Surveying the agricultural and economic prospects of the Boconó region in the Venezuelan Andes in 1841, Italian geographer Agustin Codazzi wrote of a valley where farms grew “wheat, barley, sugarcane, coffee, cotton, plantains, potatoes, maize, peas, chick peas, manioc, etc.,” for a thriving regional economy. Yet Codazzi was convinced that Boconó’s future lay in the mountainsides covered in subtropical forest that surrounded the valley: “It cannot help but advance along the path to prosperity, when the jungles that look out on the lowlands are cleared and a numerous population has built the road to the banks of the Motatán, where they will ship the produce of the mountainous heights, today uncultivated.”[1]
In succeeding decades, Boconó was to realize Codazzi’s vision, as settlers from the nearby llanos headed for the mountainsides, cleared the forest (a process that lives in local memory as tumbando montes), planted “new forests” of coffee and shade trees, and cleared more land for food crops.[2] By 1881, coffee, which figured as one of many products in Codazzi’s description, had become the principal product of a region firmly connected to the world market. With the passage of another two decades, contrary to Codazzi’s hopes, the Boconó region was already in decline, its soils worn out by the extensive cultivation of coffee.
The rapid transition from subtropical forest to thriving coffee export economy, and then to region in decline, characterized large areas of nineteenth-century Latin America.[3] Coffee was the motor of vast ecological change. Cutting down original forests and reforesting with new species transformed the natural habitat for many plants and animals. Population movements that accompanied coffee brought further changes to flora and fauna. Within as little as 40 years, soils were so depleted that large territories became useless for agriculture and were abandoned to ranchers and their cattle.
In southeastern Brazil, the expansion began relatively early in the nineteenth century, in the Paraiba Valley of Rio de Janeiro. By the last decades of the century, its soils depleted and its coffee groves in decline, Rio was displaced by São Paulo, and a process of forest clearing, coffee planting, and soil depletion was begun and repeated as planters and pioneers moved farther west across the immense plateau.[4] By the 1950s, the coffee frontier had moved into the state of Paraná, where soils were less fertile and the cycle of clearing/boom/depletion/decline was compressed into a decade.[5]
In Colombia, the move toward coffee began somewhat later and incorporated three regions in successive cycles––the Santanders from 1850 into the 1870s, the central Andean provinces of Cundinamarca and Tolima from the 1870s into the early twentieth century, and finally the colonization of the Antioquia frontier south of Medellin, where coffee boomed front the early twentieth century on.
In Costa Rica as well, coffee precipitated forest clearing and human settlement in the Central Valley, beginning in the early nineteenth century. The Venezuelan Andes and the mountains of western Puerto Rico began their coffee cycles by mid-century and were either stagnating (Venezuela) or in evident decline (Puerto Rico) by early in the twentieth. Finally, the piedmont zones of Guatemala and western El Salvador began their coffee cycles in the 1870s and 1980s and continue to this day.
In most regions, the coffee boom occurred on some sort of frontier and involved expansion into uncultivated lands, the cutting of forests, the establishment of new villages, towns, and cities, and the building of roads and railroads. In some regions, such as Boconó in Venezuela, the Santanders in Colombia, the western slopes in El Salvador, or the better known case of the Central Valley in Costa Rica, the transformation of forest into coffee farms emanated from older towns and villages that had served as the centers of regional economics. Elsewhere, as in the western interior of Rio de Janeiro, São Paulo, or Minas Gerais in Brazil, the lower slopes of Cundinamarca and Tolima or in Antioquia in Colombia, of the western slopes of Puerto Rico, coffee cultivation involved the expansion and settling of frontiers, the transformation of wilderness into settled and cultivated space, and the movement of vast number s of people.[6]
Coffee’s expansionist dynamic, which wrought such destruction, can be traced to both the physical properties of the plant and the social relations under which it was produced. Coffee is a perennial which bears fruit three to five years after planting, and is then productive for 25 years or more. In the absence of an active program of fertilizing, pruning, weeding, and pest control, coffee yields will follow a curve that peaks five to ten years after planting––due to the natural fertility of recently felled forest and the initial flowerings of young trees. It then declines steadily, and sometimes dramatically, due to the decreasing fertility of thin subtropical soils and the aging of the trees themselves.
Although coffee is a “permanent” crop, representing a long-term commitment of resources, establishing a new farm or grove is relatively simple and inexpensive. Much of the labor is invested in clearing and planting. A small farm household can maintain itself planting corn and beans between the rows of coffee trees in the early years; a plantation owner can attract laborers to a new farm by granting the right to plant food crops in the coffee groves before the coffee bears fruit. (Even mature coffee groves produce food, as it is common in many regions of Latin America to plant banana trees for shade.)
