The development of capitalist agriculture has had a wide variety of effects upon pre-existing agrarian societies in Latin America. The forms it has assumed have in part been determined by variations in such factors as climate, ecology, demographic structure and history, ethnic patterns, and land tenure. The central theme of this volume is that such variations, whilst important in explaining localized phenomena, should essentially be seen as aspects of a basic process of change from one mode of production to another in the rural sector.

This is not of course a new idea, and indeed a number of writers, especially in the fields of economic history and social anthropology, have already dealt with many of the questions of particular relevance to the theme of this volume. Broadly speaking, their various approaches can be divided into three different levels of generalization. First, there are those works principally concerned with identifying the general mode of production in contemporary Latin American agriculture, in which the argument has centred around the question of whether the social organization of agriculture is essentially feudal or capitalist. Secondly, there is a more limited amount of theoretical discussion relating to the different types of agricultural enterprise to be found in Latin America, in which the principal distinction is drawn between the hacienda and the plantation. Finally, there is a very considerable body of literature dealing with types of peasantry and rural labour, where discussion concentrates upon the role-structure of rural economic life.

The period from the 1880s to the 1920s was the ‘Golden Age’ of coffee in western São Paulo. Santos, Sao Paulo's only major port, had replaced Rio de Janeiro as the world's most important coffee export centre by 1894. Santos’ share of the world coffee market rose from less than 25 per cent in the late 1880s to 50 per cent by the first years of the twentieth century, and it continued to supply more than half the world's coffee until after the First World War. In absolute terms the annual production of the Santos coffee zone increased from an average of under two million bags during the last years of slavery to more than eight million bags in the first five years of this century, and continued at high levels thereafter. The transition from slave to free labour, the rapid expansion of coffee cultivation, and massive European immigration into Sao Paulo characterize the period here under examination – the four decades from the decline of slavery in the 1880s to the Great Depression.

The geographical area being considered is the coffee region which exported its production through the port of Santos – that is, the interior plateau west and north of the city of São Paulo. The section of the Paraíba River Valley in eastern São Paulo, together with the coffee areas of southern Minas Gerais and Rio de Janeiro state, exported through the port of Rio de Janeiro. The coffee boom in the Paraiba Valley preceded that of the western plateau and was characterized by the slave labour system.

In 1848, nearly fifteen years after the abolition of slavery, Lord Harris, the governor of Trinidad, declared in a dispatch to the Colonial Office that ‘a race had been freed, but a society had not been formed”. Harris was recognizing that although fundamental changes had been wrought in the basic rights of the black population ‘little attention had been paid to any legislation having for its end the formation of a society on true, sound and lasting principles’. The abolition of slavery had established a constitutional setting in which economic and social progress for the bulk of the population was now possible, but abolition alone could not promote that progress.

The abolition of slavery did not dissolve vested interests, established patterns of thought, or the unchallenged place of plantation owners and shareholders at the top of the hierarchy. Freedom in its basic sense did not mean for the black population freedom from the pressures and established advantages of a socio-economic system marked by massive gulfs in the distribution of wealth and privilege. Furthermore the termination of slavery did not itself bring about immediate changes in agricultural practices. The period of apprenticeship (1834–8) held the former slaves to the estates, and when release came the labouring population moved away from the plantations over a period of years. In Trinidad it would appear that labour shortages on the plantations did not become acute until the early 1840s.

As the labouring population moved off the plantations and began to establish itself as an independent peasantry it had to compete, within the framework of colonial society, with the established interests which had long controlled agriculture and land ownership.

Introduction

Whatever interest has been shown to date in the agrarian history of Latin America has largely been concentrated on the colonial period. Even so, our present knowledge about Latin American rural society before Independence remains highly uneven and fragmentary. Only a few regions or districts have been researched in depth. Vital factors such as the movements of rural prices are largely unknown. Moreover, the dependence of scholars on Jesuit documentation illustrates the scarcity of other sources to explain the internal functioning of the large estate.

