The Agrarian
Question

To understand Brazil's rural unionism, we need to distinguish among the movement's four different facets, which are not always sufficiently disentangled: opportunities for action, ability to act, ideology and program, as well as strategy. An account of Brazil's rural unionism inevitably confronts the state's central involvement in the unions' operations. All Brazilian trade unions operate within a structure highly controlled by the central state, a structure that has not been fundamentally reformed since the 1930s, despite several changes of national regime.

Anthony W. Pereira

Around 8:00 A.M. on 15 October 1988, a pickup truck filled with about fifteen striking rural workers arrived at the entrance of a sugar plantation near the town of Nazaré da Mata, Pernambuco. Led by José Patrocínio Gomes, then president of the local rural workers' union, the strikers clambered out of the pickup and came face-to-face with the plantation owner and his administrator. While many strikers clutched a cutting implement (foice) used in the fields, the administrator, a stout, dark man in a blue safari jacket and porkpie hat, brandished a double-barreled shotgun. Gomes explained to the owner and the administrator that the strikers wanted to enter the plantation and talk to the workers in the fields, in an effort to persuade them to stop work. The owner would have none of it.

A tense standoff ensued. Cowed by the shotgun, the strikers huddled by the side of the pickup as Gomes complained about the weapon. Several strikers shouted for everyone to calm down. A striker grabbed the union's megaphone and spoke to the workers, saying that he didn't want violence and that the strike committee only wanted to talk to them as comrades(companheiros). Slowly, the strikers formed a semicircle around the owner and the administrator, who issued threats in a low, tense voice. After several minutes, the situation slowly shifted. The strikers became angry. One said, "Well, he can kill one of us, but then we will certainly kill him." A young striker in a red hat and blue jacket quipped, "Who's afraid of a shotgun?" A heated discussion with the owner ensued, with much shouting and anger on the part of the workers. The owner insisted that the strike, then in its sixth day, was not obligatory and that anyone who wanted to work for him would work. On several occasions Gomes raised his arms and asked for silence so that he could negotiate with the owner, but the shouting continued. Almost unnoticed, the administrator and his shotgun began retreating, first by edging behind the owner, then by facing the group further down the road, and finally by turning his back on the assembly and walking away.

With the departure of the administrator, the plantation belonged to the workers. Led by Gomes, the strikers strode past the owner and into the fields, getting the strikebreaking cane cutters—many of them children—to stop work. They handed out leaflets that explained why they were on strike. The man with the megaphone made a short speech in favor of the strike, saying that he and his comrades were not there to pit worker against worker but to gain a better salary for all, and that it was better for them to go home and come back after the strike, when they could be assured of better wages and working conditions. With calls of "go home" to the workers in the fields, the strikers drove away in their pickup.

The incident at the plantation near Nazaré da Mata was part of an annual war of maneuver between landowners and labor in Pernambuco's sugar zone. It reflected the ability of local unions to defy the armed force of the landowners through organization, a little courage, and the power of numbers. And it reflected how the sugar workers' unions were able to force themselves back onto the local political stage after years of military rule in which such bold actions would not have been permitted. The strikers were thus capitalizing on the democratic transition, but they were also part of the transition—their actions were taken into account, if not always appreciated, by landlords, state officials, and other powerful figures involved in the regime change.

The unionists of Nazaré da Mata were part of a labor movement that represented a quarter of a million workers in Pernambuco's sugar zone (also called the forested zone or zona da mata). This strike was the eighth since their path-breaking action of 1979, when they became the first rural unions in the country to challenge the military regime with a strike. During the 1980s, their repeated mobilizations and strikes formed part of the great surge of popular pressure that influenced the Brazilian military's decision to abandon direct control over the executive in 1985. Together with the metalworkers and bank employees, they formed one of the pillars of the revitalized Brazilian labor movement whose militancy helped to democratize politics in the 1980s.

How was this militant labor movement created, despite the vigilance and repression of the Brazilian military regime? Why was it primarily a working-class rather than a peasant movement? What was the relationship between the rural trade unions and the transition to democracy that took place in Brazil in the 1980s? This book sets out to answer these questions.

