State-society relations on the agricultural frontier

Material Information

State-society relations on the agricultural frontier the struggle for credit in the Transamazônica region
Toni, Fabiano ( Dissertant )
Sanderson, Steven ( Thesis advisor )
Wood, Charles ( Reviewer )
Williams, Philip ( Reviewer )
Brown, Leann ( Reviewer )
Schmink, Marianne ( Reviewer )
McCoy, Terry ( Reviewer )
Place of Publication:
State University System of Florida
Publication Date:
Copyright Date:


Subjects / Keywords:
Agricultural land ( jstor )
Agriculture ( jstor )
Alliances ( jstor )
Capitalism ( jstor )
Cattle ( jstor )
Farmers ( jstor )
Labor movements ( jstor )
Peasant class ( jstor )
Political movements ( jstor )
Social movements ( jstor )
Agricultural credit -- Amazon River Region ( lcsh )
Agricultural credit -- Government policy -- Brazil ( lcsh )
Agricultural laborers -- Amazon River Region ( lcsh )
Agriculture and state -- Amazon River Region ( lcsh )
Dissertations, Academic -- Political Science -- UF ( lcsh )
Political Science thesis, Ph. D ( lcsh )
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
non-fiction ( marcgt )


This is a study about peasant mobilization in the Brazilian Amazonia and the relationship between the peasant movement and the State. In the early 1970s the federal government promoted the colonization of a vast area in the state of Pará. A few years later the government withdrew its support to the settlers. In the late 1970s the peasantry started organizing in Catholic communities. This movement grew and the rural workers engaged in a successful struggle to democratize labor unions in the countryside. As the regime became more democratic, the organized peasants fought for more political participation. This struggle culminated in the creation of a unique credit policy designed to support poor farmers who do not have title to their lands. This credit was channeled through peasant associations and cooperatives, as a means to strengthen the labor movement. In this work I analyzed the factors that allowed the rural workers to mobilize, as well as to adapt their mobilization strategies to a changing political environment. Also I looked at the opportunities that the state provided for the peasantry to mobilize, and how it responded to their demands. To do that I used archival research, as well as qualitative interviews with key informants. I also surveyed 60 beneficiaries of this credit policy. Different factors were crucial to the success of the movement at different points in time. Catholic identity allowed peasants to start organizing. Later, strategic alliances with other sectors of civil society became more important. Constructing the image of the peasantry as environmentally responsible farmers was crucial to gain access to subsidized credit. The outcomes of this credit policy, however, challenged this identity. Credit recipients invested the money in cattle and cleared forest for pasture. Another unforeseen consequence of this policy was the appearance of hundreds of new peasant associations in the region. Rather than strengthening the peasant movement, these associations served the interests of local bosses, state officials, and large ranchers, who gained political and economic power using public money. This association between state officials and private actors challenges the concept of state autonomy in policy making. ( ,, )
KEYWORDS: social movements, Brazil, Amazonia, Amazon, rural credit, mobilization, protest, peasants, rural workers, rural unionism
Thesis (Ph. D.)--University of Florida, 1999.
Includes bibliographical references (p. 246-254).
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
General Note:
Title from first page of PDF file.
General Note:
Document formatted into pages; contains viii, 265 p.
General Note:
Statement of Responsibility:
by Fabiano Toni.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
45261414 ( OCLC )
002456720 ( AlephBibNum )
AMG2051 ( NOTIS )


This item has the following downloads:

Full Text









I am grateful for all the support that I received from the members of my

committee. I would like to thank Dr. Steven Sanderson for all his help and support. Besides a great researcher, teacher, and advisor, he was a true inspiration as a scholar. I would like to thank Dr. Charles Wood for all his enthusiasm for my research and for giving me the opportunity to go to the field with him in 1996, when he suggested that I do research on rural credit in the Transamaz6nica. I thank Dr. Philip Williams and Dr. Leann Brown for their thorough comments and critiques during all the writing of this dissertation. I am also grateful to Dr. Marianne Schmink and Dr. Terry McCoy for their comments on the final version.

Dr. Robert Walker provided valuable suggestions and insights during the

elaboration of the questionnaire and the interpretation of the data. Stephen Perz made a valuable contribution to this work, particularly by helping me with the statistical analysis. I am grateful to both of them.

I would like to thank the University of Florida Tropical Conservation and

Development Program, the Tinker Foundation, the Ford Foundation, and the National Air and Space Administration for the research grants that enabled me to do field work. Also, I would like to thank he Empresa Brasileira de Pesquisa Agropecuria (EM[BRAPA) for all the support they gave me while I was in Brazil conducting this research. Jonas Bastos da Veiga facilitated this institutional support and I am grateful for his help. In addition, I have


also benefited from travel grants from the University of Florida Student Government, the Political Science Department, and the Center for Latin American Studies. I want to thank Marty Swilley and Debbie Wallen for all their help in the Political Science department and Margarita Gandia for making everything so easy in the Center for Latin American Studies.

Many people provided crucial help during field work in Brazil. Special thanks go to Jean-Frangois Tourrand, who offered all assistance within his reach to make this work possible and my life comfortable in Belem and Uruara. In Belem, Rui Ludovino, and Paulo and Natalia Albuquerque offered me their hospitality and friendship. I am grateful to all of them.

In Uruardi, I was lucky to meet someone like Darcisio Quanz, who guided me

through the local politics and provided all the logistic help to make my survey possible. I want to thank him for all the help and friendship. Francois "Padre Chico" Glory helped me by sharing his perspective on the social and political life of Uruardi and Pardi. I want to thank him for those insightful comments and all the pleasant conversation. Also, I want to thank Neide and Paulo Remy Machado for the long hours of entertaining, chatting, and all the great food that they shared with me.

I would like to thank all the interviewees who participated in the surveys and the key informants who shared their views on the FNO-e. Several people provided additional data on the FNO-e and the history of the labor movement in Pardi. Special thanks go to Leticia Tura (FASE), Aloisio Solino (FASE), Clarisse Leonel (IDESP, FETAGRI) Alfredo Homma (EMBRAPA), Jose Maria Trindade (BASA), and Lecyr Peixoto (LAET).

Living in the United States would have been a less pleasant experience without

good friends. I want to thank Ed Greaves, Vilma Fuentes, Shawn Bird, and Mike Kenney


for all their support during these years as a graduate student at the University of Florida. Besides all the academic support, they guided me through the American way of life with a good deal of humor. A vast Brazilian Community made Gainesville home away from home. It was a pleasure and a privilege to work with such a nice group of people to create the Brazilian Student Association at UF (BRASA).

Despite all the difficulties that Brazil faced during the years that I spent in

Gainesville, the Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (CNPq) honored its commitment to fund my studies in the United States. I will be always thankful to CNPq and I hope to fulfill their expectations about my career as a researcher in Brazil.

Family members usually do not understand why someone would bother staying in school a few more years only to see his chances of getting a good job decrease. My brothers Joao Batista and Graciliano, and my mother, Dulce B. Toni, however, always gave me all the support they could during these years in graduate school. Besides being very supportive, my wife Daniela B. Lopes has also been a student during all our life as a couple. Her intellectual support, companion, friendship, love, and care turned my life as graduate student into an unforgettable experience. I am profoundly grateful to her.








Classical Theories of Social M ovements 11
Resource M obilization 17
Political Process 24 Id e n tity 2 9

Social M ovements in Latin America 32
New Trends in the Studies of Social Movements: Toward a Synthesis 36 C o n c lu sio n s 4 4

Pluralism, Elitism, and State Autonomy . 51
M arxist Approaches to the State 62
Corporatism and the State 76

The Authoritarian Regim e 91
Development and National Security: Colonizing Amazonia 99
Extending the State Structure 101



The First Phase: Operation Amazonia (1966-19 70) 104 The Second Military Phase: The National Integration Plan (19 70-74) 106 The Third Phase: POLAMAZONA (1974-1979) 115 The F ourth P hase (1980 -P resent) 120 C o n c lu sio n s 1 2 3

4 SOCIAL ORGANIZATION AND MOBILIZATION IN THE TRANSAMAZONICA 128 The F irst Cycle of M obilization 133 The Second Cycle: Social Movements in Action the Gritos da Terra 162
Conclusions: Movement Emergence and Transformation 190

5 THE OUTCOMES OF CREDIT POLICY 194 T h e A sso c ia tio n s 1 9 5 The Relationship Between Peasants and Their Associations 202 Environmental Consequences of the FNO-e 218 C o n c lu sio n s 2 3 1

C O N C L U S IO N S 2 3 5

R E F E R E N C E S 2 4 6






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


Fabiano Toni

May 1999

Chairman: Steven E. Sanderson
Maj or Department: Political Science

This is a study about peasant mobilization in the Brazilian Amazonia and the

relationship between the peasant movement and the State. In the early 1970s the federal government promoted the colonization of a vast area in the state of Pardi. A few years later the government withdrew its support to the settlers. In the late 1970s the peasantry started organizing in Catholic communities. This movement grew and the rural workers engaged in a successful struggle to democratize labor unions in the countryside. As the regime became more democratic, the organized peasants fought for more political participation. This struggle culminated in the creation of a unique credit policy designed to support poor farmers who do not have title to their lands. This credit was channeled through peasant associations and cooperatives, as a means to strengthen the labor movement.

In this work I analyzed the factors that allowed the rural workers to mobilize, as well as to adapt their mobilization strategies to a changing political environment. Also I


looked at the opportunities that the state provided for the peasantry to mobilize, and how it responded to their demands. To do that I used archival research, as well qualitative interviews with key informants. I also surveyed 60 beneficiaries of this credit policy.

Different factors were crucial to the success of the movement at different points in time. Catholic identity allowed peasants to start organizing. Later, strategic alliances with other sectors of civil society became more important. Constructing the image of the peasantry as environmentally responsible farmers was crucial to gain access to subsidized credit. The outcomes of this credit policy however, challenged this identity. Credit recipients invested the money in cattle and cleared forest for pasture. Another unforeseen consequence of this policy was the appearance of hundreds of new peasant associations in the region. Rather than strengthening the peasant movement, these associations served the interests of local bosses, state officials, and large ranchers, who gained political and economic power using public money. This association between state officials and private actors challenges the concept of state autonomy in policy making.



This study is about the implementation of a rural credit policy by the Brazilian federal government, designed to serve colonists in Amazonia who did not have title to their lands and could not provide collateral for loans. Unionized rural workers demanded this credit (FNO-e), with some help from other sectors of civil society. The struggle for access to credit required a tremendous effort of collective action, for the workers had to gather in local, regional, and national levels to discuss their needs and to press state officials to attend to their demands. In this study I try to answer 3 simple questions: 1Why did the workers mobilized when they mobilized? 2- What factors allowed them to succeed? 3- How has this credit policy affected the rural workers movements?

The creation of this credit policy was a significant accomplishment for the peasantry because it challenged a development model implemented by the military government between 1964-85. This model sought to promote capitalist accumulation on the agricultural frontiers with little regard for the livelihood of poor peasants. During the authoritarian period, and even after the transitions to democracy, the Brazilian government and international agencies, such as the World Bank financed large scale development programs in this region. Most of these programs were aimed at promoting ranching, mining, energy production, and timber extraction.

The implementation of this model required foreign loans that contributed to a

gigantic Brazilian debt. Moreover, its social and economic benefits were minimal, whereas



their negative ecological consequences were overwhelming. Subsidies to cattle ranching rank among the most detrimental development policies adopted by the Brazilian government. Cheap credit precipitated not only a violent struggle for land between settlers and big ranchers, but also spurred deforestation, due to the low carrying capacity of the land.

The organizations and institutions that the military government created to

implement this development model in Amazonia are still functioning and command large amounts of resources. In this context, the achievements of the rural workers are all the more spectacular. The collective effort of the rural workers meant the possibility of improving their lives by gaining access to public resources that traditionally has been beyond their reach. Moreover, the struggle for the creation of FNO-e also meant an unprecedented participation of the grassroots in policy making, which is a valuable contribution to the consolidation of democracy in Brazil. Questions one and two are directly related to this relative success of the movements. In this work I will try to explain how the workers beat the odds and successfully engaged in collective action. My hypothesis (hypothesis I) is that although several factors contributed to the success of the movement, one in particular stood out: the way they framed their struggle against the state. By opposing the development model implemented by the military and constructing the image of the peasant as a sustainable alternative to this model, the rural workers were able to attract societal support and powerful allies to their cause.

In spite of these positive aspects of the struggle for access to rural credit, the

introduction of FNO-e may have created serious problems for the rural movement in the long run, for two main reasons. First, the rural workers demanded that the FNO-e be


channeled through rural unions or peasant associations. In their view, this requirement would strengthen the labor movement. After a few rounds of negotiation, state officials agreed to channel the loan through associations, but not through unions, which would be illegal. As the rural workers could not control which associations would receive the loans, their expectations to increase grassroots participation did not come true. Concerning this problem, which is related to the third question asked above, my hypothesis (II) is that the introduction of FNO-e created new opportunities for clientelism and patronage through the peasants associations.

Second, the introduction of FNO-e did not have the positive economic and

ecological impact that the leaders of the movements anticipated. It is my hypothesis (III) that this credit line promoted a sharp increase in cattle herds, particularly in the Transamaz6nica region', in the state of Par/i, where a large number of loans was released. Due to the low carrying capacity of the lands in this region, credit recipients had to aggressively clear forest for pasture. If this was really the case, it mayh undermine the workers' framing of their struggle and, thus, their ability to attract certain types of allies and supporters.

The theoretical framework to answer the questions that underlie this study,

particularly questions one and two, comes from the field of social movements studies. Scholars in this field have pointed out the role of strains in triggering mobilization the "classical approach" (Korhauser, 1959, Davies; 1969; Gurr, 1970). Supporters of the "resource mobilization" approach argue that people mobilize when they accumulate

This is the central and West portions of Pardi State, South of the Amazonas river, crossed by the Transamaz6nica highway.


resources, or enter alliances with stronger actors who give them resources to mobilize (McCarthy and Zald, 1973; McAdam, 1982; Jenkins and Perrow, 1977). Other scholars stress the importance of political opportunities that emerge due to structural and conjunctural changes in the social and political environment (Tilly, 1978; Tarrow, 1994, 1998). Followers of a fourth model of mobilization emphasize the role of social movements in establishing new group identities and fighting for their rights (Melucci, 1989; Calderon et al., 1992; Escobar, 1992; Alvarez et al., 1998). In contrast to the other approaches, these authors argue that society, rather than the state is the main locus of political struggle for social movements.

The literature on the state also helps answer these two first questions, particularly the literature that addresses the issues of state autonomy. The state is an important component of the political opportunity and the resource mobilization approaches to social movements. Most movements either press their demands on the state (which is certainly the case of the rural workers in Amazonia) or reap the opportunities to mobilize that arise when the state faces trouble, or do both. Yet, most students of social movements do not pay enough attention to theoretical perspectives on the state. It is likely that some characteristics of any given state will affect the emergence of social movements. State autonomy is one important characteristic. The degree of autonomy from society that a state has and the way it uses this autonomy will very likely affect the emergence of social movements, as well as its organization and agenda. Whether a state represents the interests of a specific class or acts as a neutral arbitrator of competition between groups also affects social mobilization.


Using these two theoretical approaches, I analyzed the history of the rural workers movement in Pardi, which culminated in the creation of FNO-e. To reconstitute the main points of the history of this movement and to answer the first two questions (hypothesis I) I used a few bibliographical references available, and archival material kindly made available by the Federation of Rural Workers Unions of Pardi and Amapi FETAGRI. Most of the data, however, came from qualitative semi-structured interviews with key informants (a complete list of interviewees is presented in Appendix B).

To answer the third question (how the introduction of FNO-e has affected the workers movement) I used the qualitative interviews, data on agricultural output for the Municipio2 of Uruardi and the Transamaz6nica region', and two surveys. The first one is a survey of FNO beneficiaries, that I conducted in 1997. The other one is a survey on land use that was carried out in the summer of 1996, also in Uruardi. The qualitative interviews and the survey were particularly useful to address the effects of credit on social organization of the peasantry (hypothesis II). The IBGE data and the surveys were particularly useful in addressing the effects of credit on land use (hypothesis III).