The present volume therefore presents an ambitious and innovative approach and shows that there is indeed some significant ongoing research being conducted into the agrarian history of the last 150 years. Naturally enough, however, the contributions vary widely, in terms of approach, methodology, and sources; and naturally enough the coverage of regions and types of agricultural production is inevitably somewhat uneven. Some are studies on the macro-level, others on the micro-level. Yet, while far from all of them are actually written by historians, almost all illustrate the historical process of modernization within the agrarian sector of Latin America in the course of the nineteenth and twentieth centuries.

The aim of the present commentary will therefore simply be to place the papers in this volume, and others presented at the Cambridge conference, within a broader historical context, and to relate them to the actual state of historical research, In turn, this double purpose will imply an evaluation of the usefulness of these essays for comparative studies and as a basis for generalizations and theoretical models.

The paradox suggested by the term ‘Andean tropics’ is indicative of the singularity of landlord-peasant relationships as they have evolved over the past four centuries in the Andean drainage basin of the Río Cauca, as in the whole of Colombia. This country was the chief gold producer of the Spanish American empire and had the highest Afro-American population of the Spanish American mainland. Its social and economic history has been firmly etched by the early numerical ascendancy of free ‘mixed bloods’ and the political and property relationships they represented. Nineteenth-century Colombia was unique, even for Latin America, in the diversity of its regionalism, factionalism, and caudillismo. Added to this, the fanatical separation of the entire society into two monopolistic political parties, not dissimilar to religious war machines, has from 1840 until very recently set it apart from the other Latin American republics.

Its composite character brought together all the features that were found separately in the other Latin American colonies. Its topography owes everything to the splayed backbone and dissected valleys of the Andes; yet African slaves, not Andean Indians, formed the basis of its wealth. But in contrast to the other slave colonies, it was not the intensive cultivation of tropical crops on large-scale plantations conveniently located near the coast that made the slaves necessary: rather, it was an economy based on mobile, scattered, and small-scale alluvial gold-mining camps. In short, it was both Caribbean and Andean in its component elements, but unique in its synthesis.

Introduction: the diversity of the Mexican hacienda

In the common usage of Mexico the term ‘hacienda’ simply meant a large estate. It designated an extensive tract of contiguous land held under single ownership, be it of an individual, a family, or an institution. Usually, it implied the presence of a casco, which included a set of buildings to house the mayordomo (administrator) or the owner, and corrals for livestock or barns for grain. It suggested some degree of formal exploitation of the territory in question, either for pasture or for cultivation. The term offered little guide to physical extension. In the Bajío it was customary to restrict its application to properties which at least encompassed a sitio de estancia de ganado menor, that is to say, 18 caballerías or about 780 hectares. Many estates, especially in the arid north, were much larger; the Tetillas in Zacatecas covered no less than 200,000 hectares. By contrast, in Tlaxcala mere farms were at times called haciendas Similarly, the term embraced all types of agricultural activity. A hacienda could be a sugar plantation, a ranch for cattle and sheep, or a farm which grew cereals, regardless of the fact that each class of enterprise differed markedly in its deployment of labour, capital, and land. To compound the problem of definition, a hacienda did not necessarily constitute a single unit of production; in many districts landowners rented a considerable portion of their land to tenant farmers. In this diversity the Mexican large estate resembled its medieval counterpart, of which Georges Duby wrote that ‘The destiny of every manor was individual and its structure unique’.

David Brading's paper constitutes the essential introduction to Part I, since it provides a useful point of departure for examining the development of the hacienda system. The haciendas of the Bajio of Mexico exemplify the economic and social difficulties encountered by pre-capitalist cereal agriculture – low profitability, unstable market conditions, and above all an underlying tension between the demands of demesne cultivation and the encroachments of tenant fanning. Indeed, in the particular case studied by Brading it seems that the landowning class had practically lost this struggle even before the great Mexican agrarian revolution of 1910–17. Jan Bazant's paper takes a similar starting point, showing the conflict between large landowners and the various categories of tenant labour settled on the periphery of their estates. However, in the haciendas of San Luis Potosí, unlike the Bajío, during the latter part of the nineteenth century the landowners were relatively successful in gradually restricting the rights of their tenants and successfully converting their peones into an increasingly impoverished class of day-wage labourers, stripped of the meagre privileges and security of the pre-capitalist agrarian society.

