha). After the officialization, 90 families
were living in the settlement; although, today just 40 families of the original
settlers are living in the area, and some land concentration started to occur.

The history of PA Alegria is similar to the history of PA Araras. The federal
government officially created this settlement in 1999. This assentamento started
with a small group of four posseiros families that were living in the area for more

than 20 years. These posseiros decided to invite four more families to live and

118

help to secure the area. The area was primary forest with some secondary
growth. The new group joined the original group of posseiros in 1997. However,
the increase in the number of people generated conflict with the landowner who
decided to use force to get the land back. The landowner had a document of
aforamento of use for one part of the land, but he argued that the whole land
belonged to him. The conflict between posseiros and landowner only stopped
when INCRA decided to intervene and evaluate the property. After three months,

INCRA expropriated the land and the settlement was to create.

119

 

Image

l

Unsupervised Classification
. (90 classes)

I

Supervised Classification
(5 classes)

I

Filter
Majority
(5x5) Window

I

Final
1 Classification

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Mosaic Cloud
Mask

 

 

 

 

 

 

 

 

Overlay Settlement
and Subset image 4—

 

 

 

 

 

l

 

 

 

 

 

Subset Deforestation
Image to »————> and
Grid Fragmentation
Calculation

 

 

 

Figure 9. Hybrid classification: deforestation and fragmentation calculation.

120

5.2.2 The Transamazon Study Site

My other study site is the Transamazon highway around the municipality
of Uruara. This city is an old colonization site located in the eastern part of the
Amazon basin between the cities of Altamira in the east and Ruropolis in the
west (Figure 6). This study area is part of the Transamazon colonization project
(Projeto lntegrado de Colonizacéo - PIC), meant to resettle landless people from
different parts of Brazil. The site was established in the 1970s by the Brazilian
government as an agrovila along the Transamazon highway. The term agrovila
was used by INCRA to define a state-directed roadside rural settlement in a
colonization project (Perz et al. 2007). At this time, the Brazilian federal
government was very active along the Transamazon highway. Consequently, the
Uruaré area was under the influence of federal policies that stimulated the inflow
of population and agricultural and urban expansion. This influence was based on
infrastructure creation (e.g., road construction), land allocation (e.g., agrovilas
formation), and agricultural price support (e.g., minimum price). All these policies
influenced the spatial dynamics of the landscape along the Transamazon
highway. For instance, the INCRA was responsible for demarcating 100-ha lots
(400 x 2,500 m and 500 x 2,000 m) for colonists and providing them with basic
services (Moran 1981; Smith 1982). Also, large parcels, so-called gIebas,
ranging up to 3,000 ha were demarcated to develop agro-industry (Caldas et al.
2007). In addition to colonization responsibilities, INCRA was in charge of
building the road system. Consequently, major roads (such as the Transamazon

highway) were important to the colonization plans.

121

The Transamazon highway crosses east—west through Uruara linking the
northeastern part of Brazil to the heart of the Amazon forest. To accomplish its
task of colonization, INCRA also constructed secondary roads running north—
south that intersect the Transamazon highway corridor perpendicularly at 5-km
intervals and extend 5-10 km from the main road. Colonists know these roads as
travessées, or settlement roads. The road system developed by INCRA is
responsible for the appearance of the fishbone pattern in the Amazon landscape,
where spacing between secondary roads did not consider any abiotic parameter,
such as slope or river network. However, in the early 19805 federal government
priorities shifted from colonization projects and road construction toward large-
scale development projects; and many colonization sites, like Uruara, were
abandoned by INCRA (Nascimento and Drummond 2003). On the one hand,
abondonment by the federal government led to a decrease in federal
infrastructure investment in colonization areas; on the other hand, these
colonization sites developed into cities. However, the change from colonization
site to city did not occur by chance. Institutional help was necessary. With the
help of many institutions and private companies, such as logging firms, the
Catholic Church, rural labor, and producer's unions, the abandonment by the
state was remedied (Toni 2003). In addition, the prices of key perennial crops
were important factors in attracting a second wave of migrants and stimulating
market expansion for agricultural production in the mid-19805 (IDESP 1990). As
a result, Uruaré became one of the most important small towns along the

Transamazon highway. However, its development was not planned and the city

122

of Uruaré evolved spontaneously. The arrival of new migrants led the colonists to
split lots that were initially used for agricultural purposes into parcels for
commercial uses. Consequently, urban parcels emerged to accommodate
migrants and services. Later,-lNCRA started to title urban parcels, and education
and public health began to develop. Thus, economic activities flourished, ranging
from farming to ranching to logging/lumber.

Farming in Uruara became the main activity with more than 6,500 rural
households, where more than 70% of the agricultural properties have an area of
< 150 ha (Veiga et al. 1996). The production in the area is divided between
annual (rice, cassava, corn, beans) and perennial crops (pepper, coffee, cacao)
and involves extensive cattle ranching (Ferreira 2001; Toni 2003; Caldas et al.
2007). Ranching in Uruaré is the second most important economic activity with

more than 250,000 head of cattle (IBGE 2007b), and timber production followed

in third with an increase from 115,000 m3 in the mid-19905 to 187,000 m3 in 2003
(Bonaudo et al. 2003). The economic growth of Uruara during the mid-19805 led
to political pressure by the community to become recognized as a municipality.
Consequently, in May of 1988, state law number 5.435 created the municipal
government, which assumed responsibility over the area. Besides some transfer
of funds (based on population size) from the federal government for basic
infrastructure development in the county, little change occurred in rural areas.
However, since the 19905, this county has grown rapidly with the immigration of
small farmers and landless people coming from many regions, principally from

the northeastern and southern regions of Brazil. The Brazilian demographic

123

census shows that Uruara had a population of 37,395 inhabitants in 1996; and it
grew to 45,201 in 2004 (IBGE 2007c). In reality, Uruara has an urban population
around 13,000 inhabitants and a rural population of 32,000 inhabitants,
occupying 10,791 km2 of land (IBGE 20070), which gives a human density
around 4 inhabitants/km2 concentrated along the Transamazon highway and the
secondary roads. This human density is low if compared with cities in more
developed areas of Brazil; however, UrUara is an agricultural frontier area with <
20 years of existence.

In sum, Uruara has experienced important changes since its initial
development, although fundamental characteristics persist, such as its rural
frontier economy based on farming, ranching, logging, and migration.
Consequently, all these characteristics have led to increased pressures on forest

conversion.

5.2.1.1 A brief history of the origins of the surveyed settlements in Uruaré

The settlement Rio do Peixe was created in 1995 by the federal agency
INCRA through INCRA’s document SR01/23 of October 25. The idea driving its
creation was to promote land reform and regularize rural households that were
occupying four glebas27 of 3,000 ha. The areas where the settlement is located
today were alienated by INCRA in 1972 and 1973 because of lack of payment
(Brandéo et al. 1998). The area of the settlement includes the travessées

180,185,190,195, and 200 and encompasses an area of around 26,234 ha

 

27 Glebas are parcels of land put aside by the federal government in the 19705 to promote
colonization through capitalized agro-industry.

124

(INCRA 2007a), although this number does not match the area calculated using
shapefiles obtained from the same agency. The settlement has the capacity to
settle 260 families on lots of 100 ha, but official numbers state that there are 241
families currently settled in the area (Table 3). The process of occupation of
assentamento Rio do Peixe is characterized by different phases. The first
occupations occurred within the boundaries of Gleba José de Souza, in the
beginning of the 19805, and later in the 19905 in terras devolutas of the Union
(Brandao et al. 1998). In fact, the first posseiros started to occupy this area in
1981 when a group decided to build picadas (trails) to demarcate 40 lots of 100
ha (the size of a rural module in the region) to plant small subsistence plots to
assure legal rights to the land (Brandéio et al. 1998). In the beginning, these
people did not move onto the lots. The lack of infrastructure and financial
resources to build houses forced the new arrivals to work as hired laborers or
sharecroppers in parcels located on the Transamazon highway. In addition, the
lack of roads, transportation, and the uncertainty in obtaining land title obliged
many posseiros to sell their rights to their lots.

The PA Tutui Norte has a different formation history, although its origin is
similar and related to the PA Rio do Peixe. This PA was created in 1999 by an
expropriation of a supposed private land (fazenda). The area of this settlement is
around 28,000 ha and began with a negotiation between INCRA and a private
landowner (or fazendeiro) that supposedly had the rights to the land. INCRA was
negotiating this land to give to posseiros, living the area for more than 10 years,

when an occupation occurred to force the settlement formation and to increase

125

the number of families to be settled. This occupation was led by posseiros
removed from PA Rio do Peixe because of lack of water in their lots. The
negotiation between INCRA and posseiros led to an unofficial parcel demarcation
of 70 to 80 ha, on average, toeach family. The PA Tutui has a capacity to settle
341 families, but official numbers state that there are currently 341 families
settled in the area (Table 3). However, the lack of infrastructure led many settlers
to abandon their parcels.

The PA Rio Trairao was formally created in 1997, and the argument that
this land had a previous owner also appeared in the key informant interviews.
The occupation of this assentamento started in 1981 with a spontaneous
occupation of primary forest. Posseiros opened trails to demarcate lots and
negotiated road building with logging companies. However, in contrast with PA
Tutui Norte, the supposed landowner donated the land for settlement formation.
This settlement has a capacity to settle 170 families in an area of 17,000 ha,
although there are currently 160 families settled (Table 3). Like PA Tutui Norte,
settlers in this area have abandoned their lots for lack of infrastructure and
delays in demarcation by INCRA.

The occupation of PA Uirapuru is similar to the other occupations. It
started spontaneously between 1981 and 1985 through the occupation of a
private area that supposedly belonged to a private company called Cotrijui.
However, this company decided to abandon the area because it was having
problems with indigenous groups, facilitating the occupation by posseiros. Key

informant interviews identified three phases in this settlement formation. The first

126

phase started in 1997 with the regularization of 140 lots. The second phase
occurred in 2001 when INCRA regularized lots for 90 families, and the last phase
occurred in 2004 with the settlement of 35 families. The number of families
settled in this period is slightly over the capacity of the settlement, designed for
262 households for an area of 18,900 ha.

The last assentamento used in this case study, PA Surubim, has its own
history linked to the region. It is one of the oldest settlements, and its formation
sometimes merges with the history of the Transamazon highway itself. The
location of the settlement starts very close to the Transamazon highway, and it
started as a spontaneous occupation of terra devoluta in 1988. The government
idea of developing a sugarcane agro-industry in the Mediciléndia municipality
attracted thousands of rural workers to this area. However, the failure of the
sugarcane industry generated a mass of rural workers that were not absorbed by
the colonization plans. Far away from their places of origin, and without financial
resources to return home, this mass of rural workers found the solution for their
problems in the occupation of land. In other words, they encroached on the forest
and started to demarcate parcels of land. Despite being one of the oldest
settlements in the area, it is not in its full occupation capacity. The PA Surubim
settlement has 1,627 families settled for an estimated capacity of 2,660

households (Table 3) in an area of 199,537 ha.

5.3 Data Collection

The data used in this dissertation is part of a major dataset collected in the

State of Para. The data collection was divided into three phases. Phase one

127

involved identifying and interviewing of key informants. In Brazil, settlement
creation is the task of the federal government agency INCRA, who is responsible
for implementing the national policy of land reform. Taking this into consideration,
in the summer of 2004 I visited Uruara to make contact with key informants. On
this visit, a short survey was elaborated and tested with posseiros to serve as
background information for future fieldwork. In the summer of 2006, I traveled to
Uruara and southern Para to identify and interview INCRA officials responsible
for settlement creation in the regions. Also, I identified and interviewed
community and union leaders responsible for supporting assentamento formation
in these two areas. These unstructured interviews allowed me to identify four
spontaneous settlements around the municipality of Uruaré (assentamentos
Trairéo, Uirapuru, Rio do Peixe, and Tutui Norte) and one settlement in the
municipality of Medicilandia (assentamento Surubim) along the Transamazon
highway. Also, eight settlements were identified in southern Para. Four
settlements were located in the municipality of Eldorado do Carajés
(assentamentos 17 de Abril, Canudos, Cabanos, and Santa Maria do Pontal),
two assentamentos in the $50 Joéio do Araguaia municipality (assentamentos 1°
de Marco and Araras), one assentamento in Paraubebas (assentamento
Palmares II), and one in Marabé (Alegria). From these eight settlements, six
were identified by key informants as receiving SMO support, and two were
identified as spontaneous settlements (assentamentos Araras and Alegria), at
least in the phase of occupation where no SMO helped in the selection of the site

or in the occupation. Thus, having identified these settlements, in the next phase,

128

l undertook in situ interviews with key settlers,28 such as presidents of settlement
associations and the oldest settlers living in the settlements. This approach took
into consideration the possibility that the interviewees did not know about the
history of the settlement. Thus, it was necessary to compare responses from
multiple interviewees in a given settlement in order to understand its entire
history. In all these interviews, I discussed settlement history, paying special
attention to how their locations were selected. I asked explicit questions in this
regard, particularly about the legal issues involved in land occupation and about
any reactions by municipal, state, or federal government (as well as by other
interested parties, such as loggers). This method yielded histories of unofficial
settlement formation by nonstate social actors, providing a local record of the
sociospatial processes that generated, expanded, and altered the landscape.
The last research phase involved satellite acquisition and land cover analysis. I
acquired Landsat imagery (TM and ETM+, paths 227, 226 and rows 62, 63),
covering a 3-year period (1986, 1991, 1999) for the Transamazon region around
the municipality of Uruara. In addition, I acquired from the federal agency INCRA
digital assentamento maps showing the location of new settlements in the

Transamazon.

 

28 I decided to use the term settlers, or landholders, instead of landless people because some of
them already have land, although they do not have the title of the land.

129

Noam: 530
o 8:5

58:58 cozsaoaxm 53:3

58:55 cofiaoaxm 8:88 8:88 E
8888 8m 86 :8 8880 <a

nwm Em S E cozwtuoaxw NQFQwo mmamc< ficcfiwmo <n.

88:88 85853 83:8

5mm E 5: Po cozmtaoaxm nt Eco __._n< mu .2. <n_

c2320 362.3

cow—£5298 o
*0 Sun .— u EoEogow

to Ban

cozom Co on»...

>u__mn_o_:32 EoEozuem

 

.EoEEoSO 53:85 2: 3 cos—3.5 .3050 no manna EoEoEow .n Bank

5.4 Sample Design

The data used in the empirical analysis comprise information reflecting
settlement characteristics taken from a settler sample survey, and deforestation
and fragmentation measures derived from the classification of satellite images. A
team of fieldworkers composed of Michigan State graduate students and
undergraduate students from the Federal University of Para, Altamira campus,
administered a 17—page survey (Appendix) during the months of July and August
2006. This survey collected information on lot history (including lot size, year of
establishment, land title, length of residence, and area in forest at arrival),
landownership history, household demography (family size, age composition,
and migration pattern), socioeconomic conditions at arrival, previous experience
with DALR, settlement history (including conflicts, violence, road building, and
land demarcation), and farming system components.

Field data collection in new settlements around Uruara is not an easy task.
To reach a new settlement, it is sometimes necessary to drive hours along the
Transamazon highway and secondary roads. The remoteness of settlement
locations made time a serious constraint on data acquisition. Also, settlement
location is not always precise. In others words, settlers do not know with
confidence where the new settlement starts and where it ends. Many settlements
are in a process of legalization, and only a rough extent of settlement
measurement exists. As a result, posseiros have demarcated their lots, but the
lots are "not officially demarcated by INCRA. Consequently, posseiros use small

areas within their lots to produce subsistence crops. Thus, settlers’ main

131

concern is whether INCRA officials will allow them to keep the area that they had
chosen to live or force them to move. The difficulties of settlement and lot
location complicate formal efforts at geospatial sample design (Griffith 2005).
Consequently, random sampling was not possible by distance (such as every
kilometer) and settler family count (every farmstead) was not viable. Counting
houses was problematic because settlers do not invest in house construction
without official confirmation of legal land status. Settler housing in some cases is
just a camp, or a wattle-and-daub house, depending on their length of residence
in the area. In addition, houses were sometimes hidden from view along the
road. For reasons such as these, the data collection was, in some cases,
opportunistic. In other words, our team conducted interviews with settlers we
could find. In other cases, we first traveled to the settlement to make contact with
community leaders. The main objective of the trip was to ask for help in locating
or gathering settlers for interviews. Community leaders have local knowledge that
facilitated meetings and created confidence between landholders and our group.
I acquired a broadly dispersed sample across settlements, although time was a
major constraint. The sample size represents between 13% and 15% of the
settlement population. Obviously, the sample possesses both systematic and
random spatial components. However, one strong characteristic associated with

settlements is homogeneity of resident populations.

132

CHAPTER 6: RESULTS AND DISCUSSIONS

In the previous chapter I developed a conceptual model to explain
spontaneous settlement formation in terra devoluta. In this chapter, I will present
descriptive statistics of a number of settlements in the State of Para to give a
general overview of their characteristics. In addition, a Logit Model will be
presented and discussed to explain the factors driving spontaneous settlement
formation in the Amazon region. I will conclude this chapter with an evaluation of
the land cover impact that spontaneous settlements are having on the landscape

in the Transamazon region, using deforestation and fragmentation analyses.

6.1 Description of Spontaneous and SMO Settlement
Characteristics

The survey results reveal how concentrated the origins of the settler
population is from the northeastern part of Brazil. The vast majority of
respondents were natives of one of the three following states: Maranhéo, Bahia,
and Ceara. However, in the spontaneous settlements, almost 70% of the
respondents came from the States of Bahia (24.2%), Maranhéo (25.6%), Ceara
(8.1%), and Para (10.9%). Considering the SMO settlement, nearly half of the
settlers came from one state, the State of Maranhéo. The city of Maraba in
southern Para is the main hub of the transportation system in the state, linking
the state to the west-central and northeastern parts of Brazil. Also, the region

borders Maranhéo, which facilitates migration through railroads.

133

Settlers in spontaneous settlements had a previous residence in a rural
area (around 70% of the settlers) before arriving on the settlement lot; while in
SMO settlements, over half (51%) of the settlers came from an urban place. The
average duration of residence for the sample varies between the two types of
settlements. In SMO settlements, the average length of residence is 8 years,
while in spontaneous settlements the length of residence is almost 11 years.
Despite this trend, settlers had arrived from everywhere as far back as 30 years
ago in spontaneous settlements, mainly in the Transamazon area. The mean age
of household head was 47 years at the time of the survey for the spontaneous
settlements and 48 years for SMO settlements. Lack of education is a serious
problem in both types of settlements. The level of education of household heads
shows that almost 90% of the settlers have between 0 and 5 years of education
(an average 2.3 years of education).

In most cases, settlements lack infrastructure. At time of arrival, 40.0% of
the settlers in spontaneous settlements reported the use of trails to access their
parcels and 48.6% reported the use of existing roads. However, existing roads in
spontaneous settlements are so precarious that in most cases they look like a
trail. In SMO settlements, 48% of the landless people reported the existence of
roads when they arrived and only 31.4% reported the existence of a trail (Table
4). For settlers, arrival at the lot is the most difficult phase because everything
needs to be prepared, including building houses. Questioned about their present

and past housing conditions, 40% of the settlers in spontaneous settlements

134

reported that they did not have a house at time of arrival, while 50% of the

settlers in SMOs reported that they did not have a house (Table 5).

Table 4. Existence of previous roads before arrival by type of settlement.

 

Road Trail Nothin Total
Freq. % Freq. % Freq. % Freq. %

Spontaneous 103 48.6 85 40.1 21 9.90 209 98.6*

SMO *
settlements 88 48.1 57 31.1 36 19.7 181 98.9

” Three respondents did not answer this question for the spontaneous settlements and two
respondents did not answer this question for the SMO settlement.

 

Settlement

 

 

 

 

 

 

 

 

 

 

 

 

 

House types vary depending on the financial resources of the household,
and in some cases vary with the household’s length of residence in the area. In
the beginning, very poor settlers build mud houses, and with time, this type of
house is substituted with a wood or brick house. Also, very poor settlers do not
arrive with any kind of personal goods. In fact, almost 68% of the settlers in
spontaneous settlements did not possess any durable goods at time of arrival. In
SMO settlements, this value was less, representing 59% of the settlers (Table 6).
Settlement formation seems to improve the possession of durable goods and
housing conditions. Brick houses become more common after recognition of
settlements by the federal government. Settlement recognition facilitates credit
acquisition through many programs, one of which is PRONAF (Programme for
Strengthening of Family Agriculture). PRONAF facilitates access to credit to
improve the living conditions of rural household families. House construction,
tools, and agricultural input acquisition are among the basic needs facilitated by

PRONAF, although demands and claims to be successful depend on the extent

135

credit acquisition and infrastructure creation.