The first survey was carried out in the town of Uruardi, where I spent two months and interviewed 60 colonists, using a structured questionnaire. All the interviewees were recipients of FNO-e and were randomly selected from a list of credit recipients provided by the association through which they get access to this credit. A copy of the questionnaire used in the survey is presented in Appendix C. The second one is a survey of

2 Municipio is the smallest administrative and political independent unity of the state.

3 These data were collected and processed by the Brazilian Institute for Geography and Statistics (Instituto Brasileiro de Geografia e Estatistica (IBGE).


241 farmers, conducted in the summer of 1996, also in Uruardi. Although the latter survey was not aimed at collecting data on the FNO-e, it contained valuable information on land use and credit in the municipio. BASA also supplied limited (due to legal constraints) information on FNO-e disbursements. The qualitative interviews and archival research were conducted between October 1997 and March 1998, in Uruardi, Altamira, and Belem.

My findings show that hypothesis I was correct. The way in which the rural workers framed their cause an alternative to the unsustainable development model implemented by the military was crucial for them to broaden their alliances. However, this explanation alone does not account for the whole history of social mobilization in rural Pardi. A fundamental factor that allowed peasants to mobilize in the first place was the assistance that they received from progressive sectors of the Catholic Church. Catholic priests provided the peasantry with fundamental human and material resources without which the movement probably would never have taken off.

Opportunities were also important for the success of the movement. The very

democratization of the country contributed to the success of the rural workers movement. By the time they mobilized, urban workers had already engaged in a struggle to break the shackles of the corporatist structure of labor representation. The example of urban workers and the support of the Catholic Church guided the rural workers in their very first task: to gain control of the labor unions that already existed in the region.

Two other opportunities made the movement stronger in terms of participation and cohesion. The first was a strike that broke out in an alcohol mill, in 1984, when the state police beat up protesters and members of the clergy. The second opportunity was the murder of a union leader, in 1991. After these two events, angry workers took to the


streets to protest, gained new allies, and dispatched representatives to the state and federal capital cities to pressure state officials.

I start this work with a discussion of the main approaches to social movements in the literature: classical approaches, resource mobilization, and identity-oriented approaches. I discuss the weaknesses and strengths of each of these approaches in chapter I. Rather than selecting any specific model, I try to combine those aspects of each one that can contribute most to the analysis of my case study.

Chapter 2 is a review and discussion of the main approaches to the state: Marxism, Pluralism/Neopluralism, and Corporatism. In this discussion I pay special attention to the issue of autonomy. In simple terms, I define autonomy as the capacity of the state to transform its preferences into policy, or conversely, to prevent societal groups to turn their preferences into policies. As said earlier, the Brazilian Authoritarian State established most of the state structure that social movements in Amazonia dealt with during a period of exceptionally high degree of autonomy.

With the onset of military rule in 1964, the state sought to promote the occupation of Amazonia. The state established a vast network of bureaucratic agencies in the region, brought legions of landless peasants to colonize the region, and usurped power from municipal and state governments. During this period the state shifted emphasis from public colonization to large scale private enterprises. This change meant the abandonment not only of the colonization plans, but also of the colonizers, or so they perceived latter. I discuss the history of the recent state-led occupation of Amazonia in Chapter 3. This chapter provides historical evidence to the theories of state discussed in Chapter 2.


Moreover, it describes the background against which the social movements in the rural areas of Amazonia emerged.

In Chapter 4, I examine the social movements that fought for the creation of the FNO-e. I begin with a historical account of the movement since its early days, when Catholic priests started organizing their communities. In the following section I stress the struggle to take over the corporatist structure of unionism and the organization, which was the main objective in the first cycle of mobilization. The following section is an analysis of the workers' annual protests in Belem that started in 1991, the Gritos. In the last section of this chapter, I use the theories of social movements presented in chapter two to explain the mobilization of the peasantry on the Transamaz6nica region.

Finally, in Chapter 5, I present and discuss the data on the outcomes of the FNO-e. Most of the quantitative data in this work is presented here. In the first part of the chapter I showed that the FNO-e helped create a large number of associations in Uruardi. Some of these associations were linked to the rural unions and other were created or controlled by opportunistic leaders. Using quantitative data form the surveys I discuss some differences in the relationship between leaders and rank and file in these two types of associations. In the following section I explore the technical aspects of this credit policy. In other words, I try to explain what the peasants did with the money they got from the bank. I also discuss the economic results and the impact that the activities financed by this credit line may have had on land use.

Before proceeding with the analysis of the case, it is necessary to explain the use of the terms peasant, rural worker, and colonist in this work, for I use all of them interchangeably, although they have different meanings. The majority of the small farmers


who live in the Transamaz6nica are colonists. The state brought most of them to this region and gave them land, which they cleared and cultivated. Even those who did not receive land from the government consider themselves colonists, at least in Uruardi. The social movement behind the creation of the FNO-e, however, represents diverse groups spread throughout Amazonia, and many of those groups are not colonists. As the movement is organized around the rural workers unions, and the leaders and active members of the movement identify themselves as rural workers, in many instances I used this term as well. Yet, some groups that were affected by this movement and by the creation of FNO-e are neither colonists nor rural workers, therefore I also used the term peasant.

The very definition of peasant is a contested domain in the social sciences and

engaging in this discussion would be beyond the scope of this dissertation. I simply adopt the term peasant as a synonym of a rural poor person. Any elaboration of this term would be based solely on my limited field experience and would not fairly represent the diversity of groups represented by the rural workers movement in the past 25 years.


THEORIES OF SOCIAL MOVEMENTS A vigorous wave of studies of social movements is now celebrating its third decade. This wave was ignited by spectacular instances of mobilization and collective action that swept the United States and Europe (and Latin America, to a lesser extent) in the late 1960s. This renewed academic interest was equally evident on both sides of the Atlantic. Besides the appeal of the social events unfolding before their eyes, social scientists also found academic motivation to address the phenomenon: the weakness of existing social movements theories.

Despite their common empirical roots, the schools that emerged in Europe and the United States were completely different, and, unfortunately, for more than a decade there was little cross fertilization between these schools. In the late 1980s, Latin American scholars engaged in social movement research. This was a delayed response, since substantial social mobilization on the continent started in the 1970s and early 1980s. Also, Latin American scholars were unable to capture the strengths and to weed out the weakness of each model. Rather they adopted some concepts of the European literature that were inappropriate for the local context. The recent literature in Latin America, Europe, and in the United States, however, points to an integration of these approaches, which apparently has reinvigorated the field.

In this chapter, I review the trajectory of social movements studies in the last thirty years. First, I review the so called "classical model" of social movements, which


dominated the field until the emergence of challenging models in the 1970s. Next, I discuss the two variants of the current American approach that substituted for the classical model: resource mobilization and political process. After that I present the European approach the identity oriented model. Following a chronological sequence, I present the Latin American interpretation of identity, developed in the 1980s upon the European model. After introducing each approach, I discuss how the recent literature is coming to common ground. Finally, drawing on this literature, I present some variables or parameters that deserve attention in case studies of social movements.

Classical Theories of Social Movements

The classical approach, which dominated the field of social movements until the 1970s, followed the social-psychological tradition of the Chicago school, according to which, social strains cause psychological changes and abnormal behavior. Rather than presenting a single theory, however, this tradition encompassed several distinct approaches that shared the following assumptions:

(1) Political action is either institutional-conventional or non-institutionalcollective. 1

(2) Non-institutional collective behavior is not guided by existing social structures and emerges to meet undefined or unstructured situations.

(3) These kinds of situation occur during periods of structural changes in society and when the organs of social control break down.

1 Simply put, institutional political behavior is action according to the rules of the game: voting, joining political parties, filing lawsuits, etc. Non-institutional action is political participation by other means (rebellions, protests, terrorism, etc.).


(4) The resulting strain, distress, frustration, and discontent lead individuals to engage in non-institutional collective action.

(5) Non-institutional action follows cycles that move from spontaneous crowd action to the formation of social movements.

(6) The emergence of these social movements within these cycles occurs through "crude modes of communication," as contagion, rumor, circular reaction, diffusion, etc.

In sum, this early approach considered social movements abnormal behavior that

emerged due to structural changes in society and the psychological strains they cause. As a result, an "epidemic" of irrationality broke out and people started resorting to "non acceptable" forms of participation (non-institutional action). Social mobilization is one of these forms. According to McAdam (1982), under the umbrella of classical theory there are five main models of collective action that seek a similar explanation of social movements: mass society, collective behavior, status inconsistency, rising expectations, relative deprivation, and Davies' J-curve.

Collective behavior theory explains social movements in terms of responses to the normative ambiguities that structural changes cause. The causes of these structural changes may be industrialization, urbanization, rapid rise in unemployment, technological changes, migration, and others. This approach is the most general of all classical models. Supporters of collective behavior do not consider any particular kind of structural change as the underlying cause of social movements. Rather, they speak in terms of general changes (Smelser, 1962; Gusfield, 1970). These changes have the power to disrupt the normative order of a given society. The breakdown of this order raises the anxiety and hostility that lead people to participate in social movements (McAdam, 1982)


According to supporters of mass-society theory, such as Kornhauser (1959), the main cause of structural strains is the absence of an extensive structure of intermediate groups through which people can integrate into social and political life. This lack of intermediate structures isolates individuals who feel alienated and anxious. In these situations, violence and irrational behavior are escape valves for the strained individuals (McAdam, 1982). This model follows Durkheim's idea of anomie, according to which modernization breaks down community solidarity and impels humans to seek new roles and identities by joining new collectivities (Tarrow, 1998).

Proponents of status inconsistency (Broom, 1959; Lenski, 1954; Gerschwender, 1964) argue that a discrepancy between a person's ranking on a variety of status dimensions (education, incomes, occupation, etc.) produces some degree of cognitive dissonance, which is upsetting and leads the individual to collective action (McAdam, 1982).

Similarly, theories of relative deprivation explain collective action in terms of

frustrated expectations of individuals or groups who are not satisfied with their material or status condition relative to others. In Ted Gurr's words: "The social-psychological potential for collective violence is a diffuse disposition toward aggressive action, a primary variable whose immediate determinants in a collectivity are the intensity and scope of Relative Deprivation" (1970: 321).

In his famous Political Order in Changing Societies, Samuel Huntington advances an argument similar to Durkheim's. In his view, societies that undergo rapid modernization are prone to a breakdown of political order. As societies change due to industrialization and urbanization, people seek new forms of participation to maintain a


high level of community. These changes demand new, stronger, and more complex political institutions, which usually develop more slowly than political participation. Increased mobilization undermines the old and anachronistic institutions and leads to confrontation (Huntington, 1968).

The J-curve theory is a variation of the relative deprivation theory. It tries to

reconcile Marx's and Tocqueville's views on revolution. According to Marx, revolutions erupt when the misery of the proletariat increases relative to the economic situation of the bourgeoisie. Tocqueville considered revolutions as the product of liberalization of previously oppressive regimes. Drawing on these two conflicting ideas, J. Davies (1969) built a model according to which revolutions are likely to erupt when a prolonged period of economic and social growth is followed by as short period of sharp reversal. During the period of prosperity people develop expectations of "continued ability to satisfy needswhich continue to rise and, during the latter [period of economic decline] a mental state of anxiety and frustration when manifest reality breaks away from anticipated reality" (Miller et al., 1977).

All the different variations of classical theory have a few common points. First, they stress the reactive nature of collective action. Structural strains cause tensions that trigger social insurgency when they reach a certain threshold. Second, although addressing collective action, these approaches focus on the psychological effects of social strains on individuals, whose reactions are the proximate cause of movement emergency. Finally, the need to resolve those psychological strains, rather than to achieve political goals is a common feature of all these approaches.


The wave of social movements that swept the United States and Western Europe during the 1960s helped discredit the classical theory of social movements. For one, as Cohen (1985) notes, the development of well-organized movements in democratic pluralist systems did not show any sign of irrationality and deviance. Most groups engaged in the confrontational politics of the 1960s had very objective and clear goals as well as strategies to reach them. Also, the actors engaged in the protests and mobilization during those years did not resemble the anomic, isolated, and alienated individuals depicted in the classical approaches.

As far as deprivation goes, it is hard to correlate all contentious groups with feelings of psychological strain due to relative deprivation, considering that many movements cut across class, gender, and race divisions. Even amidst the more socially and economically fragile subordinate classes it is hard to identify instances of particularly accentuated strain. The work by Jenkins and Perrow (1977) is exemplary in this sense. In their study of farm worker insurgencies, they claim that there is no reason to believe that workers' discontent during the 1960s was greater than during the previous decade, when there were no strikes or movement activity whatsoever. In the 1960s, they argue, wages were at their highest and the government had already extended at least a few welfare benefits to farm workers. Even more significant was the fact that linguistic and cultural cleavages were less pronounced in the mid-1960s, when Mexican descendants were finally settling and establishing their communities.

Strains are a constant in the lives of the subordinate. Nonetheless, reality shows that revolutions, riots, protests, and confrontation do not happen all the time. It is easy to associate social change and social disruption post hoc. However, as society is constantly


changing, it is difficult to predict the emergence of social movements based on structural changes. As McAdam (1982) notes, system strain may be, at best, a necessary, but insufficient cause of social movements.

Another critique of the classical approach targets its emphasis on individual

discontent as the proximate cause of social movements. As classical theorists would have it, movement participants differ from the average citizen due to their abnormal psychological profile. McAdam (1982) suggests that this characterization of rebels may serve to discredit insurgents but has little, if any, explanatory power. In any event, classical theorists fail to explain how crowds of individuals characterized by severe psychological traits get together and engage in collective action. Moreover, as McAdam points out, "impressive evidence exists that seriously challenges the assumption of individual malintegration. Especially significant are the many studies that have actually found movement participants to be better integrated into their communities than 'nonparticipants' (1982: 13). Even more important is the fact that proponents of the classical theory failed to provide evidence to support the claim that social strain is the main cause of social movements. If social change causes psychological strain, which in turn causes mobilization, they should have compared psychological strains among participants and nonparticipants. Accordingly, an analysis of social strain over time would be crucial to prove this point; however, what theorists provide are snapshots of a given situation during the eruption of social movements.

The most frustrating flaw in the classical approach is that it denies a direct link between social problems and political action. The characterization of participants as psychologically ill individuals seeking relief ignores their obvious political engagement.


The roots of this denial, as McAdam (1982) suggests, lay in an uncritical acceptance of pluralist state theory by students of social movements. The logic here is simple: If the (American) political system is really open, democratic, and responsive, those who choose not to use the institutionalized channels of politics do not behave rationally. "If, however, one rejects the pluralist model, in favor of either an elite or a Marxist view of power in America, the distinction between rational politics and social movements disappears" (1982: 19).

The critiques of classical theory outlined above are the best starting point to

understanding the paradigm that substituted for classical theory as the dominant one in the US in the 1970s: resource mobilization. As we shall see, this paradigm has evolved from an initial narrow focus on the availability of resources to a broader one, which, besides the availability of resources, considers the effects of political opportunities on collective action. Some authors treat these two approaches (resource mobilization and political process) as the same thing (Cohen, 1985); others classify them separately (Mayer, 1991). The literature shows that they emerged from the same intellectual roots. Nonetheless, for the sake of clarity, I will discuss them in separate sections.

Resource Mobilization

Pluralist theories and their limited conception of the state became discredited

during the 1960s and 1970s. Not coincidentally, during this same period social movements mushroomed in the United States. As noted above, pluralist approaches to the state considered the American political system open and responsive. But the classical


interpretation of social movements as anomalies operating within a fair political system did not convince most students of politics. Instead, many students started looking at the pluralist theory of the state as an anomaly. Soon critiques emerged. The idea that interest groups have equal power and similar chances of success when competing in a pluralist system came under attack. It became clear that some groups are stronger, better organized, and, therefore, more effective than others. These differences may reflect the political and organizational skills of each competing group. More important, though, is that they reflect the imbalances in the distribution of resources among competing groups. As McAdam puts it: "There may exist a political arena in America, but it is not the teeming convention hall depicted by the pluralists, but rather a restricted club reserved for the wealthy and powerful. Only those with sufficient political capital need apply" (1982, 20).

When citizens operate in a system that is not completely open and unbiased,

resorting to mobilization and confrontation, rather than to the institutionalized means of politics, does not seem to be irrational in any sense. To face the challenge of competing against powerful interest groups, powerless groups or classes have to create new strategies to mobilize resources that can give them some political leverage.

Although McCarthy and Zald (1973) were the first to use the term "resource

mobilization," many other authors have used a similar approach, despite some significant differences in their works. Cohen (1985), contrasting this paradigm with the classical, identifies 8 common trends in these works:

(1) Social movements must be understood in terms of a conflict model of collective action.