Arnold Bauer and Ann Hagerman Johnson deal in considerable detail with the changes in land tenure and land use during the period of expanding cereal agriculture in Chile. They show that the extension of cultivation took place mainly within the boundaries of existing estates through the conversion of previously unused land, and that there was in fact little or no change in the actual pattern of land ownership. (A similar point is made by Juan Martinez Alier in his paper on the Peruvian highlands.)

This paper presents some tentative conclusions based on the archives of a few expropriated haciendas. The material so far collected in Lima filled five or six lorries, and they will allow several books to be written by several people. A large part of the new archive consists of account books. A much smaller and sometimes very juicy part (perhaps one quarter of a lorry-load in all) consists of correspondence between hacienda managers and landowners, or hacienda managers and comunidad authorities, and so on. All this either has been catalogued or is in the process of being catalogued. It belongs almost totally to the twentieth century. A great digestive capacity will be needed to work through this material. One also needs a stomach as strong as that of a llama (the Andean cameloid), since the diet includes massacred, beaten-up, and blacklisted Indians. However, one of my tentative conclusions is that the Indians’ resistance to the expansion of the haciendas made landowners behave carefully, if not kindly.

A relevant consideration, in order to properly assess the value of the records collected in Lima, is that those haciendas which kept many records are surely not typical, modal haciendas, but they are nevertheless crucial to an understanding of the efforts made at rationalization and the resistance they provoked. One might argue that some landowners were able to expand their haciendas and rationalize their labour systems only because other, more traditional, and wiser hacendados did not follow suit in this dangerous disruption of the rules of the game.

In the late nineteenth century, two crises disturbed the Brazilian sugar economy. A serious market crisis arose when European beet sugar usurped traditional cane sugar markets. A serious social crisis arose within Brazil when the Imperial government moved to abolish slavery. This study examines the sugar planters’ means of dealing with these crises, why they failed in comparison with Cuban sugar planters, and how they nevertheless succeeded in preserving their social position. It also speculates on how the market crisis might have been overcome and how the social crisis might have been handled differently.

The Brazilian planters attempted capital improvements and reorganization of production to cope with the market crisis, but they failed. The only real solution would have been integration into a northernhemisphere market via recolonization, but that would have entailed heavy political costs. The planters coped better with the social crisis. They succeeded in transferring the losses suffered in export markets to the plantation work force in the form of depressed wages and poor working conditions. Their efforts, aided with government subsidies, perpetuated their dominance in Brazil's sugar areas. Thus ‘modernization’, understood as technological advance and the abolition of forced labour, failed to produce general changes.

The economic crisis

The market crisis affected all cane sugar producers, who had lost about one-half of the world market by the end of the century (Table 51). The shift away from cane sugar consumption did not reflect changing tastes, for cane sugar and beet sugar were good substitutes. In fact, the volume of world cane sugar production increased fivefold during the latter part of the nineteenth century.

The agrarian sector continues to present a baffling problem in Chile. Over the past decade, new tenure patterns and labour systems have been formed, reformed, and shattered through movements led by urban cadres – frequently armed with little more than passionately held assumptions. Despite the interest and energy recently focused on the countryside, remarkably little is known about the formation and nature of rural society. Usually illiterate, rural people wrote few memoirs or confessions; not owning land, rarely hired by contract, and usually disposed to toil without complaint (or having little hope of legal redress), they left little trace in the written records and are only dimly perceived through the eyes of travellers. Even in fairly recent times one is forced to infer the workers’ existence from the volume of their output or the nature of the demands placed on them. One way to get at the elusive subject of rural society is by examining the more statistically visible agrarian units – haciendas, medium and small farms – around which labour was organized. That is the aim of the present paper. We present first a statistical picture of changes in land distribution under the impact of expanding markets during the century up to 1935, and then we discuss the way the lower rural society adapted to the new conditions.