Table 5. Settler house type at arrival.

to which settlers have been able to organize themselves and pressure INCRA for

 

 

 

 

 

 

 

 

Spontaneous SMO settlement
House type At July At July
arrival 2006 arrival 2006

Wood house 34.1% 60.5% 30.1% 58.0%

Brick house 1.90% 18.7% 5.5% 27.60%
Mud house 24.00% 20.20% 14.4% 7.8%
No house 40.00% 0.60% 50.0% 6.6%
Total 100% 100% 100% 100%

 

 

 

 

 

 

With regards to water supply, most settlements have problems with
access to quality water. At arrival, just 15.4% of the settlers in spontaneous
settlements had wells. This number improved to 56% at the time of the survey. In
SMO settlements, 22% of the settlers had wells. Today, almost 65% of the
settlers in these areas have wells. However, in general, water quality did not

improve at either site.

Table 6. Settlers household possession of durable goods at time of arrival.

 

 

 

 

 

Before Now
Settlements Frequency Percent Frequency Percent
Spontaneous 144 67.9 55 25.9
SMO
settlement 108 59.0 25 13.7

 

 

 

 

 

Note: Durable goods are: stove, chainsaw, refrigerator, generator, television, car, or tractor.

136

 

6.1.1 Settlement Farming Systems and Land Tenure Characteristics

The farming systems in the two settlement types are highly diversified,
integrating annual crops (rice, beans, manioc, and corn), perennials (coffee,
cocoa, and pepper), and cattle. The general characteristics of the overall sample
show that settlements in spontaneous settlements produce more annual crops
than SMO settlements (Table 7). Many factors influence the use of a specific
farming system in a settlement, including settlement distance from the main road
and land title. The spontaneous settlements in the Uruaré region are located an
appreciable distance from the main highway, the Transamazon highway, or BR-
230 (around 41 km); while SMO settlements in southern Para are located 70 km
(on average) from PA-150. However, this distance does not adequately represent
the real transportation difficulties. Settlements located in terra devoluta, mostly in
the Transamazon region, have difficult and arduous access to the closest major
city in the area (the municipality of Uruara), requiring hours to travel a few
kilometers. Travessoes (or settlement) roads are unpaved roads that, even in the
dry season, are almost nontrafficable from a lack of infrastructure and
maintenance. Consequently, settlers usually face difficulties getting around,
especially during the rainy season, further undermining their access to health and
education and making it harder to sell their products at market. Roads in SMO
settlements, mainly located in southern Para, are better maintained, allowing
traffic during most of the year. The distance factor presumably plays an
important, though only partial, role in the choice of farming system because of its

relationship with transportation costs associated with the movement of

137

commodities (Walker et al. 2002). Consequently, settlers in spontaneous
settlements have to rely more heavily on subsistence crops than their SMO
settlement counterparts.

But the annual crop system is not the only prominent system in
spontaneous settlements. Spontaneous settlements also have more area in
perennial crops and pasture than SMO settlements (Table 7), although the
figures are not overiy expressive of this fact. At first glance, these results contrast
with expectations. However, an important distinction between these two types of
settlements exists. Spontaneous settlements have an extensive amount of
primary forest in terra devoluta, while SMO settlements in southern Para are
located in a very traditional latifundio ranching area that long ago relied on its
Brazil-nut trees. As a result, lot size is completely different between these two
types of settlement. The lot size in spontaneous settlements is, on average, over
twice the size of an SMO lot in southern Para. In other words, spontaneous
settlements (mainly in the Transamazon region) have a lot size of 81 ha on
average, while in SMO settlements the average is 35 ha. These numbers seem
to be related to population density. SMO settlements are concentrated in
southern Para, a much more populated region than the Transamazon. In
addition, land concentration in southern Para has left less land available for land
reform. Another important factor to consider when analyzing the difference
between these two types of settlement is the existence of a previous agricultural
system. The characteristic of being a traditional ranching area led settlers in

SMO settlements to dedicate less area to perennial cultivation. Many

138

expropriated farms that have subsequently been transformed into SMO
settlements had immense areas of pastures that were divided into lots. Perennial
crops (such as coffee, cocoa, and pepper) require investments that cannot be
afforded by formerly landless people. This result suggests that settlers in SMO
settlements located in southern Para prefer to (and have to) rely on a traditional
farming system based on extensive pasture rather than the planting of perennial
crops. This approach differs from spontaneous settlements where the existence
of more area in perennials can be seen as related to some market engagement.
In other words, settlers may clear primary forest with the expectation that frontier
development and maturation will bring markets and expanded opportunities for
economic activity, including the sale of their holdings (Binswanger and McIntire
1987; Rudel and Horowitz 1993; DeShazo and DeShazo 1995; Margulis 2003;

Arima et al. 2005).

Table 7. Land use by settlement type.

 

 

 

 

 

Area
Settlement (hectares)
Annual Perennial
1 Pasture
crop crop

Spontaneous 3.83 1 .89 30.5
N = 212 (4.48) (2.51) (44.07)

SMO settlement 2.26 0.24 23.82
N = 183 (2.80) (1.09) (25.98)

 

 

 

 

 

Note 1: The perennial crops include coffee and cocoa and the annual crops include rice, corn,
beans, and manioc.

Note 2: The values between parentheses are standard deviations.

Note 3: The average lot size in the SMO settlement is 35.25 ha, while in the spontaneous
settlement the average is 81.35 ha.

139

This adds another topic to the discussion: the issue of land title. The
expectations that land market will appear and that the sale of holdings can occur
suggests that landowners have land title. Thus, comparing the issue of land title
between the two types of settlements, clearly some contrast exists. In
spontaneous settlements, 34% of settlers have no documentation for their plots,
while in SMO settlements almost 84% have no land title document (Table 8).
When analyzing land reform in Brazil, two important factors need to be
considered. First, land title is the dream of every landless settler, and settlement
creation and emancipation of existing settlements is the dream of the federal
agency INCRA. Second, land title and emancipation are not the immediate
dream of SMOs struggling for land reform. Settlement emancipation can mean
that every settler has their own land and the legal right over it. Thus, the
government interprets land titling as indicating that the settlement is ready to
sustain itself. In other words, emancipation means that the federal government
will no longer invest in infrastructure creation or subsidize credit. However, SMOs
disagree with this argument; land title is not enough for SMOs. Social movement
organizations want settlements with better roads (preferably paved roads),
schools, energy, water, and credit. To achieve their dream, SMOs advise settlers
to not accept land title (key informant interview 2006), while the settlement does
not have all necessary infrastructures to bring social justice and dignity to the

people.

140

Table 8. Land title by settlement type.

 

 

 

 

 

Land Spontaneous SMO settlement
title Frequency % Frequency %
No document 72 34.12 153 83.6
5”” type °f 139 65.88 30 16.4
document
Total 21 1 100 183 100

 

 

 

 

 

 

 

Note 1: type of document could include any of the following: contract of sale, protocol,
or INCRA declaration.
Note 2: One respondent did not answer this question for spontaneous settlements.

This attitude toward title contrasts with spontaneous settlements where
the preoccupation with land title seems to be more evident. In spontaneous
settlements, almost 65% of the settlers reported to have some type of document,
such as a protocol number from INCRA or a contract of sale from a previous
owner (Table 8). Provision of collateral is a common prerequisite for commercial
bank loans, especially for medium- and long-term loans (Domer and Saliba 1981;
Caldas et al. 2007). However, land is useful as collateral only if ownership by the
borrower can be proven. Hence, documents that indicate legal ownership
enhance the owner’s access to institutional lenders, who typically lack personal
knowledge about the potential borrower and his background (Feder and Onchan
1987). Muller et al. (1994), Alston et al. (2000), and others have described the
impact of insecure land tenure on resource use in the frontier and what this
means for agricultural decisions. The results presented in Table 8 seem to
suggest that in spontaneous settlements settlers could be planting more
perennial crops given the existence of some type of land titling documentation.
However, spontaneous settlements also have more areas in annual crops; and

settlers may be assuming risk aversion behavior because, in general, land

141

ownership cannot be proven until settlement officialization. Nevertheless, this
hypothesis is borne out by empirical studies in Brazil (Homma et al. 1996;
Faminow et al. 1999) and new research is linking farming systems to household
characteristics and economic circumstances (Caldas et al. 2007). Whatever the
factors that influence land use decisions, the type of document that posseiros
retain in terra devoluta cannot be used as collateral for credit acquisition.
Although, apparently this requisite is not being applied in settlement areas. More
than 58% of the settlers in spontaneous settlements and 75% in SMO
settlements had some type of credit, varying from money to buy cattle to funds to
build houses and buy equipment. Thus, settlement creation facilitates credit even
without legal documents, probably because settlement recognition takes years to
occun

Land security is a very important factor for posseiros living in the Amazon
region. Posseiros on public lands who lack security of land ownership are subject
to possible eviction, and this type of document seems to give a kind of security
for someone that never had their own land. Even so, eviction is not overly
common in terra devoluta. Two factors can contribute to ownership acquisition:
length of residence and productive use of the land. In spontaneous settlements,
settlers have been living in settlement lots for an average of almost 11 years
(even though some settlers have been living on the same land for almost 30
years) compared to an average of 8 years of residence in SMO settlements

(Table 9).

142

An important factor that merits mention is the time period during which
land was expropriated for settlement formation. In general, spontaneous
settlements in the Transamazon region have been created in lands that were
expropriated in 1971 as part-of the federal decree 1,164 (April 1971) that
allocated a 100-km strip on both sides of federal roads in the Legal Amazon for
colonization proposes. However, in the case of SMO settlements located in
southern Para, land expropriation is more recent, starting at the end of the 19905
and continuing today. Currently, encampments are waiting to become
settlements throughout southern Para, although the process of settlement
formation is not an easy one and land expropriation can take months or years to

occur after initial occupation.

Table 9. Average length of residence in the settlements.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Settlements 8332:“ Length of residence
PA Rio do Peixe 35 12.4 (6.99)
PA Trairéo 28 10.25 (6.89)
PA Tutui Norte 24 3.54 (2.04)
PA Uirapuru 65 12.02 (7.95)
PA Surubim 22 13.59 (7.41)
PA 17 de Abril 57 8.59 (2.72)
PA 1 de Margo 27 6.09 (3.20)
PA Alexia 22 6.95 (2.84)
PA Araras 14 16.86 (4.29)
PA Canudos 23 7.02 (0.80)
PA Cabanos 22 7.06 (1.29)
PA Palmares II 30 10.97 (2.90)
PA Sta. Maria do Pontal 23 9.30 (3.51)

 

 

 

Note 1: Values between parentheses are standard deviations.

Note 2: One respondent in the PA Cabanos did not answer this question.
Note 3: The average length of residence in the spontaneous settlements is 11
years (maximum of 30) and in the SMO settlement it is 8 years.

143

Land occupations are distinct in both settlement types. In the
spontaneous settlements, 12% of the settlers said that they just occupy the land,
while 49% said that they negotiated for access to the land. Those numbers are
different in SMO settlements where 57% of the settlers said that they occupied
the land while 31% said that INCRA negotiated the land with the previous owner
(Tables 10 and 11). Here, an important distinction appears between terra
devoluta and private land. Negotiation and occupation in settlements located in
terra devoluta represent 61% of all settlers. However, negotiation occurred with
the previous owner who demarcated the land for many reasons, such as

speculation, selective logging, or productive use. In this case, settlers buy the

land through a contract that transfers the right to the land to the new posseiro.29

In contrast, in SMO settlements more than 89% of settlers affirmed that they
occupied the holdings of a previous owner, and INCRA negotiated the land and
provided it to them. Also, in spontaneous settlements almost 60% of settlers
affirmed that their lands were terra devoluta, while in SMO settlements 84%
affirmed that their lands were part of a fazenda (or farm) before they arrived
(Table 12). These results suggest that private holdings have their defenders and
INCRA acts to solve a contentious problem between landless people and private

owners.

 

29 . . . . . . . .
Many factors can Influence a posseiro decrsron to sell their land, such as Insecurity, deception,
lack of infrastructure, illness, and so on.

144

Table 10. How settlers acquired land in the spontaneous settlement.

 

 

 

 

 

 

 

 

Type of action Frequency %
Settler occupied 21 12.07
Settler negotiated 86 49.42
Other 5.1 8
Did not answer 58 33.33
Total 1 74 100

 

 

 

Note: Other includes people who inherited or exchanged land and people that
think that INCRA negotiated for the land.

Table 11. How settlers acquired land in SMO settlements.

 

 

 

 

 

 

 

 

Type of Action Frequency %
Settler occupied 126 57.53
INCRA negotiated 70 31.96
Other 3.20
Did not answer 16 7.31
Total 219 100.0

 

 

 

Note: Other includes settlers who exchanged the land.

Table 12. Land type before settlement formation.

 

 

 

 

 

 

 

 

Farm Terra devoluta
Settlements Frequency % Frequency %
Sp°man°°us 12 6.89 104 59.77
settlement
SMO settlement 184 84.02 8 3.65

 

 

Note: Fifty-eight settlers did not answer this question for the spontaneous settlements and
27 respondents did not answer this question for the SMO settlements.

145

 

6.1 .2 Settlement and Contention for Land Reform

Contention for land reform is another important distinction between both
study areas. In spontaneous settlements, only 6% of the settlers were involved in

direct action land reform (DALR) activities (such as strikes, encampments,

manifestations, frente de massa,” or private land occupation) compared with
62% in SMO settlements (Table 13). Since 1985, SMOs have used land
invasions as their main tactic in their struggle for land reform. The main goal of
SMOs is to force the federal government to execute the constitutional clause that
authorizes land expropriation for agrarian reform (Wolford 1996). Through
landless militants, SMOs spend months recruiting families in both urban and rural
places with the objective of targeting different groups of poor people with the final
goal of invading a private property that is believed to be unproductive and
therefore eligible for agrarian reform. In general, these SMOs often move in on
targeted lands with military-like quickness and discipline, looking for areas with
easy access, close to main roads and within the agricultural frontier, usually
private land of high value, as in southern Para. Here, the contentious process for
settlement formation seems to be in contrast with the Transamazon region where
contention appears to be almost nonexistent. In this area, contention occurs after
individuals acting in isolation, or disarticulated from each other, have occupied
their lands. To achieve their goals of land ownership, settlers find the need to
organize and mobilize to struggle for their rights. The collective organization and

commitments help to pressure INCRA for settlement formation (key informant

 

30 Frente de massa is a group of landless militants that work as movement recruiters and guide
occupations.

146

interviews 2007). Consequently, settlement creation affords an institutional
mechanism for obtaining land title and liberation of resources for infrastructure
creation, such as road construction and schools, among other services. Thus,
settlement formation has made landownership and social justice almost possible

for people that have historically been excluded by society.

Table 13. Settler Involvement in DALR activity by type of settlement.

 

 

 

 

Yes No Total
Settlements Frequency % Figuency % Frequency %
Spontaneous 1 1 6.32 163 93.68 1 74 100
SMO 137 62.56 81 36.99 218 99.44

 

 

 

 

 

 

 

 

 

Note: One respondent did not answer this question for the SMO settlement.

6.2 Spontaneous Settlements in the Transamazon Region

Settlement formation is not new. Settlements started to appear at the end
of the 19605 and the beginning of the 19705 as a consequence of land reform
policies. Through the National Integration Plan (PIN-Plano de Integracéo
Nacional), the federal government executed a series of acts, such as road
construction (particularly of the Transamazon highway and the Cuiaba-Santarém
highway), land expropriation (100 km on both sides of federal roads), and the
creation of INCRA with the objective of carrying on its Amazon colonization plan.
Two phases can be described in the process of colonization. The first phase
corresponds to the period of 1971 to 1974 when the State, through INCRA,
created the infrastructure conditions to settle the first colonists in the area, the

so-called colonizaca'o oficial (Moran 1981). The second phase, from 1974 to

147

1980, was characterized by a decreased pace by which land was demarcated.
Until the beginning of the 19805, the federal government was very active in
attracting colonists to the region. However, the loss of interest in colonization by
the federal government did not stop migration, and many families continued to
migrate spontaneously to the area. The lack of interest in attracting people was a
consequence of new objectives. In other words, the State changed its interest
from public colonization (colonizaca‘o oficial) to private colonization (colonizaca'o
particular), thus limiting its actions to the consolidation of existing settlements,
among them the Projeto lntegrado de Colonizacé'o de Altamira (PIC-Altamira)
(Moran 1981). This phase was the starting point of land occupation in terra
devoluta (Brandéo et al. 1998). Becker et al. (1990) argued that in this phase, the

travessc‘ies started to be elongated to augment colonization sites.

6.2.1 Characteristics of Spontaneous Settlements

The farming systems of the settlements in the Transamazon region are
highly diversified across annual crops (rice, beans, manioc, and corn), perennials
(coffee and cocoa), and cattle. Pasture is the dominant land use system and
averages about 31 ha per property. Pasture itself raises land values because of
improvements from cleared forest and preparation for agricultural use (Perz
2001). The nature of the settlements ensures that the size of the land holdings is
very similar to the original plan of the older colonization parcels. In other words,
settlers demarcated parcels as the same sizes found in already legitimated

colonization sites (i.e., 100 ha). In addition, the farming systems of settlements

148

are not much different from other colonization areas and are diversified across
the same component crops (i.e., annuals, perennials, and cattle). However, some
differences exist among the settlements. For instance, the assentamentos Rio do
Peixe, Uirapuru, and Surubim have more area in pasture than the
assentamentos Tutui Norte and Trairao. Also, these settlements have on
average more perennial crops and fewer annuals. This is probably related to the
length of residence in the parcel and distance from the main road (the
Transamazon highway). The average length of residence for the assentamentos
Rio do Peixe, Uirapuru, and Surubim is 12.4, 12.02, and 13.5 years, while the
distance is around 39, 29, and 24 km, respectively (Table 14).

These observations contrast with the other two settlements. The average
length of residence in the PA Tutui Norte and PA Trairao is around 3.5 and 10
years, respectively. In tropical frontiers, welfare is likely to depend on a highly
restricted set of items, given low incomes and lack of access to markets for
consumer goods (Walker et al. 2002). Economic resources, such as available
capital and the lack of access to markets for consumer goods clearly influence
farm decisions. The data suggest that settlers in PA Tutui Norte and PA Trairao
are more dependent on subsistence crops. These results are not different from
other studies in the Brazilian Amazon. Walker and Homma (1996) documented
region-scale switches in farm systems focused on rice production to ones with
large components of perennials, pasture, and cattle along the Transamazon
highway between the early 19705 and 1993. Although their analysis did not track

individual farms but instead was based on two different surveys, they explained

149

 

 

 

the phenomenon in terms of processes observed by CAT (1992) in southern
Para. In particular, household age and maturation of the family workforce enable
farm evolution from annual production to ranching. Also, Walker (2003) and
Caldas et al. (2007) described a two-stage process of farm evolution among

colonists in the Transamazon region.