There is, then, a “natural” tendency for coffee groves and farms to be highly productive in the first half of a 40-year cycle and in decline for the second half. Such a tendency could be offset by the maintenance and enhancement of soil fertility and the systematic replacement and renovation of coffee trees and groves. But as long as “virgin” forests are accessible and available, it can be cheaper to open a new farm than to maintain and improve an old one.[7]
Brazil is the most dramatic example of environmental devastation wrought by coffee. As large planters moved into the Paraiba Valley in the early nineteenth century, they encountered massive forests with wild figs, sucopira, jacaranda, cabiuna, jequitibá, and cedrela trees. Lianas formed a high canopy connecting the well spaced trees. To carve out coffee plantations, planters often directed their slaves to partially cut a whole group of trees on a hillside and then fully cut one tree at the top. The connecting vines would bring the whole grove down. Clear-cutting the forest in this way robbed the tapirs, pacas, and capybaras, as well as the birds that lived in the forest canopy, of a habitat.
After burning and appropriating some of the lumber for plantation houses and buildings, the charred trunks were then left pointing downhill, and the coffee seedlings were planted in vertical rows up the hillside. This allowed planters to leave the massive trunks in place (if felled horizontally, the trunks would eventually roll downhill and uproot the coffee trees). This, however, caused severe erosion.
Given the abundant land and dependent labor force, planters adopted production techniques that made for quick profits. Initial productivity depended upon the natural fertility of the forest. At harvest, trees were stripped of cherries and leaves: Pickers encircled a branch with thumb and forefinger and stripped every-thing––leaves, ripe and green cherries––into a screen. Such methods assured the productivity of labor but not of land; indeed, with the erosion caused by the vertical rows, they assured that the land would be exhausted at the end of the 20- to 30-year cycle of the coffee trees themselves.
Stanley Stein provides a vivid description of the aftermath of the coffee boom in the Vassouras region of Rio: “Abandoned to grass, and stretching up and down the rounded hills, the rachitic files of aged and unproductive coffee trees had no peaceful interment. Such vast expanses of rolling hillsides could not be left in unprofitable abandonment. On the path of those who almost a century before had moved backward toward the coast from the frontier areas of the Minas gold fields, came a second invasion of settlers front Minas. This time they did not come to learn to handle a new crop. They brought with them cattle, whose meat and milk could provision the expanding population of the capital of the Republic. In Vassouras as in other municipios of the Paraiba Valley, the new settlers found pasturage cheaper than in the southern portions of Minas. Easily they could turn loose their cattle to feed on the grasses––capim angola, capim membeca, capim gordura––spreading quickly through the untended coffee groves. As a resident expressed the change: ‘Pasture invaded the coffee groves.’”[8]
As the coffee economy’s center of gravity shifted to the western part of São Paulo, the forest/coffee/pasture cycle was attenuated somewhat by the fact that the terraroxa (purple soil) was extremely fertile and the plateau less hilly. However, the westward movement continued, and the most productive groves were found in zones of recently felled forest. Warren Dean estimates that 30,000 square kilometers of forest were felled in southeastern Brazil over the nineteenth century for the planting of coffee alone.[9]
Though coffee spawned a movement into new lands and the founding of new towns in Costa Rica, especially to the west of the central mesa, this was not fueled by soil exhaustion. Lacking a dependent labor force, the small and medium-sized growers of the central mesa fertilized their coffee trees, replaced aging groves, and maintained their patrimony. The western migration reflected the insufficiency of that patrimony for inheriting children, who would move to the west and plantfood crops, pasture, and, eventually, coffee. Deforestation had the same initial effects as in Brazil––the loss of habitat for wildlife and the contraction of floral and faunal diversity––but it did not lead to soil erosion.[10]
Costa Rica’s Central Valley is one of the few regions of Latin America where coffee became essentially a monocrop regime, as growers devoted all of their lands to coffee and bought food produced in neighboring zones.[11] Virtually alone among nineteenth-century coffee producing regions, the Central Valley presented a more uniform appearance––a sea of coffee trees.
