The Burning Season

by Andrew Revkin

  By the mid-1960s, the migration into the region had already begun of its own accord in chaotic fashion. Displaced small farmers and independent ranchers were pouring north along the Belém-Brasília Highway, in the eastern Amazon. In a 50-mile-wide corridor where the road cut through the states of Goiás, Maranhão, and Pará, millions of acres of public land were claimed, cut, and burned.

  The settlers relied on the ancient direito de posse, right of possession, which has been recognized by Brazilian law since 1850 but dates from the colonial period. Under this rule, a posseiro, squatter, who stays unchallenged on a plot for a year and a day and puts it to some “productive” use gains the right to occupy up to 250 acres —or 7,000 acres if the land is not under any local or state control. After five years, he can acquire a paper title. In the Amazon, the simplest way for a settler to prove that land was being put to productive use was to cut down the forest. The area underwent explosive change. According to a study by Roberto Santos, a Brazilian economist and lawyer, between 1949 and 1953 the state of Pará sold 86,000 acres of land. Ten years later, between 1959 and 1963, the state sold 13.8 million acres of land. In the 1970s, somewhere between 200,000 and 400,000 people moved into the region along that first highway.

  This first real land boom in the Amazon resulted in a climate of conflict, fraud, and violence that has been repeated wherever roads have been cut and continues today. Here the process of grilagem, land-grabbing through the creation of false land titles, was turned into a fine art. Offices holding real estate records mysteriously burned to the ground. Bribes were freely given and freely accepted. Sometimes titles were signed over at gunpoint. Soon, there were more titles to land in Pará than there was land. This phenomenon would spread to other parts of the Amazon. In Acre, the situation was made even more complex—and primed for fraud—by the existence of nineteenth-century Bolivian titles along with various layerings of Brazilian titles.

  By 1970, much of the land along the Belém-Brasília Highway had been developed and then abandoned, as the settlers discovered that the underlying soils were too poor to sustain agriculture. Erosion set in, resulting in what one researcher called “a ghost landscape. ”

  In theory, some limits were placed on the destruction. The Law of the Fifty prohibited a landowner from deforesting more than 50 percent of his property. But landowners soon realized that the law had a huge loophole. They could sell the uncut 50 percent to, say, a son; then the new owner could burn off half of the forest on his tract before “selling” the forested part to yet another “owner,” and so on. Other supposed controls were placed on the conversion of land to pasture. If a company wanted to create a cattle ranch on a particular tract, it had to get a certificate from local officials indicating that no peasants or rubber tappers had squatters’ rights to the land. And it had to have the new Indian agency, FUNAI, certify that no Indians occupied the site. In practice, the receipt of such certificates was more often the result of a few well-placed bribes than a field survey. Another alternative was to drive any inhabitants off the land with violence. As the pressure to develop grew during the early 1970s, the ranchers increasingly used bloody means to achieve their end.

  Most of these SUDAM projects were developed along the southern flank of the rain forest, on the Pará-Mato Grosso border. The planners assumed that the presence of certain upland areas of savanna in this part of the Amazon basin was evidence that most of the region could support grassland. And then there was Marajó, the massive island in the mouth of the Amazon. Water buffalo had been raised on its grassy plains for centuries, so obviously the rest of the Amazon could support livestock. As had happened many times before when outsiders confronted the lushness of the region, the planners were fooled into thinking that the earth supported this biological richness; they never imagined that the forests might be great green castles built on sand. Even Alfred Russel Wallace, the naturalist, was duped. He wrote: “I fearlessly assert that here the primeval forest can be converted into rich pastureland, into cultivated gardens and orchards, containing every variety of produce, with half the labor, and what is more important, in less than half the time that would be required at home.” As it turned out, of course, the planners—and Wallace before them—were wrong.