(2) There is no fundamental difference between institutional and noninstitutional collective action.

(3) Both entail conflicts of interest built into institutionalized power relations.

(4) Collective action involves the rational pursuit of interests by groups.

(5) Goals and grievances are permanent products of power relations and cannot account for the formation of movements.

(6) Collective action depends on changes in resources, organization, and opportunities for collective action.

(7) Success is evidenced by the recognition of the group as a political actor or by increased material benefits.

(8) Mobilization involves large-scale, special-purpose, bureaucratic, formal organizations.

McAdam (1982) summarizes the resource mobilization approach in a simple way: It is the study of the relationship between constant discontent over time and the increase in resources, which allow groups to mobilize (the interplay of factors 5 and 6 above). Again, the relation between resource mobilization theory and neopluralist, or elite theory is very clear. Elite theorists depict society as characterized by a sharp disparity in power between some societal elites and the masses. The effect of this imbalance is the exclusion of most segments of society from any meaningful role in the exercise of political power. As a consequence, proponents of resource mobilization shift the focus of the studies from the masses to the elite, i.e., from the resource-poor to the resource-rich. (McAdam, 1982)

In a theoretical work that outlines the principal elements of resource mobilization theory, MCarthy and Zald (1977) state the importance of elites for social movements.


According to them, "social movements tend to be very limited in their control of discretionary resources. It is only when resources can be garnered from conscience adherents that viable social movement organizations (SMOs) 2 can be fielded to shape and represent the preferences of such collectivities" (1977: 1226). Discretionary resources, they explain, are time and money. By conscience adherents they mean individuals and groups who are part of the social movement but do not stand to benefit directly from SMO goal accomplishment. In other words, if social movements are to take off, they need the active support of an outside elite. As will become clear in Chapter 4, the Catholic Church played the role of a conscience adherent in the case of peasant mobilization in Amazonia. The work of Jenkins and Perrow (1977) on the insurgency of farm workers is a good example of applied resource mobilization theory. They concluded that the powerless rural workers were able to check the elite bias of policy makers by allying with some sectors of the elite. In their words: "Sponsors then serve as protectors, insuring that the political elite remains neutral to the challenge" (1977: 266).

This approach was a direct reaction against classical theories, which considered social movements as the aggregate manifestation of irrational individuals. Also, considering that in the 1970s rational choice theory was gaining momentum in the social sciences, it comes as no surprise that students of resource mobilization embraced strategic rationality in their models.

2 Social Movements Organizations (SMOs) are organizations that mobilize their constituency for collective action guided by political goals, that is, to obtain some collective good. Together, all the SMOs of a given social movement form the Social Movement Infrastructure (SMI). The SMIs of all social movements in a society form the SocialMovement Sector (SMS) (Kriesi, 1996: 153-4). McCarthy and Zald (1977) propose


The first task resource mobilization theorists faced, therefore, was explaining how social movements overcome the dilemma of collective action (Tarrow, 1994, 1998; Foweraker, 1995). According to Mancur Olson, self-interested rational individuals will not join large groups that produce public goods. Instead s(he) calculates that others will do that and s(he) will reap the benefits anyway:

Just as it was not rational for a particular producer to restrict his
output in order that there might be a higher price for the product of his
industry, so it would not be rational for him to sacrifice his time and money to
support a lobbying organization to obtain government assistance for the
industry. In neither case would it be in the interest of the individual producer
to assume any of the costs himself A lobbying organization, or indeed a labor
union or any other organization, working in the interest of a large group of
firms or workers in some industry, would get no assistance from the rational,
self-interested individual in that industry. (1965: 11)

For Olson (1965), the only way to overcome this "free-rider" problem is by

offering selective incentives to would be members or, conversely, by imposing sanctions on those who do not cooperate. McCarthy and Zald (1977) acknowledge this problem and see the solution in organizational terms. They argue that the real task for social movement organizations is to aggregate resources and to translate these resources into action: "The resource mobilization task is primarily that of converting adherents into constituents and maintaining constituents involvement" (1977: 1221). In other words, the challenge is to make people contribute to the movement with time and money. Indeed, they agree with Olson's idea that selective incentives can affect someone's decision of whether or not to join a social movement. Using rational choice language, they explain that selective incentives play an important role in altering someone's perceptions of the costs and a similar categorization, however, they refer to all organization within a social movement


benefits of participation. Also, a gradual professionalization of social movement organizations would offset the problem of collective action.

As the early resource mobilization theorists worked within a rational choice framework, they precluded the very possibility that people may act according to their beliefs and ideologies, rather than according to strategic rationality. Participation in social movements and other forms of collective action may be a reward per se, disregarding selective incentives and social sanctions (Hirschman, 1982).

Having outlined some important aspects of the resource mobilization approach, it is possible now to discuss some of its strengths and weaknesses. Of course, the most important thing about the approach is that it rejected the stigmatization of social movements that underlay the classical theories. Instead of psychologically disturbed actors, resource mobilization considered the members of movements as rational individuals. These individuals join social movements to advance substantive political demands, rather than to alleviate their strains by joining uncontrollable mobs.

A second important departure is the recognition that social movements need and do have some degree of organization. Resource mobilization theorists introduced several analytical categories that help understand mobilization, like social movements, social movement organizations, and social movement sector. Moreover, they discuss how social movement organizations interact among themselves and with their environment. This is important not only because it reveals important organizational aspects of social movements, but also because it sheds light on the problem of collective action. Olson (1965) stressed that the free rider problem is a distinctive feature of large groups. Students

as Social Movement Industry (SMI), rather than infrastructure.


of resource mobilization, however, made an important point in showing that social movements are not necessarily large organized collectivities, but usually webs of smaller organizations and solidarity groups working together to achieve common goals.

The interaction between social movements and their environment (allies and

opponents), McAdam (1982) points out, is important because it accounts for the effect of external groups on the social movements. In his words: "According to most resource mobilization theorists, these opportunities and costs are, in large measure, structured by groups external to the movement. Accordingly, these groups command far more research attention in this perspective than in the classical model" (1982: 23).

Resource mobilization as depicted above makes a valuable contribution to the study of social movements. Nevertheless, it has a strong bias toward neopluralist interpretations of the state. On the one hand, classical theorists considered social movements irrational outbursts of collective action due to the participants' refusal to engage in institutionalized politics. Supporters of resource mobilization, on the other hand, treat social mobilization as simple attempts by organized groups to break the barriers to participation that exist in a system marked by elite politics. According to McAdam:

What is needed are several theories specifically tailored to particular
categories of action. Resource mobilization is such a theory: defensible when
applied to a certain class of collective actions, inadequate as a general
explanation of insurgency. The limits of the model's applicability stem from
the failure of its proponents to adequately differentiate organized change efforts generated by excluded groups and by established polity members.
(1982: 24)


Political Process

As a response to the pitfalls of traditional resource mobilization theory, a new

approach, currently known as the political process model, emerged during the late 1970s and 1980s. This model is more concerned with strategic interactions between social movements and their environment. Also, it stresses the political and social context, rather than the utilitarian logic of individual actors and elite participation. Yet, some students of social movements, as Cohen (1985), classify it as part of resource mobilization theory 3. Particularly important for students of political process is the reason why social movements emerge. The resource mobilization theorists somehow neglected this basic question to focus on how social movements mobilize.

To a certain extent, followers of political process pay more attention to a variable whose importance classical theory pointed out, namely, structural changes in society. Of course the political process model does not have any of the "psychologism" present in classical theory. However, it does acknowledge that changes in society may cause changes in the interests and opportunities for subordinate groups to mobilize, as well as in the ways these groups defend their interests. The works of Charles Tilly are central to this approach.

In his From Mobilization to Revolution Tilly (1978) stresses how the spread of capitalism and the establishment of nation-states changed modes of social control, and,

3 Some critics of this approach refer to this division as a split within the approach, rather than a development. As a result of this split, two competing perspectives emerged. RM I "conceives of the social movement sector in free competition with other sectors of society on an open market place of groups and ideas. RM II "is more concerned with the structure of political opportunities and with the degree of organization within the deprived social groups" (Mayer, 1991).


therefore, of social resistance. Although he does not propose any linear and deterministic model of evolution of social mobilization, he outlines some general trends in changes of modes of resistance. Tilly classifies collective action in three main categories, according to the claims that mobilizing groups assert: competitive, reactive, and proactive. Competitive claims are those involving groups that compete for the same resources. Reactive collective action involves groups that try to maintain or regain established claims when an outsider challenges them. Proactive collective actions assert group claims that have not been previously exercised.

Although Tilly warns against viewing these three modes as stages in an

evolutionary process, he does suggest that there is a long-term shift from competitive and reactive modes to the proactive one (Cohen, 1985). As the nation-states appeared and capitalism spread throughout the world, traditional modes and targets of protests changed, and new challenges, as well as opportunities, emerged. In the language of supporters of the political process model, these changes provided political opportunities for people to organize and mobilize.

Political opportunities, however, do not appear only when major structural

changes occur, as in the case of the emergence of nation-states. They are also present in less spectacular events that states face periodically. As Tarrow (1994) puts it: "Not only when reform is pending, but when institutional access opens, when alignments shift, when conflicts emerge among elites and when allies become available, will challengers find favorable opportunities." (1994: 81). In a more recent and refined definition, Tarrow (1998) explains that "by the concept of political opportunity, I mean consistent but not necessarily formal or permanent dimensions of the political environment that provide


incentives for collective action by affecting peoples'expectations for success or failure" (1998: 76-77).

Tarrow (1998) identifies five main dimensions of political opportunities: (1) the liberalization of the political system, which means increasing opportunities for participation; (2) the evidence of political realignment within the system; (3) the appearance of strong allies; (4) divisions among the elites; and (5) a decline in the state's capacity to repress dissent. All these dimensions shifted in Brazil during the transition to democracy, and certainly contributed to the emergence of the social movements studied here. I will return to this point in Chapter 4.

When challengers seize the political opportunities to mobilize, they may expand these opportunities to other groups or social movements by giving them an example to follow and by exposing the weaknesses of those they challenge. When other groups follow suit, simple mobilization may become a longer cycle of protests, one in which less mobilized and resourceful groups also resort to confrontational politics. Usually, these cycles are geographically broad, even crossing state boundaries, as occurred in the United States and Europe in the 1960s.

In many instances, social and particularly technological change worked not only to instigate mobilization, but also to provide new forms of communication and coordination (Tarrow,1994, 1998). Particularly important, in Tarrow's view, was the popularization of the printing press in the eighteenth century. Coupled with the new forms of civic association that became popular in this same period, it provided an unprecedented opportunity for individuals and groups to organize and ally. In this sense, the connection with the resource mobilization paradigm is clear here: New technologies and new forms of


association are resources that became available to discontented groups and individuals. The novelty in this approach, however, is that these resources are prior to mobilization. They worked as a catalyst of individual discontent, rather than as a lever to increase some group's power.

Political opportunity provides a powerful explanation for waves of mobilization and revolutions and also for single cases of mobilization, but one must be careful in using it. It may be too easy for any researcher to find changing structures of opportunity behind a given social movement and use them to explain the case aposteriori. In other words, political opportunity takes the risk of explaining nothing due to its power to explain everything. In the case analyzed here, it is clear that democratization created opportunities for mobilization, but was not the causal mechanism of peasant protests.

This risk of explaining everything may lie with the very assumption that grievances are a constant in society. Resource mobilization theorists in general believe that these grievances will cause mobilization when aggrieved groups accumulate enough resources or when windows of opportunity open. However, Buechler (1993) shows that new grievances reshaped and revamped contemporary women's movements in the United States. Women who experienced discrimination within the context of other social movements led the formation of the women's liberation movement (a faction of the women's movement). He argues that grievances can be as important as access to resources and opportunities in explaining social movements. Indeed, new grievances may function in a similar way as opportunities in generating mobilization, but they are analytically distinct categories. The emergence of new grievances does not necessarily involve shifting alliances, splits within elites, and increasing opportunities for participation.


Later in this chapter we will see that social groups may construct new grievances by changing the way people interpret old ones.

Another important distinction between the resource mobilization and political process approaches concerns the role of elites in mobilization. Whereas followers of resource mobilization consider elite support a key aspect of mobilization, those working within the political process approach see elites as enemies rather than allies of insurgents. According to McAdam:

Proponents of the resource mobilization model depict segments of the
elite as being willing, at times even aggressive, sponsors of social insurgency.
By contrast, the political process model is based on the notion that political
action by established polity members reflects an abiding conservatism. [T]hey
work against admission to the polity of groups whose interests conflict
significantly with their own. (1982: 38)

It comes as no surprise, therefore, that McAdam (1982) considers elements of

Marxist theory more compatible with the political process model than pluralist theory. For Marxists, the power disparity between elites and the subordinate classes is huge, but not immutable. They consider the subjective transformation of consciousness as a crucial process in the generation of insurgency, an idea that McAdam considers part of the political process model, as he proposes it, too4. As this new model of political process incorporates some aspects of the identity-oriented paradigm, we need to present the latter before proceeding.

4 Whether elites are potential allies or enemies of social movements may depend on the very definition of elites. In the case of rural workers mobilization, an economic elite (ranchers) were clearly the enemies of the rural workers. However, these workers found reliable allies in other actors, such as researchers, intellectuals and members of the clergy, who are also elites.



The mushrooming of social movements during the 1960s and 1970s did not occur solely in the United States. Europe experienced a similar phenomenon, which did not escape the attention of local social scientists. As we have seen, theories of social movements in the United States have been shaped by epistemological and theoretical developments in the social sciences during the 1970s. Europe experienced a similar process. However, whereas in the United States students of social movements built their theories upon, or responding to pluralism and economic rationality, in Europe the major influences were Marxism and structuralism (Cohen, 1985; Escobar and Alvarez, 1992; Tarrow 1994, 1998; Foweraker, 1995).

An interesting and very obvious departure from Marxism lies at the very

foundations of the European approaches to social movements (to which Cohen later attached the label "identity-oriented"). Social scientists acknowledge that these movements did not emerge following traditional class cleavages. Environmentalism, feminism, civil rights movements, gay rights, and other interests were, for the first time in history, at the core of social mobilization. Due to this change in the nature of claims, some authors designated these movements as the new social movements.

Another distinction between the identity approach and the traditional Marxist

approach is the rejection of the state as the central locus of struggle for mobilizing groups. "They [new social movements] target the social domain of 'civil society' rather than the economy or the state, raising issues concerned with the democratization of structures of everyday life and focusing on forms of communication and collective identity" (Cohen,


1985: 667). Alberto Melucci (1989), a leading scholar within the identity approach claims that contemporary conflicts develop in areas where both symbolic investments and pressures to conform are heaviest. In Gramscian terms, new social movements engage in a war of position in the realm of civil society, rather than in direct confrontations against the state.5 In Melucci's words, "contemporary movements operate as signs, in the sense that they translate their actions into symbolic challenges to the dominant codes"(1989: 12).

The associations and groups through which individuals organize are not simply the means to achieve mobilization in large scale and the consequent political gains. Rather, organization in new groups is an end by itself The assertion of group identity means the legitimization of the group in society at large. Democratization, according to this perspective is beyond the realm of the state. It has to reach state, market, and society (Cohen, 1985).

This new form of social organization, according to French thinker Alain Touraine (1985), is exclusive to modern post-industrial, or "programmed" societies (1985: 781). This new kind of society is the result of a complex set of actions that society performs in itself. In other words, society establishes new values, rules, and cultural meanings, which cause groups to mobilize to control these new symbols and meanings, or to produce new ones. This "reflexivity" of actors and the new arenas of struggle, rather than new repertoires of contention, are the factors that make new social movements new. Also, this reflexivity emerges from the arenas opened up by postindustrial society (Touraine, 1985). As Cohen points out, the circularity of the argument is obvious for, by definition,

5 In advanced industrial societies, society is strong and sustains the state. In this case, a "war of movement" to conquer the state is useless. The task for the revolutionaries is to


postindustrial society is defined as such because it generates new forms of collective action (Cohen, 1985).

At this point, Melucci and Tourraine are at odds. Melluci (1989) argues that both supporters and critics of the new social movements approach share a common epistemological limitation: They tend to regard the phenomenon of social movements as a unified empirical object. By doing so, supporters try to stress its novelty, whereas critics identify historical continuities with old movements. Melucci argues that the novelty of these movements is exactly what both sides of the debate missed; the "different relationships and meanings" of these movements (1989: 43).