150

.mmwoécema 5 E505 c2832. Emucflm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . . . . chJ. .o
.8 . 1.... .8... .. .. .8... .. ....__...
. . . . . ojom
Ammmv 8 WWW 8 Amww 8 “WW 8 38.88 8 .6 LEE
. . . . . wwIEN
.8... .. .8... .. 8 .. .8 .. .8... .. .....__._....
. . . . . 8 o
.2. . m... .. 8 .. .8 .. .. .............
.......... .. 8.. .. .8... .. 8.. .. 8...... ..
. . A.“ . . 00cm _w0L
”WWW 8 mommw 8 WWW 8 WWW 8 @8me 8 .6 chem...
. . . . . 8
.8... .. .8... .. .8. .. .8... .. .......1... ..
£8.88 8.83 $58 E8: :28
3.8 8 8.8 8 «.8. 8 8.8 8 8.8 8 87, 62mm
meEeEem mags—mace w 5 33220820 20:58.0:
8...... .. 8.... .. ...... .. .8. .. .8. ..
Ammmv 8 WWW 8 Ammmv 8 mm. w 8 ANN. ”v 8 880
wmflmv 8 Nmmflmv 8 Ammuwv 8 showy 8 Amwflmv 8 880
“mmflmv 8 Amway 8 ”mlmmv 8 Awmmv 8 ANNA-WV 8 6282
Hmwmv 8 Ammflmv 8 Ammmv 8 Ammflwv 8 Amwwv 8 .88
«WW 8 mwmv 8 Ammflmv 8 WWW 8 :me w 8 8E
8.8 8.3 as: 8.: cm:
8.0 8 8.. 8 8. F 8 8.. 8 8.0 8 58
3% on? 55m 33 cum
cams. oiEam :35. 53an :32 29.5.0: cums. 632....“ :35. oEEmm humwmumwwv
5328 E games <6 88;: 232.32; E 98.. 2:5. <6 6. o

 

 

.conaEmmcflh 05 E mozmtouoflmso E68253 msoocflcomw .3. wink

151

In phase one, they argued that colonists are in the early years of farm
formation and face a pure subsistence process with market detachment. In
phase two, colonists face commercial production with a household economy
open to markets for products, labor, and land. The results above seem to
suggest that these settlements are experiencing the same process of farm
evolution. In addition, the demographic characteristics, such as age of the
household head, length of residence, number of children, and years of education
of household head have similar results; and they are not much different from
other colonization areas (see Caldas et al. 2007). However, two points need to
be highlighted here. First, the assentamentos PA Trairéio and PA Tutui Norte
were created more than 10 years ago, but this does not mean much if the parcels
within the assentamento are not demarcated, as is the case in both of these
settlements. Settlers in these two places are not investing in perennial crops
because they are not sure where their lots will be located once demarcation
occurs. They know where they would like to live and are planting in the area, but
insecurity plays a role in land use decisions (Settler personal communications
2006). In other words, risk aversion creates strong incentives to achieve food
security through annual crops. Second, the distance of the two settlements to the
main road is more than 60 km, almost twice the distance of the other three
settlements. Consequently, transportation and market accessibility costs
increase. This lowers output prices as it increases input prices, thereby reducing

the market value of land (Nelson and Hellerstein 1997).

152

6.2.2 Land Tenure Characteristics

Table 15 explores the land title characteristics of each settlement. Today,
65.71% of the settlers in the assentamento Rio do Peixe have some type of
document that confirms their rights to the land while one-third of the settlers still
do not have any kind of land security document. The length of residence in this
settlement varies from a minimum of 1 year to 23 years with the average tenure
being 12.4 years. This suggests that land occupation is still occurring in the PA
Rio do Peixe, which is likely the case given the availability of land for up to 19
families according to lNCRA’s own estimates (INCRA 2007a). On the one hand,
PA Rio do Peixe seems to have more people with some kind of land tenure; on
the other hand, PA Trairao has more settlers without any kind of land security
document. Only 25% of the settlers in this settlement project have some type of
land document. One factor that has contributed to this problem is the fact that the

area where the settlement is located today had been previously claimed by a

logging company that argued to have a carta de sesmarias” (Key informant
interview 2004). The possible existence of conflict between the company and the
posseiros led the federal agency INCRA to act to solve the problem. Meetings
between posseiros and the logging company allowed INCRA to achieve an
agreement to demarcate the settlement, but the location of the parcels have yet
to be defined. A similar problem faced the assentamento Tutui Norte and a
similar solution was used. Even though the same process to create the

settlement was used, the Tutui Norte has more settlers with some form of

 

31 Carta de sesmaria is a land grant document from colonial times.

153

documentation. However, the document may only be a contract of buying the

right to the land from a previous posseiro. The PA Surubim and PA Uirapuru

have fewer settlers without land documentation. Assentamento Surubim is the

oldest land reform settlement in the area, created in 1988. It is also one of the

largest settlements along the Transamazon. lNCRA's report shows that its

capacity is around 2,660 families, although only 1,527 were settled there by 2007

(INCRA 2007a). Much like other settlements in the region, land claims are still

occurring in the area. In contrast, PA Uirapuru is at full capacity with 252 families

settled on 252 parcels. However, some settlers are still waiting for land title.

Table 15. Land title by settlement.

 

 

 

 

 

 

 

 

 

 

 

Some type of
Spontaneous No document document
settlements Freq. % Freq. % Frequency %
PA Rio do Peixe 11 31.42 23 65.71 34 97.13
PA Trairao 21 75.00 7 25.00 28 100
FA Tutui Norte 2 8.33 22 91 .67 24 100
FA Uirapuru 14 21.54 51 78.46 65 100
PA Surubim 4 18.18 18 81.82 22 100

 

 

 

 

Note 1: Some type of document indicates if the settler has a contract of sale, or if the settler has
an INCRA protocol number or INCRA declaration. A contract of sale is a simple document
passing the land right to another settler. An INCRA protocol number is a document that confirms
that the settler has made a petition to the federal government claiming his right to the land. An
INCRA declaration is a provisory land title.

Note 2: One respondent in Rio do Peixe settlement did not answer this question.

6.2.3 The Contentious Characteristics of Settlement Formation

Contention in the struggle for land in the Amazon region rises from a

convergence of many factors, such as political interests, social needs, and

institutional opportunities. All three factors are important for settlement formation.

154

 

However, in the case of the settlements in areas of terra devoluta, political
opportunity facilitates land occupation. This opening emerges as a consequence
of land rights law created by the Brazilian Constitution. Under this set of laws,
individuals wanting land have clearly stated rights. The Constitution establishes
that land must serve a social function, where social function is interpreted as
“being in use." In other words, the principle of “culture efetiva e morada habital”
(effective cultivation and continuous habitation), the basic principles of
usucaption, favor the occupation and use of terra devoluta. As a result, terra
devoluta in the Brazilian Amazon is presently the land most vulnerable to
occupation. The feeling that terra devoluta is free for all seems to avoid
contention in land occupation. Contentious acts in all spontaneous settlements
along the Transamazon highway are almost nonexistent (Table 16).
Approximately 97% of the settlers never were involved in DALR activity, such as

strikes, encampments, etc.

Table 16. Settler involvement in DALR activity by settlement.

 

 

 

 

 

 

 

Spontaneous '"Vgrfg m N°t 'B‘Xfl‘éw m Total
Settlements Freq. % Freq. % Freq. L%)
PA Rio do Peixe 1 2.86 34 97.14 35 100
PA Trairao 4 14.29 24 85.71 28 100
PA Tutui Norte 2 8.33 22 91.67 24 100
PA Uirapuru 3 4.62 62 95.38 65 100
PA Surubim 1 4.55 21 95.45 22 100

 

 

 

 

 

 

 

 

 

Conflict in land occupation does occur when the land has an established
owner. In the settlement Rio do Peixe, Uirapuru, and Surubim, 77%, 80%, and

64% of the respondents, respectively, stated that the area where the settlement

155

is located did not have a previous owner (Table 17). In contrast, 57% and 50% of
the respondents in the assentamentos Trairao and Tutui Norte stated that the
land had an owner before. However, the landowner may have been a previous
posseiro who sold the right to use the land or the respondents were referring to
the logging company. In the first case, a simple contract of sale had solved the
problem of occupation; while in the second case, the possible conflict between
the logging company and the posseiros was solved by lNCRA’s action as stated
earlier. Also, recognition that the land was terra devoluta avoids conflicts in land
occupation. Almost 59% of the occupants in the Surubim settlement stated that
their lands were terra devoluta before occupation. This number is more
impressive for the assentamento Tutui Norte where 91% of the respondents
affirmed that the land was terra devoluta before they arrived. The only settlement
where 17% answered that the land was a Fazenda before occupation was the
assentamento Rio do Peixe (Appendix C). However, in this case, the owner of
the Fazenda did not have the legal rights over the land (Settlers' personal
communications 2006). In other words, the Fazenda area was alienated by

INCRA by lack of payment (Brandao et al. 1998).

Table 17. Existence of previous landowner.

 

Land owner No land owner
Spontaneous before before
settlements

Total

 

Freq. % Freq. % Freq. %
PA Rio do Peixe 7 20.0 27 77.14 34 97.14

 

 

 

 

 

PA Trairao 16 57.14 12 42.86 28 100

PA Tutui Norte 12 50.00 1 1 45.83 23 95.83
PA Uirapuru 12 18.46 52 80.00 64 98.46
PA Surubim 7 31.81 14 63.63 21 95.44

 

 

 

 

 

 

 

 

 

156

Table 18. Land type before occupation.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Terra No
Spontaneous Fazenda Devoluta lnforrnation Total
settlements Freq. % Freq. % Freq. % Freq. %
PA Rio do Peixe 6 17.14 17 48.57 12 34.29 35 100
PA Trairao 1 3.58 19 67.85 8 28.57 28 100
PA Tutui Norte 0 0 22 91.66 2 8.34 24 100
PA Uirapuru 4 6.17 33 50.76 28 43.07 65 100
PA Surubim 1 4.54 13 59.10 8 36.36 22 100

 

 

Note: Four respondents did not answer this question in the spontaneous settlements

In contrast to SMO settlements found in southern Para, where contention
occurred before land occupation (or invasion from the perspective of the owner),
contention in the Transamazon region only occurred after settlers had occupied
their lands. Key informant interviews indicated that 60% of posseiros in the
Transamazon region were on their lots before settlement creation (Key informant
2004). However, to legalize their claims posseiros had to organize themselves
through associations, communities, affiliation with local syndicates, or some other
form of grouping, formal or informal. Using the constitutional clause of morada
habitual, the posseiros attempt to show that they are using and living on the land.
Organization is crucial to obtain settlement recognition. Posseiros acting together
can create structures (such as associations or cooperatives) in order to reach
decision-makers so they can influence INCRA to regularize their land holdings.
Settlement recognition can facilitate infrastructure creation, such as schools,
electricity, and credit access. However, only recognition is not sufficient to gain

these infrastructural enhancements. Settlement recognition helps to assure land

157

security, but to gain infrastructure creation it is necessary to develop a plan, the
so-called Settlement Development Plan (PDA - Plano de Desenvolvimento do
Assentamento). The PDA typically must show the reality of the settlement, such
as its history, soil and vegetation type, water quality, and so on. The PDA should
be an action plan with directions to be executed by INCRA, City Hall, and other
institutions, public or private. This plan is generally created by the settlement
community with the guidance of a specialized organization (such as an N60 or a
private institution contracted by INCRA). With the help of technical support,
posseiros define the size of their lots, the location of roads, the infrastructure
needed in the settlement, the demarcation of legally required biological and
forest reserves, and the choice of better soils for annual and perennial crops,
among other important factors. The PDA is supposed to be initiated as soon as
the families arrive in the settlement. However, in the case of terra devoluta, the
posseiros are living in the area and have already demarcated their land and built
the roads they need. Thus, the PDA plan sometimes conflicts with on-the-ground
conditions and delays posseiros who wish to have the land title to a specific
parcel. Posseiros with no previous technical orientation select their lots using
simple preferences, and these preferences can conflict with the technical
recommendations in the PDA. As a result, convincing the posseiros to change
the location of their lots is necessary if they are not suitable for planting. If
posseiros cannot be convinced to move, they can still have their lots demarcated

— but they assume the risk of failure.

158

In sum, settlement formation in terra devoluta in the Transamazon region
is seemingly “easy" to create, but these settlements face more difficulties than
SMO settlements in southern Para. In addition, their land use systems and road
building in primary forest are creating environmental problems, such as

deforestation and fragmentation, a topic that I will discuss in the next sections.

6.3 The Regression Analysis

In this dissertation, survey data have helped to develop a great deal of
contextual information and have shed light on the hypothesis that spontaneous
DALR seeks out unclaimed lands beyond the frontier, so-called terra devolufa. I
argued that, given the low land value of primary forest areas, posseiros minimize
the risk of occupying land without organizational support or assistance from the
state. But to test the hypothesis that spontaneous DALR is a consequence of
early land reform policies that stimulated in-migration to the region and created
the Ple, I use a Logit regression model (Table 19). More precisely, four Logit
models were used. In these four models, a dichotomous dependent variable was
set to 1 if an individual chose to create a spontaneous settlement or 0 (zero) if
he/she decided to create an SMO settlement. These four models were based on
the conceptual model of settlement formation outlined earlier and should have
many variables in common. However, the main difference among them is related
to the individual or family experience in contentious activities or in urban-to-rural
migration. I chose these two variables to differentiate the models because these
variables characterize most of the settlements that have support from SMOs from

those that do not. In other words, settlements with social movement support

159

have more people participating in contentious activities than settlements in terra
devoluta. In addition, social movement settlements have more people who
migrated from urban-to-rural areas than spontaneous settlements.

Thus, all independent variables were taken from the settlement survey
and included those addressed in the theoretical development of settlement
formation. The conceptual model argues that settlement efforts with SMO
support seek areas to occupy with easy access. In other words, SMO
settlements seek areas close to urban places, preferably through access to a
paved road and within the frontier. Also, in general, these areas are private
property with some type of officialized land ownership. On the one hand, SMO
settlements seek areas close to urban centers; on the other hand, the conceptual
model also argues that spontaneous settlements are formed in low land value
areas with difficult access and distant from urban places. In addition, I argue that
an endogenous process characterizes these settlements. In general, this
process occurs through a second generation of original settlers (or colonists) that
now seeks its own land beyond the frontier on terra devoluta. Consequently, I
argue that if these settlers have kin living in the colonization zone, then it
increases the probability of a spontaneous settlement formation.

Knowledge about available land in frontier areas is difficult to acquire.
Information about free land can minimize risk of land occupation and can have a
positive impact on settlement formation. Previous knowledge can be a
consequence of an informal network that transmits information outside the

region, attracting people and stimulating migration. Thus, the previous knowledge

160

variable captures information acquired about free land from friends and relatives.
Also, length of residence in the area can facilitate social capital and information
networking, thus having a positive impact on spontaneous settlement formation.
The last two independent variables in the models address contentious
experience and urban—to-rural migration. The contentious experience can be
characterized by a settler's own experience (Personal SMO Activity) or from the
experience of a settler’s kin (Kin SMO Activity). In other words, the settler may
have participated in contentious activities himself or someone from his family had
participated in DALR activities. Here, I expect to observe individuals with less
experience in contention having a higher probability in creating spontaneous
settlements rather than SMO settlements. My expectation of this variable can be
explained by the lack of infrastructure that creates difficulties in accessing and
mobilizing people in areas beyond the frontier. Also, SMOs (such as the
Movement of Rural Landless Workers) recruit more people from urban places.
Thus, I argue that people with more urban-to-rural migration will have a lower

probability to create a spontaneous settlement.

6.3.1 Results and Discussion

One of my hypotheses in this dissertation argues that information about
land availability is an important factor in explaining spontaneous settlement
formation. However, the variable that tries to capture this effect (Prior knowledge)
has no effect on spontaneous land occupation. However, knowledge about free
land can be related to length of residence in the area. In other words, individuals

with a long length of residence are expected to have more knowledge about free

161

areas. The result of the length of residence variable is robust across all models,
and suggests that long periods of residence in the frontier increase the
probability for spontaneous settlement creation. Period variables such as this
(of, age of household head) are often taken to represent human capital. In the
present setting, my interpretation is that length of residence proxies community,
or social capital, which functions as a substitute for the institutional resources
carried by SMOs.

Contrary to my expectation, the distance variable has a negative effect on
spontaneous settlement formation. In other words, settlements located farther
away from cities are less likely to be spontaneous. In general, distance variables
are related to transportation costs, because distance increases the prices of
inputs and outputs. However, in the case of settlements in the Amazon region,
attention needs to be paid to the role of distance. Spontaneous settlements in the
Transamazon region are located on average between 18 km and 52 km from the
nearest cities (Appendix E). These distances are less than those (between 17 km
to 98 km) found for SMO settlements, most of which are found in the South of
Para. No doubt, those SMOs settlements are farther from cities. However,
distance measured in kilometers does not necessarily reflect the real cost
associated with transportation given road conditions. In general, it takes less
than an hour to travel 60 km in the South of Para, in contrast to the more than 5
hours to cover the same distance in the Transamazon region. Consequently, a
pure distance measure, as used in the regressions, may not reflect actual travel

times, which is the decisive factor. Future research needs to take the condition

162

of roads into account, and accessibility of urban centers and markets, more
generally.

In the survey, the type of land that existed before settlement formation
was also investigated. The terra devoluta variable demonstrates that land in terra
devoluta is linked to the appearance of a spontaneous settlement. This result is
consistent with the hypothesized effect, and is one of the main institutional
factors differentiating spontaneous assentamentos from those created by SMOs.

Regarding a settler’s personal experience in SMO activities, the results
are in agreement with the conceptual model. Contentious experience in DALR
decreases the probability that an individual will create a spontaneous settlement,
although the kin experience has a sign contrary to expectations. Also, urban-to-
rural migration seems to have no effect on spontaneous settlement formation.
Finally, having relative living in colonization zones does not appear to influence
the formation of spontaneous settlements, as was hypothesized via life-cycle
effects. Although none of the models show significance, one of them is
borderline (model 3), and the signs of the coefficients are all positive across the
various specifications. Thus, I interpret the results in this regard as inconclusive,
and deserving of additional research.

In sum, the regression analyses confirm my hypothesis that spontaneous
settlements have some characteristics that differentiate them from SMO
settlements. Also, the results suggest and reinforce the idea that these
settlements have, in some way, been affected by early land reform that attracted

colonists to the region. The role of colonists as drivers of land cover change in

163

the Brazilian Amazon are very well known (Caldas et al. 2007; Walker et al.
2002; McCraken et al. 1999). Given that the same land use practices used by
early colonists are in use by new settlers, it is not difficult to imagine that the new
settlements could be leading to the same process of deforestation and

fragmentation, a topic that will be evaluated in the next section.

164

mo:_m> N 9w momoLEofia c0953 wo:_w> .907

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I82... 82. u 83... 853 m. 258..
88.0 88. u 888 888 N.5 A sea:
$.on New: 8.8m 8.8m 2o E
3.1%. @367 wwwdv. www.mv. .3959...— woo
com 8m 8m .8 223233 so .385

. 8.8 . .28. . am. 5 . 8. F.
98 o «88 .8 o 83 mm... o 8: mm: c «we... «cease
. m E . .8. a . .8. 5
mam o N3... nor 9 So; vow 0 mad 25M 53535.00 5 :5.
. 88.0. .
5m 0 movd .951 53.5
. 83-. . :N. E . a2.-. . .88-.
8... o «8.... RN o 8a...- Ev o Sm? «mm o 5.8. 3:925 .25..
mmod ”Anew >=>=o< 02m Ev.
. 62-. . 33-. . 8a-.
to o 48. w J .8 o 8... 7 8o o m 5. F- 5.2.3 9.5 .285.
. $03 . .88 . $3. . 88.3
.8 o 89m 08 0 83 8° 0 83 8.. o «NE 8523 «EE
. awn. . and. . 8.8 . .38
8.. o we... 8 o 83 08 o Momma 28 o new... 8528: .o 583
. 8.1 . .81 . 8i . 30.1
8.. o ~88- 8o o 80...... So o «8.0- 08 o 8...? 3.0 9 853.0
. flea . .8... . 8.8 . am. 3
3m 9 v5.0 3m o Nowd can 0 and NNN 0 3: 693355.. 5:...
n A a E20525 N A A. EflquOO N A A. EEoEooo N A a acoEEooo moi-35>
e .23.... n .28.... u .282 F .23.... .8888...

 

 

 

doze—Ea 2.95.33 *0 5.3952 ~53 .2. 638.

165

6.4 Settlement Formation and its Effects on Deforestation and
Fragmentation

Many factors have been pointed out as drivers of deforestation in the
Brazilian Amazon, ranging from the role of markets in southern Brazil to the size
of individual farming households. Human activities (e.g., farming systems,
logging, and road building) can disturb the structural integrity of the landscape
and can compromise its functional integrity by interfering with critical ecological
processes necessary for the maintenance of biodiversity (Gardner et al. 1993;
With 1999). The environmental impacts of farming systems, roads, and logging
have been well documented in the literature with specific relevance to the
Amazon (Arima 2005; Andersen et al. 2002; Walker et al. 2002; Cochrane et al.
2002; Laurance 1998; Reid and Bowles 1997; Schelhas and Greenber1996).
However, much remains to be learned about the effects of spontaneous
settlement formation in primary forest on terra devoluta where forest
fragmentation can be a critical concern. To address the ecological concern of
forest fragmentation, I made use of quantitative measurements of landscape
patterns through the application of landscape metrics. Landscape metrics come
from a specific geographical and ecological subdiscipline that attempts to
quantitatively describe landscape form and structure through both global and
local indices. Landscape ecology theory has stated that landscapes have spatial
organizations that interact between spatial elements and experience
modifiCations over time (Forrnan and Gordon 1986). Various landscape metrics

have been used to quantify aspects of spatial pattern and correlate them to

166

ecological processes (Riitters et al. 1995; O'Neill et al. 1988) and land use and
land cover change (Walker 2004). For my specific case, I selected two simple
metrics that are weakly correlated, the number of landscape patches and mean
patch size, on the hypothesis that settlement formation in primary forest
increases fragmentation. To test my hypothesis, I conducted this analysis in nine
assentamentos found in the Transamazon highway region.