In Colombia, the expansion of coffee in Cundinamarca/Tolima and the west of Antioquia also involved forest destruction, but again with effects different from the Brazilian case. In the first place, coffee did not occupy large extensions of territory, even in the most important coffee zones. It has been estimated, for example, that on a late-nineteenth-century Colombian hacienda coffee might occupy only a quarter of a hacienda’s land, the rest devoted to pasture, sugar cane, food crops, and forest reserve.[12]
Such a distribution is perhaps best understood in terms of the prevailing labor regime. Hacienda owners in Cundinamarca/Tolima were, for the most part, urban merchants based in Bogotá who invested in subtropical estates, turning hacienda affairs over to a resident administrator. The administrator would hire outside labor for the harvest, but would contract with renters to tend the coffee groves. Renters were granted access to extensive lands for their own use. On these they planted food crops and sugar cane, both for their use and for sale in an alternative local commercial economy within and beyond the confines of the hacienda. The decline of Cundinamarca estates in the twentieth century was due less to declining soil fertility than to the increasing power of the tenantry, who pressed political demands in the 1920s and 1930s.
The colonization of Antioquia was organized by entrepreneurs who reserved valley lands for cattle ranches and granted plots on the less attractive hillsides to laborers. As these new smallholders planted coffee, they created an entirely different kind of alternative economy. As in Costa Rica, these smallholders maintained the viability and fertility of their groves and did not contribute to the kind of expansive and destructive waste that characterized Brazil. In both cases, forest was cleared, but rather than a forest/coffee/pasture progression, coffee and pasture were created in proximity and in symbiotic relation.[13]
The contrast between the expansive and wasteful large estate in Brazil and the forestfelling but soil-conserving small or middle- sized farm in Costa Rica or Antioquia should not lead us to either idealize the smallholder or ignore the significant transformative and expansive pressures to which they were subject. In Costa Rica, a crisis in inheritance pushed generations of settlers into new regions. Debt provided another dynamic for spatial expansion in Costa Rica, Antioquia, and elsewhere. Smallholders were producing a commercial crop, which placed them in direct relation with urban merchants, a relation cemented by debt. Debts were paid off in coffee rather than currency; in periods of low prices, more coffee was required. Since smallholders might be in a poor position to increase the fertility of aging groves, a common response was to fell more forest and plant new groves, even though these would take several years to mature.
In considering the different environmental effects of coffee cultivation in these three countries, we can discern what Robert Redfield once called a kind of “writing on the land of the vague outlines of a social structure.”[14] Though in each of these regions the turn to coffee involved frontier expansion and forest destruction, each occurred within distinct fields of power and involved the crystallization of different class regimes. In Brazil, the expansion into the frontier was also the expansion of the great estate. Plantation owners began with estates of at least one square league (44 square kilometers). The planters in Rio depended upon a slave labor force; later planters in São Paulo, faced with the abolition of slavery, pressed the state to initiate a state-subsidized immigration program, by which millions of (largely Italian) colono families were recruited. Land and labor were cheap, and fertile forest lands seemed endlessly abundant.
Costa Rica, on the other hand, had been a thinly settled periphery of a periphery during the colonial period. The population of the Central Valley was concentrated in towns and villages. After independence, the commercial elite’s search for a profitable export commodity, and for immigrants who might grow such a crop, led them to adopt liberal homesteading laws that granted land to settlers who would agree to plant coffee. For these settlers, land was never abundant, and labor was neither dependent nor cheap. In Colombia, we find a mixture of smallholder and hacienda production. The smallholders faced pressures and adopted practices similar to those that characterized Costa Rica; the hacienda owners viewed their estates as but one of a diversified field of activities and turned the administration and organization of the estates over to their tenants.[15]
Observing the coffee regions of Latin America today, it may be hard to envision the creative destruction and expansion that reshaped landscapes in earlier generations. Some regions, such as the Boconó countryside with which we began, have experienced marked decline since the 1930s. Others, such as large sections of the São Paulo countryside, were converted to pasture, corn, rice, and other crops during the 1960s.[16] Still others, such as Costa Rica’s Central Valley, Antioquia in Colombia, western São Paulo, and the Guatemalan and Salvadoran piedmont are today the most important coffee producing regions of Latin America.
The dynamics of coffee production and its environmental effects have changed. New regions entered the coffee trade in the past half century; aside from Africa, Southeast Asia, and Oceania, which were the areas of greatest expansion, Latin American production spread to the eastern slopes of the Ecuadorian and Peruvian Andes, and to southern Mexico. And coffee is no longer the crop most directly implicated in frontier expansion and forest destruction. Physically, the coffee plant has changed little, despite the development of new strains. But the social, economic, and political contexts have changed dramatically. The outlines of quite different social structures are now written on the land.
First, the market environment is different. One of the responses to the 1930s depression was the creation of various state marketing boards that began to control international supply. The International Coffee Agreement of 1940, signed by the United States and 14 producing countries at the onset of World War II, was the first of a series of such agreements, imposing annual export quotas on particular countries. The free market environment in which merchants and farmers in various countries might respond to a market boom by cutting the forest and planting coffee has disappeared.