The most significant characteristic of Brazil's new rural trade unionism was that it was primarily a movement of workers and small farmers rather than peasants. Thus it was fundamentally different from the agrarian movement of Brazil's preauthoritarian period. In the early 1960s, under a populist regime, rapid and intense mobilization of the rural labor force in the Brazilian northeast had alarmed landowners, political leaders and the U.S. government. The initial mobilization had been conducted by Peasant Leagues, organizations of the rural poor that invoked the image of the Cuban revolution, engaged in direct actions such as land occupation, and demanded a redistribution of land "by law or by force," without compensation to large landowners. This mobilization made the agrarian question, especially in the northeast, one of the most important national issues of the day.

The Peasant Leagues were voluntary organizations with no formal link to the state; a large repertoire of collective action; many urban, middle-class leaders; and a multiclass membership with little or no central coordination of what were essentially local struggles. In response to the Leagues, trade unions concerned primarily with wage increases were founded by activists from the Communist Party and the Catholic Church. The actions of both the Leagues and the unions were condemned as subversive and unacceptable by most landowners, who supported the post 1964 military regime's policy of abolishing the Peasant Leagues and purging and repressing the trade unions.

The rural labor movement that reemerged as a political force in the 1980s was very different from its 1960s counterpart. It was formally included within the state's corporatist labor structure, had a narrow repertoire of collective action, was led mainly by professional union officials with peasant and worker backgrounds, and was centrally coordinated at the national level by CONTAG (Confederação Nacional dos Trabalhadores na Agricultura), the National Confederation of Agricultural Workers, and at the state level by state federations. While some CONTAG advisors had backgrounds in the Communist Party, the role of that party as an organized force within the movement was relatively small. The influence of the Church, on the other hand, which had been a major force in rural unionism in the 1960s, was now strongest outside the official union structure in the new social movements of the landless.

In the 1980s, CONTAG mobilized thousands of workers in legal strikes every year, including those organized by the Federation of Agricultural Workers of Pernambuco, FETAPE (Federação dos Trabalhadores na Agricultura de Pernambuco). It did pressure the federal government for land reform, but by parliamentary means. In 1988 CONTAG was the largest confederation in the Brazilian labor system, with twenty-two state federations, 2747 unions, and almost 10 million members on its books. This is significant growth for an organization that was founded only in 1963 by a few hundred newly formed unions. In Pernambuco, the number of unions in the sugar zone increased from thirty-two in 1964 to forty-five in 1988.

An adequate analysis of the Brazilian rural labor movement must therefore try to account both for the movement's reemergence in the 1980s, and for the changes in its form, size, capacities, and activities. In addition, it must explain the decline in the importance of "peasant" questions, and the emergence of new conflicts centered on the demands of rural workers and small farmers.

Rather than try to analyze the movement in toto—a very difficult task in a continent-sized country such as Brazil, with vastly heterogeneous conditions in agriculture—this book focuses on unions in Pernambuco, one of the leading segments of Brazil's rural labor movement. Pernambuco is significant because its 1979 strike made its federation an "initiator" that many other state rural labor federations later copied. Pernambuco unions also created a dynasty within the national confederation, CONTAG, in much the same way that metalworkers' unions from the ABC region of São Paulo dominated the national labor central, the CUT (Central Única dos Trabalhadores, or Unified Workers' Central). This dynasty is symbolized by the fact that from 1968 to 1989, the president of CONTAG was José Francisco da Silva, a leader from Pernambuco's sugarcane fields. Da Silva and other CONTAG leaders directly coordinated the eight strikes in the Pernambuco sugar zone between 1979 and 1988, proclaiming that the strikes were a model for rural unionists around the country. Pernambuco thus represents an "exemplary" rural labor movement whose high degree of organizational capacity calls for explanation.

Rural Labor and
Democratic Transitions

Authoritarian regimes in almost fifty countries around the world have collapsed or been reformed since 1970. These transitions have resulted in democracy when the new regime, at the very least, conducts regular, open, and competitive elections of top government officials; makes bureaucratic agencies of the state responsible to the elected parliament and/or chief executive; and guarantees a minimum of civil liberties to its citizens. The literature on democratic transitions and their consolidation is primarily concerned with explaining why and how, at the macro level, the institutional reforms necessary for democracy are brought about and preserved.