This debate has had at least one positive consequence: It has stressed the plurality and diversity of social movements concerning their objectives, constituency, targets, strategies, and meanings. The diversity of social movements demands a diversity of methodological tools in the box of social researchers. Students within the resource mobilization framework, according to Melucci (1989), fail to recognize this need for methodological and epistemological diversity. They focus on the "visible" aspects of social mobilization and collective action, and by doing so commit the mistake of "political reductionism." By political reductionism he means the limitation of analysis to the aspects of social movements that are clearly political, like protesting to confront authority. By focusing on these visible aspects of social movements, resource mobilization theorists make all movements that do not target the polity directly irrelevant, or "invisible," as Mayer puts it (1991: 175).

take control of civil society, in a "war of position"(Gruppi, 1978).


Even if new social movements are not so new, this approach has made a great contribution to the study of social movements by stressing the importance of identity, a factor neglected by other approaches. The main problem in this approach is certainly its neglect of political struggle, particularly struggle against or within the realm of the state, which is a fundamental of most social movements, traditional or "new." Also, as Foweraker notes, identity is a slippery concept. Individuals must come together to form collective identities, but how and why they get together in the first place is a crucial question that the identify-oriented approach does not address adequately6.

Social Movements in Latin America

Social sciences in Latin America were also affected by the revival of social

movement studies in the first world. The identity-oriented approach had a significant impact on the continent, initially in a somewhat naive way, which produced a wave of "romanticized" studies of social movements (Roberts, 1997). Most students rejected the emphasis on resources and strategies that characterize the resource mobilization approach, but many scholars paid due attention to the political opportunities, particularly state crises, that facilitated the emergence of movements throughout the region. The idea of identity developed within the new social movements approach was particularly well received by Latin Americans and students of social movements in the region.

6 Although Hirschman (1982) did not engage in this debate, he considers participation in collective action as a means to achieve personal satisfaction and to alleviate the frustrations of consumerism and rational individualism, which are common features of modem society.


Latin American scholars did not uncritically import Touraine's idea of reflexivity in postindustrial societies. According to Escobar (1992) and Calder6n et al. (1992), they rather adapted and reshaped it in a syncretical way. According to these authors, Latin American scholars have a tradition of syncretism that has already produced powerful research approaches, such as dependency and some versions of modernization theory.

In Latin America, it was precisely the crisis of the state that triggered the new phase of social protests (Calderon et al, 1992; Escobar, 1992; Escobar and Alvarez, 1992). The failures of the developmentalist/populist state in the Southern Cone and of the oligarchic state in Central America, and a general crisis of mechanisms of representation are particularly important components of this crisis. As opposed to the post-industrialism of European society, Escobar (1992) argues, Latin America was experiencing an "organic" crisis of modernity. Under these conditions, Latin America was not getting past a crossing of a modern era. Rather, it was skipping modernity altogether:

Social polarization, heterogeneity, and exclusion have reached
unprecedented proportions in the development era. The erosion of modernity is evident in everyday life, in the concrete behavior of people, in the economic crisis, and in the disenchantment with the modern projects of nation-building, politics, and development. What is hanging in the balance is not only politics,
progress, and democracy, but also a whole civilizational design based on
modern reason. (1992: 68)

What is clear in the studies of social movements in Latin America, however, is the rejection of the crudest versions of the resource mobilization approach. For one, this rejection is grounded on empirical and epistemological aspects of resource mobilization, particularly the inappropriateness of the pluralist theory of the state in the Latin American context. For some critics, the resource mobilization model is a blueprint for the inclusion


of middle-class movements into the American political system, a process marked by bargains and alliances, and, above all, a peaceful process. Disparity of power and income, as well as the undemocratic and repressive character of most regimes in Latin America, make normal politics difficult. What is significant, as Foweraker (1995) notes, is that, despite these (true) aspects of Latin American society and politics, most social movements in the region are peaceful and "seek democratic changes through strategic calculation" (1995: 26). It is his opinion, therefore, that Latin American theorists should not discard resource mobilization altogether.

Another critique of resource mobilization concerns epistemological aspects of this approach. Latin Americans reject the quasi teleological aspects of the approach implied in the model of economic rationality. Feminist theorists have been very critical of this model. In their view, it ignores class, gender, and race specificities that render the concept of a rational actor useless. For those feminists, the selfish rational actor is the white middle class male of western advanced societies, and has little, if anything to inform research outside this context (see Foweraker, 1995). This critique of both rational choice and grand theory is strong within postmodern academic circles. In Escobar's words:

Poststructuralist and postmodernist insights of various kinds inform
social movements theory in different parts of the world, especially in Western Europe and Latin America. More clearly in Latin America than elsewhere, the move toward a grand "theory of social movements" is actively resisted. (1992:

In general, Latin American social scientists rejected the American approach on the ground of its context specificity. Some scholars criticized the American paradigm for


failing to account for differences in the nature of the state and, therefore, of the political struggle here and there. Paradoxically, they embraced a theoretical approach that virtually ignores the importance of the state as a contender, as an arena, and as the arbitrator of politics:

But we cannot overlook the fact that the social movements of twentyfive years ago had strong state/political orientations and that, in contrast, many of today's actors are searching for their own cultural identities and
spaces for social expression, political or otherwise. (Calderon et al., 1992: 23)

By recognizing that transitions to democracy and a general crisis of representation contributed to the emergence of social movements in the region, Latin American scholars recently accepted some insights of the resource mobilization approach, particularly the idea of political opportunities, as we shall see in the next section. Nonetheless, the main focus of the Latin American scholarship during the 1980s and early 1990s was society, rather than the state. In other words, scholarship in the region chose the "invisible" side of social movements, to the detriment of the more "visible" political struggle.

The fact that scholars from distinct schools of social science have been able to identify and conceptualize different factors that affect social movements is a very significant contribution to understanding collective action. Nevertheless, the refusal to accept the insights of other approaches has hindered the studies of both social movements and collective action. The idea of competitive paradigms fits well the situation in the field of social movements during the 1980s and early 1990s. Fortunately, a new generation of studies is using the existing approaches in a complementary, rather than competitive way. The next section is a discussion of this tentative synthesis.


New Trends in the Studies of Social Movements: Toward a Synthesis

In a review of recent books on social movements in Latin America, Roberts (1997) criticizes the ideological side, as well as the optimism around the studies of social movements in the region during the 1970s and 1980s:

For a generation of left-wing academics and political activists
disillusioned by the repression of vanguard parties, the defeat of guerrilla
movements, and the political weakness and vicissitudes of organized labor, new social movements were a godsend: a new form of popular subjectivity
that aimed at a radically egalitarian and participatory sociopolitical order and
thus restored faith in the progressive march of history. (1997: 138)

The course of events, according to him, dismissed the high expectation implicit in this romanticized view of popular movements. Scholars responded to this dissonance between facts and expectations by adopting a "more sober tone." According to him, the best of recent literature is not merely praising the emergence of grassroots organizations or the opening of new spaces for social expression. This literature is "making a serious effort to understand how social movements engage the formal arenas of institutional politics and try to influence public policy"(Roberts, 1997: 139).

One important aspect of Latin American politics that greatly affects the emergence and disappearance of social movements is the very opening of the political systems and the subsequent democratization that most countries in the region have experienced. Of course, the political process model offers useful insights to the analysis of the relationship between regime transition and social mobilization. An example of this new approach to social movements in Latin America is Oxhorn' s study of shantytown organizations in Chile. While stressing the importance of creating and maintaining a collective popular identity within the popular sectors, he also scrutinizes how these popular movements translated


such a collective identity into action. The idea of political opportunity is central to his thesis that under certain circumstances, the resumption of significant base-level organizational activity is not only possible under an authoritarian regime, but also a "direct result of the authoritarian experience itself'(Oxhom, 1995: 6). By repressing traditional modes of political participation, the authoritarian state may offer new opportunities for reorganization of society along new lines, like territorially-based urban movements.

Oxhorn explores the modes and strategies of organization and mobilization of

these social movements too. He acknowledges the importance of "umbrella organizations" in protecting base-level organizations against state repression during authoritarianism. In the case of Chile, and in Latin America in general, the Catholic Church has been able to provide such protection due to two main factors. First, grass-roots movements consider the Church a trustworthy ally. Second, the Church is relatively immune from the state repressive apparatus, due to its popular support and to the Christian values embedded (at least rhetorically) in the state ideology. My study partially corroborates Oxhorns's view. The Catholic Church was certainly a trustworthy ally for the peasantry, although it was not always immune form state repression.

Another important study following these lines is Stokes' analysis of the

relationship between social movements and the state in Peru (Stokes, 1995). She identifies the advent of the corporatist State in Peru as a factor that enabled the formation of new identities among the poor:

The rise of the "classist" labor movement, the injection of the "critical
idea" of Peruvian history and society into public school curricula, the arrival in the shantytowns of legions of outside organizers with new messages about the
source of poverty and possibilities for change all of these, direct or indirect


results of the military government policies, transformed the worldviews of
large segments of the Urban poor. (1995: 116)

The central focus in Stokes' work was the role of popular culture in the emergence of social movements, but she did not limit her study to vague concepts of identity. Rather, she discussed how different political cultures lead to different strategies of organization, mobilization, and confrontation (or to the decision of not mobilizing at all). These strategies are based on normative aspects that tell the actor what kind of political action is right and what is inappropriate.

Another work that stressed the importance of culture was provided by Alvarez et al. (1998) who claim that social scientists have neglected the linkages between culture and politics. There is, according to the authors, a need to go beyond the textuality and forms of representation to understand what social movements really mean:

Culture is political because meanings are constitutive of processes that,
implicitly or explicitly, seek to redefine social power. That is, when movements deploy alternative conceptions of woman, nature, race,
democracy, or citizenship that unsettle dominant cultural meanings, they enact
a cultural politics. (1998:7)

It is important to emphasize the fact that in Latin America today all social movements enact a cultural politics. It would be tempting to restrict the concept of cultural politics to those movements that are more clearly cultural. Alvarez et al. (1998) argue that in the 1980s, this dichotomization resulted in a useless division between "new" and "old" social movements. The new movements were indigenous, ecological, women' s, gay, and human rights movements. The old movements were peasants, labor, and neighborhood movements involved in more traditional forms of struggle. They rejected


this differentiation between new and old, and along with it, the antagonism between identity and strategy that once existed in the field:

In their continuous struggles against the dominant projects of nation
building, development, and repression, popular actors mobilize collectively on
the grounds of very different meanings and stakes. For all social movements,
then, collective identities and strategies are inevitably bound up with
culture.(Alvarez et al., 1998: 6)

The authors also recognize the importance of the webs of social movements. These webs, in their view, connect, in an informal and sometimes chaotic way, a plethora of social movement organizations, individuals, churches, NGOs, sympathizers, party members and even state officials. These webs help social movements achieve their goals, particularly by deploying their "discourses and demands in and against dominant political cultures and institutions" (Alvarez et al., 1998:16) an argument very close to Melucci's idea that new social movements "operate as signs, in the sense that they translate their actions into symbolic challenges to the dominant codes" (1989: 12). When one examines the success of social movements, therefore, he or she needs to consider to what extent these alternatives, discourses, demands, and practices have penetrated society through the ramifications of these webs.

This cultural dimension of power has its roots in the ideas of Italian thinker

Antonio Gramsci, who criticized the materialist reductionism of European Marxists. The Latin American left committed the same mistake by interpreting the state as a "thing" that can be seized and controlled. This perspective ignores the roles of ideas and culture in producing and reproducing unequal relations of power. Recently, the Latin American left incorporated a Gramscian approach which, according to Dagnino (1998):


[S]tresses the confluence of three different tendencies: a renovative
critique of traditional Marxism, an emphasis on the building of democracy
with its correlated strengthening of civil society, and, in the interstices of these
two, a new approach to the relationship between culture and politics. (1998:

As state and society are not separate pieces according to the Gramscian approach, but rather a continuum, political changes have to encompass both. This organic relation between state and society is key to understanding the idea of cultural politics and how it relates to social movements. Social movements have been able to overcome limited conceptions of democracy by targeting society in general, rather than only the "political institutions as traditionally conceived" (Dagnino, 1998: 47). In a hierarchical society whose members are ranked along gender, race, class, sexual orientation, social movements have to press for changes in attitudes as well as in political practices. The struggle to change citizens' rights goes hand in hand with the struggle to assert their rights to rights before society. In other words, the demise of state authoritarianism will be complete and real only when, and if, "social authoritarianism" is eliminated (Dagnino, 1998).

These recent changes in the intellectual orientation of social movement studies in Latin America have produced some insights that should be seriously considered by any student of social movements. Contrary to the early studies that focused on identity in a narrow way, this new approach does not exclude the main focus of other approaches. Nonetheless, it still overemphasizes struggles over meanings and symbols to the detriment of confrontation in the political arena.


Clearly the theory developed within the tradition of identity has refined its

concepts and incorporated some ideas of other paradigms. It is legitimate now to ask how the proponents of the other paradigms have dealt with identity and culture.

In his introduction to a recent edited volume on social movements, McAdam

(1982) praises the increasing recognition in the field of social movements of the three main factors that affect collective action. These factors are (1) the structure of political opportunities and constraints; (2) the forms of organization available to insurgents; and (3) the collective processes of interpretation, attribution, and social construction that mediate between opportunity and action. It is worth noting that McAdam et al. (1996) did not adopt the term identity, but rather "framing process." Both terms, however, have a similar meaning. What is significant is the central position of the framing process in McAdam's model. According to him, framing is a sine qua non condition for mobilization:

If the combination of political opportunities and mobilizing structures
affords groups a certain structural potential for action, they remain, in the
absence of other factors, insufficient to account for collective action.
Mediating between opportunity, organizations, and action are the shared
meanings and definitions that people bring to their situation. At least people
need to feel both aggrieved about some aspect of their lives and optimistic
that, acting collectively they can redress the problem. Lacking either one or both of these perceptions, it is highly unlikely that people will mobilize even
when afforded to do so. (McAdam et al., 1996: 5)

The lag between the emergence of identity as a factor in the study of social movements and its adoption by followers of both resource mobilization and political process has two reasons. First, as McAdam et al. (1996) stress, the conceptual imprecision of the term identity has scared scholars away from it. Recent writings, they argue, "have equated the concept [cultural framing] with any and all cultural dimensions of social


movements" (1996: 6). The early works on identity in Latin America, as we have seen before, certainly contributed to this problem. Second, culture traditionally has been perceived as structural characteristics of societies and communities, which does not fit the more dynamic model of political process. This view has been replaced by "conceiving of culture and framing as strategically produced," which has made it more attractive to students of political process (Zald, 1996: 261). This new view of cultural framing emerged from several distinct branches of social sciences, particularly from cultural anthropology, social psychology, and dramatistic and rhetorical analysis. It was also a development within the field of studies of social movements, particularly the analysis of culture as repertoires of action (Zald, 1996).

Latin American theorists have stressed the importance of cultural politics in

creating counter-hegemony in authoritarian state-society systems. Scholars coming from the political process and resource mobilization tradition point to a related although distinct issue: competing groups. For them, social movements contest not only authorities and society in general, but also other organized groups. These opposed groups compete in mobilization contests, flexing their muscles to show who can rally more support and resources. More important, though, is the competition around meanings, aimed at establishing legitimacy and showing the rightness of their causes to society and state (Zald, 1996).

Starting in the late 1980s, American and, to some extent, European literature on social movements increasingly combined studies of political opportunities and analysis of organizational strategies of social movements with very good results (McAdam et al., 1996; Tarrow, 1994, 1998). Recently, the idea of identity and cultural politics has started


diffusing among supporters of these two approaches. The opposite is also true; proponents of identity and cultural politics are more and more accepting of insights from the other paradigms. This is good news for the study of social movements. However, empirical studies are just beginning and the development of the field will demand a lot of case studies and cross comparisons that integrate culture, strategies and opportunities. The words of McAdam are a good starting point for new studies of social movements:

Ultimately, however, the analytic utility of these three models [classic,
resource mobilization, and political process] will be determined not on their
abstract theoretic merits but on the basis of how well each accounts for particular social movements. Thus, my final objective will be, wherever
possible, to assess the degree of "fit" between the empirical implications of
these three perspectives and the data.(1982: 03)

I do agree that neither of the main approaches to social movements is theoretically superior to the others. However, the literature provides us with some clues on how to guide empirical research. There are significant differences between the "new social movements" and the "traditional" ones. Supporters of the identity-oriented approach convincingly argue that some forms of movements are targeted at society and fight in the realm of symbols and meanings, whereas other forms of movements struggle in the more traditional realm of politics. Nevertheless, a dichotomous classification of social movements along the lines of old/new or visible/invisible is neither analytically accurate nor useful. Instead, it seems more appropriate to think of the nature of social movements as a continuous between a theoretical "pure" society-oriented and its opposite: a pure polity-oriented form. Also, there is no reason to believe that some movements cannot, or actually do not, fight to change both subjective meanings and the more objective political


reality. Or yet, social movements may change their very nature during their lifetime or even keep switching their locus of action back and forth from society to the polity.