Second, the control schemes were but one feature of a ramified state presence in various coffee countries, including coffee promotion boards, buying and milling cooperatives, credit banks, and agricultural experimentation and extension programs. Since mid-century, such programs have developed more productive strains of coffee and regular regimes of fertilization, maintenance, and renovation; agricultural credit agencies have made it possible for farmers to undertake such extensive regimes. Though coffee trees are still subject to a 25-year cycle, coffee farmers are able to maintain the fertility of the soil and the productivity of their farms. However, these practices also have environmental implications: the new strains require chemical fertilizers, and the prescribed programs of grove maintenance entail heavy applications of chemical pesticides and insecticides, which contaminate soil and water resources.
Third, as large and small coffee farms turn to the newer strains of coffee, they devote increasing portions of their land to coffee groves.[17] One of the more remarkable tendencies is the decreasing importance and incidence of food production, and the consequent disappearance of alternative local economies. A number of factors have played a role here, including the growth of transportation infrastructure and distribution systems, leading to the greater availability and lower cost of imported foods.
But social and political factors have been central as well. Most importantly, in a variety of coffee producing countries, land reform laws since the 1960s granted tenants a number of rights to agricultural improvements. Estate owners and middle-sized farmers, faced with the prospect of having to compensate tenants for the groves they tended, ceased to keep tenants at all, turning instead to casual wage labor. São Paulo estate owners cast out their former colonos, but found that the Rural Labor Statute of 1963 extended the benefits and protection of labor law to rural laborers as well. They therefore turned to casual laborers in nearby towns and cities––often former colonos or their children––who are hired as gangs and purchase their food in urban markets.[18] The expulsion of tenants has had the direct effects of curtailing food production, expanding the land devoted to coffee, and reducing the diversity of cultigens in coffee-growing zones. Coffee in São Paulo has come to resemble more closely the monocrop, regions of other plantation crops.
The visitor to a coffee zone in Brazil, Colombia, Costa Rica, or Guatemala might therefore see a type of “new forest,” less diverse than the subtropical forest to be sure, but also less diverse than the new forests established in coffee zones in the nineteenth century. Today one is more likely to see a large and uniform expanse of coffee trees, as both food crops and their growers have been displaced.
However, small farmers who combine coffee with food production and treat coffee as an income supplement remain significant in many regions. On the margins of the vast groves, one still sees smaller groves, interplanted with banana trees, alongside fields of corn, beans, sugar cane, or pasture, Environments and modes of livelihood that were new in the 1980s take on the tenacity and the appearance of timeless tradition.
The new forests of the nineteenth century are no longer new; nor do they present the appearance of a devastated environment. Though coffee groves have been overrun by pasture in old production zones in Brazil, or uprooted in response to more recent political and economic developments, many of the old production zones in Latin America are still planted in coffee, and the groves are verdant. In areas that require shade, an old grove will be interplanted with flowering bucare or other trees. From a distance, a hillside planted in coffee and shade may seem to be forested; within the grove, the trees’ beauty, in flower and fruit, cannot but affect one’s senses of sight and smell. It is easy to forget that the expansion of coffee was one of the most important forces in frontier settlement and subtropical deforestation in the nineteenth and early twentieth centuries. And that coffee today continues to reduce the floral and faunal diversity that was forever lost when the old forests were felled.
ABOUT THE AUTHOR
William Roseberry teaches anthropology at the New School for Social Research.
NOTES
1. Agustin Codazzi, Resumen de la Geografía de Venezuela, Vol. 3. (Caracas: Biblioteca Venezolana de Cultura, 1940, originally 1841). Cited in William Ruseberry, Coffee and Capitalism in the Venezuelan Andes (Austin: University of Texas Press, 1993), p. 87.
2. The term “new forests” is from M. Palacios, EI Café en Colombia, 1850-1970, 2nd ed. (Mexico City: EI Colegio de Mexico, 1983), p. 178
3. Coffee consumption was introduced in Europe in the seventeenth century, and the first coffee beans were brought to the New World in the eighteenth century––first to Martinique, then Haiti, then to various Caribbean islands, the Guianas, and northern Brazil. But the nineteenth century (that is, roughly from 1830-1930) was the coffee century in Latin America, characterized by a dramatic increase in world trade (from 320 metric tons in 1770, mostly from Asia, to 90,000 metric tons in 1970 and 1,600,000 metric tons in 1920) and per capita consumption (in the United States, from 3 pounds per year in 1830 to 10 pounds in 1900 and 16 pounds in 1960). The First generalized crisis of overproduction, did not occur until the beginning of the twentieth century, causing the state of São Paulo (and later Brazil) to withhold coffee from the market, the first of a variety of state sponsored control schemes that were to dominate the trade from then on. For statistics on global production and consumption, see J. de Graaf, The Economics of Coffee (Wageningen: Netherlands, 1986), p. 26; and U.S. Department of Commerce, Coffee Consumption in the United States (Washington, D.C., 1961), p. 5.