Another aspect of regime transition has been undertheorized in this literature. That is: as regime change takes place, how do citizens at the local level claim the rights embodied by the new national institutions?

Seen from this angle, transitions cannot adequately be reduced to single, holistic narratives that homogenize the political experience of citizens within national territories. The historical reality is that both formal and effective democratic rights are usually extended variably—by gender, race, ethnicity, economic sector, region, and so on—within a nation over time. Of particular relevance to this study, the democratic rights most important to labor—to freely organize unions, bargain collectively, and strike—are not explicitly included in many macroanalyses of democratic change. They form part of a broad array of popular demands that inevitably accompany the macroinstitutional reforms of regime transition.

Because of their concern with national institutional reform, many scholars share a rather narrow view of the role of popular pressures in democratic transitions. Either popular organizations and movements are seen as unimportant, or their only relevance is their capacity and inclination to upset the carefully constructed transition, once it has been installed from above. Thus Adam Przeworski in Democracy and the Market writes: "It seems as if an almost complete docility and patience on the part of organized workers are needed for a democratic transformation to succeed." In a similar vein, Terry Karl in Dilemmas of Democratization in Latin America asserts:

No stable political democracy has resulted from regime transitions in which mass actors have gained control, even momentarily, over traditional ruling classes.... Thus far, the most frequently encountered types of transition, and the ones which have most often resulted in the implantation of a political democracy, are "transitions from above."

These and other similar comments prompted Charles Tilly to decry the "top-down models, instrumental and constructivist approaches to democratization, and short-run analyses" that currently prevail in the literature.

One reason for discounting the role of popular pressures in regime change in this literature is that the time frame analyzed is usually very short. Most studies using the notion of a democratic transition implicitly assume that the main causes of democratic change can be found in events occurring immediately before the establishment of a new regime. The focus is thus on elite bargains and pacts, the institutional compromises of high national politics, fissures within the authoritarian regime, and the like. Explanations of the formation of democratic regimes that concentrate on these events tend to stress rational-actor models and the individual decisions of the most powerful members of society. The notion of "contingency"—in which outcomes "depend less on objective conditions than subjective rules surrounding strategic choices" is a key component of these explanations.

Transitions to democracy within this perspective are seen primarily as the result of a series of conscious, planned, strategic actions. A previous generation of scholars' "bottom-up, deterministic, long-term theories" of democracy, with their focus on history, culture, and economic and political structures have been laid to one side. What really seems to count in contemporary analysis is the inclination of political elites to make pacts; in Samuel Huntington's words, in The Third Wave: "Whether democracy in fact falters or is sustained will depend primarily on the extent to which political leaders wish to maintain it and are willing to pay the costs of doing so".

Such an approach is one-sided. Popular organizations that have been shaped, at least in part, by long-term structural and institutional change have often played a significant role in the collapse or reformation of authoritarian regimes. In turn, regime change has sometimes involved concessions to popular movements. In the case of trade unions, authoritarian governments are said to "demobilize" labor, while civilian ones are expected to allow for labor's "remobilization." Institutions previously barred to workers' representatives are frequently opened up to them; some decentralization of decision making can take place; and labor's demands have occupied a more prominent place on some post-transition government agendas. Yet the role of labor in regime transitions has not won the attention that it deserves.

However, acknowledging that labor can play an important role in democratic transitions and that authoritarian regimes "demobilize" labor, while democratic ones allow for its "remobilization," is only a starting point—and a corrective to the elitism and lack of historical perspective in much theorizing about transitions. The demobilization-remobilization dichotomy oversimplifies state-labor relations. In fact, controlled labor mobilization of some type was allowed by most Latin American authoritarian regimes except in the most repressive periods. The dichotomy draws attention away from continuous political processes across and within regimes. It does little to help us understand changes in the organizational capacities and demands of labor movements, or how these might vary regionally, sectorally, and over time.