Although writers, with tiresome predictability, describe Chile as a ‘long thin land of contrasts’, many are inclined to generalize about the central nucleus. But even here the broken terrain and micro-climates produce a variety of geographic patterns that should be kept in mind.

The nature of the hacienda system: the conflict between landlord and peasant enterprises

In this paper I shall attempt to explain the changes in the Chilean hacienda system which has dominated rural society since colonial times. Although the hacienda system has until recently retained many of its traditional features, largely by maintaining its economic, social, and political predominance in the countryside, nonetheless it has experienced cumulative changes which have gradually undermined the unity of the hacienda as a system, transforming it from a large multi-farm estate, characterized by a complex of traditional landlord–peasant relationships, to a single-farm estate, characterized by farm manager–wage labour relationships. The pressures for change have come from sources largely external to the hacienda system. Among the important causes of the transformation of the Chilean hacienda system must be mentioned the changes in the market (both internal and external), the growth of the urban population, and the increasing radicalization and political organization of the urban working class and later of the peasantry. This analysis will focus on the changes which these factors provoked in the hacienda system, particularly in its labour structure.

Perhaps I should spare a few lines explaining my conceptual approach to the problem, especially in the hope that other analysts might find it useful when studying landlord–peasant relationships in regions dominated by the large landed estate. I shall adopt the ‘multi-enterprise approach’ first presented by Rafael Baraona in his study on the nature of the hacienda in the Ecuadorean Sierra. The multi-enterprise approach views the hacienda system as a conflicting unity between two types of agricultural enterprise: the landlord and the peasant.

During the 1930s, the hostility between the central mill-owners of Pernambuco and their independent cane suppliers, which had been growing since the late nineteenth century, finally reached an acute stage. An outgrowth of this was a body of literature produced by a vigorous school of local writers, who sought to assess the impact on their society of the sugar economy's development over previous decades. Their main argument concerned the excessively high social cost of modernization and the urgent need to halt the destruction of the traditional fabric of the community which it was bringing about.

According to these defenders of the old order, a serious loss was entailed by the growth of the central mills, or usinas, because in the process they were swallowing up the smaller, old-fashioned plantations, known as banguês. This had the result of gradually uprooting the centuries-old class of planters who had presided from the beginning over the nation's destiny and who embodied all that was best in its character. Of no less importance, the usinas were accused of producing a deterioration both in labour relations and in the standard of living of the rural population. The rural labourer, it was claimed, had become a pariah of the usina. Finally, the critique that attempted to vindicate the old engenhos found itself obliged in turn to defend the slave labour system with which they had been identified from the earliest times. Here it was the familiar argument that plantation slaves were often materially better off than their free counterparts locally, or than the proletariat of contemporary industrialized societies.

Introduction

Most studies of plantation systems, and especially those dealing with sugar cane economies, have concentrated on cases where the product is sold for export on the international market, and it has become more or less a commonplace to think of ‘plantation economy’ as being synonymous with ‘export economy’. However, in the case of Argentina there developed in the latter part of the nineteenth century a large and economically important plantation system whose product was destined almost entirely for a protected internal market. This sugar cane economy was based largely in the northwestern province of Tucumán, but later expanded further to the north into the provinces of Salta and Jujuy. Protected by high tariff barriers and liberally financed by banking credits, the Argentine sugar industry of the late nineteenth and early twentieth centuries provides one of the first examples of ‘import substitution’ in Latin America. However, despite these apparently modern and economically progressive features, the internal logic of the plantation system inevitably asserted itself in the field of production relations and patterns of labour recruitment, with the result that, for a certain period, a modern and technologically advanced agro-industry operated labour policies which were manifestly backward and coercive. In Tucuman these features were somewhat mitigated by factors which cannot be dealt with in this paper, but in the provinces of Salta and Jujuy archaic forms of production relations predominated until as late as the early 1940s.