The derivation of the metrics was achieved through algorithms contained
within the FRAGSTATS software package (McGarigal and Marks 1995). First,
spontaneous settlement boundaries were acquired from INCRA in digital format.
In sequence, a GIS overtay function joined the digital map with classified images.
Using an Arclnfo GRID command, I clipped the spontaneous settlements from
the 1986, 1991, and 1999 classified satellite images (Figure 10). The resulting
grids were then used as input to FRAGSTATS to compute selected landscape
metrics. The results show that overall forest cover decreases, while deforestation
and forest fragmentation increases (Appendix C). In general, all settlements
found along the Transamazon highway have experienced increases in the
number of patches cleared and the number of patches in forest, but the mean
forest patch size has decreased (Appendix C and D). However, this study
showed notable differences in magnitudes among the five settlements sampled.
For instance, the assentamento Tutui Norte is the only settlement where the
number of patches in forest does not change much in the study period (Figure
11). Also, the mean patch size of forest increases between 1986 and 1999 (Le,

from 3,554 to 3,992 ha) (Figure 12). The survey data shows that in this

167

assentamento settlers have to rely more on subsistence crops. Annuals involve
limited capital inputs, but require labor for clearing, planting, weeding, and
harvesting. Given these characteristics, annuals receive low prices in the local
market, but they also involve low risks because they produce quickly and price
fluctuations are limited (Perz 2003). Also, annual crops cannot be cultivated
continuously in most of the Amazon without capital inputs (Moran 1994; Sanchez
1994), so new land must be cleared, implying an increase in deforested land over

time (Walker et al. 2002; Perz 2003).

168

I: Settlements
[:1 Clouds
D Deforestation
- Forest
= Water Body

0 25 50 100 Kilometers

Sources: IBGE
Landsat.org
GOES

\

1999 Landsat ETM

\

 

Figure 10. Land cover along the Transamazon region, State of Paré, Brazil

169

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Tutui Norte Forest Patch Change, 1986 - 1999
350 a)
up
300 ~ ,-
3 250 «eff e;
.C
3
a .. fl-
9'5 200 l Forest
3 150 .- - - - -, ICIeared
.n
E
z 100 ~
9 N 0; K
0 -
1986 1991
Year

 

 

 

Figure 11. PA Tutui Norte - Number of Patches, 1986 - 1999. Date of
creation: 10I29I1999

170

 

 

 

 

 

 

 

 

 

l
j Tutui Norte Mean Patch Size Change, 1986 - 1999
4,100 0.6
3997-
4.000 ~~
3* 0.5
’g 3.900 —— 3g
13 ~~ 0.4 g
a 3.800 - a
8 o
-F t
5‘, 3,700 ~ t 0-3 5 +0.03%
4: .:
§ 3600 — ’3
l ' P n.
c ‘r 0.2 c
" a
g 3,500 +— g
*c 0.1
3,400 J~
3,300 — . _ 0
1986 1991 1999
Year

 

 

 

Figure 12. PA Tutui Norte - Mean Patch Size, 1986 — 1999.

Cultivation in the Amazon involves fallow periods that increase regrowth
and renew soil fertility. Although food security is also important, areas in fallow
are also apparently playing an important role. In other words, the increases in the
mean patch size of forest area could be an artifact of secondary regrowth. The
results for PA Trairao show a different history. The number of patches in forest
increases more than tenfold between 1986 and 1999, from 4 to 48 patches
(Figure 13). Also, the number of patches of cleared land increases from 13 to
210 patches (Figure 13). In addition, the mean patch size of forest decreases

dramatically from 3,916 ha in 1986 to 315 ha in 1999 (Figure 14).

171

 

Trairao Forest Patch Change, 1986 - 1999

 

250

200 -

150 ~

100 -

Number of Patches

01
O
l

 

 

 

1991
Year

 

1 999

I Forest
I Cleared l

 

Figure 13. PA Trairao - Number of Patches, 1986 - 1999. Date of creation:

10I07/1997.

172

 

 

Trairao Mean Patch Size Change, 1986 - 1999

 

 

 

 

 

 

 

 

4,500 6
4.000 ~~

—~ 5
c; 3.500 w 3
2 2
g 3,000 - -» 4 a
o o
E 5

g 2,500 -_ 3 _§ -Forest
(D m
+Cleared

f; 2.000 1* f9
8 E
g 1,500 —~ ._ 2 c
8 G
2 1,000 1— i

- 1

500 *-
0 - — 0
1986 1991 1999
Year

 

 

Figure 14. PA Trairao - Mean Patch Size, 1986 - 1999.

Much as assentamento Tutui Norte, PA Trairao does not have parcel
boundaries delimited. However, two differences exist between these settlements.
The length of residence in settlement PA Trairao is around 10 years, while in the
PA Tutui Norte the length of residence is about 3 years. Moreover, the PA
Trairao has more area in pasture. This result suggests that settlers in this
settlement are more confident that INCRA will not remove them from their
parcels. The assentamentos Surubim, Uirapuru, and Rio do Peixe show
analogous outcomes with increases in the number of patches in forest, but with

different outcomes for cleared areas (Figures 15, 16, & 17).

173

 

 

Surubim Forest Patch Change, 1986 - 1999

 

6000

5000 0

4000 ~ ,

1

3000 ‘ l Forest
l Cleared 1

2000 _

Number of Patches

1000 *

 

 

 

 

 

1986 1 991 1 999

E Year
l E

Figure 15. PA Surubim - Number of Patches, 1986 — 1999. Date of creation:
5/18/1998.

174

 

Uirapuru Forest Patch Change, 1986 - 1999

 

 

I Forest
l Cleared

 

 

Number of Patches

 

 

 

1 986 1 991 1 999
Year

 

 

Figure 16. PA Uirapuru - Number of Patches, 1986 - 1999. Date of creation:
6I1OI1997.

175

 

 

 

Rio de Peixe Forest Patch Change, 1986 -1999

l 400 e“

 

350 ~
300 -
250 -

l Forest
200 -

150 -

Number of Patches

100 ~

 

 

 

1 986 1 991 1 999
Year

 

 

Figure 17. PA Rio do Peixe - Number of Patches, 1986 - 1999. Date of
creation: 10/25/1995.

Also, the mean patch size of forest has decreased, and the mean patch
size for cleared area has showed different outcomes depending on the period.
For instance, the assentamento Surubim has showed an increase in the mean
patch size of cleared area between 1986 and 1991 (from 6.09 ha to 15.83 ha),
and a decrease between 1991 and 1999 (from 15.83 ha to 8.43 ha) (Figure 18).
Similar results for mean patch size of forest and cleared area are found for
assentamento Rio do Peixe, although with different magnitudes (Figure 19). On
the one hand, these two settlements showed similar patterns; on the other hand,
the assentamento Uirapuru showed a decrease in mean patch size of forest and

cleared area (Figure 20).

176

 

 

300

Surubim Mean Patch Size Change, 1986 -1999

 

__ ___. ___‘__ _~,‘

250 -

200 -

150 1

100 -

Mean Patch Size (Hectares)

0‘
O
1

T

 

O
1

 

1 986

1 991
Year

 

8.43

 

1 999

18

l
I

16

~~ 12

 

Mean Patch Size (Hectares)

 

- Forest 3

 

-0-Cleared I

 

Figure 18. PA Surubim - Mean Patch Size, 1986 - 1999.

177

 

 

 

Rio de Peixe Mean Patch Size Change, 1986 - 1999

 

2,000
1,800 1
1,600 1
1,400 1
1,200 1
1,000 1

800 1

600 1

Mean Patch Size (Hectares)

 

 

 

 

 

 

 

1 986 1 991 1 999
Year

12

Mean Patch Size (Hectares)

 

 

- Forest
-0- Cleared

 

 

Figure 19. PA Rio do Peixe — Mean Patch Size, 1986 - 1999.

178

 

 

 

Uirapuru Mean Patch Size Change, 1986 -1999

 

3.000

 

2.500 1

2,000 1

 

- Forest

1'500 * +Cleared

 

 

 

1 ,000 1*

Mean Patch Size (Hectares)
Mean Patch Size (Hectares)

500 1

 

 

 

 

1 986 1 991 1 999
Year

 

 

Figure 20. PA Uirapuru - Mean Patch Size, 1986 - 1999.

These three settlements are among the oldest settlements in the sample,
and consequently, pasture has increased over time and constitutes additional
land value. As a result, much deforested land in these settlements has shifted
from crops to pasture, increasing deforestation.

Clearly, from the results above, settlement formation on terra devoluta is
affecting the landscape with different patterns of fragmentation. Before
government authorization, few tracts of deforested land inside the settlements
existed. After settlement creation, infrastructure improved through credit for road
building and agriculture improvements, increasing deforestation and
fragmentation. Consequently, over time this process can bring serious

consequences to the environment. As forest patch size decreases and as

179

patches of habitat become more isolated from one another, species population
sizes dependant upon contiguous blocks of forest may decline below the

threshold needed to maintain their survival and general landscape biodiversity.

180

CHAPTER 7: CONCLUSIONS

Many factors have been identified as drivers of deforestation, and this
work does not deny this affirrnation. However, I argue in this study that much
needs to be learned about driVers of forest loss. This dissertation seeks to
address this problem in a different perspective; I try to comprehend land cover
and land use change in the Brazilian Amazdnia through a direct reference to the
underlying social and institutional circumstances that have contributed to a
source of deforestation in the basin in recent years (and shows no signs of
abating): the case of direct action land reform (Simmons 2006). This dissertation
addresses a very particular type of DALR presently affecting the most vulnerable
parts of the Amazon forest, which is found on terra devoluta, and I refer to as
spontaneous DALR.

The main objective of this research was to investigate the social
processes leading to spontaneously formed settlements (or assentamentos) and
to assess associated impacts on deforestation. To accomplish this, I defined
DALR as a form of contentious politics, and I sought to situate it in a conceptual
framework that has been elaborated by the land cover and land use change
research community. Three hypotheses were tested about spontaneous DALR
and developed with the support of anecdotal field evidence. Two were related to
factors driving the appearance of these settlements, and one related to the
environmental impacts on the landscape. More specifically, the first one stated
that spontaneous DALR occurs on unclaimed public lands beyond the frontier,

given the low land value of primary forest areas. The second hypothesized that

181

spontaneous DALR is a consequence of early land reform policies that
stimulated in-migration to the region as well as from demographic household life
cycle. The last hypothesis stated that spontaneous DALR increases deforestation
and fragmentation. Also, I argued that the underlying/proximate causation (UPC)
framework of land use and land cover change to date tended to provide
considerable generalization across highly aggregate factors, which complicates
applications to specific processes of possible interest. Despite this complication,
l adapted the UPC conceptualization to the purposes of this study by drawing
attention to assentamento formation processes and land cover change in the
Amazon basin. Thus, to achieve my goals, I conducted key informant interviews
with community leaders and government officials about settlement location
selection process and undertook a case study of 13 (or near fully) formed
settlements in the State of Para. In these settlements, I applied a 17-page survey
instrument focusing on household demography and, in particular, on familial
linkages between assentamento residents and earlier colonists in the region.
Questions addressed household demography (age cohorts, location of
immediate family members, kinship ties in the assentamento), migration history
(place of birth, pn'or residence(s), year of all municipal moves), and social capital
(organization memberships, church, syndicates, etc.). I also probed respondents
about the land occupation process. Specifically, I wanted to know if the state or
other agents (e.g., loggers, ranchers) offered resistance and whether land titles
had been secured and, if not, how long they expected to wait. Finally, i asked

whether assentamento creation liberated any federal resources, especially for

182

road building. In addition, I conducted a remote sensing analysis of deforestation
and fragmentation occurring in nine assentamentos found in the Transamazon
highway region with the objective of evaluating their impacts on the landscape.

The first and second hypotheses in this dissertation addressed the
question of spontaneous settlement formation and the role of early land reform
policies. These two hypotheses were evaluated through descriptive statistics and
regression analysis. Descriptive analysis of the data suggests that spontaneous
assentamento formation is really occurring in the Amazon region. In contrast to
SMOs settlements where social movement organizations are playing a significant
role in land occupation, the results confirmed my hypothesis that the lands where
the spontaneous settlements have been formed today were terra devoluta before
occupation. Also, key informant interviews give support to these findings through
the finding that 60% of the people in the settlements were already there before
government authorization of settlement creation (Key informant interview 2004).

I recognize that occupation of terra devoluta is not a new process; it has
been occurring since the colonial period. However, the relatively novel statement
of land rights found in the Constitution of 1988 has facilitated land occupation.
According to this set of laws, private land can be expropriated if it is not meeting
its social use function. Despite the changes meant to decrease land
concentration and modernize agriculture, Brazil still presents one of the worst
patterns of income distribution in the world, with rural poverty levels still high. As
a result, masses of rural workers are led to social marginalization. Here, the

institutional opportunity afforded by terra devoluta favors the occupation of these

183

lands by marginalized individuals. Having the perception that ten'a devoluta is a
“free good,” individuals migrate to the region and start to demarcate the land.
This migration process does not seem to occur by chance. The results suggest
that posseiros with kin in colonization zone may be involved in spontaneous
settlement formation, although none of the models show significance, one of
them is borderline, and the signs of the coefficients are positive across all
models. Even though previous knowledge of the area has no explicative power in
spontaneous settlement formation, length of residence in the settlement is an
important factor in explaining spontaneous settlement appearance. It is important
to remember that the first colonization policies of the military government in the
19708 opened the Amazon frontier and attracted thousands of migrants to the
region. This first wave of migrants arrived with their families; and after three
decades, these areas are acting as a pull factor, attracting people from different
regions. In addition, old colonist families are maturing and their growing children
are starting to search for land in order to begin their own families, thus generating
new land demands outside old sites and consequently stimulating land
occupation. This contrasts with SMOs settlements formed in southern Para. In
this region, marginalized individuals do not act in isolation. They are recruited
and convinced by SMOs to occupy private land believed to be unproductive and,
consequently, eligible for agrarian reform for a failure to fulfill its social use
function. Furthermore, the occupation of private lands creates contention.
Landowners do not simply give up their land in favor of land reform;

consequently, months of contention can occur between landless people and

184

landowners. The results clearly demonstrate this difference. Contention in
spontaneous settlements is nearly nonexistent, and when it occurs, it is ex-post.
In other words, land reform occurs after individuals acting in isolation or
disarticulated from each other-have occupied their lands. Here, the contentious
process occurs for land regularization (or regularizacéo Iatiftindiaria), while in
SMOs settlements located in the southern Para contention occurs for land
expropriation. The logit regression analysis confirmed my expectation.
Contention has a lower probability in influencing spontaneous settlement
formation. However, this result does not mean that the region containing these
settlements will always be “contention free.” Every day, a dozen migrant families
arrive in Uruaré, a county in the sample of spontaneous settlements (Key
informant interview 2004). Many of them will probably have previous experience
with DALR occupation, and this experience can affect the way that land
occupation occurs in the region. The variable in the regression model, which
measures the urban-to—rural movement, evaluated the origin of last migration
before settlement occupation. In SMOs, landless people are recruited in urban
places, generally around the area to be invaded. Migration to the Transamazon
region is still small when compared to southern Para where hundreds of landless
families arrive every day (Key informant interview 2006). As discussed in chapter
4, this part of the State has a paved transportation system that serves as a hub
linking the region to the northeast and west-central areas of Brazil. Thus, it is not
difficult to imagine that as soon as the road infrastructure along the Transamazon

improves, migration will increase. The variable distance from city projects the

185

idea that the larger the distance from cities the less the probability that an
individual will be living in a spontaneous settlement in terra devoluta. However,
improvements in road infrastructure will decrease distances, and thus, will
facilitate SMOs to recruit and mobilize people for DALR occupations.

The third hypothesis in this dissertation questioned the effects of
settlement formation on primary forest through the argument that settlement
creation in terra devoluta increases deforestation and fragmentation. Apparently,
from the fragmentation analysis presented here, settlement formation on terra
devoluta is modifying the landscape. Before government authorization for
settlement formation, the areas where the settlements are located showed few
large tracts of forest inside their boundaries. After settlement creation, forested
areas decrease and landscape fragmentation increases. However, the
magnitude of deforestation and fragmentation will vary with settlement
socioeconomic history. Government approval generates incentives for demands,
such as credit for pasture formation, cattle acquisition, and infrastructure
improvements. Despite this, government approval for settlement formation is not
a sufficient condition to stimulate deforestation and fragmentation. Posseiros
need to organize and pressure INCRA to demarcate their lots for documentation
of their land rights, which facilitates loans from credit agencies. Here, the policy
aspect will play an important role in Amazon land reform. Many studies have
shown that credit lines in colonization areas have favored cattle acquisition and
pasture formation, and there is no evidence that will change in the future. Terras

devolutas along the Transamazon highway have immense areas in primary

186

forest, and if this same approach continues, it suggests that ranch formation will
increase deforestation and fragmentation in this region. From a policy
perspective, it would be wiser to increase credit lines for perennials crops, such
as cocoa, rubber trees, and fruit trees (e.g., cupuacu, Theobroma grandiflorum).
This suggestion seems to be an “old-fashioned” idea. However, the failure of this
policy in the past is linked to bad infrastructure and lack of technical assistance.
The excuse that a market does not exist for these products sounds false. What
make these products noncompetitive with others are their higher transportation
cost, low productivity, and lack of technical assistance.

In conclusion, if the goal of the land reform program in Brazil is to bring
social justice and dignity to the poor, then it is time to recognize past mistakes
and adapt land policy to the new reality in the Amazon that takes into
consideration the environmental problems that current laws are causing. If we do
not act now, the future of the poor in the region will not change; and the same
cyclic processes of land occupation and degradation will occur until no forest will

remain to support life in the region.