Regarding the case of rural labor in Brazil, therefore, the social science literature seems to contain two major gaps. The first is in theorizing about the relationship between rural labor and transitions. The vast literature on the role of the peasantry in revolution has not generated an equally rich body of work on the role of peasants (and other rural actors) in negotiated changes of regime. The second gap lies in historical approaches that look at long-term change, not just short-term interactions; these are rather thin on the ground. As Karl comments, what is needed is a

path-dependent approach which clarifies how broad structural changes shape particular regime transitions in ways that may be especially conducive to (or especially obstructive of) democratization . . . [and how] such structural changes become embodied in political institutions and rules which subsequently mold the preferences and capacities of individuals during and after regime changes.

Some scholars have adopted such an approach, showing, for example, how state practices and capabilities changed over a long period, whether during the authoritarian regime or before its foundation, to shape the contours of the democratic transition. This book shares this perspective, arguing that the political institutions regulating rural labor during the Brazilian regime transition embodied a massive structural change: the decline of the peasantry.

How did this change occur? In Latin America, the rural working class did grow in agro-export regions, but much more slowly than in either manufacturing or mining sectors. This is because plantations had previously coexisted uneasily with peasant plots. Plantation owners in northeast Brazil and elsewhere extracted labor first from indigenous people, then from African slaves, and then from the "reconstituted peasantries" that formed in and around the plantations after the abolition of slavery. Peasant labor was extracted through sharecropping and other nonmonetary arrangements; wage labor tended to come quite late, if at all, in the process of capitalist development. When it did come, wage labor was frequently mixed with precapitalist relations of production, creating a "peasantariat," or labor force with mixed peasant and proletarian characteristics.

To the extent that a rural working class emerged free of dependent relations with landlords involving land, this new class was likely to demand inclusion in the political system. The rural working class has a strong interest in gaining the democratic rights to assemble, speak out, bargain collectively, go on strike, and vote, because with these rights it has a much better chance of being able to improve wages and working conditions. Rural workers are also, in many cases, much better able to act on this desire for political freedom than peasants, tied as the latter often are to landlords. However, when rural workers surmount the high barriers to collective action in the countryside and engage in autonomous organization, they face the likelihood of employer repression. According to a recent comparative study of democratization, in the cases reviewed "the landed upper classes which were dependent on a large supply of cheap labor were the most consistently anti-democratic force. In turn, landlord repression may engender a messianic or maximalist political orientation among rural workers, as was the case with Brazil's Peasant Leagues. In the context of labor-intensive agriculture and incipient working-class formation, therefore, mobilization from below is likely to lead in the short term to revolutionary or authoritarian violence rather than to democracy.

The tensions generated by agricultural modernization, however, can enhance the prospects for agrarian democracy. This dynamic is analyzed clearly by Jeffrey Paige in Agrarian Revolution. When hacienda agriculture undergoes capitalist transformation, he argues, labor declines as a percentage of landowners' costs, making it easier for landlords to accept bargaining with labor organizations and the possibility of wage increases. Rural workers, for their part, begin to resemble their industrial counterparts. As patron-client ties are replaced by the cash nexus of the marketplace, workers move away from the plantation and the supervision of the landlord and break out of their former isolation. Their scope for organization increases, and in the absence of repression they do not adopt maximalist positions. Eventually, they will lose their peasant attachments to land and tend to develop a militant but reformist labor movement that negotiates for higher wages and better working conditions. A democratic class pact, rather than revolution or authoritarianism, is likely to emerge. Rural workers will join their urban counterparts in pushing for democratic political inclusion.

However, such an outcome is not guaranteed—Paige's analysis should be interpreted probabilistically rather than as a universal law. The structure of landholding, the history of past conflicts, and the cultural predispositions ("frames") of the local population, among other factors, influence the likelihood of a democratic class pact in the countryside. More specifically, the organization of agricultural production also varies significantly from one crop to another. Therefore, this book draws on an eclectic variety of works in political economy that focus on the distinctiveness of collective action within economic sectors. Understanding the sector—in this case, sugar—within which the unions are embedded is very important, because the sector imposes additional, specific structural constraints on unions' collective action aside from those that exist in the broader social setting.