  If all of these ranches had succeeded, the human, environmental, and economic costs of the Amazon development scheme might have been a little less painful. But most of the ranches never came close to reaching their planned potential. For one thing, SUDAM’s hasty approval process neglected to assess carefully the nature of the underlying soil. The surveys that had been done were on too large a scale to discern the wide variations in Amazonian soil quality. (This was long before Brazil had ever heard of environmental impact reports. Even though such reports are now required for major Amazonian development projects, there is still no public debate, and projects frequently still receive a rubber stamp.)

  As pointed out by Susanna Hecht, an agronomist who surveyed soils in the southern Amazon in the 1970s and who has since become one of the world’s experts on the mismanagement of the region, in some areas the use of heavy equipment compressed the soil until it baked into brick-hard laterite. In others, the nutrient-hungry capim grass quickly sopped up many essential minerals. If fertilizers had been used, and other self-fertilizing plants mixed in with the grasses, the soil might have remained productive, but few ranches employed agronomists who might suggest such innovations, and the government failed to provide any technical assistance. And everywhere there was the incessant invasion of fields by inedible weed species and brush. The typical result was that after five years, pasture that might have supported one head of cattle for every 2.5 acres—a bad ratio to begin with—would drop to one head of cattle for every 10 acres. Soon thereafter, the land would simply be abandoned.

  Most of the failures were downplayed in government statistics, and word that the Amazon in general was a bad place for raising cattle did not make it back down south. For some of the investors moving north to start cattle ranches, the low yield and bad soil were irrelevant; they were more interested in the rapidly rising value of the land itself than in the beef that might be raised on it. And everyone liked the idea of free money from the subsidized rural credit offered throughout Brazil. Often loans were available for ranchers at interest rates far below the rate of inflation. (In 1975, for example, when inflation was running at 35 percent a year, the interest rate charged on these loans was 7 percent.)

  For every ranch funded by SUDAM, there were nine more that were created by independent ranchers, according to a report by Dennis Mahar, a World Bank economist. The independents were motivated by either the easy credit or the cheap land. Many of them were small farmers who chose to raise cattle because cattle require little labor, once the land is deforested, and they are a self-transporting commodity—they can walk to the slaughterhouse. (Even among the rubber tappers, a few head of cattle are seen as desirable—as both a hedge against inflation and a way of carrying rubber out of the forest.)

  The impact on the Amazon of the rush to raise cattle was calculated by Robert Repetto of the World Resources Institute. By 1980, 35,156 square miles of forest had been converted to pasture, or 73 percent of the total deforested area of 47,718 square miles. When all of the subsidies and tax breaks and operating costs of the ranches were tallied, it turned out that the incentives to bring cattle ranching to the Amazon cost Brazil $2.9 billion. And after all of that effort, the Amazon was (and still remains) a net importer of beef.

  The ranches did nothing to boost regional employment, either. Studies show that the big ranches employed only one person for every 600 to 700 acres of pasture. And those who did work on the ranches could hardly make a decent living. Except for boiadeiros (cowboys) and pistoleiros, most ranch labor was picked up from contractors called gatos, cats, who would scour the flophouses in Amazonian towns, pay off the debts of unemployed migrants, and force them into indentured labor until they could repay the gatos.

  One rancher after another d
iscovered the disappointments and hazards of raising cattle in the Amazon. Along with erosion and bad soil, they had to cope with malaria, which often felled a third to a half of a ranch’s labor force. Malaria tended to thrive in areas of fresh deforestation, where puddles formed and mosquitoes normally restricted to the treetops descended to the ground. Typical is the story of a São Paulo businessman named Carlos Vilela de Andrade, who bought 100,000 acres of untouched forest a few miles from the Belém-Brasília Highway in 1966. His workers quickly cut and burned the trees, but by 1975 weeds and brush were swamping the pastures and the soil was deteriorating. His initial euphoria had vanished. By 1979, he was broke; a corporate partner in the enterprise had taken over his shares and was trying to sell the ranch, with no takers.