187

APPENDICES

_.o.va1t.mo.vat .ofiva. .0::0>0m :. 00:0 05 0o 5.0805 0 m. 5.00.00? .0 000:. :0 0. 3.5000 05:0..0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

05:30:90.
:1- 11..- we- 1.1+ 1:- a. 098.0 .08 a
0:00 tfla
0o :o_toao._n.
:5. a. .08 5
:1 00.000200 0.9.0
00?. 0:0 0oo>>
500000.800
1...... :1- 1.1+ 1...- :o no. 05. 5%.“.va
:. 00:0:0
.008“.
+ + 1 1 1 1 1 1 c. 00?.
.Cfi it! fix! 01¢. .10.? it. .11! .1... z... .911. Omflwwh°%oo
00 0.050
- .. . - - 5.3 00“....
0:0 0.00
11+ 11+ 111 111 11+ :1 :1 05:00.0
0.3.3.090.
05:00.0 050%.:
5+ 1.1+ 20.. m
0 E000. 0:0 0:06
0 0009.500
000... 0:04 don. don. 5.0:00 . 002 009
0:53.. 0.000 0000.0 .051 :09... :02”. ~0m0> 0000:: 00>0n. 00:30.0
0:200“. 0.00.:0> 00%“qu
£800“. 0:005:00 000 0 <
m0.n0_.0> .obcoo gnawwmwg «:0Eco..>:m 2900“. “0500.).
£000 :0N0E< 05 :_ 0 :05 .0>oo 0:0. 0:0 00: 0:0. 00 0.00oE :. 000: 00.00:? 00:0.me

< X_02mmn_<

189

3.3111693: ”Suva... .0::0>00 :_ 00:0 05 3 5.0820 0 0. 3.000009 .0 .x00:. :0 0. 3.5000 05:00.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

05.30:? £080
.8 00:00.0 00:5,...
111+ 0:01. 0:0
00 :oanoi 5.80:0
.00
00+ 0.1+ :1 08+ 0.0+. «0+ Imwmrv UCNJ—
098.0 .080.
no 5390 ..0 00
30
0:0
- + - -80: 23 cm 92
.1... 0:. fl. 85m0_0
“—0 £390
000... 0:01. don. den. 3.0:00 . 009 009
0:503 0.000 0000.0 .05”. :09: .021 u0m0> 0000:: 00>0n. 00:00.0 0.00:0> 00.030
0.900“. . .
00.n0..0> 65:00 0.:00000E00 0:200“. 0:200”. 000.00.). 0:00:300 00000500.
10.000 0:0E:o:.>:m
3:80 5000 :o~0E< 05 :_ 0 :0...o :0>oo 0:0. 0:0 00: 0:0. 0o 0.009: :. 000: 00.n0...0> 20:0Exm

190

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

05000
1 1 000::
0:+ 2.... i.... m: 3 0:+ 005 0:0.
0000:0200
0.0.::0._00 88:
1 - 1 000::
0:+ 1.. 0:+ 0: 0:+ 1 00 ._0 0:0. 00.“.0200
0000:0900 0:0 ”“00”...
0.03::0 . .<
1 - 1 1 000::
w: 1 2. wt... 3+ 0:. 0000 0:0—
000200.00
11+ 0:+ 1+ 0:+ t- 0:08.095
0000000000 _ . 0 “Maw...“
00.< 00.050503
000005000 88:
0:0. N0.50.01
000500.00 0:0 .000”.
002000.
000.000:
15.0, .03 .08:
00801 00052
11+ 11+ 0:- 0:1 0.0050 00 0 0: N
03.02000... 0 .0 0
E0585.
0:- 0:+ 0:1 0:1 E00,. 00
0:.0> .000...
00:00.00.
0.0.0 .0050 0.0000 0:0..
5&3 £002. .639: 28> 00< 0? 29.0 E”. .60 8:900 o_nm_.m> wwmwwww
0:00:0000
0.0.00... 0.0.00“. 0:090.
0.0.00“. 020050.:0010000 E 0E: 0..>:m. 000.002

 

 

3:00. :.000 505.4 05 :. 00:0:0 .000 0:0. 0:0 00: 0:0. 00 0.000E :. 000: 00.00t0> 0000:0000

191

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.02002.
.0503:
- .0080 .30:
11+ 111 2+ 8+ .1 03053 55.: 0:0
000000 .30.:000..m
.000... 5:05.).
+ + + + 1 00.00:.
«.1 f... a at. it. _Nuo 1-1
0500030:
8 .000:
- 1 0:..:0.0 :. .5055. 0:0
11 11 00>.0>:. 0. .30.:0:0..0
.:00:0000. 000:..-
00:00:;
59 :.
0:+ 0:+ 0:+ 0:+ 0:... 000005000
00. 00
0. I .008:
:0 E00
flu...” 20 8.052
0:+ 0:+ 0:+ 0:1 1+ 0 0. 0 0. 0.. 0 0 00 0.5.00
00.000:
.00..
00:00.00.
0N.m 00:00 00:00 0:01.
0 00 00 E0: .0 00:0 0.
:0m:0.. 5. >> E0020 0.00> 00< < . .O 00.“. .. w 0 .0 0.:0...0> WManwmn.
0:00:0000
0.0.00“. 0.0000”. E00<
0.200“. 0.:00.00E00-0.0om E 0E : 0..>:m 00000.2

 

 

3:00. :.000 :0N0E< 0:: :. 001:0:0 .000 0:0. 0:0 00: 0:0. 00 0.000E :. 000: 00.:0:0> 0000:0300“:

192

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

so .005. .000 F.
-- a...
.0 00.0.00... $0000
.8 .85.. .000 F.
11+ 0:1 0:+ 52.0.0120 5020 0:0
.0 00.0.00... 0.0.0 $0000
000000005 A .
.0 0.00.5.0. 0.0mm?
0..- .8505 .0 .0
0:.03 .0 0:0 “050.”.
0.522: 890
.30 «00.0.
00+ m:: .01 its ”CI ENE—LG . Awg Fv
.0 00.0.00... .0 .0 .8000
00.8 :.
.101 001 00+ «5... :1 it... mwgm DEW—
000500.00
0.3.000
0.0+ 3+ 0+ .31.. 0:+ 0C1 000C!
00.0 0cm. Camp.
000500.00 :0:0.n.
0.05:0.00
00+ 00+ 111 0:... 11+ 1.11 .09.:
00.0 0:0.
000500.00
00:”..0000. :..00>> 0~.0 .00:00 0.0300 00< 995 0:0.. ..00 00:05.0 00. 3
59.3 0.08.: 98> 09.. . . .00. . . 0.8....3 Bwnowm.
500:0000
0.0.00". 05000050012000 0.0.00“. 0.0.00“. 500<
50E:0..>:m. .0502

 

 

3:00. 5000 50:2. 0:. c. 00005 .0>00 0:0. 0:0 003 0:0. .0 0.000.: :. 0003 00.:0..0> 0000:0306

193

 

0C+

£+

0:.-

0+

0200.200
00.<

NOON

..0 .0 000.00

 

0:+

0C+

€50.50
.0.0

0.00.. n v
30:0“. 0.0
0:0 0:30>.
0.0:.0

 

0:.-

arc-0+

3.0 0.00»
\- A >50:0.
5005000
0:0
000.5.
500.0“.

.38..
30:82.0
0:0 :0...

 

2.1+

0:+

0:+

0:+

0:1

0C1

0.03::<

5.3
0.0.5200.

 

0:+

0C.-

0:.-

0:1.

0:+

00+

0.500 5.3
0.055.

 

01+

0:1

0:+

0C1

00...

0:+

0.05:0.00

:55
0.0594

 

0:+

0.1+

0:1

05+

0:+

0.000 :55
0.05:0.00.

 

0.1+

0:+

0:1

0:.-

0+

0.05:0.00.

 

0:+

0C...

0:+

0:+

£+

0.300

.0000.

..0 .0 .0502.

 

00:00.00.
.0
59.0..

 

5.003

0~.0
b.0020.

 

 

.850
0.00>

 

0.0300
004.

 

00<

 

:.0..O

0003
.00.

 

=00

00:05.0

 

 

0.0.00". 05000050012000

 

0.050....
50E:0..>:m

 

0.0.00“.
.050...

 

0.:0..0>
500:0000

 

00.020
000300“.
5.09..

 

 

3:00. 5000 50:2. 0:. :. 00:0:0 .0>00 0:0. 0:0 003 0:0. .0 0.000... :. 0003 00.:0..0> 5000:0506

194

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.5.0020; .000:
6520.0 20.3.2 0:0
10+ :. 00>.0>:. 56:80.5
”20050000 .0005
53 c.
:+ 0:... 00.00.0000
2... 0......
.0>...0 00:.0 .
0:0 00.022.
+ 0:- 00.00.0000 00 0:000
. 00.0.00: .
.02...
0.3.00
.00:3
0:... 0:+ 00.0 9.0.
000200.00
0.02:0.00 A .
0:+ 0:- .00:3 .0 _%:0W_.
00.0 0:0. .00. : m
000200.00 0:0 ”“0”...
0.03::0 . .<
- - .00:3
0: 0: 00.0 0:0.
000200.00
:+ .+ 20.5002.
.000:
0:+ .+ :+ 00.00.0000 ..00> 0:0
00.< .0x0...0:00>w
00.00.0000
0:+ 0+ 0:0. Emma.”
000200.00 0: 0. .0030
..0.00< .000. :0.0..:0 :020>> :05. 0N5 00.035
. . 0
.500 :00 0 .0; .09.... 0 0 .0 2.20“. .cwwmmmwo 000:8.
0.0.000 .00003505 0.0.000 0200502000200 20m<

 

 

.200. £000 :0N0E< 0:. :. 090:0 .0>00 0:0. 0:0 003 0:0. .0 0.0002 :. 0003 00.0009 b0.0:0.0xm.

195

 

.10...

0+

0.200 :25

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.0.2900 .080.
:+ 0:+ 20.20.00 ..0 .0 0.03
:+ 0- 0.000
000.0..000c.
.0 0.00.20. 800..
.00.20:0 0.030.0
0+ 050: .0 0:0.0x:0.0
0.50020. .6000
.30 .008.
0:+ 0:+ 1- 002.0 ._ 0 .0 %th
.0 00.0.00:
00.0. :.
0:+ 0:+ t.- ..- :- 00.0 0:0.
000200.00
00.0
0:- 0:- 0:+ -+ 0:+ 0.3.000
000200.00
00.0 000..
0:+ t- :+ 0:- :+ .03::0 :0:0.0
000200.00
0:+ 2.- :+ :+ :+ 0.02:0.00
.00:3
00.0 0:0.
000200.00
:+ 0:- 0:+ 0:+ £00..
:0:0.0 0:0
04%“. £280....
.>:0.3.2
..0.00< .000. 5.220 :020>> :05. 0~.0 00.030
.60. .090 0.... 09.: u .0 0 2.50“. 0.00mm> 000:8.
0.0.000 .0:0..3...0:. 0.0.000 0200002000600 200: 00 209.

 

 

3:00. :.000 0802.0 0:. :. 00:0:0 .0>00 0:0. 0:0 003 0:0. .0 0.0002 :. 0003 00.:0..0> bo..0:0.0xw

196

 

f0...

man

9.:

mt:

fa...

00.09800
00.<

NOON

.0 .0 00200

 

mc+

mCI

0.4+

w:-

32030
.0_0

0.00.. .- v
30:00 0.0
0:0 0:39..
0.025

 

90:

wt:

in?

wc+

0.03::<

 

mc+

1+

0.02:0.00

 

5+

£+

wc:

mp...-

0.3.000

 

Ix,

w:-

.1:

8.0 0.00..
.- A >..o__0.
0000:0000
0:0
E0200.
$030

0000.
39.820
0:0 :00

 

we...

.3...-

0.02:0.00
5.;
0.03::<

.080.

._N am 00x_m>>

 

.000...
.58.

 

.090

 

0......

.30..
09:.

 

5.220
u

 

:020>>
a.

 

:05.
0.

 

000
20:00

 

 

0.0.000 .0:0_.3._.0:_

 

0.0.000 02%.3200-2000

 

0.00..0>
200:0000

 

8020

0003000
2090

 

 

.200. 2.000 :0N02< 0;. :. 00:0:0 .0>00 0:0. 0:0 003 0:0. .0 0.0002 :_ 0003 00.0000> E0.0:0_0xm

197

001.

003.
005.

006.

007.

008.

009.
010.
011.
012.
013.
014.

015.
016.

017.

APPENDIX B

DALR - Household Surveys

Questionnaire number: 002. Date:

 

Interviewer: 004. Reviewer:

 

GPS type:

 

Position:
5.1
5.2

 

Settlement name:

 

Settlement was officially created?

No Yes

 

7.1 When:

 

Municipality:

 

Gleba:
Lot number:
Secondary road number (Travesséo):

 

 

 

Distance (Transamazon/City):

 

Name of the interviewee:

 

Position of interviewee:
1. Owner
2. Spouse
3. Son/Daughter
4. Other (specify 09., sharecropper):

 

Name of the owner?

 

Where does the owner live?

 

 

 

 

1. Lives in the lot: Yes No
2. Lives in the city: State:
3. Lives in another lot: Municipality: State:

4. How many lots does the owner have:

 

House where the family is living

198

Before arrival After arrival

Wood walls
Brick walls
Mud walls
Well

AAAA Z
vvvv O
AAA/N _<
vvvv (D
AAAA Z
vvvv O
AAAA .<

VVVV (D

Questions about history of the lot

018. Year acquired?
019. How long have you lived in this settlement?
020. What is the size of this lot? Hectares Alqueires
021. Does the owner have any type of land document?

Yes No

 

 

022. If Yes, what is the type of the document?
( )Definitive title (Year received: )
( )Provisional title (Year received: )
( ) Contract of sale
( ) Other (Specify: )
( )No document
023. If No, when do you expect to receive documentation?
( )6 months to 1 year
( ) 1 year to 2 years
( )2 years to 5 years
( ) More than 5 years

024. How was the physical condition of the lot at arrival?

1.__ (%) Primary forest
2. (%) Some open areas but nothing cultivated
3._ (%) Some areas in production

3.1___ annual crops (ha)

3.2 perennial crops (ha)

3.3 pasture (ha)
3.4 secondary forest (ha)

 

 

4. (%) Other characterization (specify: )
99. _ NA
025. Do you have water in the lot?
Yes No NA
026. If Yes, how do you use the water?
0. irrigation of crops
1. cattle and other animals
2. to drink

199

£1

 

 

 

3 do not use why?

 

9'9. NA
027. Were you owner since the beginning?
0. No 1. Yes

 

028. If Not, 1. How many owners did the lot have before you?
2. What year was this lot bought for the first time?

Landownership history (urban-rural link)

029. Does the owner of this lot have rural or urban properties in the State of

 

 

 

Para? 0. No 1. Yes 99. NA
030. If Yes, what is the type of the property?
1. residential property
2. business
3. Other (specify : )
031. Does the owner of this lot have other lots or rural properties in the State?
0. No 1. Yes 99. NA
032. If Yes, how many rural properties do you have?
(# of properties)

033. What is the area of all rural properties that you have (please, include this
lot)?
1. Hectares 2. Alqueires 99. NA
034. How many properties did the owner have that were sold?
Year Size of the property (Hectares):

 

 

Socio Economic Background:

035. How many families and persons are living in this lot today?
1. families 2. persons 3. children

036. Beginning with yourself and then your spouse, please provide the age, sex,
and number of years of education of each member of your household, along with
their relationship to you:

Age Sex Relationship Years of school

 

 

 

 

 

 

 

 

 

 

 

 

.U‘PP’N.“

 

 

 

200

0373. How many days of labor did you pay for in the past year?

037b. How many days labors did you exchange last year?

 

038. How many days of labor did members of your household work for others in
the past season?

039. Where were you born?
a. State

 

b. Municipality 3. City
040. When were you born?

a. Year

 

b. Month

 

041. When you arrived here, did you bring any equipment?

Had upon arrival Today

No Yes No Yes
a. Chainsaw ( ) ( ) ( ) ( )
b. Generator ( ) ( ) ( ) ( )
0. Cocoa dryer ( ) ( ) ( ) ( )
d. Tractor ( ) ( ) ( ) ( )
e. Truck ( ) ( ) ( ) ( )

042. When you arrived here, which kind of durable good did you have?

Had upon arrival Today

No Yes No Yes
a. Stove ( ) ( ) ( ) ( )
b. Sewing machine ( ) ( ) ( ) ( )
c. Refrigerator ( ) ( ) ( ) ( )
d. Television ( ) ( ) ( ) ( )
e. Parabolic antenna ( ) ( ) ( ) ( )

201

Bicycle ( )
Car ( )
Radio ( )
House in the city ( )

2" :0 . "
o o
AAAA
AAAA
vvvv
vvvv

Migration Questions

043. Please give the year and month (or your age) when you moved to another
place of residence, along with the state, municipality, and type of residence
(rural/urban):

Year-month Municipality Rural/urban

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F.-
m
v-O-
(D

999°35’39”?wa

 

 

Social/political capital Questions

044. Do you (or the owner) participate in a rural syndicate in the municipality?

0. No 1. Yes 99. NA

045. Do you (or the owner) participate in any association in the municipality?
0. No 1. Yes 99. NA

046. Do you (or the owner) participate in any cooperative in the municipality?
0. No 1. Yes 99. NA

047. Are you (or the owner) a member of a church in the municipality?
0. No 1. Yes 99. NA

202

048. Please list all of the organizations that you are involved in (Le. rural workers
union — STR).

Name Type Date Office held

 

 

 

 

 

 

 

 

9959.”?

 

 

DALR Experience:

049. Have you been involved in DALR activities?
0. No 1. Yes 2. Which type 99.
(If No, go to question 050)

DALR When Where Duration Results A
event r State/

 

* Type of event: 1=protest; 2=Encampment; 3 Settlement

A. Who organized the occupation?
(1) Nobody

(2) City hall

(3) Rural union
(4) INCRA

(5) MST

(6) CPT

(7) Other: specify

B. How did you learn about this occupation?

(1) Through parents/relatives

(2) Friends

(3) SMO (MST, CPT, STR, etc.) Which one:

(4) Government agency (INCRA, City hall) Which one:
(5) Commentary in the city (sounds car and advertisements)
(6) Other: specify

 

C. How did you choose this area to occupy? Which factors were important in
the choice?

(1) Lot is close to parents/relatives house

(2) Roads

(3) Lot w/ valuable trees

203

(4) Close to the city
(5) Other: specify

 

050. Has, or is, anyone in your family been involved in DALR activities?
0. No 1. Yes 99. NA

(If No, go to question 051)

 

DALR When Where Duration Results A
event Month/year State/city

 

 

 

 

 

 

 

 

 

 

 

 

* Type of event: 1=protest; 2=Encampment; 3 Settlement

A. Who organized the occupation?
(1) Nobody

(2) City hall

(3) Rural union
(4) INCRA

(5) MST

(6) CPT

(7) Other. specify

History of the settlement

051. How did you learn about this occupation?
(1) Through parents/relatives
(2) Friend
(3) SMO (MST, CPT, STR, etc.) Which one:
(4) Government (INCRA, City hall) Which one:
(5) Comments in the city
(6) Other: specify

 

 

052. How did you choose this area to occupy? Which factors were important in
the choice?

(1 )Lot is close to parents/relatives house

(2)Roads

(3)Lot w/ valuable trees

(4)Close to the city

(5)Other: specify

 

053. What was this area before occupation?

204

054.

055.

056.

057.

058.

059.

060.

061.

062.

063.

1. Fazenda( ) 2. Terra devoluta( ) 3. Other( )specify:

Do you have roads in the lot?

0._No 1. Yes
If Yes, who built the road?

(1) The owner

(2) City hall

(3)Logger

(4) INCRA

(5) Other: specify
When you arrived here, the settlement had:

1._Trail 2. Road 3. Nothing
Who built the trail/road?

(1) The owner

(2) City hall

(3)Logger

(4) INCRA

(5) Other: specify
Did this lot have a landowner before your arrival?

0. No 1. Yes 2. Who
If Yes, how did you get this lot?

(1) You just occupied the lot

(2) You made a deal with the landowner

(3) Other: specify
Who helped you to create the settlement?

(a) Nobody

(b) City hall

(c) STR

(d) INCRA

(e) MST

(f) CPT

(9) Other: specify

 

 

 

 

 

Explain the process:
What did you do to create the settlement?

(a) Occupied the area

(b) Strike

(c) Encampment in front of INCRA

(d) Encampment in the road

(e) Other: specify
Were you a member of Frente de Massa?

O. No 1. Yes

Explain:
Did you join any association before or after lot demarcation?

(. ) Before ( ) After ( ) NA

Why did you join the association? Explain:

 

 

 

 

Who created the association? Explain

 

205

064.

065.
066.

067.

068.

069.

070.

071.

072.

073.

074.

075.

Do you think that was difficult to demarcate this lot?
0. No 1. Yes
Explain:

 

 

 

 

 

 

 

 

How were the lots divided? Who divided them? Explain:
Did you have any conflict with others in demarcating the lot?
0. No 1. Yes
Who? ~
Do you follow any environmental or land use rule?
0. No 1. Yes
Explain:
If yes, which rules do you follow? Do you agree with these rules?
0. No 1. Yes
Explain:
The settlement creation helped you to obtain credit for:
(1) House
(2) Roads
(3) Energy
(4) School
(5) Other

(6) It did not help
Do you want your children to have land here?
0. No 1. Yes
If not, why not?
Do you have any relative or parent living the old colonization area?
0. No 1. Yes
How many?
if yes, who?
( ) Father/Mother Where?
( )Brother/Sister Where?
( ) Uncle/Aunt Where?
( ) Other: specify

 

 

 

 

 

 

 

 

Do you have relative or parent living here in the secondary road?
0. No 1. Yes
How many?

If yes, who?

( ) Father/Mother Where?
( )Brother/Sister Where?
( ) Uncle/Aunt Where?
( ) Other: specify

What do you think about land reform in Brazil? Explain?

 

 

 

Agriculture experience

076.

What was your parent’s occupation?
0. Farmer 1. Business man 2. Other

206

Urban Rural Specify:

 

077. Years respondent practiced agriculture:
078. Agricultural (or ranching) education/training?
0. No 1. Yes 99. NA
What type?
How many months/years?
079. Where do you get your information about planting/harvesting?
( ) Parents/relatives
( ) Friends
( )SMO (CPT, MST, OTC, STR)
( )INCRA, City Hall
( ) School
( ) Other: specify
( ) Nobody
080. Did anybody living here receive credit to buy equipment, commercialize, or
soil conservation practices?

 

 

 

 

 

 

 

 

 

 

 

0. No 1. Yes 99. NA
081. If yes, please, describe the objective of the credit?

1.

2.
082. Do you receive technical assistance?

0. No 1. Yes
083. If yes, from who?

1. 2.

3. 4.
084. if yes, did you receive incentives to use?

No Yes

Curvas de Nivel ( ) ( )

Soil cover ( ) ( )

Terracing ( ) ( )

Crop rotation ( ) ( )

Mulch ( ) ( )

Other ( ) ( ) Which one?
085. Using these practices did you observe any improvement in production?

0. No 1. Yes 99. NA
086. If yes, where did the improvement occur?

( )com ( ) beans ( ) rice ( ) pasture

( )other: specify
087. Do you have any interest in reforestation?