The very nature of a given social movement is, therefore, an empirical question that demands research. The case I study in this work, however, is close to the polityoriented spectrum of social movements that I suggested above. Rural workers directly confronted the state to increase their access to state resources and to policy making. Therefore, resource mobilization and political opportunity approaches are more useful in guiding the analysis of the case.


The case of peasant mobilization in Transamaz6nica is simple: the movement

targeted the state which, according to one movement leader, "brought us to the middle of nowhere and, all of a sudden, abandoned us leaving all their promises unfulfilled." Their struggle was aimed at gaining access to tangible state resources, such as schools, health care, transportation infrastructure, and rural credit. The underlying idea of resource mobilization that social movements are an extension of politics by other means is extremely relevant in this case. Peasants who live in the frontier, in a very delicate ecosystem, are among the least enfranchised groups in Brazil. Their capacity to mobilize and disrupt normal politics is one of their greatest assets, and the strategies they use to deploy it are key to understand their struggle. Also, although the focus of the movement was not cultural, but rather materialistic, the peasants engaged in a struggle to establish their rights and importance as a social group. By working on their image of food producers, environmentally friendly farmers, they attracted new allies and social


supporters to their cause. In other words, identity may have played an important role in this movement, but in a rather strategic way. For these reasons, the resource mobilization/political opportunity approach provides an adequate framework for the case studies here.

Rucht (1996) claims that external factors profoundly affect not only the

emergence, but also the very organization of social movements. Simply put, he considers that both mobilization and organization are dependent variables, whereas political opportunity is the independent variable. Moreover, he advocates the use of structures as the independent variable, for they are more stable and comparable than simple opportunities.

Structure has three contextual aspects: cultural, social, and political. The cultural context refers to the attitudes, values, and behavior of individuals who may or may not support the movement (or, using Rucht's words, be consonant or dissonant with the movement). The social context is the embedding of the social movement in its social environment. The environment may facilitate or restrict the formation of collective identities, the establishment of support networks, and even the socialization of potential movement supporters. Finally, the political context deeply affects the overall likelihood and nature of political opportunities. The main variables of the political context, which can facilitate are the formation of collective identities are: (1) access to participation; including both formal and informal channels to influence political decisions (2) the policy implementation capacity of authorities; (3) alliance structure; and (4) the conflict structure, which is the constellation of adversaries of the social movements. To these variables Tarrow (1998) adds (5) the repressive capacity of the state.


A study of social movements should pay attention to how these structural aspects, under normal conditions, facilitate or constrain the emergence of social movements, as Rucht (1996) argues. As the structural context affects the organizational structure of social movements, the five aforementioned variables will determine the type of movements that will likely emerge under certain circumstances.

The case studied here is complex because the political structure underwent profound changes during the period analyzed. In fact, these changes represented opportunities for the rural workers to mobilize. To make the analysis clearer, I discuss at length the political environment within which the movement operated in Chapter 3. In Chapter 4, I discuss some of the other structural variables (cultural and social) while reconstructing the history of the movement.

Kriesi (1996) proposes a model for analyzing the evolution of social movements. In this model he stresses four parameters of organizational development: (1) organizational growth and decline- the changes in size of the social movement sector, and the variation of resources within a given social movement organization (SMO); (2) internal structuration as an immediate consequence of the resource flow, a SMO can experience formalization, professionalization, internal differentiation and integration; (3) external structuration refers to the integration of the SMO with its constituency, allies, and authorities; and (4) goal orientation and action repertoire as a SMO evolves, it tends to change its goals. Many times, the very maintenance of the organization becomes its main goal, which implies a conservative accommodation of the movement.

Variations in these variables can lead a SMO to four different evolutionary paths:


(1) Institutionalization a SMO becomes institutionalized when it stabilizes its resource

flow, the development of its internal structure, its relations with external actors, its

goals, and repertoire of action. At this point, the SMO resembles a party or an interest


(2) Commercialization is the transformation of a SMO into a service organization. This

may occur on purpose, or as an intended consequence of an increased emphasis on the

use of selective incentives to maintain and increase membership.

(3) Involution is a change toward an exclusive emphasis on social incentives, like mutual

help and solidarity. In this case the SMO becomes something resembling a club, selfhelp group, or voluntary association.

(4) Finally, the SMO may reinvigorate mobilization to achieve its goals, undergoing a

process of radicalization.

To the four paths suggested by Kriese, one could add co-optation by establishing or reinforcing existing patron-client relations. Stokes (1995) stresses the importance of this path in her study of social movements in Peru. She defines clientelism as

[A] dyadic relation between two unequal actors in which the superior
actor trades goods and services for other valued goods (political support, labor) from the inferior actor. Although the exchange is beneficial to both parties in some limited ways, it does not enhance the power of the inferior
actor. Clientelism is thus a strategy for dominant groups to retain their
position of superiority. (1995: 112)

These patron-client relations may develop between movement leaders and rank and file, between movement leaders and state authorities, or between movement leaders and power brokers that mediate the relationship between the movement and the state. State officials may seek to promote patron-client relations either as a way to sabotage a social


movement or as a rent seeking strategy. As will become clear in Chapter 4 and Chapter 5, the introduction of rural credit created new opportunities for patronage and clientelism; however, it would not be fair to say that the movement was co-opted. We shall return to this point later.

It is important to stress that the evolution of social movements is not unilinear or unidirectional. It can take many paths and shift back and forth from one path to another. The rural workers movement in Pard experienced periods of radicalization followed by institutionalization, and then back to radicalization. At the time of this research the movement was at a crossroads, due to the ambiguous outcomes of its most recent achievement: access to rural credit. Therefore, it is important to assess the outcomes of social movements. Relative success or failure of a movement has to be measured against the formal goals of the movement, as well as the expectations of its constituency. The very evolution of a social movement may be linked to its performance. As Hirschman points out, frustration with public participation may lead individuals to exit collective action and to shift their attention and energy back to their private lives. Conversely, the fulfillment of a movement's demands may demobilize its members and kill the movement, or lead it to a path of commercialization or involution, as described above.

The discussion of theories of social movements suggests that, at least in cases like the one studied here, the state has an overwhelming effect upon the emergence, durability, and achievements of social movements. Yet, the approaches presented here fall short of providing acceptable accounts of the relationship between the state and social movements. This relationship between state and social movements is a major gap in the political science and sociology literature. State theorists, on the one hand, "have largely focused on


those who hold and wield power rather than studying their challengers. Social movements scholars have focused primarily on those who are contesting power rather than their relations with the powerful" (Jenkins, 1995: 15).

The common critique that the resource mobilization approach is based on a narrow US-centered neopluralist interpretation of the state has its merits. As much as social movement theory, state theory is a contested field of political science, and a state-centered view of social movements must account for the very nature of the state. I do not have the ambition to bridge this gap here, but a study of peasant mobilization in the Transamaz6nica region does require some theoretical as well as empirical study of the state. The relationship between the peasantry and the state was central to every aspect of this case, and an uncritical acceptance of the neopluralist assumptions implicit in the resource mobilization approach would cripple this study. In the next chapter, I outline the major points of the debate on the state. Chapter 3 is a historical analysis of contemporary state action in the Brazilian Amazon.



Popular mobilization does not occur in an institutional vacuum. As soon as social movements arise, they must deal with the state, which regulates acceptable forms of mobilization, and has discretionary power to repress societal elements that threaten its authority. Moreover, movements aimed at gaining access to the state policy-making structure and state resources, like the one I study here, must adapt themselves to the very structure they deal with. The state, therefore affects social movements during their interaction, and, in extreme cases, the outcomes of popular mobilization may depend on the state as much as on the attributes of the social movement itself.

As we have seen in the previous chapter, the approach that accounts best for the relation between state and social movements resource mobilization and its variations entails some problematic assumptions about the state. The nature of the state can account for the opportunities and constraints that social movements face, and, therefore, for the diversity of strategies they deploy in distinct contexts. More important, however, is the fact that states vary in their capacity and willingness to implement or to resist implementing policies once a compromise is reached. The pluralist and elitist assumptions embedded in the resource mobilization approach fail to capture this "supply side" of policy making. Even though recent works recognize the importance of structural and institutional factors (Rucht, 1996; Tarrow, 1998), the literature on social movements largely ignores the academic debate on the state, which is crucial to understanding the emergence and


evolution of social movements, as well as the outcomes of the policies they demand. This chapter presents a discussion on the main theories of the state and the issue of state autonomy and how they relate to social movements. I begin with the pluralist and elitist approaches to the state, which are not only the bases of the resource mobilization model, but also the very focus of the debate on state autonomy. After that, I review the literature on Marxist approaches to the state, and discuss how it relates to both this debate and to the theories of social movements. Finally, I do a similar analysis of the corporatist model of the state.

Pluralism, Elitism, and State Autonomy

Until the early 1980s, the state would rarely appear in academic works as a

meaningful independent actor or an organizational structure, much less as an independent variable in the studies of politics. Nettl (1968) argues that the state vanished as a conceptual variable due to epistemological as well as ideological and geographic factors. From an epistemological perspective, the concept of the state, as it is commonly used in studies of international relations, proved to be problematic. The idea of sovereignty, which is in the core of the definition of the modern state, came under attack and was pushed to the margin of political science by Frederick Watkins, who said it was a "limiting concept an ideal-typical situation that had to be qualified in all sorts of ways" (Nettl, 1968:560). For him, in the late 1960s, the problem of sovereignty was, for social scientists, a "dead duck." Even more problematic was the very idea of nation-state. Although there were exceptions to the idea of one nation corresponding to one state (like Switzerland), the exceptions almost became the rule recently. In Nettl's words: "If the entry of the third


world onto the stage of modern socio-scientific consciousness has had one immediate result or should have had, it is the snapping of the link between state and nation" (1968: 560).

The geographical and ideological factors that Nettl referred to are the shift of the epicenter of social science research from Europe to the United States. As Almond (1988), drawing upon Garson's work (Garson, 1974), points out, "[P]olitical science grows largely reactively [. ], rather than through the self-conscious testing of analytical models" (1988: 854). This shift meant the "contamination" of political science by the "statelessness" of American Political Science Scholarship, which was a refection of the very "statelessness" of the American political system. Skocpol (1985) corroborates this view. She claims that the "statelessness" of American political science scholarship was embedded in the American and British liberal ideology:

As long as capitalist and liberal Britain, and then capitalist and liberal
America, could plausibly be seen as the unchallengeable "lead societies," the
Western social sciences could manage the feat of downplaying the explanatory
centrality of states in their major theoretical paradigms for these paradigms
were riveted on understanding modernization, its causes and direction. And in Britain and America, the "most modern" countries, economic changes seemed
spontaneous and progressive, and the decisions of governmental legislative
bodies appeared to be the basic stuff of politics. (1985: 6)

The major theoretical paradigms that Skocpol referred to are the pluralist and

neopluralist traditions. It is important to note, however, that these two schools are not the only "society-centered" approaches in the field of empirical democratic theory. Social corporatism and Marxism also have societal constraints as their main core assumptions. Nonetheless, pluralism has dominated this field in the United States, at least until the 1960s.


The main theoretical focus of pluralism is group competition and its effects on

policy-making. According to pluralists, the state does not play any role other than

summing up the preferences of the competing groups and allocating resources according

to these calculations. In Nordlinger's words:

As portrayed by pluralism, civil society is made up of a plethora of
diverse, fluctuating, competing groups of individuals with shared interests.
Many effective political resources are available to them: numbers, votes,
organization, money, expertise, information, social status, a positive public
image, access to mass media, and control over economic resources. The
liberal state is guided and constrained by the resource-weighted parallelogram of societal preferences; its authoritative actions are the "resultant of effective
access by various interests." With voters choosing the elected officials they
prefer, elected officials competing for popular support and tailoring their
public actions to calculations of electoral advantage, elected and appointed
officials being lobbied at numerous access points and pressured at permeable
decision sites, and public policies and group cooperation in implementing them
under these conditions the state consistently responds to societal demands.
Public officials are even wary of acting contrary to the interests of "potential
groups." (1981: 151-2)

Social preferences emerge as a resultant of vectors of group preferences, which

Nordlinger (1981) calls the resource-weightedparallelogram of social preferences.

Pluralists are concerned with problems of allocation, rather than those of rule and control.

Many pluralists ignore the role and preferences of public officials and the structure of the

state. The function of the state, according to their perspective, is to assure that the game

of politics among the social actors is played fairly, and that the results of these games

become public policies (Krasner, 1984).

This utopian view of the political system helps one understand the classical

approach to social movements discussed in the previous chapter. If the political system


were really open, democratic, and unbiased, as pluralists claimed, there would be no rational explanation for noninstitutional political action.

A variation of the pluralist tradition is neopluralism, or elite theory. Neopluralists are less sanguine about the equality of access to the political system. In their view, competing groups do not have the same power and influence on public officials. A novelty in this approach is the concept of "effective groups." These are groups that know which resources are effective to influence officials and to constrain the state. Moreover, these groups control such resources and use them at the expense of competing groups. Effective groups are usually those well-organized wealthy groups with clear interests in specific policy issues. According to Nordlinger, due to the effectiveness of their resources, as well as a liberal bias of public official towards organized groups, "the state enfranchises 'the most interested and best organized' business, labor, agricultural, trade and professional associations as the exclusive representatives of particular social interests"(1 981:45).

This interpretation clarifies the emphasis of the resource mobilization approach on accumulation of resources and strategic alliances with more powerful actors. Following an elitist interpretation of the state, resource mobilization theorists concluded that the only way for small groups to reap political gains was by leveraging their power to the point that they become acceptably organized and, therefore, competitive.

This bias towards organized and powerful groups caused a general disillusionment and demoralization in U.S. politics during the 1960s and 1970s. As a consequence, scholars lost their confidence in the pluralist model, leaving room for the more realistic neopluralist perspective and other less society-centered approaches (Almond, 1988).


Also, abrupt social and economic changes after World War II contributed to the decline of society-centered models. Keynesian macroeconomic management, in which state expenditures are central, became the norm in advanced societies, even in the United States. Also, the end of colonial empires gave birth to myriad new states, which soon realized that the pathway of Western liberal democracy was unfeasible for them in the short term (Skocpol 1985).

As a consequence of these changes, the state came to the forefront of politics and the pluralist model came under criticism, accused of "societal reductionism." Almond (1988) vehemently refuses this accusation. He claims that critics of pluralism fail to consider important pluralist works in their literature reviews, overlooking studies that recognize some degree of government autonomy. After reviewing part of this pluralist literature, he concludes that:

Though this literature does by and large stress the importance of
interest and pressure groups in policy-making, it clearly does not support the
reductionist thesis. Autonomous government agencies are present and
important throughout this literature. The pluralist "paradigm" is not the onesided one of Skocpol, Krasner, and others, but rather a two-directional one
with the state influencing the society as well as the society influencing the
state. (1988: 866)

Some pluralists do emphasize the role and leadership of government officials in policy-making. However, their role has been depicted as constrained by "societal forces, commanding resources that are derived from a variety of public and private sources" (Krasner, 1984: 229). He also points out that this pluralist view gained momentum in the 1960s, as part as the "behavioral revolution" that swept the social sciences in the United States. Behavioralism rejected the primacy of formal institutions and rules in the political


process and emphasized individual preferences. Society substituted for formal legalism, which prevailed in American political science during the first half of the twentieth century. For behavioralists, political outcomes are determined by preferences and power capabilities of societal actors.