  Along with encouraging private development, the government’s other major ploy to attract people to the Amazon was to build a vast network of roads. To shunt some of the newcomers away from the overburdened lands in the southeastern Amazon, the government hastened to complete the extension of BR-364, the highway that eventually snaked its way from the south of Brazil up to Acre. It was the stretch built from Cuiabá, the capital of Mato Grosso, to Pôrto Velho, the capital of what is now the state of Rondônia, that sparked most of the destruction. Even though the road was not paved until 1980, through the 1970s the population of the area grew at a dizzying pace.

  Most of the migrants came from the south. Between 1970 and 1980, there was a net emigration of two and a half million people from Paraná, for example. Most went to the urban slums, but when word made it back to the south that there were substantial areas of fertile soil along BR-364 in Rondônia, the result was natural enough. Between 1970 and 1985, the population of Rondônia exploded from less than 100,000 to more than 730,000, with most of the new arrivals coming from the south. The state could not begin to support this flood of people—nor could the road. During the rainy season, buses were bogged down for days at a time. Along sparsely populated sections of roadway, there were reports of people dying in their bus seats. And for those who made it to Rondônia, the fertile land was long gone; INCRA was overwhelmed with applications for land. Those who grew tired of standing in line simply headed out beyond the planned grid of feeder roads and burned their way into the forest.

  In 1970, not satisfied with the existing network, the government created a new program of road building and colonization that was intended to accelerate the development of the Amazon further. The so-called National Integration Plan was conceived by the military without any advice from Brazil’s Congress. The plan was abruptly announced in June, after President Emílio Garrastazul Médici made a carefully orchestrated visit to the northeast at the height of an intense drought. There he found the justification to build a dirt highway to nowhere—the Transamazon—after seeing hordes of malnourished children and masses of homeless, hopeless peasants. Médici told reporters, “Nothing in my whole life has shocked and upset me so deeply.”

  The project called for the construction of a 3,400-mile highway running due west from the parched northeast bulge of Brazil, paralleling the main trunk of the Amazon River about 300 miles to the south. Other highways would run north and south. The goal was to lure 100,000 families into the Amazon within six years. It was hoped that 75 percent of those migrants would come from the northeast. Each family would be given title to a 250-acre lot, a prefabricated house, and a minimum salary, around $30 a month, for the first six to eight months. The road building would be coordinated with offers of land, cash, seeds, schools, and health posts overseen by INCRA. Villages, towns, and even small cities would be built from scratch. The government calculated the ancillary benefits: the process of building the roads would create jobs and the construction might uncover mineral deposits. To help sell the plan, the government devised the slogan “A land without people for a people without land.”

  After an initial burst of interest fueled by a barrage of propaganda, the effort fizzled and died. By 1980, only eight thousand families had settled in INCRA lots along the highway. Families who were accustomed to farming the open land of the south and northeast were flown by jet into the rain forest, where they scratched confusedly at the poor soil and watched as unfamiliar pests and weeds hindered their efforts. There was no consistent market for the crops they managed to raise. The oil crises of the 1970s made it too expensive to transport goods and crops from the Amazon. Many colonists were felled by malaria. And the integration program failed to attract the very people it sought: less than a third of the settlers came from the northeast. Some of them began to call the Transamazon the “Transmisery” highway.

  While the organized colonization scheme floundered, the road served a chaotic stream of thousands of settlers who cut their way into the forests and became posseiros. For every INCRA settler, there were four unofficial homesteaders. Ironically, many in this group fared better than those who moved within the system. According to Emílio Moran, an anthropologist who studied the highway project from its inception to the present, about 30 percent of the settlers were seasoned ribeirinhos or rubber tappers who followed the roads just as they once followed an Amazon tributary. By studying the character of the forest, they were able to locate distinctive plant communities that indicated the presence of terra roxa, a particularly fertile soil, and they staked their claims on these buried riches.