My argument is that something like the trajectory from hacienda to capitalist agriculture described by Paige occurred in the Pernambuco sugar zone in the 1960s and 1970s, and that this facilitated the emergence of a labor movement capable of negotiating a local democratic class pact during the national regime transition of the 1980s. Brazil as a whole evolved from a rural to an urban nation in this period; while the majority of the population lived in the countryside in 1960, by 1980 almost 70 percent was urban. Pernambuco urbanized as well. In its sugar zone, traditional landlord-peasant arrangements such as tenancy, sharecropping, and moradia (in which peasants had the right to cultivate a small plot on the landlord's estate in return for several days of labor per week) declined. Meanwhile, larger, more efficient mills increasingly dominated production, leading to a reduction in the total number of mills. Wage employment increased, as did the size of the rural working class. In pursuit of their own interests, rural labor organizations played a vital role in demanding the inclusion of all those excluded by policies of conservative modernization in the countryside. Agrarian transformation laid the basis for local acceptance of the democratic transition in the 1980s.

Rural Labor
in Brazil

To understand Brazil's rural unionism, we need to distinguish among the movement's four different facets, which are not always sufficiently disentangled: opportunities for action, capacities to act, ideology and program, and strategy. The argument made here is that opportunities for action are explained mainly by macropolitical factors at the level of the national regime; capacities to act are influenced primarily by large historical processes—changes in social structure and the institutional form of representative institutions; and ideology and strategy are explained by the members', and especially the leaders', learning through experience and through local traditions and institutions. Part II of this book mainly concerns the capacity to act, while Part III covers ideology and strategy. Some background information on opportunities for action is given throughout the text.

An account of Brazil's rural unionism inevitably confronts the state's central involvement to the unions' operations. State agriculture policies influence the kinds of popular protest that arise in the countryside. Reciprocally, the state's responses to popular protest determine the political consequences of that protest. More specifically, all Brazilian trade unions operate within a structure highly controlled by the central state, a structure that has not been fundamentally reformed since the 1930s, despite several changes of national regime. The key features of this union structure, discussed in more detail in chapter 5, are a unitary system of representation or unicidade (only one union allowed per category of worker in a given territory); a mandatory union tax (imposto sindical or contribuição sindica1) deducted from employees' pay and passed on to unions by the state; and a hierarchy of union organizations, ranging from the local union in a município (equivalent to a county) through state federations and national confederations. Like most other Latin American union structures, Brazil's is a mandatory closed shop enforced through a state-controlled, hierarchical, corporatist, and bureaucratic structure. Thus, the political opportunities available to unions fluctuate with changes in the composition of the national regime. Furthermore, rural unions operate in agriculture, a portion of the economy highly regulated by the state in Brazil, as it is in many other countries.

The state in Brazil and in Latin America more generally has promoted the modernization of agriculture, which has contributed to some of the most important structural changes in the region over the last forty years. Agricultural modernization refers to the application of scientific knowledge and capitalist rationality to the ancient art of farming. It involves the transformation of agriculture from a subsistence activity or a productive system in which profits derive primarily from land and labor to one in which profits accrue mainly from capital investment and in which land or labor productivity, or both, are increased through the use of new inputs.

The most dramatic increase in agricultural modernization in the advanced capitalist countries took place after World War II. There, rural labor became relatively scarce, due to the growth of urban industry. This forced agricultural producers to adopt capital-intensive forms of production. Chemical fertilizers and pesticides, new planting and harvesting machines, and genetically engineered plants facilitated a quantum leap in agricultural productivity. These new technologies caused relatively little social dislocation because they were introduced in societies whose rural labor force had already become largely urban.

In many parts of the Third World, these new agricultural technologies were rapidly adopted in very different social environments. The new technology spread, despite the relative scarcity of capital and abundance of labor. This occurred for various reasons, including transnational pressures, state subsidies to large estates and skewed prices, a highly unequal distribution of land, the desire of large farmers to insulate themselves from rural union organizing, and the prestige associated with modern forms of agriculture. Unlike the case in advanced capitalist countries, the labor surplus often produced by high-technology, capital-intensive agriculture in the Third World could not readily be absorbed by the industrial and service sectors. The agricultural revolution in the Third World thus contributed to, rather than mitigated, social strains. Peasants were uprooted from the land without a corresponding increase in available urban employment.