The discussion of behavioralism is important to put Nordlinger's work in

perspective. On the one hand, he supports the idea of state autonomy, challenging the view that the state responds solely to societal stimuli. On the other hand, his view of state and society preferences as the resolution of parallelograms does not challenge behavioralist assumptions. He considers the state as a group of individuals, rather than a conjunction of individuals, institutions, and structure. According to him, individuals make decisions; offices and agencies do not. He defines the state in a Weberian manner:

It refers to all those individuals who occupy offices that authorize
them, and them alone, to make and apply decisions that are binding upon any and all segments of society. Quite simply, the state is made up of and limited
to those individuals who are endowed with society-wide decision-making
authority. (1981: 11)

In sum, Nordlinger defines state as government and bureaucracy. By doing so, he claims to be avoiding misleading anthropomorphization and reification fallacies, that is, treating organizations as human beings. In his view, institutions and organizations cannot make decisions or express preferences; only human s can. Therefore, he responds to pluralists by granting autonomy to the state parallelogram of preferences, but he does not accept the state as anything other than actors that have personal preferences and control certain amounts of resources.


Other supporters of the state-centered perspective reject this definition of state as government. Stepan (1978: xii) argues that "the state must be considered as more than the government." It is the continuous administrative, legal, bureaucratic and coercive systems that attempt not only to structure relationships between civil society and public authority in a polity but also to structure many crucial relationships within civil society as well." In a comparative study of development in Brazil, Zaire, and East Asian newly industrialized countries, Peter Evans (1989) concludes that the differences in their achievements were related to the very structure of the state. Developmental states, he claims, "[A]re embedded institutions historically as well as social structurally" (1989: 583).

In spite of his timid definition of the state, Nordlinger's idea of autonomy does

represent a significant departure from pluralism. He believes that state officials are able to translate their resource-weighted preferences in what he call "authoritative actions." Autonomy is, therefore, related to "the degree to which any and all state preferences coincide with authoritative actions" (1981: 25). He defines authoritative actions as:

[B]oth the public policies that have been adopted and those that have
not, including those whose consideration has not even been placed on the formal or informal agenda. Authoritative actions are not limited to highly
visible ones that have a great impact upon many individuals, such as the
enactment of new social welfare legislation; authoritative interpretations and applications of narrowly focused regulations are also included. (Nordlinger,
1981: 13)

Autonomy and capability are distinct concepts. A state can be strong but not autonomous and vice-versa. Capability refers to the resources available to the state (Nordlinger, 1981). Nordlinger accepts the relational notion of state strength (in relation to society). States that act on their preferences with social support are stronger than those


that must overcome societal opposition. Also, he claims that states that can change the societal preferences are also strong, even if they cannot act until they change the parallelogram of societal preferences. Another important feature of his theory is that autonomy is not restricted to the instances when society opposes the state. Autonomy can also reflect decisions made by the state without societal resistance. If society neither opposes nor supports some policy initiative, the state is, by definition, acting autonomously by translating its internal preferences into authoritative actions.

According to the state and society preferences and the authoritative actions of the state, Nordlinger builds a typology of state autonomy in a simple matrix, reproduced in Figure 2.1. The top half of the chart shows the cases when the state acts according to its preferences. The lower box in the left, the lack of state autonomy. In this case, societal preferences prevail.

Type I autonomy explanations are those in which the state and society preferences are divergent, but society does not succeed in imposing its preferences on the state. Accordingly, the state does not feel constrained by this divergence to the extent that it decided not to translate its preferences into authoritative actions.

Those situations in which the state does not face opposition from society, and do translate its preferences into authoritative actions form the type III autonomy. Authors who build their perspective on state autonomy within the Marxist tradition usually ignore this kind of autonomy. Those authors consider the state to be autonomous only when it makes decisions against the preferences of the dominant classes. An example of this approach is Skocpol's States and Social Revolutions. In this work, she discusses how


states act against the long term interests of the dominant class when its position in the international system is challenged (Skocpol, 1979).

State and society preferences are: Divergent Nondivergent


preferences and authoritative actions are:


Figure 2.1: Types of state autonomy, according to Nordlinger (1981).

The discussion of type II autonomy is perhaps the most important contribution in On the Autonomy of the Democratic State. In these cases, the state preferences prevail and the state also uses its resources to change societal preferences and to hide itself from societal constraints. In type II cases of state autonomy, "public officials purposefully bring about a shift in the societal parallelogram of resource-weighted" (Nordlinger, 1981: 99). Marxists and pluralists alike have ignored, in general, three important factors. The first one is the regular availability of resources and opportunities for public officials to change

Type 11
state Type III
autonomy state
explanations autonomy


Societal constraint explanations


Type I
State Autonomy Explanations


societal preferences. Second, they overlook the willingness of state officials to use those resources and capabilities to enhance their autonomy. Finally, they do not seriously consider the likelihood of success of state officials when they attempt to interfere with those societal preferences (Nordlinger, 1981).

In cases of type II autonomy, public officials can deploy four strategies to transform divergent societal preferences into non-divergent ones. The first strategy consists of persuading actors with divergent preferences. If officials cannot change their preferences, they at least try to instill enough uncertainty to make their opponents withdraw from the policy-making process. The second strategy focuses on avoiding or limiting the use of resources by societal forces to hinder state autonomy. A third strategy is aimed at bringing indifferent societal actors to side with the state, as an effort to counter-weight the resources of those who oppose the state preferences. Whereas this strategy focuses on calling the attention of certain social groups to the policy issue under discussion, the fourth strategy consists of increasing the level and weight of resources employed by the supporting groups to offset the resources of those who oppose the state.

This discussion of state autonomy contradicts the pluralist view that the state is neutral by introducing the preferences of state officials in the analysis of policy making. Even the less naive elitist view of the state is challenged, for it denies the neutrality of the state but does not account for the role of state officials in shaping social preferences. Of course the idea of autonomy poses a great challenge to the resource mobilization approach to social movements. If state officials can change the preferences of social actors, any study of social movements need to pay attention to state preferences, as well as to the preferences of the movement itself. In the following chapter I discuss how state


preferences deeply affected the recent colonization of Amazonia and became the very target of the social movements that appeared in the 1970s and later.

Although Nordlinger presents an interesting typology of state autonomy and

makes a great contribution by discussing the mechanisms of state manipulation of societal interests, he does not address institutional development. On the Autonomy of the Democratic State does not explain how states became what they are. The book is concerned with the impact of state upon society. Other political scientists (Stepan, 1978; Evans, 1989), however, have addressed the institutional and evolutionary aspect of the state. Both approaches are complementary, rather than antagonistic. As Skocpol (1985), a leading figure in what became known as the "bringing the state back in" movement, stresses:

The point is that policies different from those demanded by societal
actors will be produced. The most basic research task for those interested in
state autonomy surely is to explore why, when, and how such distinctive
policies are fashioned by states. (1985: 15)

And then she comments on the importance of the institutional aspects of the state:

This second approach may be called "Tocquevillian," because Alexis
de Tocqueville applied it masterfully in his studies. In this perspective, states matter not only because of the goal-oriented activities of state officials. They
matter because their organizational configurations, along with their overall patterns of activity, affect political culture, encourage some kinds of group
formation and collective political actions (but not others), and make possible
the raising of certain political issues (but not others). (1985: 21)

Krasner divides the literature on the state that appeared in the late 1970s and 1980s according to its temporal dimensions. There are studies on the autonomy of the state and studies on the congruity of the state and its environment. The issue of autonomy


is a temporally static one, in which the state is viewed as an exogenous variable. Congruity, on the other hand, has a temporal dynamic. Researchers following this trend focus on changes in the domestic and international environment and their impacts on the state structure. Besides that, those studies also pay attention to the effect of the state on its environment during periods of "normality." The main question researchers seeking to understand this temporal dynamic ask is "how do institutional structures change in response to alterations in domestic and international environments and then in subsequent time periods influence these environments" (Krasner, 1984: 224).

This idea of congruity is relevant to students of social movements who follow the political opportunity approach. However, the literature on state autonomy goes beyond the literature on social movements in the sense that it presents a two-way interaction between state and society. Whereas students of political opportunities look at how changes in the state configuration affect the emergence of social movements, students of the state are also concerned with the possibility that social movements can also reshape the state. The obvious exception is Charles Tilly, who has studied long-term interactions between state and society (Tilly, 1978).

Marxist Approaches to the State

Critics of society-centered approaches criticize pluralism and Marxism equally. The Marxist view of the state commonly has been reduced and simplified to what some authors call instrumentalism. According to this view, the state is the executive committee of the bourgeoisie, or a tool or instrument of ruling-class domination. As Carnoy (1984) points out, Marx himself never produced a single coherent theory of politics, and thus, of


the state, in his writings. His political ideas surface in his discussions on political economy;

however, they were spread in so many sources and vague enough to leave room for

subsequent interpretations by his followers and critics. Of course some authors would

dispute this description of Marx on the state. Thomas is one of them, to whom:

It does not follow from this that, all on its own, ruling-class theory
constitutes a viable theory of the state. But to say that Marx had no real theory of
the state because of the evident shortcomings of ruling-class theory would be a
serious mistake [.] there is in Marx's writings a theory that is viable. It is viable not because of Marx's various dogmatic utterances but despite them. It has to be elicited, drawn out and retrieved from its writings, not read into them. (1994: 1011)1

Indeed, Carnoy acknowledges that, despite the possibilities of diverse

interpretations, there are some fundamental Marxist ideas (in Marx's own works as well as

in most of his followers') that frame the debate on Marx and state theory:

First, Marx viewed the material conditions of a society as the basis of its
social structure and of human consciousness. The state is a product of the relations
of production, not from the general development of the human mind or from the collective of men's will. In Marx conception, it is impossible to separate human
interaction in one part of society from interaction in another: the human
consciousness that guides and even determines these individual relations is the
product of the material conditions-the way things are produced, distributed, and
consumed. (Carnoy, 1984: 46)

At this point, Marx is at odds with Hegel, who considered the state as a rational

structure that involves a just, ethical relationship of harmony among elements of society.

The Hegelian state is eternal, not shaped by historical conditions, as opposed to the

1 Although claiming that there is a theory of the state in Marx's writings, Thomas concedes that these works are rather full of "complexity, ambivalence, and equivocation" (1994, 11).


Marxist state, which is a product of the historical and material development of a society (Carnoy, 1984).

A second common trait of the Marxist interpretation of the state is the rejection of the idea that the state represents the common good. Rather, Marx identified it as the expression of the class structure and class relations, which are, again, conditioned by the mode of production. Contrary to Classical and Liberal views (such as the ones by Hegel, Locke, Rousseau, Hobbes, and Smith) the state does not represent the "social collectivity," nor does it stand above the interests of particular groups. According to Marx and other Marxists, the state is not above class struggles; rather, it is deeply engaged in them.

This second characteristic of Marxist thought has not always been so clear in Marx's own works. In his early writings, Marx, deeply influenced by the political and social conditions in Germany, agreed with Hegel's idea of a "communal state," noting that it had to be a democratic state. He conceded that the state could have some autonomy from society. However, he did not accept Hegel' s view that this was a universal trait of the state. He based this early acceptance of a state that was not class-bounded on historical conditions, as Carnoy explains:

For young Marx, then, the state had some life of his own, separated from
civil society, having its own particular interests. Given conditions in Germany at the time, it is not unusual that Marx should see the state in this way: there was a
separation of the state on the one hand and a rising civil society of the bourgeoisie
on the other. The state was not an instrument of the bourgeoisie. In absolutist Prussia, the state was still in the hands of a precapitalist ruling class, with very
different social values from those of the increasingly powerful bourgeoisie.(1984:


Draper (1977), quoted in Camoy (1984), explains that:

This Prussian state was indeed forced to exercise control over the
aristocracy itself, it was no longer the simple feudal state, but the Beamtenstaat of
absolute monarchy-the state of the functionaries, who had to keep a rein on all
classes in order to keep the growing antagonisms from pulling society apart.
(1977: 169)

According to Jessop, Marx's early works depict the state as "an irrational abstract system of political domination which denies the social nature of man and alienates him from genuine involvement in public life" (1982: 7). His early view fell short of anything resembling an instrumentalist theory of the state. Contrary to Hegel, who considered the bureaucracy in the modem state a universal class seeking to realize the universal interest of society, Marx argued that universal interest is a pure abstraction. Moreover, Marx pointed out that if, on the one hand, the state elite represents private interests, on the other hand, the bureaucracy may become another interest group among many others that seek to influence state decisions (Jessop, 1982).

An early version of a class theory of the state was presented by Engels in his The Condition of the Working Class in England. In this study, he argues that the English bourgeoisie secures control over the working class by controlling the state. Engels and Marx later elaborated this new approach in The German Ideology and in the Manifesto of the Communist Party. It is important to stress, however, that rather than a coherent theory of the state, what those authors presented was a "complex array of ideas and arguments unified (if at all) through their common concern with the general framework of historical materialism" (Jessop, 1982).


The struggles within the modern capitalist state, according to Marx and Engels,

represent the struggles within society itself, particularly class struggles. The state mediates contradictions between individuals and the community; as the bourgeoisie dominates the community due to its control over the means of production, it also dominates the state. The state, therefore, is not a result of a class plot. Rather, it reflects the balance of power within the community: "Hence, the state does not exist owing to the ruling will, but the state which arises from the material mode of life of individuals has also the form of a ruling will" (Marx and Engels, 1964: 358, quoted in Carnoy, 1984: 48-49).

Also, as evidence that Marx did not take the instrumentalist view to extremes,

Jessop (1982) shows how he stressed the effects of different types of state on the balance of class forces in different societies. Marx developed this argument in The Civil War in France, wherein he explains that "the state is a system of political domination whose effectiveness is to be found in its institutional structure as much as in the social categories, fractions, or classes that control it" (Jessop, 1982: 27). In other words, Marx grants some autonomy to the state instead of simply collapsing it into society.

Even though the state was not a class plot, it was not a structure of universal

representation of class interests. Most states pretended to be representative of all classes, but in fact protected the interests of the capitalist class, and this false idea of universal representation was, according to this Marxist view, a central ideological function of the state. According to this Marxist tradition, collective action will emerge when the subordinate classes are in fully developed contradiction with its antagonists. In the case of western societies, this would be possible after capitalists had deprived workers of their tools and driven them into large factories. In this situation, unions could provide


organizational resources, and raise the class consciousness necessary for mobilization. It is worth noting that Marx and Engels considered liberal democracy and the liberal state as an institution and a structure that could be useful to the workers if they captured them; hence the idea that capitalism paves the road to socialism. As workers often failed to rise against their oppressors, Marx blamed them for developing a "false consciousness," instead of a true class consciousness. He ignored splits within the labor movement as well as tacit support some sectors gave to capitalists in exchange for welfare benefits and integration into capitalist democracy (Tarrow, 1994;Jenkins, 1995).

If someone is to blame for the mistaken Marxist instrumentalist view of the state, Lenin is probably the right person. For Lenin, the state is an organ of class rule, whose main function is to moderate class conflict by controlling the lower classes. According to him, petty bourgeois ideologists distorted this fundamental principle of Marxism by depicting the state as an organ of class conciliation. He claims that class conciliation is impossible, and the dominant class uses the state to perpetuate class oppression. In his view, in a society without class struggle, there would be no need for the state. In fact, in this kind of society, the state could not exist altogether. Contrary to Marx and Engels, Lenin (1929) considered that the capitalist state had no use for the lower classes:

If the state is the product of the irreconcilable character of class
antagonisms, if it is a force standing above society and "separating itself gradually
from it," then it is clear that the liberation of the oppressed class is impossible
without a violent revolution, and without the destruction of the machinery of the
state power, which has been created by the governing class and in which this
"separation" is embodied. (1929: 116)2

2 This was Lenin's second attack on the "betrayers" of Marx and Engels directed to Kautsky. Contrary to the "petty bourgeois ideologists", Kautsky acknowledged that the state was a mechanism of class domination, but he also distorted Marxism by denying that revolutionary forces had to destroy this mechanism.


Thomas (1994) claims that the Russian revolution deeply affected Lenin's works, particularly The State and Revolution, wherein he exposed his main ideas. Critics of Lenin argue that he was overwhelmed by the events surrounding him when he wrote the book and, as a result, it is more of a revolutionary blueprint than a piece of scholarship:

He [Lenin] was of course able to back-reference what were largely
organizational imperatives by invoking phrases drawn from the available writings
of Marx and Engels, but such invocations were necessarily tendentious and almost
incantatory. Marx and Engels were not invoked in a spirit of calm, scholarly
objectivity; concepts and phrases were torn from their contexts and made to carry
talismanic force under the impress of events. (Thomas, 1994, 123)

In conclusion, Lenin does not leave room for any form of state autonomy by affirming that the capitalist state apparatus is of no use for the working class, "for the capitalist state apparatus is organized structurally in form and content to serve the capitalist class, and cannot be taken over by the working class to serve its ends"(Camoy, 1984, 59). A new society, one with no class struggle (due to the suppression of the bourgeoisie), will require and imply a new state. The dictatorship of the proletariat, in this case, substitutes for the dictatorship of the bourgeoisie.