 

 

 

 

 

0. No 1. Yes 99. NA
088. If yes, which kind of tree would you use?

1. 2. 3.
089. Do you raise cattle?

0. No 1. Yes 99. NA

207

090. If yes, how many different pasture areas do you have in your lot? What is
the size of each area and condition?

 

Pasture Pasture Pasture Pasture Pasture Pasture
1 2 3 4 5 6

 

# of cattle
heads
Area
(ha)
Condition
(8)
Mineral
salt/vaccin
e
(Y/N)
Rent to
other -
2005
(YIN)
(a)Code: AL: active clean pasture; As: Dirty pasture; B: abandoned pasture;

C: Pasture in reform

 

 

 

 

 

 

 

 

 

 

 

 

 

091. In case you rented any pasture area did you receive income from it?
0. No 1. Yes How much (R$ /head/month)
99. NA

092. Agriculture - land use during 2005
Annual crops Lot 1 Lot 2 Lot 3 Lot 4
Corn
Hectares
Production
Fertilize (f)
Pesticide (p)
Rice
Hectares
Production
Fertilize (f)
Pesticide (p)
Bean
Hectares
Production
Fertilize (f)
Pesticide (p)

 

 

 

 

 

 

 

 

 

 

 

208

 

Manioc
Hectares
Production
Fertilize (f)
Pesticide (p)

 

Vegetables
Hectares
Production
Fertilize (f)
Pesticide (p)

 

Water melon
Hectares
Production
Fertilize (f)
Pesticide (p)

 

Sugar-cane
Hectares
Production
Fertilize (f)
Pesticide (p)

 

Other
Hectares
Production
Fertilize (f)
Pesticide (p)

 

 

 

 

 

 

 

093. Agriculture — perennials land use during 2005

 

Perennial crops Lot 1 Lot 2 Lot 3 Lot 4

 

Coffee

Planted Hectares
Year planted

Total harvest (unit)
Sale+exchanged (unit)
Price (unit)

 

Cocoa

Planted Hectares
Year planted

Total harvest (unit)
Sale+exchanged (unit)
Price (unit)

 

Banana

Planted Hectares
Year planted

Total harvest (unit)

 

 

 

 

 

 

 

209

 

Sale+exchanged (unit)
Price (unit)

 

Citrus

Planted Hectares
Year planted

Total harvest (unit)
Sale+exchanged (unit)
Price (unit)

 

Rubber tree

Planted Hectares
Year planted

Total harvest (unit)
Sale+exchanged (unit)
Price (unit)

 

Coflon

Planted Hectares
Year planted

Total harvest (unit)
Sale+exchanged (unit)
Price (unit)

 

Cashew

Planted Hectares
Year planted

Total harvest (unit)
Sale+exchanged (unit)
Price (unit)

 

Guarana

Planted Hectares
Year planted

Total harvest (unit)
Sale+exchanged (unit)
Price (unit)

 

Black pepper
Planted Hectares

Year planted

Total harvest (unit)
Sale+exchanged (unit)
Price (unit)

 

 

Other

Planted Hectares
Year planted

Total harvest (unit)
Sale+exchanged (unit)
Price (unit)

 

 

 

210

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

094. How many years do you leave the land in fallow?
095. Do you sell your produce (or production) in the market?
0. No 1. Yes
096. If yes, how do you transport the product to the market?
( ) Own car
( ) Bus
( ) Kombi van
( ) Carro de Iinha
( ) Other
( ) NA
097. If not, why? Would you like to sell your product in the market in the future?
098. Do you have problem with loss of soil fertility or erosion?
0. No 1. Yes 99.
099. Why or why not?
100. Do you use fire to clear or prepare your land?
0. No 1. Yes 99.
With what frequency do use fire?(how many times a year?)
When? ( Month?)
101. Have you had problems with fire in your property in the last 10 years?
0. No 1. _Yes 99. NA
Year # of fires Burned area Crops/infrastructure Fire/causes
(ha) (a) (b)
1997
1998
1999
2000
2001
2002
2003
2004
2005
2005
Code:
(a) CA: annual crops; CP: perennial crops; PA: pasture; CP: secondary forest
CE: fence; PR: buildings; VE: vehicle; OU: other (specify: )
(b) VZ: fire from neighbor; FP: own fire; OU: other (specify: )
Forest

102. When you arrived in the lot:

a. how much deforestation did the lot have?

211

 

b. Who opened the area?

Open area (after arrival) Year Who opened

 

 

 

 

 

 

 

 

 

PPN.‘

 

 

 

Plan for the future (next 2 years)

103. If you had to decide your plans for the next two years, which changes would
you like to make in your lot?

104. What would you do with your annual crops?

 

 

 

 

 

0. keep the same area
1. increase the area
2. decrease the area
9. NA
105. What would you do with your perennial crops?
0. keep the same area
1. increase the area
2. decrease the area
99. NA
106. What would you do with your pasture area?
0. keep the same area
1. increase the area
2. decrease the area
99. NA

 

107. In case you want to increase the area in annual or perennial crops, where
would you increase the area?

 

0. secondary forest
1. forest area
99. NA

108. What is your plan for your cattle?
0. keep the herd size
1. increase the herd size
2. decrease the herd size
99. NA

 

109. In the next two years, what would you like to do?
stay in the same lot

move to another lot in the Para State
move to the city

other:

1.
2.
3.
4.

110. In the next two years, what would you like to do?

212

1. sell the lot
2. expand agriculture production
3.
4. leave the lot as inheritance

keep the production the same

213

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Simon madmwdw m fimmm nmémvdw 3.0m w¢.m_.m.mN 3.:on o0 OE (n.
Nwéomé omdoméw N90; mmfinmdw N_..mor afiwwodw amen...”— <m
v _..oo¢ 0 flow v.3 madam 91.39%? m flow mmdvivv who...— og< (n.
ENMFN 94.3%.: Nwdws mm.oww.N_. NNdNN mfiwmmdv _aw =33... (n—
Nnénoém Q: fiwmr nmdmmér mm. F 5K: owdwwdv mmdum. Fm? E_n:..:m (n—
wwfimmfi wosomdw hmdomd K. F 5.9. mv. 5m Ft. Fm @5925 (a.
vodm omhmm. Pm and co. 5m. 5 mad onéwm. 5 3.52 35h (n.
mmdmm coda _..m_. 9.. EN Svafir NMNN undoofiw ambit. <n.
NNKCMN v wodm ”.mwfi mo. Sum; v mmhomdw mud FNA v 3.. FNVMVNJ 930n— ou 0.1 (n—
o; a: a: a... a: a:

3coEo to»

.5520 $20... 3320 .33“. 3.620 $20... «soocflcram

89. poor camp

 

 

 

 

 

EoEoEom 3 $5822 5:98.960

0 X_Dzmn_n_<

214

cozmoho
ho 2mm

 

mEoEoEow 3 $523 2.:on ho LonEzc 6E0: 8.8%ch

cos—No.5
we San

 

£58933 3 $22 E 3:28 .6 LonEzc ”oEos. 383:3

mEoEoEom
msoocuucoam

mace—=23:
wsoocuuconm

215

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8 .o- 0 to 8.0 mm... 8.0 8.0 002 0:02 .33... <0
8... 8.0 8.0 v: 8.0 8.2 82 800.0 <0
8...- 8... 8... 8.0. 8.... :0 002 08.80 00 20 <0
8.0 2.0 8.0 0.2 8.0 3.2 B2 0.00 22 <0
8.0 8.2 8.2 00.0 8.2 3.0. 52 .30 .33... <0
0.0- 8.0 8.2 «0.00. 82 0 tom 32 0500.5 <0
8.0- 80 8.0 5.0 8.0 N: 0.02 000.0; <0
8.... 8.0 8.2 0.0.0 8.2 5.0 82 3.00 on 0.0 <0
:- 88 8.2 2.0 8.2 8.0 82 53:50 <0
00.3 0200.0 0200.0 5.8 0200.0 028.0
028.04 82 32 028.04. 32 82 ”um“. nuannfiufim
A05 05m :00... :00—2
000080.000 .3 050 00.00 0008 00000.0 ”00.0.0.2 00000000...
00.. «8.0 08.0 F- 08.0 «8.0 82 232 .33. <0
80. 20 0 8 N8- 20 23 82 800.0 <0
Sm- R... 08 8- 50 0E 82 8.000 on 0.0 <0
80.0- 8%... 08... 00.0.0- 08.0 m 8.2 B2 0.00 0.2 <0
~8- 02 08 3 F- 08 .2 B2 .30 .33 <0
08. 8 En mam- Em 80.0 52 0500.... <0
08..”- m 5 v8.0 %- v8.0 0 5.0 32 80.0... <0
:R- B 08 20 - 08 3.: 82 8.00 on 0.0 <0
2 2- 8. wow 8- mom 8.. 82 53:50 <0
8.3 0330 .830 3.8 .830 .830
.8304 82 32 .8304 32 82 “wfiumo mucuufiuwfim
05 005 :00... :00—2

 

 

 

 

 

0.0060200 .3 050 00.00 000:. .0000. ”0050.). 00000000..

216

APPENDIX D

 

 

Uirapuru Forest Change, 1986 - 1999

 

25.000

1
Y

20,000

 

1

1 5,000
I Forest

I Cleared

 

 

 

Hectares

1

10.000

1

5,000

 

 

 

1986 1 991 1 999
Year

 

 

 

PA Uirapuru - Land Cover Change, 1986 - 1999

217

 

Trairao Forest Change, 1986 - 1999

 

 

 

 

I Forest
I Cleared

 

 

 

Hectares

 

 

 

 

1 991
Year

 

 

 

 

 

PA Trairao - Land Cover Change, 1986 - 1999

218

 

 

Rio de Peixe Forest Change, 1986 - 1999

 

30,000
25,000 "
20,000 0

15,000 J

Hectares

10.000 -

 

 

I Forest
I Cleared

 

 

5,000 —» ,

 

 

 

1991 1999

Year

 

 

 

 

 

PA Rio do Peixe - Land Cover Change, 1986 - 1999

219

 

 

 

Surubim Forest Change, 1986 - 1999

9

160.000 14:91——

  

138 a‘

    

140,000
120-000 ,
100,000

80,000 0

I Forest
I Cleared

Hectares

60,000 -
40,000
20,000 -

 

 

0 -

1986 1 991 1 999
Year

 

 

 

PA Surubim - Land Cover Change, 1986 - 1999

220

 

Tutui Sul Forest Change, 1986 - 1999

 

16,000

 

14,000

10,000 0
8,000 -

Hectares

6.000 ~
4.000 —
2.000 _

12,000 *7

.—

 

 

   
  
  
  
  

 

 

I Forest
I Cleared

 

 

 

 

1 986 1 991 1 999
Year

 

PA Tutui Sul - Land Cover Change, 1986 - 1999

221

 

 

Alto Para Forest Change, 1986 - 1999

 

 

 

I Forest
I Cleared

 

Hectares
N
S"
O
O
O

 

 

_x _I. N
P 5" P
O O O
O O O
O O O
l 1 l
T T I

 

 

 

 

 

Year

 

 

 

PA Alto Para - Land Cover Change, 1986 - 1999

222

 

Placas Forest Change, 1986 - 1999

 

 

 

Hectares

 

 

1 986 1 991
Year

 

 

 

 

I Forest
I Cleared

 

 

 

PA Placas - Land Cover Change, 1986 — 1999

223

 

 

 

Rio de Pedras Forest Change, 1986 - 1999

 

30,000

 

 

25,000 ~

20,000 -

 

I Forest
I Cleared

15,000 *

Hectares

10,000 —l'

5,000 -

 

 

 

 

 

 

Year

 

 

 

PA Rio de Pedras - Land Cover Change, 1986 - 1999

224

 

Tutui Norte Deforested Area, 1986 - 1999

l A“
1 60 955"

 

Hectares

 

 

 

1 986 1991 1 999
Year

 

 

PA Tutui Norte — Land Cover Change: Cleared Area, 1986 - 1999

 

Tutui Norte Forested Area, 1986 - 1999

e
31 990 . ‘6” “9°
, .5-

1,09"

 

31,980
31.970 '-
31.960 3

31,950 ,

Hectares

31.940 '
31,930
31 .920

 

 

 

31,910 ,
1986 1991 1999

Year

 

 

 

PA Tutui Norte — Land Cover Change: Forest Area, 1986 — 1999

226

 

 

 

 

 

APPENDIX E

Distance to major urban center from assentamento centroid.

 

Assentamento Distance to City (km) City Name

17 de Abril 97.80 Maraba
1st de Marco 20.10 Maraba
Alegria 17.80 Maraba
Araras 26.30 Maraba
Cabanos 97.80 Maraba
Canudos 97.80 Maraba
Palmares ll 98.00 Maraba
Santa Maria do Pontal 64.10 Maraba

Surubim 29.10 Medicilandia
Rio do Peixe 18.60 Uruara
Trairéio 52.40 Uruara
Tutui Norte 49.70 Uruara
Uirapuru 28.30 Uruara

 

227

 

BIBLIOGRAPHY

Agnew, J. A. 1987. Place and politics. Boston: Allen and Unwin

Aldrich, S. P., R. T.Walker, E. Arima, M. Caldas, J. Browder, and S. Perz. 2006.
Land cover and land use change: Processes of social stratification in
agricultural expansion. Economic Geography 82 (3): 265—88

Allen, J. C., and D. F. Barnes. 1985. The causes of deforestation in developing
countries. Annals of the Association of American Geographers 752163-84

Almeida, A. L. O. 1992. The colonization of the Amazon. Austin: University of
Texas Press

Almeida, A. L. 0., and J. S. Campari. 1995. Sustainable settlement in the
Brazilian Amazon. Oxford University Press, New York

Alston, L.J., G.D. Libecap, and B. Mueller. 2005. How interest groups with limited
resources can influence political outcomes: information control and the
Landless Peasant Movement in Brazil. Institute of Behavior Science.
Working paper EB2005-0005

Alston, L. J., G.D, Libecap, and B. Mueller. 2000.Land Reform Policies, the
Sources of Violent Conflict, and Implications for Deforestation in the
Brazilian Amazon. Journal of Environmental Economics and Management
39(2): 162-188

Alston, L.J., G. D. libecap, and R. Schneider. 1993. The settlement process,
property right, and land use on the Brazilian Amazon Frontier: Preliminary
results and lessons from US economic history. Paper presented at the All-
University of California Economic History Conference, April, Pasadena,
CA

Alvarenga, O. M1985. Manual de Direito Agrario. Rio de Janeiro, R.J.: Forense

Alvarez, S.E., E. Dagnino, and A. Escobar. 1998. Cultures of Politics/Politics of
Cultures: Re-Visioning Latin American Social Movements. Boulder, CO:
Westview

Amsberg, J.V. 1998. Economic Parameters of Deforestation. The World Bank
Economic Review 12 (1 ): 133-53

228

Andersen, L. 1996. The Causes of Deforestation in the Brazilian Amazon.
Journal of Environment and Development 5(3): 309- 328

Andersen, L., C. W. J. Granger, E. J. Reis, D. Weinhold, and S. Wunder. 2002.
The Dynamics of Deforestation and Economic Growth in the Brazilian
Amazon. London: Cambridge University Press

Angelsen, A., and D. Kaimowitz. 1999. Rethinking the Causes of Deforestation:
Lessons from Economic Models. World Bank Research Observer 14 (1 ):
73-98

Arima, E., P. Barreto, and M. Brito. 2005. Pecuréria na Amazénia: Tendencies e
implicacé’es para a conservacéo. Belem: Imazon

Arima, E.Y., R.T. Walker, S.G. Perz, M.M. Caldas. 2005. Loggers and forest
fragmentation: Behavioral models of road building in the Amazon basin.
Annals of the Association of American Geographers 95: 525-541

Assies, W. 1994. Urban Social Movements in Brazil: A Debate and Its Dynamics.
Latin American Perspectives Vol. 21 (2): 81 -1 05

Bachrach, P. and M. Baratz. 1973. Power and Poverty: Theory and Practice.
New York: Oxford University Press

Barbier, E. B. 2001. The economics of tropical deforestation and land use: An
introduction to the special issue. Land Economics 77 (2):155-171

Barbier, EB, and J.C. Burgess.1996. Economic Analysis of Deforestation in
Mexico. Environment and Development Economics 1(2): 203-240

Barros, R. P., C.H. Corseuil, S. Cury, and P. G. Leite. 2001. Aben‘ura econémica
e distribuicé'o de renda no Brasil. Proceedings of the Workshop on Trade
Liberalization and the Labor Market in Brazil, Brasilia, UnB/IPEA

Bawa, KS, and S. Dayanandan. 1998. Causes of tr0pical deforestation and
institutional constraints to conservation. In Tropical Rain Forests: a wider
perspective. Eds. G.T. France, N. Brown, B. Sharpe, M.J. Eden, S.M.
Ross, Kamaljit S. Bawa, S. Bass, A. Grainger, C. Clubbe, A. Juniper, and
EB. Goldsmith, 175-198. London, UK: Chapman & Hall

Bawa, KS, and S. Dayanandan.1997. Socioeconomic factors and tropical
deforestation. Nature. Vol. 386: 562-63

Becker, B. K.. M.H.P. Miranda, L.O. Machado1990. Fronteira Amazonica :
Questées sobre a Gesta'o do Territorio. Brasilia: UNB/UFRJ

229

Bernard, 8., and R. de Koninck. 1996. The retreat of the forest in Southeast Asia.
A cartographic assessment. Singapore Journal of Tropical Geography 17
(1 ): 1 -14

Bilsborrow, R.1994. Population, development and deforestation. Some recent
evidence. In Population, Environment, and Development. Eds.
Department of Economic and Social Information and Policy Analysis, 117-
134. New York: United Nations

Binswanger, H. P., and McIntire, J. 1987. Behavioral and material determinants
of production relations in land-abundant tropical agriculture. Economic
Development and Cultural Change 36:73—99

Bockstael, N. E. 1996. Modeling economics and ecology: The importance of a
spatial perspective. American Journal of Agricultural Economics 78:1 168-
1 180

Bonaudo, T., S. Doris, B. Mertnes, and J. F. Tourrand. 2003. Indicateurs de
pression en vironnementale selon un degré d'anthropisation croissante.
CIRAD (unpublished draft), Montpellier, France

Boucek, B., and E. F. Moran. 2004. Inferring the Behavior of Households from
Remote Sensed Changes in Land Cover: current methods and future
directions. In: Spatially Integrated Social Science, ed. M. F. Goodchild and
DG. Janelle. New York: Oxford University Press

Brandao Jr., A., and C. Souza Jr. 2006. Desmatamento nos Assentamentos de
Reforma Agraria na Amazénia. O Estado da Amazdnia, n° 7. Belém:
Imazon

Brandéo, H.V., L.A.R. Corréa, M.P. Ribeiro, P.P.M. Araujo. 1998. Diagnéstico da
Agricultura Familiar 6 Plano de Desen volvimento do Projeto de
Assentamento Rio do Peixe. Projeto LUMIAR. Uruara, Para

Branford, 8., and J. Rocha. 2002. Cutting the wire: The story of the landless
movement in Brazil. London: Latin America Bureau

Browder, J. 1994. Surviving in Rondonia: The Dynamics of Colonist Farming
Strategies in Brazil's Northwest Frontier. Studies in Comparative
International Development 29 (3): 45-69

Browder, J. 1988. Public policy and deforestation in the Brazilian Amazon. In

Public polices and the misuse of forest resources. ed. R. Repetto and M.
Gillis, 247-298. Cambridge: Cambridge University Press

230

Brown, K., and D.W. Pearce. 1994. The Causes of Tropical Deforestation. In
The economic and statistical analysis of factors giving rise to the loss of
the tropical forests. ed. K Brown and D. W. Pearce, London: University
College London Press

Bunker, S. 1985 Understanding the Amazon: Extraction, Unequal Exchange, and
the Failure of the Modern State. University of Illinois Press, Urbana

Burns, T., E. L. Kick, and D. A. Murray. 1994. Demography, development and
deforestation in a world-system perspective. International Journal of
Comparative Sociology 35 (3/4): 221 -39

Caldas, M., R. Walker, E. Arima, S. Perz, C. Wood, S. Aldrich, and C. Simmons.
2007. Theorizing Land Cover and Land Use Change: The Peasant
Economy of Amazonian Deforestation. Annals of the Association of
American Geographers Vol 97 (1 ): 86-110

Calderbn, F., A. Piscitelli, and J.L. Reyna. 1992. Social Movements: Actors,
Theories, Expectations. In The Making of Social Movements in Latin
America: Identity, Strategy, and Democracy, ed. A. Escobar and SE.
Alvarez, 19-36. Boulder: Westview Press

Campanhole, A. 1969. Legislacéo Agraria. Séo Paulo: Atlas

Campos, A. S. 2002. O confronto em Eldorado de Carajés. Belém: Promev
Grafica e Editora

Capistrano, A. D., and CF. Kiker. 1995. Macroscale economic influences on
tropical forest depletion. Ecological Economics 14 (1 ): 21 -9

Cardoso, R. C. 1983. Movimentos sociais urbanos: balance critico. In Sociedade
e politica no Brasil po’s 1964. eds. Bernardo Sorj and Maria H. T. de
Almeida, 215-239, 850 Paulo: Brasiliense

CAT (Centro Agro-Ambiental do Tocantins). 1992. Elementos de analise do
funcionamento dos estabelecimentos familares do regiéo de Maraba.
Unpublished manuscript. CAT, Maraba

Chander, G. and B. Markham 2003. Revised Landsat-5 TM radiometric
calibration procedures and postcalibration dynamic ranges. Geoscience
and Remote Sensing, IEEE Transactions on 41(11): 2674-2677

Chase, J. 1999. Regional Studies of Labour Relations, Regulation and
Globalization Exodus Revisited: The Politics and Experience of Rural Loss
in Central Brazil. Sociologia Ruralis 39 (2):165-185

231

Chayanov, A. V. 1925. Peasant farm organization. In A. V. Chayanov on the
theory of the peasant economy, ed. D. Thomer, B. Kerblay, and R. E. F.
Smith, 33—269. Homewood, IL: Richard D. Irwin

Chomitz, K.M., and T. S. Thomas. 2003. Determinants of Land Use in Amazénia:
A Fine-Scale Spatial Analysis. American Journal of Agricultural Economist
85 (4): 1016-028

Chomitz, K. M., and D. A. Gray. 1996. Roads, land use, and deforestation: A
spatial model applied to Belize. The World Bank Economic Review 10 (3):
487-512

Cline,W.R . Economic consequences of a land reform in Brazil. Amsterdam:
North-Holland, 1970.