The modernization of agriculture has been part of the larger transformation of rural societies into predominantly urban ones in Latin America. This change has been rapid and dramatic: a process that took half a century in many advanced capitalist countries occurred in much of Latin America in a mere twenty years. The traditional Latin American large estate (latifúndio) has become, in many places, a capital-intensive enterprise. At the same time, the peasantry has been greatly reduced, while a large class has emerged of mostly seasonal rural workers and small farmers, alongside an even larger class of former peasants in the cities who are self-employed or employed in the informal sector.

This de-peasantization is an economic, political, and cultural phenomenon. De-peasantization is a more accurate term than proletarianization because not all former peasants become full-time wage earners. In the economic sphere, rural and urban labor markets become more tightly integrated, and increased transportation and communication links bridge the urban-rural gap. Politically, de-peasantization is usually accompanied by the incorporation of former peasants by the state, but not as subsistence cultivators. Culturally, the distance between the rural and urban areas is reduced by the spread of schooling and literacy, television and radio, frequent migration from rural areas to cities and other rural areas and back again, and the relocation of many large landowners to cities.

The structural change induced by agricultural modernization did not lead invariably to the formation of only two, highly polarized classes. Agrarian growth may produce the immiseration of all but a handful of rich farmers, but if applied indiscriminately and mechanically as a kind of universal law of capitalist development in agriculture, this formula misses important, subtle forms of differentiation in the countryside. Even in the Brazilian northeast, a region considered economically backward, expansion and improved production techniques in the 1970s led not just to the dislocation of many, but to increased opportunities for small landowners (and an expansion in their numbers), more permanent jobs for laborers, and the growth of an intermediate strata of technical and supervisory personnel such as agronomists, chemists, economists, and machine operators. These various categories of people had divergent interests. Development in agriculture can thus lead to complex patterns of differentiation that are not as simple as class polarization between a small group of capitalists and a mass of destitute workers.

This study argues that it is precisely because the rural labor force was not a mass, but rather a highly heterogeneous and changing collection of interests and identities, that the challenges faced by the rural labor movement during the democratic transition were so complex. While capitalist modernization mainly benefited large landowners and was detrimental to the poor, many of whom lost access to land, people who fell between these two poles in the production system frequently benefited from change as well. In the process, trade unions that had been organized around the single identity of the "rural worker" became arenas of contention between small farmers, wage laborers, and a declining number of subsistence cultivators.

The complexity of structural change in the countryside is related to the unions' problems of representation. Structural change can lead to conditions that favor union formation. But the types of collective action that unions engage in, the ideologies they espouse, and the demands they make, are not necessarily determined by social structure. For these issues, cultural and political factors—local cultural frames, the background of the leadership, its alliances and values, political opportunities, the nature of state institutions—may have significant weight. Disentangling union foundation, capacity, strategy, and ideology seems crucial if one is to avoid teleological reasoning that attaches inevitable political consequences to economic change.

With these distinctions in mind, we can see that CONTAG's claim to represent all progressive forces in the countryside was problematic. The confederation could not be all things to all its members, and a stable, national political consensus did not exist within its member unions. Given the great heterogeneity of interests represented by the unions, and the shifting opportunities they faced, it was not always clear what a "progressive" political line was at any point. There were certainly internal debates about this. For example, the labor central CUT (Central Única dos Trabalhadores, or Unified Workers' Central), founded in 1983, frequently clashed with CONTAG leaders over the confederation's leadership style, position on union reform, and tactics on the land question. There were also criticisms of the unions from without, on the part of new social movements of the landless such as the Church's Pastoral Land Commission (CPT, Comissão Pastoral da Terra) and the Landless Workers' Movement (MST, Movimento dos Trabalhadores Rurais Sem Terra). All of these organizations claimed to speak for the rural poor against the large landholders and existing state policies, but their analyses and tactics differed markedly, revealing the contested nature of representation in this period.

This article was excerpted from The End of the Peasantry—The Rural Labor Movement in Northeast Brazil, 1961-1988 by Anthony W. Pereira, University of Pittsburgh Press