Italian thinker Antonio Gramsci is commonly credited with being the first scholar to systematize a Marxist science of political action. As opposed to Marx, who stressed the primacy of political economy in shaping society, Gramsci gave primacy to politics, granting it some autonomy from the economy. An underlying preoccupation in Gramsci's works is the locus of power in the political system. To some extent, he agreed with Lenin by considering that a significant amount of power lies in the state, where it takes the form of domination. However, he also acknowledged that power also lies in a network of social institutions, which, sometimes, have no direct connection to the state (Thomas, 1994).


Marx and Engels considered the state as a product of society. Society is the

decisive element, whereas state is the subordinate one. In Marxist terms, society is the structure, the state is the superstructure, and both coexist in a dialectical relation. The essence of the structure are the relations of production, which produce a juridical and institutional order: the state (superstructure). Gramsci had a different view. In Carnoy's words:

For both Marx and Gramsci, civil society is the key factor in understanding
capitalist development, but for Marx civil society is structure (relations in
production). For Gramsci, on the other hand, it is superstructure that represents
the active and positive factor in historical development; it is the complex of
ideological and cultural relations, the spiritual and intellectual life, and the political
expression of those relations that become the focus of analysts rather than
structure.(1984: 66)

These cultural and intellectual relations that Carnoy referred to are at the core of Gramsci's theory and are the factor that most differentiates it from Leninism. The state is not simply a tool for class domination. In reality, if the bourgeoisie is able to spread its ideas, norms, and values throughout society, it will not need to resort to violence to control the subordinate classes. The cultural and intellectual relations among classes, in this case, lead to the internalization of class domination. This is what Gramsci called hegemony.

Beyond a simple intellectual response to the utilitarian view of the state, the concept of hegemony reflects institutional and historical differences among European countries. Whereas Lenin was particularly influenced by what he saw in tsarist Russia, Gramsci's built his theory upon the experiences of Western European representative democracies. Gramsci did not substitute hegemony and consent for coercion and


domination. Rather, he argued that both modes of social control could exist, at the same time, in different states.

The preponderance of civil society over the state in the West can be
equated with the predominance of 'hegemony' over 'coercion' as the fundamental mode of bourgeois power in advanced capitalism. Since hegemony pertains to civil society, and civil society prevails over the state, it is the cultural ascendancy of the
ruling class that essentially ensures the stability of the capitalist order. For in Gramsci's usage here, hegemony means the ideological subordination of the
working class by the bourgeoisie, which enable it to rule by consent. (Anderson,
1977: 45)

This definition emphasizes the role of society in the political process, to the

detriment of the state, in a clearly Marxist fashion. Later, Gramsci admitted that society was not the only locus of hegemony; rather, state and society were responsible for ideological control. In this new formulation, however, he confused the attributions of state and society by arguing that these two spheres of public life share both force and consent. Repression, or the use of force, however, is restricted to the state, not an attribute of a normal society. As an attempt to fix this analytical problem, Gramsci finally merged state and society in a larger unity with hegemonic and coercive powers'. Anderson claims that due to this change "the very distinction between state and society is canceled. which undermines any scientific attempt to define the specificity of bourgeois democracy in the West" (1977: 34).

Although largely criticized, this lack of differentiation between state and society

had some supporters. Althusser was a radical follower of Gramsci in his claim that society

3 These definitions appeared in The Prison Notebooks. Anderson suggests that the second definition wherein both state and society have repressive powers may have been the result of Gramsci's living in Italy during the fascist years, when militias were common and the state lost the monopoly of use of force.


and state cannot be distinguished one from another4. In his view, any institution, private or public, is part of the state ideological apparatus: family, church, education, trade unions, parties, newspapers, etc. This definition made a distinction between a representative democracy and a fascist regime not only impossible, but also unnecessary (Anderson, 1977). Rather than individuals or institutions, what determines this lack of boundaries between state and society is ideology. Such a determinism comes as no surprise considering Althusser was a radical structuralist (which Gramsci, of course, was not).

With this theory of ideology, Althusser constructs a mechanism by which
individuals willingly subject themselves to an ideology (Gramsci's hegemonic "consensus"), and it is this subjugation that defines them in the society itself
Inherent in the ideology is the necessary ignorance of the reality that the ideology
represents, and this reality is, in the last resort, the reproduction of the relations of
production and the relations deriving from them (Carnoy, 1984: 92).

The installation of the state ideology is the real stake in the class struggle. It is not by seizing the state that a class establishes its supremacy; rather, it is by installing its ideological state apparatus. Anderson argues that Althusser may have been deeply influenced by the Chinese Cultural Revolution, which, in last instance, was aimed at destroying the existing ideological state apparatus and installing a new one.

If Althusser, on the one hand, did not differentiate private institutions and

organizations from the state, Gramsci, despite his late view of state and society as one single entity, never went that far. He did not follow the Chinese revolution from a comfortable distance as Althusser did; rather, he lived in and was a victim of a fascist state. He realized, drawing on his own life experience that there was a clear difference

4 Whereas Gramsci's work focused on political process, Althusser idealized the state apparatus.


between a bourgeois state with relative influence on society and a fascist state that controlled society by means of force.

Despite the absence of a clear and definitive gramscian interpretation of the relation between state and society, students of the state usually consider his societycentered view as predominant. Some students of social movements, as we have seen in the previous chapter, definitely consider the realm of superstructure as the main battlefield of social movements, at least the so called new social movements (Alvarez et al., 1998). Even though followers of this trend do not go as far as Althusser to erase the borders between state and society, they do, at least implicitly, claim that changes in society at large are as important, or perhaps even more important, than changes in the "visible" aspects of the polity, namely the state. In fact, the students of politics of culture do reject the Althusserian structural determinism embedded in his view of ideology. According to this Althusserian interpretation, the relations of production (infrastructure in the classical Marxist tradition) are the source of ideology, which shapes the superstructure (the state). Under this perspective, material conditions determine cultural politics, stripping it from the autonomy that some students of social movements believe it has (Dagnino, 1998). If the relationship between state and society is the opposite, as Gramsci argues, then culture, not material relations, shapes politics.

Later developments in the Marxist tradition, however, returned to the state as a relatively independent actor. Poulantzas (1978) acknowledges the repressive and ideological aspects of the capitalist state, but he rejects the reduction of the state to these two dimensions (force and ideology). Poulantzas argues that this Althusserian view, despite its radicalism, in fact masks a more critical view of the state.


The exercise of power is split between two groups: the repressive and the
ideological state apparatuses. This apportionment diminishes the specificity of the
economic state apparatus by dissolving it into the various repressive and
ideological apparatuses; it thus prevents us from locating the state network in
which the power of the hegemonic fraction of the bourgeoisie is essentially
concentrated; and it obscures the character of the modalities required to transform
this economic apparatus in the transition to socialism as distinct from those required to transform the repressive and ideological apparatuses.(Poulantzas,
1978: 33)

In other words, Poulantzas argues that the state has a more active role in producing the conditions that enable capitalism to reproduce itself. The state has a stronger link to the dominant classes than other Marxists supposed. For Poulantzas (1978: 77), it is the state that "represents and organizes the long-term interests of a power block, which is comprised of several class fractions[. ] and which sometimes embraces dominant classes issuing from other modes of production that are present in the capitalist social formation." As the dominant classes are not homogeneous, the state has relative autonomy from them, so that it can ensure the organization of the interests of the bourgeoisie under the leadership of monopoly capital (the leading fraction of the dominant class).

Another important neo-Marxist who addressed state theory was Offe. He

considered the state as a crisis manager-an institution that emerged in capitalist society in response to the crises that naturally arise from the contradictions of capitalism. These crises were the result of "increasing socialization of labor (the incorporation of wage labor) and continuing private appropriation (surplus extraction by capitalists)"(Camoy, 1984:13 1). These crises trigger both market responses (emergence of oligopolies and monopolies) and state intervention.


The state best represents the ruling class when it does not act upon the will of this class, but rather upon its own routines and formal structures. Specific interests of a sector of the bourgeoisie tend to deepen the cleavages within it, also diminishing the state ability to intervene and solve crises. The state does not have a selective bias that purposefully excludes non bourgeois citizens from its structure, nor a structural constraint that impedes it from adopting strategies that hurt the interests of the capitalist class. For Offe (1973), what drives the capitalist state to work to the benefit of the dominant class is its need to survive and to reproduce itself.

The capitalist state, by definition, does not have direct control over the

accumulation process, which is the sole responsibility of private actors. As the state depends on private accumulation to survive, by means of collecting taxes on profits and wages, it has, therefore, the authority and the mandate to create the conditions that will promote private accumulation. Nevertheless, even though capitalist interests underly state strategies, representatives of the capitalist class do not take direct control of the state apparatus not only because of class cleavages, as discussed above, but also because of a concern with legitimacy. As Carnoy points out:

Since the personnel of the state apparatus do not have a power base of
their own, they need some mandate for action derived from some alternative
source of power. This mandate for action must come from the concept of the state
representing the common and general interests of society as whole.(1984: 134)

In Offe's own words:

This is to say that the state can only function as a capitalist state by
appealing to symbols and sources of support that conceal its nature as a capitalist


state; the existence of a capitalist state presupposes the systematic denial of its
nature as a capitalist state. (1973: 127)

Of course these two imperatives of the state are relatively contradictory, imposing limits one on the other. At the same time, the capitalist state has to facilitate investment by private actors, as well as ensuring that the working class is employable and employed. Block (1977) calls this linkage between the state and the capitalist class the main structural mechanism of class control. He notes, however that there are other subsidiary mechanisms that make the state serve capitalist interests. This structural mechanism, according to Przeworski and Wallerstein (1988), is false in a static sense, but true in a dynamic one. According to them, any distribution of consumption between workers and capital owners is compatible with continual private investment once an appropriate set of taxes and transfers is in place. However, the anticipation by capitalists of any changes in this distribution cause them to reduce investments and slow down the economy.

The first subsidiary mechanism is direct pressure on the state, through influence channels. Although the capitalist class lacks a collective consciousness, individual capitalists have a very clear idea of what their short-term interests are, and they will press the state to realize those interests. These influence channels include the recruitment of ruling-class members into public offices, the formation of lobbying groups, campaign financing, bribing public officials, and others. In many instances, this kind of pressure runs against the reproduction of capitalism, and it is counter-weighted by the structural mechanism (Block, 1977). A second subsidiary mechanism is cultural hegemony. Block follows Gramsci in his definition: "the widespread acceptance of certain unwritten rules about what is and what is not legitimate state activity" (Block, 1977: 14).A government


that violates these rules runs the risks of losing popularity and even legitimacy, which, of course, discourages a democratic government from adopting anti-capitalist measures.

The idea of structural dependence of the state on capital implies a high degree of state autonomy. Even though the state is subject to hegemonic ideas and has to protect the interests of the capitalist class, at the same time it adopts policies that run counter to these interests and ideas. In the following chapter I show that the Brazilian authoritarian state conforms to this Marxist discussion of state autonomy. At some points the state acted with a high degree of autonomy from society. In other instances, representatives of the capitalist class had ample access to the state decision making structure. In both cases the military government sought to promote private accumulation of capital. If the state has autonomy to veto particular interests of the most powerful class, it certainly has an overwhelming power over the lower ones, especially if their demands pose any apparent challenges to economic performance. This autonomy has obvious consequences to the social movements approaches. Contrary to what followers of resource mobilization claim, strategic alliances with more powerful groups will not necessarily result in incorporation into politics. If the state perceives these alliances as a threat to economic performance, it will use its power to neutralize them.

Corporatism and the State

Corporatism is an applied form of what Alfred Stepan (1978) calls an organicstatist approach to the state. This approach, he argues, is "not as textually and historically specific as classic Marxism or Liberal Pluralism," yet has an intellectual lineage that runs "through Aristotle, Roman Law, medieval natural law and into contemporary Catholic


social philosophy" (1978, 27). An interesting characteristic of this organic statist tradition is that it is built upon a normative, rather than descriptive or analytical view of the state.

Liberal pluralist views of the state are descriptive in the sense that they evolve from what liberals consider a basic trait of human beings, namely individualism and selfinterest. Accordingly, Marxists develop their theories from a descriptive analysis of the dominant mode of production and its consequential class struggles. Organic-statist interpretations, on the other hand, do not start from accounts of what society is, but rather from what society ought to be.

What society ought to be, according to prominent organic-statist thinkers, is a well integrated and functional system wherein different parts coexist harmonically and perform differentiated tasks. The earliest intellectual conceptualization of this kind of social integration is St. Paul's letter to the Christians at Corinth, when he suggested that society should be an integration of separate parts, much like the human body (Wiarda, 1997). This integration is achieved through the insertion of humans in communities a basic feature of all texts that influence corporatist approaches.

Aristotle had a similar view of mankind. According to him, man's nature can only be fulfilled within a community. Men have, consequently, a natural tendency to form political associations. The best way to functionally organize society, therefore, would be along the "natural" existing groups and divisions of society. Participation in the "polis" is a fundamental part of human life, even prior to the family and to the individual (Stepan, 1978). A third influence on the organic statist approach comes from ancient Rome. As in Greece, political participation was a fundamental characteristic of society, but whereas the Greek practiced direct citizen participation, the Romans had a system of indirect


representation. This system granted representation to the existing social groups, such as the military, professional associations, and religious institutions, often organized under colegios. In a typical corporatist way, the groups entitled to participate in the political system were monopolistic in character. In other words, one and only one group represented each meaningful segment of society. Moreover, all these groups operated under the tutelage of the state.

The collapse of the Roman empire buried corporatist practices and theories for

centuries. It was not until the late middle ages that groups began organizing again, due to economic changes that brought about professional guilds, more structured military, and religious orders. The decay of the feudal system opened new opportunities and needs for corporate representation. The eventual prevalence of the absolutist state, however, soon closed the opportunities for substantive participation.

It was not until the late eighteenth century that the corporate statist ideology reemerged within the realm of political philosophy. German romanticists rescued this tradition and strongly rejected the individualism that was characteristic of the French and American revolutions. Individuals could not claim their part in the formation of the will of the state directly, but only as a member of a Stand. The romanticists manifested their support to the central tenets of the organic state approach by stressing the importance of a strong central authority the Monarch but not an absolutist one. Checks and balances on the authority of the monarch are not in the hands of a parliament or any other structure of direct representation. Rather, the most capable members of the "Estates" should surround the king and represent their groups' interests. A confrontation between the monarch and his subjects would certainly lead to this undesirable form of government.


Representation through "Estates" was the only effective way to mediate these relations (Landauer, 1983).

It is interesting that up to this point, lineage and economic function were closely correlated. The proponents of organic state, therefore, simply accepted the traditional modes of voting and representation, based on hereditary status. As social and economic development unfolded, economic function and hereditary status became increasingly separated. The corporatist literature reflected this change by gradually focusing on economic function and abandoning the concept of hereditary rights (Landauer, 1983).

As romanticism started fading, so did theoretical studies of corporatism. Economic development, however, made social problems more acute and, by the 1860s, proposals of corporatist remedies to these problems reemerged. Also, this corporatist revival was a response to two emerging ideologies of last century: Marxism and liberalism. Following the communitarian tradition, corporatist arrangements stressed the importance of the group, rather than individual, as in the liberal ideology. The organization of politics according to traditional social divisions would bring peace instead of the class struggles that Marxists predicted.

Again, the Catholic church was a strong source of corporatist ideology. In 1881, Pope Leo XIII commissioned a study of corporatism and how it related to the Catholic doctrine. A group of thinkers and of theologians met at the University of Freiburg in 1884, and produced the first definition of corporatism:

A system of social organization that has as its base the grouping of
men according to the community of their natural interests and social functions
and, as true and proper organs of the state they direct and coordinate labor
and capital in the common interest. (Wiarda, 1997)


This work had a direct influence on Pope Leo XIII's Rerum Novarum encyclical which, for the first time in the catholic doctrine, recognized the rights and importance of working men. This document advocated the dignity of labor conditions, as well as the incorporation of the working force into politics. Of course, ideally, this incorporation should follow corporatist guidelines. Indeed, labor initially organized following these corporatist tenets, under the tutelage of the Catholic and Protestant churches. By the turn of the century, labor unions had spread throughout Europe, competing with socialist unions. Also, other forms of religious associations mushroomed: youth associations, business organizations, women's associations, and even Christian democratic parties (Wiarda, 1997).