  In the end, the Transamazon and the other roads of the National Integration Plan cost the Brazilian government about $39,000 for every settler relocated. No vast mineral riches were uncovered. The landscape of the Transamazon was not the flat plain the planners had presumed from inadequate surveys; it proved to be a nightmare of short, steep hills, resembling the corrugated tin roof of a settler’s hut. At every low spot, the rainy season turned the road into a muddy, malarial swamp. As the settlers cleared and farmed the hills, what little soil remained quickly washed away.

  Moreover, the resettlement schemes did not serve as a population pressure valve. According to a study by Charles Wood and John Wilson, social scientists at the University of Florida, Gainesville, of the estimated 17 million small farmers and peasants who were forced to leave the farmland in the south in the 1970s, only 766,000 ended up in the Amazon—less than 5 percent of the total. The vast majority wound up in the tin and brick favelas that now surround São Paulo, Rio de Janeiro, Curitiba, and other southern cities, magnifying urban problems such as crime, unemployment, and drug abuse. Those few who did head north ended up in a similar setting. Once they abandoned their land, most moved into the cities along the rivers, creating a new galaxy of tropical favelas that is still expanding.

  Despite the failure of the first two strategies for Amazonian occupation—private investment and internal migration—the planners were not dissuaded from trying one more time. In 1973, dozens of Brazil’s wealthiest businessmen were flown to the Amazon by the planning ministry and shown great tracts of undeveloped land that had been made accessible by the new roads. They were wooed with a new package of incentives called Polamazonia, which involved the creation in 1974 of fifteen economic “poles” that would act like magnets to draw industry and people into targeted regions.

  One of the biggest poles was the Carajás mining project, which eventually grew to encompass 300,000 square miles, fully a sixth of Brazil’s portion of the Amazon basin. This project was centered around a chain of low mountains 340 miles south of the port of Belém. In 1967, a geologist working for U.S. Steel had discovered that this Serra dos Carajás range was rich in iron ore. Later it was shown to be the largest known deposit of iron ore in the world—roughly 18 billion tons of 66 percent pure ore—enough to make Brazil a leading producer for several centuries to come. The mountains also contain copper, manganese, nickel, and gold, among other minerals. The iron mine itself was developed carefully, to minimize environmental damage, but related projects became environmental disasters.

  Through the 1970s, Brazil created a $60 billion plan to develop the region, funded by loans from Japan, the United States, Euro
pe, and the World Bank. It included the construction of a 534-mile railway to the coast, a series of hydroelectric dams, ten new cities to house workers, and a network of improved roads to encourage smelters and other industries to move into the area. One destructive offspring of the Carajás mines was pig-iron smelters, plants that extract crude iron from the ore with heat. But their huge furnaces are fueled by charcoal. Eventually, the plan is to make charcoal by burning the wood from plantations of eucalyptus trees, which grow quickly. But right now, the kilns that produce the charcoal are being fed wood that is obtained the cheapest possible way—by cutting down great tracts of rain forest. If the number of smelters grows as fast as projected, the charcoal kilns will consume as much as 900 square miles of forest per year.

  Another pole of the Polamazonia project was a series of hydroelectric dams such as Tucurui, a 12-mile arc of concrete built across the Tocantins, one of the largest tributaries of the Amazon. The reservoir that formed behind this dam when its sluices were closed in 1984 has replaced 1,461 square miles of rain forest with a stewing lake full of 50 million tons of rotting timber. The original plan was for the timber to be harvested before the flood. But the firm created by the government to do the job became mired in corruption and the basin was flooded before any wood could be saved. A handful of adventurous entrepreneurs have since taken to harvesting some of the valuable trees submerged in the lake using scuba gear and hydroserras, chain saws converted to underwater use.

  Normally, hydroelectric power is considered “clean” power, but in this case the rotting vegetation in the reservoir is emitting thousands of tons of methane each year. Like the carbon dioxide that is given off when trees burn, methane is a “greenhouse” gas, one that traps the sun’s energy and may contribute to global warming.