Cohen , D and R. Portes. 2003. Dealing with destabilizing ‘market discipline’.
Center for Economic Policy Research. Discussion Paper Series.
International Macroeconomics No. 4280.London, UK

Cohen, J.L. 1985. Strategy or Identity: New Theoretical Paradigms and
Contemporary Social Movements. Social Research 52: 663-716

Coimbra, C. 1981. A Revolucéo de 30 no Para: analise, critica e interpretaga'o da
histéria. Belém: Conselho Estadual de Cultura.

Coomes, O. T. 2004. Rain forest ‘conservation-through-use’? Chambira palm
fibre extraction and handicraft production in a land-constrained
community, Peruvian Amazon. Biodiversity and Conservation 13: 351-360

Coomes, O. T., F. Grimard, and G. J. Burt. 2000. Tropical forests and shifting
cultivation: Secondary forest fallow dynamics among traditional farmers of
the Peruvian Amazon. Ecological Economics 32:109-124

Costa Porto, J. da. 1978. Sistema sesmarial no Brasil. Brasilia,D.F.:
Universidade de Brasilia

Cox, K. R.1997. Spaces of globalization: Reasserting the power of the local. New
York: Guilford

Davies, J.C. 1969. Toward a Theory of Revolution. In Studies in Social
Movements, ed. B. McLaughlin, 85-109. New York: The Free Press

Davis, DE. 1999. The power of distance: retheorizing social movements in Latin
America. Theory Soc. 28(4):585—638

232

Deacon, RT. 1994. Deforestation and the rule of law in a cross-section of
countries. Land Economics 70:414-430

DeShazo, R. P., and J. R. DeShazo. 1995. An economic model of smallholder
deforestation: A consideration of the shadow value of land on the frontier.
In Management of tropical forests: Towards an integrated perspective, ed.
0. Sandbukt, 153-66. Oslo: Center for Development and the Environment,
University of Oslo

Domer, P. and B. Saliba. 1981. Interventions in Land Market to Benefit the Rural
Poor. University of Wisconsin. Land Tenure Center, Research Paper 74

Eckstein, S. 2001. Epilogue. Where Have All the Movements Gone? Latin
American Social Movements at the New Millenium. In Power and Popular
Protest. Latin American Social Movements, ed. Susan Eckstein. Berkeley:
University of California Press

Ehrhardt—Martinez, K. 1998. Social Determinants of Deforestation in Developing
Countries. A Cross-National Study. Social Forces 77(2): 567-86

Emmi, M. F. 1999. A oligarquia do Tocantins e o dominio dos castanhais. Belem:
UFPAINAEA

Entwisle, B., and P. C. Stern. 2005. Population, land Use, and Environment:
Research Directions. National Research Council. National Academy
Press, Washington, DC.

Escobar, A. 1992. Culture, Economics, and Politics in Latin American Social
Movements Theory and research. In The Making of Social Movements in
Latin America: Identity, Strategy, and Democracy, ed. A. Escobar and SE.
Alvarez, 62-85. Boulder: Westview Press

Escobar, A., and SE. Alvarez. 1992. Introduction: Theory and Protests in Latin
America Today. In The Making of Social Movements in Latin America:
Identity, Strategy, and Democracy, ed. A. Escobar and SE. Alvarez, 1-15.
Boulder: Westview Press

European Space Agency (ESA).1994.Modeling deforestation process: A review,
Trees Series B, Research Report # 1. Luxembourg: Office for Official
Publications of the European Community

Evers, T. 1985. Identity: The hidden side of new social movements in Latin

America. In New Social Movements and the state in Latin America, ed.
David Slater. [Amsterdam] CEDLA : Cinnaminson, N.J., USA

233

Eyerrnan, R., and A. Jamison. 1991. Social Movements: A Cognitive Approach.
University Park: Penn State University Press

Faminow, M. D., C. Dahl, S. Vosti, J. Witcover, S. Oliveira, and C. Carpentier.
1999. Smallholder risk, cattle and deforestation in the western Brazilian
Amazon. World Animal Review 93: 1 6-23

FAO (2007). Food and Agriculture Organization. FAO Statistics Division. Last
accessed: 12/18l2007. Available from:

http://wwwfao.orgjes/esslcensuslginiltable1 .asp

Feamside, P. M. 2001. Land-tenure issues as factors in environmental
destmction in Brazilian Amazonia: the case of Southern Para. World
Development 29 (8): 1361 -1 372

Feder, G. and T. Onchen. 1985. Land Ownership Security and Capital
Foundation in Rural Thailand. Washington, DC: World Bank Research
Unit

Femandes, B. M. 2001. The Occupation as a form of Access to Land. Paper
presented at the meeting of the Latin American Studies Association,
Washington, DC

Femandes, B. M. 2000. A Formaca’o do MST No Brasil. Petrépolis: Vozes

Femandes, B. M. 1999. Donos de terras: Trajetorias da Uniao Democratica
Ruralista—UDR Belem, Para, Brazil: Universidade Federal do
Para/Nucleo de Altos Estudos Amazénicos

Ferreira, LA. 2001. Le role de l’élevage bovin dans la viabilité agro-ecologique et
socio-économique des sytemes de production agricoles familiaux en
Amazonie brésilienne - le cas d’Uruara (Para, Brésil), INA-PG, Paris

Fonnan, R.T.T., and M. Gordon. 1986. Landscape Ecology. John Wiley & Sons,
New York, NY, USA

Fox, J., R. R. Rindfuss, S. J. Walsh, and V. Mishra, 2003. People and the
Environment: Approaches for Linking Household and Community Surveys
to Remote Sensing and GIS. London, UK: Kluwer Academic Publishers

Foweraker, J. 1995. Theorizing Social Movements. London and Boulder: Pluto
Press

Foweraker, J. 1981. The struggle for land: A political economy of the pioneer

frontier in Brazil, 1930 to present. Cambridge, UK: Cambridge University
Press

234

Furtado, C. 1959. Formacéio econémica do Brasil. RJ: Fundo de Cultura
Gamson, W. A. 1975. The Strategy of Social Protest. Homewood, Ill.: Dorsey

Gardner, R. H., R. V. O'Neill, and M. G. Turner. 1993. Ecological implications of
landscape fragmentation. In Humans as components of ecosystems:
subtle human effects and ecology of population areas, eds. S. T. A.
Pickett and M. G. McDonnell, 208-226. Springer-Venag, New York

Geist, H. J., and E. F. Lambin. 2002. Proximate Causes and Underlying Driving
Forces of Tropical Deforestation. BioScience 52 (2): 43-150

Geist, H. J., and E. F. Lambin. 2001. What drives tropical deforestation? LUCC
Report Series No. 4. Louvain-Ia-Neuve: CIACO

Geoghegan, J., L. Pritchard Jr., Y. Ogneva-Himmelberger, R. Roy Chowdhury, S.
Sanderson, B. L.Tumer II, 1998. Socializing the pixel and pixelizing the
social in land-use and land-cover change. In People and Pixels: Linking
Remote Sensing and Social Science. Eds. D. Liverman, E. Moran, R.
Rindfuss, P. Stern, 51-69. Committee on the Human Dimensions of Global
Environmental Change. National Research Council. National Academy of
Science Press, Washington, DC

Godoy, R., S. Groff and K. O’Neill. 1998a. The Role of Education in Neotropical
Deforestation: Household Evidence from Amerindians in Honduras.
Human Ecology 26 (4): 649- 675

Godoy, R., M. Jacobson, J. De Castro, V. Aliaga, J. Romero, and A. Davis.
1998b. The Role of Tenure Security and Private Time Preference in
Neotropical Deforestation. Land Economics 74 (2): 162-70

Godoy, R. M. Jacobson, and D. Wilkie. 1998c. Strategies of Rain-Forest Dwellers
Against Misfortune: The Tsimane’ Indians of Bolivia. Ethnology 37 (1 ): 55-
69

Godoy, R., D. Wilkie. and J. Franks. 1997a. The effects of markets on neotropical
deforestation: A comparative study of four Amerindians societies. Current
Anthropology 38:875-878.

Godoy, R., K. O’neill, S. Groff, P. Kostishack, A. Cubas, J. Demmer, K.
Mcsweeney, J. Overman, D. Wilkie. N. Brokaw, and M. Martinez. 1997b.
Household determinants of deforestation by Amerindians in Honduras.
World Development 25 (6):977-987

235

Griffith, D. A. 2005. Effective geographic sample size in the presence of spatial
autocorrelation. Annals of the Association of American Geographers
95(4): 740-60

Gusfield, JR. 1970. Protest, Reform, and Revolt. New York: John Wiley and
Sons

Gutman, G., A. C. Janetos, C. 0. Justice, E. F. Moran, J. F. Mustard, R. R.
Rindfuss, D. Skole, B. L. Turner II, and M. A. Cochrane. 2004. Land
Change Science: Observing, Monitoring, and Understanding Trajectories
of Change on the Earth’s Surface. Boston: Kluwer Academic Publishers

Hamilton, R. 2001. Mass society, pluralism, and bureaucracy: explication,
assessment, and commentary. Publisher Westport, Conn.: Praeger

Harrison, S. 1991. Population growth, land use, and deforestation in Costa Rica,
1950-1984. lnterciencia 1 6: 83-93

Hecht, SB. 1985. Environment, development, and politics: Capital accumulation
and the livestock sector in Eastern Amazonia. World Development 13:
663-684

Hellman, J.A. 1992. The study of new social movements in Latin America and the
question of autonomy. In: The Making of Social Movements in Latin
America: Identity, Strategy, and Democracy, ed. A. Escobar and SE.
Alvarez, 52-61. Boulder: Westview Press

Hoffman, R. 1998. A estrutura fundiaria no Brasil de acordo com o cadastro do
INCRA: 1967 a 1998. Working paper. INCRA/UNICAMP

Homma, A., R.T. Walker, R. Carvalho, A. Conto, and C. Ferreira. 1996. Razbes
de risco e rentabilidade na destruibao de recursos florestais: 0 case de
castanhais em lotes de colonos no Sul do Para. Revista Econémica do
Nordeste 27 (3): 515—35

Homer-Dixon, T. 1999. Environment, scarcity, and violence. Princeton,N.J.:
Princeton University Press

Howe, G, and D. Goodman.1992 Smallholders and Structural Change in the
Brazilian Economy: Opportunities in Rural Poverty Alleviation. San Jose:
IFADIIICA

IBGE. 2007a. lnsituto Brasileiro de Geografia e Estatistica. Série Histdricas do

Brasil. Last accessed:12/02l07. Available from:
www.ibge.gov.br/home/estatisticas/popuIacaoldefaulttab historicas.sht_rn_

236

IBGE. 2007b. lnsituto Brasileiro de Geografia e Estatistica. Pesquisa Pecuaria
Municipal. Last accessed:12/02l07. Available from:

http://www.sidra.ibge.gov.br/bda/pecualdefaultasg

IBGE. 2007c. lnsituto Brasileiro de Geografia e Estatistica. Contagem da
Populaca'o. Last accessed:12l02l07. Available from:

http://www.ibge.gov.br/homelestatisticalpopulacaolcontagem2007/default.
shtm

IBGE. 2007d. lnsituto Brasileiro de Geografia e Estatistica. Projetos Criados.
Last accessed: 12/1 7/2007. Available from: h_ttp:/lwww.incragov.br/

IDESP. 1990. Instituto de Desenvolvimento Econémico-Social do Para. Uruara.
Belém: IDESP.

INCRA. 2007a. Instituto Nacional de Colonizacae Reforma Agréria. Diretoria de
Obtencao de Terras e Implantacao de Projetos de Assentamento,
Relatorio 227. Brasilia, D.F.

INCRA. 2007b. Instituto Nacional de Colonizacao e Reforma Agraria. lndice de
Gini - O Brasil Desconcentrando Terras, Brasilia, D.F. Last
accessed:11/10/O7. Available from: [mm/wwwjncragovbrl

INCRA. 2005. Instituto Nacional de Colonizacao e Reforma Agraria. Relarério da
Ouvidaria Agréria 09/2005, Brasilia, D.F. Last accessed:10/21/07.
Available from: h_ttp:/lwww.incra_gmv§_rl

INCRA. 2003. Instituto Brasileiro de Colonizacao e Reforma Agraria. Relatério de
Gestéo. Last accessed: 09l07l07. Available from: http://www.incrigpvbrl

INCRA. 1996. Instituto Nacional de Colonizacao e Reforma Agréria. Mapa
Fundiério do Brasil, Brasilia, D.F. Last accessed:11/10/07. Available from:
mpzllwwwjncragovbr/

INCRA. 1995. Instituto Nacional de Colonizacéio e Reforma Agraria. Relatdrio de
Atividades INCRA - 1985/ 1994. Brasilia, D.F. Last accessed: 10/11/07.
Available from: http://www.incragovbr/

Irwin, E.G., and J.Geoghegan. 2001. Theory, data, methods: developing spatially
explicit economic models of land use change. Agriculture, Ecosystems
and Environment 1764:1-18

Jenkins, J.C. 1983. Resource Mobilization Theory and the Study of Social
Movements. Annual Review of Sociology 9: 527-53

237

Jenkins, J. C. and, C. Perrow. 1977. Insurgency of the Powerless: Farm Worker
Movements, 1946-1972. American Sociological Review 422249-67

Jones, D. W., V. H. Dale, J. J. Beauchamp, M. A. Pedlowski, and RV. O'Neill.
1995. Farming in Rondonia. Resource and Energy Economics 17:155-188

Kaimowitz, D. and A. Angelsen.'1998. Economic, models of tropical
deforestation: A review. Bogor, Indonesia: Center for International
Forestry Research

Kant, S., and A. Redantz. 1997. An econometric model of tropical deforestation.
Journal of Forest Economics 3 (1 ): 51 -86

Kimsey, M. 1991. A spatial analysis of the causes of tropical deforestation.
University of Georgia, Department of Geography (Ph.D. Thesis)

Komhauser, W. 1959. The Politics of Mass Society. Glencoe, IL: The Free Press

Kotscho, R. 1981. O Massacre Dos Posseiros: Conflicto de Terras no Araguaia-
Tocantins. Brasiliense, Séo Paulo

Krutilla, K., W.F. Hyde, and D. Bames.1995. Periurban deforestation in
developing countries. Forest Ecology and Management 74: 181-95

Lamarao, P. 1977. Legislacé’o de Terras do Estado do Para, 1890-1977. Vol 2.
Belészrafisa.

Lambin, E.F. 1997. Modelling and monitoring land-cover change processes in
tropical regions. Progress in Physical Geography 21 (3): 375-93

Lambin, E.F. 1994. Modelling deforestation processes: A review. TREES Series
B: Research Report 1. Luxembourg: Office of Official Publications of the
European Community

Lambin, E.F., H. J. Geist, and E. Lepers. 2003. Dynamics of Land-Use and Land-
Cover Change in Tropical Regions. Annual Review of Environment and
Resources 28:205-241

Lambim, E.F., B.L.Tumer ll, H.J.Geist, S.B.Agbola, A.Angelsen, J.W.Bruce,
O.Coomes, R. Dirzo, G. Fisher, C. Folke, P.S. George, K. Home-Wodd, J.
Imberson, R. Leemans, X. Li, E.F.Moran, M. Mortimore, P.S.
Ramakrishnan, J.F. Richards, H. Skanes, W. Steffen, D.G.Stone, U.
Svedin, T.A. Veldkamp, C. Vogel, and J.Xu. 2001. The causes of land use
and land cover change: Moving beyond myths. Global Environment
Change: Human and Policy Dimensions. Vol. 11 (4): 261—269

238

Lang, K. and Lang, G. E. 1961. Collective Dynamics. New York: Thomas Y.
Crowell Co

Laurance, W.F., M.A. Cochrane, S. Bergen, P.M. Feamside, P. Delambnica, C.
Barber, 8. D’Angelo, and T. Femandes. 2001. The future of the Brazilian
Amazon. Science 291, 438-439

Laurance, W. F. 1998. Forest fragmentation may threaten genetic diversity.
Bioscience 48 (9): 784

Lenski, G.F. 1954. Status crystallization: A non-vertical dimension of social
status. American Sociological Review 19:405-413

Lima, R. C. 2002. Pequena historia territorial do Brasil: sesmarias e terras
devolutas. Goiana: Editora Universidade Federal de Goias

Liverman, D., E. F. Moran, R. R. Rindfuss, and P. C. Stern. 1998. People and
Pixel: linking remote sensing and social science. Washington, D. C.:
National Academy of Press.

Mainwaring, S. 1989. Grassroots Popular Movements and the Struggle for
Democracy: Nova lguacu. In Democratizing Brazil, ed. A. Stepan, 168-
204. New York: Oxford University Press

Mahar, D. J. 1989. Government polices and deforestation in Brazil's Amazon
region. Washington: World Bank

Mahar, DJ 1979. Frontier development policy in Brazil: A study of Amazonia.
New York: Praeger Publishers

Margulis, S. 2003. Causes of deforestation in the Brazilian Amazon, World Bank
Working Paper Series. Washington, DC: World Bank

Martine, G. 1993. The Phases of Agricultural Modernization in Brazil. In
Population-Environment Dynamics. eds. G. D. Ness, W. D. Drake and S.
R. Brechin, 167-186. Ann Arbor: University of Michigan Press

Martins, MD. 2006. Learning to Participate: The MST Experience in Brazil. In
Promised land: Competing visions of agrarian reform. eds. P. Rosset, R.
Patel, and M. Courville, 265-276. IFDP, NY: New York

Massey, D. 1984. Spatial Divisions of Labor: Social Structures and the Geography
of Production. New York: Methuen.