It is important to stress that, in the beginning of the century, corporatism had a

progressive side, for it advocated the incorporation of laborers into politics. On the other hand, it also was conservative in the sense that the state should dictate the scope and depth of this incorporation. Moreover, the conservative character of corporatism is evident when it is compared with the growing Marxist and anarchist trends of the labor movement, in the early decades of the twentieth century.

If corporatism had been mostly restricted to the realm of philosophy and ideology until the early twentieth century, it became a substantive mode of political practice during the inter-war period. At this point, it is important to retrieve Stepan's distinction between corporatism and organic statism. The latter refers to a normative approach to the state, as we have discussed above. The former is a system of government that tries to follow the guidelines or the rationale of organic statism. To the same extent that no state has ever come close to a pure form of communism, corporatist states have never approximated


pure organic-statism (Stepan, 1978). Despite the deviations from the theoretical guidelines of organic statism, there are some characteristics that define a corporatist state. According to Schmitter:

Corporatism can be defined as a system of interest representation in
which the constituent units are organized into a limited number of singular,
compulsory, noncompetitive, hierarchically ordered and functionally
differentiated categories, recognized or licensed (if not created) by the state
and granted a deliberate representational monopoly within their respective categories in exchange for observing certain controls on their selection of
leaders and articulation of demands and supports. (1974: 93-94)

Stepan (1978) argues that this organization of society in functional groups, under the tutelage of the state, is a reaction by elites to crises. By controlling the state and reshaping the relationships between sectors of civil society and the state, these elites overcome crises in an orderly way. Although Stepan's idea is not unanimous among students of corporatism, it has its appeal as an explanation of emergence of corporatist regimes in the first half of the twentieth century. The failure of parliamentary liberal regimes before World War I, the Bolshevik revolution in 1917, the economic crisis of the 1930s, and the escalation before World War 11 were all critical elements that may have affected the emergence of corporatist experiences throughout the world. Be that as it may, Stepan's point is that corporatism is a "policy output," rather than a "social input" (1978: 47).

Although Italy under Mussolini was a defacto authoritarian regime, it started as a corporatist experiment5. Mussolini himself changed the electoral laws to concentrate Stepan rejects the common misinterpretation of fascism as a form of corporatism. Corporatist regimes are responsive and, to some extent, accountable to those groups that are recognized by the state. Totalitarian regimes do not have this commitment to any group. Totalitarianist regimes do not integrate functional societal sectors, but rather try to


power in the hands of the majority party, and then passed other constitutional arrangements that ultimately substituted a corporate organization for the party system. Accordingly, Hitler's Germany also fell short of a corporatist regime and exhibited the same vicious traits of Italian fascism. The idea of building corporatist structures, however, also emerged out of the economics of the war efforts.

In France, Henri Petain also sought a corporatist formula to solve economic problems; however, his plans were frustrated by World War 11. Austria was another country that experienced corporatism as a remedy against crises. In this case, a constitutional crisis led the country to a civil war and then to the corporatist regime of Kurt Von Schuschnigg, between 1934 and 1938. In Portugal, Antonio de Oliveira Salazar, influenced by his social Catholic background, also resorted to corporatist arrangements to overcome economic crisis, although with limited success (Landauer, 1983).

Portugal is an interesting case in point, because, as Landauer stresses, Catholicism had a deep effect on Salazar. However, the adoption of a corporatist system in Portugal should not support the idea that Catholic societies are prone to this form of government. This cultural determinism is exactly the underpinning of another approach to corporatism. Authors like Wiarda (1973, 1997) do not explain corporatism as a proactive policyresponse to crises, but rather as a historical continuity based on "medieval institutional conceptions of harmony, the common good, and the state" (Stepan, 1978: 52).

Wiarda (1997) claims that after extensive research in Latin America, Portugal, and Spain, during the 1960s, he

subjugate them. Also, corporatism seeks a peaceful integration of functional groups, whereas fascism has been associated with popular control through paramilitary groups and


[F]ound many of the same persistent and characteristic features,
namely organic, integralist, top-down, statist, and corporatist institutions and
practices that stood at considerable odds with the generally prevailing
orthodoxy that proclaimed liberalism and pluralism as the model of
developmentalism. After finding corporatism, organic statism,
authoritarianism, patrimonialism, and these other features so omnipresent in so
many regimes, the author determined that there was a distinct model of
corporatism or organic statist (the terms were often used interchangeably)
development "out there" that was particularly characteristic of Iberia and
Latin America. Moreover, this model stood in marked contrast to the liberalpluralist and Marxist models, both of which had proved inadequate for a full
understanding of Latin America. (1997: 58)

Stepan (1978) refutes this cultural/historical explanation on empirical grounds. Historical continuity is a fallacy, for corporatism was in fact stronger in Latin America during the 1970s than it was in the 1950s, or in the liberal period of the late nineteenth century. Also, he argues that the common Iberian heritage fails to explain why countries like Colombia and Ecuador, where the Iberian-Catholic ethos is as strong as anywhere else in the continent, did not develop corporatist structures and practices.

Wiarda (1997) argues that corporatist culture is a source of resistance against change in state structure and institutions in Latin American societies. Stepan (1978) responds to this conclusion by pointing out that abrupt changes led by elites were exactly what brought corporatist arrangements to power in contemporary Latin America.

Rogowski and Wasserspring (1971) present a more society-centered variant of this approach to corporatism. They share with Marxists and pluralists the idea that society determines politics. Societies that are segmented will, in their view, naturally produce corporatist regimes. The case that is closest to their theoretical definition is Mexico. Stepan, however, shows how the state and political elites reshaped and even formed

even a romanticized view of violence (Stepan, 1978).


groups in Mexico, rather than being a product of societal forces. Historical experiences in Mexico and elsewhere show the role of the state or state leaders in marking the boundaries between groups and selecting and tutoring their leaders (Stepan, 1978).

Schmitter (1974) recognizes the importance of societal forces in some corporatist regimes. He divides corporatism in two types: societal corporatism and state corporatism. Societal corporatism refers to systems that are democratic and produce corporatist arrangements in a contractual way. These arrangements are characteristic of some industrialized countries that have "relative autonomous multilayered territorial units, open competitive electoral processes and party systems; ideologically varied, coalitionally based executive authorities" (1974: 105). Sweden, Switzerland, and Denmark are some examples of these countries. State corporatist systems are usually present in countries with a more centralized structure of power, with less competitive or non-existent electoral systems, and ideologically exclusive executive authorities. In other words, in authoritarian regimes.

Societal corporatism, or democratic corporatism, or yet consociationism, has its origins in historical conditions that facilitated bargaining and the formation of coalition, rather than in cultural characteristics of a given society. Katzenstein (1985) argues that European states successfully adopted democratic corporatist measures in the 1930s and 1940s as a direct response to the crisis of the inter-war period.

Democratic corporatism is a policy aimed at achieving a stable, bourgeoisdominant regime. These changes are a response to economic crises and have as a rationale the need to include subordinate classes and status groups into the political process. This


kind of response is typical of societies where pluralism slowly decays. Conversely, state corporatism is, according to Schmitter:

[C]losely associated with the necessity to enforce 'social peace,' not
by coopting and incorporating, but by repressing and excluding the
autonomous articulation of class demands in a situation where the bourgeoisie
is too weak, internally divide, externally dependent and/or short of resources
to respond effectively and legitimately to these demands within the framework
of the liberal democratic state. (1974: 108)

Corporatism, therefore, in any of its variations, is a political system antagonistic to pluralism. Both state or social corporatist models rule out the possibility of universal access to politics and group competition. Social movements that operate within these systems cannot be fully explained in terms of resources and alliances, following the resource mobilization approach.

From this discussion, I do not conclude that the resource mobilization approach fits only pluralist states. Rather, what it suggests is that studies of social movements must account for the political context in which these movements are embedded. The underlying assumptions of resource mobilization cannot be taken for granted even in pluralist regimes, much less in authoritarian, corporatist, or transitory ones. Whether and to what extent the state is autonomous is a relevant empirical question. Rather than subscribing to any model, I, therefore, try to assess the Brazilian state on empirical grounds. I do have to say, however, that I reject Nordlinger's methodological individualism and follow an institutional approach to the state.

It is important to stress again that corporatism is, as Stepan (1978) puts it, a

"policy output." Contrary to Marxism and pluralism, it is not an analytical, but rather a normative approach to the state. Therefore it is not paradoxical that Marxist theories may



help explain a state that adopted corporatist policies, as it is certainly the case of the Brazilian military state between 1964 and 1985.



The Brazilian authoritarian regime promoted a spectacular occupation of

Amazonia between 1966 and 1985. The rationale behind this endeavor was promoting economic growth and ensuring national security. Although these goals underlined the whole period of occupation, the means to accomplish them varied during the military rule. These changes were the result of an interplay of three variables: the results of the governmental projects (measured against their own standards security and economic growth), the interests of the regime's allies, and the interests of the peasantry in the region (which usually opposed the state and confronted its capitalist allies). During the 21 years that the military stayed in power, there were four phases of occupation of Amazonia demarcated by different economic and security policies.

In a first phase of state-led development of the Amazonia, peasants did not

represent a threat to the capitalist interests in the region. Migration into Amazonia was spontaneous and slow. Squatters occupied small tracts of land in few sites close to the Belem-Brasilia highway, in Southern Para. As long as the government managed to keep indian groups inside restricted reservations, the capitalist groups met little competition for access to land. Contractors supplied cheap labor to do the work necessary to start the new enterprises. Also, the investors pressed the state to speed up the approval of their projects and to build the transportation infrastructure necessary for their operations. This phase


started in 1966, with a program know as Operaqdo Amaz6nia, and ended in 1970, when the government started promoting immigration and public colonization of the region.

The construction of the road system and a massive influx of migrants to the region marked the second phase of frontier expansion. The sudden increase in the population changed the relations between the capitalist groups and peasantry. Competition for land became fierce, and the entrepreneurs changed their demands on the state to protect their interests. At this point, the expansion of the road network, contrary to what happened in the first phase, would most likely hurt their interests, as more roads would increase migration and competition for land. Also, squatters and land speculators usually accompanied the peasantry moving into the region, making the distribution of land and titles even more difficult.

For the owners of large enterprises, in this phase, more important than

appropriation of material resources was the establishment of institutional mechanisms to protect them: "Emphasis shifted from the 'economical and physical' occupation of the area to a concern for the 'legal and juridical' occupation of state lands by private investors" (Pompermayer, 1984: 435). Particularly important for the owners of large estates was gaining titles to their land before migrants settled in the vicinities. If the government did not respond immediately to their claims, they argued that the success of the existing projects, as well as the development of new ones, would be compromised. This phase lasted for five years (1970-74). After that, the government shifted policies again.

In the third phase, the Amazonian entrepreneurs sought to consolidate their

business and retain large tracts of land. The greatest challenge to their interests was the contingent of workers they once brought to the region to open their ranches and then


fired. By that time, there was a large population of rural workers struggling in the Amazon, particularly in southern Pardi. With help from the Catholic Church, these workers started invading large ranches frequently, posing another threat to the continuation of capitalist enterprises in the region. This time the investors called on the state to guarantee the security of their investments, which was usually accomplished by use of force. There was no single event that marked the end of this phase; however, changes in policy became more noticeable in the late 1970s.

In the fourth and last phase of recent occupation of Amazonia the state at the same time withdrew support for private enterprises and tried to tighten social control of the peasantry. In this phase, SUDAM cut support to failing projects as the state faced economic trouble. On the other hand, the military intensified its presence in southern Par/i, where land conflict was becoming a pressing issue. Even after the transition to a civilian government, the military remained in command of Amazonian development and security policies. At the same time, however, the subordinated classes came to organize in the region, and slowly got an active voice in policy making. This phase outlived the authoritarian government and still endures.

This division of the recent history of occupation of Amazonia is arbitrary. These four phases roughly correspond to the role of the state as promoters of economic growth in the region. As we will see later, the fourth phase could be easily divided in several other phases. The withdraw of governmental support was not the only significant characteristic of this phase, and perhaps not even the most important one. During this phase that the


military crated GETAT1, which represented an unprecedented instance of state autonomy. At the same time, the authoritarian government lost popular support and its capacity to repress mobilization and free speech. In 1985 the military returned power to civilians. At the same time the military tried to maintain its control over Amazonia, by successfully lobbying for the creation of a huge occupation proj ect along Amazonia's Northwest border Calha Norte, which was created in 1986. In 1988 Brazilian congress approved a new constitution that granted subordinated groups and minorities new rights. In 1992 Brazil hosted the United Nations' Conference on Environment and Development UNCED. All these were critical events that marked this fourth phase of colonization. Some of them were particularly relevant to social movements and will be discussed later in this chapter.

I start this chapter with an overview of the military intervention that began with a coup in 1964. In this section I explain the general doctrine that guided the authoritarian regime. After that, I briefly discuss the differences between previous attempts to develop Amazonia and the military intervention. The next sections are discussions on the four aforementioned phases of occupation of Amazonia. In the end of the chapter, I discuss some theoretical aspects of the recent history of state intervention in the region. Since there are many relevant phases, names, and dates in this recent history of Amazonia, I summarized some of them in the last page of this chapter (Figure 3.1).

1 Grupo Executivo de Terras do Araguaia e Tocantins (Executive Group for the Araguaia Tocantins Lands).


The Authoritarian Regime

On March 31, 1964, the Brazilian military overthrew the constitutional

government in a quick coup that met little resistance. By the time of the coup, the country was experiencing economic turmoil, which included hyperinflation and negative growth rates. Political and ideological cleavages had deepened, particularly after the inauguration of vice-president Jodo Goulart's term in office, following the resignation of president Jdnio Quadros, in August, 1961. Goulart's support from leftist groups and his populist rhetoric and policies fed the fear of communism and left some elites uneasy, including the military.

The military intervention occurred after a period of increasing polarization of politics. Although the radical left was small and weak, the threat of a leftist revolution scared the upper classes and mobilized right wing extremists: "[T]he economic stakes of political brinkmanship, as perceived by the propertied classes, had risen very high by 1964" (Skidmore, 1973: 4. Italics as in the original text)

Most of the military and parts of the upper classes were also discontented with the economic performance of the country and the economic policies of Jodo Goulart's government. In 1964, Brazil was experiencing a negative growth of per capita income, as well as out of control inflation. Also, the country was on the verge of default on its foreign debt. Solving these problems would require unpopular orthodox economic measures, which Goulart's populist government was not willing to implement. Even if the government wanted to adopt hard measures, there was little reason to believe that it could successfully do that within the existing democratic institutions, or so thought the military.


Economic adjustment was, therefore, a rationale for military intervention. Due to its disregard to popular support and democratic procedures, the authoritarian regime should have been able to carry out the reforms, and, after that, return power to civilians. The first military government (1964-67), headed by President Humberto de Alencar Castello Branco, engaged in an unpopular fight against inflation endorsed by the International Monetary Fund. Two respected liberal economists Otivio Gouveia de Bulh6es and Roberto Campos led a team of technocrats, in this battle. First, they cut increases in currency supply, slashed public expenditure, froze the government-set minimum wage, cut subsidies and protection to domestic businesses, and increased the prices of public services. Still more impressive was the government's capacity to go against interests on both sides of the political spectrum. On the one hand, it overcame the economic nationalism that had been enshrined during Goulart's term in office. On the other hand, the government got rid of tax shelters and loopholes that prevented it from collecting income tax from a large part of the population (mostly higher classes)2.

As a result of these policies, the country was first engulfed by a deep recession that was particularly damaging to the industrialized state of Sao Paulo. By the end of Castello Branco's term, however, the technocrats had brought inflation and federal deficit down, and reestablished appropriate conditions for resuming economic growth3.

2 Between 1967-69, the number of individual income tax payers grew from 470,000 to 4,000,000 (Skidmore, 1973).

3 In the first quarter of 1964 annual inflation was above 100%. It went down to 66% in 1965, 410% in 1966, and 310% in 1967. The federal deficit declined from 4.2 % of GDP in 1963 to 1. 1% in 1966 (Skidmore, 1973).