4. S. Stein, Vassouras, A Brazilian Coffee County, 1850- 1900: The Roles of Planter and Slave in a Plantation Economy, 2nd ed. (Princeton: Princeton University Press, 1985); W. Dean, Rio Claro: A Brazilian Plantation System, 1820-1920 (Stanford: Stanford University Press, 1976); W. Dean, “Deforestation in Southeastern Brazil,” Global Deforestation and the Nineteenth-Century World Economy, ed. Richard P. Tucker and J. F. Richards (Durham: Duke University Press, 1983), pp. 50-67.
5. M. Margolis, The Moving Frontier (Gainesville: University of Florida Press, 1973).
6. The older regional centers had generally been marginal to their respective colonial economies, relatively inaccessible to port cities, located in mountainous terrain, and not so densely populated. The coffee market made possible the expansion of export production into these zones. The shelled bean could be stored for long periods without spoilage, and it had a high value per kilogram, therefore able to support the refictivety high transport costs from distant or inaccessible interiors. Local growers or merchants could organize mule trains over the mountains into port cities, and the revenues from coffee could support road and railroad projects to “open up” and expand the frontiers of settlement. See L. Bergad, Coffee and the Growth of Agrarian Capitalism in Nineteenth Century Puerto Rico (Princeton: Princeton University Press, 1983), p. 38.
7. This depends on the costs of land and labor, of course, which in turn depend on social and political factors.
8. Stein, Vassouras, p. 286.
9. Dean, “Deforestation,” pp. 62-63. This was not the sole factor contributiong to forest destruction, however. To this must be added the clearing of forest for pasture and the planning of other crops ter local consumption. Also, the settling of the frontier and the establishment of towns and cities ever further west created a demand for lumber and firewood. And the construction of 6,000 km of railroads in southeastern Brazil presented a huge and continous demand for lumber for cross-ties and, most importantly, fuel. See ibid., pp. 63-64.
This discussion also relies on Stein, Vassouras, and Dean, Rio Claro. For a dramatic tuitional account of the effects of fuel demands for transport on forest resources, see G. García Márquez, Love in the Time of Cholera (New York: Penguin, 1988), pp. 336-337.
10. Essential sources on Costa Rica include C. Hall, El Café y el desarrollo Histórico Geográfico (San Jose, 1976); L. Gudmundson, Costa Rica Before Coffee (Baton Rouge: Louisiana State University Press, 1986); M. Samper, Generations of Settlers: Rural Households and their Markets on the Costa Rican Frontier (Boulder: Westview, 1991).
11. Hall, EI Café, y el Desarrollo. This reflects the fact that the micro region of the Central Valley is relatively small and accessible to other regions. In a large region like São Paulo, growers elected instead to grant food plots to tenants within the estate as part of their compensation package.
12. Palacios El Café en Colombia, p. 209.
13. In both Cundinamarca and Antioquia, pasture lands were also neccessary for mules to transport coffee to port towns. Essential sources on Colombia include Palacios, El Café en Colombia; M. Arango, Café e Industria: 1850-1930 (Bogotá, 1977); M. Jiménez, “Traveling Far in Grandfather’s Car: The Life Cycle of Central Colombian Coffee estates. The case of Viotá, Cundinamarca (1900-1930)” Hispanic American Historical Review, No. 69, (1989); C. Bergquist, Labor in Latin America: Comparative Essays on Chile, Argentina, Venezuela, and Colombia (Stanford: Stanford University Press, 1986), pp. 274-375.
14. R. Redfield, The Little Community/Peasant Society and Culture (Chicago: University of Chicago Press, 1960), p. 30.
15. Further comparisons of Brazil, Costa Rica, and Colombia can be found in W. Roseberry, “La Falta de Brazos: Land and Labor in the Coffee Economics of Nineteenth Century Latin America,” in Theory and Society No. 25 (1991) pp. 351-382.
16. V. Stolcke, Coffee Planters, Workers, and Wives: Class Conflict and Gender Relations on São Paulo Plantations, 1850-1980 (New York: St. Martin’s, 1988), p. 100.
17. In addition, the new strains of coffee are planted much more densely, diminishing the space between rows available for food production in new groves.
18. Stolcke, Coffee Planters, Workers, and Wives, pp. 108-132.