Massey, D., and J. Allen. 1984. Geography Matters. Cambridge: Cambridge
University Press

239

McAdam, D. 1 982. Political Process and the Development of Black Insurgency,
1930-1970. Chicago: University of Chicago Press

McAdam, D., S. Tarrow, and C. Tilly. 2001. Dynamics of Contention, Cambridge:
Cambridge University Press

McAdam, D., and D. A. Snow. 1997. Social Movements: Readings on Their
Emergence, Mobilization, and Dynamics. Los Angeles, California: Roxbury

McCarthy, JD, and MN. Zald. 1977. Resource Mobilization and Social
Movements: A Partial Theory. American Journal of Sociology 82: 1212-
1 241

McCarthy, JD, and MN. Zald. 1973. The Trend of Social Movements in
America: Professionalization and Resource Mobilization. Morristown, NJ:
General Learning Press

McCracken, S. D., E. Brondizio, D. Nelson, E. Moran, A. Siqueira, and C.
Rodriguez-Pedraaz, 1999. Remote sensing and GIS at farm property
level: Demography and deforestation in the Brazilian Amazon.
Photogrammetric Engineering and Remote Sensing 65:1311-1320

McGarigal, K., and B.J. Marks. 1995. FRAGSTA TS. Spatial Analysis Program for
Quantifying Landscape Structure. USDA Forest Service General
Technical Report

Medeiros, L.S. de. 2002. Movimentos Sociais, disputas policas e reforrna agréria
de marcado no Brasil. Rio de JaneirozEditora da Universidade Rural do
Rio de Janeiro

Melluci, A. 1980. The new social movements: a theoretical approach. Social
Science Information, Vol. 19, 2

Mendonca, N. D. 1976. A propriedade rural no processo de urbanizacéo na zona
da campanha rio-grandense. In A propriedade rural, Vol. III, Anais do VIII
Simpésio Nacional de Professores de Histbria, Séo Paulo, Brasil

Merton, Robert K1968. Social Theory and Social Structure. New York: Free
Press

Merton, Robert K. 1938. Social Structure and Anomie. American Sociological
Review 3: 672-82

240

Meyer, OS, and Staggenborg, S. 1996. Movements, Counten'novements, and
the Structure of Political Opportunity. American Journal of Sociology Vol
101(6): 1628-60

Miller, B. A. 1994. Political empowerment, local-central state relations, and
geographically shifting political opportunity structures: Strategies of the
Cambridge, Massachusetts, Peace Movement. Political Geography 13 (5):
393—406

Monteiro, B. 1980. Direito Agrario e Processo Fundiério. Rio de Janeiro, R.J.:
PLG Comunicacéo.

Moran, E. F. 1981. Developing the Amazon. Bloomington, IN: Indiana University
Press.

Moran, E., and E. Brondizio, 1998. Land-Use Change After Deforestation in
Amazonia. In People and Pixels: Linking Remote Sensing and Social
Science, eds. D. Liverrnann, E.F. Moran, R.R. Rindfuss and PC.
Stem,94-120. National Academy Press, Washington DC

Moran, E., E. Brondizio, P. Mausel, and Y. Wu. 1994. Integrating Amazonian
vegetation, land use, and satellite data. BioScience 44(5):329-338

Morrison, D. E. 1978. Some Notes toward Theory on Relative Deprivation, Social
Movements, and Social Change. In Collective Behavior and Social
Movements, ed. Louis E. Genevie. Itasca, Ill.: Peacock

MST. 2006. Movimento dos Trabalhadores Rurais Sem-Terra. Biblioteca de
Dados. Last accessed: 09l02l07. Available from:
http://www.mst.org.br/mstl

Muller, C., and B. Mueller. 2006. The evolution of agriculture and land reform in
Brazil, 1960 — 2006. Paper prepared for the conference in honor of
Werner Baer. University of Illinois, Dec. 1-2

Mueller, B., L. Alston, G. D. Libecap, R. Schneider. 1994. Land, property rights
and privatization in Brazil. The Quarterly review of Economics and
Finance 34:261-280

Murali, KS, and R. Hedge. 1997. Patterns of tropical deforestation. Journal of
Tropical Forest Science 9 (4): 465-76

Murphy, L., R. Billsborrow, and F. Pichén. 1997. Poverty and Prosperity among

Migrant Settlers in the Amazon Rainforest Frontier of Ecuador. The
Journal of Development Studies 34(2): 35-66

241

Nascimento E. P. and J. A. Drummond. 2003. Amazénia: Dinamismo Econdmico
e Conservacéo Ambiental. Brasil, RJ.: Garamond

Nelson, G. C., V. Harris, and SW. Stone. 2001. Deforestation, Land Use, and
Property Rights: Empirical Evidence from Darién, Panama. Land
Economics 77 (2):187-205

Nelson, G. C., and D. Hellerstein. 1997. Do roads cause deforestation? Using
satellite images in econometric analysis of land use. American Journal of
Agricultural Economics 79:80—88

Oberschall, A. 1973. Social Conflict and Social Movements. Englewood Cliffs,
NJ: Prentice Hall

O’Neil, R.V., J.R. Krummel, R.H. Gardner,G. Sugihara, B. Jackson,D.L.
DeAngelis, B.T. Milne, M.G.Tumer, B. Zygmunt, S. Christensen, V.H.
Dale, and R.L. Graham. 1988. Indices of landscape pattern. Landscape
Ecology 2:63-69

Oxhom, PD. 1995. Organizing Civil Society: The Popular Sectors and the
Struggle for Democracy in Chile. University Park: Pennsylvania State
University Press

Painter, M., and W.H. Durham. 1995. The Social Causes of Environmental
Destruction in Latin America. Ann Arbor: The University of Michigan Press

Palo, M. 1994. Population and deforestation. In: The causes of tropical
deforestation, the economic and statistical analysis of factors giving rise to
the loss of tropical forest. eds. K. Brown and D. Pearce, 42-56. University
Collge London Press, London

Pan, W., and R. E. Bilsborrow. 2005. The use of a multilevel statistical model to
analyze factors influencing land use: A study of the Ecuadorian Amazon.
Global and Planetary Change 47: 232-52

Pan, W., S. Walsh, R.E. Bilsborrow, B. Frizzelle, C. Erlien, and F. Baquero. 2004.
Farm-level models of spatial patterns of land use and land cover dynamics
in the Ecuadorian Amazon. Agriculture, Ecosystems and Environment,

101 (2-3): 117-34

Pastore,J., Dias, G.L.S., and Castro, M. 1976. Condicionantes da produtividade
da pesquisa agricola no Brazil. Estudos Econémicos vol. 6(3): 147-181

Perz, S. 2003. Social Determinants and Land Use Correlates of Agricultural

Technology Adoption in a Forest Frontier: A Case Study in the Brazilian
Amazon. Human Ecology 31 (1 ):1 33-165

242

Perz,S. 2001. Household demographic factors as life cycle determinants of land
use in the Amazon. Population Research and Policy Review 20 (3): 159-
86

Perz, S.G., M.M. Caldas, E.Y. Arima, and RT Walker. 2007. Socio-spatial
processes of unofficial road-building in the Amazon: socioeconomic and
biophysical explanations. Development and Change 38, 529-551

Perz, S., and R. T. Walker. 2002. Household Life Cycles and Secondary Forest
Cover among Smallholders in the Amazon. World Development 30(6):
1009-27

Petras, J. 1998. The new revolutionary peasantry: The growth of peasant-led
opposition to neoliberalism. 2 Magazine, October

http://www.mstbrazil.org/Qetras1098.html.

Pfaff, A. S. P. 1999. What drives deforestation in the Brazilian Amazon? Journal
of Environmental Economics and Management 37:26-43

Pichon, F. J. 1997. Colonist land allocation decisions, land uses and
deforestation in the Ecuadorian Amazon frontier. Economic Development
Cultural and Change 35 (3): 707- 44

Prewitt, K. and Stone, A. 1973. The Ruling Elites: Elite Theory, Power and
American Democracy. New York: Harper and Row

Rangel, L.A, J.Andrade, and J.A. Divino. 2007. Crescimento Econbmico e
Desigualdade de Renda no Brasil de 1991 a 2000 - Uma Analise das
Areas Minimas Comparaveis. Textos para Discusséo. Rio de
JaneirozlPEA

Rau, V. 1982. Sesmarias medievais portuguesas. Lisboa: Editoral Presence

Reid, J.W. and LA. Bowles. 1997. Reducing the impacts of roads on tropical
forests. Environment 39, 10-1 7

Reis, E.. and R. Guzman. 1994. An economic model of Amazon deforestation. In
The causes of tropical deforestation: The economic and statistical analysis
of factors giving rise to the loss of tropical forests, ed. K. Brown and D.
Pearce, 172-192. London: University College of London Press

Rezende, 6.0. 2006. Pobreza e Desigualdade no Brasil: 0 papel Adverso das

Politicas Trabalhistas, Fundiaria e de Credito Agricola. Working paper, Rio
de Janeiro, R.J.: IPEA

243

 

Richards, J.F., and RP. Tucker. 1988. World Deforestation in the Twentieth
Century. Durham: Duke University Press.

Riitters, K.H., R.V. O’Neil, C.T. Hunsaker, J.D. Wickham, D.H. Yankee, S.P.
Timmons, K.B. Jones, and BL. Jackson. 1995. A factor analysis of
ladscape pattern and structure metrics. Landscape Ecology 10:23-40

Rindfuss, RR, and RC. Stern. 1998. Linking Remote Sensing and Social
Science: The need and the Challenges. In People and Pixels: Linking
Remote Sensing and Social Science, eds. D. Liverman, E.F. Moran, R.R.
Rindfuss, and PC. Stem, 1-27. Washington, DC: National Academy
Press

Rios, J.A. 1950. A posse e o posseiro no Brasil. Digesto Econémico, numero 11,
setembro

Roumani, A., and Coirolo, L. 2005. Brazil: Innovation increases land access and
incomes of poor rural families. World Bank. En Breve: series of notes,
May, No. 70

Rudel, T. K. 1989. Population, development, and tropical deforestation: A cross-
national study. Rural Sociology 542327-38

Rudel, T., and J. Roper. 1997. The Paths to Rain Forest Destruction: cross-
national patterns of tropical deforestation, 1975-90. World Development.
25(1): 53-65

Rudel, T., and J. Roper. 1996. Regional Patterns and Historical Trends in
Tropical Deforestation, 1976-1990: a qualitative comparative analysis.
Ambio 25(3): 160-66

Rudel, T., and B. Horowitz. 1993. Tropical deforestation: Small farmers and land
clearance in the Ecuadorian Amazon. New York: Columbia University
Press

Sanchez, P.A.1994. Alternatives to slash and burn: a pragmatic approach for
mitigating tropical deforestation. In Agricultural Technology: Policy Issues
for the International Community, ed. J.R. Anderson, 451-479.CAB
lntemational/World Bank, Washington, DC

Santos, R. 1984. Law and Social Change: the problem of Land in the Brazilian

Amazon. In Frontier Expansion in Amazonia. eds. M. Schmink and C.
Wood, 439-462. Gainsville: University of Florida Press

244

Santos, R. 1979. A Fronteira Amazonica e a Reformulacéo da Politica de Terras.
Boletim da Associacao Brasileira de Reforma Agraria 9:l (Janeiro-
Fevereiro)

Sawyer, D. R. 1984. Frontier expansion and retraction in Brazil. In Frontier
Expansion in Amazonie, eds. M. Schmink and C. Wood, 180-203.
Gainsville: University of Florida Press

Schelhas, J., and R. Greenber. 1996. Forest patches in tropical landscapes.
Washington, DC: Island Press

Schmink, M., and C. Wood.1992. Contested frontiers in Amazonia. New York:
Columbia University Press

Schmink, M. and C. Wood. 1984. Frontier Expansion in Amazonia. Gainsville:
University of Florida Press

Scott, A. 1990. Ideology and the New Social Movements. London, Routledge

Serrao, E. A. S., and A. K. O. Homma. 1993. Country profiles: Brazil. In
Sustainable agriculture and the environment in the humid tropics, 265-
351. Washington, DC: National Academy Press

Serrao, E. A. S., D. Nepstad, and R. T. Walker. 1996. Upland agricultural and
forestry development in the Amazon: sustainability, criticality and
resilience. Ecological Economics 1 8: 3-1 3

Sewell, W.H. 2001. Space in contentious politics. In Silence and voice in the
study of contentious politics. Ed. R.R. Amizade, J.A. Goldstone, D.
McAdam, E.J. Perry, W.H. Sewell, S. Tarrow, and C. Tilly, 51-88.
Cambridge: Cambridge University press

Silva, J.G. da. And B. Kohl. 1984. Capitalist ”Modemization" and Employment in
Brazilian Agriculture, 1960-1975. Latin American Perspectives Vol. 11(1):
1 1 7-136

Silva. J. G. da. 1981.Modernizacao dolorosa: estrutura agraria, fronteira agricola
e trabalhadores rurais no Brasil. Rio de Janeiro: Zahar

Silva, LC. 1996. Terras Devolutas e Latifundio: Efeitos da Lei de 1850.
Campinas, SP: Editbra da Universidade Estadual de Campinas

Simmons, C. S. 2006. Collaborative Research: Brazil’s Direct Action Land

Reform Movement: Environmental Impacts and Socio-spatial Dynamics.
National Science Foundation Proposal, Washington, DC

245

 

Simmons, C. S. 2005. Territorializing land conflict: Space, place, and contentious
politics in the Brazilian Amazon. Geojournal 64(4):307—1 7

Simmons, CS. 2004. The Political Economy of Land Conflict in the Eastern
Brazilian Amazon. Annals of the Association of American Geographers 94
(1 ): 183 — 206

Simmons, C.S., Walker, R., Arima, E., Aldrich, S., and Caldas, M. 2007. The
Amazon Land War in the South of Para. Annals of the Association of
American Geographers Vol. 97(3): 567—592

Simmons, C.S., S. Perz, M. Pedlowski, and L. G. T. Silva. 2002. The Changing
Dynamics of Land Conflict in the Brazilian Amazon: the Rural-Urban
Complex and its Environmental Implications. Urban Ecosystems 6: 99-122

Skole, D., W. H. Chomentwoski, W. A. Salas, and A. D. Nobre. 1994. Physical
and Human Dimension of Deforestation in the Amazon. BioScience 44(5):
314-22

Smelser, N.J.1962. Theory of Collective Behavior. New York: The Free Press

Smith, Nigel. 1982. Rainforest corridors: The Transamazon colonization scheme.
Berkeley: University of California Press

Soares, P. R. R. 2000. A modemizacao agropecuaria na regiéo Sul do Rio
Grande do Sul. Coloquio de Geografia Rural de Espan'a, Lleida:
Universidade de Lleida

Southgate, D., R. Sierra, L. Brown. 1991. The Causes of Tropical Deforestation
in Ecuador: A Statistical Analysis. World Development 19 (9): 1145-151

Sponsel, L.E., T. N. Headland, and R. Bailey. 1996. Tropical Deforestation: the
human dimension. New York: Columbia University Press

Staeheli, L. A. 1994. Empowering political struggle: Spaces and scales of
resistance. Political Geography 13 (5): 387—91

Stam, O. 1992. I dreamed of foxes and hawks: reflections on peasant protest,
new social movements, and the rondas campesinas of northern Peru. In
The Making of Social Movements in Latin America: Identity, Strategy, and
Democracy, ed. A. Escobar and SE. Alvarez, 89-111. Boulder: Westview
Press

Stern, PC, 0. R. Young, and D. Druckman. 1992. Global environmental change:

- Understanding the human dimensions. Washington, DC: National
Academy Press

246

 

 

Sydenstricker Neto, J., and S. A. Vosti. 1993. Household size, sex composition,
and land use in tropical moist forests: Evidence from the Machadinho
Colonization Project, Rondénia. Brazil. Unpublished manuscript

Ta rrow, S. 1998. Power in Movement: social movements and contentious
politics. Cambridge University Press

Tilly, C.1978. From Mobilisation to Revolution. Reading, Ma.: Addison- Wesley
Publishing Company

Toni, F. 2003. Uruara: Pecuarizacao na Fronteira Agricola. In Municipios e
Gestéo Florestal na Amazénia, eds. Fabiano Toni and David Kaimowitz,
175-217. Natal, Brasil: AS Editores

Toni, F. 1999. State-society relations on the agricultural frontier: The struggle for
credit in the Transamazonia region. Ph.D. diss. Department of Political
Science, University of Florida, Gainesville

Treccani, G. D. 2001. Violéncia e grilagem: lnstrumentos de aquisica'o da
propriedade da terra no Para. Universidade Federal do Para: Belem

Tucker, RP, and J.F. Richards. 1983. Global Deforestation and the Nineteenth
Century World Economy. Durham: Duke University Press

Turner, B. L. II, and S. B. Brush. 1987. Comparative farming systems. New York:
Guilford

Van de Steeg, J. A., G. Sparovek, S. Ranieri, B. Lima. 2006. Environmental
impact of the Brazilian Agrarian Reform process from 1985 to 2001.
Scentia Agricola 63 (2): 176-83

Vasconcellos, J.M.P. de. 1860. Livro das terras. Rio de janeiro: Eduardo 8.
Henrique Lamemmert

Veiga, J.B., Tourrand, J.F., Quandz, D. 1996. A pecuaria na fi'onteira agricola da
Amazénia: caso do municlpio of Uruara, PA, na regiéo da Transamzénica
Bele’m, Documentos de pesquisa 87, EMBRAPA-CPATU

Walker, RT. 2004. Theorizing Land-Cover and Land-Use Change: The Case of
Tropical Deforestation. International Regional Science Review 27(3): 247-
70

Walker, RT. 2003. Mapping Process to Pattern in the Landscape Change of the

Amazonian Frontier. Annals of the Association of American Geographers
93(2): 376-98

247

Walker, RT. 1987. Land Use Transition and Deforestation in Developing
Countries. Geographical Analysis 19(1): 18-30

Walker, RT, and Solecki, B. 2004. Theorizing Land Cover and Land Use
Change: The Case of the Florida Everglades and its Degradation. Annals
of the Association of American Geographers 94(2): 311-28

Walker, R. T., S. Perz, M. Caldas, and L. G. T. Silva. 2002. Land use and land
cover change in forest frontier: the role of household life cycles.
International Regional Science Review 25(2): 169-99

Walker, R. T., E. F. Moran, and L. Anselin. 2000. Deforestation and cattle
ranching in the Brazilian Amazon: external capital and household process.
World Development 8 (4)2683-99

Walker, R. T., and Homma, A. 1996. Land use and land cover dynamics in the
Brazilian Amazon: An overview. Ecological Economics 18 (1 ): 67-80

Walsh, S. J., and Crews-Meyer, K. A. 2002. Remote Sensing and GIS
applications for linking people, place, and policy. Boston: Kluwer
Academic Publishers

Weinmann, J. 2007. The sustainability of electricity sector liberalization in Latin
America. Journal of Cleaner Production 15 (2): 1 12-1 14

Weinstein, BS. 1980. Prosperity without development: The Paraense Elite and
the Amazon Rubber Boom, 1850-1920. PhD. diss., Yale University

Welch, C. 2006. Keeping Communism Down on the Farm: The Brazilian Rural
Labor Movement during the Cold War. Latin American Perspectives. Vol
33(3):28-50

Williams, M. 1994. Forests and Tree Cover. In Changes in Land Use and Land
Cover: A Global Perspective. eds. W. B Meyer and BL. Turner II, 97-124.
Cambridge, UK: Press Syndicate of the University of Cambridge

Williams, M. 1990. Forests. In The earth as transformed by human action: global
and regional changes in the biosphere over the past 300 years. ed. B.L.
Turner ll, 179-202. Cambridge, U. K.: Cambridge University Press

Wilson, J. 1973. Introduction to social movements. New York: Basic Books

With, K. A. 1999. Is landscape connectivity necessary and sufficient for wildlife
management? In J. A. Rochelle, L. A. Lehmann, and J. Wisniewski,

248

editors. Forest fragmentation: wildlife and management implications, 97-
115. Brill, The Netherlands

Wynia, G. W. 1990. Brazil: Populists, Authoritarian, and Democrats. In The
Politics of Latin American Development. Ed. G.W. Wynia, 214-246,
Cambridge: Cambridge University Press

Wolford, W. 2005. Agrarian Moral Economies and Nee-liberalism in Brazil:
Competing world-Views and the State in the Struggle for Land.
Environment and Planning, A, 372241-61

Wolford, W. 2004. This Land ls Ours Now: Spatial Imaginaries and the Struggle
for Land in Brazil. Annals of the Association of American Geographers 94
(2): 409 - 24

Wolford, W. 2003a. Families, fields, and fighting for land: The spatial dynamics of
contention in rural Brazil. Mobilization: An International Journal 8 (2): 201—
1 5

Wolford, W. 2003b. Producing community: The MST and land reform
settlements in Brazil. Journal of Agrarian Change 3(4): 500—20

Wolford, W. 1996. Edible Ideology? Survival Strategies in Brazilian Land-Reform
Settlements. Geographical Review 86(3):457-461

Wood, C. H., and D. Skole. 1998. Linking Satellite, Census, and Survey Data to
Study Deforestation in the Brazilian Amazon. In People and Pixels: Linking
remote sensing and social science. Ed. D. Liverman, E. F. Moran, R.R.
Rindfuss, and P. C. Stem, 70-93. Washington, DC: National Academy
Press

Wunder, S. 2000. The Economics of Deforestation: the example of Ecuador.
London, UK: Macmillan Press Ltd

249

 

      

M G

llIll11111111111111l